Organ Transplantation A Clinical Guide

Organ Transplantation A Clinical Guide Edited by Andrew A. Klein Consultant, Anaesthesia and Intensive Care, Papworth Hospital, Cambridge, UK Clive J. Lewis Consultant Cardiologist and Transplant Physician, Papworth Hospital, Cambridge, UK JorenC.Madsen Director of the MGH Transplant Center, Section Chief for Cardiac Surgery, and W. Gerald and Patricia R. Austen Distinguished Scholar in Cardiac Surgery, Massachusetts General Hospital, Boston, MA, USA CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao˜ Paulo, Delhi, Tokyo, Mexico City Cambridge University Press The Edinburg Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521197533

c Cambridge University Press 2011

This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press.

First published 2011

Printed in the United Kingdom at the University Press, Cambridge

A catalogue record for this publication is available from the British Library.

Library of Congress Cataloguing in Publication data Organ transplantation : a clinical guide / edited by Andrew Klein, Clive J. Lewis, Joren C. Madsen. p. ; cm. Includes bibliographical references and index. ISBN 978-0-521-19753-3 (hardback) 1. Transplantation of organs, tissues, etc. 2. Transplantation immunology. I. Klein, Andrew. II. Lewis, Clive J., 1968– III. Madsen, Joren C., 1955– [DNLM: 1. Organ Transplantation. 2. Transplantation Immunology. WO 660] RD120.7.O717 2011 617.95 – dc22 2011002165

ISBN 978-0-521-19753-3 Hardback

Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Every effort has been made in preparing this book to provide accurate and up-to-date information which is in accord with accepted standards and practice at the time of publication. Although case histories are drawn from actual cases, every effort has been made to disguise the identities of the individuals involved. Nevertheless, the authors, editors and publishers can make no warranties that the information contained herein is totally free from error, not least because clinical standards are constantly changing through research and regulation. The authors, editors and publishers therefore disclaim all liability for direct or consequential damages resulting from the use of material contained in this book. Readers are strongly advised to pay careful attention to information provided by the manufacturer of any drugs or equipment that they plan to use. Contents

List of contributors vii Foreword xi Preface xiii List of abbreviations xv

Section 1 – General 10 Management during surgery 88 Kate Drummond and Andrew A. Klein 1 Historical perspectives 1 John Dunning and Sir Roy Calne 11 Postoperative care and early complications 94 2 Immunological principles of acute Mandeep R. Mehra rejection 9 FadiG.Issa,RyoichiGoto,andKathrynJ.Wood 12 Long-term management and outcomes 102 3 Immunosuppression: Past, present, Hari K. Parthasarathy and Clive J. Lewis and future 19 Vineeta Kumar and Robert S. Gaston 13 Pediatric heart transplantation 112 Jacob Simmonds and Michael Burch 4A Major complications – cancer 31 Bimalangshu R. Dey and Thomas R. Spitzer 4B Major complications – pathology of Section 3 – Lung chronic rejection 38 14 Recipient selection 122 Yael B. Kushner and Robert B. Colvin J.S. Parmar 4C Major complications – infection 46 15 Living donor lobar lung Camille Nelson Kotton transplantation 128 5 Organ donor management and Hiroshi Date procurement 53 16 Surgical procedure 133 Edward Cantu III and David W. Zaas Faruk Ozalp,¨ Tanveer Butt, and Stephan V. B. S c hu e l e r Section 2 – Heart 17 Management during surgery 138 David Ip and Peter Slinger 6 Recipient selection 63 R.V. Venkateswaran and Jayan Parameshwar 18 Postoperative care and early complications 145 7 Donor heart selection 70 Vlad Vinarsky and Leo C. Ginns KiranK.KhushandJonathanG.Zaroff 19 Long-term management and 8 Ventricular assist devices 76 outcomes 155 DavidG.HealyandStevenS.L.Tsui Paul Corris 9 Surgical procedure 83 20 Pediatric lung transplantation 164 R.V. Venkateswaran and David P. Jenkins Stuart C. Sweet and Samuel Goldfarb

v Contents

Section 4 – Liver 33 Pediatric kidney transplantation 278 Khashayar Vakili and Heung Bae Kim 21 Recipient selection 173 Alex Gimson 22 Living donor liver transplantation 182 Section 6 – Other abdominal organs Koji Hashimoto, Cristiano Quintini, and 34 Pancreatic transplantation 286 Charles Miller Dixon B. Kaufman 23 Surgical procedure 190 35 Pancreatic islet transplantation 295 SimonJ.F.HarperandNevilleV.Jamieson Heidi Yeh and James F. Markmann 24 Peri-operative care and early 36 Intestinal transplantation 303 complications 199 Stephen J. Middleton, Simon M. Gabe, John Klinck and Andrew J. Butler Neville V. Jamieson, and Andrew J. Butler 25 Long-term management and outcomes 212 William Gelson and Graeme J.M. Alexander Section 7 – Other 37 Composite tissue allotransplantation: 26 Pediatric liver transplantation 220 Face transplantation 313 Hector Vilca-Melendez and Giorgina Maria Siemionow and Can Ozturk Mieli-Vergani 38 Hematopoietic stem cell transplantation 320 Section 5 – Kidney Charles Crawley and Thomas R. Spitzer 27 Recipient selection 231 39 Corneal transplantation 330 Ernest I. Mandel and Nina E. Tolkoff-Rubin Yvonne H. Luo and D. Frank P. Larkin 28 Sensitization of kidney transplant recipients 238 Nick Pritchard Section 8 – The transplant service 40 UK and European service – legal and 29 Live donor kidney donation 248 operational framework 335 Arthur J. Matas and Hassan N. Ibrahim ChrisJ.RudgeandAxelO.Rahmel 30 Surgical procedure 253 41 US transplant service – legal and Paul Gibbs operational framework 347 31 Peri-operative care and early Walter K. Graham, Richard S. Luskin, and complications 258 Francis L. Delmonico Lorna Marson and John Forsythe 42 Conclusions 355 32 Long-term management and Clive J. Lewis outcomes 265 Sharon Mulroy and John D. Firth

Index 357 Color plate section falls between pages 181 and 182.

vi Contributors

Graeme J.M. Alexander, MA, MD, FRCP Charles Crawley Consultant Hepatologist, Addenbrooke’s Hospital Consultant, Department of Hematology, Cambridge, UK Addenbrooke’s Hospital, Cambridge, UK

Heung Bae Kim, MD Hiroshi Date, MD Assistant Professor of Surgery, Harvard Medical Professor, Department of Thoracic Surgery, Kyoto School, and Director, Pediatric Transplant Center, University Graduate School of Medicine, Children’s Hospital Boston, Boston, MA, USA Kyoto, Japan

Michael Burch Francis L. Delmonico, MD Lead Transplant Consultant, Consultant Cardiologist, Medical Director, New England Organ Bank, and Great Ormond Street Hospital for Sick Children, Professor of Surgery, Harvard Medical School, London, UK Massachusetts General Hospital, Boston, MA, USA

Andrew J. Butler, MD Bimalangshu R. Dey, MD, PhD Consultant, Department of Transplantation Surgery, Bone Marrow Transplant Program, Department of Addenbrooke’s Hospital, Cambridge, UK Medicine, Massachusetts General Hospital, and Harvard Medical School, Boston, MA, USA Tanveer Butt, FRCS Department of Cardiopulmonary Transplantation, Kate Drummond, MD TheNewcastleuponTyneHospitalsNHSFoundation Fellow, Anaesthesia and Intensive Care, Papworth Trust (NUTH), Freeman Hospital, High Heaton, Hospital, Cambridge, UK Newcastle upon Tyne, UK John Dunning, MD Roy Calne, MD Consultant Cardiothoracic Surgeon, Papworth Yeah Ghim Professor of Surgery at the National Hospital, Cambridge, UK University of Singapore, Singapore John D. Firth, DM, FRCP Edward Cantu III, MD Consultant Physician and Nephrologist, Cambridge Associate Surgical Director of Lung Transplantation, University Hospitals Foundation Trust, University of Pennsylvania, Philadelphia, PA, USA Cambridge, UK

Robert B. Colvin, MD John Forsythe, MD Department of Pathology, Massachusetts General Consultant Transplant Surgeon, Royal Infirmary of Hospital, and Harvard Medical School, Boston, Edinburgh, Edinburgh, UK MA, USA Simon M. Gabe Paul Corris, MB, FRCP Lennard Jones Intestinal Failure Unit, St Mark’s Professor, Department of Respiratory Medicine, Hospital, Northwick Park, Harrow, Freeman Hospital, Newcastle upon Tyne, UK London, UK

vii Contributors

Robert S. Gaston, MD, MRCP David G. Healy, PhD, FRCSI (C-Th) Endowed Professor, Transplant Nephrology, and Honorary Fellow, Department of Cardiothoracic Medical Director, Kidney and Pancreas Surgery, Papworth Hospital, Cambridge, UK Transplantation, Division of Nephrology, University of Alabama at Birmingham, Hassan N. Ibrahim, MD Birmingham, AL, USA Department of Medicine, University of Minnesota, Minneapolis, MN, USA William Gelson, MD, MRCP David Ip, MBBS, FANZCA Consultant Surgeon, Department of Transplant Surgery, Addenbrooke’s Hospital, Cambridge, UK Anaesthesia Fellow, Toronto General Hospital, Toronto, ON, Canada Paul Gibbs Fadi G. Issa, MA, BMBCh, MRCS Consultant Surgeon, Department of Transplant Clinical Research Fellow, Nuffield Department of Surgery, Addenbrooke’s Hospital, Cambridge, UK Surgery, University of Oxford, Oxford, UK Alex Gimson Neville V. Jamieson Consultant Physician, Department of Hepatology, Consultant Transplantation and HPB Surgeon, Addenbrooke’s Hospital, Cambridge, UK University of Cambridge and Addenbrooke’s Leo C. Ginns, MD Hospital, Cambridge, UK Medical Director, Lung Transplantation, Massachusetts General Hospital, and Associate David P. Jenkins, MB BS, FRCS (Eng), Professor of Medicine, Harvard Medical School, MS (Lond), FRCS (CTh) Boston, MA, USA Consultant Cardiothoracic Surgeon, Papworth Hospital, Cambridge, UK Samuel Goldfarb, MD Attending Physician, Division of Pulmonary Dixon B. Kaufman, MD, PhD Medicine, Children’s Hospital of Philadelphia; Professor and Chief, Division of Transplantation, Medical Director, Lung and Heart/Lung Transplant Department of Surgery, University of Wisconsin, Programs, and Assistant Professor of Pediatrics, Madison, WI, USA University of Pennsylvania School of Medicine, Philadelphia, PA, USA Kiran K. Khush, MD, MAS Division of Cardiovascular Medicine, Department of Ryoichi Goto, MD Medicine, Stanford University, Stanford, CA, USA Clinical Research Fellow, Nuffield Department of Surgery, University of Oxford, Oxford, UK Heung Bae Kim, MD Director, Pediatric Transplant Center, Department of Walter K. Graham, JD Surgery, Children’s Hospital Boston, and Associate Executive Director, United Network for Professor of Surgery, Harvard Medical School, Organ Sharing, Richmond, VA, USA Boston, MA, USA Simon J.F. Harper Clinical Lecturer in Transplantation, University Andrew A. Klein, MD of Cambridge and Addenbrooke’s Hospital, Consultant, Cardiothoracic Anaesthesia and Cambridge, UK Intensive Care, Papworth Hospital, Cambridge, UK

Koji Hashimoto, MD, PhD John Klinck, MD Department of Surgery, Cleveland Clinic, Cleveland, Consultant Anesthetist, Addenbrooke’s Hospital, OH, USA Cambridge, UK viii Contributors

Camille Nelson Kotton, MD Mandeep R. Mehra, MBBS, FACP, FACC Clinical Director, Transplant and Immuno- Division of Cardiology, Department of Medicine, compromised Host Infectious Diseases, University of Maryland School of Medicine, Infectious Diseases Division, Massachusetts General Baltimore, MD, USA Hospital, Boston, MA, USA Stephen J. Middleton, MA, MD, FRCP, FAHE Vineeta Kumar, MD Consultant Physician, Department of Assistant Professor of Medicine and Director, Gastroenterology, Addenbrooke’s Hospital, Transplant Nephrology Fellowship Program, Division Cambridge University, Cambridge, UK of Nephrology, University of Alabama at Birmingham, Birmingham, AL, USA Giorgina Mieli-Vergani, MD, PhD, FRCP, FRCPCH Alex Mowat Professor of Paediatric Hepatology, Yael B. Kushner, MD Institute of Liver Studies, King’s College London Department of Pathology, Massachusetts General School of Medicine, London, UK Hospital, and Harvard Medical School, Boston, MA, USA Charles Miller D. Frank. P. Larkin, MD, FRCPI, FRCOphth2 Director of Liver Transplantation, Cleveland Clinic, Consultant Surgeon, Moorfields Eye Hospital, Cleveland, OH, USA London, UK Sharon Mulroy, MD Clive J. Lewis, MB, BChir, MRCP, PhD Locum Consultant Nephrologist, Department of Consultant, Department of Cardiology, Transplant Renal Medicine, Addenbrooke’s Hospital, Cambridge, Unit, Papworth Hospital, UK Cambridge, UK Faruk Ozalp,¨ MRCS Yvonne H. Luo, MA, MRCOphth Department of Cardiopulmonary Transplantation, Specialist Registrar, Moorfields Eye Hospital, TheNewcastleuponTyneHospitalsNHSFoundation London, UK Trust, Freeman Hospital, High Heaton, Newcastle upon Tyne, UK Richard S. Luskin President and CEO, New England Organ Bank, Can Ozturk, MD Waltham, MA, USA Dermatology and Plastic Surgery Institute, Cleveland Clinic, Cleveland, OH, USA Ernest I. Mandel, MD Clinical Fellow in Medicine, Brigham and Women’s Jayan Parameshwar, MD, MPhil, FRCP Hospital/Massachusetts General Hospital, Boston, Consultant Cardiologist, Advanced Heart Failure and MA, USA Transplant Programme, Papworth Hospital, James F. Markmann, MD, PhD Cambridge, UK Chief, Division of Transplantation, Department of J.S. Parmar, BM, PhD, FRCP Surgery, and Clinical Director, Transplant Center, Consultant Transplant Physician (Respiratory), Massachusetts General Hospital, and Professor of Transplant Unit, Papworth Hospital, Papworth Surgery, Harvard Medical School, Boston, MA, USA Everard, Cambridge, UK Lorna Marson Hari K. Parthasarathy, MD, MRCP Senior Lecturer in Transplant Surgery, University of Transplant Unit, Papworth Hospital, Edinburgh, Edinburgh, UK Cambridge, UK

Arthur J. Matas, MD Nick Pritchard, BSc, MBBS, PhD, FRCP Department of Surgery, University of Minnesota, Consultant Physician, Department of Renal Minneapolis, MN, USA Medicine, Addenbrooke’s Hospital, Cambridge, UK

ix Contributors

Cristiano Quintini, MD of Medicine, Harvard Medical School, Boston, Department of General Surgery, Cleveland Clinic, MA, USA Cleveland, OH, USA Steven S.L. Tsui, MA (Cantab), MD, FRCS (Eng), Axel O. Rahmel FRCS (C-Th) Medical Director, Eurotransplant International Consultant Surgeon and Director of Transplantation, Foundation, Leiden, The Netherlands Papworth Hospital, Cambridge, UK

Chris J. Rudge, FRCS Khashayar Vakili, MD National Clinical Director for Transplantation, Instructor in Surgery, Harvard Medical School, and Department of Health, London, UK Department of Surgery, Children’s Hospital Boston, Stephan V.B. Schueler, MD, PhD, FRCS Boston, MA, USA Consultant Surgeon, Department of Cardiopulmonary Transplantation, The Newcastle R.V. Venkateswaran, MS, MD, FRCS-CTh upon Tyne Hospitals NHS Foundation Trust, Fellow, Department of Cardiothoracic Surgery, Freeman Hospital, High Heaton, Newcastle upon Papworth Hospital, Papworth Everard, Tyne, UK Cambridge, UK Maria Siemionow, MD, PHD, DSC Hector Vilca-Melendez, MD, PhD Professor of Surgery, Dermatology and Plastic Consultant Surgeon, Institute of Liver Studies, King’s Surgery Institute, Cleveland Clinic, College Hospital, London, UK Cleveland, OH, USA

Jacob Simmonds Vladimir Vinarsky, MD Specialist Registrar, Department of Cardiothoracic Clinical Fellow in Medicine, Massachusetts General Transplantation, Great Ormond Street Hospital, Hospital, Boston, MA, USA London, UK Peter Slinger, MD, FRCPC Kathryn J. Wood, DPhil Professor of Anesthesia, Toronto General Hospital, Professor of Immunology, Nuffield Department of Toronto, ON, Canada Surgery, University of Oxford, Oxford, UK Thomas R. Spitzer, MD Heidi Yeh, MD Director, Bone Marrow Transplant Program, Assistant in Surgery, Massachusetts General Massachusetts General Hospital, and Hospital, and Instructor in Surgery, Harvard Medical Professor of Medicine, Harvard Medical School, School, Boston, MA, USA Boston, MA, USA StuartC.Sweet,MD,PhD David W. Zaas Medical Director, Pediatric Lung Transplant Medical Director, Lung Transplantation, and Program, and Associate Professor of Pediatrics, Vice-Chair for Clinical Practice, Department Department of Pediatrics, Washington University, of Medicine, Duke University, Durham, NC, St.Louis,MO,USA USA Nina E. Tolkoff-Rubin, MD Jonathan G. Zaroff, MD Medical Director for Renal Transplantation, Division of Cardiology, Kaiser Permanente, Massachusetts General Hospital, and Professor San Francisco, CA, USA

x Foreword

The ever expanding nature of transplantation means This text provides state-of-the-art knowledge from that a book aimed at encompassing all aspects of all experts in their respective fields. As such, it will transplant subspecialties would be vast. Instead, this become an essential companion for anyone involved book focuses on the clinical aspects of transplanta- in transplantation, especially those at the beginning of tion. It provides a concise yet comprehensive guide to their careers. It will be available as an e-book, and in the art and science of caring for transplant patients. the traditional print form. I am sure that you will enjoy, It will undoubtedly provide an excellent resource for Organ Transplantation – A Clinical Guide. physicians, surgeons, anesthesiologists and, indeed, all transplant practitioners – medical and non-medical. It Thomas E. Starzl, MD, PhD will also be of interest to patients and their families Professor of Surgery and Distinguished Service because it is written and presented in an easy-to-read Professor, University of Pittsburgh format.

xi

Preface

The field of solid organ transplantation has developed pool, and those who are potential donors are becom- enormously in the last three decades and what was pio- ing older and heavier so that the number and quality of neering surgery has now become routine. Outcomes retrieved organs is falling. The reduction in the tradi- are no longer considered in terms of 1-year survival, tional donor pool has encouraged clinicians to look at but clinicians and patients are looking to 20 years additional sources of donors, including living donors and beyond. The current success of transplantation is and donors after circulatory death. These approaches based on many different factors: developments in sur- will go some way toward mitigating the impact of the gical technique, better immunosuppression, improved shrinking traditional donor pool; however, the widen- anesthetic and intensive care, improved microbiology, ing gap between need and supply does bring into focus and close collaboration between – all those involved in the moral, ethical, and legal implications of the intro- the transplant pathway have contributed. duction of policies for what is, effectively, the rationing However, there are still many problems to be over- of life-saving organs. come and success has brought its own challenges. Transplantation remains a high-risk procedure The adverse impact of immunosuppression – such and its risks have to be balanced against those of as increased risk of some cancers and infections, ongoing medical management. Donated organs are increased cardiovascular and cerebrovascular disease, not free of risks of transmission of cancer or infec- diabetes and renal failure – have not yet been avoided tion and should be considered “second hand” rather by the development of more effective and specific than new. Recipient’s expectations must be managed agents; tolerance remains elusive, although inducing appropriately. An excessive focus on outcomes and operational tolerance is perhaps less distant now than avoidance of risk will encourage risk-averse behav- it was a decade ago. In many situations, recurrent ior by clinicians and may inhibit some surgeons disease is yet to be overcome. Most transplant recip- from remaining in this challenging field. Therefore, ients still have a reduced life expectancy compared unless regulation is maintained at an appropriate level, with the normal population and so clinicians are now over-monitoring will ultimately adversely affect the focussing on maintaining the quality and length of recipient. life. The future of transplantation is, for the moment, Overcoming many of the technical barriers to secure and there is little doubt that the need for transplantation has increased the number of people transplantation will continue to exceed the supply of who could benefit from transplantation and high- organs.Althoughmanyproblemshavebeenover- lighted the need for more donors. Donation rates vary come, many challenges remain. We are encouraged between countries and many factors contribute to this by the progress in immune tolerance, regenerative variation: cultural, logistical, financial, legal, and med- medicine, organ support, and even xenotransplanta- ical. The success of initiatives to reduce premature tion. However, there is much yet to be learned and death from road accidents and cardiovascular and then applied to patients. The race between perfect- cerebrovascular disease are of course hugely welcome ing the process of organ transplantation-fabrication buthaveresultedinareductioninthepotentialdonor on one hand and the curing of diseases that lead

xiii Preface

to organ failure and the need for transplantation plants and better prepared to embrace the exciting on the other, is on. Fortunately, however, no matter future of our field. which side wins, it the patient who is ultimately the Andrew A. Klein, MD victor. Consultant, Cardiothoracic Anaesthesia and This book, with contributions from experts in the Intensive Care, Papworth Hospital, Cambridge, UK broad field of transplantation from all over the world, provides an authoritative account of where transplan- James Neuberger, DM, FRCP tation has come from, where it is now, and where it Honorary Professor of Medicine, University of might go in the future. The state-of-the-art knowledge Birmingham, and Associate Medical Director, contained within this volume will help make all who Organ Donation and Transplantation, NHS Blood read it better caregivers to recipients of organ trans- and Transplant, Bristol, UK

xiv Abbreviations

6-MP 6-mercaptopurine BKV BK virus A2ALL Adult-to-Adult Living Donor Liver BLT bilateral lung transplantation Transplantation Cohort Study BMD bone mineral density ABOi ABO incompatible BMI body mass index ABVC doxorubicin, bleomycin, vinblastine, BNP B-type natriuretic peptide and dacarbazine BODE body mass index, airflow obstruction, ACAID Anterior chamber–associated dyspnea, and exercise capacity immune deviation BOS bronchiolitis obliterans syndrome ACD acquired cystic disease BTT bridging to transplant ACE angiotensin-converting enzyme C3 complement component 3 ACOT Advisory Committee on C4d complement component 4d Transplantation CAN chronic allograft nephropathy ACR acute cellular rejection CAV cardiac allograft vasculopathy ACT activated clotting time CAV coronary artery vasculopathy ADCC antibody-dependent cell-mediated CBD common bile duct cytotoxicity CBP cardiopulmonary bypass ADH anti-diuretic hormone CD3 cluster of differentiation (CD) 3 AHR acute humoral rejection molecule aHUS atypical HUS CDC complement-dependent cytotoxicity aICB atraumatic intracranial bleed cross-match AIDS acquired immune deficiency CDR chronic ductopenic rejection syndrome CF cystic fibrosis AIH autoimmune hepatitis CHOP cyclophosphamide, doxorubicin, ALF acute liver failure vincristine, and prednisone ALG anti-lymphocyte globulin CHR chronic humoral rejection AMR antibody mediated rejection CIT cold ischemic time AP-1 activator protein 1 CKD chronic kidney disease APC antigen-presenting cell Cmax maximum concentration APOLT auxiliary partial orthotopic liver CMV cytomegalovirus transplantation CNI calcineurin inhibitor ARB angiotensin receptor blocker CNS central nervous system ARDS acute respiratory distress syndrome CO carbon monoxide ATC anti-thymocyte globulin COPD chronic obstructive pulmonary ATG antithymocyteglobulin disease ATN acutetubularnecrosis CPB cardiopulmonary bypass ATP adenosine triphosphate production CPEX cardiopulmonary exercise testing AV atrioventricular CR complete remission AZA azathioprine CRP C-reactive protein BASM biliary atresia splenic malformation CRT cardiac resynchronization BCC basal cell carcinoma therapy

xv List of abbreviations

c-SMAC central supramolecular activation FEF25–75% forced expiratory flow 25–75% cluster FEV1 forced expiratory volume in 1 second CT computed tomography FLR future liver remnant CTA composite tissue allotransplantation FRC functional residual capacity CTL cytotoxic T cell FSGS focal segmental glomerulosclerosis CTLA-4 cytotoxic T-lymphocyte antigen 4 FVC forced vital capacity CTTR Cincinnati Transplant Tumor GBM glomerular basement membrane Registry G-CSF granulocyte colony-stimulating factor CVP central venous pressure GDA gastroduodenal artery CY cyclophosphamide GERD gastroesophageal reflux disease CyA cyclosporine GFR glomerular filtration rate D donor GI gastrointestinal D– no diarrhea GRWR graft-to-recipient body weight ratio D+ diarrhea associated GvHD graft-versus-host disease DAG diacylglycerol GvL Graft versus leukemia DBD donation after brain death HAR hyperacute rejection DC dendritic cell HARI hepatic artery resistance index DCD donation after cardiac death HAT hepaticarterythrombosis DEXA dual-energy X-ray absorptiometry HBV hepatitis B virus DGF delayed graft function HCC hepatocellular carcinoma DLCO diffusing capacity of carbon HCV hepatitis C virus monoxide HF heart failure DLI donor lymphocyte infusion HFSS Heart Failure Survival Score DM1 type 1 diabetes mellitus HHS Department of Health and Human DM2 type 2 diabetes mellitus Services DSA donor specific antibodies HHV human herpes virus DTC donor transplant coordinator HHV-8 human herpes virus type 8 DTH delayed-type hypersensitivity HIV human immunodeficiency virus DVT deepvenousthrombosis HLA human leukocyte antigen EBV Epstein-Barr virus HLAi HLA incompatible ECD extended criteria donor HLT combined heart–lung transplantation ECMO extracorporeal membrane HLT heart–lung transplantation oxygenation HMGB-1 high-mobility group box 1 protein EC-MPS enteric-coated mycophenolate HPC human progenitor cell sodium HPV human papillomavirus EF ejection fraction HRQOL health related quality of life eGFR estimated glomerular filtration rate HRSA Health Resources and Services EHBA extra-hepatic biliary atresia Administration ELITE Efficacy Limiting Toxicity Elimination HSC hematopoietic stem cells ENT ear, nose, and throat HSCT hematopoietic stem cell ER endoplasmic reticulum transplantation ESC embryonic stem cell HSP heat shock protein ESRD end-stage renal disease HSV herpes simplex virus ETT endotracheal tube HT heart transplant EVL everolimus HT heart transplantation EVLP ex vivo lung perfusion HTA Human Tissue Authority FasL Fas ligand HUS hemolytic uremic syndrome FCXM flow cytometry cross-match IBMIR immediate blood-mediated FDA US Food and Drug Administration inflammatory reaction FEF forced expiratory flow ICAM-1 intercellular adhesion molecule 1 xvi List of abbreviations

ICD implantable cardioverter-defibrillator LDLT living donor lung transplantation ICOS inducible T-cell costimulator LFA-1 leukocyte function-associated ICP intra-cranial pressure antigen 1 ICU intensive care unit LFA-1 lymphocyte-function associated IE islet equation antigen-1 IF/TA interstitial fibrosis and tubular LT liver transplantation atrophy LT lung transplantation IFN interferon LV left ventricle IFN␥ interferon ␥ LV left ventricular Ig immunoglobulin LVAD left ventricular assist device IgA immunoglobulin A LVEDP left ventricular end-diastolic IgG immunoglobulin G pressure IHD ischemic heart disease LVH left ventricular hypertrophy IL interleukin MAC membrane attack complex IL-2 interleukin-2 MAG3 mercaptoacetyltriglycine IL-2R interleukin-2 receptor MAP mitogen-activated protein IMP inosine monophosphate MAPK MAP kinase IMV inferior mesenteric vein MBD mineral bone disease iNO inhaled nitric oxide MCSD mechanical circulatory support INR international normalized ratio devices INTERMACS International Registry for MELD Model for End-Stage Liver Disease Mechanically Assisted Circulatory MHC major histocompatability complex Support MHV middle hepatic vein IP3 inositol triphosphate miH minor histocompatibility IPAH idiopathic pulmonary arterial MMF mycophenolate mofetil hypertension MMR measles, mumps, and rubella IPEX immune dysregulation, MPA mycophenolic acid polyendocrinopathy, enteropathy MPGN membranoproliferative X-linked glomerulonephritis IPF idiopathic pulmonary fibrosis MPSC Membership and Professional IPTR International Pancreas Transplant Standards Committee Registry MRI magnetic resonance imaging IRI ischemia/reperfusion injury MRR medical record review ISHLT International Society for Heart and MRSA methicillin-resistant Staphylococcus Lung Transplantation aureus iTreg inducible Treg mTOR mammalian target of rapamycin ITU intensive treatment unit NAPRTCS North American Pediatric Renal IV intravenous Trials and Collaborative Studies IVC inferior vena cava NFAT nuclear factor of activated IVIg intravenous immunoglobulin Tcells IVUS intravascular ultrasound NF-␬B nuclear factor-␬B Jak Janus kinase NHL non-Hodgkin’s lymphoma JCV JC virus NHS National Health Service KS Kaposi’s sarcoma NHSBT NHS Blood and Transplant LAS lung allocation system NK natural killer LCR late cellular rejection NKT natural killer T LDH lactate dehydrogenase NO nitric oxide LDLLT living donor lobar lung NOD non-obese diabetic transplantation NODAT new-onset diabetes after LDLT living donor liver transplantation transplantation

xvii List of abbreviations

NODAT new-onset diabetes after RAS renal artery stenosis transplantation rATG rabbit ATG NOTA National Organ Transplant Act RBCs red blood cells NRM non-relapse mortality RCC renal cell carcinoma nTreg naturally occurring Treg RCT randomized controlled trial NYHA New York Heart Association RER respiratory exchange ratio OB obliterative bronchiolitis RIC reduced intensity conditioning OKT3 muromonab-CD3 ROS reactive oxygen species OLV one-lung ventilation RR relative risk OP osteoporosis RSV respiratory syncytial virus OPO organ procurement organization RV right ventricle OPTN Organ Procurement and RV right ventricular Transplantation Network RVAD right ventricular assist device PA pulmonary artery RVI respiratory viral infections PAK pancreas after kidney transplant SA sinoatrial PAP pulmonary artery pressure SARS severe acute respiratory syndrome PAS Periodic Acid-Schiff SCC squamouscellcarcinoma PCA patient-controlled analgesia SCID severe combined immunodeficiency PCR polymerase chain reaction SFSS small-for-size syndrome PCWP pulmonary capillary wedge pressure SIOPEL International Childhood Liver PDA UK Potential Donor Audit Tumour Strategy Group of the PDGF platelet-derived growth factor International Society of Paediatric PEEP positive end expiratory pressure Oncology PGD primary graft dysfunction SIRS systemic inflammatory response PGE2 prostaglandin E2 syndrome PGI2 prostacyclin SLT single lung transplantation PK pharmacokinetic SMA smooth muscle antibody PKC protein kinase C SMA superior mesenteric artery PML progressive multifocal SMV superior mesenteric vein leukoencephalopathy SOT solid organ transplant PMN polymorphonuclear neutrophil SPK simultaneous pancreas and kidney PN parenteral nutrition transplant ppm parts per million SRL sirolimus PRA panel-reactive antibody SRTR Scientific Registry of Transplant PRAISE Prospective Randomized Amlodipine Recipients Survival Evaluation SRTR US Scientific Registry of Transplant PRR pattern recognition receptor Recipients PSC primary sclerosing cholangitis SVC superior vena cava p-SMAC peripheral SMAC SVR systemicvascularresistance PT prothrombin time T3 triiodothyronine PTA pancreas transplant alone T4 tetraiodothyronine PTLD post-transplant lymphoproliferative TA transplant arteriopathy disease TAC tacrolimus PTM post-transplant malignancy TAH total artificial heart PUVA ps oralen and UVA therapy TBBx transbronchial biopsy PVR pulmonary vascular resistance TBI total body irradiation PVT portal vein thrombosis TCD T-cell depleted QoL quality of life TCMR T-cell mediated rejection Rrecipient TCR T-cell receptor RAD everolimus TEA thoracic epidural anesthesia xviii List of abbreviations

TEE transesophageal echocardiography TR tricuspid regurgitation TEG thromboelastography Treg regulatory T cell TG transplant glomerulopathy TSH thyroid-stimulating hormone TGF␤ transforming growth factor ␤ UCLA University of California, Los Angeles ThhelperTcell UKELD United Kingdom Model for TIVA total intravenous anesthesia End-Stage Liver Disease TLC total lung capacity UNOS United Network for Organ Sharing TLCO transfer coefficient for carbon UTI urinary tract infection monoxide VA arteriovenous TLI total lymphoid irradiation VAD ventricular assist device TLR Toll-like receptor VBDS vanishing bile duct syndrome TMA thrombotic microangiopathy VOD veno-occlusive disease Tn troponin VRE vancomycin-resistant Enterococcus TNF tumor necrosis factor faecalis TNF-␣ tumor necrosis factor alpha VTE venous thromboembolism TNFR TNF receptor VV venovenous TPEX therapeutic plasma exchange VZV varicella zoster virus TPG transpulmonary gradient WHO World Health Organization TPG transpulmonary pressure gradient WU Wood units

xix

Section 1 General Chapter Historical perspectives

1 John Dunning and Sir Roy Calne

Key points ancient literature of China and India. The first descrip- tion of a skin transplant is contained in the Sushrutta r Successful techniques for vascular manuscripts dating from around 450 BC. The tech- anastomoses developed at the end of the nique described found use in Europe during the Mid- nineteenth century made the transplantation dle Ages in the hands of the Italian surgeon Gas- of internal organs possible. r pare Tagliacozzi. He used it for the reconstruction The first successful human allograft, a of damaged noses, frequently a result of syphilitic corneal transplant, was performed in 1905. r injury, using a skin flap from the forearm. At the time The recognition that the body’s reactions to hewrotethat“thesingularnatureoftheindividual foreigntissueledtothefailureofallograft entirelydissuadesusfromattemptingthisworkon transplantation gave rise to the new another person.” Perhaps he had already attempted discipline of immunology. r the repair using allogeneic donors (transplantation The discovery that cyclosporine, a metabolite between genetically disparate individuals) prior to his from the fungus Tolypocladium inflatum,is successful autograft (transplant of tissue in the same 300 times more active against the individual). Although the technique was new to the proliferation of splenic lymphocytes than people of the time, the concept of tissue transplanta- against other cell lines changed the face of tion was well established among Europeans following transplantation. r the legend of a total leg transplant by Saints Damon As transplantation has become more and Cosmos illustrated by artists such as Fra Angelico successful in terms of survival, quality of life, and sculpted by Donatello. Such legendary optimism and cost benefit, the demand for donor was not rewarded clinically until much later, but it is organs has increased so that it is now greater certain that interest in skin grafting was revived due than supply. to the substantial need for treatment of the gross leg ulcers prevalent in the nineteenth century as a result of Transplantation of organs represents the pinnacle of injury from syphilis, nutritional deficiency, and burns. medical achievement in so many different ways. It Greatadvancesweremadewiththeobservationsof epitomizes the multi-disciplinary team approach to theFrenchPhysiologistPaulBert,whorecognizedthe patient care. It has a foundation in refined surgical importance of graft neovascularization and described technique, supported by an understanding of com- the success of autografting in comparison with the fail- plex immunological events, and requires a complex ures of allografting. approach to pretransplant assessment and postoper- It was the ophthalmic surgeons who really led the ativecareofmultipleorgansystems.Yetinsome way to successful allografting with the transplanta- respects it also represents a failure: the inability to tion of corneal grafts. Samuel Bigger reported what repair diseased organs such that the only way forward was probably the first successful full-thickness corneal is to cast aside the worn out tissue! allograft when he performed an operation on a blind The idea of organ and tissue transplantation is pet gazelle while he was a prisoner in Egypt in 1835. not new, and reference to it may be found in the He replaced the cornea, apparently with good results.

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

1 Section 1: General

Attempts to reproduce this success continued through Mathieu Jaboulay attempted the technique in man, and the latter part of the nineteenth century, and with tech- in 1906 he reported his observations in Lyon Medical. nical improvements and increasing frequency of tri- His attempts used a pig kidney in one patient and a als, the results with animal corneal grafts improved goat’s kidney in a second, with the organs implanted steadily. Finally, in 1905, the first successful human in the cubital fossa and anastomosed to the humeral corneal allograft was performed. Although therapeu- artery and cephalic vein. Ultimately both attempts tic transplantation of the cornea became firmly estab- failed as a result of vascular thrombosis, but the kid- lished as part of ophthalmic practice from this time, neys did start to diurese initially. there was no theoretical explanation why corneal graft- It quickly became apparent that whereas autografts ing should be successful whereas the grafting of other generally succeeded, allografts and xenografts mostly organs and tissues was not, nor of the observation that failed. Although the technical problems of the oper- from time to time corneal grafts were rejected. ation had largely been sorted out, it was clear that It was not until Alexis Carrel and Mathieu Jaboulay “fromabiologicalstandpoint...theinteractionsofthe developed successful techniques for vascular anasto- host and of the new organ are practically unknown.” moses at the end of the nineteenth century that the The increasing understanding that the resistance to transplantation of internal organs became possible. foreign grafts was caused by systemic factors led to Many different animal models were used with attempts the repeated suggestion that an immune response of to transplant almost every organ, but the kidney was the “anaphylactoid type” was somehow responsible for the first organ to which this technique was repeatedly graft rejection. It was recognized that research had now applied. to be directed toward understanding the body’s reac- Carrel remained a prominent contributor to the tions to foreign tissue, and so from experimental trans- field of transplant surgery throughout the early 1900s, plantation in the early part of the century, the two moving from France to the United States, where his new disciplines of vascular surgery and immunology collaboration with Guthrie led to significant contri- emerged. butions to vascular surgery with the development of Other landmarks were reached throughout the techniques for venous patching of arteries and the early years of the twentieth century, with growing use of cold storage to protect tissue for reimplanta- understanding of skin grafts used to treat burns, and tion up to 20 hours from its procurement. The result with Voronoy transplanting the first cadaveric human of their labors was a series of 35 papers describing kidney in 1933. His recipient was a 26-year-old woman their experimental achievements in a wide variety of who had attempted suicide by swallowing sublimed animal models for transplantation. However, it was mercury. This led to uremic coma. The kidney was pro- not until 1908 that survival became extended when cured from a 60-year-old man who died following a Carrel performed a kidney transplant in a dog with fracture of the base of the skull. The operation was per- survival of the graft for several years. With the survival formed on April 3, 1933, with the renal vessels anas- of grafts beyond a few hours, the opportunity to study tomosed using Carrel’s technique to the femoral ves- tissue histologically emerged, and by 1905, parenchy- sels and the kidney placed in a subcutaneous pouch, mal infiltration by “round cells” and arterial lesions with externalization of the ureter. Local anesthetic was were recognized. used. The donor was known to be blood group B, and Of course human donors were not available at the recipient blood group O. The grafted kidney did this time, and all organs transplanted were obtained diurese for a while, but unfortunately the patient died from animals so that a mixture of pig, goat, mon- 2dayslater. key, and sheep xenografts (transplantation between Despite the demonstration of second-set skin graft species) were undertaken in human patients with acute rejection in man as early as 1924 and the successful renal failure. Although none of these attempts were exchange of skin between identical twins in 1927, no successful, the last attempt by Neuhof in 1923 was useful generalizations were made to further elucidate particularly encouraging, with the recipient surviv- the immunological mechanisms involved. The prac- ing for 9 days. It demonstrated clearly that throm- tice of corneal grafting continued, but it seemed to be bosis and hemorrhage from vascular anastomoses accepted that the transplantation of other tissues and was not inevitable. Although most attempts to per- organs was impractical, and there was a lull in activ- form organ transplantation were made in animals, ity among surgeons for the next 20 years, with further

2 Chapter 1: Historical perspectives interruptions to the field brought about by the Second Patients subjected to total-body irradiation frequently World War. succumbed to infection, aplasia, and cancer. The area of skin grafting became of greater impor- The introduction of chemotherapy to supplement tance for the treatment of war burns and other irradiation and allow dose reduction improved the injuries, and the death from kidney disease also outcomes further, and in 1960, William Goodwin provided impetus to focus once more on kidney introduced methotrexate and cyclophosphamide to transplantation. Short-term success in the late 1940s the field of living related transplantation and treated was reported by a number of individuals, including an episode of rejection with prednisolone. Then, in Voronoy, and David Hume working in Boston. Both London in the mid 1950s, the prolongation of survival transplanted kidneys into patients with uremic coma of renal allografts in dogs by the anti-leukemia drug that diuresed for a number of days, before stopping and 6-mercaptopurine (6-MP) heralded clinical immuno- being removed again. The technique was not seen as suppression and azathioprine (AZA), a derivative of replacement therapy but a method of stimulating a 6-MP, was found to be slightly better experimentally. recovery reflex in the native diseased kidney. How- Although 6-MP was used briefly with irradiation, it ever, as the immunological basis of rejection became was rapidly abandoned because of significant toxic- established, scientific interest in organ transplanta- ity. The use of AZA in clinical kidney transplanta- tion waned until effective immunosuppressive regi- tion was originally disappointing, but when corticos- mens were found. teroids were added, this immunosuppressive regimen resulted in some long-term clinical renal allograft suc- cesses from the early 1960s. Abdominal organ transplantation Further understanding of transplant immunology Transplantation of abdominal organs has been a long- was gained with insights into the human leukocyte term success story, with patients surviving 40 years antigen (HLA) system and histocompatibility. Cross- with excellent function in their original grafted organs. match techniques became established through the The success of clinical allograft transplantation began 1960s, and understanding of the “transfusion effect” with transplantation of kidneys between identical was also gained (Opelz and Terasaki), whereby previ- twins by Murray and colleagues at Peter Bent Brigham ous transfusion appeared to confer protection for the Hospital in Boston in 1956. This was an outstanding transplanted organ. achievement and demonstrated clearly that the kid- In the 1960s, experimental transplantation of the ney would withstand the trauma of removal, periods liver, pancreas, intestines, and heart led to a clari- of ischemia, and then the procedure of transplanta- fication of the technical requirements involved, and tion into another individual of the same species. The in 1963, Starzl in Denver carried out the first clini- fact that identical twins would not be able to reject cal liver transplant. Unfortunately, the results of this skin grafts and the experimental auto-transplantation clinical series were dismal, and Starzl self-imposed a of the kidney in the dog enabled the group in Boston moratorium until 1967, when he resumed clinical liver to proceed with the clinical operation with reasonable transplantation, having in the meantime improved optimism. Unfortunately, a twin donor would not be the surgical technique and the assessment of graft available for most patients dying of kidney failure, and function and prevention of rejection. The first ortho- the immunological barrier between individuals proved topic liver transplant in Europe was performed in to be an enormous biological problem. Cambridge by Calne in 1968. For nearly 10 years, For more than a decade, clinical kidney transplan- Denver and Cambridge were the only two centers tation was the only form of organ grafting that was with regular programs of clinical liver transplantation. seriously studied and yielded some success. The identi- There were a few outstandingly good results, but many cal twin experience was reproduced, and conditioning disappointments. Patients were referred for operation of the recipient with total-body irradiation was applied too late, and anti-rejection therapy was still in the to kidney grafting between donor and recipient who processofdevelopmentusingmodifiedregimensof were not twins. This was based mainly on experimen- AZA, steroids, and polyclonal anti-lymphocyte serum. tal work with bone marrow transplantation; however, In addition to rejection, sepsis, biliary, and vascular in the clinic the results were disastrous, except in two complications and recurrence of the patient’s own dis- cases of kidney grafting between non-identical twins. ease often resulted in failure. During this uncertain

3 Section 1: General

and disappointing phase of development, the vascular- require renewal of their immunosuppressive regimen ized pancreas was also transplanted and shown to be of drugs. Confidence in the surgery and immunosup- capable of curing diabetes in a few patients. However, pression gradually increased. many patients suffered from complications of leakage A variety of complicated organ graft procedures of pancreatic enzymes, causing inflammation and fatal were reported, including small bowel on its own (1988) sepsis. and in combination with liver and other organ grafts. A watershed in organ transplantation was the The first combined heart, lung, and liver transplant discovery of the immunosuppressive properties of was performed by Wallwork and Calne in 1987 at Pap- cyclosporine (CyA), a metabolite from the fungus worth (Cambridge, United Kingdom), with survival of Tolypocladium inflatum,byJeanBorelworkingin thepatientformorethan10years. the Sandoz laboratories. CyA was 300 times more There is now a move toward minimization of active against the proliferation of splenic lymphocytes immunosuppression and tolerance. Alemtuzumab than against other cell lines. Experimental and clin- (Campath), an extremely powerful anti-lymphocyte ical application of CyA transformed the attitude of antibody developed in Cambridge by Waldmann and previously sceptical clinicians to organ transplanta- colleagues, has induced “prope or almost tolerance” tion. Calne’s paper published in The Lancet in 1979 when used as an induction agent followed by main- described its use in 32 kidney transplants, 2 liver tenance immunosuppression with half-dose CyA, transplants, and 2 pancreatic transplants and showed rather than a full dose of three drugs. Of the original improved 1-year functional survival of kidney trans- series of kidney transplantation patients treated in plants from below 50% to approximately 80%. It was Cambridge, 80% have never received steroids, and introduced to clinical immunosuppressive regimens their quality of life has been excellent after more worldwide in 1982 and radically improved the sur- than 10 years of follow-up. This immunosuppressive vival of heart, kidney, liver, and pancreas recipients. regimen has reduced complications of anastomotic About10centershadsoldieredoninthepre-CyAera, leakageinpancreastransplants,withencouraging but after the introduction of CyA, there were soon results. more than 1000 centers. The improved results led to an Pancreas grafting can never be a treatment for expanding mismatch of numbers of available donors all diabetics, but when transplanted together with a to potential recipients seeking a life-saving organ kidney in patients with diabetic renal failure, pan- graft. creas transplantation has produced excellent long- Unfortunately, the nephrotoxic side effects of CyA term results. A move toward islet transplantation to led to late renal failure in many cases. Hopes that there avoid the major operation has had some early encour- might be a dosage window in which rejection could be aging results. This is a field in which stem cell and/or controlled and side effects avoided were only realized gene therapy may well lead to fruitful developments in in a minority of cases. However, the concept was estab- the future. lished of combining immunosuppressive drugs with the objective of obtaining added immunosuppressive effect but reducing the individual side effects. Thus Cardiothoracic transplantation AZA, CyA, and steroids became a standard immuno- While the field of kidney transplantation research and suppressant regimen. experimentation moved rapidly into the clinical arena, The liver proved to be less susceptible to rejec- progress was not so rapid for the transplantation of tion than other organs. This had been anticipated by other organs. The first heart transplant described in experimentsinpigsandrats.Inanimportant“patient- the literature was performed in 1905 by Carrel and led clinical study,” a group of patients from Denver Guthrie. The heart, transplanted from one dog into stopped taking their maintenance immunosuppres- a heterotopic position in the neck of another dog, sion without telling their doctors. Although lack of continued to beat for 2 hours. This model demon- compliance is a common cause of organ graft failure strated that it was possible to transplant a heart with duetorejection,asurprisingnumberofyoungpatients all four chambers pumping blood. More importantly, with liver transplants did well long-term. A number it demonstrated that the heart could be removed from of patients, in whom immunosuppression was stopped its blood supply and sutured into the circulation of a for medical indications, usually infection, also did not second animal and still recover its normal organized

4 Chapter 1: Historical perspectives contractile pattern. This brought into focus the con- ered from anesthetic, stood up, and drank, but died cept of “preservation” of the heart during its ischemic 15 hours later, an event attributed to superior vena period. caval thrombosis. Other workers were pursuing the Further reference to transplantation of the mam- same goal but were less successful. malian heart was made in 1933 by Mann and col- By the early 1950s, it was well established that car- leagues at the Mayo clinic, who were seeking a den- diac transplantation was technically feasible, and stud- ervated heart model. They made contributions to the ies were undertaken to clarify the physiology of car- area of preservation, advising that ventricular disten- diac transplantation. However, the move to orthotopic sion and coronary air embolism should be avoided; transplantation had not been achieved, and this was made observations on the general behavior of the largely due to the difficulties associated with the trans- transplanted heart; and made the first observations on fer phase, when the recipient’s own heart had been the phenomenon of cardiac allograft rejection, noting removed, and the problems associated with protec- that “histologically the heart was completely infiltrated tion of the donor heart during transfer. These prob- with lymphocytes, large mononuclears and polymor- lems were addressed in a report published in 1953 in phonuclears.” They concluded that “the failure of the which the operative technique was simplified by trans- heart is not due to the technique of transplantation but planting a heart–lung block, thus reducing the num- to some biologic factor. ...” ber of anastomotic connections, and the problems of Interest in cardiac transplantation waned until recipient preservation and myocardial protection were 1951, when workers at the Chicago Medical School solved as both animals were “placed in an ordinary reported their experience with a slightly modified beverage cooler for the production of hypothermia.” Mann preparation. They were interested in the possi- Using these techniques and arresting the recipient cir- bility of transplanting organs as a treatment modality culation for up to 30 minutes, the authors reported for end-stage disease, but their experiments, although successful transplantation in three dogs, with survival elaborate, were disappointing, with a maximum sur- of up to 6 hours. vival of only 48 hours. It was apparent to them that The recognition of the value of hypothermia asa “the greatest deterrent to long survival of the heart protective medium was important, but a further step is the biologic problem of tissue specificity” and con- wasmadetowardthepossibilityofclinicaltransplanta- cluded that “a transplanted heart...must be consid- tion with the development of the heart–lung machine, ered,atpresent,afantasticdream,anddoesnotfall pioneeredbyGibbonandattributedlargelytothetech- within the scope of the present considerations.” The nical expertise of the famous pilot Charles Lindbergh. Mann preparation continued to be used by various This allowed the circulating blood to bypass com- investigators to evaluate the transplanted heart, and pletely the patient’s own heart and lungs, allowing an Downie, working at the Ontario Veterinary College, extended operative period. reported excellent results, which he attributed to the The result of these innovations was that in 1958, use of penicillin and appropriate commercial suture thefirstorthotopichearttransplantswereperformed, material. Demikhov published results in 1962 in which and further steps were taken toward clinical transplan- an intrathoracic heterotopic heart continued to beat tation with the development of a simplified operative for 32 days. The long survival of this graft strengthened technique (Lower and Shumway), which removed the his belief that failure of transplanted organs was not necessity of individual venous anastomoses. The recip- due to immunological factors, but to simple technical ient left atrium was circumscribed, leaving a cuff of problems. tissuetosewtothedonorleftatrium,arelativelysim- The successful intrathoracic transplantation of the ple anastomosis compared with the complex multi- heart without interrupting the circulation led to the ple anastomoses of four pulmonary veins. The cavae idea that a cardiac allograft might be able to assume were reconnected with synthetic tubes, and the arter- some of the normal circulatory load. Demikhov led ies were simply sutured end to end. Recipient circula- the way, performing 22 such auxiliary heart trans- tion was maintained with the cardiopulmonary bypass plants between 1951 and 1955. The donor heart was machine, but hypothermia was not required for either implanted, and when fully resuscitated, the great ves- donor or recipient. Donor organs were ischemic for sels of the native heart were ligated so that the donor between 25 and 32 minutes, and the longest support heartassumedthefullload.Onesuchanimalrecov- of circulation by the allografts was 20 minutes.

5 Section 1: General

Further experiments in the late 1950s established In 1967, the first human-to-human orthotopic that orthotopic transplantation was technically possi- heart transplant was performed by Christian Barnard ble, and advances in the surgical techniques used were in South Africa. The patient was a 54-year-old man described. An important paper published in 1960 inte- suffering from ischemic cardiomyopathy who received grated the developments of the previous decade into the heart of a 16-year-old female donor. He recov- a single method for orthotopic transplantation, and ered from the operation but on the 18th day suc- five of eight consecutive canine transplant recipients cumbed to pseudomonas pneumonia. On the day that survived for between 6 and 21 days, eating and exer- he died, Barnard performed a second transplant, and cising normally in the postoperative phase. This was this recipient survived 594 days. the first description of a truly successful procedure in Following the initial efforts of Barnard in Cape which the circulation was maintained by the trans- Town and Kantrowitz in New York, 102 cardiac trans- planted organ. plants had been performed in 17 countries by the However, technical ability to perform the trans- end of 1968. The early results were discouraging, and plant operation is clearly not all that is required. In by 1970, there were only a few centers persevering. Lower’sseries,noneofthedogsreceivedimmuno- Gradually the problems were dealt with, and by 1978 suppression, and they all died as a result of rapid the 1-year survival rate had risen from 22% to 68%, myocardial failure due to the massive infiltration with with a return to normal function in 90% of these round cells and interstitial hemorrhage. Lower and patients. This was a time of real growth for clinical Shumway concluded that “if the immunologic mech- heart transplantation, with many reports of the early anismsofthehostwerepreventedfromdestroy- results, infectious complications, and the hemody- ing the graft, in all likelihood it would continue to namics of the transplanted heart. The indications and function adequately for the normal lifespan of the contraindications became clearly defined, and donor animal.” management was described. A further significant step was taken in 1965 when A further great advance was made by Philip Caves, Lower reported the use of the surface electrocar- who devised the bioptome for obtaining repeated diograph as a marker of rejection episodes. A volt- transvenous endomyocardial biopsies to detect car- age drop was seen during rejection episodes, which diac allograft rejection, and by Margaret Billingham, was reversible with the administration of methylpred- who described a histological system for grading the nisolone and azathioprine. With this test as a guide rejection reaction seen in these specimens. Further to the intermittent administration of immunosuppres- improvements were to be seen with the introduction of sive therapy, survival of 250 days was achieved in an rabbit antithymocyte globulin for the prevention and adult dog. treatment of acute rejection. As the concept of brain- Thus there had been a step-wise progression over stem death became accepted and methods of long- the years providing the solution to many of the most distance procurement were developed, together with difficult problems faced in transplanting the heart, and donor organ-sharing networks, donor organs became in 1964, Shumway wrote that “only the immunologi- more readily available, ensuring the continued practice cal barrier lies between this day and a radical new era of clinical transplantation. in the treatment of cardiac diseases.” Others clearly felt that the time was already right to undertake car- Combined heart and lung diac transplantation in man, and a planned approach was made toward this goal at the University Hospital transplantation in Jackson, Mississippi, in 1964. Legal and logistic rea- Demikhov developed a method of heart–lung trans- sonsmeantthatthefirstmantoreceiveahearttrans- plantation in dogs in the 1940s, but it was not revis- plant was to receive the heart of a large chimpanzee, ited until 1953, when Marcus and colleagues at the and not that of another man. The suture technique of Chicago Medical School described a technique for het- Lower and Shumway was used, and although the oper- erotopic heart–lung grafting to the abdominal aorta ation was technically successful, the heart was unable and inferior vena cava in dogs. Disappointingly, how- to maintain the circulatory load, and about 1 hour after ever, failure to resume normal spontaneous respira- cardiopulmonary bypass, attempts at further support tion was noted by a number of groups. Later primates were abandoned. were found to develop a normal respiratory pattern

6 Chapter 1: Historical perspectives following complete denervation with cardiopulmo- and cost benefit, the demand for donor organs has nary replacement. A Stanford series showed survival increased so that it is greater than supply. For example, forwellover5yearsafterheart–lungallografttrans- there were 454 thoracic organ transplants performed plants in primates, allowing Reitz and colleagues in the United Kingdom in the year ending Decem- to perform the first successful human heart–lung ber 1992, but at the end of the same year, the num- transplant in a 45-year-old woman with end-stage ber of patients on the waiting lists for cardiac and primary pulmonary hypertension in 1981. They uti- pulmonary transplantation had grown to 763. Thus lized a technique that preserved the donor sinoatrial even if no more patients were accepted onto the lists, node and eliminated the potential for caval anasto- it would take nearly 2 years to clear the back-log of motic stenosis. Subsequently, “domino” transplant was potentialrecipients.Theflawinthisargumentisthatof developed, in which the healthy heart of a heart–lung these potential recipients, approximately 25–30% will recipient is itself donated for grafting in a cardiac die on the waiting list before suitable organs become transplant recipient. available. It is worth noting that the patients who are accepted for transplantation are the tip of the iceberg; Lung transplantation many are not referred, and for every patient who is accepted, there are two or three who are rejected, but Experimental lung transplantation developed in paral- who might have benefited from transplantation if there lelwithheart–lungtransplantation.Metrasdescribed were a limitless donor pool. important technical concepts, including preservation The annual need for kidneys in the United King- of the left atrial cuff for the pulmonary venous anas- dom is estimated at between 2500 and 4000, whereas tomoses and reimplantation of an aortic patch con- a recent audit of intensive care units in England sug- taining the origin of the bronchial arteries to pre- gested an absolute maximum number of 1700 potential vent bronchial dehiscence in 1949. The technique was donors. Even if all these patients were consented for technically difficult and did not gather widespread donation and were medically suitable, there would still acceptance. Transection of the transplant bronchus be a deficit in supply compared with the demand. The close to the lung parenchyma was advocated in the demandcanbeexpectedtocontinuetorise,whereas 1960s by Blumenstock to prevent ischemic bronchial thenumberofpotentialdonorsmaybeexpectedto necrosis. Further surgical modifications to prevent fall as factors such as seat-belt legislation and better bronchial anastomotic complications included tele- trauma care reduce the pool of patients declared brain- scoping of the bronchial anastomosis, described by stem dead. Veith in 1969, and coverage of the anastomosis with The indications for transplantation are widening, an omental flap, described by the Toronto group and although kidney, liver, heart, and even lung trans- in 1982. The first human lung transplant was per- plantation is now seen as routine, the necessary skills formed in 1963 by Hardy and colleagues at the Uni- are being developed to transplant other organs, such versity of Mississippi; however, the patient only sur- as the small intestine, pancreas, face, hand, and uterus. vived for 18 days. It was only in 1986 that the Clearly this stretches the donor pool beyond its limit. first series of successful single lung transplants with Other solutions to the donor shortage must be long-term survival were reported from Toronto (with sought if transplantation is to be extended to treat the first patient undergoing transplantation in 1983). all those in need. Recent trends have seen increased En-bloc double lung transplantation was performed use of living related donors for kidney transplantation, by Patterson in 1988 but was later superseded by and although renal transplant surgeons have used this sequential bilateral lung transplantation, described resource for a long time, the potential to use livers by Pasque and colleagues in 1990. Subsequently, (first performed in 1989) and lungs from live related Yacoub introduced live lung lobar transplantation in donors has only recently been explored. The potential 1995. hazards for the donor of such procedures have stimu- lated fierce ethical debate. Living related donation will Indications and refinements never solve the problem entirely, and the fact that such There has been a steady growth in the number of trans- drastic measures can be considered and indeed put plants performed, and as transplantation has become into practice underlines the severity of the donor organ more successful in terms of survival, quality of life, shortage.

7 Section 1: General

Another recent development has been the use Calne RY. Inhibition of the rejection of renel homografts in of organs procured from individuals who die with- dogs by purine analogues. Transplant Bull 1961; 28: out ever meeting brainstem death criteria. In these 445–61. patients, once cardiac activity has ceased, kidneys, Calne RY, Friend PJ, Moffatt S, et al. Prope tolerance, liver, and even lungs may be removed and used for perioperative campath IH, and low-dose cyclosporin transplantation as a result of advanced preservation monotherapy in renal allograft recipients. Lancet 1998; 351: 1701–2. techniques. However, despite the first successful heart transplant being performed using a donor of this Calne RY, Rolles K, White DJ, et al. Cyclosporin A initially as the only immunosuppressant in 34 recipients of nature, there has been no widespread adoption of the cadaveric organs; 32 kidneys, 2 pancreases, and 2 livers. non–heart-beating donor for cardiac transplantation. Lancet 1979; 2: 1033–6. Organ transplantation may be supplemented or Calne RY, Williams R. Liver transplantation in man. even replaced in due course using totally artificial Observations on technique and organization in five organs. The only implantable device that finds clin- cases. BMJ 1968; 4: 535–50. ical use at present is the artificial heart. The range Hamilton D. Kidney transplantation: a history. In Morris, of devices available and their apparent complexity PJ (ed). Kidney Transplantation.NewYork:Grune& underline the difficulties encountered in replacing Stratton, 1988, pp. 1–13. a relatively simple biological organ with mechani- Medawar PB. The behaviour and fate of skin autografts cal substitutes. Fundamental problems such as power and homografts in rabbits. JAnat1944; 79: 157– supply, thrombosis, infection and biocompatibility of 76. mechanical surface-blood interfaces remain, but these Merrill JP, Murray JE, Harrison JH, Guld WR. Successful obstacles may be overcome in due course to allow long- homotransplantations of the human kidney between term function. However, the replacement of those identical twins. JAMA 1956; 160: 277–82. organs with more complex metabolic functions is Merrill JP, Murray JE, Harrison JH, et al. Successful more difficult, and complete replacements for the kid- homotransplantations of the human kidney neys, lungs, and liver are still a long way distant. between non-identical twins. JAMA 1960; 262: The field of organ transplantation has grown mas- 1251–60. sively over the last hundred years. It has been made Murray JE, Merrill JP, Harrison JH. Renal possible by developments in individual disciplines, homotransplantation in identical twins. Surg. Forum supported by growth in our knowledge and under- 1955; 6: 432–6. standing of individual organ system physiology and Murray JE, Merrill JP, Harrison JH, Wilson RE, Dammin pathology. It remains a challenging and rewarding GJ. Prolonged survival of human kidney homografts by immunosuppressive drug therapy. NEngJMed1963; activity. However, successful as it is, transplantation is 268: 1315–23. not without problems, and it would not be possible at all it were not for the death, often in tragic circum- Report of the Ad Hoc Committee of the Harvard Medical School to examine the definition of brain death. JAMA stances, of patients who are suitable for organ dona- 1968; 205: 339–40. tion. Frequently the donors are young people who have Roitt IM, Greaves MF, Torrigiano G, Brostroff J, Playfair met an unexpected accident, or suffered a catastrophic JHL. The cellular basis of immunological responses. medical event such as subarachnoid hemorrhage, and Lancet 1969; ii: 366–71. their death is always an emotionally charged event. Schwartz RS, Damashek W. Drug-induced immunological Our reliance on the goodwill of the donor’s relatives to tolerance. Nature 1959; 183: 1682. make available their organs in order that others may Starzl TE, Marchioro TL, Von Kaulla KN, et al. live is somewhat perverse, yet it is central to the suc- Homotransplantation of the liver in humans. Surgery cess of transplantation. Gynecology and Obstetrics 1963; 117: 659–76. Starzl TE, Marchioro TL, Huntley RT, et al. Experimental Further reading and clinical homotransplantations of the liver. NY Acad Barnard CN. Human cardiac transplantation: an evaluation Sci 1964; 120: 739–65. of the first two operations performed at Groote Schuur Wallwork J, Williams R, Calne RY. Transplantation of Hospital, Cape Town. Ann Cardiol 1968; 22: 284–96. liver, heart, and lungs for primary biliary cirrhosis Calne RY. The rejection of renal homografts. Inhibition in and primary pulmonary hypertension. Lancet 1987: dogs by using 6-mercaptopurine. Lancet 1960; 1: 417. 182–4.

8 Section 1 General Chapter Immunological principles of acute rejection 2 Fadi G. Issa, Ryoichi Goto, and Kathryn J. Wood

Key points between genetically disparate (allogeneic) individuals, the immune system recognizes and reacts with the r The immune response to a transplant isa foreign antigens of the other individual (alloantigens) consequence of a complex interplay between on the transplant (allograft) to cause rejection. This the innate and adaptive immune systems. r rejection response is the result of interplay between The adaptive immune system mounts a thehostinnateandadaptiveimmunesystems.The highly destructive, sustained, and specific innate response is mediated by cells and molecules that attack on the transplant through recognition include macrophages, dendritic cells (DCs), granulo- of foreign antigens, activation of T cells, cytes (neutrophils, basophils, and eosinophils), natu- expansion of donor-reactive lymphocytes, ral killer (NK) cells, and the complement cascade, as and infiltration of allografts with effector well as proinflammatory cytokines and chemokines lymphocytes. r (chemoattractant cytokines). It represents a preformed Immunosuppressive drugs are required to defense that is immediately available until a specific prevent the immune system from destroying response can be mounted by the adaptive immune sys- the transplant. The majority of tem. The innate response is less specific than the adap- immunosuppressants act to inhibit T-cell tiveresponseandwillbeinducedevenifatransplant responses. r has been performed between genetically identical indi- Current immunosuppressive regimens have viduals (isograft), simply as a result of implanting or improved the short-term but not the transplanting the cells or organ. Adaptive immunity is long-term survival of organ transplants. The mediated by lymphocytes (T and B cells) and displays broad immunosuppressive activity of these slower kinetics than the innate response. However, drugsisassociatedwithserious the adaptive response is specific to foreign antigens complications, such as an increased risk of (alloresponse) and is therefore not activated by iso- malignancies and opportunistic infections. r grafts. Although the innate immune response is impor- Anidealsolutiontobothrejectionandthe tant for the initiation of the alloresponse and can ini- complications of immunosuppression is the tiate tissue damage, it cannot alone cause rejection (in induction of tolerance. Research on other words, the complete destruction of the tissue). achieving tolerance clinically is most On the other hand, the adaptive immune response promising in the fields of mixed chimerism is more damaging and is essential to rejection. The and regulatory T-cell therapy. importance of the adaptive response is reflected in the observation that animals experimentally depleted of T The immune system has evolved to clear the host of cellscannotrejectallografts. invading microorganisms and its own cells that have This chapter outlines the events involved in the become altered in some way, such as infected cells or adaptive and innate immune responses to a transplant mutated tumorigenic cells. The immune system rec- and the subsequent mechanisms of rejection, conclud- ognizes such cells as “foreign” and the molecules they ing with current clinical and experimental strategies to express as antigens. When organs are transplanted protect transplants from immune-mediated damage.

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

9 Section 1: General

Initiation of rejection donor cells and, to a lesser extent, minor histocom- patibility (miH) antigens. In humans, the MHC com- The immune system is frequently exposed to harm- plex is termed the humanleukocyteantigen(HLA)sys- less (and sometimes beneficial) foreign antigens that tem. miH antigens are peptides derived from other do not require an aggressive effector response, such molecules that are mismatched between the donor and as gut flora. The context in which such foreign anti- recipient and are presented by host MHC molecules gens are encountered is important in dictating the to host T cells. miH antigens alone cannot cause magnitude of the immune response. For instance, rapid rejection. However, when multiple miH are mis- theactivationofleukocytesinaninflammatoryenvi- matched, rejection can be as rapid as when MHC anti- ronment augments the immune response. In trans- gens are mismatched. miH mismatches alone may be plantation, these inflammatory signals can be pro- present in transplants between siblings with identi- vided by the surgical trauma, the oxidative stress of cal MHC molecules, leading to slow rejection of these ischemia/reperfusion injury (IRI), and brain death. transplants. Indeed, the innate immune response is mediated by There are two pathways by which foreign anti- cells that express invariant pattern recognition recep- gens are recognized by T cells. The more common tors (PRRs), such as Toll-like receptors (TLRs), that or natural one is called the indirect pathway.Anti- recognize altered endogenous molecules on the allo- gens, such as viral antigens, are first processed by graft produced as a result of tissue injury by reactive host APCs and then presented to host T cells by self- oxygen species (ROS), heat shock proteins (HSP), or MHC molecules on the APCs. In the transplant set- high-mobility group box 1 protein (HMGB-1) or as ting, the indirect pathway occurs when APCs process a direct consequence of donor brain death. Activa- and present donor HLA antigens to host T cells within tion of innate immune cells by TLR ligation results in self-MHC molecules. The TCR-CD3 complex on host the production of “danger” signals such as chemokines T cells recognizes unique features of the small pro- and preformed P-selectin (CD62P), which help recruit cessed donor HLA peptides (epitopes) in the context of anddirecthostleukocytesintothetransplantsite. self-MHC. The second pathway of allorecognition, the Macrophages release cytokines such as tumor necro- direct pathway, is the dominant pathway in transplan- sis factor (TNF) ␣, interleukin (IL)-1, and IL-6, which tation and occurs when T cells react directly with intact contribute to the inflammatory environment and assist donor HLA antigens. By way of comparison, T cells in the activation of other leukocytes. On recognition of that react to peptides derived from a nominal antigen inflammatory signals, antigen-presenting cells (APCs) (indirect pathway) are estimated to be less than 0.1% such as DCs in the allograft migrate to the draining of the total T-cell repertoire, whereas a much higher lymphoid tissues, where they present antigen to host frequency(about10%)ofTcellsreacttoanMHCmis- Tcells,leadingtoanadaptiveimmuneresponse. matched transplant (direct pathway). The recognition of foreign antigens by naive host Following organ transplantation, donor-derived (recipient) T cells (allorecognition,otherwiseknown “passenger” APCs residing in the donor organ and as signal 1) is a principal step in the rejection process. expressing large amounts of donor HLA antigens Allorecognition in the presence of costimulation (oth- migrate out of the transplant into the draining lym- erwise known as signal 2) results in the activation and phoid tissue, where they interact with host T cells expansion of T cells that recognize the mismatched via the direct pathway. With time after transplanta- donor alloantigens (alloreactive T cells). Alloreactive tion, passenger APCs diminish in number, and the TcellsorchestratethedevelopmentofTcellswith direct pathway becomes less important. In contrast, effector activity that can either have direct destructive the indirect pathway of allorecognition is maintained activity against the transplant or promote and amplify and remains active for as long as the transplant is B-cell function and other elements of the innate present. The direct pathway is therefore theoretically and adaptive immune response that can damage the more active during acute allograft rejection, whereas transplant. the indirect pathway becomes more important later in Allorecognition is mediated by the T-cell receptor chronic allograft rejection. (TCR), which is associated with the cluster of differen- A newly recognized third pathway, called the tiation (CD) 3 molecule (TCR-CD3 complex). TCRs semidirect pathway, may also be involved in allorecog- on host T cells bind to antigens encoded by genes nition. It occurs when intact donor HLA antigens are of the major histocompatibility complex (MHC) on 10 Chapter 2: Immunological principles of acute rejection

Figure 2.1 Allorecognition in transplantation occurs via the direct, indirect, and semi-direct pathways. Indirect allorecognition occurs when T-cell receptors (TCR) of T cells engage donor major histocompatibility complex (MHC) molecules that have been processed and presented by host antigen-presenting cells (APC) and presented in the context of self-MHC. Direct allorecognition is the recognition of intact MHC molecules on donor-derived passenger APCs by T host T cells. In semi-direct allorecognition, intact donor MHC molecules are transferred to recipient APCs by direct cell-to-cell contact or membrane fusion. These intact foreign molecules are then recognized by host T cells. CD4+ T cells engage MHC class II, whereas CD8+ T cells engage MHC class I. The effector response develops after allorecognition and T-cell costimulation by professional APCs. Activated T cells clonally expand, differentiate, and infiltrate the allograft. CD4+ T cells are polarized to a T helper (Th)1, Th2, or Th17 phenotype, depending on the local cytokine environment. Each Th subtype is associated with a distinct effector response. CD8+ T cells receive stimulation from activated CD4+ T cells and in turn produce interferon ␥ (IFN␥). Both activated CD8+ T cells and activated Th1 cells have the potential to become cytotoxic, predominantly employing perforin/granzyme and Fas/Fas ligand (FasL) to kill target cells, respectively. Th1 cells also produce IFN␥ and promote a delayed-type hypersensitivity (DTH) response from macrophages with the ensuing production of inflammatory molecules such as nitric oxide, tumor necrosis factor (TNF)␣, and reactive oxygen species. Th2 cells activate B cells to produce alloantibodies, which mediate complement activation or antibody-dependent cell-mediated cytotoxicity (ADCC). The hallmark of a Th17 response is neutrophil recruitment. See text for further details. physically transferred to the membrane of host APCs typically results in T-cell unresponsiveness, or anergy. and are then recognized by host T cells. Host APCs Costimulatory molecules are divided into two fami- appear to acquire intact HLA molecules from exo- lies: the B7 family, of which the prototype receptor- somes secreted by donor APCs or through cell-to-cell ligandpairareCD28(ontheTcell)andCD80/86 contact. The relative contribution of this pathway to (B7.1/B7.2, on the APC), and the TNF and TNF recep- allograft rejection is not clear. Figure 2.1 illustrates the tor (TNFR) family, best characterized by CD40 (on pathways of allorecognition and subsequent effector the APC) and CD154 (CD40L, on the T cell). Other mechanisms. costimulatory T-cell/APC pairs include CD27/CD70, Allorecognition alone is insufficient to promote inducible T-cell costimulator (ICOS or CD278)/ICOS T-cell activation. The second essential signal is cos- ligand, 4–1BB/4–1BB ligand, OX40/OX40 ligand, and timulation, which is provided by the interaction of CD279/CD274. Signaling via CD28 lowers the thresh- pairs of cell-surface molecules present on T cells and old for T-cell activation and increases the expression APCs. Absence or blockade of costimulatory signals of the T-cell growth factor (leukocytotropic) IL-2 by

11 Section 1: General

stabilizing the mRNA species, thereby promoting T- to generate the secondary messengers inositol triphos- cell proliferation and resistance to apoptosis. During phate (IP3) and diacylglycerol (DAG). IP3 leads to an immune response, activated T cells also upregu- the release of stored calcium from the endoplasmic late expression of cytotoxic T-lymphocyte antigen 4 reticulum (ER) and activation of phosphatase cal- (CTLA-4, CD152), a molecule that has close homol- cineurin, which dephosphorylates the transcription ogy to CD28 but has an inhibitory effect on T-cell acti- factor nuclear factor of activated T cells (NFAT), allow- vation. CTLA-4 has a higher affinity for CD28 than ing it to translocate to the nucleus. Generation of DAG CD80/86 and is able to attenuate immune responses results in the activation of the transcription factor by competing for CD28. CD28 signaling also upreg- nuclear factor-␬B(NF-␬B). The MAPK cascade also ulates expression of other costimulatory molecules, leads to the generation of transcription factor activa- such as CD154 (CD40 ligand), which, on ligation with tor protein 1 (AP-1). The action of these transcrip- CD40, activates APCs, leading to increased expres- tion factors alters the expression of many genes, and in sion of B7 family molecules and therefore a greater particular leads to upregulation of IL-2 and the high- ability to activate further T cells. The balance of pos- affinity IL-2 receptor (IL-2R) ␣-chain (CD25) required itive and negative signals transmitted through cos- for T-cell growth. Large amounts of IL-2 and other timulatory molecules to the T cell ultimately deter- leukocytotropic cytokines are produced and act to pro- mines whether the T cell will be activated or become vide further signaling to promote cell cycle progres- anergic. sion, clonal expansion, and differentiation of activated The interface of a T cell with an APC in which Tcells. both the TCR-MHC and costimulatory molecule inter- Activated lymphocytes also upregulate chemokine action occurs is termed the immunological synapse. receptor expression, allowing them to activate further This immunological synapse forms a “bull’s eye” struc- leukocytes and subsequently infiltrate the allograft. ture with a central supramolecular activation clus- The process by which leukocytes migrate into the graft ter (c-SMAC) containing the TCR-MHC complex, is termed leukocyte recruitment. Leukocyte recruit- surrounded by a peripheral SMAC (p-SMAC) ring ment is enhanced by vasodilation and endothelial acti- containing adhesion molecules, such as leukocyte vation in the vicinity of the transplant. Chemokines function-associated antigen 1 (LFA-1) on the T cell that have been released from the allograft become teth- bound to intercellular adhesion molecule 1 (ICAM-1, ered to the activated endothelium, providing a sig- CD54) on the APC. nal gradient recognized by passing leukocytes. When Within the cell membrane biphospholipid layer are leukocytes bind to activated endothelium, further cholesterol-rich regions that have been termed lipid adhesions are made between integrin molecules on the rafts. Certain membrane-bound molecules are prefer- leukocyte and endothelial adhesion molecules such as entially associated with lipid rafts, in particular, those ICAM-1, which result in arrest of the leukocyte and with lipophilic attachments to the cell membrane. In extravasation into the transplanted organ. resting T cells, the TCR-CD3 complex is not usu- Followingtheclearanceofapathogenbynaivecells ally associated with lipid rafts and is therefore unable of the adaptive immune system (the primary response), to interact with other signal transduction molecules a small number of antigen-specific T and B cells sur- found within these lipid rafts. During the formation of vive as memory cells that are able to mount a rapid an immunological synapse, clustering of signaling and response in the event of reintroduction of the same adhesion molecules occurs as a result of multiple TCRs pathogen (the secondary or memory response). The binding to MHC peptide on the surface of the APC. immune response to a transplanted organ results from A reorganization of the cell membrane subsequently the stimulus of both naive and memory alloreactive occurs, allowing TCR-CD3 complexes to integrate into Tcells.UnlikenaiveTcells,memoryTcellscansur- lipid rafts. This facilitates downstream signaling by vive in the absence of antigen and can be activated in placing the TCR-CD3 complex in close proximity to the absence of costimulatory molecules essential for signal transduction molecules, which are then acti- naiveTcells.MemoryTcellsmaybepresentdueto vated by phosphorylation. The end result is the activa- prior exposure to alloantigen during pregnancy, from a tion of the intracellular Ras and Rac mitogen-activated previous transplant, or from a blood transfusion. How- protein (MAP) kinase (MAPK) pathways and hydroly- ever, memory cells capable of responding to alloanti- sis of membrane phosphatidylinositol 4,5-biphosphate gen may also be present in individuals even without

12 Chapter 2: Immunological principles of acute rejection prior exposure to that antigen. This occurs through also recently been implicated in allograft rejection. the operation of three mechanisms: cross-reactivity Notably, although CD4+ activity is triggered in an through molecular mimicry from prior infectious antigen-specific manner, the effector mechanisms of agents, bystander proliferation following lymphope- allograft destruction are non-specific. nia, or heterologous immunity. Properties of mem- Th1 cells express the transcription factor T-bet ory cells not only include more rapid and efficient and produce IFN␥, TNF␣,andIL-2,whichresult responses to previously encountered antigen, but also in the activation of CD8+ cytotoxicity, macrophage- a resistance to apoptosis (programmed cell death) due dependent delayed-type hypersensitivity (DTH), and to the upregulation of anti-apoptotic molecules such as the synthesis of immunoglobulin (Ig) G2a by B cells Bcl-2 and Mcl-1. These characteristics confer an espe- (which activates complement), all of which contribute cially detrimental role for memory cells in rejection. to allograft rejection. Furthermore, Th1 cells express Fas-ligand (FasL), enabling them to exhibit cytotoxic activity. The Th1 DTH response is a nonspecific effec- The effector response tor mechanism that induces the production of medi- ators such as nitric oxide, ROS, and inflammatory The adaptive immune system arachidonic acid derivatives such as prostaglandin E2, Two types of T cells, identified by the cell surface mark- thromboxane, and leukotrienes from macrophages. ers CD4 and CD8, are active in rejection. CD4+ Tcells Th1-mediated effects have been shown to directly are activated by MHC (HLA) class II molecules, which affectgraftphysiologybyalteringcellpermeabilityand have two transmembrane domains and are expressed vascular smooth muscle tone and are implicated in the by APCs. Functionally, CD4+ Tcellsareusuallyhelper early stages of rejection, otherwise known as acute cel- T cells (Th) and are therefore often referred toas Th lular rejection. cells.CD4+ Th cells can differentiate into several sub- Th2 cells express the transcription factor GATA- types, including Th1, Th2, or Th17. In contrast, CD8+ 3 and secrete IL-4, IL-5, IL-9, IL-10, and IL-13, T cells are activated by MHC (HLA) class I molecules, which activate B cells (inducing Ig class switching) which differ structurally from MHC class II molecules and eosinophils to promote graft rejection primar- in that they have only one transmembrane domain. ily through the humoral immune response. B cells CD8+ T cells often have cytotoxic activity and are utilize surface Ig as an antigen receptor, internaliz- therefore known as cytotoxic T lymphocytes (CTLs). ing alloantigens that are processed and presented in Class I MHC is expressed, albeit at varying levels, by conjunction with class II MHC molecules. Antigen- all nucleated cells. As discussed in the previous section, specific recognition and costimulatory signaling from alloreactive T cells are activated only after an immuno- activated CD4+ T cells is required for the activation logical synapse is formed with an APC that pro- and differentiation of primary and memory B-cell vides the appropriate MHC and costimulation signals. responses that result in plasma cell generation and APCsthatareabletoprovidethesesignalsaretermed the production of alloantibodies. This results in immunostimulatory, or “professional,” and constitu- B-cell–induced antibody-mediated rejection (AMR), a tively express MHC class II (e.g., DCs, macrophages, phenomenon that is increasingly recognized as prob- and B cells). TLR ligation on DCs induces upregulation lematic in transplantation. AMR appears to be con- of costimulatory molecules and MHC class II, thus tributory in 20–30% of acute transplant rejection enhancingtheabilityoftheseAPCstoactivateTcells. episodes and up to 60% of chronic allograft dysfunc- So-called non-professional APCs express MHC class II tion cases. Antibodies directed against donor HLA only on stimulation with a cytokine, such as interferon molecules, ABO blood group antigens, or endothe- (IFN) ␥ (e.g., fibroblasts and endothelial cells). lial cell antigens may be generated during the immune CD4+ Th cells are critical for allograft destruc- response to the allograft, or in the case of antibodies tion. The type of Th response is determined bythe to endothelial cells, may be pre-existing at the time cytokine microenvironment in which APC and T-cell of transplantation. Patients with detectable anti-HLA interactions take place. Both cell-mediated immunity, antibodies at the time of transplantation have sig- driven by Th1 cells, and humoral immunity, driven nificantly worse graft survival rates than patients by Th2 cells, are independently capable of causing who are not sensitized, and the development of anti- allograft destruction. IL-17–producing Th17 cells have HLA antibodies in previously non-sensitized patients

13 Section 1: General

following transplantation is highly predictive of early B. In addition, CTLs are able to upregulate cell surface graft failure. AMR may be subdivided into hypera- expression of FasL and secrete soluble mediators such cute, acute, and chronic. Hyperacute AMR is a rare as TNF␣. Perforins polymerize and insert into the tar- event, occurring when recipients have preformed anti- get cell membrane, forming a pore that facilitates the body directed against allogeneic MHC molecules or entry of granzyme B and other compounds into the ABO isoagglutinins expressed on the graft endothe- cell.GranzymeBisaproteasethatisabletoinitiate lium.Itisdefinedbyrejectionoccurringwithin apoptosis by several mechanisms, including activation 24 hours of reperfusion and is characterized by imme- of caspase cascades. Binding of FasL to Fas on the tar- diate or near-immediate loss of graft function sec- getcellsurfaceisalsoabletotriggerapoptosisbyacti- ondary to complement-mediated thrombosis within vating caspases. the allograft vascular supply. Modern cross-matching techniques have made hyperacute rejection extremely rare, whereas acute AMR and chronic AMR remain The innate immune system problematic. Acute AMR occurs around the same Due to the intimate relationship of the adaptive and time as acute cellular rejection and is likely due to innate immune responses, many of the aspects of the a recall response of B cells that have been sensitized innate immune response have already been discussed. by a previous antigen encounter during pregnancy, a This section discusses the mechanisms of action of two bloodtransfusion,oraprevioustransplant.Chronic innate immune system components that have not been AMR is increasingly seen as a contributor to late attri- fully covered, the complement cascade and NK cells. tion of allografts that succumb to chronic graft dys- The complement cascade is a proteolytic cascade function. The hallmark of AMR is the activation of that generates a range of effector molecules: C5a and complement and membrane attack complex (MAC) C3a are chemoattractant molecules that assist leuko- formation, leading to target cell lysis. Positive histolog- cytes in migrating toward the allograft; C3b, C4b, and ical staining for complement 4d (C4d) in biopsies is their fragments opsonize cells (thus targeting them therefore indicative of AMR. Cell killing by antibody for destruction by phagocytes, e.g., macrophages and may also occur via a mechanism termed antibody- neutrophils) and facilitate antigen presentation and dependent cell-mediated cytotoxicity (ADCC), in which T-cell activation. The terminal components of the cas- NK cells or macrophages recognize and kill target cells cade, C5b-9, result in the formation of the MAC in the that have been coated in antibody. target cell membrane, inducing cell lysis. Apart from Th17 cells express the transcription factor ROR␥t being activated by immunoglobulin, complement and produce IL-17, IL-21, and IL-22, which act alone can be activated as a result of IRI, cytomegalovirus and synergistically with other cytokines to promote infection (CMV), and anti-lymphocyte antibody neutrophil recruitment to the site of rejection. In treatment. mouse experimental models, neutralization of IL-17 NK cells kill target cells in an identical manner to has been shown to reduce the features of vascular acute CTLs but do not possess antigen-specific TCRs and rejection of aortic allografts and to significantly extend do not require activation. NK cells express a variety of the survival of cardiac allografts. In a vessel allograft receptors that regulate their activity. Self-cells are able model, graft-derived IL-1 has been shown to promote to deliver an inhibitory signal to NK cells, whereas IL-17 production from alloreactive T cells, enhanc- infected or malignant cells cannot deliver this signal ing the production of the proinflammatory cytokines and are subsequently killed. Allogeneic cells are also IL-6, CXCL8, and CCL20. Further research is required unable to deliver an inhibitory signal to NK cells and to fully clarify the relative contribution of Th17 cells to theoretically should be destroyed, which is the case rejection. in bone marrow transplantation. Until recently, NK CD8+ Tcellscanbeinvolvedintransplantdestruc- cells have not been shown to contribute significantly tion via cytotoxic activity leading to cell death. Acti- to solid organ rejection. However, recent results vated CTLs migrate to the graft site, where they are demonstrate that NK cells can contribute to chronic able to identify their target cells in the graft by recog- rejection, at least in experimental models of heart nition of allogeneic class I MHC molecules. Once a transplantation. NK cells have also been shown to kill target cell is located, CTLs release granules contain- donor-derived APCs, in theory reducing the relative ing cytotoxic molecules such as perforin and granzyme contribution of direct allorecognition to rejection and

14 Chapter 2: Immunological principles of acute rejection

Figure 2.2 This figure illustrates the targets of some common immunosuppressants in clinical use. The majority of immunosuppressants target T cells, apart from rituximab (which targets B cells) and alemtuzumab (which targets most nucleated bone marrow–derived cells). See text for further details. AP-1: activator protein 1; APC: antigen presenting cell; CD: cluster of differentiation; FK-506: tacrolimus; IL: interleukin; IL-R: interleukin receptor; MMF: mycophenolate mofetil; mTOR: mammalian target of rapamycin; NFAT: nuclear factor of activated T cells; NF␬B: nuclear factor-␬B; PI3K: phosphoinositide 3-kinase; PkC␪: protein kinase-C␪. promoting tolerance. Further studies are needed to anisms of action in relation to the immunobiology, determine when NK cells may be harmful and when whereas clinical use is explored in the following chap- they may be beneficial to long-term graft survival. ter. Broadly speaking, immunosuppressants can be The tempo and timing of rejection is defined in divided into those that act on intracellular targets immunological terms and divided into hyperacute affecting signal initiation (such as antimetabolites and rejection, acute rejection, and delayed graft dysfunc- macrolides) or signal reception (such as mammalian tion. Hyperacute rejection, as discussed, is a rapid target of rapamycin [mTOR] inhibitors), and those act- event caused by preformed antibodies against allo- ing on extracellular targets (such as antibodies and geneic MHC molecules or ABO blood group antigens. fusion proteins). Corticosteroids cannot be placed into Acute rejection is characterized by a sudden deterio- one of these classes, as their effects are widespread. ration in transplant function over days to weeks and is Figure 2.2 summarizes the mechanisms of action of predominantly secondary to acute cellular rejection or common immunosuppressive drugs currently in clin- acute AMR. Delayed graft dysfunction, often referred ical use. to as chronic rejection, is a term that encompasses Corticosteroids act by binding to cytoplasmic glu- long-termdamagetotheorgancausedbothbythe cocorticoid receptors, altering the expression of multi- immune system and toxicity of immunosuppressive ple cytokines and inflammatory mediators by targeting agents and is often characterized by fibrointimal pro- the transcription factors NF-␬BandAP-1.Molecules liferation of intragraft arteries. affected include IL-1, IL-2, IL-3, IL-6, TNF-␣, IFN-␥, leukotrienes, and prostaglandins, as well as Modulating the immune system to several chemokines. Corticosteroids therefore possess both immunosuppressive and anti-inflammatory prevent rejection effects. At high doses, corticosteroids can have Immunosuppressive therapy can be credited with the receptor-independent effects. Side effects of corticos- vast improvements in transplant survival over the past teroid therapy include weight gain, hyperlipidemia, 50 years. This chapter explores the underlying mech- osteoporosis, and glucose intolerance.

15 Section 1: General

Antimetabolites such as azathioprine (AZA) and hypotension, rigors, and pulmonary edema, although mycophenolate mofetil (MMF) interfere with DNA it more commonly results in milder signs such as a synthesis and cell cycle progression, therefore impair- pyrexia and flu-like symptoms. This has led to the ing the clonal expansion of T cells. MMF is a more virtual abandonment of OKT3 for clinical use. lymphocyte-specific drug than AZA, which also has Newer monoclonal antibodies include alem- bone marrow suppressive effects. AZA is metabo- tuzumab, rituximab, basiliximab, and daclizumab, lized in the liver into 6-mercaptopurine, which is then which target specific T-cell surface proteins. Alem- incorporated into DNA, inhibiting purine nucleotide tuzumab is a humanized monoclonal antibody against synthesis with widespread effects on gene transcrip- human CD52, present on most mature nucleated tion and cell cycle progression. Sir Roy Calne orig- bone marrow–derived cells. Alemtuzumab therefore inally introduced 6-mercaptopurine as an experi- depletes T and B cells both centrally and peripherally, mental immunosuppressive therapy. AZA was subse- monocytes, macrophages, NK cells, and some granu- quently found to be less toxic than 6-mercaptopurine locytes. Evidence also suggests that it may expand the and was therefore pioneered as a clinical therapy. regulatory T-cell population, and it is likely to deplete MMF is metabolized in the liver into mycophenolic memory T cells as well. A single dose exerts a deple- acid, a non-competitive reversible inhibitor of inosine tional effect as profound and prolonged as multi-dose monophosphate (IMP) dehydrogenase, an enzyme administration of ATG. Recovery of these cells to required for purine generation. IMP dehydrogenase normal levels can take years after administration of inhibition has downstream effects on DNA and RNA alemtuzumab, and it is therefore reserved for special synthesis. circumstances in a select group of transplant recip- Calcineurin inhibitors (CNIs) are a subset of the ients for induction immunosuppression or for the macrolide compounds (so called because their activ- treatment of rejection episodes. Rituximab is directed ity depends on the structural presence of a macrolide against CD20, present on most mature B cells, and ring) and include cyclosporine and tacrolimus (FK- is useful for the treatment of AMR. Basiliximab and 506 or Fujimycin). Both drugs bind cytoplasmic daclizumab (which only differ slightly in structure) are immunophilins to form complexes that inhibit cal- humanized monoclonal antibodies directed against cineurin, a phosphatase enzyme in the T-cell sig- CD25, which is present on activated T and B cells. nal transduction pathway. Inhibition of calcineurin These antibodies bind and inhibit the high-affinity prevents the translocation of the transcription factor alpha chain of the IL-2 receptor (CD25), which is NFAT to the nucleus. Effects include inhibition of the expressed in greater density by antigen-activated T production of the cytokines IL-2, IL-4, TNF␣,and cells.ThustheyarethoughttotargetonlythoseTcells IFN␥, as well as the downregulation of costimulatory involved in rejection, avoiding the more generalized molecules such as CD154. immunosuppression and adverse effects associated mTOR inhibitors include sirolimus (rapamycin) with ATGs. and everolimus. These drugs act by binding and The advances in immunosuppression have inhibiting mTOR, which has a critical role in cytokine improved short- and medium-term graft survival receptor signal transduction, specifically in relation rates and reduced the rates of acute rejection, but to IL-2, IL-4, and IL-15. These cytokines act through this has not been followed by a comparable reduction mTOR to induce the production of proteins that are in long-term graft dysfunction rates. Furthermore, necessary for progression from the growth phase to the the immunosuppressive regimens currently used are DNA synthesis phase of the cell cycle and are therefore not ideal as they are non-specific, required lifelong, critical for T-cell clonal expansion. and risk the development of opportunistic infections Antibodies can be polyclonal, such as anti- andtumorsintransplantpatients.Thereistherefore thymocyte globulin (ATG) directed against multiple substantial research into strategies that may allow epitopes of antigens on human lymphocytes, or mon- a reduction or complete withdrawal of immuno- oclonal, such as OKT3 directed against human CD3ε. suppression with improved long-term outcomes in Both antibodies can initially activate lymphocytes, transplantation. Long-term graft acceptance with inducing the release of cytokines and leading to normal function in the complete absence of immuno- a“cytokinereleasesyndrome.”Thismaymanifest suppression with otherwise normal immune responses as a severe systemic inflammatory response with is known as tolerance and is the “holy grail” of trans-

16 Chapter 2: Immunological principles of acute rejection plantation immunology research. The hallmark of chronic graft dysfunction. Some experimental tech- tolerance is donor-specific immune hyporesponsive- niques induce Tregs in vivo by employing lympho- ness. Current experimental and early clinical strategies cyte depletion around the time of transplantation in to induce tolerance center on the use of regulatory T conjunction with a donor-specific antigen challenge, cells (Tregs) and the induction of chimerism. such as a donor-specific blood transfusion. Other tech- Tregs are a population of T cells with profound niques generate Tregs for therapy ex vivo by isola- suppressive or regulatory capabilities. Tregs physio- tion of nTregs from peripheral or cord blood and logically act to maintain immune tolerance against subsequent in vitro expansion, or by conversion of self-antigens and to provide negative feedback for non-Treg cell types to iTregs under certain in vitro immune responses that may become detrimental to cytokine and antigen environments. Several clinical the host. Patients with defects in the master tran- studies are currently running to test the safety and scription factor of Tregs, FoxP3, develop the devas- efficacy of Treg therapy for graft-versus-host disease tating autoimmune disease IPEX (immune dysreg- (GvHD) after bone marrow transplantation. To date, ulation, polyendocrinopathy, enteropathy X-linked), no trials in solid organ transplantation have been demonstrating the importance of Tregs in maintaining undertaken. Early reports from bone marrow trans- immune homeostasis. Because Tregs are able to sup- plantation trials have demonstrated that Tregs may be press effector responses in an antigen-specific manner, efficacious at inhibiting the development of GvHD there is potential for these cells to be used as a ther- without affecting the crucial graft-versus-tumor effect apy to suppress immune responses to an allograft while of treatment. keeping all other effector responses intact. Many types During T-cell development in the thymus, T cells of suppressive leukocytes exist, but the most studied that are strongly reactive to host MHC are deleted by populations are the naturally occurring Tregs (nTregs) aprocesstermednegative selection. This physiologi- that develop in the thymus and express CD4, CD25, cal process has been harnessed experimentally for the and FoxP3, and the inducible Tregs (iTregs) that are induction of tolerance to foreign antigens, whereby induced in the periphery under particular conditions hematopoietic complete chimerism (the replacement of cytokine and antigen exposure and that express of all host hematopoietic cells with donor-derived CD4 and FoxP3. Another population of regulatory stem cells) through myeloablative therapy and donor- Tcells,theCD4+ Tr1 cells, have also been described. derived bone marrow transplantation results in the These cells can be induced in the periphery and pro- repopulation of the host thymus with donor-type duce the suppressive cytokines IL-10 and transform- DCs that delete donor-reactive T cells. A number of ing growth factor ␤ (TGF␤) in a FoxP3-independent successful clinical cases have been reported whereby manner. patients with hematological indications for bone mar- Tregs suppress effector responses at multiple levels, row ablation who also require renal transplantation by directly inhibiting CD4+ and CD8+ T-cell activa- have received a bone marrow transplant and a kid- tion and proliferation as well as by modulating APC ney transplant from the same donor, resulting in long- function. Other targets of Tregs include B cells, NK term donor-specific tolerance. Nevertheless, the mor- cells, natural killer T (NKT) cells, and mast cells. bidity and mortality of myeloablative therapy and risk Mechanisms of Treg suppression include the cytoly- of GvHD in most transplant patients makes this mode sis of target cells by perforin and granzyme B; the of therapy unacceptable to those without a hemato- secretion of the inhibitory cytokines IL-10, TGF␤,and logical indication for bone marrow ablation. On the IL-35; and the consumption of IL-2 in the surround- other hand, mixed chimerism, in which donor cells ing environment by their high-affinity CD25 recep- represent a varying proportion (but not 100%) of the tors, therefore depriving naive and effector T cells of total hematopoietic pool, is a more promising area of this growth factor. Furthermore, Tregs express CTLA- research. Mixed chimerism can be established using 4, which, as described previously, prevents the costim- non-myeloablative conditioning regimens, therefore ulatory interaction of CD80/86 with CD28. maintaining immunocompetence and reducing the Various studies have demonstrated the ability of risk of GvHD. Tregs to induce long-term graft survival experimen- Two promising clinical trials utilizing mixed tally, with some studies even demonstrating inhibi- chimerism for the induction of tolerance have been tion of transplant arteriosclerosis, a manifestation of performed. An initial trial enrolled six patients

17 Section 1: General

with renal failure consequent to multiple myeloma, ity of patients, reducing immunosuppression to a min- a hematological malignancy. Patients received non- imal level would offer many advantages in terms of myeloablative bone marrow transplants and renal reduced complications of long-term drug therapy. This transplants from an HLA-identical sibling followed state, in which graft function is maintained in the pres- by a donor leukocyte infusion as treatment for both ence of low doses of non-toxic immunosuppression, the multiple myeloma and renal failure. These patients has been termed prope tolerance and may represent a successfully accepted their renal transplants long- more realistic goal. term without any immunosuppression. Following this study, a similar approach was piloted in five patients Further reading without a hematological malignancy. Patients received Brent L. A History of Transplantation Immunology.San an HLA-mismatched haploidentical related donor Diego, CA: Academic Press/Elsevier, 1996. bone marrow transplant along with a renal trans- Ginns LC, Cosimi AB, Morris PJ. Organ Transplantation. plant from the same donor. Four patients in the trial Oxford: Wiley-Blackwell, 1999. currently maintain graft function after weaning from KingsleyCI,NadigSN,WoodKJ.Transplantation their initial immunosuppression (follow-up 2–5 years tolerance: lessons from experimental rodent models. postweaning). However, one kidney transplant was Transpl Int 2007; 20: 828–41. lost due to acute AMR, leading to a modification Paul WE. Fundamental Immunology. Philadelphia: in the trial protocol to include B-cell depletion with Lippincott Williams & Wilkins, 2008. rituximab. Warrell DA, Cox TM, Firth JD. Oxford Textbook of Although the attainment of tolerance is an ideal Medicine. Oxford: Oxford University Press, 2010. solution, whether this can be achieved in each and Wood KJ, Sakaguchi S. Regulatory T cells in transplantation every transplant recipient is unknown. For the major- tolerance. Nat Rev Immunol 2003; 3: 199–210.

18 Section 1 General Chapter Immunosuppression Past, present, and future 3 Vineeta Kumar and Robert S. Gaston

Key points a basis for understanding newer therapeutics on the horizon. r The modern era of transplantation was made possible by the introduction of azathioprine andcyclosporinetopreventacuterejection. Immunosuppression: past r Modern immunosuppression regimes utilize induction therapy with polyclonal or The AZA era (1962–1981) monoclonal antibodies directed against T cells followed by maintenance therapy. Long-term survival of allografts in humans first occurred with the introduction of azathioprine (AZA), r Clinical trials suggest that tacrolimus, a modification of 6-mercaptopurine (6-MP) in the mycophenolate mofetil, and corticosteroid as early 1960s. It was recognized that most renal recipi- maintenance are better at preventing acute ents experienced, usually within the first month after rejection. engraftment, a rejection “crisis,” comprised of graft r Newagentshavethepotentialtoprovide tenderness, fever, reduced urine production, and ris- more effective immunosuppression by ing blood urea. These crises could be ameliorated with targeting different immunobiological high doses of corticosteroids, often requiring repeated pathways. administration. Ultimately, it was recognized that the best patient outcomes were fostered with concomi- Clinical solid organ transplantation became a real- tant daily or alternate-day use of smaller corticosteroid ity with the serendipitous recognition that use of an doses together with AZA, and experience in those isograft from a genetically identical living donor cir- early years defined the proper dosing regimen (1.5– cumvented immunological responses. However, the 3 mg/kg/day) for AZA. In the absence of ongoing rejec- ability to utilize solid organ transplants as therapy tion, renal function was well preserved. However, as for large numbers of patients with end-organ failure many as half the allografts failed within a year of trans- is a direct consequence of the development of phar- plantation, and opportunistic infections (thought to be macological immunosuppression. The Nobel Prizes the consequence of high-dose steroids) were a com- awarded to Hitchings and Elion, Medawar, Murray, mon cause of mortality. and Dausset bear witness to the interplay of sur- Production and administration of anti-lymphocyte gical skill, immunological understanding, and bold globulin (ALG) or anti-thymocyte globulin (ATG) therapeutics that still resides at the core of clinical emerged during this same period, with source (rabbit transplantation. To the newly initiated struggling to or equine) and immunizing agent (harvested thymic understand the whys and wherefores of current tissue or cultured lymphoblasts) often dependent on immunosuppression, lessons learned in the past help resources available at individual transplant centers. simplify the process. Furthermore, future therapies Earliest use of these agents was as adjunctive treatment are likely to evolve from equally important experience for rejection “crises,” although by the mid 1970s, sev- acquired in the present. This chapter focuses on cur- eral centers were administering ATG or ALG prophy- rent practice, as informed by past experiences and as lactically at the time of transplantation to either reduce

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

19 Section 1: General

or delay onset of rejection episodes. The first multi- in the United States was the report in a prominent center randomized, controlled trials in kidney trans- journal of progressive CyA nephrotoxicity in recip- plantation were two studies of prophylactic ATG con- ients of cardiac transplants, with onset of end-stage ducted in the late 1970s and early 1980s: the Upjohn renal disease (ESRD) in an alarming number. This equine ATG (Atgam) trial in the United States and the resulted in two major changes in approach to CyA- Medical Research Council trial in the United King- based immunosuppression. The first was rediscovery dom. Both produced similar results: overall outcomes of AZA, and “triple” therapy, with low-dose AZA (graftandpatientsurvival)werenobetterwithpro- and corticosteroids utilized to reduce the maintenance phylactic ATG, but treated patients tended to have CyA requirement to 8–12 mg/kg/day, with equally fewer rejection episodes, occurring later after trans- effective control of rejection and better preservation of plantation, and with better preservation of kidney kidney function. The second was based on the realiza- function. tion that, unlike assumptions based on animal mod- With or without prophylactic ATG, kidney trans- els, CyA could be effective when introduced late, sev- plantation during this period was challenging due to eral days or more after engraftment, under effective rejectionontheonehandandimmunosuppressive ALG-based alternative prophylaxis. Thus sequential or complications on the other, but with enough long-term quadruple immunosuppression was born, with non- survivors to push forward. It was during this period nephrotoxic ALG induction allowing delayed admin- that chronic dialysis emerged as the “default” ther- istration of nephrotoxic CyA (in even lower doses) apy for end-stage renal disease, with transplantation until after adequate allograft function could be estab- limited more by poor outcomes than any other vari- lished. Although some remained committed to imme- able. Successful transplantation of other solid organs diate CyA use in higher doses either alone or with (liver, lung, heart) was even more challenging dur- steroids, the most effective, widely used protocols con- ing this era of relative ineffective immunosuppres- tained some or all of these modifications. sion, though surgical technique and organ preserva- Parallel to these developments were advances tion became ever more sophisticated. in a new technology: monoclonal antibodies pro- duced via the mouse hybridoma technique of Kohler and Milstein (recipients of yet another Nobel The cyclosporine era (1982–1995) Prize). Muromonab-CD3 (OKT3), an early anti–T- Early use of cyclosporine (CyA) in animals and lymphocyte monoclonal antibody, was compared with humans as monotherapy (with doses commonly rang- corticosteroids in a prospective, randomized trial for ing between 20 and 50 mg/kg/day) seemed effective treatment of an initial rejection crisis, with better in preventing acute rejection crises, but was associ- efficacy in reversing the episode, but little impact on ated with significant graft dysfunction among other recurrent rejection and only marginal improvement in disabling toxic effects. Ultimately, the drug was made graft survival at 1 year. On the basis of this trial, OKT3 available to several groups for clinical trials, some at was released for use in the United States in 1986 and individual centers and others multi-center, each of hailed as the first of many monoclonal antibodies that which modified dosing and administration in some would enable delivery of targeted immunosuppression fashion. Simply using lower doses to minimize tox- to transplant recipients. icity seemed effective in some patients; others com- These advances in immunosuppression enabled bined lower doses with long-term corticosteroids (at solid organ transplantation to emerge as a major ther- lower doses than previously thought necessary) or apy for end-organ failure. In kidney transplantation, ALG to produce excellent outcomes, with rejection triple or quadruple immunosuppression resulted in 1- rates cut to 50% rather than near 100%, and 1-year year allograft survival routinely in excess of 80%, with renal graft survival of 70–75%. These approaches even rejection rates of 40–50%. OKT3 became an alternative seemed to work in extra-renal transplantation, with to polyclonal ALG for induction at some centers. Those successful cardiac and hepatic transplantation in ever- who experienced rejection most often were treated ini- growing numbers of patients. Some even deemed AZA tially with high-dose steroids; “steroid-resistant” rejec- “obsolete.” tions were treated with a second course of OKT3 Thereremained,however,oneoverriding,nag- or ALG. Long-term toxicities, including renal dys- ging issue: nephrotoxicity. Coincident with its release function, seemed manageable. In addition, financial

20 Chapter 3: Immunosuppression profits associated with manufacture and use of CyA SRL was administered in fixed doses of 2 mg and 5 mg and OKT3 (Sandoz and Ortho Biotech, respectively) daily, in combination with CyA and corticosteroids. laid to rest the notion that transplantation as an In these trials, although SRL reduced the risk of acute enterprise was too limited in scope to justify private rejection compared with AZA or placebo, it was asso- investment in novel therapeutics. The ensuing decade ciated with significant complications (nephrotoxicity, witnessed the development and approval of at least a impaired wound healing, hyperlipidemia) that have, in half dozen novel agents, ushering in the era of what practice, limited its utility. some have called modern immunosuppression. By the mid 1990s, antibody induction had become less popular due to the disappearance of locally pro- duced polyclonal ATG/ALG, the cost and relative Modern immunosuppression (1995–) unreliability of the Upjohn equine polyclonal (Atgam), Tacrolimus (TAC), known during its developmental and the lack of effective alternatives. In 1995, OKT3 phase as FK506, was discovered in a soil sample from was the primary induction agent utilized, but only Japan in 1984. Early development occurred largely at a in a quarter of kidney transplant recipients. The sec- single center (the University of Pittsburgh), with use ond half of the decade, however, witnessed introduc- as a single agent or in combination with low-dose tion of several new biologic preparations into the mar- corticosteroids and primarily in liver transplantation. ketplace, including Thymoglobulin (rabbit ATG) and US Food and Drug Administration (FDA) approval of alemtuzumab, which was developed as an anti-tumor TAC (initially for liver transplantation) came in 1994 agent, but quickly became widely used in transplan- based on two large multicenter phase 3 trials (one in tation. Availability of these agents has coincided with the United States and one in Europe) involving over novel understanding and clinical approaches to induc- a thousand liver transplant recipients. In both studies, tion immunosuppression. acute rejection episodes occurred less often with TAC, Thus transplant pharmacotherapeutics entered the though only the European study documented a sur- new millennium with nearly all the tools that com- vival benefit. A subsequent trial in kidney transplanta- prise current immunosuppressive protocols. It must tion (comparing TAC versus CyA in combination with be stated that use of these potent new agents would ALG induction, AZA, and prednisone) with similar have been nearly impossible without commensurate results (less rejection, especially in African Americans) advances in anti-infective therapies, notably anti- led to approval in kidney transplantation. virals (ganciclovir and valganciclovir) and anti-fungals Mycophenolate mofetil (MMF) was a new modi- (liposomal amphotericin, imidazoles, etc.). Current fied preparation of an older agent (mycophenolic acid, practice in immunosuppression is the product of pro- MPA) that enhanced its absorption and stability. After tocols that have evolved over the last decade in an dosing and safety were established in a small prelimi- attempt to provide the most effective therapy with the nary study, three large phase 3 studies (1500 patients) least adverse impact on patients. were conducted in North America, Europe, and Australia, each showing a 50% decline in frequency of acute rejection in kidney transplant recipients treated Immunosuppression: present with several different CyA-based regimens, with effi- Current practice involves utilizing the agents available cacy documented for most patients at a 2-g daily in combination protocols, a theme that has defined dose. Pooling of data from the three studies docu- clinical immunosuppression almost since its incep- mented additional benefit from MMF in improving tion: drugs with differing mechanisms of action are allograft survival at 1 and 3 years. These data led to administered to inhibit immune responses at multiple FDA approval and release in 1995, with a subsequent steps in the rejection cascade. This approach is thought registration trial also documenting efficacy in cardiac toenableuseoflowerdosesofeachindividualagentto transplantation. minimize toxicity, while also most effectively prevent- Sirolimus (SRL) was discovered in a soil sam- ing or treating allograft rejection. Induction is now a ple collected in a Pacific island, Rapa Nui. Originally widely accepted concept, referring to immunosuppres- developed as an anti-fungal agent, it was later found to sion administered in the initial peri-transplant period, have immunosuppressive and antiproliferative prop- usually antibody-based, to prevent early rejection and erties. In the pivotal trials that led to FDA approval, set the stage for long-term therapy. The primary

21 Section 1: General

purpose of induction therapy is immediate estab- syndrome or potentially anaphylaxis. Thus rATG is lishment of effective immunosuppression, with sec- most commonly administered after pretreatment with ondary goals of delaying administration of potentially corticosteroids, acetaminophen, and anti-histamine. nephrotoxic agents (calcineurin inhibitors, CNI) and Serum sickness is infrequent with this preparation, but modifying immune responses to enhance long-term can occur in some cases. Common early adverse effects engraftment. Maintenance immunosuppression is the include leukopenia and thrombocytopenia, with the long-term therapy required to ensure allograft sur- latter considered an indication for dose reduction. vival, administered with the dual intentions of avoid- Long-term side effects generally reflect burden of over- ing both immunological injury and drug-related all immunosuppression. Despite its cost (as much as toxicity. USD 1000 per dose), rATG is the most commonly used induction agent in the United States. Alemtuzumab is available in the United States Induction afterFDAapprovalastherapyforchroniclympho- Induction agents can be sub-classified as depletional cytic leukemia; its use in transplantation has grown or non-depletional. Depletional agents including in part due to its low cost relative to other deple- polyclonal (ATG) and monoclonal (alemtuzumab) tional agents. The potent and long-lasting effects of preparations reduce lymphocyte populations in alemtuzumab have been viewed as enabling minimiza- both central and peripheral lymphoid tissue, with a tion protocols in which one or more maintenance long-lasting impact on absolute lymphocyte counts. agents (usually corticosteroids or CNIs) are avoided. Non-depletional agents modulate immune responses Several single and multi-center studies support its without altering lymphocyte counts and include utility in this regard. Various dosing regimens have basiliximab and OKT3. been used, including repetitive dosing of alemtuzumab Polyclonal ATG (or ALG) is manufactured by asmonotherapyafterkidneyorpancreastransplan- immunizing animals with crude preparations of tation. Most commonly, alemtuzumab is adminis- human thymocytes (or cultured lymphoblasts). Serum tered at the time of transplantation as a single 30- from the animals (horses or rabbits) is then recovered mg intravenous dose. There appear to be few adverse and purified to its immunoglobulin G (IgG) compo- effects associated with infusion; most long-term side nent. Not only is there activity against lymphocytes, effects are a consequence of its potent, long-lasting but these preparations also contain non-specific IgG lymphodepletion. reactivity against platelets and other leukocyte recep- The only current agent actually approved for tors. Administration in humans is thought to induce induction usage is basiliximab, available since 1997. both complement-dependent cytolysis and apopto- Compared with placebo, and combined with CyA- sis, resulting in lymphodepletion that may require based regimens that did not include MMF or SRL, 6–9 months to return to baseline. A depletional effect this agent, when administered at the time of trans- on other lymphoid cells (including B lymphocytes and plantation, reduced the incidence of acute rejection natural killer [NK] cells) may also contribute to ATG byhalf,withfew,ifany,significantadverseeffects. efficacy. Although shown unequivocally to reduce incidence Rabbit ATG (rATG), which received FDA approval of acute rejection compared with placebo, efficacy of for treatment of acute rejection after kidney trans- basiliximab has been thought by some to be limited to plantation, is now more commonly used as an induc- relatively low-risk patients. In a subsequent ran- tion agent. Dosage and timing of administration are domized trial, with a study population enriched for widely variable, but a usual dose of rATG is 0.5– high-risk patients (African Americans, highly sensi- 1.5 mg/kg/day for a total of 3–5 doses. The maxi- tized), basiliximab was shown to be less effective than mal total per treatment course is generally limited rATG in preventing rejection, although graft survival to 10 mg/kg, with larger doses thought to be asso- was identical among treatment groups. A more recent ciated with increased risk of complications, notably trial found more effective prevention of acute rejection post-transplant lymphoproliferative disease (PTLD). with alemtuzumab compared with basiliximab in a Infusion-related side effects are seen with the first or corticosteroid-free protocol. Thus use of basiliximab second dose, generally within the first few hours of tends to be limited to relatively low-risk patients. infusion, and are typically related to cytokine release As the only antibody preparation FDA-approved

22 Chapter 3: Immunosuppression

Table 3.1 Calcineurin inhibitors (CNI) and drug interactions Effect Drugs Drugs that increase CNI levels Calcium Channel Blockers: diltiazem, nicardipine, verapamil Antifungals: fluconazole, itraconazole, ketoconazole, voriconazole Macrolides: erythromycin, azithromycin, clarithromycin Gastric acid suppressants: lansoprazole, rabeprazole, cimetidine Others: methylprednisolone, allopurinol, colchicine, bromocriptine, metoclopramide, amiodarone, grapefruit juice, INH, protease inhibitors, statins Drugs that reduce CNI levels Anti-tubercular drugs: rifabutin, rifampin Anti-convulsants: carbamazepine, phenobarbital, phenytoin Others: octreotide, ticlopidine, St. Johns wort, orlistat, bosentan, nafcillin, Drugs levels increased by CNI Statins, methotrexate Drugs levels decreased by CNI Digoxin for induction therapy, however, basiliximab is also 10 ng/ml. Significant drug–drug interactions impact commonly used in registration trials of novel serumlevelsofbothTACandCyA(Table 3.1). immunosuppressants. Basiliximab is administered Studies comparing efficacy between TAC and CyA intravenously as a fixed-dose 20-mg infusion on days have consistently shown little difference in patient 0and4aftertransplantation.Ithasnotbeenshownto andallograftsurvival,withmost,however,indicat- be effective in reversing established rejection. ing less acute rejection, and although both drugs are nephrotoxic, higher glomerular filtration rate (GFR) is demonstrated (at least in the short term) with TAC usage. In general, TAC demonstrates a more favor- Maintenance immunosuppression able cardiovascular disease risk profile (in terms of Despite the potency of induction therapy, it is gener- less hyperlipidemia and less hypertension) than CyA. ally accepted that without long-term immunosuppres- Alternatively, although both agents are associated with sion, allografts would fail. Although TAC and MMF, glucose intolerance and new-onset diabetes after trans- among others, have been shown to reverse established plantation (NODAT), incidence is generally greater rejection at times, they are, in general, considered to with TAC than CsA. Cosmetically, more hirsutism be less potent than induction agents (or high-dose and gingival hyperplasia occur with CsA, whereas a corticosteroids), and their primary purpose is to pre- small percentage of TAC-treated patients experience vent acute rejection with a tolerable profile regarding alopecia. adverse effects. MMF and MPS MMF dosing was derived from the phase 3 trials noted CNI(TACandCyA) earlier. From this experience, recommended oral or TAC has been viewed by many, but not all, as an intravenous dosage for MMF is 1000 mg admin- improvement over CyA. These agents are not used istered twice daily, with doses not to exceed 3000 together, but one or the other forms the basis of most mg/day. It is recommended that MMF be admin- immunosuppressive protocols. CyA is usually initiated istered on an empty stomach due to clinical evi- at a dose of 4–8 mg/kg/day, with target trough lev- dence that food decreases MPA maximum concentra- els (depending on the assay used) of 100–300 ng/ml. tion (Cmax) by 40%. Extensive safety data are avail- Current practices with TAC typically involve a start- able from the clinical trials of MMF, with the most ing dose of 0.05–0.15 mg/kg/day and a maintenance common adverse events related to the gastrointestinal dose of 0.05–0.1 mg/kg/day (6–36 months after trans- tract: diarrhea, nausea, bloating, dyspepsia, and rarely plantation), especially when used in conjunction with gastrointestinal bleeds. Enteric-coated mycopheno- MMF. There is a wide range of acceptable TAC levels late sodium (EC-MPS) is thought to be therapeuti- for solid organ transplant recipients; however, when cally equivalent to MMF at comparable doses. Claims used in combination with MMF, data indicate thera- of fewer gastrointestinal side effects with EC-MPS peutic efficacy at a trough level between 5 ng/ml and compared with MMF have not been substantiated

23 Section 1: General

by randomized trials, although some patients clearly 2 mg/day is recommended with a targeted trough level benefit. Other adverse effects of both MMF and between5and15ng/ml.Whenusedinnon-CNIcom- AZA include hepatotoxicity, bone marrow suppres- binations, some recommended higher trough levels sion (anemia, thrombocytopenia, and leukopenia), (10–25 ng/ml), although side ffects are more com- and macrocytosis. The manufacturers of MMF do mon with higher levels. SRL, administered concomi- not recommend use during pregnancy, although some tantly with CyA, seems to exacerbate the nephrotoxic- centers report successful outcomes of pregnancy ity and increases the risk of acute rejection compared on MMF. Switching immunosuppression in prepara- with combinations. Apart from this nephrotoxic effect, tion for pregnancy has inherent risks, in particular, common reasons for discontinuation of SRL include rejection. hyperlipidemia, proteinuria, mouth ulcers, peripheral As noted above, MMF and EC-MPS dosing were edema, pneumonitis, impaired wound healing, lym- developed empirically in combination with CyA. It is phocele, and anemia/thrombocytopenia. Everolimus, now apparent that pharmacokinetic (PK) profiles vary also known as RAD during development, is another considerably when both preparations are used in com- mammalian target of rapamycin (mTOR) inhibitor bination with other agents, including TAC and SRL. It with a shorter half-life. Its more benign side effect pro- is now known that CyA inhibits enterohepatic recir- file may represent usage in phase 3 trials at a relatively culation of MPA, lowering plasma levels and overall lower dose and target levels than SRL. exposure by as much as 50%, and that bioavailabil- SRL, when administered concomitantly with TAC, ity increases dramatically in most patients over the shows less potential to exacerbate CNI-related adverse first 4–6 weeks after transplantation. The effect canbe effects.Inseveralrandomizedtrials,outcomeswiththe highly variable, although a recent study documented TAC/SRL combination appear comparable to those of intra-patient pharmacokinetic variability as very low. TAC/MMF and superior to those of CyA/SRL. SRL or Some have advocated monitoring of MPA drug lev- everolimus used in a non-CNI combination has been els as a guide to proper dosing, with AUC measure- used de novo following transplantation (although with ments more predictive of outcomes than trough lev- greater risk of acute rejection) but may be useful late els. A recent consensus panel recommended that for after transplantation when there is significant nephro- standard-risk patients, MPA monitoring may not be toxicity and may allow some recovery or delay pro- necessary. Empiric doses of 3000 mg/day during the gression to ESRD. There is also the suggestion that peri-transplant period, with subsequent reduction to the anti-proliferative effects may be useful in patients 2000 mg/day, provided adequate MPA exposure for with evidence of vasculopathy or after treatment for most CyA-treated patients. For TAC-treated recipi- malignancy. ents, an initial dose of 2000 mg/day, with subsequent reduction to 1500 mg/day, was adequate for most. In patient sub-groups deemed at higher risk for rejection Prophylaxis of acute rejection (e.g., diabetics or others with altered gut absorption; The primary goal of current immunosuppressive pro- elevated immunological risk; or undergoing CNI or tocols is to prevent acute rejection, a goal achiev- corticosteroid withdrawal), MPA monitoring may be able with numerous combinations. Most common beneficial in establishing the correct dose. Because of approaches begin with an induction agent in the peri- the altered PK profile associated with its enteric coat- operative period, accompanied by high-dose corti- ing, the relationship of trough levels to overall MPA costeroids. Appropriate doses of maintenance agents exposure with EC-MPS is not reliable. are instituted during this period; once desirable ther- apeutic levels of maintenance agents are achieved, induction is discontinued. Given the multitude of SRL agents available, there is wide center-to-center and Although approved as a fixed-dose adjunct to CyA, geographic variation, driven by drug availability, cost, SRLisseldomusedinthisfashioninthepresent and physician preference. day. Trough level monitoring, with appropriate dose Standard of care in kidney transplantation in the adjustment, is now recommended. A loading dose United States is use of antibody induction (78% of of 4–6 mg/day followed by a maintenance dose of recipients) with either rATG or basiliximab (given

24 Chapter 3: Immunosuppression

Table 3.2 Immunosuppression use for induction and daclizumab induction with low-dose CyA, TAC, or maintenance in the United States SRL. Patients in the TAC group had the fewest episodes 2007 1998 of acute rejection after 1 and 3 years, with better allo- Functioning graft at initial discharge 16 252 12 002 graft survival and better preservation of renal func- tion than the other groups. Serious adverse events were With immunosuppression information 15 884 11 933 most common among patients in the SRL group, which Induction (%) rATG (Thymoglobulin) 43 0.2 alsohadthemostrejectionand,despitetheavoidance Basiliximab/daclizumab 26 17 of CNI, inferior renal allograft function. NODAT and Alemtuzumab 9 0 diarrhea were more common in patients treated with ATGAM/OKT3 2 22 Information not recorded 20 61 tacrolimus. Beyond current standards, trends in maintenance Maintenance regimen (%) TAC + MMF/MPA + steroids 51 20 immunosuppression are to minimize or avoid those TAC + MMF/MPA 28 0.6 agents most associated with long-term toxicity, notably CsA+ MMF/MPA + steroids 5.4 48 corticosteroids and CNIs. In renal transplantation, CsA + Ever/sirolimus + steroids 1 3 CsA + Anti-metabolite + steroids 0.1 9 minimization is perceived to be more dependent on depletional induction than standard therapy, with Source: Organ Procurement and Transplantation Network/SRTR Data as of May 1, 2008. rATG and alemtuzumab as the agents of choice. A recent randomized multicenter study indicated that early corticosteroid withdrawal can best be accom- plished with the least risk of acute rejection in the limited availability of daclizumab), TAC, MMF, and intermediate term (3 years) with either rATG or alem- corticosteroids (prednisone, tapered to 5–10 mg/day tuzumab as compared with basiliximab. Alemtuzumab by 4–6 weeks after transplant). Induction is less com- was associated with fewer infectious complications mon in lung (62%), heart (54%), small bowel (49%), than rATG and was administered at a lower cost. It is in and liver (25%) according to 2007 data from the US this context that alemtuzumab use is increasing rapidly Scientific Registry of Transplant Recipients (SRTR). in the United States. Overall, use of antibody induction is less common in A 5-year, randomized, double-blind study com- Europe. Although TAC is the preferred CNI in 80% paring early corticosteroid withdrawal to maintenance of patients, CyA is still widely used at some centers steroid therapy (5–10 mg/day) after antibody induc- in the US and Europe. Likewise, although MMF is tion with concomitant TAC/MMF was recently com- used in 87% of kidney recipients, some choose SRL or pleted. Overall, there was no difference in graft or AZA instead. Triple therapy remains the standard in patient survival after 5 years. However, there was sig- thoracic transplantation, with TAC/steroids (and no nificantly more acute rejection (27% versus 17%) in the induction) the most common approach in hepatic steroid-free group, with more interstitial fibrosis and transplantation (Table 3.2). tubular atrophy (IF/TA) in for-cause biopsies taken Numerous studies support current therapies in fromthesamegroup.However,somemetabolicben- kidney transplantation. Mourad and colleagues stud- efits did accrue to the steroid-free group, with less ied rATG induction in an open-label multi-center NODAT and better preservation of bone histology. studyusingaTAC-basedregimeninkidneytrans- Overall, about 25% of US kidney recipients are now on plant recipients. Compared with a non-induction pro- steroid-free protocols. tocol, there was comparable graft survival between As documented in the ELITE-Symphony trial, CNI the two approaches but less rejection in the rATG minimization or avoidance has, to the present time, group. Recently, the Efficacy Limiting Toxicity Elimi- been perceived as relatively ineffective because of nation (ELITE)–Symphony study, a global project that the inability of alternative agents to prevent rejec- involved 1600 subjects in 83 centers randomized to tion. With risk of acute rejection highest in the first four different protocols, was completed. All groups 90 days after transplantation, and risk of SRL adverse received standard dosing of MMF and corticosteroids. effects most daunting during the same period, sev- The control group was standard dosing of CyA with- eral studies of planned conversion from CNI to SRL out antibody induction. The other groups all received have been conducted. Although one study in liver

25 Section 1: General

and one in kidney documented little associated rejec- of AMR is often less simple and may be co-existent tion and improved renal function with conversion to with cell-mediated and chronic injury. The benefit of SRL, others have indicated excessive rejection compli- the protocols outlined above (IVIg, TPEX, with or cated by proteinuria, anemia, and graft failure. It would without rituximab) become less well defined, though appear that widespread adoption of CNI minimization still often implemented. Several novel therapies (see or avoidance will not occur until late after transplant, next section) are currently undergoing investigation. when the risk of rejection is lower or more effective alternative agents become available. Generic drugs Generic substitution of immunosuppressants remains Treatment of acute rejection a topic of considerable debate within the transplant The traditional approach for treatment of acute rejec- community. Clearly, the costs associated with life-long tion in organ transplantation is high-dose corti- maintenance therapy can be daunting for patients and costeroid followed by anti-lymphocyte therapy for payers alike, with some grafts clearly lost as a conse- “steroid-resistant” rejection. In renal transplantation quence of financial challenges. AZA and prednisone this approach is falling rapidly out of favor following were the first agents to transition to generic status, the introduction of the Banff schema using histologic with little or no apparent negative impact. The emer- findings to define therapeutic approaches to rejection. gence of generic CyA preparations in the late 1990s Acute cellular rejection of the milder variety (Banff 1A was complicated by confusion, uncertainty regard- or B) is generally treated with intravenous methylpred- ing financial impact, and an extremely slow transi- nisolone (10 mg/kg/day for 3 days) up to a total of 1– tion away from branded preparations despite several 3 g. More severe acute cellular rejection or refractory studies indicating no adverse impact of the switch. In acutecellularrejection(incompleteclinicalresponse the past 2–3 years, generic formulations of MMF and to IV corticosteroids) generally becomes an indica- TAC have been introduced with bioequivalence data tion for use of antibody therapy, with rATG (dosed but no published experience in transplant recipients. similarly to induction protocols) as the most com- Criteria for approval of generics in the United States monly used agent. Strategies for treating recurrent or are single-dose trials in fed and fasted control subjects persistent rejection include drug changes (converting demonstrating bioequivalence in terms of absorption AZA to MMF and CyA to TAC; increasing baseline of active ingredient with major PK measures, with steroid dose), photopheresis, and total lymphoid a 90% confidence interval within 0.8 and 1.25 ratio irradiation. with reference product. Despite aggressive introduc- The relatively recent appreciation of a differ- tion of these agents into the community and some ent phenotype of acute rejection (antibody-mediated confusion on the part of patients, there appears to rejection, AMR) has resulted in a different approach to be little negative impact on patient outcomes to this therapy for patients with this diagnosis. Until AMR is point. fully characterized and understood, the protocols for In 2001, The American Society of Transplanta- treatment remain variable and not evidenced-based. tion invited experts to review generic immunosup- Treatmentistargetedatremovingantibodies(usually pressants. Alloway and colleagues published a sum- anti-HLA) from circulation, inhibiting antibody bind- mary statement (American Journal of Transplantation, ing to cells, and/or manipulating antibody production. 2003) that underscored the welfare of the individ- Early AMR within the first 3–6 months after transplan- ual patient as the preeminent concern and supported tation is usually treated with a combination of plasma- the availability of efficacious less expensive immuno- pheresis (with therapeutic plasma exchange, TPEX) suppressive medications as helping improve patient followed by low-dose intravenous immunoglobulin adherence. They further endorsed the FDA process (IVIg) at 100 mg/kg for a total of 3–5 cycles. Some as appropriately rigorous, with FDA-approved generic centers use a single dose of rituximab at 1000 mg or narrow therapeutic index immunosuppressive agents 375 mg/m2 at the end of treatment cycles to reduce fur- appearing to provide adequate immunosuppression to ther B-cell maturation. Others utilize very high-dose low-risk transplant patients. However, they cautioned IVIg (2 g/kg, with a second dose 1 month later). As that the standards may not be adequate for higher time elapses following transplantation, the phenotype risk patients, and approval of new generic preparations

26 Chapter 3: Immunosuppression should include studies in at-risk patient populations. now being devoted to antibody-mediated injury and The International Society for Heart and Lung Trans- lymphoid tissue of B-cell lineage (Figures 3.1 and 3.2). plantation released a Consensus Statement on generic immunosuppression in 2009. The recommendations Small molecules cautioned against use of generics in patients with crit- ical drug dosing requirements and that regulating Agents now under development in this category follow agencies should approve only those generics studied in the tradition of most current maintenance immuno- with concomitant interacting medications or in trans- suppression: oral agents that will be ingested on a daily plant recipients. In addition, the transplant recipient basis in combination with other agents. They differ and center should have greater vigilance to adverse from current therapeutic approaches by targeting dif- sequelae and closer therapeutic drug monitoring fol- ferent pathways, with hopes of better efficacy and fewer lowing conversion. side effects. Tafocitinib (CP-690550) Immunosuppression: future With the discovery of the importance of Janus kinases As implied in the previous section, despite the (Jak) in mediating cytokine-driven lymphocyte acti- remarkable advances of the last half-century, cur- vation, and that congenital deficiency of Jak3 results rent immunosuppression remains a less than ideal in severe combined immunodeficiency in children, approach to support solid organ transplantation. Each partial inhibition of this pathway became an attrac- agent is associated with its own set of toxicities in tive target for immunosuppression. Given widespread addition to the shared adverse consequences of long- expression of Janus kinases in other tissues, especially term immunosuppression: infection and malignancy. hematopoietic, specificity for Jak3 was essential, as well As also noted above, drug minimization to avoid as the ability to inhibit lymphocyte activation in a these adverse effects now appears directly associated dose-dependent fashion. Both of these prerequisites with immunological graft failure as we increase our havebeenmetbythePfizercompoundtafocitinib. understanding of previously unrecognized immune In rheumatoid arthritis, doses ranging from 5–30 mg responses. It seems unlikely that additional permuta- twice daily are effective in improving symptomatol- tions of currently available therapies will be able to ogy, with relatively few adverse effects. A pilot study alter this dynamic. Although the promise of manipu- in renal transplant recipients comparing two dosage lating immune response at the time of transplantation regimens of tafocitinib in a CNI-free protocol that to induce long-term graft acceptance (tolerance induc- included basiliximab, MMF, and steroids documented tion) remains an ultimate goal, techniques to apply comparable rates of acute rejection after 6 months to these approaches on a widespread basis still seem far a TAC-based control group. However, the higher-dose in the future. tafocitinib patients (30 mg twice daily) appeared over- In the near future, therefore, further improvement immunosuppressed, with complications including BK in outcomes for transplant recipients is likely to reflect nephropathy and cytomegalovirus disease. There also progress along two fronts. First is the increasing abil- appeared to be additive hematopoietic toxicity in com- ity of new approaches to define subgroups of transplant bination with MMF and a trend to higher plasma lipid recipients at different immunological risk using novel levels. Additional studies are underway to better define biomarkers and microarrays. Although such immuno- proper dosing and the most beneficial drug combina- logical monitoring has been attempted throughout the tions for use with this novel agent, with an initial phase history of the field, these new technologies seem on 3 trial being organized at the time of writing. the verge of moving from the laboratory to the clinic and providing a mechanism to define who does and Sotrastaurin (AEB071) does not need potent immunosuppression. The sec- Protein kinase C (PKC) occupies a central point ond approach likely to contribute to better outcomes in mechanisms of T-cell activation via both the is development of novel immunosuppressant pharma- T-cell receptor and costimulation pathways; PKC cotherapies. Discovery of new agents is informed by activity is crucial for production of inflammatory our evolving understanding of how immunological cytokines, including interleukin-2 (IL-2) and inter- processes injure allografts, with substantial attention feron ␥. Sotrastaurin, by inhibiting PKC, blocks T-cell

27 Section 1: General

Figure 3.1 Novel biologics and small molecules targeting cell surface receptors and intracellular pathways of the T cell. PKC: Antigen Presenting Cell protein kinase C; CN: calcineurin. Reproduced from Vincenti F and Krik AD, What’s next in the pipeline, American Journal of ICAM MHC Transplantation, 2008, with permission from Efalizumab CD40 CD05 John Wiley & Sons Inc. LFA3 CDEC CD4 Anti-CD40 CO114

CD154 CTLA4 TCR CD2 CD28 LFA1

AEB071 PKC CN TNF IL2 Transcription

T Cell

Figure 3.2 Novel monoclonal antibodies and fusion receptor proteins that deplete and/or block activation of B cells. Reproduced from Vincenti F and Krik Belimumab AD, What’s next in the pipeline, American Atacicept APRIL C Ataclcept D1 BAFF Journal of Transplantation, 2008, with 3 (BLyS) permission from John Wiley & Sons Inc. T Cell 4 BR3FC

C Anti-CD40 D4 0 BCMA BR3 Epratuzumab TAC i

CD22 BCR

syK NIK CD19 IKK Ink pl3K Ras p3 8 MHCII AME-133 PKCII

Ocrelizumab CD Activation/Survival 2 0

Ofatumumab B Cell

activation in a CNI-independent fashion and in pri- sotrastaurin with MPS and corticosteroids as an alter- mate models has been shown to prolong kidney and native to TAC-based therapy were disappointing due heart allograft survival. Results from two large phase to excess acute rejection. However, patients receiving 2 trials in de novo kidney allograft recipients using a TAC/sotrastaurin-based regimen for 90 days prior to

28 Chapter 3: Immunosuppression substitution of MPS for TAC experienced little acute Belatacept rejectionandarelativedearthofadverseevents.A Belatacept is a novel biologic fusion protein com- third phase 2 trial with sotrastaurin in combination posed of the non-binding portion of IgG fused to with everolimus is underway, with more trials planned the extracellular domain of cytotoxic T-lymphocyte using sotrastaurin with a variety of agents in kidney antigen 4 (CTLA-4). As mentioned earlier, CTLA-4 and liver transplantation. has a high affinity for CD80/86 and therefore prevents the CD80/86-CD28 interaction required for costimu- Bortezomib lation, raising the threshold for activation of T cells, effectively downregulating the immune response. The Bortezomib is a reversible inhibitor of the 26S protea- agent is administered intravenously at intervals rang- some in mammalian cells. This proteasome degrades ing from 1–4 weeks in clinical trials. Two large phase ubiquitinated proteins, thus playing an essential role 3 trials of belatacept-based maintenance therapy have in maintaining cell homeostasis. Inhibition of this pro- recently been completed. In both studies conducted in teasome prevents targeted proteolysis, leading to cell differing patient populations all thought to be at rela- death, especially in rapidly dividing cells. It is admin- tively low immunological risk, maintenance therapy istered intravenously and is approved by the FDA for with two different regimens of belatacept dosing was treatment of multiple myeloma, shown to be effec- compared with CyA in regimens that also included tive in decreasing plasma cell burden in some refrac- basiliximab induction, along with MMF and corticos- tory cases. Its use in kidney transplant recipients has teroids. Despite a higher incidence of acute rejection been anecdotal, primarily as therapy for refractory with belatacept, the newer agent seemed associated antibody-mediated rejection. with equivalent graft and patient survival at 1 and 2 years and better allograft function. However, although the overall rates of infection and malignancy Biologics were similar among treatment groups, PTLD occurred It is now common to use biologics, such as poly- more frequently among belatacept-treated subjects, clonal or monoclonal antibodies, for a short time primarily those who were naive for Epstein-Barr virus as induction or treatment of acute rejection. Under before transplantation. Data after 1 year indicate a development are several monoclonal antibodies, usu- possible beneficial impact of belatacept on anti-donor ally humanized, that target specific cell surface pro- antibody formation. teins to alter immunological responses for long-term Patients in the phase 2 belatacept trial were offered administration. This is analogous to increasing use of enrollment in an extension, allowing follow-up in a agents such as etanercept and adalimumab in autoim- small cohort of patients beyond 5 years after trans- mune diseases; in fact, some of the agents studied in plant. There did not appear to be an increasing inci- transplantation are already marketed for these indi- dence of rejection, graft failure, or malignancy in these cations. Efalizumab is a humanized monoclonal anti- patients, with a stable GFR over time. Additional small body directed against lymphocyte-function associated pilot studies have been performed using belatacept in antigen-1 (LFA-1), an adhesion molecule expressed combination with either TAC or SRL, with favorable on multiple cell populations including T cells, B cells, short-term outcomes. FDA approval of belatacept for and NK cells. LFA-1 is important in facilitating cell routine usage in the United States is anticipated in migration and homing. It had been used for psori- 2010. asis and showed promise in clinical trials for solid organ transplant recipients but was withdrawn from the market when a small number of psoriasis patients Alefacept developed progressive multifocal leukoencephalopa- Alefacept is a dimeric, humanized fusion protein that thy (PML), a fatal central nervous system infection targets the extracellular CD2 receptors expressed on caused by JC polyomavirus, after 3 years on the T lymphocytes and inhibits the CD2/LFA-3 interac- drug. There are no current trials using efalizumab, tion, thereby inhibiting lymphocyte activation. T cells and existing patients have been transitioned to other of the memory effector phenotype may be more sensi- immunosuppression. tive to the effect than other T-lymphocyte populations.

29 Section 1: General

Treatment with alefacept reduces the circulating total data indicate that eculizumab is highly effective in CD4 and CD8 counts. As CD2 is also expressed on preserving allograft function in highly sensitized kid- the surface of other lymphoid tissue (such as NK cells ney allograft recipients, at least for intermediate-term and some B cells), there may be additional effects as periods. well. Alefacept, available for subcutaneous injection, is already FDA-approved as treatment for chronic plaque psoriasis, with a known safety profile as monother- Further reading apy.Aglobalphase2trialtodefinetheproperdosing Brennan DC, Daller JA, Lake KD, Cibrisk D, Del Castillo D. Rabbit antithymocyte globulin versus basiliximab in and explore combination therapy in kidney transplan- renal transplantation. NEnglJMed2006; 355: 1967– tation has recently completed enrollment. 77. Eculizumab Ekberg H, Tedesco-Silva H, Demirbas A, et al. Reduced exposure to calcineurin inhibitors in renal Eculizumab is a recombinant humanized monoclonal transplantation. NEnglJMed2007; 357: 2562–75. IgG that specifically binds, with high affinity, tothe Kirk AD. Induction immunosuppression. Transplantation complement component, C5, and prevents develop- 2006; 82: 593–602. ment of the terminal complement complex, C5b-9. Kuypers DR, Le Meur Y, Cantarovich M, et al.Consensus It thus blocks complement-mediated cell lysis and is report on therapeutic drug monitoring of mycophenolic FDA approved for treatment of paroxysmal nocturnal acid in solid organ transplantation. Clin J Am Soc hemoglobinuria (intravascular hemolysis). It is also Nephrol 2010; 5: 341–58. an intravenous preparation, administered at intervals Scientific Registry of Transplant Recipients. 2008 Annual of 1–4 weeks. Anecdotal use in transplantation is to Data Report. Available at: www.ustransplant.org prevent (in patients at high risk) or treat antibody- (accessed on January 14, 2011). mediated rejection. By blocking the terminal effec- Vincenti F, Kirk AD. What’s next in the pipeline. Am J tor component of the injury response, preliminary Transplant 2008; 8: 1972–81.

30 Section 1 General Chapter Major complications – cancer

4A Bimalangshu R. Dey and Thomas R. Spitzer

Key points Recurrence of pre-existing malignancy is an important consideration when selecting transplant r Malignancy may occur due to recipients. Studies in renal and cardiothoracic organ transplantation in patients with pre-existing transplant patients with pre-existing malignancies malignancy, transmission from donor to suggests a recurrence rate of approximately 20%, with recipient,anddenovointherecipient the majority occurring in patients who remain disease following transplantation. Transplant free for more than 5 years. Tumor recurrence is more outcomes are adversely affected by recipient common following previous cancer of lung, skin, malignancy. r bladder, pancreas, and lymphoma. The length of the Incidence of de novo malignancy increases surveillance period until true remission is achieved cumulatively following transplantation, with has to be balanced against the known outcomes for 20% after 10 years and nearly 30% after individual cancers, their grade, and the degree of 20 years, and is three to four times more organ failure. common in transplant recipients compared Transmission of donor malignancy to a recipi- with age-matched controls. r ent has been described in the Cincinnati Transplant Cutaneous and lip malignancies are the most Tumor Registry (CTTR) and United Network for common malignancies, comprising Organ Sharing (UNOS) Registries, often with sub- approximately 36% of all post-transplant sequent poor outcome. Close consideration should neoplasms. r be given to unexpected donor malignancy during Post-transplant lymphoproliferative disorder work-up from a detailed history and during organ is Epstein-Barr virus driven, with an procurement. Non-melanoma skin cancers, primary incidence of 1–5% after renal transplant and cerebral malignancies (except astrocytomas, glioblas- 5–15% after lung, small bowel, and tomas, and medulloblastomas) and those who remain multi-organ transplants. r cancer-free more than 5 years following treatment are Careful surveillance for occurrence of generallyconsideredtohaveanacceptablerisk. malignancy, investigation of symptoms, and Recipients are at high risk of developing de novo early treatment are important in cancers, both epithelial and lymphohematopoietic post-transplant care. malignancies, due to complex interactions of several factors, among which oncogenic viruses and impaired Malignancy in recipients of solid organ transplants immunity as a result of chronic immunosuppression (SOT) occurs due to transplantation in patients with are the major causes. The prevalence of post-transplant pre-existing malignancies, transmission from donor malignancies (PTM) ranges from 4–18%. The inci- to recipient, and de novo following transplantation. dence of malignancy increases with the length of The outcomes for recipients who develop malignancy follow-up after transplantation, with a cumulative can- following transplantation are significantly worse, and cer incidence of 20% after 10 years and nearly 30% after therefore close consideration must be given to donor 20 years. Overall, the incidence of cancer is increased and recipient selection. three- to four-fold in transplant recipients compared

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

31 Section 1: General

Table 4A.1 Incidence of neoplasms in general population an Australian study showed a linear increase in skin versus Cincinnati Transplant Tumor Registry patients cancers with the length of time after transplantation, General Transplant approaching 70% at 20 years. The increase is mostly Neoplasm population recipients seen with squamous cell carcinoma (SCC). The inci- Lymphoma 5% 24% dence of SCC in transplant recipients is 40 to 250 times higher than in the general population, basal cell can- Carcinoma of the lip 0.2% 6.7% cer (BCC) 10 times higher, and malignant melanoma Kaposi’s sarcoma 0.02% 5.6% 5 times higher. Precancerous keratoses, Bowen’s dis- Carcinoma of vulva/perineum 0.5% 3.5% ease, and keratoacanthomas are also seen and can Carcinoma of the kidney 2.3% 4.8% be treated with topical 5-fluorouracil or tretinoin. Hepatobiliary carcinoma 1.5% 2.4% Melanomas comprised 5.4% of skin cancers in the Reproduced from Ginns LC, Cosimi AB, and Morris PJ, Immuno- CTTR in contrast to the incidence of 2.7% in general suppression in Transplantation, 1999, p. 771, with permission from population. BCC outnumbers SCC in the general pop- John Wiley & Sons Inc. ulation by 5 to 1, but in transplant recipients SCC out- numbers BCC by 1.8 to 1.0. The characteristics of SCC with age-matched controls in the general population. differ in many ways from those seen in the general Cancer has become the second cause of late mor- population. In the general population, skin cancers tality following transplantation after late graft fail- cause only 1–2% of all cancer deaths, the great majority ure. Fast progression, unfavorable prognosis, and poor from melanoma. SCC in transplant recipients is much response to treatment are the general characteristics more aggressive, accounting for the majority of lymph of post-transplant tumors. Once a diagnosis of PTM node metastases, and approximately 5% of patients die is made, staging evaluation should be done specific to fromtheirSCC.Allograftrecipientswithskincancers the diagnosis. aremorelikelytohaveother,moreaggressivetypes This chapter is based on the study of data collected of neoplasia than transplant recipients without skin by CTTR and the available literature published by both cancer. Specialist dermatology input is required, but North American and European organ transplant cen- treatment options include reduction in immuno- ters.Wehaveattemptedtoreviewthecharacteristics suppression (especially stopping anti-metabolites), of the most important de novo malignancies in organ cryotherapy, and surgical excision. allograft recipients, although it is outside the scope of this chapter to discuss diagnosis and treatment of less common malignancies. Kaposi’s sarcoma The incidence of Kaposi’s sarcoma (KS), which isa Types of de novo malignancies human herpes virus type 8 (HHV-8)–related cancer, in allograft recipients is markedly increased, comprising The most frequent cancers in transplant recipients are approximately 5.6% of all neoplasms, in contrast to its skin and lip cancer, solid organ malignancies, and incidence of 0.02–0.07% of all cancers in the general post-transplant lymphoproliferative disorder (PTLD). population.TheriskofdevelopingKSintransplant Table 4A.1 illustrates the incidence of certain cancers recipients is more than 100-fold compared with age- in SOT recipients in contrast to that in the general pop- matchedcontrolsinthegeneralpopulation,withthe ulation (Table 4A.1). risk being much higher in allograft recipients who are Arab, Italian, Jewish, black, or Greek, likely reflecting Skin and lip cancers the greater prevalence of HHV-8 infection. Reddish- Cutaneous and lip malignancies are the most com- blue macules or plaques in the skin or oropharyngeal mon neoplasms, comprising approximately 36% of all mucosa in allograft recipients, particularly in those at PTMs, appearing on sun-exposed areas, particularly risk,mustbeinvestigatedforKS.Onestudyshowed of the head and neck and upper limbs, especially in that KS was the most common cancer in renal recipi- light-skinned individuals with blue eyes and blonde ents in Saudi Arabia, comprising up to 76% of all can- or red hair. The incidence of non-melanoma skin can- cers. In the CTTR, a little over 50% had non-visceral cers in renal recipients, according to a Dutch study, was disease confined to the skin, conjunctiva, or orophar- 10% at 10 years, rising to 40% after 20 years. Similarly, ynx. The rest had visceral KS mainly involving the

32 Chapter 4A: Major complications – cancer gastrointestinal tract, lungs, or lymph nodes, but other than one third of PTLD cases develop within 2 years organs may also be affected as well. The relatively after transplantation, and approximately 20% develop short period of latency between transplantation and 10 or more years following transplantation. Available the diagnosis of KS suggest that KS development in histologic analysis of tumor tissue in the Ann Arbor transplant recipients is associated with rapid reactiva- study showed that 62% of PTLD cases were EBV pos- tion of latent HHV-8 infection. itive, and EBV-positive tumors occurred sooner after According to the CTTR, 54% of patients with non- transplantation than EBV-negative tumors (mean, visceral KS achieved complete remission (CR) follow- 29 months versus 66 months). ing treatment, including 32% of these CRs occurring EBVisahumanherpesvirusthatinfectsmorethan following reduction in immunosuppressive therapy. 90% of adults. EBV is highly immunogenic, and dur- Patients with visceral KS fared worse, with only 27% ing primary infection, a vigorous cellular and humoral achieving CR following therapy. Surprisingly, 63% of immune response occurs, with the cellular component CRs in visceral KS resulted from a decrease or discon- consisting of both CD4 and CD8 T cells, which con- tinuation of immunosuppressive therapy only. Fifty- trol both primary infection and the periodic reacti- six percent of patients with visceral KS died. Inter- vation that occurs in all EBV-seropositive individuals. estingly, 16 of 39 renal allograft recipients in whom After clearance of primary infection, EBV persists in immunosuppression was reduced or stopped had long- B cells, establishing latent infection. Immunosuppres- term stable renal function, 21 lost their allografts to sive agents in allograft recipients disrupt the balance rejection, and 2 had impaired renal function. between latently infected B-cell proliferation and the EBV-specific T-cell response, allowing the increased population of latently infected B cells to develop into Lymphoproliferative disorders PTLD. Several lines of evidence support a causative PTLD, mostly driven by Epstein-Barr virus (EBV), role of EBV in the pathogenesis of PTLD. First, EBV- is a potentially fatal complication of transplantation naive patients who experience primary infection after and spans the spectrum from early lesions, including transplantation are at the highest risk for PTLD. Sec- infectious mononucleosis-like PTLD, to frank mono- ond, both EBV genomes and gene products are found clonal malignant lymphoma. In the CTTR, of a total of frequently in the tumor cells of PTLD. Third, EBV 1560 lymphomas, only 42 (2.7%) were Hodgkin’s lym- canexpresstypeIIIlatencyinBcells,whichisthe phoma, compared with a 10% incidence of lymphomas most promiscuous, antigenic, and growth-promoting in the general population; multiple myeloma and plas- stimulator of B cells. Based on the data on histologi- macytoma constituted only 58 cases (3.7%) in contrast cal detection of EBV in tumor cells in the Ann Arbor to a 19% incidence of all lymphomas in the general study, 66% of patients with diffuse large B-cell lym- population. The great majority of post-transplant lym- phoma and 60% of patients with polymorphic PTLD phomas were non-Hodgkin’s lymphoma (NHL), mak- were EBV positive. EBV positivity was detected in ing up 93.6% of the cases, as compared with 71% in the 100% of patients with Hodgkin’s lymphoma and 75% general population. of patients with multiple myeloma/plasmacytoma, In transplant recipients, PTLD is predominantly whereas 75% of patients with Burkitt’s lymphoma were of recipient origin. Multiple single-center reviews sug- EBV negative. gest two important risk factors for the development of In transplant recipients, PTLD is predominantly PTLD: the degree of immunosuppression and negative of recipient origin. In the CTTR, 86.3% of PTLD arose EBV recipient serology, i.e., development of primary from B lymphocytes and 13.3% were of T-cell origin. infection after transplantation. A recent study from Histologically, PTLD includes a heterogeneous group Ann Arbor, Michigan, showed that transplant recipi- of disorders ranging from reactive, polyclonal hyper- ents with lymphoma, compared with all other patients plasia to aggressive NHL. The World Health Organi- who received allografts, were more likely to have neg- zation classifies PTLD into four categories: (1) early ative pretransplant EBV serology (48% versus 7%) and lesions such as plasmacytic hyperplasia and infectious to have received an organ from an EBV-positive donor. mononucleosis-like lesion; (2) polymorphic PTLD, The incidence of PTLD is lowest after renal trans- whichiscomposedofamixtureofsmall-andmedium- plantation (1–5%) and highest after lung, small bowel, sized lymphocytes, immunoblasts, and plasma and multi-organ transplants (5–15%). Overall, more cells; (3) monomorphic PTLD, which has sufficient

33 Section 1: General

architectural and cytopathic atypia to be diagnosed tion assays was highly predictive of EBV-PTLD. Sub- as frank lymphoma; and (4) classic Hodgkin’s sequent studies showed that only 50% of patients lymphoma-like PTLD. A great majority of all types are with elevated EBV-DNA developed PTLD. Up to 50% associated with EBV. Despite this and previous efforts of transplant patients may have an elevated EBV- to standardize the pathological classification of PTLD, DNA after transplantation, but only a small subset neither histology nor clonality consistently predicts will develop PTLD. Serial monitoring of EBV-DNA is outcome. PTLD may occasionally present as more important to distinguish patients with stable-elevated diffuse disease that is more difficult to diagnose and EBV-DNA from the patients with increasing EBV- may be mistaken for a fulminant sepsis syndrome. DNA. A rising EBV-DNA load is more predictive Although extralymphatic disease occurs in 24–48% of PTLD than stable-elevated EBV-DNA load. Com- ofNHLpatientsinthegeneralpopulation,extran- bined monitoring of EBV-DNA and EBV-specific CTL odal disease is present in 69–79% of NHLs in allograft responses has been found to be more predictive of sub- recipients. Central nervous system (CNS) is one of the sequent development of PTLD. most common extranodal sites, being involved in 21% of lymphoma patients in the CTTR. Of these cases, Treatment of PTLD 9% were patients with systemic disease involving both Reduction of immunosuppression CNS and other organs, whereas 12% were confined There is increasing evidence that reducing immuno- to the CNS only. This 12% CNS-only PTLD contrasts suppression may decrease the development of PTLD in markedly with the general population, in which there high-risk patients, i.e., patients with a rising EBV-DNA is only a 1–2% incidence of primary CNS involve- load. With overt PTLD, reducing immunosuppres- ment. Macroscopic or microscopic allograft involve- sion has yielded variable results. Single-center studies ment can occur in 23% of NHLs in SOT recipients. report overall response rates of up to 75% in patients According to the Ann Arbor study, frequent involve- treated with this modality alone or in combination ment of extranodal sites (79%), poor performance sta- with surgery, and one study showed complete remis- tus (68%), elevated lactate dehydrogenase (LDH; 71%), sions in 20% of patients. If EBV is actively promot- and advanced-stage disease (68%) are all common at ing B-cell proliferation at the time of PTLD diagno- the time of diagnosis of PTLD. sis, as is believed to be the case in PTLD that occurs early after transplantation, reducing immunosuppres- Diagnosis of PTLD sion appears to have a better therapeutic influence than Patients with PTLD may often be asymptomatic, in late PTLD. whereas others may present with fever, night sweats, weight loss, lymphadenopathy, and so forth. Occasion- EBV-specific cytotoxic T cells ally, PTLD may present as a diffuse disease involv- Donor-derived EBV-specific cytotoxic T cells (CTLs) ing multiple organs and running a fulminant clinical given to the high-risk patients after HSCT have been course. As the gastrointestinal tract is often involved found to be highly effective as prophylaxis, and they by PTLD, any signs suggesting perforation, peritonitis, have also proved to be effective as treatment for or intestinal obstruction should be pursued seriously PTLD in more than 80% of patients. In solid allograft for a conclusive diagnosis. The possibility of CNS lym- recipients, there are additional challenges because phoma must be suspected when a transplant recipient patients remain on long-term immunosuppression. develops neurological symptoms, including a change However, studies have demonstrated that it is possi- in mental status. ble to generate autologous EBV-specific CTLs from As PTLD may evolve progressively from a poly- transplant recipients, which restores EBV-specific clonal entity to a more aggressive monoclonal vari- immunity, albeit short-term, that controls disease ant, early diagnosis is important. Therefore, there has progression. Long-term persistence of EBV-specific been interest in designing predictive assays to iden- immunity has not been feasible in patients who remain tify patients who are at high risk for developing PTLD. on immunosuppression. Initial studies in recipients of T-cell–depleted (TCD) hematopoietic stem cell transplantation (HSCT) indi- CD20 monoclonal antibody (rituximab) cated that an elevated serum EBV-DNA measured Because EBV-driven neoplastic B cells routinely by quantitative polymerase chain reaction amplifica- express CD20 antigen, rituximab, a humanized

34 Chapter 4A: Major complications – cancer

Treatment of B-Cell PTLD Figure 4A.1 An algorithm for the treatment of B-cell post-transplant lymphoproliferative disorder (PTLD). BMT: bone marrow Aggressive Indolent transplantation; CR: complete remission; CTX: Polymorphic Monomorphic chemotherapy; Int: Intermediate; IPI: low IPI Int to high IPI International Prognostic Index; SCT: stem cell Post-BMT 1 2 transplantation; Tx: treatment. Antiviral Rx Rituximab, Antiviral Rx Reduction in Immunosuppression Reduction in Immunosuppression,

< CR Progression Follow CR CR Follow Recurrence

3 4 Systemic CTX +/- Rituximab Autologous SCT Experimental Tx Follow CR

anti-CD20 monoclonal antibody, has become an vage therapy has been associated with an overall attractive approach to eliminate EBV-transformed response rate of up to 75% in patients with NHL-type malignant cells. Rituximab has significant activity PTLD. Similarly, patients with Hodgkin’s lymphoma- against PTLD in transplant recipients, with response like PTLD can be treated with doxorubicin, bleomycin, rates of 44–100% in several small series. A recent vinblastine, and dacarbazine (ABVD) chemotherapy. phase 2 clinical trial that included patients who had Some centers are reporting high response rates using reduction of immunosuppression as their only previ- chemotherapy as first-line treatment in addition to ous therapy, but excluded patients with CNS PTLD, reduction in immunosuppression. reported a response rate of 44% at day 80. However, a long-term follow-up study of transplant recipients treated with rituximab demonstrated that 57% had Hepatocellular carcinoma progressive disease at 12 months, and another study In the CTTR, hepatocellular carcinoma (HCC), which showed recurrence of PTLD in 50% of patients. represents the great majority of hepatobiliary tumors in transplant recipients, constitutes 2.4% of all PTMs, Surgery and radiotherapy compared with an incidence of 1.5% in the general population. Two epidemiological studies have shown If the staging evaluation confirms localized disease a 20- to 40-fold increased incidence of HCC in trans- confined to one site, then radiation and/or surgery can plant recipients compared with age-matched controls. be an effective approach. In one study on the outcome Hoffmann et al.fromtheJohnsHopkinsUniver- of PTLD occurring late after transplantation, patients sity School of Medicine demonstrated an incidence with localized disease achieved sustained remissions of HCC of 6.5 per 100 000 person-years among kid- following surgery or radiotherapy and reduction of ney, heart, and lung (non-liver) recipients and 25 immunosuppression. per 100 000 person-years among liver recipients. HCC incidence among non-liver recipients was indepen- Chemotherapy dently associated with hepatitis B virus (HBV) sur- PTLD is generally chemotherapy-sensitive after pro- face antigenemia, hepatitis C virus (HCV) infection, gression or failure to respond to rituximab and reduc- and diabetes mellitus. Among liver recipients, HCC tion of immunosuppression, and cyclophosphamide, incidence was associated with advancing age, male doxorubicin, vincristine, and prednisone (CHOP) sal- sex, HCV infection, and diabetes mellitus. In the

35 Section 1: General

Johns Hopkins study, although HCC incidence in Carcinoma of the vulva and perineum liver transplant recipients was overall elevated, the The incidence of carcinomas of vulva/perineum in overall incidence of HCC among non-liver recipi- allograft recipients is 3.5% in the CTTR compared with ents was similar to that of the general population, its 0.5% incidence in the general population. In gen- but increased among those with HCV and HBV eral, males outnumber females by more than 2:1 in antigenemia. regard to the distribution of all PTM. In contrast, with carcinomas of the vulva, perineum, or anus, females outnumber males by 2.5:1. Approximately 50% of Renal cell carcinoma patients have a history of preceding condyloma acumi- Carcinoma of the kidney in the CTTR comprised 4.8% natum, which should be regarded as a premalignant of all PTMs in comparison with a 2.3% incidence in lesion. Many patients can be successfully treated with the general population. Renal cell carcinoma (RCC) local or extensive surgery, but 12% of patients in the represents approximately 80% of all renal tumors, CTTR died from their malignancy. whereas 12% are represented by transitional cell car- cinoma. The incidence of RCC in transplant recipi- ents is increased by 30- to 40-fold over its incidence in Other PTMs the general population. Most malignancies in kidney The Australia and New Zealand Dialysis and Trans- transplant recipients arose in their own diseased kid- plant Registry noted a 289-fold increased incidence neys, whereas 9% were found in the kidney allografts; of endocrine cancers, a 5.6-fold increase in leukemia, 24% of RCCs were discovered at the time of native a 2.5-fold increase in cancers of the digestive tracts, nephrectomy for hypertension or other reasons. One and a 2-fold increase in pulmonary tumors. Recently, predisposing cause of cancers in renal allograft recip- a study from Spain showed an increased risk of acute ients is acquired cystic disease (ACD) of the native myeloid leukemia in liver allograft recipients. Malig- kidneys, which occurs in 30–95% of patients receiving nancies have been reported from most organ sys- hemodialysis and which is known to be complicated by tems in addition to those discussed already. Sarcomas, RCC. breast carcinoma, bladder, and bowel cancers are par- ticularly seen after transplantation. Tobacco- and alcohol-related cancers Severalserieshavenotshownanappreciablyhigher PTMs in children incidence of lung and head and neck malignancies Post-transplant malignancies in children are differ- in allograft recipients as compared with the general ent from those in the general childhood population population. However, recently, Medina et al.from and from those observed in adult transplant recipi- SpainexaminedPTMsinliverallograftrecipients ents. In a recent study from France in which 1326 chil- anddemonstratedanincreasedprevalenceoflung, dren underwent SOT, 80 patients (6%) were diagnosed oropharyngeal, laryngeal, and esophageal cancers. with PTMs; EBV-related PTLD was the most common, This may be explained by the fact that a good propor- comprising 80% of all tumors, with 52% of cases occur- tion of their patients underwent liver transplants due ring within the first 6 months after transplant. PTLD to alcoholic liver disease and they also had significant was associated with primary EBV infection in 81% of exposure to tobacco. Lung cancer is the most common cases and EBV reactivation in the remaining 19% of non-cutaneous malignancy after heart transplantation patients. The prevalence of PTLD in SOT children in and often occurs in patients with a significant tobacco this study was 5%, but it was disproportionately higher history. Although the success of surgical treatment (18%) among combined liver and small bowel recipi- for early-stage lung cancer is good, most present with ents. EBV-negative lymphomas were rare and occurred metastatic disease, despite frequent surveillance post- more evenly throughout the 10-year period after trans- transplant, with poor survival despite treatment. There plantation. Skin cancer is the second most common have also been reports of tobacco-related lung can- malignancy after PTLD, and melanomas comprised cers following transplantation of lungs from known 16% of all skin cancers in children compared with 5% smokers. among adults.

36 Chapter 4A: Major complications – cancer

Management of PTM aggressive behavior than do similar cancers in the general population. In principle, reduction or cessa- Prevention and surveillance tion of immunosuppressive therapy, as clinically fea- sible, must be considered a central component of the In principle, the level of immunosuppression should therapeutic strategy for PTM. Cessation or reduction be kept as low as possible in order to maintain good of immunosuppression may lead to significant anti- allograft function. Whenever possible, viral infec- tumor responses, including sustained remissions of tion should be prevented in immunocompromised PTLD and KS, but such treatment by itself rarely patients. Because the incidence of HCC is increased causes regression of epithelial tumors. in transplant recipients, mainly in non-liver recipients, andasthisisfrequentlyassociatedwithHBVinfec- tion, vaccination of potential organ allograft recipi- ents with the HBV vaccine may prevent HCC. Because Conclusion human papillomavirus (HPV) infections, suspected Malignanciesofbothepithelialandlymphohe- of having a pathogenetic role in carcinomas of the matopoietic cell origin are common, often life- vulva, perineum, and cervix, are transmitted sexually, threatening complications of transplantation and patients should use barrier methods of contraception underscore the delicate balance between providing and undergo HPV vaccination where available. Pre- adequate immunosuppression to prevent allograft malignant lesions such as condyloma acuminatum or rejection and allowing sufficient restoration of cervical dysplasia should be treated adequately to pre- immune surveillance to prevent causative mecha- vent progression to invasive cancers. Light-skinned nisms of these PTM (such as oncogenic viruses). patients should avoid excessive sunlight exposure. In More targeted immunosuppressive strategies in the many institutions, antiviral drugs such as acyclovir or future will hopefully lessen the global immunosup- ganciclovir are used prophylactically to prevent viral pression that complicates conventional anti-rejection infections including EBV infection, but whether or approaches and diminish the risk of these malignan- not this prophylactic strategy will reduce the inci- cies. In the meantime, diligent preventive strategies, dence of EBV-related malignancies, including PTLD, early diagnosis (with, e.g., monitoring of EBV-DNA) remains debatable. Reduction in immunosuppression leading to successful preemptive treatment, and in response to a rising EBV viral load, which indicates aggressive treatment of established PTM will remain a high risk of developing PTLD, has reduced the inci- the cornerstones of management of this serious dence of PTLD in pediatric liver transplant patients. complication of transplantation. Understanding the increased risk of malignancy of transplant recipients, careful surveillance and screen- ing for selected malignancy should be undertaken. Further reading Careful examination is required at follow-up, includ- Heslop H. How do I treat EBV lymphoproliferation. Blood ing examination of skin, mucosa, and lymph nodes. 1009; 114: 4002–8. In some centers periodic screening investigations Hoffmann CJ, Subramanian AK, Cameron AM, et al. include fecal occult blood, flexible sigmoidoscopy, Incidence and risk factors for hepatocellular carcinoma colonoscopy, prostate-specific antigen, mammogra- after solid organ transplantation. Transplantation 2008; phy, chest X-ray, and cervical smear. Regular derma- 86: 784–90. tology review with biopsy of suspicious skin or lip Medina EM, Romero CJ, Camara AG, et al. Malignancy lesions should also be undertaken. Symptoms or sus- after liver transplantation: cumulative risk for picious findings on routine examination and testing development. Transplant Proc 2009; 41: 2447–9. require further investigation, imaging, biopsy, or spe- Navaro MD, Lopez-Andreu M, Rodriguez-Benot A, et al. cialist referral as necessary. Cancer incidence and survival in kidney transplant patients. Transplant Proc 2008; 40: 2936–40. Penn I. Neoplastic complications of organ transplantation. Treatment In: Ginns LC, Cosimi AB, Morris PJ (eds). PTM has a poor prognosis in general, and cancers Transplantation. Blackwell Science, Oxford, 1999, in allograft recipients frequently demonstrate a more pp. 770–86.

37 Section 1 General Chapter Major complications – pathology of chronic rejection 4B Yael B. Kushner and Robert B. Colvin

Key points attempt herein to convince the reader that none of these attributes are necessarily true. r Chronic rejection is a major cause of late allograft loss in solid organ transplants and is due to repeated or sustained alloimmune Definition graft injury. Chronic rejection is defined as the slowly evolv- r T-cell mediated and antibody-mediated ing injury and response of a solid organ graft rejection as well as innate immunity can all to persistent or recurrent alloimmune attack. Pro- lead to chronic rejection as a common gression of chronic rejection occurs over weeks to outcome. months to years and depends on two main fac- r The final pathway leading to late graft tors. The first is the intensity of the alloimmune loss likely includes non-immune vascular response and the second is the resistance of the pathology such as atherosclerosis, organ to the assault. All tissue structures within hypertension, and chronic drug organsarepotentialtargets;however,mostcom- toxicities. monly affected are the arteries (transplant arterio- r Treatment for chronic rejection is limited pathy or vasculopathy), capillaries (chronic humoral by a lack of agents that affect antibody rejection), and the epithelium. Organ-specific targets production, tissue repair, and include the glomeruli (transplant glomerulopathy), fibrosis. biliary tree (vanishing bile duct syndrome), and bron- r Originally thought to be irreversible and chioles (bronchiolitis obliterans). Chronic rejection untreatable, the lesions of chronic rejection is a major cause of late graft injury and mortality. can be responsive to certain regimens and Other causes of late graft injury, such as infection possibly show resolution over time if the and recurrent disease, can generally be distinguished offending insult is removed. histologically. r New techniques are needed to assess the activity of chronic rejection and the molecular pathways involved and to direct Transplant arteriopathy targeted therapy. Definition Transplant arteriopathy (TA) is the main lesion of Chronic rejection is a major problem and the most chronic rejection. TA describes the effects of the common cause of late graft injury in solid organ trans- immune system on arteries in solid organ trans- plantation. It causes late graft loss in approximately plants. It is variously termed chronic allograft vas- 25% of recipients of kidney, heart, and lung allografts culopathy, sclerosing transplant vasculopathy, acceler- and is a significant factor in liver allografts. Chronic ated atherosclerosis, chronic allograft arteriopathy,and rejection is widely regarded as difficult to diagnose, transplant arteriopathy.Transplantarteriopathy(TA) of obscure etiology, untreatable, and irreversible. We is our preferred designation, because the disease is

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

38 Chapter 4B: Major complications – pathology of chronic rejection

there is intimal fibrosis but a relative paucity of elas- tic fibers in the neointima, with occasional infiltrating mononuclear cells. In contrast, in native organs with atherosclerotic vascular disease, there is duplication of the internal elastica and few inflammatory cells (Fig- ures 4B.2A and 4B.2B). In TA, lipid deposition is usu- ally modest and consists of foamy macrophages abut- ting the internal elastic, whereas it is more substantial and diffuse in atherosclerosis. In addition, atheroscle- rosis typically affects vessels in an eccentric distribu- tion and tends to spare the small arteries, in contrast to TA, which is concentric and frequently affects small vessels. Although many authors have noted these dif- fering features, it has been shown that there is great synergy between immune and non-immune vascu- lar injury. For example, conditions such as hyperc- holesterolemia are also risk factors for TA, and pro- tocol biopsies have shown evolution from lesions that resemble TA to those resembling atherosclerosis over Figure 4B.1 Chronic, active TA can be identified by the time. Certain authors feel that transplant patients are inflammatory infiltrate that is typically under the endothelium and accompanied by intimal fibrosis. This artery from a renal allograft simply at risk for accelerated arteriosclerosis for vari- shows T cells within an arterial wall and focally lifting the ous reasons and that these lesions would all fall under endothelium of an affected vessel stained for CD3 by the guise of immune and non-immune mediated TA. immunohistochemistry. Antibody also may play a role in the pathogenesis of TA, as shown in experimental studies. See color section. Pathogenesis and experimental models restricted to arteries, and this nomenclature does not Intimal proliferation is thought to be due to increased implicate a specific mechanism. numbers of myofibroblasts or smooth muscle cells as well as increased deposition of extracellular matrix. These proliferating myofibroblasts stain with alpha- Pathological diagnosis smooth muscle actin. The myofibroblasts/smooth The stages of TA have been well documented histolog- muscle cells express endothelin-1, iNOS, ␤-actin (a ically. Typically, the first stage detectable is endarteri- marker of synthetic myocytes), and the anti-apoptotic tis, with inflammatory cells (primarily T cells) infiltrat- molecule Bcl-xL. The extracellular matrix material is ing and lifting off vascular endothelium (Figure 4B.1). composed of collagen, fibronectin, tenascin, biglycan, Endarteritis in a previous biopsy is a strong risk factor decorin, and acid mucopolysaccharides and is believed for later TA. to be in part produced by the myofibroblasts them- Following endarteritis, concentric intimal fibro- selves. Collagen IV and later I and III are deposited sis ensues, with eventual vascular luminal narrowing first against the internal elastica and then progressively leading to graft ischemia and failure. The lesions of toward the vessel lumen. Both the inflammatory cells TA are most prominent in larger caliber arteries but and about half of the spindle-shaped myofibroblasts extend to peripheral arterial branches, sparing only are of recipient origin. arterioles. Activity of the lesion is suggested by loose The immunological basis of TA is under active stroma, density of the lymphocytic infiltrate, stromal investigation. The belief that the process is immuno- proliferation, and the occasional presence of fibrin. logically mediated is based on the observations that Diagnostic difficulty is encountered when attempt- TA rarely arises in autografts. Three separate and pos- ing to distinguish chronic TA from atherosclerotic sibly synergistic pathways have been identified: T cells, lesions that develop in patients with organ transplants antibody-mediated injury, and natural killer (NK) due to hypertension or other risk factors. Certain fea- cells. Macrophages, although unlikely primary respon- tures are thought to be helpful in this regard. In TA, ders to donor antigen, are thought to be important

39 Section 1: General

(A) primarytargetcellinTAisprobablytheendothelium, although the medial smooth muscle cells may also be affected. T cells are sufficient immunological instigators of TA, as judged by animal studies, and provide the first etiologic pathway of TA. TA lesions readily arise in set- tings without antibody production or NK cell activity. This has been shown in B-cell deficient allograft recip- ients, murine male-to-female heart grafts, and other minor histoincompatible grafts. In vitro, T-cell super- natants from patients with chronic rejection enhance the proliferative effects of cultured smooth muscle cells via platelet-derived growth factor (PDGF). T cells in the lesions of human TA express a cytotoxic phenotype (perforin, GMP-17). In murine models lacking cyto- toxic molecules perforin or granzyme B, the lesions of TA are inhibited. Blocking the alternative cyto- toxic pathway with soluble Fas also inhibits TA in rats. Thus repeated or sustained T-cell mediated rejection, as evidenced in the kidney by inflammatory cells infil- (B) trating renal compartments (interstitial inflammation, tubulitis, endarteritis), is sufficient to cause chronic rejection. Antibodies, the second pathway of TA, are also suf- ficient to produce TA. Acute humoral or antibody- mediated rejection (AHR) is typically diagnosed in the presence of serum donor-specific antibodies (DSA), complement split-product complement component 4d (C4d) deposition in capillary walls as seen on immunofluorescence or immunohistochemistry, and histological signs of humoral rejection, such as inflam- matory cells in peritubular capillaries or fibrinoid necrosis of vessel walls. It is thought that repeated or sustained episodes of AHR can lead to chronic rejec- tion. Antibody-mediated rejection is independently linked to chronic rejection as well: DSA to major histocompatibility complex (MHC) antigens (class I and II) are associated with transplant glomerulopa- Figure 4B.2 Contrasting appearances of TA and thy(TG)inthekidneyandacceleratedprogression non-immunologic arteriosclerosis can be appreciated in sections of TA. In non-human primates, TA in renal allografts stained for elastic fibers. (A) An artery from a renal allograft with TA shows intimal fibrosis without an increase in elastic fibers and with is strongly associated with prior DSA and C4d depo- infiltrating inflammatory cells. (B) An artery from a native kidney sition in the graft. In vitro antibody to MHC class with arteriosclerosis shows duplication of the elastic lamina, termed I promotes endothelial proliferation via upregulation fibroelastosis, and few inflammatory cells. A and B, Weigert elastic stain. See color section. of FGF-1 receptors and RhoA and phosphorylation of Src. Endothelial proliferation is inhibited by sim- secondary participants in all three pathways. The tar- vastatin or complement component 3 (C3) exoen- get antigens are mainly major or minor histocompat- zyme, which also inhibit RhoA. In Rag-1 knockout ibility antigens. The possibility of other autoantigens mice that lack T and B cells, adoptive transfer of suchasvimentinandcollagenVhasbeenraised,but DSA causes TA over 3–4 weeks in heart allografts. In theevidenceisnotyetconvincingtotheseauthors.The humans, TA is often associated with C4d deposition in

40 Chapter 4B: Major complications – pathology of chronic rejection

Figure 4B.3 In chronic antibody-mediated rejection CHR, C4d is Figure 4B.4 In CHR, multi-lamination of peritubular capillary walls deposited in peritubular capillary walls, as shown in this occurs in renal allografts, best shown by electron microscopy. This immunofluorescence micrograph of an allograft kidney cryostat is a manifestation of repeated episodes of endothelial injury and section stained with anti-C4d monoclonal antibody. CHR is well repair mediated by antibodies. Similar changes occur in glomerular documented in the kidney but not yet in other organs, although capillaries, where it is termed transplant glomerulopathy.Seecolor it probably exists. See color section. section.

Chronic humoral rejection in myocardial capillaries. However, animal studies indi- cate that complement fixation is not necessary for the kidney antibody-mediated TA. This may explain the variable The concept of chronic humoral rejection (CHR) was associationofC4ddepositionandTAinhumans. introduced with diagnostic criteria by Mauiyyedi and The third pathway of chronic rejection involves the colleagues in studies of kidney allografts and was soon innate immune system and NK cells. This has been confirmed and expanded by Regele. In the kidney, most clearly demonstrated in animal studies in which the hallmarks of chronic antibody-mediated injury MHC compatibility and immune reactivity can be bet- include C4d deposition in peritubular and glomeru- ter controlled. The first evidence came from murine lar capillary walls, multi-lamination of the basement models that had become tolerant of donor skin tissue membrane of peritubular and glomerular capillaries, by neonatal exposure to donor cells or later by induc- and the presence of circulating DSAs. The combina- tion of mixed chimerism. Curiously, these recipients tion of these features define CHR in the Banff schema. showed no acute rejection of donor strain hearts, but C4ddepositionincapillarywalls,asseenby later developed TA. In parent to F1 heart grafts, in immunofluorescence or immunohistochemistry, is which T or B cells are tolerant to the graft (self) and pathognomonic for humoral rejection, both acute and only NK cells are expected to respond to the lack of self chronic (Figure 4B.3). Because C4d is cleared by allo- on the graft, TA also developed. In these same models, grafts in 1–2 weeks, visualization of C4d deposition TA could be inhibited by blocking NK cells. In Rag- serves as a marker of current immunological activity. 1 knockout mice (who lack T or B cells), when par- Multi-lamination of peritubular capillary walls is ent to F1 grafts are performed, viral infection promotes thought to arise from repeated endothelial injury the ability of NK cells to produce TA. These results are and healing deposition of new basement membrane particularly interesting in conjunction with the clinical material (Figure 4B.4). This multi-lamination is best evidence that viral infection (cytomegalovirus, CMV) appreciated on electron microscopic studies, but can is a risk factor for TA in the kidney and in the heart. also be highlighted by special stains that have an

41 Section 1: General

eventual loss of capillary density, and this contributes to secondary ischemic injury. The loss of capillaries may make it more difficult to interpret C4d stains because fewer target vessels are present. Other, less specific histological features of CHR may be present in the kidney, such as interstitial fibrosis, tubular atrophy, and TA. Interstitial fibrosis and tubular atrophy, a manifestation of any chronic injury to these renal compartments, can frequently be accompanied by lymphoid aggregates. Although TAisthoughttobeamajorfactorinchronicrejec- tion, it is not considered in the definitive histological diagnosis of chronic rejection because chronic vascu- lar injury due to ischemia, hypertension, atheroscle- rosis, drug toxicities, and more can all have a similar histological end point. The majority (57%) of late kidney graft losses are associated with DSA and/or C4d deposition, and theriskofsubsequentgraftfailureissignificantly Figure 4B.5 Transplant glomerulopathy (TG) is defined by worse after a C4d+ biopsy. Increased transcription duplication of the glomerular basement membrane, easily appreciated when severe by light microscopy in sections stained of endothelial-specific genes has been detected in with PAS. This is highly associated with circulating anti-donor patientswhohaveDSAandiscorrelatedwithan antibodies and deposition of C4d. However, the pattern of injury adverse outcome, even in the absence of detectable can be due to other causes, such as thrombotic microangiopathy. See color section. C4d. Thus antibody-mediated rejection is a significant clinical barrier to long-term success of kidney allo- grafts. It is likely that CHR also occurs in other organs affinity for the basement membrane, such as Periodic and will be one day be identified by appropriate clini- Acid-Schiff (PAS) or silver stains. A similar etiology copathological studies. is thought to underlie duplication of the glomerular basement membrane (GBM), which, when found in allografts, is termed transplant glomerulopathy Chronic rejection in the liver (TG) (Figure 4B.5). Repeated or sustained endothelial Although liver allografts are relatively resistant to injury leads to deposition of new basement membrane rejection, features of chronic rejection have been material within the capillary loops and eventual dupli- described, primarily consisting of TA (difficult to cation of the GBM. This can be easily observed by detect in needle biopsies) and immune destruction of light microscopy as well as electron microscopy. The the biliary tree, termed vanishing bile duct syndrome finding of GBM duplication/TG is not pathognomonic (VBDS). VBDS is manifested in biopsies or liver resec- for CHR. Other causes of chronic endothelial injury, tions by loss of bile ducts within portal tracts and even- such as thrombotic microangiopathy (TMA), also lead tual signs of biliary obstruction such as cholestasis, but to GBM duplication in allografts and native kidneys. without ductular proliferation (Figure 4B.6A). VBDS Because transplant recipients can often develop TMA also occurs in native livers under certain conditions due to drug toxicities as well as recurrent disease (lymphoma, drug toxicity, and viral infection); there- (membranoproliferative glomerulonephritis [MPGN] fore,itmayormaynotbeadirecteffectofimmunolog- pattern) in addition to CHR, TG has an etiologic ical rejection. Bile duct epithelium manifests molec- differential diagnosis when seen on a biopsy. Other ular markers of senescence (p21) in early chronic findings, such as intraluminal thrombi (TMA), rejection, a process that can be reversed by conver- immune-complex deposits (recurrent or de novo sion from cyclosporine to tacrolimus. Transient lob- glomerulonephritis), and C4d (CHR), can be helpful ular hepatitis may also be a feature of chronic rejec- when assessing TG in renal allografts. Repeated tion and is potentially reversible, although VBDS antibody-mediated capillary injury in CHR causes and TA (Figure 4B.6B) are resistant to current therapy.

42 Chapter 4B: Major complications – pathology of chronic rejection

(A) (B)

Figure 4B.6 Chronic rejection in the liver affects bile ducts and vessels. (A) In vanishing bile duct syndrome, a manifestation of chronic rejection in the liver, there is loss of bile ducts in portal tracts with associated cholestasis. (B) Chronic allograft vasculopathy in an hepatic artery shows prominent foam cells in the intima. Foam cells, which are macrophages filled with lipid, are characteristic of chronic rejection and are probably related to hyperlipidemia. Photomicrographs courtesy of Dr. V.A. Marcus, McGill University (A) and A.J. Demetris, University of Pittsburgh (B). See color section.

Features predicting irreversibility included bile duct imaging or catheterization, or histologically at the time loss of greater than 50%, bridging perivenular fibro- of explant or autopsy. In addition, because the car- sis, and the presence of foam cell clusters within portal diac vasculature is a terminal circulation, patients with tracts. The role of antibodies in hepatic chronic rejec- severe CAV are in immediate danger, as opposed to tion is not clear, although DSA is associated with late renal transplant patients who could feasibly return to graft loss, and several studies have detected endothelial dialysis. In the heart, TA is felt to be multi-factorial and C4d in liver grafts with ductopenic chronic rejection attributable to both cell-mediated and humoral fac- and TA. tors, as well as non-immune injury due to hyperten- sion, metabolic derangements, perioperative ischemia, Chronic rejection in the heart and infection. Interferon ␥ (IFN␥)isthoughttobe a critical mediator of CAV, originating from lympho- The study of rejection in cardiac allografts has been cytes, macrophages, and even medial smooth muscle greatly limited by the difficulty in obtaining real-time cells. histological material for examination. TA is the main lesion in heart transplants that leads to late graft loss, ischemia,anddeath;intheheart,byconventionitis Chronic rejection in the lung termed cardiac allograft vasculopathy (CAV). TA in The lesions of chronic rejection in the lung consist the heart has a similar appearance as in other organs, of TA and a lesion occluding airways that is termed with concentric intimal fibrosis leading to luminal obliterative bronchiolitis (OB). OB describes edema- occlusion (Figure 4B.7). However, for practical rea- tous myofibroblastic obliteration of membranous and sons, TA and luminal occlusion can only be assessed by respiratory bronchioles, leading to partial or complete

43 Section 1: General

Figure 4B.7 Chronic rejection in heart allografts is generally best appreciated in angiographic studies, such as intravascular ultrasound, because the diagnostic lesion is in the small-to-large Figure 4B.8 In addition to TA, chronic rejection in the lung results arteries not typically sampled in endomyocardial biopsies. TA is in edematous myofibroblastic obliteration of a bronchiolar lumen, manifested in the heart by the same features as in other organs, termed obliterative bronchiolitis, which can be identified in biopsies namely, concentric intimal fibrosis leading to luminal stenosis and and has characteristic effects on pulmonary function (Courtesy of eventual occlusion with intramural chronic inflammation, as shown Dr. E.J. Mark, Massachusetts General Hospital). See color section. in this explant sample (Courtesy of Dr. J.R. Stone, Massachusetts General Hospital). See color section.

luminal occlusion (Figure 4B.8). The fibrosis can also the lung is similar to that found in other solid organ be more hyalinized, with an eccentric or concentric transplants, with intimal fibrosis and a possible link to distribution, and is often associated with fragmenta- CHR. tion and destruction of the smooth muscle and elas- tica of the airway wall. Mononuclear inflammation is Comments on treatment and frequently detected within the obliterative fibrosis. At times,theonlycluetothepreviouspresenceofanair- reversibility way is a focus of dense laminated fibrosis adjacent to Several issues impede the treatment of chronic rejec- a pulmonary artery. The pathogenesis of OB is prob- tion: diagnostic accuracy, detection of activity, and ably a mixture of immune and non-immune mecha- availability of appropriate therapy. The availability of nisms. In miniature pigs with MHC identical, minor- C4d staining and solid-phase antibody testing has mismatched lung allografts, classic OB lesions develop. improved the sensitivity and specificity of diagnosis Although certain studies have shown evidence of acute of CHR. Standard histopathology, sometimes coupled antibody-mediated lung rejection, pulmonary CHR with immunophenotyping of inflammatory cells, is per se has not yet been specifically defined. How- helpful in the diagnosis of T-cell–mediated lesions. ever, DSA and autoantibodies have been associated However, there is no test analogous to C4d to show the with OB. OB was more frequent in patients with anti- T cells’ direct effect on graft tissue. When contemplat- HLA antibody and was preceded by circulating anti- ing treatment of CHR, presence or absence of lesional body. Pulmonary alveolar septal C4d and C3d depo- activity (i.e., C4d deposition) is of great importance. sition predicted graft failure and was correlated with For T-cell–mediated rejection, perhaps the best mea- bronchial wall fibrosis. In murine models, when anti- sure of activity would be detection of cytokines and class I antibodies were introduced into native lungs via cytotoxic molecules in tissue (e.g., granzyme B, IFN␥, tracheal lavage, lesions resembling OB were induced. and T-bet); however, this is not yet routinely available. Experimental studies therefore support a possible role Immunohistochemistry or immunofluorescence may for antibodies in chronic rejection in the lung. TA in prove useful, especially when multiple markers can be

44 Chapter 4B: Major complications – pathology of chronic rejection combined. Although staining for cytokines themselves Further reading is difficult, other possibilities include cytokine target Colvin RB. Pathology of chronic humoral rejection. Contrib ␥ detection (e.g., IDO for IFN ,andSmad3fortrans- Nephrol 2009; 162: 75–86. forming growth factor ␤) and cell proliferation mark- GastonRS,CeckaJM,KasiskeBL,et al. Evidence for ers (Ki67). antibody-mediated injury as a major determinant of late A major limitation is the current lack of effec- kidney allograft failure. Transplantation 2010; 90: 68– tive treatment for chronic rejection. Agents in cur- 74. rent use are primarily directed against acute rejection Hill GS, Nochy D, Loupy A. Accelerated atherosclerosis: a and thereby reduce recurrent or sustained episodes of form of transplant arteriopathy. Curr Opin Organ acute rejection as a cause of eventual chronic rejec- Transplant 2010; 15: 11–15. tion. They therefore indirectly have beneficial effects Mitchell RN. Graft vascular disease: immune response on the prevention of chronic rejection. Cellular path- meets the vessel wall. Ann Rev Pathol Mech Dis 2009; 4: ways involved in chronic rejection need to be defined 19–47. at the molecular level. Missing from our current reper- Sis B, Halloran PF. Endothelial transcripts uncover a toirearedrugsthatwillhaltantibodyproductionatthe previously unknown phenotype: C4d-negative level of the plasma cell, inhibit myofibroblast-related antibody-mediated rejection. Curr Opin Organ fibrosis, and restore cell longevity and tissue repair. Transplant 2010; 15: 42–8. Further exploration of additional and largely unknown Weigt SS, Wallace WD, Derhovanessian A, et al. Chronic pathways of chronic rejection is needed. Only when allograft rejection: epidemiology, diagnosis, we have investigated these pathways and developed pathogenesis, and treatment. Semin Respir Crit Care Med appropriate targeted agents will we be in a position to 2010; 31: 189–207. determine whether or not lesions of chronic rejection Zhang X, Reed EF. Effect of antibodies on endothelium. Am are indeed treatable and potentially reversible. J Transplant 2009; 9: 2459–65.

45 Section 1 General Chapter Major complications – infection

4C Camille Nelson Kotton

Key points A timeline of infection after transplantation has r Infections are among the most common been described (Figure 4C.1). In the first month complications after transplantation and after organ transplant, infections tend to be related greatly increase the morbidity and mortality to the surgical procedure and hospital stay and of transplantation and decrease graft include wound infection, anastomotic leaks and survival. ischemia, aspiration pneumonia, catheter infection, r Improved understanding of various and Clostridium difficile colitis. They are more likely infections, diagnostics, therapeutics, and to be due to resistant pathogens in this popula- prevention has allowed for improvement of tion with frequent health care exposure, including outcomes after infection in transplant methicillin-resistant Staphylococcus aureus (MRSA), recipients. vancomycin-resistant Enterococcus faecalis (VRE), and non-albicans Candida. Donor-derived infections and r Prophylactic measures and medications can recipient-derived infection due to prior colonization significantly decrease the risk of infection with agents such as Aspergillus or Pseudomonas may after transplantation. present in this phase. r Pretransplant evaluation for latent infections The next 5 months are often the period when the and optimization of vaccination can help classic opportunistic infections occur; their risk can mitigatetheriskofinfectionafter be mitigated or delayed by prophylaxis and increased transplant. by intensified immunosuppression, leukopenia, or immunomodulatory (and usually endogenous) viral Infections are among the most common complications infections. Although relatively standard prophylaxis after transplantation and greatly increase the morbid- (e.g., trimethoprim-sulfamethoxazole and an antivi- ity and mortality of transplantation while decreas- ral [acyclovir or ganciclovir or related products]) ing graft survival. New infections occur from acqui- mitigates the risk of many infections, certain infec- sition of infections in the hospital, from the organ tions are still seen, including BK reactivation, hep- transplant or blood product donor, or in the com- atitis C infection (HCV), adenovirus and influenza munity. Reactivation of latent infections encompasses infections (especially seasonally), and Cryptococcus another significant number of infections. In gen- neoformans infection. Without prophylaxis, numerous eral, the intensity of immunosuppression is con- additional infections are seen (especially those in the sidered highest for a year after solid organ trans- herpes virus family and hepatitis B [HBV]), as well plant and for 2 years after hematopoietic stem cell as Pneumocystis jiroveci pneumonia and infections transplant. Clinically focused guidelines on diagno- with Listeria, Nocardia, Mycobacterium tuberculosis sis, treatment, and prevention of many infections (TB), Toxoplasma, Strongyloides, Leishmania,andTry- after solid organ transplant have recently been pub- panosoma cruzi (T. cruzi; the etiologic agent of Chagas lished by the Infectious Disease Community of Prac- disease). tice of the American Society of Transplantation The stable and relatively healthy organ trans- (2009). plant recipient who is more than 6 months out

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

46 Chapter 2: Major complications – infection

Nosocornial, technical Donor-Derived Activation of latent infection Community-acquired Infection (donor or recipient) (relapsed, residual, opportunistic)

Dynamic assessment of risk of infection Transplantation

Common Infections in Solid-Organ Transplant Recipients

Recipient-Derived <1 Month 1–6 Months >6 Months Infection Infection with antimicrobial- With PCP and antiviral (CMV, HBV) Community-acquired pneumonia, resistant species: prophylaxis: urinary tract infection MRSA Polyornavirus BK infection, nephropathy Infection with aspergilus, atypical VRE C. difficile colitis molds, mucor species Candida species (non-albicans) HCV infection Infection with nocardia, rhodo Aspiration Adenovirus infection, influenza coccus species Catheter infection Cryptococcus neoformans infection Late viral infection: Wound infection Mycobacterium tuberculosis infection CMV infection {colitis and Anastomotic leaks and is chemia Anastomotic complications retinitis} Clostridium difficile colitis Without prophylaxis: Hepatitis (HBV, HCV) Donor-derived infection Pneumocystis HSV encephalitis

(uncommon): Infection with herpesviruses (HSV, Community-acquired (SARS, West Nile virus infection) HSV, LCMV, rhabdovirus, VZV, CMV, EBV) JC polyomavirus infection (PML) (rabies), West Nile virus, HBV infection Skin cancer, lymphoma (PTLD) HIV, Trypanosome crazi Infection with listeria, nocardia, toxo- plasma, strongyloidies, leishmania, Recipient-derived infection T. cruzi (colonization): Aspergillus, pseudomonas

Figure 4C.1 Trends in the timings of infection after organ transplantation. Reproduced with permission from Fishman JA, Infection in solid-organ transplant recipients, N Engl J Med 2007; 357: 2601–14. C Massachusetts Medical Society. All rights reserved. from transplant tends to develop community-acquired for latent infections such as T. cruzi, Coccidioides, or routine infections, including urinary tract infec- and Strongyloides. Those seronegative for measles, tions, upper respiratory infections and pneumonia, mumps, rubella, hepatitis A and B, and varicella gastroenteritis, and varicella zoster virus (VZV). should undergo important pretransplant vaccination, Infections with unusual and opportunistic pathogens as some are with live viral vaccines that can not such as Aspergillus,unusualfungi,Nocardia,and begivenaftertransplantwhenarecipientissig- Rhodococcus are still seen. In the era of effective nificantly immunosuppressed, leaving them possibly prophylaxis with valganciclovir, an increased risk of permanently vulnerable to potentially life-threatening late cytomegalovirus (CMV; occurring more than infection(s). 6 months after organ transplant) has been noted, Atypical presentation of infection is more com- primarilyinthefewmonthsafterprophylaxishas moninimmunosuppressedhosts.Clinicalpresenta- been stopped. Other late viral infections include poly- tions may be subtler, yet the patients more ill com- omavirus infections (from BK, causing nephropa- pared with normal hosts. For example, transplant thy primarily in renal transplant recipients, or JC, recipients with West Nile virus infection are much causing progressive multifocal leukoencephalopathy) more likely to have clinical illness and succumb to and Epstein-Barr virus (EBV)–related post-transplant infection. Clinicians need to expand their differential lymphoproliferative infections. diagnoses when caring for transplant recipients. Diag- Pretransplant evaluation can help mitigate the risk nosis of emerging, novel, and atypical pathogens is of some infections, especially latent ones. Knowledge especially challenging in this vulnerable population, as of serostatus for CMV, EBV, HBV, and HCV can has been seen with cases of lymphocytic choriomenin- help optimize post-transplant management. Potential gitis virus, tuberculosis, Chagas disease, strongyloidi- transplant recipients should be screened for latent TB, asis, and numerous others. either by skin testing or use of the QuantiFERON- The importance of donor-derived infections has TB Gold (in vitro test for TB and latency). Recipi- been increasingly recognized in recent times. Such ents from or in endemic regions should be screened infection occurs in at least 1% of deceased donor

47 Section 1: General

organ transplants and are significantly higher in some cytes, endothelium, and other tissues, and the gamma organs, e.g., lung transplants. Although some trans- herpes viruses (EBV and HHV-8) are latent in lym- mission of some infections is expected, such as CMV phoid tissue. Disseminated infection from any of the and EBV, others have been a surprise to clinicians human herpes viruses can be life-threatening. caring for patients. Such unanticipated donor-derived Numerous other types of viruses cause disease infections range from viruses such as rabies, lympho- in transplant recipients. Respiratory viruses such as cytic choriomeningitis, and West Nile virus, to bac- influenza, respiratory syncytial virus (RSV), aden- teria including bacteremias and tuberculosis, fungi ovirus, parainfluenza, and human metapneumovirus including cryptococcosis and histoplasmosis, and par- are common and may present more subtly or with asites including T. cruzi (causing Chagas disease) fulminate disease. Hepatitis viruses (primarily B and and Strongyloides. Enhanced appreciation of donor- C) are common causes for liver transplantation and derived infections has resulted in better screening and also common complications after transplant, predom- use of diagnostics. inantly as reactivation of latent infections. In addition, Post-transplant infections can be mitigated by pre- the primarily zoonotic hepatitis E has been reported as ventative methods as described next in individual sec- an emerging pathogen in transplant recipients. Most tions, as well as routine vaccinations, intake of clean adults have latent infection with the polyoma viruses food and water, preventative measures during times BK and JC. Although BK is predominantly a pathogen of outbreaks (as with severe acute respiratory syn- in kidney transplant recipients, it can cause disease drome [SARS] and H1N1 influenza), visits with travel in other transplant recipients. Risk of BK reactivation medicine specialists prior to visiting high risk regions, relates directly to the intensity of the immunosuppres- safersexualpracticesfornon-monogamousrecipients, sion, and early diagnosis of BK replication and sub- and guidance on better tattoo acquisition. sequent reduction in the immunosuppressive regimen largelyabrogatestheriskofBKnephropathy,which generally has poor outcomes in kidney transplant Viruses recipients, with high rates of graft loss. JC virus causes Viral infections are the most common type of infection progressive multifocal leukoencephalopathy, which is after transplantation. They encompass a broad array of often mortal but fortunately is also fairly rare. Numer- viruses, ranging from herpes to respiratory to hepati- ousotherviruseshavebeenshowntocausediseasein tis and other viruses. In addition to the direct effects transplant recipients, including parvovirus B19, West (i.e., clinical syndromes) caused by viruses, they can be Nile virus, lymphocytic choriomeningitis virus, and quite immunomodulatory, especially CMV, HCV, or others. EBV,resulting in both inflammation (perhaps mitigat- Viral diagnostics have improved exponentially ing graft tolerance) as well as increased immunosup- with the onset of molecular techniques. Viral culture pression (increasing the risk of infection from other is progressively being replaced by more rapid and spe- opportunistic pathogens). Because many of the impor- cific molecular assays. Within a matter of hours, vari- tant viruses after transplantation remain latent, their ous amplification methods can document active repli- ongoing management is a fine balance between opti- cating viral infections. In the era of quantitative assays, mal levels of immunosuppression and reactivation of trends in infection can be followed over time, as is infection. seen with serial assays for response to CMV treatment The human herpes viruses (HHV) are the most or for BK viremia/viruria or EBV viremia. Molecular common etiologic agents of infection after transplan- diagnostics have given us powerful assays for infec- tation. The family includes eight viruses: herpes sim- tions that were previously very challenging (or even plex type 1 and type 2 (HSV-1, -2), VZV, EBV, CMV, impossible)todiagnoseinthispopulation,suchaspar- the roseola-like human herpes virus 6 and 7 (HHV-6, vovirus B19, HHV-6 and -7, and others. Knowledge of -7), and human herpes virus 8 (HHV-8, the etiologic serostatus for some viruses, such as CMV, EBV, and agent of Kaposi’s sarcoma). The alpha herpes virus the hepatitis viruses, can be helpful in management. family (HSV-1, -2, VZV) establishes latent infections Immunohistochemistry can be very helpful for various primarily in sensory ganglia, whereas the beta herpes h e r p e s i n f e c t i o n s , i n c l u d i n g H S V, V Z V, E BV, C M V, viruses (CMV, HHV-6, -7) maintain latency in leuko- and HHV-8.

48 Chapter 2: Major complications – infection

Effective treatment of viral infections may involve mal management of CMV after solid organ transplant a multi-pronged approach: use of antiviral agents, have been published. Because the severity of many reduction of immunosuppression when possible, viralinfectionsrelatestotheintensityofimmuno- and augmentation of immunity through the use of suppression, careful management of the regimen can immunoglobulins and sometimes adoptive infusions result in a decreased risk of viral infections, espe- of CMV-specific T cells. Common antiviral treatments cially with latent viruses. CMV immunoglobulin and include the acyclovir family (including famciclovir and HBV and VZV hyperimmune globulins have been valacyclovir, primarily for HSV and VZV infections), shown to be effective in preventing infection in certain ganciclovir (with the oral prodrug valganciclovir for settings; repleting recipients with hypogammaglobu- CMV and other infections), foscarnet (predominantly linemia with intravenous immunoglobulin can reduce for resistant CMV), cidofovir (for resistant CMV, BK their net risk of infection. In addition, periodic moni- virus, and others), and ribavirin (for RSV and other toring has been shown to be helpful, as with CMV,EBV, less common infections). Hepatitis B has numerous and BK viruses. Vaccination against influenza, hepati- antiviral agents for treatment, and hepatitis C is pri- tis A and B, human papillomavirus, and other viral marily treated with ribavirin and interferon. Reducing pathogens can provide additional protection. the intensity of immunosuppression (even transiently) may allow for more rapid clearance of a viral infection. Although not always well evidence-based, repleting Bacteria recipients who have hypogammaglobulinemia with Bacterial infections occur at increased frequency in intravenous immunoglobulin may help clear infec- the vulnerable transplant recipient. They range from tion. Some centers use CMV immunoglobulin in routine infections such as urinary tract infections, seronegative recipients with active disease. The novel pneumonias, and bacteremias to more exotic infec- use of adoptive infusions of CMV- or EBV-specific tions with Nocardia, Rhodococcus, Listeria,andother T cells has been shown to be effective especially in pathogens. Frequent exposure to health care settings hematopoietic stem cell recipients and increasingly in increases the risk of resistant pathogens, including organ transplant recipients. MRSA, VRE, Pseudomonas, Stenotrophomonas,and Prevention of viral infections can be achieved others. TB reactivates at much higher rates in those by the use of antiviral medications, judicious use of with renal and hepatic failure, as well as in the post- immunosuppression, occasional use of immunoglob- transplant period. ulins, careful monitoring, and vaccination. The most Diagnosis of bacterial infections still relies heavily common antiviral medications used for prevention on culture techniques. In order to optimize the diag- include the acyclovir family, primarily for HSV and nostic yield of cultures, clinicians should notify the VZV prophylaxis, and (val)/ganciclovir, which is laboratory when usual organisms are suspected, such targeted at CMV prevention but also successfully as Listeria, Rhodococcus,TB,andNocardia.Expand- decreases the rates of other herpes virus infection ing the standard panel of antibiotic sensitivity at the to varying extents, as well as anti-hepatitis B agents. time of initial diagnosis may help with subsequent The duration of prevention varies considerably among therapy, especially given the increased risk of drug institutions, but many programs use antivirals for 3– interactions and side effects, partly due to concomi- 6monthsafterorgantransplant(especiallywithhigh- tant use of multiple medications (i.e., increased risks risk situations where the donor is CMV seropositive of leukopenia, nephrotoxicity, etc.). Molecular diag- and the recipient seronegative, or with lung trans- nostics are emerging as a diagnostic methodology for plant), often with either a ganciclovir family agent to bacterial infections. Serologic techniques tend to yield prevent CMV or an acyclovir-type drug to prevent dis- diagnoses less frequently in this population due to seminated VZV infection. Although some organ trans- a more muted immunological response. Histopathol- plant centers use antiviral agents in certain cohorts ogy, especially with special stains for microorgan- of patients at risk for CMV (termed universal pro- isms, can sometimes be helpful in achieving a phylaxis), other use preemptive therapy,wheretreat- diagnosis. ment is begun only when routine monitoring tests Treatment often begins in febrile or ill-appearing show evidence of active infection. Guidelines for opti- transplant recipients with empiric antibacterial

49 Section 1: General

therapy, which should be chosen based on local Fungal diagnostics utilize dedicated fungal stains epidemiology. This approach appears to be justified and culture as well as detection of fungal antigens in by the significant incidence of bacteremia in the blood, urine, and other fluids. Fungi may be harder post-transplant period and by the concomitant high to grow in culture and harder to diagnose than other mortality rate when appropriate treatment is delayed, pathogens. A high level of suspicion, as well as mul- especially in infections caused by certain pathogens. tiple diagnostic approaches, is imperative in the diag- Transplant patients are at higher risk for resistant nosis of these potential more elusive pathogens. Some pathogens, and the empiric antibiotic choice should pathogens such as Candida will usually grow on rou- reflect this. Once a culture diagnosis has been made tine culture, whereas others require fungal culture and antibiotic susceptibilities are available, the antibi- media to promote growth. Fungal antigens, including otic regimen may be tailored. Because of the increased the 1,3-␤–D-glucan, galactomannan, and cryptococ- rates of resistance, intravenous (IV) therapy is more cal assays, have increased diagnostic capacity in recent common in this population, which requires the use of times. The 1,3-␤–D-glucan assay can be positive with prolongedIVaccess.Ingeneral,armveinsshouldbe a variety of fungal pathogens ranging from Candida avoided and preserved for future hemodialysis access to Aspergillus and numerous others including Fusar- in this population at higher risk for chronic kidney ium spp., Trichosporon spp., Saccharomyces cerevisiae, disease. Optimizing the drainage of collections via the Acremonium, Coccidioides immitis, Histoplasma capsu- use of radiographically or surgically placed drains may latum, Sporothrix schenckii, Blastomyces dermatitidis, help clear infection and prevent recurrent infection. and Pneumocystis carinii. The galactomannan antigen Preventative measures include eliminating any nidi hasbeenusedonavarietyofspecimensandisrela- of infection (such as IV catheters, skin defects that tively specific for Aspergillus. Serology may sometimes encourage abscess formation, etc.) and optimizing foci be helpful (i.e., Coccidioides). Histopathology can of recurrent infections (i.e., urinary tract infections). also provide diagnostic input, especially with special The use of trimethoprim-sulfamethoxazole after stains. transplant to prevent Pneumocystis has the additional Treatment of fungal infections may involve use advantage of preventing other bacterial infections, of one or more antifungal agents, as well as surgical ranging from Streptococcus to Listeria to Nocardia debulking (especially with zygomycosis and some- and many others. Vaccination against Streptococcus times aspergillosis). Candida infection may be treated pneumoniae, Clostridium tetani, Corynebacterium with an azole (primarily fluconazole) or with an diphtheriae,andotherbacterialpathogensmay echinocandin (i.e., micafungin, caspofungin, anidu- provide additional protection. lafungin). Depending on the individual pathogen, thefilamentousfungalinfectionsaretreatedwithan amphotericin B product, often a lipid-based one for Fungi better tolerability, or with a higher level azole, such Invasive fungal disease, particularly aspergillosis, are as voriconazole or posaconazole. The echinocan- significant causes of morbidity and mortality in trans- dins are sometimes used in salvage regimens or plant recipients and are among the most dreaded as part of a multi-drug regimen, albeit with very of the infectious complications. Although infections little available data for multi-drug regimens in this from Candida tend to be more manageable, infections setting. Antifungal susceptibilities are increasingly such as zygomycosis have very high mortality rates. being used to guide optimal treatment, as is thera- Iron overload in liver transplant recipients may explain peutic drug monitoring for the higher level azoles. their propensity to develop more fulminant disease There are important drug interactions between the presentation and a higher risk of disseminated disease immunosuppressive agents (especially tacrolimus, due to a number of opportunistic infections, including cyclosporine [CyA], and sirolimus) and the azoles invasive aspergillosis, cryptococcosis, and zygomyco- (especially the higher level ones), necessitating reduc- sis. Although these have a predilection to occur in the tions in doses of the immunosuppressive agents. P. early post-transplant period, they may also occur years jiroveci istreatedwithagentssuchastrimethoprim- later, for example, Pneumocystis jiroveci and Crypto- sulfamethoxazole, clindamycin, primaquine, ato- coccus neoformans, which may cause late meningitis. vaquone, and others.

50 Chapter 2: Major complications – infection

Preventing fungal infections involves a combina- plant tourism); increases in leisure tourism to endemic tion of avoidance measures, filtered air systems in regions by transplant recipients; and decreases in CyA- hospitals, recognition of existing infection or colo- based immunosuppressive regimens and the increased nization, and evidence-based targeted antifungal pro- use of newer drugs that lack the anti-parasitic effects of phylaxis. Fungal spores are ubiquitous in the environ- CyA metabolites. ment.Itisveryrarelypossibletoreliablydistinguish Diagnosis of parasitic infections in solid organ community-acquired from nosocomial aspergillosis. transplant recipients can be complex. Depending on Transplant recipients should wear gloves while gar- the parasite, a variety of techniques may be used, dening or touching plants or soil, and they should ranging from rapid diagnostics on stool, peripheral avoidinhalingorcreatingsoilordustaerosolsthatmay blood smears (Babesia,malaria,T. cruzi), special stains contain fungal spores. They may wish to wear well- and microscopic exam of various specimens or tissues fitting face masks (FFP2/N95 masks) in some unavoid- (blood,stool,biopsy),culture,serology(whichmay able situations and should always wash their hands be less helpful in this population), and histopathology. after such contact and care for skin abrasions or cuts Molecular diagnostics, when available, can be quite sustained during soil or plant contact. Invasive fun- helpful.Useofclinicalmeasuressuchaseosinophilia gal infections in the explanted lungs, more common may be muted in this population, in whom the in those with cystic fibrosis, are often not recognized immunosuppressive regimen (especially steroids) may before lung transplantation and have been associated cause false-negative results. with poor outcomes; similarly, airway colonization Treatment of individual parasitic infections can with fungus in other organ recipients may blossom involve medications that may interact with trans- into a full infection after transplant; thus knowledge plant medications, or have significant side effects, and of culture data at or before the time of transplant may should be used carefully. Immunocompromised hosts help target therapy and mitigate infection. Antifungal are more likely to have relapses of certain parasitic prophylaxis varies broadly among transplant centers; infections (i.e., Babesia, T. cruzi, Strongyloides)and in general, most centers would only give prophylaxis should be monitored after treatment. against Aspergillosis and other fungi for a relatively Parasitic infections can be prevented by avoiding brief period, if at all. Many centers recommend pro- ingestion of contaminated food and water (predomi- phylaxis against P. jirov ec i , often with trimethoprim- nantly for intestinal pathogens), by screening recipi- sulfamethoxazole, atovaquone, or dapsone. Adherence ents for latent infection prior to transplant as well as to avoidance measures, acknowledging their limita- organ and blood product donors in endemic regions tions, combined with antifungal prophylaxis, is likely (i.e., toxoplasmosis, Chagas disease, malaria, babesio- to be the most effective approach to prevent invasive sis), and by use of preventative medications such as fungal disease. trimethoprim-sulfamethoxazole (used chronically to prevent T. gondii infection) or ivermectin (to treat active or latent Strongyloides). Once a more common Parasites infection after solid organ transplant, toxoplasmosis Parasitic infections tend to be less common than has become a largely preventable disease in the era of the previously mentioned pathogens. The more clini- trimethoprim-sulfamethoxazole or atovaquone pro- cally significant parasites include Toxoplasma gondii, phylaxis. Use of anti-malarial prophylaxis in endemic Strongyloidesstercoralis,T.cruzi,Leishmania,and regionsisrecommendedforalltransplantrecipients. intestinal parasites. The incidence of parasitic infec- Avoidance of tick and other insect bites will also tion is anticipated to increase in transplant recipients decrease risk of transmission. due to multiple factors, including increases in active Infections tend to occur in fairly predictable phases organ transplant programs in geographic areas where after solid organ transplant. Although many of the parasitic infections are highly prevalent; increases in classic opportunistic infections occur in the first travel and migration of donors and recipients from 6 months, during what is usually the period of most endemic areas, with latent or asymptomatic infec- intense immunosuppression, the risk of such infection tions, as well as patients from developed countries remains for the duration of time that the recipient is on undergoing transplantation in endemic areas (trans- immunosuppressive medications. The risk of infection

51 Section 1: General

is decreased by the use of prophylaxis and augmented Ison MG, Hager J, Blumberg E, et al. Donor-derived disease by the use of more potent immunosuppression (both transmission events in the United States: data reviewed in the induction and maintenance phases, as well as by the OPTN/UNOS Disease Transmission Advisory during treatment of rejection), allograft rejection, con- Committee. Am J Transplant 2009; 9: 1929–35. comitant infections, leucopenia, and technical surgical Jong EC, Freedman DO. Chapter 8: Advising Travelers with issues. Specific Needs: The Immunocompromised Traveler. 2010 Health Information for International Travel. 2010 ed, Atlanta, GA: Centers for Disease Control and Prevention. Available at: http://wwwnc.cdc.gov/travel/ Further reading yellowbook/2010/chapter-8/immunocompromised- American Society of Transplantation Infectious Diseases traveler.aspx (accessed January 14, 2011). Guidelines 2nd Edition. Am J Transplant 2009; 9: Kotton CN, Kumar D, Caliendo AM, et al.International S1–S281. Consensus Guidelines on the Management of Fishman JA. Infection in solid-organ transplant recipients. Cytomegalovirus in Solid Organ Transplantation. NEnglJMed2007; 357: 2601–14. Transplantation 2010; 89: 779.

52 Section 1 General Chapter Organ donor management and procurement 5 Edward Cantu III and David W. Zaas

plantation continues to grow at a faster rate than the Key points number of available donors. r Traumatic brain injury accounts for one The number of potential donors in the United third of all trauma-related deaths; however, States is believed to be between 10 000 and 14 000 per only 20% of these potential donors become year, with actual donation rates at about 26 donors actual donors. per million population. The United Kingdom remains r Somatic organ donor identification, relatively low at 15 donors per million population, management, and procurement requires although steps are in place to try to increase rates intensive coordinated effort in order to of donation. Spain has the highest rate of cadaveric abrogate the negative consequences of organ donation of any country, with an annual donor brainstem death and avoid donor loss. rate of 30–35 donors per million population. This is r Physiological derangements during the due to a unique organizational model that has been process of brain death are induced via diffuse implemented in other countries, with similar though vascular regulatory perturbation and somewhat less success (see Chapters 40 and 41). The metabolic cellular injury. percentage of donors utilized for lung transplant in r Care of potential donors is goal-directed, the United States is only 19%, significantly lower than with emphasis on minimizing further brain most other countries. Concerns about the risk for poor injury by maximizing cerebral perfusion and outcomes create conservative practice styles, but also preventing or treating systemic morbidity. increase the number of deaths on the waiting list. r Interest in non–heart-beating donors and ex Reasons for limited recovery of organs include vivo perfusion systems continues to grow (1) failure to identify potential donors, (2) inability to and may make significantly more organs obtain consent for donation, (3) donor loss due to car- available. diovascular collapse, and (4) unsuitability of organ(s) for transplantation. This chapter discusses the phys- iological changes of brain death, the management of In many patients with end-stage organ dysfunction, these complex patients, the organization of the recov- solid organ transplantation has emerged as the treat- ery, and new technologies that may allow increased ment of choice; however, the paucity of suitable organ number of organs available for transplantation. donors has placed severe limits on the number of transplants being performed. Given the limitation of Physiology of brain and available organ donors and number of deaths on the waiting list, many transplant programs have expanded brainstem death selection criteria to include older, higher risk donors Brain death has been defined as the complete and irre- in order to offer transplants to more patients. Despite versible cessation of all brain and brainstem function. the increased use of higher risk donors, outcomes Diagnosis is usually made by clinical examination have continued to improve as experience has accrued. documenting no brainstem reflexes and a lack of Unfortunately, the number of patients awaiting trans- spontaneous respiratory effort. Several confounders

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

53 Section 1: General

must also be ruled out, including hypothermia Cardiovascular changes (Ͻ32◦C) and drug or metabolic intoxication. How- Cardiac function is reduced after brain death, which ever, the legal definition does vary from state to state has been attributed to catecholamine cardiotoxicity. and country to country. The myocardial damage and morphological changes As the brain is dying, physiological derangements seen after brain death (myocytolysis, contraction band are induced via two major mechanisms, the first being necrosis, subendocardial hemorrhage, edema, and through a diffuse vascular regulatory perturbation interstitial mononuclear cell infiltration) are similar to and the second through a diffuse metabolic cellu- those described after high-dose catecholamine infu- lar injury. Severe alterations in endocrine function, sion. Further, these changes could be abrogated by immunology, and coagulopathy also commonly man- total cardiac sympathectomy before cerebral insult. ifest. These diffuse physiological insults contribute to The extreme sympathetic surge seen after vagal nuclear numerous complications that compromise end-organ death leads to profound increases in circulating cat- function and result in significant donor loss; there- echolamines and have been shown to correlate with fore, a thorough understanding of the events lead- the speed of ICP elevation. In animal models eval- ing to and following brain death must be clearly uating rate of increase in ICP, it has been demon- understood. strated that slow gradual increases in ICP resulted in Most lethal brain injuries follow a common path- 175-fold increase in epinephrine and a 40-fold increase way whereby a patient suffers brainstem death sec- in norepinephrine plasma levels, with only mild ondary to sudden or gradual increases in intracranial ischemic injury to the myocardium, whereas rapid pressure (ICP). The mass effect from the initial injury increases in ICP resulted in 750-fold and 400-fold self-potentiates, with further progression of ischemia, increase in epinephrine and norepinephrine plasma venous congestion, and edema subsequently leading levels accompanied by extensive ischemic injury to to more mass effect. The brainstem is compromised the myocardium. This sympathetic surge increases by arterial compression when it herniates through the myocardial oxygen demand and results in large foramen magnum, further promoting ischemia and increases in intracellular calcium, leading to impaired edema, resulting in continued increases in ICP until high-energy phosphate production and increased free- cessation of cerebral blood flow occurs. radical generation, leading to further cellular damage. A characteristic sequence of physiological changes Morphometric analysis has demonstrated contrac- occurs as ischemia propagates caudally from the cere- tion band necrosis in 89% of inpatients without previ- brum to the spinal cord. Pontine ischemia leads to a ous heart disease who have succumbed to brain death. mixed sympathetic and vagal stimulation resulting in Furthermore, myocardial gene expression in animal hypertension and bradycardia, commonly referred to models has also demonstrated significant perturbation as the “Cushing reflex.” This, in turn, is followed by an after brain death. The sympathetic storm seen initially unopposed sympathetic stimulation as ischemia pro- followed by massive loss of vasomotor tone are related gresses more caudally to the distal medulla oblongata, by time course and severity. Experiments evaluating involving the vagal nuclei. This period is characterized velocity of ICP and time to cardiovascular collapse by severe hypertension and tachycardia, often referred noted that in the rapid elevation group, death occurred to as “sympathetic storm.” Finally, as herniation is within 1 hour, whereas in the gradual elevation group, completed and ischemia of the spinal cord occurs, animals survived until the experiments were termi- thereisaprogressivelossofspinalsympatheticstim- nated at 3 hours. Further studies by other groups dis- ulation, leading to a loss of all vascular tone, hypoten- covered that rapid induction of brain death resulted in sion, and cardiovascular collapse. Homeostatic more precipitous decrease in systemic vascular tone, control pathways mediated by the hypothalamus and which could be mitigated by volume infusion. pituitary axis are also lost, resulting in endocrine Hemodynamic instability seen after brain death and cellular metabolic dysfunction. Although many is also consequent to loading conditions imposed lethal brain injuries follow these typical changes, on the heart. Additional evidence from animal stud- there is some variability with respect to time course, ies has shown that profound vasodilation and after- speed of deterioration, and time spent at each stage. load reduction results in reduced coronary perfu- This is likely due to etiology of brain injury, clinical sion pressure, relative intravascular hypovolemia, and treatments, and biological variability.

54 Chapter 5: Organ donor management and procurement decreased preload. In canine cross-circulation studies, trophil recruitment and alveolar disruption seen in if loading conditions and coronary perfusion pressure the lung. In rats, traumatic brain injury has demon- were kept constant, no cardiac dysfunction was seen. strated increased lipid peroxidation and intracellular Coronary blood flow is reduced secondary to dys- membrane damage in type II pneumocytes. Addition- function of endothelial-dependent vasodilatory mech- ally, structural damage to tracheal-bronchial epithe- anisms after brain death, but the significance of this lium (and possibly airway defense mechanisms) has findingisnotsettled.Itislikelythatbecauseofalter- been seen and may help explain why there is such a ations in loading conditions, profound vasodilatation, high airway complication rate after severe head injury. and loss of autoregulatory control in coronary arteries, deterioration of cardiac function is even seen in donors without pre-existing cardiovascular disease. Endocrine changes Many hormonal changes have been noted after brain death and are consequent to the loss of the Pulmonary changes hypothalamic–hypophyseal axis. The most common The physiological changes of brain death have direct and immediate manifestation is the early deple- and indirect effects on lung function. The direct tion of anti-diuretic hormone (ADH) and develop- effects are related to neurogenic pulmonary edema and ment of diabetes insipidus. Greater than 75% of inflammatory acute lung injury. The indirect effects are brain death donors have undetectable levels of ADH. secondary to respiratory complications of severe brain This leads to inappropriate diuresis of dilute urine, injury (pneumonia, aspiration, pulmonary contusion, resulting in severe hypovolemia, hyperosmolarity, and and barotraumas). Neurogenic pulmonary edema was hypernatremia. initially recognized at autopsy in combat casualties The anterior pituitary receives some of its blood who had traumatic brain injuries. This constellation supply from the inferior hypophyseal artery, a branch of pulmonary microvascular hemorrhage and pul- off the extradural internal carotid artery. In animal monary edema has also been documented in patients models, brain death results in rapid decrease in free who experience sudden death after intracranial hem- triiodothyronine (T3) from diminished peripheral orrhage. The pathological mechanism results from the conversion of tetraiodothyronine (T4) and impaired sympathetic storm seen in brain death. Unopposed thyroid-stimulating hormone (TSH) secretion. catecholamine excess results in systemic vasoconstric- Decreased circulating T3 is thought to play a role tion, which in turn increases left ventricular afterload in the deterioration of cardiac function after brain and end-diastolic pressure. Accordingly, the left atrial death. Furthermore, studies in pigs and baboons pressure rises, resulting in shunting of blood to the demonstrated that decreased T3 levels are associated central compartment. Together with increased pul- with changes in cellular energy utilization due to the monary vasoconstriction, pulmonary capillary pres- conversion from aerobic to anaerobic metabolism. sures experience dramatic increases, resulting in Administration of T3 in animal studies has been pulmonary edema and hemorrhage due to the signif- shown to enhance cardiovascular function with abro- icantly elevated hydrostatic pressure and mechanical gation of hemodynamic collapse and improvement in disruption of the endothelium. myocardial contractile function. However, in human There is also some evidence that ␣-adrenergic studies there have been conflicting results, and a receptor stimulation may cause an increase in pul- recent clinical trial demonstrated no efficacy with monary capillary permeability independent of its respecttocardiacperformanceorheartretrievalrate hemodynamic effects. Several studies have demon- with infusion of intravenous T3. strated that ␣-adrenergic receptor blockade reduced Plasma cortisol levels are relatively preserved in systemic inflammation and preserved capillary- most brain dead donors; however, the capacity to alveolar membrane integrity. increase secretion with adrenocorticotropic hormone Acute inflammatory lung injury has been exper- stimulation appears attenuated. The clinical signifi- imentally induced after cerebral hemorrhage in ani- cance of this is not entirely known. mal models. Circulating levels of tissue factor and Immediately after brain death, plasma insulin lev- increased intracellular adhesion molecule expression els begin to decrease, leading to increased extracel- have been documented, resulting in progressive neu- lular glucose concentration, shift toward anaerobic

55 Section 1: General

metabolism, and acidosis. Coupled with the ongo- Optimization of donor organ function ing resuscitation with free water to combat hyperna- tremia that typically occurs, this insulin deficiency for donation after brain death (DBD) can quickly lead to profound increases in osmolarity, excessive diuresis, and hypovolemia. Initial resuscitative measures Traumatic brain injury occurs every 7 seconds and Inflammatory and immunological changes results in death every 5 minutes, making it the largest cause of brain death in the United States. Traumatic A generalized inflammatory state has been docu- brain injury accounts for one third of all trauma- mented after severe brain injury and brain death, related deaths. Unfortunately, only 20% of these poten- with increased serum levels of several cytokines such ␤ tial donors become actual donors. This is in part the as interleukin (IL)-6, IL-8, IL-1 ,IL-2Randtumor result of marked pathophysiological changes experi- necrosis factor alpha (TNF-␣). Increased expression of ␣ enced by the brain-injured patient and the extremely IL-6 and TNF- in donor hearts has been correlated labor-intensive management required to support these with donor heart dysfunction. Furthermore, increased patients prior to organ recovery. These patients are cytokine levels in blood and kidneys of brain death typically treated in the intensive care unit, according donors are predictive of worse short- and long-term to evidence-based guidelines advocated by the Brain outcomes as compared with living unrelated donors. Trauma Foundation in the United States. Care is goal- Diffuse cytokine-stimulated inflammatory responses directed, with emphasis on minimizing further brain have been documented in all organ systems evalu- injury by maximizing cerebral perfusion and prevent- ated. Upregulated adhesion molecule expression and ing or treating systemic morbidity. This may include leukocyte adhesion and activation has been demon- butmaynotbelimitedtosupportingbloodpressure, strated. This increased cytokine release results in fur- reducing ICP,maintaining euthermia and euglycemia, ther inflammation and increase major histocompat- maintaining adequate intravascular volume, and cor- ibility complex (MHC) expression on donor cells, recting anemia and coagulopathy. The goal during this which increases its immunogenicity. This has been phase of care is brain protection. confirmed by more frequent higher grade rejection Once brain death has been diagnosed, the empha- occurrence rates in organs from donors after brain sis of patient care changes based on patient and fam- death than those of living donors. ily preference. The challenge is to recognize lethal Although increased cytokine levels have been doc- or irreversible brain injury prior to other end-organ umented, there remains uncertainty regarding the cell damage. During the 1990s, potential donor identi- of origin of these cytokines. It is likely that con- fication rates improved significantly. This improve- tributions from multiple sources (endothelial cells, ment was likely the result of implementation of the platelets, circulating leucocytes, and brain) result in Medicare Conditions of Participation requirements the pro-inflammatory states seen. It is also likely of hospitals, which required deaths to be reported that the hemodynamic perturbations seen with severe to organ-procurement organizations for evaluation brain injury and brain death lead to ischemic changes and that request for organ donation be made by within end organs, which also contribute to the inflam- specially trained personnel. This also, in part, has matory state. contributed to the improvement in conversion rate (ratio of actual donors/potential donors) seen from Other systemic perturbations 33% in 1990 to 66% in 2008. Increasing conver- Temperature regulation is significantly altered, with sion rate is one way to address the current organ loss of hypothalamic control. Hypothermia is not shortage. uncommon in brain death donors consequent to peripheral dilatation and decreased metabolic rate. Severe brain injury results in blood–brain barrier com- Post-consent donor management promise and release of large amounts of tissue factor. Consent to organ donation realigns care to that of Circulating tissue factor coupled with systemic inflam- end-organ preservation and prevention of somatic mation and endothelial activation results in consump- donor loss. Prior to consent, treatment may require tive coagulopathy in up to 28% of brain death donors. many strategies that result in end-organ damage and

56 Chapter 5: Organ donor management and procurement

HD Optimized

Respiratory Optimized MAP ≥ 60 mmHg Ye s Ye s EF > 45%

Continual Reassessment FIO2, < 40%. PAO > 100 mmHg CVP 6-8 mmHg; U/O 1mL/kg/hr PCO2 35-40 mmHg, pH 7.35-7.45 PAP< 30 cm H2O, PEEP 5-10 cm H2O No TV 6-8 mL/kg IBW, Sat >95% No Pulmonary Arterial Catheter

Preload Contractility After load CXR/Chest CT Bronchoscopy CVP 6-8 mmHG CI ≥ 2.4 L/min/m2 MAP > 60 mmHg CWP 6-12 mmHG SVR 800-1200 dyn/sec/cm5

Fluid Diuretics Inotropes Vasopressor Beta Agonist Antibiotics

Respiratory Optimized HD Optimized Ye s MAP > 60 mmHg Continual Ye s FIO2 < 40%, PAO2 > 100 mmHg CVP 6-8 mmHg Reassessment PCO2 35-40 mmHg, pH 7.35-7.45 U/O 1mL/kg/hr PAP < 30 cm H2O, PEEP 5-10 cm H2O TV 6-8 mL/kg IBW, Sat > 95%

No No

Hormone Replacement Therapy

Bolus Infusion Recruitment Preload Pulmonary Hygiene Maneuvers CVP 6-8 mmHg T3 4 μg 3 μg/hr CWP 8-12 mmHg T4 20 μg 10 μg/hr Methyl- prednisolone 15 mg/kg Q 24H Unrecoverable Suitability Vasopressin 1 U 0.5-4 U/hr

Consider Ex Vivo Unrecoverable Stability Perfusion

Figure 5.1 Donor management algorithm (modified from Wood et al., 2004). increase the risk of organ dysfunction after trans- of loss increases with time from brain death. There- plant (Figure 5.1). The period between consent for fore, care must be delivered in a timely, efficient, donation and organ procurement can be tumul- and standardized manner. The following sections will tuous, and it requires intense medical management specifically address donor management concepts in to avoid somatic donor loss. Up to 20% of poten- sequence, which ordinarily are addressed at the same tial donors are lost during this timeframe, and risk time.

57 Section 1: General

Cardiovascular management taken when administering large amounts of glucose- containing free water when treating hypernatremia, As previously discussed, brain death may result in because this may precipitate hyperglycemia, resulting severe cardiovascular perturbations as a consequence in osmotic diuresis, further hypovolemia, and elec- of cardiac dysfunction and loss of vasomotor tone. trolyte disorders. In these situations, insulin infusion Therefore, the central goal in donor management is to should be maintained to achieve target glucose levels maintain hemodynamic stability with good end-organ between 80 and 150 mg/dl. Also, care must be taken perfusion while maintaining euvolemia with minimal when administering hydroxyethyl starch, which has to no inotropic or pressor support. Generally, care been implicated in injury to renal tubular cells and occurs within the intensive care unit, where arterial early renal graft dysfunction. Given the high incidence and central venous pressure monitoring is available. of temperature dysregulation seen in donors, all fluids Continual reassessment is the rule, with targeted mean should be warmed to 37◦C. arterial pressure ≥60 mmHg, central venous pressure Euvolemic donors with hypotension require (CVP) 6–8 mmHg, and urine output ≥1 ml/kg/hr. vasoactive drug support. The choice between inotrope Echocardiographic assessment should be obtained andpressorisdependentoncardiacfunction.If in all potential heart donors. This provides valuable cardiac function is preserved and the donor is expe- information regarding heart function and structural riencing loss of vasomotor tone, then dopamine, abnormalities that may complicate donor manage- vasopressin, or phenylephrine can be used based ment and/or preclude heart donation. It is impor- on the specific clinical scenario. The use of any one tant to recognize that early poor cardiac function particular agent is not based on randomized clinical may improve with time and should not in and of trials, but rather on individual opinion, which varies itself preclude heart donation. Many centers, but par- widely. However, norepinephrine has traditionally ticularly in patients with depressed cardiac function been avoided, and vasopressin has traditionally been (ejection fraction Ͻ45%) or hemodynamic instability, the first choice in this clinical scenario. That being use pulmonary artery catheterization to assess ven- said, in most donors, homodynamic stability can be tricular loading conditions (ideally capillary wedge maintained with volume replacement and moderate pressure 8–12 mmHg, CVP 6–8 mmHg), cardiac con- doses of dopamine (0–5 ␮g/kg/min). Requirements tractile function (cardiac index ≥ 2.4 l/min/m2), and higher than 10 ␮g/kg/min may require the addition vasomotor tone (systemic vascular resistance 800– of additional vasoactive support and suggest serious 1200 dyne/s/cm5); see Figure 5.1.Thisaggressive cardiac dysfunction. approach has resulted in significant improvements There are no guidelines for specific combinations in hemodynamics, somatic donor salvage, and organ of vasoactive drugs. Clearly, combinations of drugs recovery rates. that work synergistically and decrease dosages of Hypotension is the most common clinical sce- each agent (thereby reducing the ␣-adrenergic effects) nario encountered, with incidence of up to 80% and are beneficial. With respect to kidney transplanta- the most common precipitator of cardiac arrest and tion, this strategy has resulted in decreased rates of somatic loss. Twenty percent of donors may remain acuterejectionandimprovedgraftsurvival.Cate- hypotensive despite vasoactive drug support, with this cholamine excess, as seen during sympathetic storm being more common in donors who are hypovolemic or when administered to experimental animals in or who have diabetes insipidus and who are not on very high concentrations, results in damage to the vasopressin. Therefore, it is imperative to continu- myocardium, as previously described. For this reason, ously reevaluate volume status, contractile function, many donors who are on high doses of catecholamines and vasomotor tone. are not used by cardiac surgeons. There is some evi- With low CVP and hypotension, volume should dence, however, that the combination of dopamine and be administered. Colloid or crystalloid should be used norepinephrine, even at high doses, has resulted in based on clinical scenario. For example, if the hema- excellent outcomes in heart recipients. Unfortunately, tocrit is less than 30%, then packed red blood cells there are also data that demonstrate increased inci- should be transfused. If the donor is hypernatremic dence of primary graft dysfunction, reduced right ven- and acidotic, bicarbonate buffered (50 mmol/l) half- tricular contractility, and reduced 1-year survival with normal saline should be administered. Care should be use of norepinephrine. Currently, there are no data

58 Chapter 5: Organ donor management and procurement conclusively demonstrating superiority of one vaso- the result of the myriad of pulmonary complications pressor over another. However, the dominant opin- that occur in this patient population, which include ion is the norepinephrine should be studiously avoided aspiration pneumonia, trauma, inflammatory medi- due to its ␤-adrenergic effects. ated acute lung injury, and neurogenic pulmonary In the event that hemodynamic stability cannot edema. However, standardized goal-directed therapy be achieved with volume optimization and combina- has improved conversion rates without jeopardizing tion vasopressor therapy, hormone replacement ther- other organs for recovery. apy should be initiated early. As previously discussed, Most guidelines advocate a lung-protective strat- the hypothalamic–pituitary axis becomes dysfunc- egy with judicious use of fluid for resuscitation and tional, resulting in depletion of ADH and insulin aggressive pulmonary toilet in order to optimize anddecreasesinthyroidhormoneandrelativecor- gas exchange and maintenance of adequate hemody- tisol levels. Early practice utilized a combination of namics. Low-pressure ventilation with end-inspiratory vasopressin, insulin, thyroid hormone, and steroids. pressures Ͻ30 cm of water, tidal volumes of 6–8 ml/kg, Today, insulin is begun prior to brain death and vaso- moderate amounts of positive end-expiratory pressure pressin is begun once diabetes insipidus is recog- (5–10 cm of water), and the lowest fraction of inspired nizedorasapressoragentforhypotension.Although oxygen possible that achieves PaO2 Ͼ 100 mmHg useofthyroidhormoneiscontroversial,inthesce- are recommended. Bronchoscopy, frequent suction- nario in which donor loss is inevitable, if hemody- ing, and recruitment maneuvers counteract mechan- namic stability is not achieved, thyroid hormone is ical reasons for poor gas exchange and have also been reasonable. Dramatic improvement in hemodynam- demonstrated to improve lung recovery. ics and acid–base status and improvement in the rate Early management of the somatic donor often of organ recovery have been demonstrated with T3 requires adjustments to minute ventilation in order infusion (4-␮g bolus followed by 3 ␮g/hr). Similarly, to normalize carbon dioxide levels (often resulting steroid administration (methylprednisolone 15 mg/kg in a rise in carbon dioxide), reduce barotrauma, bolus repeated every 24 hours) may enhance vascular and improve oxygen extraction (allow oxyhemoglobin reactivity and is associated with improved short- and curve to shift to the right). Furthermore, judicious long-term outcome for most transplanted organs. use of diuretics may be necessary to decrease CVP to Arrhythmias are common and difficult to treat. 6–8 mmHg and/or pulmonary capillary wedge pres- They can be a result of catecholamine surge during sure to 8–12 mmHg, as the lungs will be exquisitely herniation or initiation of pharmacological drug sup- sensitive to small hydrostatic pressure changes given port or as a terminal event after brain death 48– the inflammatory and neurogenic lung injury so com- 72 hours later. Atrial and ventricular tachyarrhythmias mon in brain death donors. Additionally, evidence can be treated with amiodarone. Bradyarrhythmias from animal and ex vivo studies suggests that inhala- do not respond to atropine and must be treated with tionofabetaagonistmayacceleratetherateofalveo- epinephrine or isoproterenol. Cardiac arrest is treated lar fluid clearance. Steroids abrogate the inflammatory as in non–brain-dead individuals with advanced car- effects of brain death and have been demonstrated to diac life support protocols because recovery of cardiac reduce fluid accumulation in the lung and be the most function will allow for organ donation. significant predictor of successful lung donation in multivariate analysis of lung procurement. All somatic donors should receive steroids as soon as possible after Respiratory management brain death. The medical optimization of respiratory function in brain death donors has not been rigorously evalu- ated by clinical trials. The medical suitability of lungs Organization of procurement for procurement lags significantly behind that of all After brain death has been certified and consent has other solid organs. Published results from one insti- been obtained, the recovery time is determined and tution comparing lung procurement with that of all procurement teams are notified. Aggressive donor care other solid organs using the same potential donor continues in the operating room. The donor is placed population demonstrated a conversion rate of 17% supine on the operating table with arms tucked and is and 70%, respectively. This huge disparity is mainly prepped from chin to knees. A midline incision from

59 Section 1: General

sternalnotchtopubisismade.Whatfollowsnextisa surgeon is notified of cross-clamp time and expected description of the thoracic recovery. time of arrival. The sternum is divided, the pericardium opened, and a pericardial cradle created. The heart and lungs are inspected and assessment of organ suitability is Abdominal organs communicated to the implanting surgeon. Mobiliza- Maintenance of cardiac output following confirmation tion of the superior vena cava (SVC), inferior vena of death allows a significant proportion of dissection cava (IVC), intra-atrial groove, aorta, and pulmonary to be performed prior to cold perfusion with preserva- artery is performed. Once the abdominal procure- tion solution (“warm dissection”). This has the distinct ment teams have completed their dissection, 30 000 advantage that important structures are much easier U of heparin is administered intravenously. Purse- to identify and can be prepared with greater precision, string sutures are placed in the ascending aorta and minimizing organ damage and shortening dissection pulmonary artery and perfusion cannulas are placed. following perfusion (“cold dissection”), during which The azygous vein is ligated and divided. Five hun- suboptimal cooling of the organs may occur. A mid- dred micrograms of prostaglandin E1 is injected into line laparotomy and sternotomy is performed and can the pulmonary artery. The SVC is ligated or clamped be extended using transverse abdominal incisions to distaltotheazygousvein.Theleftheartisvented maximize surgical access. A full exploratory laparo- through the left inferior pulmonary vein (or the left tomy is undertaken to exclude intra-abdominal sepsis atrial appendage if the lungs are being procured) and or malignancy, and any suspicious lesions are biopsied the IVC is vented at the pericardial reflection. The aor- and sent for urgent histological analysis. Early assess- tic cross clamp is applied and the heart and lungs are ment of renal and liver quality is performed, in partic- ◦ perfused with 4 C preservation solutions. Crushed ice ular the degree of steatosis and, in cases of donor death is placed over both the heart and lungs. Care is taken secondary to trauma, exclusion of liver injury. The pro- to ensure no dilatation of the heart occurs and the curement of the liver is discussed in detail in Chap- lungs blanche. After arrest has occurred, completion ter 22. Both kidneys are inspected for transplant suit- of the cardiectomy takes place. The SVC is divided ability; however, if only one kidney is required, the left distal to the azygous, taking care not to injure the is preferred because the renal vein is longer, making right pulmonary artery. Next, the IVC is divided at recipient transplantation easier to perform. Multiple the pericardial reflection. The aorta and pulmonary renal arteries exist in approximately 20% of potential arteries are then divided, with sufficient length for sub- donors, and removal of the kidney with a single renal sequent transplantation of both the heart and lungs. artery is preferred (cadaveric donor nephrectomy is The apex of the heart is then elevated, and midway discussed further in Chapter 29). between the coronary sinus and the pulmonary veins, the left atrium is divided. This division is carried around the left atrium, maintaining adequate cuff for Non–heart-beating donors both heart and lung teams. The heart is then removed Despiteeffortstoexpandthedonorpool,therestill andinspectedandpreservationsolutioninfusedinto remains a significant shortage of donor organs. In an the coronary arteries prior to placement in two cold effort to make more organs available for transplan- saline–filled bags and then a hard container of ice for tation, the transplantation community, government transport. regulators, and United Network for Organ Sharing Ventilation continues with room air. The lung have reintroduced policies for the donation of organs team then retrograde flushes the pulmonary vein with after cardiac death. Historically, organs were recovered the cold preservation fluid to remove any pulmonary from non–heart-beating donors, also termed donation emboli. Next the trachea is identified and divided and after cardiac death (DCD); however, the introduction stapled with the lungs at almost maximal inflation, 2–3 of the concept of brain death decreased enthusiasm for rings above the carina. Lastly, heavy scissors are used this population of donors, given the numerous uncer- to quickly divide all posterior mediastinal tissue just tainties involved. Recently, mounting pressure to make anterior to the esophagus. The lungs are then removed, more organs available has renewed interest in DCD inspected, and bagged for transport. The implanting (see Chapter 41).

60 Chapter 5: Organ donor management and procurement

Ex vivo perfusion the additional costs incurred in the static cold storage group for graft-related complications. In our current era, more organs are procured from As a result of the impressive results in kidney older, marginal, and non–heart-beating donors than preservation, interest has also increased in liver, pan- ever before. From 1988–2009, there has been a 630% creas, heart, and lung ex vivo preservation systems. increase in donors greater than 50 years of age and Studies with warm blood perfusion of beating donor an 800% increase in donors who experienced a car- hearts have shown promising early results, allowing diovascular cause of death. Given the increased use of longer distance procurement and potentially further higherriskdonorsandDCD,improvementsinorgan assessment of marginal hearts. Given the increas- preservation techniques are required to help expand ing pressure for more transplantable organs and the the number of suitable donors. extremely low conversion rates seen in lung trans- Today, most centers use static cold storage as the plantation, considerable research has been directed preferred organ preservation method. Static cold stor- in evaluating and optimizing donors prior to DBD age begins after cold perfusion of a preservative solu- and DCD. Early experiments in dogs demonstrated tion (varies by center and organ) to displace blood that lungs transplanted with 1 hour of non-ventilated components within the vasculature and to achieve a warm ischemic time could function well with good hypothermic equilibrium within the organ. In the early gas exchange. Additional studies by the same group 1960s, cooling alone was demonstrated to be an effec- demonstrated that up to 4 hours of ventilated warm tive preservation strategy. Metabolic rate is halved by ◦ ischemic time was tolerable. Encouraged by these each 10 C drop in temperature, reaching its nadir at ◦ results, since 1995, several clinical groups have trans- about 10% of normal at 4 C. Unfortunately, impaired + + planted DCD lungs from Maastricht category III activity of the Na /K ATPase results in passive donors using ex vivo lung perfusion (EVLP) rigs to first sodium entry to the cell and consequent cell swelling. assess the organ. Total published experience of cate- Cell swelling, acidosis secondary to metabolic shifts, gory III somatic donors to date is about 100 patients and generation of reactive oxygen species have led worldwide. to many developments in organ preservation solu- Interest in EVLP assessment and optimization tions. The preservation solutions currently available has continued to increase. Described systems allow for use vary with respect to impermeants, buffers, for visual evaluation, hemodynamic and ventilatory electrolyte compositions, reactive oxygen species scav- mechanics, and gas exchange to be followed over time. engers, and additives. Each solution has been opti- The circuit consists of a blood reservoir, a pump, a mized for specific organ use. The perfect universal leukocyte filter, a gas exchanger, an in-line blood gas preservation solution remains to be discovered and monitoring system, and a heat exchanger. Perfusion has therefore led many groups to investigate ex vivo cannulas are connected to the main pulmonary artery perfusion. andtheleftatrium,thetracheaisintubatedusinga Much of the conceptual basis for current practice standard endotracheal tube, and the lungs are placed originated from the Medical College of Georgia’s pio- in a sterile box. Perfusion is maintained at pressures neering work on hypothermic ex vivo renal perfu- ◦ no greater than 20 mmHg at 37 C with varying perfu- sion systems. In 2006, 20% of all kidneys were pre- sion solutions and additives. Successful normothermic served using a hypothermic ex vivo perfusion system. perfusion has been described up to 12 hours without In extended criteria kidney donors, static cold storage inducing edema and preserving stable pulmonary vas- hasbeendemonstratedtoincreasetheriskofgraftdys- cular resistance, gas exchange, and mechanics. function and loss. Given these limitations, a prospec- The University of Toronto has recently presented tive randomized controlled study was performed in their outcomes in more than 20 patients who under- 336 consecutive donors in which one kidney was pre- went lung transplant with donor lungs treated with served using the static cold storage techniques and the EVLP for 12 hours. Initial results showed that graft other kidney was preserved using a hypothermic ex function and short-term survival were not signifi- vivo perfusion system. At 1-year follow-up, the ex vivo cantly different from those of DBD. These early reports perfusion grafts experienced less delayed graft func- have increased enthusiasm for the introduction of tion and improved graft survival. Further, the upfront EVLP in an increasing number of centers to improve cost of the perfusion system was more than offset by utilization of marginal lung donors.

61 Section 1: General

These reports open the possibility of extended Jenkins DH, Reilly PM, Schwab C W. Improving the evaluation and repair of marginal donors and poten- approach to organ donation: a review. World J Sur 1999; tially further expanding the donor pool. Treatment of 23: 644–9. lungs during EVLP has great potential to significantly Smith M. Physiologic changes during brain stem improve graft function. It has been shown that treat- death–lessons for management of the organ ment of ex vivo human lungs with an adenoviral vector donor. JHeartLungTransplant2004; 23: S217– encoding for IL-10 decreased inflammatory cytokine 22. expression and led to significant improvements in Wood KE, Becker BN, McCartney JG, D’Alessandro AM, graft function. Clearly more research is needed, but Coursin DB. Care of the potential organ donor. NEnglJ there remains conservative optimism that many more Med 2004; 351: 2730–9. organs will be rehabilitated and transplanted. Yeung JC, Cypel M, Waddell TK, Van Raemdonck D, Keshavjee S. Update on donor assessment, resuscitation, Further reading and acceptance criteria, including novel Bugge JF. Brain death and its implications for management techniques–non-heart-beating donor lung retrieval and of the potential organ donor. Acta Anaesthesiol Scan ex vivo donor lung perfusion. Thorac Surg Clin 2009; 19: 2009; 53: 1239–50. 261–74.

62 Section 2 Heart Chapter Recipient selection

6 R.V. Venkateswaran and Jayan Parameshwar

Key points the recipient’s point of view, it could be argued that the organ should be allocated to the most ill patient. r Advances in the management of heart failure From the point of view of the wider community of HF have significantly improved the outcome of patients, society in general, and the donor family, it patients with this condition. r mightbearguedthattheorganisgiventotherecipi- Heart transplantation remains the best ent predicted to have the best long-term outcome after treatment for selected patients with advanced a transplant. In practice, a compromise between these heart failure. two positions is required. r The number of donor hearts available for transplantation continues to fall; careful assessment of potential candidates for Recipient selection transplantation is therefore crucial to ensure Patients with New York Heart Association (NYHA) optimal use of this scarce resource. class IIIB and class IV HF are best discussed with r Pulmonary vascular resistance is a the local heart failure/transplant center to opti- continuous variable for risk of an adverse mize medical management and to consider high-risk outcome after transplantation. non-transplant surgery where appropriate. Examples r Cardiopulmonary exercise testing with include valvular heart disease and reversible ischemia measurement of gas exchange is generally that could be managed by conventional surgery. It is thought to be the best method of assessing imperative that every effort is made to optimize med- functional capacity in the ambulant heart ical therapy. Particular attention must be paid to fluid failure patient. restriction and elimination of potential contributors such as excessive alcohol intake. Every attempt must be made to maximize the dose of prognostically beneficial Heart transplantation (HT) is the treatment of choice agents such as angiotensin-converting enzyme (ACE) for selected patients with advanced heart failure (HF). inhibitors and beta-blockers. Patients who may benefit Despite the success of the procedure and improve- from cardiac resynchronization therapy (CRT) should ments in long-term recipient outcome, only a small be identified and referred for bi-ventricular pacing. fraction of advanced HF patients can be treated by this Patients with chronic HF should be referred before modality. Unfortunately, the number of heart trans- they develop significant renal and hepatic dysfunc- plants performed worldwide every year is declining. tion and irreversible pulmonary hypertension. Finally, This trend is seen in Europe, particularly in the United it is important to continually reassess the response to Kingdom, whereas the number of hearts transplanted changes in therapy. in the United States remains fairly static (Figure 6.1). The increasing numbers of patients referred for trans- plantation has lead to a major imbalance between sup- Indications for heart transplantation ply and demand. Mortality on the HT waiting list is Patients with NYHA class IV heart failure despite also increasing. Allocation of a scarce resource (the best medical therapy will generally derive prognos- donor heart) requires two different perspectives. From tic benefit from heart transplantation. Patients with

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

63 Section 2: Heart

300

Donor 250 244 Transplants Patients on waiting list 203 208 200 176 181 180 165 166 180 158 158 154 157156 148 141 150 140 128 133 127 130 Number 135 100 106 111 105 106 110 88 95 93 50

0 1999- 2000- 2001- 2002- 2003- 2004- 2005- 2006- 2007- 2008- 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Ye a r

Figure 6.1 Deceased donor heart program in the United Kingdom, April 1, 1999–March 31, 2009. Number of donors, transplants, and patients on the active transplant list at March 31, 2009.

Table 6.1 Indications for heart transplantation Table 6.2 Absolute and relative contraindications to heart transplantation Heart failure Ongoing symptoms of HF at rest or minimal exertion despite optimal Active infection (including chronic active viral infections, e.g., medical therapy. Functional capacity HIV, HBV) measured by peak oxygen uptake on exercise less than 14 ml/kg/min (or Symptomatic peripheral or cerebrovascular disease 50% predicted peak VO2 normalized for Diabetes mellitus with end-organ damage (nephropathy, age and sex). For patients receiving neuropathy, proliferative retinopathy) beta-blockers, a value of 12 ml/kg/min has been recommended. Coexistent or recent neoplasm (within 5 years) Ͻ Recurrent admission History of recurrent admission to Severe lung disease (FEV1 and FVC 50% predicted) hospital with worsening HF despite Renal dysfunction: estimated glomerular filtration rate compliance with optimal medical Ͻ40 ml/min/1.73 m2 (consider combined heart and kidney therapy. transplantation) Ischemia Refractory ischemia not amenable to Recent pulmonary thromboembolism revascularization associated with severe impairment of left ventricular Pulmonary hypertension: pulmonary artery systolic function. pressure Ͼ60 mmHg, transpulmonary gradient ≥15 mmHg, and/or pulmonary vascular resistance Ͼ5 Wood units. Arrhythmia Recurrent symptomatic ventricular arrhythmia associated with severe Psychosocial factors including history of non-compliance impairment of ventricular function. with medication, inadequate social support, drug or alcohol abuse Obesity:bodymassindexϾ34 or weight Ͼ140% of ideal body NYHA class III heart failure need careful assessment weight of prognosis. Table 6.1 shows the usual indications for Age: usually Ͼ70 years heart transplantation; Table 6.2 outlines the usual con- traindications. Most of these are relative rather than FEV1: forced expiratory volume in 1 second; FVC: forced vital capacity. absolute contraindications.

Ventricular dysfunction function have well-controlled heart failure and do not Patients who require heart transplantation invariably require listing for a transplant. Although the sever- have severe ventricular dysfunction, but the converse ity of ventricular dysfunction and increasing ventric- is not true; many patients with severe ventricular dys- ular volume is a predictor of adverse outcome in HF,

64 Chapter 6: Recipient selection there is no cut-off level of ejection fraction (EF) or been suggested that in obese patients, a lean body mass left ventricular dimension that mandates listing for adjusted peak VO2 of Ͻ19 ml/kg/min is used to guide transplantation. listing.

Cardiopulmonary exercise testing Right heart catheterization Invasive right heart pressure measurement gives Exercise capacity is known to correlate with prognosis important information regarding the filling pressures in advanced HF. Assessment of this variable by history and cardiac output in patients with HF. It is also essen- is usually expressed in terms of NYHA class. An objec- tial to measure the transpulmonary pressure gradi- tive measure of functional capacity is essential to com- ent (TPG) and pulmonary vascular resistance (PVR). pare patients and to assess response to treatment. The Chronic HF is associated with a high left ventricular 6-minute walk test is a simple tool to assess exercise end-diastolic pressure (LVEDP), which in turn leads to capacity in advanced HF, but cardiopulmonary exer- pulmonary venous and pulmonary arterial hyperten- cise testing (CPEX) with measurement of gas exchange sion. This is usually reversible following transplanta- is generally thought to be the best method of assess- tion.However,acuterightventricular(RV)failureisan ing functional capacity in the ambulant HF patient. important cause of peri-operative mortality following This tool was introduced into the HF clinic in the heart transplantation (up to 20% of early deaths), and 1980s by Weber and Janicki and was rapidly adopted it was recognized early in the 1970s that a high PVR by transplant centers to guide listing for heart trans- was a predictor of this complication. The donor RV is plantation following a publication by Mancini and particularly susceptible to injury following brainstem colleagues in 1991. Patients with advanced HF were death and organ retrieval, and there is a complex inter- stratified based on peak VO levels: (1) peak VO 2 2 action between PVR and RV dysfunction. Ͻ 14 ml/kg/min and suitable for HT; (2) peak VO2 There is evidence that when HF therapy achieves Ͻ 14 ml/kg/min but with contraindications to HT; low filling (near normal) pressures with a cardiac index (3) peak VO Ͼ 14 ml/kg/min One-year survival 2 . Ͼ2 l/min/m2, medium-term prognosis is good enough was 94% for patients with peak VO ≥ 14 ml/kg/min 2 to defer listing for transplantation. PVR is a contin- compared with 74% for patients with peak VO 2 uous variable for risk of an adverse outcome after Ͻ 14 ml/kg/min who were suitable for HT. In patients transplantation, as discussed later. The exact levels of whowereunfitforHT,the1-yearsurvivalrateonmed- pulmonary artery (PA) pressure, TPG, and PVR that ical treatment was only 47%. Touse the measured peak preclude transplantation are uncertain. A reasonable VO for prognostic purposes, it is important to estab- 2 guide would be a combination of PA systolic pres- lish that the patient has achieved the anaerobic thresh- sure Ͼ60 mmHg, TPG Ͼ 15 mmHg, and fixed PVR old. In practice, a respiratory exchange ratio (RER) of above 5 Wood units (WU). Reversibility, as defined by Ͼ1.05 is used to define a maximal CPEX. the acute response to a pulmonary vasodilator, may be The widespread introduction of beta-blockade and used to refine selection of patients, but there is no con- implantable cardioverter-defibrillators (ICDs), which sensus on the best agent or the fall in PVR that pre- improve survival but have little effect on exercise dicts a satisfactory outcome. Longer term reversibility capacity, has prompted revision of the level of peak studies, e.g. with intra-aortic balloon pumping and a VO that is used to guide listing for transplant. There 2 phosphodiesterase inhibitor (e.g., enoximone or milri- have also been advances in the interpretation of CPEX. none), may be more useful in practice. It is important In the presence of a beta-blocker, it has been suggested to repeat right heart catheterization while patients wait that a level of 12 ml/kg/min is used. Normalizing peak for a suitable donor heart; this helps to identify those VO forageandgenderisparticularlyusefulforyoung 2 in whom the PVR is rising, but also those patients who patients and women. It may also be preferable to use improve on continued medical therapy and may there- alevelof50%predictedratherthananabsolutevalue fore no longer require transplantation. of peak VO2 to guide listing. When a patient is unable to perform a maximal CPEX, the ventilatory efficiency orminuteventilationtocarbondioxideproduction Role of HF prognostic scores relationship (VE/VCO2 slope) may be used. A slope Severalsurvivalscoreshavebeendevelopedtohelp Ͼ35 is used as a determinant for listing. Finally, it has define prognosis in HF. The best studied scoring

65 Section 2: Heart

system in the context of predicting the need for HT immunological activity that occurs with aging. is the Heart Failure Survival Score (HFSS). HFSS This immune senescence has also been linked with includes CPEX in addition to serum sodium, rest- increased risk of development of opportunistic infec- ing heart rate, ischemic etiology, QRS duration, EF, tions and post-transplant malignancies. In patients andmeanarterialbloodpressure.Thescoreiscal- who are older than 65 years, the decision regarding culated as the absolute value of the sum of prod- listing should be individualized; careful attention to ucts of the prognostic variables and their computed comorbidity, including renal function, pulmonary coefficients. Based on final HFSS, the patients are function, and general fitness, helps to define a group graded as low risk (≥8.10), medium risk (7.20–8.09), with acceptable outcome. and high-risk (Ͻ7.20). This score was developed and validated in patients undergoing transplant evalua- tion and has been extensively re-validated. On con- Malignancy tinued medical therapy, patients in the three risk Patients with a recent history of malignancy (Ͻ5years) strata had a 1-year survival of 88%, 60%, and 35%, are generally not considered suitable for HT. Potential respectively. However, the HFSS was developed before recipients with some hematological malignancies in widespread use of CRT, ICDs, and beta-blockers. remission with very low rate of recurrence may be con- Introduction of these treatments has altered the prog- sidered for HT within shorter time periods. However, nostic variables used in computing the score. Nev- each patient should be assessed carefully, and discus- ertheless, a number of studies have evaluated the sions should involve oncologists to establish the long- role of HFSS in the current therapeutic era and term prognosis. If the expected survival for patients confirmed the validity of the score in assessing with the malignancy in question exceeds the median prognosis. survival following heart transplantation, it may not The Seattle HF Score may be easier to use in the be considered a contraindication. The possible effect heart failure clinic and can guide the physician in refer- of immunosuppression on the malignancy also needs ring a patient to a transplant center at the appropriate to be taken into account. Low-grade tumors such as time. It is a 21-variable model derived from data col- prostate carcinoma may not be a contraindication lected during the PRAISE (Prospective Randomized to HT. Amlodipine Survival Evaluation) study and validated on subsequent clinical trial data. Five of the seven vari- ables of the HFSS are included in this model (heart rate and peak VO2 are not). There is relatively good concor- Renal dysfunction dance between the two models. Renal dysfunction is common in patients with advancedHF.Itisamarkerofadverseoutcome Comorbidity (relative in patients on optimal medical therapy; severe renal dysfunction also affects outcome after HT. contraindications) Serum creatinine will often rise in the immediate post-transplant period, in part due to the effects of Age cardiopulmonary bypass. Hemodynamic instability The long-term outcome following HT has been shown and the introduction of a calcineurin inhibitor com- to be worse with increased recipient age (discussed fur- pound this problem. If renal size is normal and there is ther in Chapter 12). However, chronological age alone no proteinuria, an estimated glomerular filtration rate should not be considered an absolute contraindication of above 40 ml/min/1.73m2 is considered adequate to to transplant evaluation. Patients greater than 70 years list patients for HT. Occasionally a trial of IV inotropic of age have generally been reported to have a worse therapy to increase cardiac output is warranted; if survival following transplantation. Older patients run there is a substantial fall in the serum creatinine, the a higher risk of post-transplant malignancy and renal patientcanbelistedforHT.Thedurationofabnormal dysfunction as compared with younger recipients. renal function may also be an important variable. In contrast, the incidence of rejection is lower in There may be an indication in individually assessed older recipients. This is perhaps due to reduced cases for combined heart and kidney transplantation.

66 Chapter 6: Recipient selection

Obesity and complex anatomy. Despite these difficulties, long- term outcome for patients who survive the HT oper- Patients with obesity (Ͼ140% of ideal body weight or ation is excellent. Although these patients should not body mass index Ͼ3.5) have increased morbidity and be denied transplant evaluation, their risk should be mortality in the medium term. The Cardiac Transplant carefully assessed and consideration given to concen- Research Database identified this variable as a risk trating expertise in a few centers. factor for infection and death. Ideally, such patients should be strongly advised and helped to lose weight before listing for transplantation. Smoking and alcohol abuse Smokers have a decreased survival following HT, which in one study was apparent at 4 years. They have Chronic viral infection a higher incidence of cardiac allograft vasculopathy Infection with human immunodeficiency virus (HIV), (CAV) and malignancy than non-smokers. Smoking hepatitis B virus (HBV), and hepatitis C virus (HCV) cessation, ideally for 6 months, before transplantation has generally been considered a contraindication to is required by many programs. A recent history of alco- heart transplantation. The reasons include viral multi- holism or drug abuse is usually considered an abso- plication in the face of immunosuppression, the poten- lute contraindication to transplantation because of the tial for dysfunction of other affected organs (liver and association with non-compliance in such patients. kidney), toxicity of antiviral drugs, and the possible association of viruses (including cytomegalovirus and hepatitis viruses) in the acceleration of coronary artery HLA tissue typing and sensitization disease. The advent of effective therapy for HBV and The most important alloantibodies are directed against HIV has led to this position being challenged by sev- human leukocyte antigen (HLA) and are formed in eral groups who hold that the outcome in selected response to foreign HLA molecules, often at the time patients is no worse than that of the general popula- of blood product transfusion, previous transplanta- tion of HF patients. Data on long-term outcome in any tion, or pregnancy. All potential recipients require tis- of these groups are lacking (particularly in the con- sue typing to identify any preformed HLA antibod- text of HT), but there are several reports of accept- ies, which would result in hyperacute rejection at the able short- and medium-term outcome in small num- time of transplant and should therefore be avoided. bers of patients. Although there is no consensus on the Potential recipients with very high levels of antibody best approach, HIV patients who have normal CD4 will inevitably wait longer for a compatible donor counts and no evidence of the acquired immune defi- heart and may therefore never undergo transplanta- ciency syndrome (AIDS)–related complex may be con- tion, even in the context of a separate national “sensi- sidered for transplantation. Similarly, patients who are tized patient” list. Recipients for heart transplants are HBV surface antigen–positive but have no detectable not prospectively HLA matched with donors as would viral load by polymerase chain reaction may have an be performed for renal transplantation. This is due acceptable outcome after transplantation. In addition, to concerns about smaller numbers of donor hearts, some centers advocate the use of hearts from HCV- ischemic time, and clinical urgency of recipients. Out- positive donors for transplantation into HCV-positive comes transplanting across HLA-A, -B, or -DR appear recipients. to improve with increasing matching (risk reduction of 1 or 2 matches, 0.8 and for 4–6 matches, 0.6), although the DR mismatch appears to have the great- Congenital heart disease est short- and long-term impact on outcome. In the Growing populations of young adults who have sur- presence of HLA antibodies, traditionally, prospective vived palliative surgery for congenital heart disease are complement-dependant cross-match and flow cytom- likely to be referred for transplant evaluation in the etry cross-match would be performed by the tissue- next 10 to 20 years. The operative mortality and 1- typing laboratory prior to proceeding with a heart year survival for these patients is inferior to that of transplant. In many centers, this has been replaced other recipients. This is due to the operative difficul- by virtual cross-match, which allows prediction of a ties caused by adhesions, increased risk of bleeding, significant antibody-mediated immune event between

67 Section 2: Heart

donor heart and recipient and can thus be used in portion of donors will also have a long wait and, in pretransplant risk assessment and organ allocation. some cases, are unlikely to receive a donor heart. These Desensitization of presensitized patients with donor- factors must be taken into account when the decision reactive HLA to reduce antibody levels to levels that to list for transplant is considered. They may also drive make transplantation safe have been reported prior earlier use of mechanical circulatory support at a time to heart transplantation (discussed in Chapter 33). when end-organ function is preserved and the risk of Evidenced-based protocols and outcome data remain surgery is lower. to be defined. Re-transplantation Combined heart–kidney and Early re-transplantation in patients with multiple heart–liver transplantation episodes of acute rejection has been associated with a Combined heart–kidney transplantation with allo- poor outcome. In selected patients with CAV and HF, grafts from the same donor is regularly but rarely late post-transplant acceptable outcomes have been performed (1.4% of the total heart and kidney trans- achieved. However, in general, survival is worse than plantation experiences). Since the initial report of in patients receiving their first transplant. Allosensi- simultaneous heart–kidney transplantation with allo- tization complicates donor organ allocation in many grafts from the same donor in 1978, multiple solid of these patients. In an era of decreasing donor organ transplantation has evolved into an acceptable organs, there is also the ethical issue of whether approach, with satisfactory short and intermediate re-transplantationisjustified.Incarefullyselected results. In carefully selected patients, long-term sur- patients,wefeelitisreasonabletoconsiderre- vival is similar to that of separate heart and kid- transplantation, but patients must be made aware at ney transplantation, although rejection can occur in initial evaluation that, for most, this will not be an both allografts individually. No HLA matching is per- option. formed in heart or combined heart–kidney transplan- tation. Combined heart–liver transplantation has been Bridging to transplant with increasingly performed, but data on patient and graft mechanical circulatory support outcomes remain limited. A recent study described 47 = The percentage of donor hearts implanted in patients cases of combined heart–liver (n 41) and heart– = whoareclassedasin“urgent”needofatransplant liver–kidney transplantation (n 6) reported to the (usually patients who are inotrope dependent in crit- United Network for Organ Sharing registry over an ical care units) is rising as the overall number of 18-year period. Indications include cardiac disease donor organs has declined. These patients are at risk with cardiac cirrhosis, amyloidosis, and hemochro- of sudden deterioration and have a high short-term matosis. One-year patient, heart, and liver graft sur- mortality rate. Mechanical circulatory support may be vival rates were 84.8%, 84.8%, and 82.4% at 1 year and needed in many of these patients to bridge them to a 75.6%, 75.6%, and 73.5% at 5 years, respectively. Com- heart transplant. Assessment of the appropriateness of bined heart–liver transplantation is a viable option, such therapy should ideally be carried out at the same with outcomes comparable to those of single-organ time as transplant evaluation (Discussed further in recipients. Chapter 8). Further reading Estimated waiting time for Anyanwu AC, Rogers CA, Murday AJ. Intrathoracic organ a transplant transplantation in the United Kingdom 1995–99: results from the UK cardiothoracic transplant audit. Heart The estimated waiting time for patients after listing 2002; 87: 449–54. depends on blood group, allosensitization, and patient Kirklin J, Banner NR, Jessup M (eds). Advanced Heart height and weight. In general, patients who carry Failure. ISHLT monograph series,Vol.3,Part1&2. blood group O and who are of large build have a longer Addison, TX: International Society for Heart and Lung wait. Patients who are sensitized to a significant pro- Transplantation, 2009.

68 Chapter 6: Recipient selection

Mancini D, Lietz K. Selection of cardiac transplantation cardiac transplant candidate. JHeartLungTransplant candidates in 2010. Circulation 2010; 122: 2006; 25: 1024–42. 173–83. Taylor DO, Stehlik J, Edwards LB, et al. Registry of the Mehra MR, Kobashigawa J, Starling R, et al. Listing criteria international society for heart and lung transplantation: for heart transplantation: International Society for Heart twenty-sixth official adult heart transplant report. J and Lung Transplantation guidelines for the care of Heart Lung Transplant 2009; 28: 1007–22.

69 Section 2 Heart Chapter Donor heart selection

7 KiranK.KhushandJonathanG.Zaroff

Key points Donor selection r Availability of donor organs limits heart Absolute contraindications transplantation rates today. Donor sepsis and severe infections r Judicious liberalization of graft acceptance criteria would increase transplant rates. Patients expiring from overwhelming infection have r The risks of transplantation with a marginal traditionally been excluded from donor evaluation due graft must be weighed against the risk of to potential transmission of pathogens (discussed fur- remaining on the transplant waiting list. ther in Chapter 4C). A retrospective study of 599 donorswhoseorganswereusedfortransplantation r Many criteria previously thought to be identified 46 donors (7.5%) with positive blood cul- “absolute” contraindications to transplantation are now considered “relative” tures and 25 donors (4.5%) with positive urine cul- tures. A total of 179 patients received organs from contraindications. these contaminated donors, including 11 heart trans- r Marginal donor hearts may be safely utilized plant recipients. Three of the heart recipients grew for high-risk transplant recipients who would organisms in the postoperative period that were sim- not otherwise be suitable candidates for heart ilar to those found in the corresponding donors; transplantation. however, no patient suffered significant morbidity and mortality from these infections. These data sug- gest that in the era of rapid and accurate identi- Advances in surgical techniques, postoperative care, fication of pathogens in the transplant donor and and immunosuppression have led to greatly improved effective antibiotic therapy, contamination of organs survival following cardiac transplantation in the past should not be an absolute contraindication to graft two decades. The success of heart transplantation, utilization. however, has led to a large disparity between the num- ber of available donor organs and the number of patients on the waiting list. Thus wait-list mortality Donor hepatitis B and C virus infection remains unacceptably high, at up to 30 deaths per 100 Hearts from donors infected with hepatitis C virus patient-years for critically ill (status 1A) recipients. (HCV) carry a substantial risk of transmission to the Numerous proposals have therefore been put forward recipient, as evidenced by a report of five HCV-naive to liberalize donor organ acceptance criteria. Although recipients who were transplanted with hearts from use of “extended-criteria donors” presents some risk HCV-positive donors, of whom four developed to the recipient, this risk must be balanced against evidence of infection. A subsequent study examined the risk of remaining on the waiting list. Efforts to outcomes of 261 heart transplants involving an HCV- increase donor heart utilization are therefore needed, positive donor. This analysis revealed higher 1-, 5-, but caution must be exercised, as early allograft failure and 10-year mortality in recipients of HCV-positive accounts for up to 25% of deaths in heart transplant donor hearts. These transplant recipients were more recipients. likely to die of liver disease and cardiac allograft

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

70 Chapter 7: Donor heart selection vasculopathy (CAV), suggesting a viral trigger for this data suggesting that donors with subcapsular renal cell common post-transplant complication. carcinomas may be considered. Given these precau- This discovery led to interest in other viral infec- tions, the current risk of donor-related tumor trans- tions, including hepatitis B virus (HBV). An analysis mission appears to be extremely low. These data sup- of coronary angiograms from 13 patients in whom the port a careful risk/benefit assessment by the transplant heart transplant (HT) donor or recipient was positive team, reviewing donor cancer history and the recip- for HBV also revealed a significant increase in CAV ient’s urgency of need for transplantation in order to at 1 year post transplant (31% versus 5% in control safely expand the donor pool. patients), suggesting that HBV may be a trigger for the development of CAV. Despite the possible association Donors exposed to myocardial toxins with CAV, there is currently no evidence that recip- Carbon monoxide (CO) poisoning causes pathological ients of cardiac allografts from HBV-positive donors myocardial changes from tissue hypoxia. Use of car- have reduced survival. Currently, grafts from HBV diac allografts from CO-poisoned donors can result in core antibody–positive donors may be safely accepted early recipient deaths due to graft failure. Since adop- for heart transplantation into HBV-negative recipi- tion of standardized donor management protocols, ents, followed by post-transplant prophylactic lamivu- longer donor management periods have allowed ade- dine to prevent viral transmission. quate time to evaluate and reject irreversibly damaged organs. Several centers have since published results of Donor human immunodeficiency virus (HIV) infection small case series (n = 5–7) of heart transplantation There is a paucity of data regarding transmission of with organs from CO-poisoned donors, demonstrat- the HIV virus in the setting of heart transplantation. A ing acceptable recipient outcomes. Given the limited 1991 multi-center review of 88 HIV-infected recipients data available, caution is warranted when evaluating of solid organ transplants in whom HIV was presum- potential grafts from CO-poisoned donors. Ideally, the ably transmitted via organs or peri-operative blood donor should be hemodynamically stable for at least transfusions showed that 28% of the recipients devel- 36 hours from the time of poisoning in order to select oped acquired immune deficiency syndrome (AIDS), organs with a low risk of graft failure. and 80% of these patients died of AIDS-related com- Case reports have described the transplantation of plications. Another 10% of recipients developed HIV- hearts from donors poisoned with tricyclic antidepres- related diseases. In the majority of cases, transplan- sants with satisfactory graft function. Successful heart tation occurred before routine donor screening for transplantation has also been reported after donor HIV antibody began. With current screening prac- death from poisoning with barbiturates, cyanide, tices, HIV transmission by transplantation is rare, and methanol, serotonin antagonists, and 3,4-methylene- donor HIV seropositivity is an absolute contraindica- dioxymethamphetamine (“Ecstasy”). Given the rela- tion to graft utilization. tively small number of cases, consensus guidelines are unlikely to be proposed for evaluation of donors Donors with extracranial malignancies exposed to myocardial toxins. Nevertheless, it is rea- Studies of donor-related tumor transmission to trans- sonabletoconcludethatheartsfromthesedonors plant recipients usually distinguish between central should not be summarily rejected (Figure 7.1). nervous system (CNS) and non-CNS donor malig- nancies. The risk of cancer transmission in HT from Relative contraindications donors with a CNS malignancy appears to be very low, provided that there are no detectable metastases. Advanced donor age Non-CNS malignancies, however, do have a significant The acceptance cut-off for donor age has increased risk of transmission from donor to recipient in HT significantly in the setting of the donor organ short- as in other organs, with multiple reports of transmis- age. In the early days of HT, the upper limit for donor sion of melanoma, choriocarcinoma, and cancers of age was 35 years. However, older donors are now used the lung, kidney, and breast. Currently, most surgeons frequently, with hearts from donors aged 50–55 years will accept grafts from non-skin and non-CNS cancer being accepted at many centers, especially for older donors with a 5–10-year “disease free” interval, using recipients. Donors ≥ 50 years of age now account for this time period as evidence for “cure.” There are also approximately 12% of all cardiac donors. The effect

71 Section 2: Heart

Absolute Relative Safe Figure 7.1 Donor contraindications to graft utilization: a changing paradigm. ∗Consider for HCV-positive recipients. Traditional criteria ∗∗ Active infection Assuming other cardiovascular factors HIV Donor age < 40 years Extracranial malignancy are favorable. HIV: human immunodeficiency virus; LV: left Hepatitis C* Normal cardiac anatomy Myocardial toxins ventricular. Normal LV function Advanced age West Nile Virus No infection Undersized heart No coronary artery disease LV hypertrophy Size matched to recipient LV dysfunction New criteria Prolonged ischemictime Hepatitis B Coronary artery disease Non-traumatic death Donor substance abuse Elevated Troponin l**

of donor age liberalization on post-transplant out- that undersized donor hearts can be safely utilized comes has been debated, and the International Soci- for heart transplantation, even in recipients with high ety for Heart and Lung Transplantation Registry has pulmonary vascular resistance, provided that post- reported increased recipient mortality (discussed fur- operative care is tailored to maximize cardiac output ther in Chapter 12). An analysis of recipient survival at and reduce the risk of acute RV failure. As is often the Columbia University transplant program showed the case, the decision regarding whether to accept an a 30-day mortality rate of 5% in recipients of hearts undersized donor heart must be informed by individ- from donors less than 40 years of age, 13% in recipi- ual donor and recipient characteristics. ents of hearts from donors 40–50 years of age, and 22% in recipients of hearts from donors more than 50 years Donor left ventricular hypertrophy ofage.Afterthefirst30days,however,theydidnot Recent case series report a 15–30% prevalence of find any association between advanced donor age and left ventricular hypertrophy (LVH) in donor hearts recipient mortality. These early mortality data must be accepted for transplantation, suggesting that this is a interpreted in light of the significant risk of death while fairly common finding. It is unclear, however, whether awaiting heart transplantation. A risk/benefit analysis increased left ventricular (LV) wall thickness detected performed by the same authors concluded that expan- during the donor evaluation period represents tran- sion of the donor age to greater than 40 years is more sient myocardial edema in the setting of brain death, or beneficial than indefinitely remaining on the waiting actual myocyte hypertrophy. Furthermore, the impact list, especially for critically ill patients. of donor LVH on graft and recipient outcomes remains in question. Initial reports suggesting that donor LVH Undersized donor hearts portends reduced recipient survival included very few Traditional rules dictate that cardiac allografts should donor hearts with LVH (n = 6–9), and LVH was fre- only be accepted from donors within 20–30% of the quently diagnosed by electrocardiography or quali- recipient’s weight. However, undersized donor hearts tative interpretation of echocardiograms. Later series have been used successfully with excellent long-term studied larger numbers of donors (n = 47–62) and outcomes. In a prospective study of 14 undersized used echocardiographic measurements, but defini- donor hearts, investigators demonstrated an increase tions of LVH varied (e.g., LV septal wall thickness Ͼ in left ventricular mass and internal dimensions post 1.1–1.4 cm). Thus it is not surprising that results transplant, suggesting that the donor left ventricle vary widely between studies. In donors with a history adapts to the larger recipient circulation. of hypertension, LVH had an unfavorable impact on One area of concern lies in transplanting under- late survival, suggesting that those patients had more sized hearts into recipients with preoperative pul- advanced changes in myocardial structure and dias- monary hypertension, due to the risk of acute right tolic function. Finally, several studies have reported ventricle (RV) failure. Several case series demonstrate an interaction between donor LVH and prolonged

72 Chapter 7: Donor heart selection ischemic time, which in combination portend poor Prolonged ischemic time recipient outcomes. Our need to transfer organs between centers has spurred innovations that allow for longer cold ischemic time (CIT). Current practice generally uses Donor left ventricular dysfunction aCITthresholdof4hoursforcardiacallografts. LV dysfunction is the most frequently cited reason for Data from the Collaborative Transplant Study, which non-utilization of potential cardiac allografts. A review included 104 centers in 24 countries, demonstrated of non-transplanted donor hearts managed by the Cal- a significant survival difference at 3 years between ifornia Transplant Donor Network demonstrated that hearts preserved for ≤ 4 hours (73%, n = 7589) LV dysfunction was the cause for graft refusal in 26% compared with those preserved for more than 4 hours of cases. (65%, n = 630) (p Ͻ 0.001). In an analysis of the The pathogenesis of brain death–induced LV dys- United Network for Organ Sharing (UNOS) database, function is complex, but acute catecholamine toxicity the authors found a significant interaction, with a likely plays a central role. Regardless of the underly- greater tolerance for prolonged ischemic times among ing mechanism, there is evidence that brain death– grafts from younger donors, especially those less than induced cardiac dysfunction is highly reversible. Serial 19 years of age. In aggregate, these data support an echocardiograms performed on 13 brain dead organ extension of the CIT to more than 4 hours for young donors with an initial LV ejection fraction of less than donors with excellent graft function. In older donors, 50% demonstrated an improvement in LV function in efforts should be made to minimize the CIT, especially 12 of the 13 donors after extended donor management. given the concern that endothelial injury during Hearts from these donors were subsequently accepted the ischemic period could predispose grafts to the for transplantation, with a recipient survival rate of development of CAV. Newer, improved preservation 92% at 16 months. solutions and continuous organ perfusion may allow Encouragingly, single-center studies have reported further extension of ischemic times. good results after transplantation of donor hearts with LV dysfunction. Although an analysis of risk factors Donor coronary artery disease affecting recipient survival in a multi-institutional With the increasing acceptance of older donor hearts study found that diffuse echocardiographic wall for transplantation comes the likelihood that a sig- motion abnormalities are an independent risk factor, nificant number of these hearts will have coro- with 22% recipient mortality in the early post- nary atherosclerosis. Several case series have been transplant period, this study was performed before published describing coronary artery bypass graft- the era of extended donor management and tailored ing performed at the time of the transplant surgery, hemodynamic protocols. More recent evidence has demonstrating that the surgery is technically feasi- demonstrated the reversibility of brain death–induced ble. However, there appears to be a high risk of early LV dysfunction, suggesting that these hearts may be graft failure. The University of California, Los Angeles acceptable for transplantation. (UCLA) heart transplant group demonstrated that the The debate regarding whether to accept a graft with prevalence of CAV at 3 years post transplant appears LV dysfunction for heart transplantation is likely to to be higher in recipients of hearts with donor coro- continue. Evaluation of donor hearts with depressed nary artery disease than in recipients without donor function is difficult, because none of the commonly atherosclerosis (25% versus 4%; p Ͻ 0.001). CAV did used tests, such as electrocardiography, echocardiog- not appear to preferentially develop at sites of pre- raphy, and coronary angiography, distinguish between existing donor lesions, but rather in a diffuse distribu- reversible ischemic injury and irreversibly damaged tion, perhaps secondary to coronary endothelial dys- tissue. Such cases are best evaluated in the context of function in older donor hearts. This observation has donor management guidelines discussed in Chapter been reported by other centers, suggesting that the use 5. Furthermore, extended donor management times of donor hearts with coronary atherosclerosis is associ- enable organ procurement organizations to assess for ated with development of CAV in the transplant recip- reversibility of graft dysfunction, such as with serial ient;however,itisunclearwhetherthistranslatesto echocardiograms, and may therefore improve graft significant differences in quality of life, graft function, utilization rates and recipient outcomes. or survival.

73 Section 2: Heart

Elevated donor troponin (Tn) levels performed in donors who die from aICB or trau- The lack of echocardiography and angiography at matic brain injury. Innovative cardioprotective strate- many hospitals where donors are evaluated under- gies aimed at preventing graft dysfunction in this set- scores the need for simpler methods for expedited and ting, such as the use of short-acting anti-hypertensive accurate evaluation. The use of serum markers that medications and CO-releasing preservation solutions, reliably predict graft function and survival is there- are needed. fore attractive as a simple, non-invasive screening tool. Circulating cardiac Tn levels are highly sensitive and specific markers of myocardial cell injury and are Donor substance abuse therefore often measured during organ donor assess- There is widespread use of substances such as cigarettes ment; however, their utility in donor selection remains and alcohol in the donor pool. Several retrospective controversial. Riou and colleagues initially observed studies have attempted to determine whether donor that an elevated cardiac troponin T level was associ- substance abuse is associated with adverse recipient ated with a severe decrease in LV function in brain- outcomes. dead patients and expressed concern that this finding Significant alcohol consumptionϾ ( equivalent of could represent irreversible myocardial cell damage. 60 ml 100% alcohol daily for ≥3 months) is present Small-scale retrospective and prospective studies in in 15–25% of the donor pool, and concern exists due heart transplantation have yielded conflicting results; to the association between heavy alcohol consump- although most confirm that elevated Tn levels are asso- tion and cardiomyopathy. However, in a relatively large ciated with graft LV dysfunction, some show an asso- study of 437 heart transplants, De La Zerda noted the ciation with early graft failure, whereas others failed opposite effect: chronic alcoholism in donors appeared to find an association between elevated donor Tn lev- to be protective, with higher recipient survival. The els and recipient mortality. All studies had major lim- data regarding cocaine use is similarly contradictory. itations, and well-designed prospective studies with High catecholamine levels in the setting of cocaine use serial, standardized Tn assays are needed to determine may cause coronary and systemic vasoconstriction, whether there is a Tn threshold level above which grafts myocardial microinfarction, increases in heart rate, may be unsuitable for transplantation. systemic arterial pressure, and myocardial demand. Great concern therefore exists regarding the use of car- diac grafts from cocaine-abusing donors. A recently Donor cause of death published analysis of the UNOS Registry, however, is Experimental studies in animals have shown that reassuring: In 9217 adult heart transplant recipients, acute increases in intracranial pressure as a result the incidence of donor cocaine use was 10%, with of traumatic brain injury or atraumatic intracra- no difference in mortality or development of angio- nial bleeds (aICB) cause a dramatic upregulation of graphic CAV at 1 and 5 years. pro-inflammatory cascades and an exaggerated cate- Donor cigarette smoking is a cause for concern cholamine response compared with other modes of given the link between smoking and endothelial dys- brain death. The combination of this catecholamine function, atherosclerosis, and other pathological pro- “storm” and upregulation of pro-inflammatory medi- cesses. This is especially concerning given the high ators is associated with hemodynamic impairment prevalence of smoking in the donor pool, with esti- and persistent myocardial dysfunction. Several studies mates of up to 76%. Of interest is a prospective study have suggested that donor brain death caused by aICB by Rickenbacher in which coronary angiography with is a potential risk factor for early recipient mortality. intravascular ultrasound was performed to identify ArecentanalysisoftheUNOSdatabasereportedthat donor factors that predispose recipients to CAV. These recipients of donors who died from cerebrovascular investigators found a significant association (p Ͻ 0.02) accidents have mildly worse survival when compared between donor smoking history and coronary artery with recipients whose donors died from head trauma intimal thickening within the first year after transplan- (relative risk = 1.1; 95% confidence interval, 1.01– tation, suggesting that early intimal thickening reflects 1.12). Given the current donor organ shortage, mode a predisposition to myointimal proliferation mediated of death should not preclude grafts from acceptance; by smoking, which is then accelerated following the rather, careful evaluation of graft function should be initiation of the alloimmune response.

74 Chapter 7: Donor heart selection

Non-standard donors and the to quantify the impact of a single donor risk factor on recipient survival, because few randomized trials alternate list with higher-risk donor organs have been conducted, Due to the severe donor organ shortage, with long due to ethical and practical considerations. Similarly, recipient waiting times, “non-standard” or “marginal” the combination of donor and recipient features may donor hearts are increasingly being used for higher also play an important role. Moreover, the decision risk recipients and critically ill patients, leading to about whether or not to proceed with transplantation an expansion of both the donor and recipient pools. must be weighed against the risk of remaining on the The UCLA group first reported experience with such transplant waiting list. Given these considerations, it is a listing policy. In their scheme, recipients who oth- encouraging that use of non-standard donors appears erwisewouldhavebeenexcludedfortransplanta- to be safe and provides acceptable long-term results. tion (due to advanced age, renal insufficiency, re- transplantation, or peripheral vascular disease) were Further reading matched with non-standard donor hearts that other- Ardehali A, Bresson J, Fonarow GC, et al. Use of two wise would not have been used for transplantation recipient lists for adults requiring heart transplantation. (due to donor coronary artery disease, illicit drug use, JThoracCardiovascSurg2003; 125: 49–59. HBV, HCV, LV dysfunction, high inotropic require- Armstrong WF, Babcock WD, D’Alessandro A, et al. Ͼ ment, LVH, age 55 years, small size, or a reused Consensus conference report: maximizing the use of transplanted heart). Although early recipient mortality organs recovered from the cadaver donor: cardiac was higher in this alternate group than among patients recommendations, March 28–29, 2001, Crystal City, Va. on the standard transplant list, conditional mid-term Circulation 2002; 106: 836–41. outcomes were comparable. Recently, a review of Aurora P, Christie JD, Dobbels F, et al. Registry of the all heart transplants reported to UNOS from 1999– International Society for Heart and Lung 2005 (n = 13 024, including 347 alternate list trans- Transplantation: Twenty-sixth Official Adult Heart plants) revealed higher morbidity and mortality in Transplant Report – 2009. JHeartLungTransplant2009; 28: 1007–22. the alternate-list group, with a 5-year survival rate of 51.4% compared with 75.1% in the standard transplant Edwards NM, John R, Lietz K, et al. Outcomes in cardiac transplant recipients using allografts from older donors group. Even with worse outcomes, however, this strat- versus mortality on the transplant waiting list: egyoffersamediansurvivalof5.2yearstopatients Implications for donor selection criteria. JAmColl with end-stage heart disease who otherwise would be Cardiol 2004; 9: 1553–61. expected to live for less than 1 year. Fishman JA, Grossi PA. Donor-derived infections in solid organ transplant recipients. Am J Transplant 2009; 9: Conclusion S19–26. Large SR, Oduro A, Potter CD, et al. Transforming the Liberalizing donor criteria has already saved or at least “unacceptable” donor: outcomes from the adoption of a extended the lives of many individuals who would oth- standardized donor management technique. JHeart erwise have died while awaiting organs. It is difficult Lung Transplant 1995; 14: 734–42.

75 Section 2 Heart Chapter Ventricular assist devices

8 DavidG.HealyandStevenS.L.Tsui

Key points Mechanical circulatory support devices (MCSD) include the use of extra corporeal circulatory support, r Ventricular assist devices (VAD) are implantable ventricular assist devices, and total artifi- increasingly used in the management of acute cial hearts. These devices are increasingly used in the cardiogenic shock to support the circulation management of acute cardiogenic shock to support the until either myocardial recovery (bridge to circulation until either myocardial recovery (bridge to recovery) or a more definitive treatment is recovery) or a more definitive treatment is decided decided upon (bridge to decision). upon (bridge to decision). In patients who are decom- r The type of device used is influenced by the pensating from chronic heart failure, MCSD is used acuteness of the patient and the projected as a bridge to transplantation or as a permanent form duration of support required. of therapy in those with contraindications for heart r Preoperative preparation should focus on transplantation (HT). For the purposes of this chapter, optimizing end-organ function and right weconfineourfocusontheroleofventricularassist ventricular function. devices (VAD) in HT. r Overall survival with left ventricular assist device placement (not device-specific) is 74% at1year.Thiscompareswith50%1-year History survival for bi-ventricular assist device Attempts to develop VADs was spearheaded by support. Michael DeBakey, who performed the first clinical r The most common causes of death following VAD implant in 1963 (Table 8.1). By 1966, he achieved VAD insertion are cardiac failure, infection, the first successful bridge to recovery in a post- and neurological events. cardiotomy HF patient who was weaned from a para- corporeal device after 10 days of support. The first total artificial heart (TAH) was used by Denton Coo- Chronic heart failure (HF) remains a persistent global ley in 1969. This is noteworthy not only as a histor- problem despite advances in preventive and therapeu- ical medical event, but also as a legal milestone, as tic developments. Considering that the primary func- the patient’s wife filed an action claiming failure of tion of the heart is to act as a blood pump, it might consent in relation to an experimental procedure. It be expected that such a role could easily be replaced remains a reference case in issues of consent. The early with contemporary technology. Numerous attempts to enthusiasm for MCSD was fueled by optimism, rather create artificial blood pumps have been made since than by clinical outcomes. There were expectations the early 1960s. The initial hope of developing a pump that once minor technical challenges were overcome, that would completely replace the function of the heart success would inevitably follow. An Artificial Heart hasproventobechallenging.However,significant Program was formed in the United States with the progress has been made, particularly in recent years, aim of developing an artificial heart by 1970. The con- and there are now a number of reliable mechanical temporary thinking included an implantable nuclear replacements for the failed heart. energy source to power the device long term. This

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

76 Chapter 8: Ventricular assist devices

Table 8.1 Milestones in the evolution of mechanical Table 8.2 INTERMACS classification system circulatory support INTERMACS Date Event Level Patient status NYHA 1953 Gibbon performs first ASD closure with aid of 1 Critical cardiogenic shock IV cardiopulmonary bypass circuit 2 Progressive decline IV 1963 DeBakey (Texas) – first ventricular assist device case 3 Stable but inotrope IV 1967 Bernard (Cape Town) performs the first successful dependent heart transplantation 4 Recurrent advanced HF Ambulatory IV 1969 Cooley (Texas) implants first total artificial heart 5 Exertion intolerant Ambulatory IV 1971 First patient discharged home with MCS 6 Exertion limited IIIB 1982 De Vries (Utah) implants a total artificial heart as destination therapy 7 Advanced NYHA III III 1984 First successful bridge to transplantation with a Novacor LVAD 1985 First survivor for 1 year on MCS outcomes, confidence has grown, and these devices 1995 First (Berlin) successful VAD bridge to recovery are increasingly used in the clinical management of advanced HF. 1996 First survivor for 2 years with MCS 1999 First rotary pump implantation 2001 REMATCH trial: Randomized trial comparing optimal medical therapy against pulsative VAD as destination Indications therapy The predominant indication for mechanical circula- 2009 HeartMate II Study: Randomized trial of pulsatile tory support is to augment the systemic circulation in versus continuous flow VAD as destination therapy thepresenceofafailingnativeheart.Themediansur- vival after diagnosis of heart failure is 3 years in med- developmental phase spawned a number of com- ically managed patients. A decision regarding elective mercially available devices such as the Thoratec placement of a VAD in the chronic HF setting must paracorporeal VAD, the Novacor left ventricular assist balance the predicted survival without VAD, the like- device (LVAD), the HeartMate, and Abiomed devices. lihood of a timely heart transplant, and their survival However, although the initial goal was long-term ontheVAD.TheInternationalRegistryforMechani- replacement of the heart with a mechanical pump, cally Assisted Circulatory Support (INTERMACS) has problems with device durability and device-related developed a classification system of HF that best iden- morbidity resulted in a switch in focus from develop- tifies their urgency and status and is tailored to the ing a device as an end point or destination therapy to indications of MCS (Table 8.2). a device to support an HF patient to an allograft HT In this system, MCSD is considered for INTER- (bridging to transplant, BTT). The first BTT was per- MACS level 1–4 patients. To date, there is no evidence formed in 1978, but it wasn’t until 1984 that the first supporting the use of MCSD in INTERMACS level 5 survivor was seen. These first-generation devices were patients. The choice of device used is influenced by large and generally could only be sited in recipients the acuteness of the patient and the projected dura- above a certain size. The recipient was also trapped in tion of support required. When the etiology suggests the hospital due to the large and unwieldy drive con- rapid recovery is possible (e.g., fulminant myocardi- soles. The need for smaller implantable devices, with tis) or when the patient’s systemic suitability for a per- control systems that facilitated return to community manent device is doubtful (e.g., possible neurological living, motivated the next generation of devices. In dysfunction), support with a temporary device is war- addition, improving device reliability and reducing ranted. This allows bridge to recovery or bridge to deci- morbidity were important priorities. To achieve these sion regarding the placement of a longer-term VAD, aims, the focus shifted away from total heart replace- listing for transplantation, or withdrawal of support. ment to ventricular assistance, and from pulsatile In less urgent situations and when the patient has been to continuous-flow devices, e.g., Thoratec Heart- fully assessed, a long-term implantation device can be Mate II, HeartWare HVAD. With improving clinical sited from the outset.

77 Section 2: Heart

Patient selection are also associated with RV impairment. RVAD use is associated with higher rate of bleeding and throm- Careful selection of patients is a cornerstone in a boembolic complications. In a HeartMate II series, successful VAD program. Worldwide, 78% of VAD 6% of LVAD patients required additional RVAD sup- recipients are male; 51% of patients are aged between port. In cases in which the initial strategy was to site 40 and 59 years and 18% between 19 and 39 years. an LVAD but RV dysfunction becomes an impedi- The majority are sited in an acute phase, with 30% ment, late conversion from a uni-ventricular LVAD placedinthepresenceofcriticalcardiogenicshockand to bi-ventricular support is associated with poorer 40% in the case of progressive decline. BTT candidates outcomes. Bi-ventricular support as an initial ther- must be suitable for transplantation or likely to be apy is indicated for patients with arrhythmia likely to transplant candidates after a period of VAD support, threaten the function of the remaining ventricle if a e.g., patient with raised pulmonary vascular resistance uni-ventricular approach is used, or those with multi- (PVR) or borderline renal function. Patient selection organ failure associated with right heart dysfunction. for transplantation is discussed in Chapter 6.Among The final decision to use bi-ventricular support should those being considered for destination therapy, an be made intraoperatively, with direct inspection of the approachsimilartothatofhearttransplantassessment right ventricle and the aid of transesophageal echocar- is required. Ideally the patient should demonstrate sin- diography (TEE). gle organ failure, with minimal or reversible hepatic or renal impairment. A number of scoring systems are available to predict survival following VADplacement. Classification MCSD are classified according to the duration of sup- Ventricular assist devices port, flow characteristics, and/or pump mechanism. VADs provide either left-, right-, or bi-ventricular sup- Classification based on support duration reflects the port. A left-sided VAD (LVAD) withdraws oxygenated durability of the pump and ease of placement. Gen- blood from the left atrium or LV and returns it to erally, short-term devices can be inserted quickly and the aorta; a right-sided VAD (RVAD) draws venous are usually paracorporeal or extra-corporeal. They are blood from the right atrium or right ventricle (RV), only designed for days or weeks of support but cost and returns it to the pulmonary artery. In general, it a fraction of the price of longer term devices. Longer is preferable to cannulate the ventricle for VAD inflow, term devices are usually implanted inside the chest or as this provides superior ventricular decompression, the anterior abdominal wall. They have portable con- avoids ventricular stasis, and affords higher VAD flow trollers that are designed to allow patient discharge rates. from hospital. The flow generated by these blood pumps may Bi-ventricular versus uni-ventricular be pulsatile or non-pulsatile. The first-generation devices produce flow by volume displacement, and support therefore, they generate pulsatile flows (e.g., Thoratec The output of an LVAD is dependent on adequate RV PVAD and IVAD, Novacor, HeartMate XVE). These function to propel sufficient blood across the lungs devices are relatively bulky and noisy in operation into the left heart chambers for the LVAD to pump. due to the multiple moving parts. The latter predis- Likewise, an RVAD can only provide benefit if the poses to wear and tear and therefore finite durabil- native left ventricle can match the output of the RVAD. ity. Second- and third-generation devices are rotary If both native ventricles are failing, two VADs are blood pumps that produce continuous flows with required in order to provide bi-ventricular assistance reduced pulsability. These devices are smaller, quiet to support the circulation. in operation, and do not require valves or compli- Pre-implantation factors can often predict the suc- ance chambers. Instead, without a valve, there is the cess of uni-ventricular support with LVAD alone. potential for regurgitant flow if the device is stopped. Poor RV stroke work index, older age, and non- Second-generation devices have mechanical bearings ischemic cardiomyopathy predict poor RV function (e.g., Jarvik 2000, HeartMate II). Third-generation in the post-implantation period and a high require- devices, however, do not require mechanical bear- ment for inotropic support. High pulmonary pressures ings. Instead, the impeller is either suspended by

78 Chapter 8: Ventricular assist devices magnetic levitation (e.g., Berlin Incor, Terumo Dura- Table 8.3 Potential complications of LVAD use Heart) or hydrodynamic suspension (e.g., HeartWare Air embolism HVAD). Bleeding Thromboembolism Peri-operative considerations Neurological injury Preoperative preparation Right ventricular failure in the setting of LVAD placement Hemolysis Preoperative preparation should focus on optimiz- ing end-organ function and RV function. Ideally, all Infection sources of sepsis are eradicated. Optimization of renal Device failure function may require administration of inotropes and/or intra-aortic balloon pump to augment cardiac must be secured externally to minimize movement and output and renal perfusion. RV function can also be trauma to the exit sites. improved by normalizing fluid status to achieve right Once returned to the intensive care unit, patients atrial pressures Ͻ12 cmH2O with diuretics or veno- must be closely monitored for early complications. venous hemofiltration. In the emergency situation, the Coagulation defects should be corrected without wait- use of a short term MCSD might enable recovery of ing for signs of significant bleeding. RV failure can be end-organ function prior to committing to a longer- precipitated by excessive LVAD flow or elevated PVR. term device (bridge). Nutritional status should also be Therefore,itisadvisabletolimittheLVADflowrate optimized. in the first few days to avoid overwhelming the RV. It is essential to avoid factors that may predispose to Implantation increases in PVR, such as hypoxia and acidosis. Anti- coagulation is usually omitted in the first 48 hours The implant procedure should be covered with broad- and is only started when the patient has stopped bleed- spectrum prophylactic antibiotics. Anti-fibrinolytic ing (test tube drainage Ͻ30 ml/hr for 3 consecutive agents may reduce the risk of postoperative blood loss hours). Rising right atrial pressures associated with (e.g., tranexamic acid 2 g IV followed by an infu- a fall in pump flow are signs of impending RV fail- sion of 1 g/hr). TEE is used to confirm aortic valve ure or tamponade. RV failure could be confirmed with competence and to exclude the presence of a sep- TEE, which would demonstrate full right-sided cham- tal defect. Normothermic cardiopulmonary bypass is bers and empty left chambers. Under these circum- used, but the lungs are kept ventilated throughout the stances, it is important not to increase the pre-load fur- bypass period, often with the addition of nitric oxide ther with more fluid infusions. Immediate treatment (NO) at 5–20 parts per million to minimize atelecta- consists of a combination of inotropes and pulmonary sis and pulmonary vasoconstriction respectively. The vasodilators. If the situation does not respond readily, use of NO has been shown to reduce the need for early consideration should be given to re-exploring the RV support. The patient is filtered on bypass to main- patient and addition of right ventricular assist device. tain hemoglobin concentration greater than 10 g/dL and base excess within ± 2 mEq. The VAD cannu- lae are implanted on a beating heart, thus avoiding Operative outcomes aortic cross-clamping and cardiac ischemia. The peri- Data from the 2010 INTERMACS Register showed cardial space is flooded with carbon dioxide so that an overall 1-year survival of 74% after LVAD place- gas entrained into cardiac chambers can dissolve more ment (not device specific). This compares with 50% readily. The RV is supported by inotropes as appro- 1-year survival for Bi-VAD support. In selected series priate. Heart rate is optimized with temporary pacing using only rotary LVAD, 1-year survival rates of 74% at 90–100 bpm. SVR is maintained between 800 and and 85% have been achieved for DT and BTT, respec- 1000 dyne.sec.cm−5 using an infusion of vasopressin tively. Bleeding and infection are the most common or with an alpha agonist. Thorough de-airing is con- early complications (Table 8.3). Neurological events firmed prior to wean from bypass. At the end of the are most likely to occur in the first 1–2 months after procedure, careful hemostasis is crucial to minimize implant. The most common causes of death are car- hemorrhage. The percutaneous cannula or driveline diac failure, including right ventricular failure and

79 Section 2: Heart

ventricular arrhythmias, infections, and neurological and neuroendocrine responses. There is much inter- events. At 1 year after implant for bridge to transplant, est in the systemic effects of reduced pulsativity with 52% have undergone heart transplant, 35% remain continuous flow devices. So far, there does not appear alive on LVAD support, and 12% have died on support. to be any detriment to renal and hepatic functions even after years of support with such devices. However, Physiological response to reduced systemic pulsatility appears to be associated with a higher incidence of gastrointestinal bleeding. VAD – remodeling of heart failure Outpatient management Cardiac response The benefits of angiotensin-converting enzyme (ACE) Hospital discharge inhibitors in HF through afterload reduction leading Early mobilization is essential to successful outcomes. to ventricular remodeling are well established. A more Many candidates have been in the intensive care unit direct unloading of the LV with a VAD offers similar for prolonged periods prior to device implant, with remodeling potential. An effective LVAD will unload significant physical deconditioning. Physiotherapy the LV,which results in restoration of pressure-volume involvement and patient motivation will be required relationships, reduction in end-diastolic ventricular at an early stage. Although full fitness may take some dimensions, reduction in cardiac mass, and improved timetorestore,theremustbeaprogramtoopti- ␤ contractility and cardiac index. Normalization of - mize independence and to provide education for the adrenergic receptor function is also seen. Histology patients and their caregivers to facilitate hospital dis- samples taken at the time of LVAD placement com- charge. This can be a staged process, with intermediate paredwiththosetakenatthetimeofsubsequenttrans- dischargetoanearbyhostelpriortodischargehome. plantation have shown reversal of myocardial hyper- Outpatient management represents more efficient trophy, attenuation of the myocardial apoptotic pro- use of health care resources and is of high impor- fileseeninHF,andimprovedmyocardialmitochon- tance for patient quality of life. Prior to discharge, drial function. There is reduced ventricular expression patients need to be mobilizing independently, feed- of atrial natriuretic peptide and reduced tumor necro- ing adequately, have good pain control, and be able sis factor alpha production. Cases of complete myocar- to manage the device in conjunction with their home dial recovery have also been reported, allowing LVAD support. Arrangements must be in place to provide explant. VAD patients in the community access to advice and assistance round-the-clock. Regular outpatient follow- Pulmonary response ups are arranged at intervals. Improved LV unloading has sequential effects on the pulmonary circulation. VADs have been shown to Patient education reduce pulmonary hypertension, with both systolic and diastolic pressures falling significantly. The pul- The patient and their family must be trained in manag- monary capillary wedge pressure is reduced, as are the ing the device (including emergencies) and prescrip- PVR and transpulmonary pressure gradient. This may tion medications. This requires a period of familiar- render a patient with prohibitive pulmonary hyperten- ization prior to discharge, during which the patient sion secondary to LV failure to HT candidacy. should manage the device, power supply, and all their own medications. The patient and their caregivers should be comfortable with battery and controller Systemic responses changes prior to discharge. Theimprovedcardiacoutputhasbeneficialeffects on end organs and on systemic neurohormonal and cytokine signaling. Improved hemodynamics lead to Monitoring improvements in renal and hepatic function and facil- Monitoring of patients implanted with a continuous itate physical exercise and rehabilitation. There is a fall flow device can be challenging for the unfamiliar. in plasma renin, angiotensin II, plasma epinephrine, With a weak or absent pulse, blood pressure can often and norepinephrine levels and reductions in cytokine only be measured with a blood pressure cuff and a

80 Chapter 8: Ventricular assist devices hand-held Doppler. Interpretation of the device con- with an LVADfor 160 days at the Berlin Heart Institute trol display and alarm record is likely only possi- demonstrated such an improvement in cardiac func- ble by those familiar with the device. Monitoring for tion that it became possible to remove the LVAD. This hemolysis with bilirubin, hemoglobin, and plasma free started the bridge to recovery program, with good suc- hemoglobin should be performed at intervals. Anti- cess reported both in Berlin and in London using the coagulation will require careful monitoring with INR Harefield Protocol. Successful weaning from the LVAD measurements. is predicted by LV size (Ͻ50 mm LV end-diastolic diameter) and ejection fraction. Quality of life Destination therapy Most patients report significant improvement in symptoms to New York Heart Association (NYHA) Long-term permanent MCS for patients not suit- class I or II. Patient satisfaction and improved objec- able for transplantation is sometimes called destina- tive measures are noted in small studies. Patients can tion therapy. This is perhaps the ultimate goal of travel freely by road or by plane. Some countries still MCS development. It may be appropriate in selected impose restrictions on driving for patients implanted patients with INTERMACS 1–4 (NYHA IV) heart with a VAD. Contact with water will damage the failure status on optimum medical therapy. The first device, and so VAD patients are not allowed to swim piece of good evidence for destination therapy came or bathe. Showering is performed ideally with a hand- from the REMATCH trial with the Thoratec Heart- held shower head to avoid wetting the power supply. mate VE. This demonstrated a 1-year and 2-year sur- Some VAD manufacturers provide special equipment vival rate of 52% and 23%, respectively, in the VAD to protect the driveline and controller during shower- group compared with 25% and 8%, respectively, in ing. Sporting activities are limited to land-based sports the medical therapy group. Subsequent comparison that are non-contact. As a result, some VAD recipients of the pulsatile HeartMate XVE with a continuous- remain dissatisfied with living with a VAD implant. flow device (Heartmate II) showed 2-year survival Psychological support is often helpful and neces- rates for the Heartmate XVE of 24% and the Heart- sary, as there are reports of self-harm among VAD mate II of 58%. However, there is currently no trial patients. data to support the use of LVADs in NYHA class III patients. Transplantation after VAD therapy The future The first successful bridge to a heart transplant with a VAD was performed in a patient at Stanford Uni- Science fiction has often used the theme of the bionic versity. Although technically challenging, the 1-year man of the future. MCS is certainly in that realm but post-transplant survival for patients bridged to trans- is now of our time. The latest generation of devices plant with an LVAD is similar to that of those with- offer clinical reliability, and the limitation of their out an LVAD (discussed further in Chapter 12). Dur- widespread application is largely due to cost and a lim- ing the procedure, the re-opening and dissection phase ited pool of expertise in their use. The use of VAD in isthemostchallenging.Itishelpfuliftheimplanting transplantation will always be limited by the number surgeon used a membrane under the sternum over- of available hearts. Unless the number of donor hearts lying any cannula and device. A good operative note increases, using VADs to bridge patients to trans- or diagram is useful, but a preoperative chest com- plant will only serve to select which patient receives a puted tomography is helpful to identify VAD and con- heart. The more likely scenario is that VADs will see duit positioning. Some operators prefer initiating car- greater application outside of transplantation as des- diopulmonary bypass by femoral or axillary cannula- tination therapy. However, as the duration of support tion prior to sternotomy. for bridge to transplant increases, these two indica- tions are becoming increasingly blurred. Using VAD as bridge to recovery is a very exciting prospect and Myocardial recovery and VAD removal deserves further study. However, the proportion of LVAD support can result in significant ventricular VAD patients successfully recovered and explanted remodeling, as previously discussed. In 1995, a patient remains small at present. Adjuncts in regenerative

81 Section 2: Heart

medicine, including gene therapy and cell transplanta- this area in the Western world? Perhaps a bigger chal- tion, may enhance and realize the full potential of this lengeistheethicalandmoraldilemmaoftheappro- approach. priate healthcare resource allocation in the world. With regard to device technology, further improve- mentsaredesirable.Amodularcomponentsystem Further reading would enable surgical replacement of failing parts Feller E, Sorensen E, Haddad M, et al. Clinical outcomes are rather than replacement of the entire VAD system if a similarinpulsatileandnonpulsatileleftventricularassist problem develops. The size of the VADs has reduced device recipients. Ann Thorac Surg 2007; 83: 1082–8. considerably, but further size reduction may confer Kirklin J, Naftel D, Kormos R, et al. Second INTERMACS additionalbenefit.Someofthesearenowsmallenough annual report: More than 1,000 primary left ventricular to be used as a pair of devices for Bi-VAD support. assist device implants. JHeartLungTransplant2010; 29: The Achilles heel of all currently available LVAD sys- 1–10. tems is the need for a percutaneous driveline. Elimi- Liang H, Lin H, Weng Y, Dandel M, Hetzer R. Prediction of nating such a requirement is likely to minimize infec- cardiac function after weaning from ventricular assist tion risks, remove the onerous burden of driveline devices. JThoracCardiovascSurg2005; 130: 1555–60. management, and improve body image and the psy- Lietz K, Long J, Kfoury K, et al. Outcomes of left ventricular chological impact of the device and, therefore, qual- assist device implantation as destination therapy in the ity of life of VAD recipients. The development of fully post-REMATCH era: Implications for patient selection. Circulation 2007; 116; 497–505. implantable VADs hinges on a better power delivery system and control mechanism. Improvements in bat- McMurray J. Systolic heart failure. NEnglJMed2010; 362: 228–38. tery technology will also enable a patient to charge overnightwhilesleepingandmobilizeforthefullday Ochiai Y, McCarthy P, Smedira N, et al. Predictors of severe on a small and light battery pack. The main challenge, right ventricular failure after implantable left ventricular assist device insertion: analysis of 245 patients. however, remains affordability. These devices are most Circulation 2002; 106: I198–202. likely to be used in an older patient population, possi- Slaughter M, Rogers J, Milano C, et al. Advanced heart bly retired and not contributing to the state economy failure treated with continuous-flow left ventricular and without independent income to cover the cost of assist device. NEnglJMed2009; 361: 2241–51. their treatment. The potential number of people who Tsui S, Parameshwar J. Shock due to catastrophic loss of might benefit from such devices is unknown. Can we myocardium. In Banner N, Jessup M (eds). ISHLT afford to offer this to the entire HF population, and Monograph Series: Advanced Heart Failure. Philadelphia: what is the opportunity cost of resources directed into Elsevier, 2009, pp. 509–23.

82 Section 2 Heart Chapter Surgical procedure

9 R.V. Venkateswaran and David P. Jenkins

Key points tion. Assessment and management of potential donor r Median sternotomy is the standard approach hearts is discussed in Chapter 5. for heart transplantation. r The bicaval technique of implantation is Technique of orthotopic heart most commonly employed. r Previous sternotomy makes the procedure transplantation considerably more complicated and may lead to increased postoperative bleeding. Preparation r Ventricular assist devices and implantable The patient is painted with standard antiseptic solu- defibrillator devices should be removed tions and draped. Both groins should be prepared and concurrently. draped for access to the femoral vessels either for can- nulation or for insertion of intra-aortic balloon pump. In the current era, many recipients have cardiac resyn- Heart transplantation, like any other procedure, chronization therapy (CRT) devices with or without requires careful selection of the recipient and donor implantable cardioverter-defibrillators (ICDs). These graft. It is medically and ethically vital to select recip- should also be included in draping so that after heart ients who need the most and have the best chance implantation, these leads and the generator can be of good outcome. It also requires an excellent coor- removed at the end of the procedure. dination and communication between the donor and recipient teams. The surgical procedure and immedi- ate postoperative care are discussed in this section. Incision and cardiopulmonary bypass Median sternotomy is the standard approach for Preoperative preparation heart transplantation. The pericardium is opened Donor to recipient matching is based primarily on and the edges sutured to the skin edges. The plane ABO blood group compatibility. Other variables that between aorta and pulmonary artery is separated using mayplayaroleinoptimallymatchingdonorandrecip- diathermy. In difficult circumstances, the aorta may ients are age, height, weight, and gender. In general, be encircled with a nylon tape to get easy access to the height disparity between the donor and recipi- ascending aorta for cannulation. Following systemic ent should be restricted to within 10%. Recipients’ heparinization, the ascending aorta is cannulated as invasive hemodynamic parameters such as transpul- high as possible just below the origin of innominate monary gradient (TPG) and pulmonary vascular resis- artery. The superior and inferior vena cavae (SVC and tance (PVR) must be taken into account when accept- IVC) are cannulated separately for venous drainage. A ing a marginal heart donor. Recipients with more than size 24 or 28 right-angled cannula is preferred for the 10% of panel reactive antibodies in serum suggesting SVCandasize30or32malleablestraightcannulafor presensitization to alloantigen usually require cyto- theIVC.Caremustbetakenduringcannulation,espe- toxic or virtual cross-matching prior to transplanta- cially if the CVP is very high. Following cannulation,

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

83 Section 2: Heart

cardiopulmonary bypass is initiated, and the patient is any venting incisions made at retrieval are oversewn. A cooled to 30oC. Both cavae should be encircled using curvilinear incision is made on the lateral wall of right nylon tapes and snared to avoid draining air into the atrium from the IVC opening toward the right atrial bypass circuit. appendage. This reduces the risk of injuring thesinoa- trial node. The SVC is oversewn. Recipient cardiectomy Using a long length of double-ended 3–0 Prolene, the anastomosis is started at the left atrial appendage Explantation of the heart is commenced after making of the donor heart to the recipient’s left atrial cuff at sure that the donor heart is within 20 minutes from the the level of the left superior pulmonary vein. After two recipient hospital. The aorta is clamped and the cavae or three stitches, the heart is lowered into the pericar- snared, but cardioplegic arrest is unnecessary. How- dial cavity over a cold swab, and further cold swabs are ever, sometimes with a large heart, venting is useful placed anteriorly. A Prolene stay suture at the level of to aid cardiectomy, either via the left ventricular (LV) the left inferior pulmonary vein on both donor and apex or right superior pulmonary vein to keep the field recipient left atrium can aid accurate approximation clear. An alternating current fibrillator is used by some and reduce the risk of torsion. The posterior lateral left to induce ventricular fibrillation. atrial cuff is sutured, and then the second arm of the The right atrium is opened along the atrioventric- 3–0 Prolene is used to complete the anterior medial left ular groove, leaving an adequate cuff along the lat- atrial cuff. If a right superior pulmonary vein vent was eral wall toward the appendage. The interatrial sep- used prior to explantation, the same could be used to tum is opened along the fossa ovalis and blood drained vent the left ventricle of the donor heart to avoid any fromtheleftatrium.Theincisionisextendedcra- distension during the implantation. A cold saline irri- nially around the roof of the left atrium. Then the aorta gation or a cold jacket may be used to keep the heart and pulmonary arteries are divided immediately dis- cold during the rest of the implantation. tal to their valves. The interatrial septum is further Next the pulmonary artery anastomosis is com- divided caudally into coronary sinus, and the incision menced, using a continuous 4–0 Prolene suture. Care follows the course of the sinus around the heart. The is taken to avoid any twist, and the length should be heart is explanted, leaving a large cuff of left atrium by kept short to avoid any kinking. Toreduce the ischemic dividing left atrium close to the posterior annulus of time, the anterior wall of pulmonary artery anastomo- the mitral valve. Both atrial appendages are excised to sis can be completed after reperfusion of the heart. The reduce the risk of postoperative thrombus formation. aortic anastomosis is next completed using a continu- This should leave a cuff of right and left atrium and ous 4–0 Prolene suture. Any discrepancy in size should the two major arteries. The left and right atrial cuffs be adjusted on the anterior wall so that any suture line can be further fashioned to match the size of the donor bleeding can be controlled easily. A 4–0 Prolene purse heart. string suture is placed on the ascending aorta, and an aorticrootventcannulaisinserted.Thecoldsalineor Implantation of donor heart the jacket is removed, and the patient is placed in the Trendelenburg position. The heart is filled with blood, Biatrial technique and thorough deairing is done through both aortic Lower and Shumway initially described the technique root and pulmonary vein vent if present. Carbon- of biatrial anastomosis, which has been widely used in dioxide insufflation of the surgical wound may be used cardiac transplantation. The heart is removed from the to aid deairing. The cross-clamp is then removed, and ice box and prepared on a table in a cold saline con- the perfusionist is advised to open the arterial circuit tainer. It is examined for any damage during retrieval, with a leucocyte filter. and the interatrial septum is examined for any septal During reperfusion, the anterior wall of the pul- defect or persistent foramen ovale, which may then be monary artery anastomosis is performed, if not com- closed. The aorta and pulmonary artery are separated. pleted earlier. The right atrial anastomosis is com- The left atrium is fashioned for anastomosis by inci- pleted using continuous 3–0 Prolene with the heart sions between the origins of the pulmonary veins to perfused and beating to reduce “warm” ischemic time. create a circular cuff of atrium to match that of the The patient is re-warmed, and the caval snares are recipient. The left atrial appendage is inspected, and removed.

84 Chapter 9: Surgical procedure

Following the completion of all anastomoses, the domized trials are unlikely to be done to evaluate the heart is reperfused and the patient is rewarmed. Dur- outcome between both techniques due to a major shift ing this time, ventilation of the lungs may be recom- to the bicaval technique of implantation. menced. Both atrial and ventricular temporary pacing wires are inserted and connected to a dual-chamber Special considerations pacing device. Temporary pacing and isoprenaline infusion are useful to keep a higher heart rate in the Although the surgical technique of heart transplanta- denervated heart. The patient is weaned off from car- tion is simple, there are certain specific circumstances diopulmonary bypass, and the heart is decannulated in which the operation can be technically demanding and hemostasis is achieved. The right pleural cavity is and require careful planning to get the best outcome. opened widely to create a large window to avoid post- operative pericardial effusion. Redo-sternotomy Patients who have had previous cardiac surgery either Bicaval technique forischemicorvalvularheartdiseaseareincreasingly referred or listed for heart transplantation. In these In the early 1990s, the bicaval technique of heart trans- patients, the recipient dissection and cardiectomy will plantation was first introduced, and currently, it is take longer, and adjustment of timings is necessary. most widely used technique. The recognition of com- Theretrievalofthedonorheartmayhavetobedelayed plications related to atrial enlargement in the biatrial until the recipient team has completed the dissection anastomosis led to the development of the bicaval and and are ready to institute cardiopulmonary bypass. total heart transplant techniques to preserve atrial size. Some units advocate exposure of the femoral vessels The donor retrieval team should dissect the entire in particularly complex cases in the event that cannu- length of the SVC and ligate the azygos vein in order to lation for bypass is necessary. Following sternotomy, get adequate length of SVC. After confirming the ade- the aim is directed toward dissecting the aorta and quacy of the length of donor SVC, the recipient right right atrium and both cavae to establish cardiopul- atrial cuff is excised, leaving separate SVC and IVC monary bypass. The rest of the dissection can be per- cuff for bicaval anastomosis. The caval anastomosis can formed after establishing bypass. It is preferable to use be constructed after reperfusing the heart during sys- diathermy dissection to have a dry field. Once the dis- temic rewarming. A sump sucker inserted through the section is complete, the remaining operation is carried SVC orifice into the right atrium would aid in dry field out as a standard procedure. during IVC anastomosis, which is performed using 3– 0 Prolene suture. The SVC anastomosis is finally com- pleted using 4–0 or 5–0 Prolene sutures. Care is taken Explantation of ventricular assist devices to avoid any twisting and narrowing by purse-string Ventricular assist devices (VADs) are more commonly effectontheSVCanastomosis.Totalhearttransplanta- used as a bridge to transplantation, and many patients tion involves excision of the recipient heart, including wait for heart transplantation with a functioning VAD the left atrium, leaving two separate pulmonary venous or are listed for urgent procedure due to VAD-related cuffs for each side and anastomosis of the pulmonary complications. The procedure is technically more chal- venous cuffs and separate caval anastomosis. lenging in these circumstances, and we recommend These techniques are more technically demanding 2 to 3 hours dissection time to explant the recipient and may lead to increased implantation and warm heart. During the initial VAD procedure, it is advis- ischemic time. However, several series of retrospec- able to avoid dissection around the aorta and pul- tive studies have shown that the bicaval technique monary artery and both cavae to minimize pericar- has been associated with reduced hospital stay; lower dial adhesions. Use of silastic membrane around the incidence of atrial arrhythmia, pacemaker require- VAD inflow cannula and partial closure of the peri- ment, and diuretic dependence; early return to sinus cardium at the time of initial surgery may help reduce rhythm; and reduced incidence of postoperative tri- adhesions. Good communication is even more essen- cuspid regurgitation and RV dysfunction. However, tial between the donor and recipient team to avoid a the long-term outcome of heart transplantation does long ischemic time. A preoperative computed tomog- not differ between both techniques. Prospective ran- raphyscanofthechestmayaidincleardemarcationof

85 Section 2: Heart

the VAD outflow graft and its relationship to the pos- recipient SVC. The donor retrieval team should dissect terior sternal table. the SVC, including the innominate vein, for tension- Atredosternotomy,careistakentoavoidinjuryto free SVC anastomosis. In patients following previous the drive line, which normally crosses the midline at systemic to pulmonary artery shunts or bidirectional the lower end of sternum, and the VAD outflow graft, Glenn procedure, the pulmonary arteries may be nar- which may lie anteriorly beneath the sternum. If there rowed and require surgical correction of stenosis at the is difficulty in identifying the aorta, the VAD outflow time of heart transplantation. graft can be used for arterial cannulation. Once the cavae and aorta are dissected, cardiopulmonary bypass Left-sided SVC is established in the usual manner and the VAD is turned off and the outflow graft isolated and clamped. Patients with left-sided SVC pose difficulty in estab- The VAD itself can be removed with the heart, but if lishing adequate venous drainage and cardiopul- time is short, it can be removed following implanta- monary bypass. It may only be detected at the time of tion of the donor heart. If there is any infection around surgery due to the absence or smaller size of the recip- the driveline, it is dealt with at the end of the proce- ient SVC. It may be associated with a small or absent dure after closure of the sternotomy wound to avoid innominate vein. If detected, the left SVC should be any contamination. cannulated to avoid venous congestion of the brain and left side of the upper torso. After explantation of the heart, the transplant operation is performed as usual. Heart transplantation for congenital The donor SVC is anastomosed to the innominate vein heart disease high up. If the innominate vein is absent, then the Duetoadvancesinthemanagementofcongenital donor team should retrieve the heart, including the heart disease, many patients present in early or later innominate vein, and this can be used to create a new adult life with end-stage heart failure and are listed SVC innominate vein junction. At the end of proce- for heart transplantation. These patients may have had dure, the left SVC cannula is removed and it is over- multiple operations in the past, and dissection and sewn. explantation of their heart can be a challenge. Surgi- cal planning due to the variability of anatomical find- Domino heart transplantation ings is essential with the aid of appropriate radiolog- In patients undergoing heart–lung transplanta- ical imaging. There are a number of particular prob- tion, especially small cystic fibrotic recipients, their lems encountered with heart transplantation in this explantedheartcanbeusedfortransplantation.This group of patients. Recipients with atrial switch palli- procedure is called domino heart transplantation. ation (Mustard/Senning procedure) for transposition Technically, the surgical procedure is not different of great arteries in the earlier era pose particular dif- from that of standard heart transplantation. However, ficulty for venous cannulation. In these patients, the the SVC may not be long enough in the explanted atrial baffle prevents the insertion of venous cannula heart and may require biatrial anastomosis in the through the right atrium. The SVC is preferably can- domino recipient. The outcome following domino nulatedhighupnearthejunctionwithinnominate heart transplantation has been shown to be com- vein. After establishing cardiopulmonary bypass using parable to that of cadaveric donor transplantation. single venous cannula, the atrium can be opened and Heart–lung transplantation and domino heart trans- the baffle incised to allow the insertion of IVC can- plantation have largely been superseded by bilateral nula. An alternative is cannulation of the femoral vein. sequential lung transplantation. In some patients, the pulmonary artery lies posterior to the aorta, and it is preferable to have an adequate length of pulmonary artery to allow tension-free anas- Heterotopic transplantation tomosis. In addition, bleeding from the anastomo- This procedure was rarely performed in the recent sis will be difficult to control due to the poor access. era but may have a place in the modern era to In patients with single ventricular palliation (Fontan make use of small hearts that might otherwise not be type procedure), heart transplantation is specifically used to maximize organ utilization. Heterotopic HT complicated, in particular by the short length of the may be performed as either a bi-ventricular or left

86 Chapter 9: Surgical procedure ventricular assist configuration. The donor heart is ientsarepreferablymanagedinanisolatedintensive placed usually in the right side of the chest and anas- care unit bed to avoid cross-infection. Careful con- tomosed to the recipient’s heart, which remains in sideration should be given to the adequacy of cardiac situ. For bi-ventricular support, after institution of output in maintaining oxygen delivery to the tissues, cardiopulmonary bypass, the posterior pericardium is bleeding, collections, pneumothorax, and position of divided to create space for the donor heart. The left the monitoring lines. The patient is slowly warmed atrialanastomosisisfirstperformedbyplacingthe using warming blankets and inotropes are continued donor heart in the chest with the right ventricle facing to maintain stability. After achieving stability for a few posteriorly. After completing left and right atrial anas- hours,thepatientmaybewokenupfromtheanesthe- tomosis, the donor aorta is connected to the recipient’s sia for planned extubation. aorta using side-biting clamp. The pulmonary artery anastomosis is performed last, and both this and the Further reading aortic anastomosis may require vascular grafts due to Christie JD, Edwards LB, Aurora P, et al. The Registry of length limitations. Heterotopic transplantation allows the International Society for Heart and Lung much more leniency on the donor and recipient mis- Transplantation: Twenty-sixth Official Adult Lung and matching.Astherecipientheartisnotexcised,this Heart-Lung Transplantation Report – 2009. JHeartLung could be particularly useful in patients with elevated Transplant 2009; 28:1031–49. TPG and PVR. However, patients may continue to suf- Jacob S, Sellke F. Is bicaval orthotopic heart transplantation ferfromsymptomsofanginaorthromboembolism superior to the biatrial technique? Interact Cardiovasc from the diseased heart. Thorac Surg 2009; 9: 333–42. Luckraz H, Charman SC, Parameshwar J, et al. Are non- brain stem-dead cardiac donors acceptable donors? J Immediate postoperative Heart Lung Transplant 2004; 23: 330–3. management Potter CD, Wheeldon DR, Wallwork J. Functional assessment and management of heart donors: a rationale Postoperative management will be discussed in more for characterization and a guide to therapy. JHeartLung detail in the following chapter. Post-transplant recip- Transplant 1995; 14: 59–65.

87 Section 2 Heart Chapter Management during surgery

10 Kate Drummond and Andrew A. Klein

Preoperative considerations Key points Heart transplantation is considered emergency r Commencing an infusion of an inotropic surgery, and there is often little time for extensive agent before induction of anesthesia may evaluation in the immediate preoperative period. In improve hemodynamic stability. order to determine their need for transplantation, all r Donor heart failure is not uncommon, and patients will have had an extensive multidisciplinary mechanical support may be required in order assessmentpriortobeinglistedfortransplantation.In to separate successfully from CPB. order to identify and plan for all potential anesthetic r Management of right ventricular dysfunction complications, some centers review potential heart includes optimization of fluid status, transplant recipients in preanesthetic assessment judicious lung ventilation strategies, and use clinics. Regardless of this, there will be some degree of of pulmonary vasodilators. assessment required prior to induction of anesthesia. r Transesophageal echocardiography may aid A quick medical history, examination, and review of diagnosis and guide management of recent investigations may be all there is time for prior hypotension after heart transplantation. to induction. Patients presenting for heart transplantation by theirnaturehaveseverecardiacdisease.Manyhave subsequent end-organ dysfunction due to chronic low Heart transplant surgery is almost inevitably emer- perfusion. Serum chemistry, electrolytes, and liver and gency in nature, happening often with minimal notice renal function should be checked, along with hema- and out of hours. It requires a rapid response to min- tological and clotting studies. Blood should also be imize ischemic time for the transplanted organ, and taken for blood typing and antibody screening. Cross- good communication between the retrieval and the matching of blood can be difficult in these patients due operating team is essential. Optimal timing of anes- to the presence of antibodies, so the anesthetist should thesia and surgery will ensure that the recipient is liaise with the laboratory ensure there is enough com- prepared and ready for cardiopulmonary bypass as patible blood products on site. In addition, consid- the donor organ arrives. From the time the anesthetic eration should be given to the cytomegalovirus sta- team is alerted to the possibility of organ transplan- tus of donor and recipient when ordering blood for tation and the arrival of the organ, the patient must transfusion. be assessed and prepared and anesthesia induced and It is not uncommon for patients to be on long-term maintained. Following transplantation itself, the peri- therapeutic anti-coagulation due to the high incidence operative team must be prepared to deal with a num- of atrial fibrillation and risk of atrial thrombus. Liver ber of potential complications that may arise in this dysfunction due to hepatic congestion may also com- complex surgical group after weaning from cardiopul- plicate this. Surgery should of course proceed with- monary bypass (CPB). This chapter covers the preop- out delay, but correction in the immediate post-CPB erative considerations and reviews the intraoperative period may be required with fresh-frozen plasma or management of heart transplant patients. prothrombin complex concentrate.

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

88 Chapter 10: Management during surgery

Patients with severe heart failure are often on many to induction for both invasive blood pressure monitor- drugs, including diuretics, angiotensin-converting ing and access to arterial blood sampling. enzyme (ACE) inhibitors and calcium antagonists. Central venous access and introducers for pul- Many of these drugs interact with anesthesia and monary artery catheters are often placed after induc- should be taken into account. tion of anesthesia for patient comfort, but may be Preoperative pulmonary hypertension is a signif- placed in advance of this if clinically indicated. The icantriskfactorforpostoperativerightheartfail- internal jugular approach allows easier access dur- ure. Elevated preoperative pulmonary artery pressure ing surgery but, although uncommon, may interfere (PAP) is associated with increased postoperative com- with future endomyocardial biopsies due to vascular plications, including death. Although this relationship thrombosis. The pulmonary artery catheter should be is not linear, the information obtained from routine placed so that it lies within the superior vena cava preoperative right heart catheterization can assist the (SVC) so that it does not interfere with surgical bicaval anesthetist in predicting and preparing for problems anastomosis. Patients with severe heart failure may be with right heart dysfunction after bypass. transferred to the operating theatre straight from the Any history of previous surgery should be clari- intensivecareunitandalreadyhavetheselinesand fied, as at least 30% of patients will have had a previ- monitors in situ. The anesthetist should check their ous sternotomy incision. Usual precautions for rester- position and patency prior to surgery, as access will be notomy should be taken, including obtaining large- limited once the patient is positioned and draped and bore venous access, placing external defibrillator pads, surgery has commenced. All lines should be placed and having cross-matched blood checked and available with strict aseptic technique due to the risk of infec- in the operating theatre prior to commencing surgery. tion from immunosuppression. In some cases the femoral vessels may be prepared for Cerebral oximetry by using near-infrared spec- cannulation before commencing resternotomy. troscopy is increasing in use in cardiac anesthesia. Preoperative left ventricular assist device (LVAD) It has the advantage of being non-invasive compared use presents a number of unique problems for the with other methods such as jugular venous bulb satu- anesthetist,butmostimportantlypreloadmustbe ration monitoring. Prolonged cerebral oxygen desatu- maintained as hypovolemia can result in inadequate ration (SrO2) is associated with a higher risk of post- ventricular assist device (VAD) output and hypoten- operative cognitive dysfunction and longer hospital sion. Patients with VADs are usually on some form of stay in general cardiac surgery populations. Despite anti-coagulation that will need to be managed. this, the ideal treatment strategy for when confronted Implanted defibrillators and pacemakers are com- with low SrO2 is still unknown. The use of depth mon in this patient group due to the high inci- of anesthesia monitors, such as the bispectral mon- dence of malignant supraventricular and ventricu- itor, although not in routine use, can be a useful lar arrhythmias. Implantable defibrillators should be additional tool for the titration of anesthetic agents turned off immediately prior to induction of anaes- and has also been shown to correlate with cerebral thesia to prevent inadvertent discharge during surgery. oxygenation. External defibrillator pads should be applied. Pace- There is often inadequate time to observe conven- makers should be reprogrammed to asynchronous tional fasting periods in transplant surgery. If this is (fixed rate) mode. the case, a modified rapid sequence induction should be performed. Preoperative use of prokinetics such as metoclopramide or drugs to lower gastric pH such Intraoperative considerations as sodium citrate or proton pump inhibitors are use- ful. Cricoid pressure may be indicated if gastric emp- Induction and maintenance tying is considered likely to be incomplete, in which Following pre-anaesthetic assessment, induction of case either suxamethonium or a longer acting mus- anesthesia should be performed after placement of cle relaxant such as rocuronium may be administered. essential monitoring. Large-bore intravenous access Both have been used successfully in heart transplant should be inserted, and pulse oximetry and ECG surgery. monitoring applied. Commonly placed in radial or Patients with severe heart failure are depen- femoral arteries, arterial lines should be inserted prior dent on preload for ventricular filling and elevated

89 Section 2: Heart

sympathetictonetoensurecardiacoutputandtissue Preparation to come off bypass perfusion. Acutely unwell patients may already be on Preparation to discontinue CPB should be undertaken significant doses of inotropic agents preoperatively. as per any other cardiac case. The anesthetist should In other patients, commencement of an infusion of ensure that the patient’s electrolytes, pH status, and inotrope such as dopamine or epinephrine prior to oxygenation are adequate. Ventilation should be com- induction is prudent and may minimize hypotension menced gently to reinflate the lungs and aid in the and hemodynamic instability prior to commencement deairing process. TEE examination looking for resid- of CPB. ual air should be performed. The transplanted heart Although most anesthetic agents have been suc- is denervated, so reflex-mediated heart rate responses cessfully used in cardiac transplantation, all will unfa- will be absent. Baseline heart rate is usually around vorably alter hemodynamics. Traditionally, opioid- or 100 beats per minute in a denervated heart. Any benzodiazepine-based induction is used, with fentanyl ischemia or damage to the sinoatrial (SA) or atrioven- 5–10 ␮g/kg and midazolam 0.1–0.2 mg/kg commonly tricular (AV) nodes may result in severe bradycar- used. Ketamine has also been used at a low dose of dia. Indirectly acting chronotropic agents will be inef- 1–2 mg/kg. Anesthesia can then be maintained with fective, so direct-acting agents such as isoproterenol total intravenous anesthesia (TIVA), volatile agents, at 0.005 to 0.05 ␮g/kg/min are recommended for or a combination of both. Institutional or individ- chronotropic support. In addition, temporary epicar- ual anesthetist preference will ultimately decide which dial pacing wires may be used with AV sequential pac- drug combination is used. Regardless of the agents ing at approximately 90–100 beats per minute, as it used, dosing should be judicious and titrated carefully, may take several minutes for spontaneous rhythm to as overdose is easy in patients with low cardiac out- resume after release of the aortic cross-clamp. Fail- put state. Changes in preload, systemic vascular resis- ure of the return of SA node activity or adequate AV tance (SVR), and heart rate must be aggressively man- conduction requires permanent pacemaker implanta- aged with fluids, inotropes, and/or vasopressors. The tion early after transplant in 5–10% of patients after use of intraoperative transesophageal echocardiogra- HT following biatrial anastomosis, with a lower rate in phy (TEE) can be a useful guide to hemodynamic man- bicaval transplants. agement as well as diagnosing intracardiac thrombi, which are common in this population. Infection control and maintenance of aseptic tech- nique is essential due to infection risk. Most trans- Pharmacological support plant centers have protocols including prophylactic Initial pharmacological support is required during the antibiotics and pre- or intraoperative administration periodofweaningfromCPB,andthisinitialman- of immunosuppression including induction. agement is described here, with ongoing support and Since the voluntary withdrawal of aprotinin in choice of agent discussed in the following chapter. 2007, there has been much discussion in the litera- Mean arterial pressure should be kept above 60 mmHg tureabouttheuseofanti-fibrinolyticagentsincar- with vasopressor or inotropic agents as clinically indi- diac surgery. Currently, two lysine analogues are avail- cated. Ideally, short-acting, titratable agents should be able and commonly used (epsilon amino-caproic acid used. A mean arterial pressure greater than 80 mmHg and tranexamic acid). The Society of Thoracic Sur- is undesirable, and a short-acting vasodilator should geons Blood Conservation Guideline Task Force rec- be available for use if the need arises. Following sep- ommends the use of these anti-fibrinolytic agents as aration from bypass, taking care not to disrupt suture part of a comprehensive strategy to reduce blood loss lines, a pulmonary artery catheter can be passed into and transfusion requirements in routine and high- the pulmonary artery. The inotrope of choice will risk cardiac surgery. Many centers have also incor- depend on patient variables derived from this, TEE poratedtheiruseintocardiactransplantsurgery. evaluation, and institutional and the individual anes- There is a paucity of evidence regarding potential com- thetist’s practice. plications of the lysine analogues currently in use; Care must be taken with the use of systemic vaso- however, the risk of thrombotic complications appears pressors, because although they increase MAP and to be low. coronary perfusion pressure, they may also increase

90 Chapter 10: Management during surgery pulmonary vascular resistance and worsen right heart Right heart dysfunction function. Noradrenaline, adrenaline, dopamine, and Acute RV dysfunction is the most common cause of phenylephrine have all been used successfully in car- difficulty weaning from CPB, and this initial manage- diac transplantation surgery. Phenylephrine is most ment to aide weaning is described here. The etiology likely to worsen right ventricular (RV) function in is multifactorial: pulmonary hypertension is common patients with chronic pulmonary hypertension. Vaso- inthispopulation,andelevatedPVRwillbeworsened pressin has been used to elevate SVR and has minimal temporarily by the release of vasoactive substances effects on PVR. associated with cardiopulmonary bypass. The donor Isoproterenol is a non-selective beta-agonist. heart, although able to increase contractility, may be Along with its chronotropic properties, it has unable to cope with such a high afterload. Surgical inotropic activity and produces systemic and pul- manipulation, tricuspid or pulmonary insufficiency, monary vasodilatation and has been advocated as the denervation, and ischemia may all contribute to right inotrope of choice after cardiac transplantation. If heart dysfunction. Additionally, although temporary, used, it should not be discontinued abruptly, as PVR the RV is more commonly affected by intracoronary will rapidly rise to baseline levels. Dobutamine, a syn- air, as the right coronary artery is most anterior. This thetic catecholamine with predominantly beta-agonist can cause acute and catastrophic right heart ischemia effects and minimal alpha-agonist activity, is also a and failure until the air is expelled from the coronary useful agent when aiming to avoid vasoconstriction. vessel. Phosphodiesterase III inhibitors such as milrinone Right heart failure results in a reduction in pul- and enoximone have been used successfully due to monarybloodflowandleftventricular(LV)preload. their positive inotropic action coupled with vascular In addition, dilation of the failing RV causes the inter- smooth-muscle relaxing activity. These drugs are ventricular septum to bulge leftward, impinging on LV particularly useful, as this group of patients often have filling. The overall result is a reduction in cardiac out- beta-adrenergic receptor downregulation. Enoximone put and aortic root pressure. Reduction in coronary is incompatible with many other commonly used blood flow ensues and further ischemia occurs, lead- cardiac drugs and should be infused via a dedicated ing to a downward spiral of RV function. line. It is not uncommon for a combination of agents Right heart failure can be identified using TEE, to be required in order to achieve optimal cardiac and which will show a dilated, poorly contracting right vascular conditions. ventricle. The left ventricle will be underfilled, with a septum bulging leftward in systole. There may be Mechanical support severe tricuspid regurgitation secondary to a dilated tricuspid annulus. Mean PAP will be elevated, typically If cardiac function following bypass remains inade- above 25–30 mmHg. quate after implementation of all other strategies, the TreatmentofRVfailureinvolvestemporarysup- use of mechanical support should be considered. An portive measures until the ventricle recovers. In order intra-aortic balloon pump will assist in supporting the to maximize RV function, a number of strategies are left ventricle predominately but will also aid the right useful: ventricle by ensuring adequate perfusion pressure for 1. Optimizing ventilation to minimize PVR the coronary vessels and limiting ischemia. Temporary 2. Optimizing fluid status and RV preload left, right, or bi-ventricular mechanical assist devices, together with extracorporeal membrane oxygenation 3. Managing PVR to minimize RV afterload (ECMO), have been used successfully in cases of 4. Pharmacological and mechanically supporting RV extremely poor ventricular function after transplanta- function tion. The particular device used will depend on which 5. Preservation of coronary blood flow by ventricle is failing. Their use should be considered ear- maintaining aortic root pressure lier rather than later to avoid prolonged periods of organ hypoperfusion and ischemia. The heart usually Ventilation strategies for RV dysfunction recovers quickly, and nearly half of these devices may Although there is little evidence for any specific venti- be weaned within 4 days. lation strategies in preventing RV decompensation, it

91 Section 2: Heart

would seem reasonable to avoid factors that increase effects. Rebound pulmonary hypertension may occur pulmonary artery pressure (PAP) and employ strate- when it is withdrawn. Nitric oxide has been show gies known to reduce PAP. Avoiding hypoventilation to reduce PVR immediately after heart transplanta- by providing adequate minute ventilation and inspired tion by 50%, whereas PGE1 decreased resistance only oxygen is essential. Hypercarbia increases systolic PAP 10%. predominantly by increasing PVR. High inspired oxy- gen concentrations should be used to avoid hypoxic pulmonary vasoconstriction in the presence of low Transesophageal echocardiography alveolar oxygen concentrations. Although this reduces Recent updated guidelines from the American Society hypoxemia from shunting, it results in elevated PVR. It of Anesthesiologists and the Society of Cardiovascu- is also stimulated by acidosis and inhibited by alkalo- lar Anesthesiologists on TEE recommend its use in all sis. The use of positive end expiratory pressure (PEEP) open heart procedures. A complete examination of the in cardiac transplantation remains controversial. PVR heart should be performed prior to bypass in order to is minimal at functional residual capacity (FRC). The assess the likelihood of potential complications follow- judicious application of PEEP reduces atelectasis and ing transplantation of the donor heart and to look for aids in maintaining FRC. Excessive PEEP or high infla- intracardiac thrombi. If any thrombi are identified, the tion pressures, however, will cause a rise in PVR and surgeon should be alerted in order to avoid manipula- RV dilation. tion that may cause embolization. After CPB, the TEE should focus on the ventricu- lar function, particularly that of the RV, as described Fluid management in RV dysfunction earlier. It is often more practical to determine preload Fluid management in patients with a failing RV statuswithTEEratherthatintermittentpulmonary is critical, as the heart will be preload-dependent. artery catheter measurements. Filling can be under- Hypovolemia will result in reduced preload and RV taken with TEE guidance in order to avoid ventricular output, whereas any large increase in fluid volume over distension. The surgical anastomoses of the main may result in RV distension at the expense of LV pulmonary arteries should be evaluated with Doppler preload and output. Observation of the heart directly pressure gradients. Mild mitral, tricuspid, and pul- through the sternotomy, TEE, and CVP or pulmonary monary regurgitation is not uncommon after trans- artery catheter monitoring are all useful measures to plantation, but aortic regurgitation should be cause for employ when determining the volume of fluid to be concern. Distortion of the atria, depending on surgical administered. technique used, may cause significant mitral regurgita- tion. A higher incidence of systolic anterior motion of Selective pulmonary vasodilators the anterior mitral leaflet has also been described after for RV dysfunction heart transplantation. The use of pulmonary vasodilators in cardiac trans- plantation has been well studied. A number are avail- Management of coagulopathy able in current practice. Intravenous prostaglandin E2 Significant coagulopathy is not uncommon following (PGE2) is a potent pulmonary vasodilator that under- CPB in heart transplantation procedures. This may goes extensive metabolism in pulmonary vessels. It be due to residual preoperative anticoagulation, CPB- is widely used to facilitate weaning from CPB, as it induced platelet dysfunction, dilution of clotting fac- reduces PVR and improves RV function. Vasodilata- tors, and hypothermia. tion and hypotension due to its effects on SVR can limit Heparinshouldbereversedwithprotamineonce its use. Inotrope or vasopressors are often required to the patient is stable and off bypass. It should be admin- counteract this. istered slowly due to the risk of pulmonary vasocon- Another commonly used selective pulmonary striction and its effect measured using activated clot- vasodilator, inhaled nitric oxide, produces vasodi- ting time (ACT). Thromboelastography (TEG) can be lation of smooth muscle in pulmonary vessels by used to determine clotting status and has the advantage increasing production of cyclic GMP.Systemic absorp- that it can be performed while still on bypass (by using tion is minimal, resulting in fewer systemic side a cup with heparinase added to negate the effect of

92 Chapter 10: Management during surgery heparin) in order to allow time for thawing of fresh- Hand-over frozen plasma and cryoprecipitate if necessary. Formal Finally, there should be a careful and thorough hand- clotting studies should also be performed once CPB over to the team taking over the patient’scare following has been discontinued. transfer to the intensive care unit. All observations and Clotting factor concentrates, fibrinogen concen- interventions should be documented for later review if trates, and recombinant single factors (e.g., factor necessary. VIIa) can be considered in patients with clotting defi- ciencies who cannot tolerate administration of large fluid volumes. Platelet count and hemoglobin should Further reading be maintained as per institutional transfusion guide- Shanewise J. Cardiac transplantation. Anesthesiol Clin lines. Input from an experienced hematologist can help North Am 2004; 22: 753–65. guide management of any coagulopathy. Meticulous Thys DM, Abel MD, Brooker ER, et al. Practice guidelines surgical technique will aid in reducing blood loss. The for perioperative transesophageal echocardiography: an updated report by the American Society of surgical field should be thoroughly checked for bleed- Anesthesiologists and the Society of Cardiovascular ing points before the chest is closed and the patient Anesthesiologists Task Force on Transesophageal leaves the operating theatre. Echocardiography. Anesthesiology 2010; 112: 1084–96.

93 Section 2 Heart Chapter Postoperative care and early complications 11 Mandeep R. Mehra

Key points on achieving hemodynamic stability in the setting of renal dysfunction, pulmonary hypertension, fluid r In the immediate postoperative period, close overload, unexpected bleeding, and immunological attentionmustbepaidtohemodynamic interactions. stability by focusing on preventing right ventricular failure and maintaining chronotropic competence. Early allograft-related r Infection prophylaxis is essential and hemodynamic events structured protocols must be devised, Normal systolic function of the left ventricle (LV) targeted against superficial candida infection, coupled with restrictive physiology is typical; how- cytomegalovirus, toxoplasmosis, and ever, the function of the right ventricle (RV) can often anti-protozoal infections. r be variable. This is highly dependent on donor and Induction immunosuppression may be recipient size match, the pulmonary vascular resis- helpful when primary graft failure ensues, tance (PVR) of the recipient, cold ischemic time, and renal dysfunction forces a delay in aggressiveness of fluid loading. Furthermore, damage calcineurin inhibitor initiation, or a high risk to the susceptible RV can also ensue from air embolism for rejection is evident, as with a positive or technical difficulties such as pulmonary artery cross-match. torsion. r Statinsshouldbeusedearlyanduniversally Importantly, cardiac allografts that display excel- in the post-transplantation phase to improve lent early function can suffer a functional decline outcomes from rejection and cardiac over the first 6–24 postoperative hours. This may be allograft vasculopathy. due to development of RV failure or development of r Two distinct types of rejection exist, with myocardial edema that leads to restrictive physiology distinct allograft outcomes and responses to by diminishing diastolic filling and subsequent poor therapy. stroke volume. In this setting, the denervated cardiac allograft often is required to maintain overall cardiac output by a reliance on increased heart rate, and thus During procurement and implantation, the cardiac bradycardiaisverypoorlytolerated. allograft suffers a series of insults, ranging from the Bradycardia can occur from the effects of pre- effects of donor brain death and cold ischemic injury existing medications (particularly amiodarone), pro- during transport and then subsequent ischemia- longed cold ischemia, and even direct injury to the reperfusion injury as engraftment ensues. Immedi- sinoatrial (SA) node. The rhythm in these circum- ately, the denervated cardiac allograft encounters a stances is most likely an accelerated junctional rhythm, hostile circulation in the recipient, with aberrant neu- and such SA node dysfunction is usually transient, rohormonal reflexes and an inability to match its func- with full recovery occurring in 2–6 weeks. How- tion to the hostile vasculature. As such, the prin- ever, a permanent pacemaker may be required in ciple focus of early postoperative management is rare instances (Ͻ5%). The incidence of SA node

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

94 Chapter 11: Postoperative care and early complications dysfunction is low when a cavo-caval technique as Maintenance of hemodynamic stability opposed to a bi-atrial technique for transplantation is used. Temporary atrioventricular (AV) pacing (using 1. Choice of inotropic therapy: Isoproterenol has surgically implanted temporary wires), isoproterenol, been the time-honored agent used to provide andevenoralbeta-2receptoragonistscanbeused hemodynamic support. This is because of its to maintain cardiac output by increasing the heart chronotropic and pulmonary vasodilating rate in this situation. The general clinical recommen- properties in addition to inotropism. However, dation is to maintain the heart rate between 90 and in the setting of a cavo-caval transplant, the 110 bpm; however, a higher rate is recommended chronotropic property may cause excess if significant donor under-sizing with respect to the tachycardia because the incidence of SA node recipient (donor weight Ͼ30% below recipient weight) dysfunction is reduced by that technique. The occurred. combination of dobutamine with temporary The cardiac allograft may require inotropic sup- pacing may allow for similar outcomes without port for the first 24–72 hours as it begins to replen- the excess tachycardia, but effects on pulmonary ish itself and adapt to the new circulation. Early vasodilation are less forthcoming. Dopamine on, myocardial contractility may be decreased due to alone may not be effective as it requires an intact edema. In addition, myofibril necrosis can also occur, nerve supply to the cardiac allograft. an event that can be picked up by serum biomarker 2. Unique vasodilators: Prostacyclins, such as PGE1 assessment. or PGEI2, are used to maintain low pulmonary Primary graft failure in the absence of hyperacute pressures and may improve renal perfusion. rejection after cardiac transplantation is said to occur However, when systemic hypotension is a if more than two inotropes, or the need for mechan- concern, inhaled nitric oxide may be used. ical circulatory support, is required early (24 hours) 3. Systemic hypotension: More often than not, after engraftment. Isolated RV failure is more com- systemic blood pressure is reduced after cardiac mon than bi-ventricular failure. This is typically exhib- transplantation. This is due to the presence of ited by elevated right atrial pressure ≥20 mmHg with acidosis and cytokine-related vasodilatation. a left atrial pressure ≤12 mmHg, elevated PVR, and Continuous infusion of vasoconstrictor such as resultant decreasing cardiac output. Typically, these phenylephrine, norepinephrine, or vasopressin are accompanied by instability of the peripheral circu- may be required. The guanylate cyclase inhibitor lation and hypotension. methylene blue has been successfully used for the Primary cardiac allograft failure accounts for 40% treatment of catecholamine-refractory of mortality within 30 days of heart transplantation vasoplegia. (HT) and 18% of mortality for the second through the 4. Mitral and tricuspid valve regurgitation: Early 12th months. General risk markers include preoper- after transplantation, it is not uncommon to ative use of a mechanical circulatory support device, discover mild grades of mitral insufficiency. This female donors, and “marginal” donor hearts (e.g., is due to poor coaptation of the mitral leaflets as a positive viral serology, older donor, diabetic donor, result of loss of diastolic function of the left atrium reduced cardiac function, coronary artery disease, and development of mechanical dyssynchrony. moderate to severe left ventricular hypertrophy, and Active management is usually not necessary. On high preoperative catecholamine requirements). the other hand, the occurrence of tricuspid valve Preoperative support of the recipient circulation disease is vastly more common, a result of the by mechanical assist devices appears to significantly effects of damage to the right ventricle and in the increase the risk of post-transplantation primary graft setting of pulmonary hypertension. In situations failure. The reasons underlying this observation are of persistent hemodynamic compromise and unclear but may have to do with the general inflam- severe dilation of the tricuspid annulus, valve matory response provoked by the prolonged opera- repair can be considered. In small series, this has tion that is required to separate the recipient from the been shown to improve cardiac allograft mechanical device. Whether other articulated associa- outcomes. Typically, the RV remodels favorably tions with primary cardiac failure truly influence sur- and the tricuspid regurgitation either reduces or is vival is uncertain. eliminated by 1 year.

95 Section 2: Heart

5. Pericardial effusion: Although commonly worsening of pre-existing diabetes, stress-induced encountered in one fifth of transplanted hearts, uncovering of hyperglycemia, or this not usually hemodynamically significant and corticosteroid-induced hyperglycemia. Because resolves within 4–6 weeks after transplantation. advanced diabetes can cause worsening of acidosis Vigilance for a pericardial hematoma and and increased propensity for infection, it is compressive physiology must be maintained and important to control blood sugars within a range resternotomy may be required. of Ͻ200 mg/dl blood glucose. Importantly, metforminshouldbewithheldbecauseitcan precipitate lactic acidosis. The development of Atrioventricular arrhythmias hypoglycemia must be avoided because in Despite the physiological insults and pathological non-transplant settings, hypoglycemic episodes injury accorded to the donor allograft, the frequency are associated with worse cardiac outcomes. of AV arrhythmias is relatively low. Atrial fibrillation 2. Renal failure: Many patients have pre-existing is more common than atrial flutter and occurs with a renal failure. Additionally, the exposure to frequency of 7–25%. It should be noted that this fre- hemodynamic instability and fluid shifts in quency is lower than the anticipated frequency in com- addition to the pro-inflammatory effects of parison with traditional cardiac surgery, and although cardiopulmonary bypass worsen kidney function. thereasonsarenotclear,theuseofsteroidshasrecently Strategies include optimizing RV function and been shown to lower the risk of atrial fibrillation in reducing exposure to nephrotoxic drugs and generalcardiacsurgery.Becausethesepatientsare immunosuppressant. If renal replacement therapy ubiquitously given steroid therapy, one can speculate is required (1–15% of patients) early after that this may be an operative factor in determining a transplantation, outcome is quite poor (mortality lower than expected occurrence of atrial arrhythmias. up to 50%). Despite this tendency to worse Ventricular arrhythmias are even uncommon and typ- outcome, consideration should be given to early ically seen in Ͻ2% of cases. When they occur, the renal replacement therapy if there is renal failure etiology can usually be traced to excessive patholog- in the context of RV dysfunction and rising right ical ischemic injury in the allograft or to a causative atrial pressures. drug exposure. Although infrequent, some principles 3. Infection prophylaxis: Early after transplantation, oftherapymustbekeptinmind.Digitalisanditsana- bacterial infections predominate, and therefore, logues do not work effectively due to vagal denerva- institution-specific flora and sensitivity-based tion. Beta-blockers can slow the rate of atrial arrhyth- regimens must be used. Prophylaxis is similar to mias, but in the early postoperative period, their that of other transplants, as described in Chapter use can worsen RV failure and hemodynamics. Some 4C. Importantly, cytomegalovirus (CMV) intensify anti-rejection therapy when atrial arrhyth- infection is a critical target for prophylaxis, and mias are encountered. However, it is rare for rejection the appropriate regimen is based on the donor to be a culprit in those arrhythmias occurring within and recipient serological status. Those at high risk the first 2 weeks of engraftment. Typically, a rapid of primary infection (donor positive/recipient reduction in catecholamine exposure coupled with negative) or reactivation (recipient positive) anti-arrhythmic therapy is usually effective. Amio- should receive targeted prophylaxis with either daronecanbeusedinthesepatients,butdrugslikecal- oral valganciclovir (900 mg daily) or initial cium channel blockers (negative inotropism), sotalol 4–6 weeks of intravenous ganciclovir (beta blockade and renal dysfunction), and dofetilide (5–10 mg/kg/day). Some centers use (multiple drug interactions) should be avoided. CMV-specific immunoglobulin initially, and this has support from observational trials in reducing Special early perioperative infection-related complications and also the occurrence of cardiac allograft vasculopathy. Oral circumstances therapy is usually continued for 3 months. 1. Diabetes mellitus: Postoperative hyperglycemia is Similarly, therapy to prevent toxoplasmosis and an important occurrence and may be due to pneumocystis jiroveci infection using

96 Chapter 11: Postoperative care and early complications

trimethoprim-sulfamethoxazole–based therapy, the destruction of the therapeutic antibodies by or, if allergic, dapsone or pyrimethamine is also the recipient’s anti-mouse antibodies (daclizumab: advocated. At the time of hospital discharge, 10% murine, 90% human protein; basiliximab: 30% anti-fungal prophylaxis to prevent murine, 70% human protein) and the development mucocutaneous candidiasis should be initiated of serum sickness associated with mouse-, rabbit-, or with nystatin (4–6 mL [400 000–600 000 units] horse-derived proteins. Alemtuzumab has been asso- 4 times daily, swish and swallow) or clotrimazole ciatedwithprolongedleukopeniainrenalandlung lozenges (10 mg). transplant recipients, and few multi-center data exist in cardiac transplantation. Of the polyclonal antibod- ies,ATGhasthemostsupport.Inrecentyears,investi- Clinical principles of gators have evaluated a shorter ATG course (5 versus 7 immunosuppression days) or adjustment of ATG dose to achieve a lympho- Development of a rational immunosuppressant cyte count below Ͻ100/␮l. Shorter duration of ATG regime requires assessment of the donor and recipient therapy was associated with higher rejection rates. In immunological match as well as comorbidities likely contrast, adjustment of ATG doses according to T- to interact with the immunosuppressants within cell counts was associated with lower rejection rates the host environment (see Chapter 3). Following as well as lower or fewer ATG doses. Although there HT, the use of intraoperative and peri-operative is little universal support for using induction therapy, corticosteroids remains the mainstay of early therapy. it should be considered in special circumstances in Typically, high-dose corticosteroids are administered which a calcineurin inhibitor (CNI) delay is desired before induction of anesthesia and during reperfusion due to renal insufficiency. However, at this time, CNI- after implantation. Thereafter, induction therapy may free protocols cannot be considered a standard of care or may not be used, followed by the establishment in de novo HT. In situations in which the risk of early of a calcineurin-based regimen (cyclosporine or rejection is high (sensitized patients with a positive tacrolimus) with an adjunctive agent (mycophenolate cross-match) or when primary graft failure has ensued, mofetil, mammalian target of rapamycin [mTOR] the use of induction agents may be particularly useful. inhibitors, or azathioprine). In all cases, steroids are prescribed with intent to rapidly wean off by 6 months. Calcineurin inhibitors and Up to 44% of all cardiac transplant recipients are able to wean off steroids completely by 1 year. adjunctive therapy The major randomized clinical trials of immuno- suppression in heart transplantation have allowed Induction therapy us to alter our clinical management options, but The concept of induction therapy applies to the use a close examination of these data suggest that one of very intensive immunosuppression designed to can interpret the findings broadly. On an intent-to- deplete entire cell lines in an effort to abrogate the treat basis, these trials have not shown differential immune response. Unlike in other organ transplants, effects on survival with various immunosuppressive the use of polyclonal or monoclonal antibodies in HT regimens. Although the 1998 mycophenolate mofetil is unproven. Muromonab-CD3 (OKT3), the old mon- (MMF) trial resulted in a large shift in replacing aza- oclonal antibody of choice, has now been replaced by thioprine with this agent, the data show that MMF anti–interleukin-2 (IL-2) receptor blockers (used in did not improve 1-year survival compared with aza- approximately 30% of induction cases). More com- thioprine (AZA). However, in this study, random- monly, the new polyclonal preparations (Thymoglob- ization occurred preoperatively, and 11% of recipi- ulin, antithymocyte globulin [ATG]) have replaced the ents never received the study drug. When the anal- antibodies used in the past (ATGAM). Two CD25 ysis was restricted to patients who received at least (IL-2 receptor blockers) specific chimerical mono- one dose of MMF (treated-patient analysis), 1-year clonal antibodies, daclizumab and basiliximab, have survival was greater in the MMF than in the AZA been designed to reduce the limitations of non-human group (6.2% versus 11.4%; p = 0.031). Two tri- antibodies. The inclusion of human proteins prevents als of CNIs within Europe and the United States

97 Section 2: Heart

comparing tacrolimus (TAC) and cyclosporine (CyA)- and15ng/mlduringtheearlypostoperativeperiod based immunosuppression found similar rejection (days 0–60), between 8 and 12 ng/ml for the next 3–6 rates. More recently, a three-arm trial comparing months, and between 5 and 10 ng/ml in stable patients regimens of TAC/MMF, TAC/sirolimus (SRL) and 6monthsafterHT. CyA/MMF showed that both TAC-based regimens Therapeutic drug monitoring for mTOR inhibitors, were associated with significantly lower 6-month rates SRL, or everolimus (EVL) using trough concentra- of any treated rejection than the CyA/MMF regi- tion levels is recommended. Levels should be mea- men. Furthermore, TAC/MMF-treated patients had sured at least 5 days after adjustment of the dose, when lower rates of cellular rejection (International Soci- a new steady state is achieved. When used in com- ety for Heart and Lung Transplantation [ISHLT] grade bination with CyA, the optimal trough target levels Ͼ3A) and of any treated rejection than the CyA/MMF- rangeforEVLisbetween3and8ng/ml.Thecorre- treated subjects. Another trial suggested that the com- sponding optimal trough level range for SRL is 4 to bination of TAC/MMF is ideal across ethnic popula- 12 ng/ml. The optimal levels for these agents with TAC tions, with evidence of enhanced survival in African- remain uncertain, and the safety of this combination American heart transplant recipients. Thus the combi- remains in doubt. Routine therapeutic drug monitor- nation of TAC/MMF appears to possess the most opti- ing of mycophenolic acid levels to adjust MMF doses mum risk-benefit ratio in treating cardiac transplant is not recommended. recipients and may therefore represent the drug strat- egy of choice. Statins Three-hydroxy-3-methylglutaryl coenzyme A (HMG- Therapeutic drug monitoring CoA) reductase inhibitors, or statins, function as both lipid-lowering and immunomodulating agents. Statins Monitoring of therapeutic drug levels is important but are used early post-transplant irrespective of lipid lev- there is some controversy in how best to monitor the els in an effort to employ their immune-modifying target levels of CNIs. At present, 2-hour post-dose properties. Evidence with pravastatin and then subse- (C2) levels should not replace 12-hour trough (C0) quently with simvastatin has suggested that this may concentrations for routine monitoring of CyA expo- represent a class effect of immunomodulation and sure in most patients, but may be useful in selected anti-inflammatory effects. In aggregate, the evidence patients in whom a better characterization of the supports a reduction in mortality, decrease in hemody- pharmacokinetic profile of CyA is desired. Measure- namically compromising rejection, and amelioration ment of 12-hour trough CyA concentration is the of cardiac allograft vasculopathy. One must remain recommended form of therapeutic drug monitoring vigilantforthedevelopmentofsympathyorrhab- for routine clinical use. In general, when used in domyolysis with statins in this vulnerable population, conjunction with azathioprine, or MMF preparations, particularly with simvastatin, due to interaction with the average CyA trough concentration target using CNIs. the Abbot TDX assay (or equivalent) is 325 ng/ml (range 275–375 ng/ml) for the first 6 postopera- tive weeks, 275 ng/ml (range 200–350 ng/ml) for Clinical principles in rejection weeks 6–12, 225 ng/ml (range 150–300 ng/mL) for surveillance, detection, and months 3–6, and 200 ng/ml (range 150–250 ng/mL) from month 6 onward, although this needs to be management balanced against renal function. Measurement of 12-hour trough concentration for Typesofrejection twice-daily TAC and a 24-hour trough concentration Hyperacute rejection, which occurs within hours of for once-daily TAC is the recommended drug moni- engraftment,isuniformlyfatalbutveryrare,due toring method for routine clinical use. The therapeutic to immunological matching and monitoring. On the range of TAC levels varies depending on concomitant other hand, acute cellular rejection (ACR) is most drugs, toxicity concerns, and time after HT. In gen- common in the first 6 months after heart transplan- eral, when used in conjunction with AZA or an MMF, tation and is a predominantly T-cell–mediated phe- TAC trough concentration targets range between 10 nomenon. It occurs in 20% of recipients in the first

98 Chapter 11: Postoperative care and early complications Screen - Clinical evidence of graft dysfunction2 Unresolved Questions - Histologic findings of capillary endothelial Should screening for AMR be 1 swelling, macrophage activation, 1 +/− interstitial edema and/or hemorrhage3 routinely done early on? 2 Should subclinical AMR be disregarded?

3 Are histologic findings alone adequate to screen for AMR? Absent Present 4 What specimens and stains should be used?

Confirm Which analytes should be measured? - Positive immunofluorescence 4 5 When and how often should or immunoperoxidase DSA be monitored? - √ serum for donor-specific antibodies5 Should DSA be considered a risk factor or diagnostic criterion for AMR?

6 Should the serverity of AMR AMR 0 (EMB negative for AMR)6 AMR 1 (EMB positive for AMR)6 be graded?

Figure 11.1 Understanding antibody-mediated rejection (AMR) in cardiac transplantation. DSA, donor-specific antibodies; EMB, endomyocardial biopsy. Reprinted with permission from Kfoury AG, Hammond ME, Controversies in defining cardiac antibody-mediated rejection: need for updated criteria, J Heart Lung Transplant 2010; 29: 389–94. year, and this frequency is declining as a result of with symptoms of allograft failure, and echocardiog- improvements in the intensity of immunosuppres- raphy will signal the presence of myocardial thicken- sants. More recently acknowledged as a clinically valid ing (due to edema) or restrictive physiology. In overt entity, acute antibody-mediated rejection (AMR) is less rejection, systolic dysfunction ensues, signaling the common than ACR, occurring in approximately 10% immense gravity of this rejection episode. of patients, but is more dreaded because it is accompa- Early after transplantation, particularly in the first nied by hemodynamic compromise, is difficult to treat, 3 months when the risk of rejection is highest, inva- and leads to late sequelae, particularly cardiac allo- sive biopsies are recommended at decreasing intervals. graft vasculopathy. Acute AMR consists of antibodies Thus weekly biopsies are advised in the first month, directed against donor vascular endothelial antigens. fortnightlyinthenextmonth,andthenmonthlythere- It is not uncommon for both ACR and AMR to coex- after until 6 months. After that a lower frequency is ist. The allograft demonstrates recruitment of neu- undertaken, usually at 3 monthly intervals until the trophils, interstitial edema, and intravascular throm- end of the first year. There is much controversy regard- bus and subsequent myocyte necrosis. The presence ing the value of performing routine biopsy beyond the of immunoglobulin (IgG, IgM, or IgA), complement first year of transplantation (Figure 11.2). An invasive fragments (C3d, C4d, C1q), or CD68-positive cells biopsy is reassuring but is not without risk. Compli- (macrophages), as well as the appearance of circulating cations may occur in 2–3% of procedures, including de novo anti-donor human leukocyte antigen (HLA) access-related vascular or bleeding problems, pneu- antibodies, are sentinel peripheral and allograft events mothorax, arrhythmias, heart block, or even tricuspid that form the signature for AMR (Figure 11.1). valve injury. Pericardial tamponade can occur due to right ventricular perforation. Monitoring for rejection Because the signs and symptoms of rejection are few, Classification of rejection surveillance with invasive endomyocardial biopsy is The histology of the allograft is evaluated patho- advised. In advanced cases, the patient may present logically based on a uniform ISHLT grading scale,

99 Section 2: Heart

Period of Immunological Adjustment (I) followed by immediate return to the baseline dose is quite adequate in treating the episode to histological resolution. Background immunosuppression must Period of Allograft Adaptation (II) be adjusted by evaluating trough levels, and attempts to wean corticosteroids must be slowed down. Most episodes of ACR resolve within 3 weeks. Interestingly, Period of Allograft Maintenance (III) high two thirds of ACR episodes that are untreated tend to resolve spontaneously as well. This data accrues from ACR risk low

≤ 2 mos later phases of transplantation, and most clinicians > 2-6 mos > 6–12 mos would rather err on treating during the early phase of > 12 mos transplantation. However, it is perhaps prudent to not I. Intense Vigilence II. Gene-based Risk Stratification III. Clinical and Functional Evaluation over-immunosuppress mild episodes, as the incidence of infection may increase. An assay that measures Figure 11.2 Phases of cardiac adaptation and time-dependency of cardiac rejection. Reprinted with permission from Kfoury AG, activated T lymphocytes by assessing adenosine Hammond ME, Controversies in defining cardiac triphosphate (ATP) production (Cylex) may hold antibody-mediated rejection: need for updated criteria, JHeart promise in allowing clinicians to evaluate the balance Lung Transplant 2010; 29:599–602. of immunosuppression and infection risk. The treat- ment of AMR is quite distinct from that of ACR, and initially described in 1990 and subsequently updated although the scope of this chapter is to not detail these in 2005 to include a definition for AMR. Due to intra- issues, it is important to provide some clinical impli- and inter-observer variability in the determination of cations of available therapy. There are three distinct the different grades of mild or moderate rejection and goals of AMR therapy, which include protection of the theobservationthatgrades1and2weremostlyself- allograft from further injury, reduction of circu- limiting, the revised heart allograft rejection grading lating donor-specific antibodies, and suppression system in 2005 divided the various grades; grade 0 of production of these deleterious antibodies. The (no cellular rejection) was now named grade 0R (“R” allograft is typically protected by the administra- added to reflect the revised 2005 scale). The interme- tion of high-dose steroids and therapy targeted diate grades of 1A, 1B, and 2 were re-classified as grade to prevent thrombosis. Circulating antibodies are 1R, or mild acute cellular rejection. Grade 3A was re- decreased by the use of plasmapheresis, typically used classified as grade 2R, moderate acute cellular rejec- 3–5 consecutive times followed by intravenous tion, and grades 3B and 4 were re-classified as grade IgG; cytolytic induction antibodies such as ATG 3R, severe acute cellular rejection. In addition, AMR and rituximab are then used to diminish T- and was recognized as a clinical entity, and recommen- B-cell responses. In the late phase, suppressive therapy dation was issued for determination of its presence with cyclophosphamide or, in recalcitrant cases, (AMR1) or absence (AMR0). Non-invasive monitor- photopheresis or total lymphoid irradiation may ing for rejection has been attempted, and more recently be used. Close vigilance for re-emergence of circu- a gene based biosignature has shown promise. How- lating antibodies is needed, and newer approaches ever, this particular signature should not yet be applied using complement inhibitors or intensive B-cell in the early post-transplant period, within the first modulating drugs such as bortezomib are being 6months. studied. The long-term outcomes for these patients deserves further study because it is uncertain whether these anecdotally derived treatments influence late Management of rejection outcomes from AMR. Early after transplantation, even asymptomatic allograft histology restricted ACR is treated with Further reading augmentation using corticosteroids. Although most Costanzo MR, Dipchand A, Starling R, et al. The use a high-dose intravenous pulse of steroids daily International Society of Heart and Lung Transplantation for 3 days, in the absence of allograft failure, oral Guidelines for the care of heart transplant recipients. J steroids at a dose of 1 mg/kg of prednisone for 3 days Heart Lung Transplant 2010; 29: 914–56.

100 Chapter 11: Postoperative care and early complications

Kfoury AG, Hammond ME. Controversies in defining heart transplantation. JHeartLungTransplant2009; 28: cardiac antibody-mediated rejection: need for updated 213–25. criteria. JHeartLungTransplant2010; 29: 389–94. Uber PA, Mehra MR. Induction therapy in heart Kobashigawa J, Mehra M, West L, et al. Report from a transplantation: is there a role? JHeartLungTransplant consensus conference on the sensitized patient awaiting 2007; 26: 205–9.

101 Section 2 Heart Chapter Long-term management and outcomes

12 Hari K. Parthasarathy and Clive J. Lewis

management, cardioplegia, and intensive care, but par- Key points ticularly immunosuppression with the introduction of r Heart transplantation has excellent cyclosporine. The International Society for Heart and long-term survival, with 50% of patients Lung Transplantation (ISHLT) Registry is the largest living 10 years, and significant improvement dataset for heart transplantation worldwide and shows in quality of life. the half-life to be 10 years following transplant, with a r Early graft failure, rejection, and infection half-life conditional on surviving the first year of 13 contribute most to mortality in the first year. years (Figure 12.1). Importantly, there is also signifi- Cardiac allograft vasculopathy and cant improvement in functional status and quality of malignancy contribute most to late mortality. life, with more than 90% of recipients having no func- r Complications continue to limit long-term tional limitations during the first 7 years, more than survival following heart transplantation. 50% working or retired at 1, 3, and 5 years, and re- r The incidence of vasculopathy at 10 years hospitalization in fewer than 25% of patients 1 year after heart transplant is 50% and causes late after heart transplant. graft dysfunction, myocardial infarction, Between 1982–1991 and 2002–2007, survival arrhythmia, and sudden cardiac death. increased by 4% at 1 month and 6% at 6 months r Malignancy, severe renal dysfunction, and in spite of higher risk donor and recipient profiles. metabolic syndrome (hyperlipidemia, The improvement in early mortality following heart hypertension, diabetes and obesity) are transplantation has led to a longer half-life, from common comorbidities both early and late 8.8 to 10.5 years, in the two decades above. Short after heart transplantation. and long-term survival is also determined by the underlying diagnosis: cardiomyopathy (84% 1-year survival), coronary artery disease (82%), valvular Since Christiaan Barnard performed the first human heart disease (78%), congenital heart disease (76%), heart transplantation over 40 years ago, more than and re-transplantation (68%), and this effect persists 85 000 heart transplants have been performed world- in subsequent years. Survival has greatly improved in wide. On average, more than 5000 heart transplants recent patient cohorts undergoing re-transplantation, are undertaken every year in more than 225 cen- mainly attributed to careful patient selection and ters. Understanding the short- and long-term survival improvement in post-transplant management. is important to determine the place of transplanta- Disappointingly, however, long-term complica- tion within the success of other medical and surgi- tions continue to limit survival, with an attrition rate cal therapies. In the early years, survival after heart of 3–4% annually in all decades, including the most transplantation was limited to the medium term, with recent. Cumulative causes of death from the ISHLT mortality due to rejection and severe infection (1-year Registry following transplantation are shown in Fig- survival 30% in 1967–1973, 60% in 1974–1980). Sub- ure 12.2. Understanding causes of death after trans- sequently, survival has increased with each decade, plant continues to have a significant impact on direct- although the major improvement has been during the ing preventative, surveillance, and treatment strategies first year after transplant due to advances in donor to reduce the incidence of long-term complications.

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

102 Chapter 12: Long-term management and outcomes

HEART TRANSPLANTATION Figure 12.1 Survival following heart transplantation in adult transplant recipients Kaplan-Meier Survival from ISHLT Registry, 1982–2008. From JHeart (January 1982 – June 2008) Lung Transplant, 2010; 29(10): 1083–141. 100 Reproduced with permission from the Half-life = 10.0 years International Society for Heart and Lung Conditional Half-life = 13.0 years Transplantation. 80

60 N = 80,038 N at risk at 23 years = 124 40 Survival (%) Survival 20

0 0123456789101112 13 14 15 16 17 18 19 20 21 22 23 Years

ADULT HEART TRANSPLANT RECIPIENTS: Figure 12.2 Cumulative incidence of leading causes of death in adult transplant Cumulative incidence of Leading Causes of Death recipients from ISHLT Registry, 1992–2008. (Transplants: January 1992 – June 2008) From J Heart Lung Transplant, 2010; 29(10): 10% CAV Acute Rejection 1083–141. Reproduced with permission from Malignancy (non-Lymph/PTLD) Graft Failure the International Society for Heart and Lung 9% CMV Infection (non-CMV) Transplantation. 8% 7% 6% 5% 4% 3% 2% 1% 0% Incidence of Cause-Specific Deaths 012345678910 Time (years)

Mortality following heart 32% of deaths 5 years following transplantation, fol- lowed by malignancy (23%) and non-CMV infections transplantation (10%). The outcome following transplantation can be use- fully divided into short (30 day and 1 year) and long- term mortality (Ͼ1year).Themajorcausesofdeath alter with the length of survival after transplant (Fig- Mortality during the first year ure 12.3). Within the first 30 days post-transplant, graft Various factors contribute to increased early graft fail- failure is the leading cause of death (41%), followed by ure and morality, including changing donor and recip- multi-organ failure (13%) and non-cytomegalovirus ient profiles in recent yearsTable ( 12.1). Continu- (CMV) infection (13%). Between 31 and 365 days, ous risk factors for worse outcome include increas- non-CMV infection is the leading cause and accounts ing recipient age and body mass index (BMI), increas- for 30% of the deaths, followed by graft failure (18%) ing donor age and BMI, smaller transplant center and acute rejection (12%). Long-term mortality is due volume, longer ischemic time, borderline pulmonary to coronary artery vasculopathy (CAV) and late graft wedge pressure, and rising bilirubin, creatinine, and failure (likely due to CAV), which together account for pulmonary vascular resistance (PVR).

103 Section 2: Heart

ADULT HEART TRANSPLANT RECIPIENTS: Figure 12.3 The relative incidence of causes of death in adult transplant Relative Incidence of Leading Causes of Death recipients over time from transplant, ISHLT (Deaths: January 1992 – June 2009) Registry, 1992–2008. From JHeartLung 50 CAV Acute Rejection Transplant, 2010; 29(10): 1083–141. Reproduced with permission from the Malignancy (non-Lymph/PTLD) Infection (non-CMV) International Society for Heart and Lung Graft Failure 40 Transplantation.

30

20

10 Percentage of Deaths Percentage

0 0-30 Days 31 Days to >1 Year to >3 Years to >5 Years to >10 to 15 >15 Years (N = 3,771) 1 Year 3 Years 5 Years 10 Years Years (N = 1,075) (N = 3,675) (N = 2,848) (N = 2,448) (N = 5,592) (N = 3,070)

Donor selection Acute cellular rejection contributes to up to 40% of early death following transplant and is especially Patient selection for transplant remains the most common in the first month and months 2–6 and is important predictor of outcome, with the adage unusual after the first year, but the risk of fatal rejection “poor donor and poor recipient leads to poor out- never disappears completely. Some centers continue comes.” Optimization and careful selection of the to biopsy annually in subsequent years, but studies donor (Chapter 7) prior to harvesting the organ will have shown no difference in outcome without annual improve outcomes. Careful selection and optimiza- surveillance. Histological rejection grade 2R requires tion of sick heart failure patients is critically impor- treatment, especially when associated with allograft tant to reduce the chance of early and late mortality. dysfunction (ISHLT Consensus on Diagnosis of Heart A range of treatments from conventional heart fail- Rejection, 2005). Treatment of lower grades of rejec- ure drugs to cardiac resynchronization therapy (CRT) tion (1R) is controversial but is warranted if recurrent and implantable cardioverter-defibrillator (ICD) mean within the first 6 months. Acute cell-medicated rejec- that patients are often referred with very poor car- tion may be treated conventionally as for other organs diac reserve and are too sick to remain ambulant while with high-dose steroid, augmenting or switching oral awaiting transplantation. Patients may be optimized immunosuppression, intravenous (IV) antithymocyte in high-dependency wards with inotrope, intra-aortic globulin (ATG)/muromonab-CD3 (OKT3), plasma- balloon pump, ultra/hemofiltration, and mechanical pheresis, and total lymphoid irradiation, as discussed support to improve otherwise adverse prognostic indi- in Chapter 11. cators for mortality after transplant. Antibody or humoral rejection is increasingly rec- ognizedasanimportantformofrejectionevenwith- out evidence of acute cellular rejection and leads to Rejection reduced graft survival and increased incidence of early vasculopathy. Strategies for treatment include ritux- Hyperacute rejection is rarely seen after heart trans- imab, plasmapheresis, and immunoglobulin. Risk fac- plantation and is usually due to a rapid over- tors for rejection are shown in Table 12.2. whelming immune attack on the graft due to pre- formed antibodies directed against human leukocyte antigen (HLA) or ABO blood group systems but may also occur with non-HLA antibodies. Avoiding trans- Infection plant against known recipient HLA antibodies and Infection is the cause of death in 10–15% of patients ABO-incompatible transplants should avoid hypera- duringmonth1andtheleadingcauseofdeathin cute rejection. the first year. Decreased resistance to infection arises

104 Chapter 12: Long-term management and outcomes

Table 12.1 Donor and recipient risk factors associated with Table 12.2 Risk factors for acute cellular rejection increased early graft failure and mortality 1 year following heart transplantation Younger recipient age Donor factors Black or Hispanic recipient Increasing age (25% of recipients in current era, RR = 1.5 age 40 CMV mismatch years, 2.4 age 50 years) HLA mismatch Decreasing BMI Older donor age Increasing allograft ischemic time (RR = 1.4 after 5 hours) High TPG (may represent poor cardiac reserve to hemodynamic High-dose inotropic support stress of rejection) TPG: transpulmonary gradient. Reduced systolic function “Marginal” or extended criteria donors (especially in higher risk recipients) Recipient factors occur in up to 20% of recipients but does not pre- Significant donor/recipient size mismatch (especially female dict worse outcome, although there may be a higher donor to male recipient) incidence of early vasculopathy. Antibiotic prophy- Younger and older age (RR = 1.33 for 18-year-old; survival 77% laxisshouldroutinelybegiveninthepostoperative over 70 years vs. 84% 30–39 years) period. Longer term prophylaxis is given for pneu- Increasing BMI (RR = 1.3 over BMI of 34) mocystis and toxoplasmosis with cotrimoxazole (may Decreasing transplant center volume (RR = 0.9 for Ͼ 55 bestoppedwhenlevelsofimmunosuppressionare transplants/year) lower); CMV (unless donor and recipient CMV neg- Increasing pretransplant serum bilirubin (RR = 1.3 Ͼ 3.5 mg/dl) ative) with IV ganciclovir then oral valganciclovir for and creatinine (RR = 2.1 Ͼ 2.5 mg/dl or dialysis requirement) 3–12 months; and herpes viruses with acyclovir if Pulmonary vascular resistance Ͼ 3–5 Wood Units (PVR Ͼ 5 not on valganciclovir. Physicians need a systematic increases mortality by 3% and 5% at 1 and 5 years; survival approach to diagnosing infection, with consideration improved by donor over-sizing) given to time since transplant, donor infection, recip- Ͼ Trans pulmonary gradient 15 ient serology, degree of immunosuppression (recent High right atrial pressure treatment for rejection), and recent exposure. Ongoing Higher rates of diabetes mellitus post-transplant surveillance by specialists is impor- Higher rates of nicotine use tant to detect unusual infection. Up-to-date immu- Recipient panel-reactive antibody (PRA) Ͼ 10% nization should be checked pretransplant and continue Congenital heart disease (RR = 2.3) or coronary artery (except live vaccine) as usual post-transplant, includ- disease (RR = 1.2) ing annual influenza, pandemic influenza, and pneu- Ventilated patients (RR = 1.6) mococcal vaccination. Dialysis at transplant (RR = 1.6) Recent infection Survival after the first Higher incidence of temporary (RR = 3) or long-term left ventricular assist device (LVAD, RR = 1.25) post-transplant year Compatible but non-identical ABO matching (RR = 1.2) Coronary vasculopathy and malignancy account for LVAD: left ventricular assist device; RR: relative risk. more than 50% of the mortality at 5 years. The risk fac- tors for 1-year mortality remain powerful predictors for5-yearoutcomes(Table 12.3). The factors affecting from recipient older age, high or low BMI, lung dis- 5-year survival also affect the longer term survival to ease, pretransplant ventilation, smoking history, dia- 10 and 20 years or longer. It is important to note that betes, previous intervention (sternotomy, mechani- the data collected for these analyses are not as com- cal support), and critically illness. As in other organ prehensive as that of the modern era, and the num- recipients, following heart transplantation, bacterial ber of patients reaching more than 20 years is small and viral infections are common, but consideration (144 at 22 years in the ISHLT dataset). Changes in should be given to protozoal and fungal infection, recentyearsthathaveimprovedverylong-termsur- which carry the highest mortality. CMV infection may vival include modern immunosuppression, statin use,

105 Section 2: Heart

Table 12.3 Risk factors affecting 5-year survival conditional on Table 12.4 Comorbidities 10 years following heart surviving the first year post-transplant transplantation Donor factors Angiographic CAV, 52% Older age (RR = 1.35 age 55 years) Severe renal insufficiency, 14% Recipient factors Dialysis, 4.7% Allograft vasculopathy within first year Renal transplant, 1.5% Rejection episodes in first year (especially if graft dysfunction) Hypertension, 97% Ventilator at the time of transplant Hyperlipidemia, 93% Diabetes Diabetes, 39% Coronary artery disease Malignancy, 30% Age (U-shaped: RR = 1.5 age 20, RR = 1.4 age 65 years) High BMI Table 12.5 Risk factors for development of cardiac allograft vasculopathy Decreasing transplant center volume (RR = 0.8 Ͼ 55 transplants/year) Donor factors Re-transplantation Hypertension Ischemic time (RR = 1.36 Ͼ 5 years) Active infection BMI: body mass index; RR = relative risk. Increasing donor age Recipient factors systematic post-transplant care/surveillance, and bet- Pretransplant coronary artery disease ter management of renal dysfunction. Mismatches at DR locus/class 1 mismatch Decreasing recipient age Long-term complications after Antibody-mediated rejection heart transplant Recurrent cell-mediated rejection The long-term complications following heart trans- CMV infection plantation are similar to those of other organ Smoking history (particularly post-transplant) transplants and include vasculopathy and compli- cations of immunosuppressants. Consideration is (see Chapter 4B). It may be regarded as an accelerated given here to complications and management that form of atherosclerosis, but is usually a diffuse pro- are specific to the cardiac allograft; other principles cess affecting the entire coronary artery tree, includ- of pathophysiology and management are discussed ing epicardial, intra-myocardial branches, and veins. elsewhere. Table 12.4 illustrates the incidence of This process is distinct from that of the usual coronary complications 10 years after heart transplant (ISHLT atherosclerosis in which the lesions are often eccen- 2009 Registry data). tric and more focal, the elastic lamina is extensively involved, and calcification is common. Donor-related Cardiac allograft vasculopathy coronary atherosclerosis may be recognized early fol- lowing transplantation but tends to have a differing CAV or chronic rejection is a leading cause of graft pathophysiology, slower progression, and better prog- failureandmortalityfollowinghearttransplantation, nosis than CAV. The risk factors for developing CAV with the incidence approaching 8% at 1 year, 31% at in addition to those involved in traditional atheroscle- 5 years, and 52% at 10 years after heart transplanta- rosis (hypertension, diabetes, dyslipidemia) are shown tion. There is a small reduction in CAV incidence in in Table 12.5. themostrecentera,possiblyduetoimprovedpatient selection, post-transplant management, and in partic- ular the widespread use of statins. CAV has a complex Pathophysiology pathophysiology, and the detailed mechanisms are The process of CAV involves coronary endothelial largely unknown but involve both immune-mediated cell damage since this is the barrier between recip- endothelial damage and conventional atherosclerosis ient blood carrying immune cells and donor tissue

106 Chapter 12: Long-term management and outcomes in addition to being a site of donor antigen pre- Table 12.6 ISHLT nomenclature for cardiac allograft sentation (see Chapter 4B). Endothelial injury may vasculopathy be immune (HLA mismatch, antibody- and cell- ISHLT CAV0 (Not significant) No detectable angiographic mediated rejection) or non-immune mediated (e.g., lesion due to ischemia/reperfusion injury, atherosclerotic ISHLT CAV1 (Mild) Angiographic left main (LM) 50%, or primary vessel with risk factors, CMV infection). maximum lesion of 70%, or any branch stenosis 70% (including diffuse narrowing) without Clinical presentation and diagnosis allograft dysfunction. The clinical presentation of CAV is often atypical, and ISHLT CAV2 (Moderate) Angiographic LM 50%; a single primary vessel 70%, or isolated diagnosis may be challenging due to denervation of branch stenosis 70% in the transplanted heart. Patients often lack chest pain branches of two systems, and suffer from silent myocardial ischemia, although without allograft dysfunction. re-innervation may occur to a variable degree a few ISHLT CAV3 (Severe) Angiographic LM 50%, or two or years after transplant. Presentation may therefore be more primary vessels 70% stenosis, or isolated branch asymptomatic (surveillance angiography) or present stenosis 70% in all three with new-onset heart failure, arrhythmia, myocardial systems; or ISHLT CAV1 or CAV2 infarction, syncope, or sudden cardiac death. Follow- with allograft dysfunction (defined as LVEF 45% usually in ing exclusion of acute cellular rejection, the diag- the presence of regional wall nosis of CAV must be considered in the presence motion abnormalities) or of signs/symptoms of heart failure, ECG changes, evidence of significant restrictive physiology. changes in systolic or diastolic function on echo, arrhythmia (especially ventricular), bradycardia, and LVEF: left ventricular ejection fraction. heart block. Regular surveillance for CAV is per- formed, often annually and using a variety of different andisassociatedwithadverseoutcomes:intimalthick- investigations. Coronary angiography has remained ening of ≥0.5 mm in the first year after transplant the “gold standard” for diagnosis of CAV since the appears to be a reliable surrogate marker for subse- early days of heart transplantation. quent development of CAV and mortality up to 5 years Angiography has the advantage of being widely after heart transplantation. available but may under-diagnose CAV because of the Other diagnostic investigations include non- diffuse and concentric nature of vessel involvement. invasive computed tomography–based angiography Careful comparison of previous angiograms may help (sensitivity and specificity remain low, and distal to identify the progressive loss of the lumen calibre and branch vessel anatomical definition is poor); coronary smaller branches. In 2010, the ISHLT published a stan- flow reserve quantification (abnormal endothelial dardized nomenclature for CAV (Table 12.6). Lesions responses associated with vasculopathy); non-invasive may be classified as type A (discrete, tubular, or mul- imaging and perfusion studies (nuclear, magnetic tiple stenoses), type B1 (abrupt ending with distal dif- resonance imaging [MRI], and echo stress testing); fuse narrowing and obliterated vessels), type B2 (grad- Myocardial biopsy, immune-based markers, gene and ual tapering with remnant distal lumen), and type protein biomarkers (B-type natriuretic peptide [BNP], C (narrowed irregular and blunt ending distal vessels). troponin, high-sensitivity C-reactive protein [CRP]). Intravascular ultrasound (IVUS) of at least one epi- None of these investigations are recommended for cardial vessel over 40–50 mm appears very sensitive to CAV surveillance due to lack of sufficient sensitivity or identify early CAV with very good negative predictive specificity for diagnosis and issues of reproducibility value. IVUS can be used to assess various parameters, or standardization of performance. including intimal thickness and luminal and exter- nal elastic membrane cross-sectional areas. IVUS may provide evidence of subclinical CAV but probably does Outcomes in CAV not add incremental information when the angiogram CAV reduces short-term survival and remains the appears normal. Increasing intimal thickness predicts main cause of death in long-term survivors after development of angiographic CAV, guides treatment, cardiac transplantation. A cohort of patients has a

107 Section 2: Heart

PATIENT SURVIVAL AFTER REPORT OF CAV AND Figure 12.4 Survival following diagnosis of CAV from ISHLT Registry, 1994–2008. From J PATIENT SURVIVAL IN PATIENTS WITHOUT CAV* Heart Lung Transplant, 2010; 29(10): 1083–141. (Transplants: April 1994 - June 2008) Reproduced with permission from the 100 International Society for Heart and Lung 90 No CAV (N = 15.547) Transplantation. 80 CAV (N = 5,706) 70 60 50 40 30 Survival (%) Survival 20 p < 0.0001 10 0 02468113579011 Time after Report of CAV* (Years)

more rapidly progressive CAV, with overall survival iber of vessels and carry a higher risk of re-stenosis. reduced by approximately 7% 1 year after diagnosis Conventional cardiac surgery on a transplanted heart and increasing to 11% 5 and 10 years after diagnosis carries higher risk due to associated comorbidities, (Figure 12.4). A worse prognosis appears to be related including previous sternotomies, renal dysfunction, to early development of CAV, rapid progression, and hypertension, immunosuppression, and poor wound more severe disease (widespread, multi-vessel, and healing. However, the only definitive treatment for lesions Ͼ70% luminal diameter). established CAV is re-transplantation. Treatment Metabolic syndrome Because the development of CAV is generally slow, The incidences of various metabolic syndrome risk fac- early recognition of the presence of disease may tors including hypertension, obesity, diabetes mellitus, allow early intervention and prevent graft dysfunc- and hyperlipidemia are increasingly seen after heart tion. Rapid progression of CAV may also be recog- transplantation. The following sections give consider- nized, allowing consideration of re-transplantation. ation to aspects of these complications, which are spe- Prevention of CAV by modification of conven- cific to patients after heart transplantation. tional atherosclerotic risk factors is the current ther- apeutic strategy, specifically control of hyperten- sion, hyperlipidemia, obesity, and diabetes melli- Systemic hypertension tus; abstinence from smoking; and exercise. Poten- Systemic hypertension is seen in the majority of tial therapeutic strategies to treat CAV are listed in patients after transplantation and often develops early Table 12.7, although few have clinical evidence in (73% in year 1, 93% by year 5). The risk factors for human heart transplantation. Among other therapeu- hypertension in the general population apply post- tic agents, newer immunosuppression regimens may transplant but are more likely to be pre-existing in reduce the immunological risk for CAV and have heart transplant recipients, leading to increased inci- anti-proliferative effects. In symptomatic CAV, med- dence. The use of CNIs as described earlier is directly ical treatment is similar to that of traditional coro- linked to the development of post-transplant hyper- nary artery disease. Revascularization has no proven tension. prognostic value but may provide symptomatic relief Mechanisms include both neurohormonal and in selected cases. In any case, percutaneous coronary sympathetic nervous activation and direct cyclo- interventional and surgical treatments are often not sporine (CyA)-induced renal vasoconstriction. CyA possible in view of the diffuse nature and small cal- stimulates the release of endothelin and thromboxane

108 Chapter 12: Long-term management and outcomes

Table 12.7 Potential therapeutic strategies of treatment of through left ventricular hypertrophy and both systolic cardiac allograft vasculopathy and diastolic dysfunction. Treatment follows usual Antihypertensives guidelines for management of hypertension; however, Calcium antagonists patients are often resistant to treatment, and multi- ple drug combinations are needed. Calcium antago- ACE inhibitors nists have the advantage of also reducing CAV and Alpha and beta-blockers are not nephrotoxic. Diuretics may also be useful due Antioxidants to the increasing circulating volume after heart trans- Omega-3 plant, especially in the context of right heart dysfunc- Fish oils tion early after transplant, or in the presence of tricus- Vitamins E, B6 pid regurgitation. Beta-blockers exacerbate autonomic dysfunction and prevent chronotropic response to Folic acid exercise and have traditionally been avoided, although Antiplatelet they may be used safely in selected patients. Aspirin Clopidogrel Antiproliferative Low-molecular-weight heparin Hyperlipidemia mTOR inhibitors Heart transplant patients require tight control of Antivirals hyperlipidemia due to the impact on CAV. There is a high incidence of hyperlipidemia, with 58% of patients Ganciclovir/valganciclovir affected 1 year and 88% 5 years after heart transplan- Endothelial protection tation. Causes include immunosuppressants, high- Better myocardial preservation cholesterol diet, excessive weight gain (possibly related Strategies to reduce ischemia/reperfusion to reversal of ghrelin resistance), and genetic factors Strategies to reduce effects of brainstem death (familial hypercholesterolemia), which may be present pretransplant, particularly in patients with an ischemic Immunosuppression etiology. Pravastatin has been shown to reduce choles- MMF terol after heart transplant but also to reduce the inci- mTOR inhibitors dence of CAV and should be used first line. Other Anti-cytokines/adhesion molecules statins and lipid-lowering therapies (fibrates, nicotinic Lipid modification acid) may be useful but should be used with cau- tion to avoid rhabdomyolysis. Because CAV predicts Statins long-term survival after heart transplant, considera- Apheresis tionshouldbegiventoalterationinimmunosuppres- Re-transplantation sion if lipid reduction is suboptimal. Revascularization Percutaneous coronary intervention Coronary bypass surgery Transmyocardial laser revascularization Hyperglycemia Diabetes mellitus and glucose intolerance are com- A2. The end point is renal vascular bed vasoconstric- mon after transplantation due to steroid and CNI use. tion and salt and fluid retention. Other causes include The incidence of diabetes after heart transplantation corticosteroids (salt and fluid retention), loss of auto- is 28% at 1 year and 36% at 5 years, although the nomic regulation due to cardiac denervation, and lack incidence remains higher in those with an ischemic of nocturnal dip of blood pressure. Abnormal respon- cardiomyopathy pretransplant. Usual treatment and siveness of the renin–angiotensin–aldosterone system surveillance for diabetes should be instigated, but met- to salt and fluid retention is seen in post-transplant formin should be avoided in patients with renal dys- patients. Hypertension leads to allograft dysfunction function.

109 Section 2: Heart

Table 12.8 Proportion of malignancies after heart transplant in Chronic renal dysfunction 10-year survivors (ISHLT 1994–2008) Acuteandchronicrenalfailurearecommonafterheart Malignancy % transplantation. Due to the 50% increase in relative risk of 1-year mortality for creatinine Ͼ2.5 mg/dl, con- Skin cancer 72 sideration should be given to combined heart–kidney Lymphoma 8 transplant in carefully selected patients with mod- Prostate 2 erate to severe renal dysfunction. The incidence of Adenocarcinoma 1 severe renal dysfunction (creatinine Ͼ2.5 mg/dl, dial- ysis, or renal transplant) is 7–12% at 1 year and 40% at Lung 1 10 years after transplant. Over the last decade, the rate Bladder 4 of development of severe renal dysfunction seemed to Kaposi’s sarcoma 1 have lessened significantly. This could be attributed to Breast 2 better patient selection, improved management, and Cervical 2 modern immunosuppression regimens. Risk factors Colon 1 include pre-existing renal dysfunction from chronic heart failure, diabetes, hypertension, and atheroscle- Renal 1 rosis, but CNI usage is undoubtedly the major fac- Indeterminate 5 tor for chronic renal dysfunction. CNI nephrotoxicity is discussed elsewhere but results in glomerulosclero- after heart transplant. In common with other trans- sis, tubulo-interstitial fibrosis, and obliterative afferent plants, the major malignancies are skin cancers and arteriolopathy. post-transplant lymphoproliferative disorder (PTLD), Prevention by minimizing CNI levels may be help- which together form about 80% of malignancies in 10- ful balanced against the risk of rejection, although year survivors and may present early after transplant higher levels of immunosuppression are often required (Table 12.8). after heart transplantation compared with other trans- A reduction in late malignancy can only be planted organs. Other strategies include minimizing achieved by lower levels of induction and maintenance exposure to any additive medication with nephro- immunosuppression, induction of tolerance, and ear- toxic potential, using CNI-sparing immunosuppres- lier detection through close surveillance. Improved sion (mycophenolate mofetil [MMF]– and mam- immunosuppressive regimes in the modern era may malian target of rapamycin [mTOR]–based regimens), explain the lower incidence of malignancies seen over secondary prevention (tight control of blood pres- the last decade. Recurrence of pre-existing treated sure, diabetes, and cholesterol), and use of calcium malignancy in heart transplant recipients, following channel blockers where appropriate (afferent arterio- careful selection, is similar to that of other organ trans- lar vasodilatation). Care should be taken to minimize plants at 20%. The pathogenesis, risk factors, and treat- use of contrast agents for various investigations, espe- ments are similar to those of other transplanted organs cially for coronary angiography. Progression to dialy- and are discussed elsewhere, but it is worth noting sis predicts poor outcome after heart transplant, and that the incidence of malignancy, particularly PTLD, if allograft function remains well preserved and other is somewhat higher in cardiac transplantation due factors would predict reasonable medium-term prog- to the generally higher levels of immunosuppression nosis(e.g.,absenceofCAV),thenrenaltransplantation required to prevent rejection. Urological and lung can- should be considered. cersarealsocommonafterhearttransplant,inpartdue to the high incidence of smoking prior to transplant. Malignancy after heart transplant Malignancy is a major cause of late morbidity and Late cardiac allograft dysfunction mortality after heart transplantation, with mortality increasing cumulatively over time and accounting for Arrhythmia approximately 10% of deaths at 5 years and 20% Atrial arrhythmia late after heart transplant is com- at 10 years. Cumulative prevalence is approximately mon due to atrial enlargement (in biatrial anas- 3% at 1 year, 15% at 5 years, and 30% at 10 years tomosis), allograft dysfunction, and late tricuspid

110 Chapter 12: Long-term management and outcomes regurgitation. Consideration should always be given gated,althoughthereisnoevidencebaseafterheart to the possibility of acute rejection in the presence of transplant. No place has been established for cardiac arrhythmia. The loss of atrial contribution to ventricu- resynchronization or implantable defibrillator therapy. lar contraction is particularly important in transplant More important perhaps is the recognition of end of patients. Management algorithms follow usual treat- life, counseling the patient, and access to palliative ment for atrial arrhythmia, although beta-blockers care. maybelesswelltoleratedandelectrophysiologicalpro- cedures are more anatomically challenging. Ventricu- lar arrhythmia is usually pathological and suggests the Quality of life after heart transplant presence of CAV and allograft dysfunction. Late brad- There is excellent improvement in quality of life in yarrhythmia due to sinus or AV node dysfunction is addition to the improved mortality following heart unusual (1–2% of patients) and may also be related to transplantation. In common with other transplants, acute rejection or CAV. Permanent pacing to preserve there remain problems with the demands of a complex atrial contraction is preferred. medical regimen, altered body image due to immuno- suppression, stress about uncertainty of complica- tions, unfulfilled expectations about exercise capac- Tricuspid regurgitation ity, anxiety/depression, and other psychosocial diffi- Mild and moderate late tricuspid regurgitation (TR) is culties. Involvement of psychological and psychiatric common after heart transplant and is usually slowly support is especially important following heart trans- progressive. Severe TR requiring treatment or surgi- plantation. cal intervention is more unusual, seen in 1–5% of patients 5 years after transplant. Causes include mul- tiple endomyocardial biopsies, tricuspid annular dis- Further reading tortioninthebiatrialanastomosis,rightheartdilata- Costanzo MR, Dipchand A, Starling R, et al. The tion (in pulmonary hypertension), and torsion of International Society of Heart and Lung Transplantation atria during cardiac contraction. Diuresis is sufficient Guidelines for the care of heart transplant recipients. J Heart Lung Transplant 2010; 29: 914–56. therapyinmostpatients,butsurgicalrepairmaybe of benefit in patients with severe right heart failure Kirklin JK, Young JB, McGiffin DC. Heart Transplantation. Philadelphia: Churchill Livingstone, 2002. symptoms. Mehra MR, Crespo-Leiro MG, Dipchand A, et al. International Society for Heart and Lung Cardiac allograft failure Transplantation working formulation of a standardized nomenclature for cardiac allograft vasculopathy. JHeart Systolic and diastolic heart failure are seen late after Lung Transplant 2010; 29: 717–27. transplant, most commonly due to CAV,hypertension, StehlikJ,EdwardsLB,KucheryavayaAY,et al. The Registry and left ventricular hypertrophy, although acute rejec- of the International Society for Heart and Lung tion should always be considered because impaired Transplantation: Twenty-seventh official adult heart function may improve with treatment. Treatment transplant report – 2010. JHeartLungTransplant2010; with conventional heart failure medication is insti- 29: 1089–103.

111 Section 2 Heart Chapter Pediatric heart transplantation

13 Jacob Simmonds and Michael Burch

Key points appear to be important in progression, r ABO-incompatible transplantation is although the exact cause is unknown. possible in infants and young children. The precise level of isohemagglutinins at which it Heart transplantation remains the only realistic ther- is safe to perform ABO mismatch transplant apeutic option for children with end-stage heart dis- remains unclear; however, a dilution of 1:16 ease. Although the first pediatric heart transplant was may be used. r performed by Kantrowitz in New York in 1967 only 3 In ambulatory children with chronic heart days after Barnard’s celebrated first adult transplant, Ͻ failure a VO2 max of 50% of predicted, it was soon understood that the discipline was not a blunted blood pressure and heart rate mere extension of transplantation in adults, but rather responses, and a B-type natriuretic peptide a distinct clinical entity with its own set of principles, Ͼ level of 300 pg/ml are all associated with a techniques, and goals. Of course, there is much overlap poor outcome and can be considered in the with adult transplantation, and much of what we know transplant listing process. and practice is a direct extrapolation of that much r Transplant for congenital heart disease has a larger experience. However, although broad surgical varied peri-operative risk. Previous Senning techniques, postoperative intensive care, and ongoing and Mustard patients have a relatively low pharmacological management share much in common risk, but Fontan patients have a relative risk between the age groups, there are vital differences that ofdeathof8.6comparedwithother must be taken into account if a center is going to pro- congenital patients. duce a successful pediatric transplant program. r The risk for Fontan patients appears greater if It is also important to state that pediatric heart the circulation is failing because of transplantation is itself a very heterogeneous subspe- pulmonary resistance rather than ventricular cialty. Whereas approximately 90% of adult transplants failure. are performed for cardiomyopathy (either ischemic or r Post-transplant lymphoproliferative disease non-ischemic), up to one in three pediatric recipients is the predominant malignancy in pediatric has a diagnosis of congenital heart disease. The use transplantation. Epstein-Barr viral load on of the word “diagnosis” in the singular form is some- polymerase chain reaction may be helpful in whatofamisnomer,however,sincethissubgroupof diagnosing incipient lymphoma or guiding patients is further subdivided into a wide variation of treatment, although this remains unproven. congenital heart defects, with the possibility of mul- r Cardiac allograft vasculopathy is the main tiple previous surgical interventions and wide diver- cause of late graft loss and the major reason sity in clinical status. This variability in anatomic sub- for re-transplantation in children, as in strate must be thought about carefully and on an indi- adults. Inflammatory factors such as vidual by individual basis by the surgical team prior cytomegalovirus and recurrent rejection to heart transplant listing; it is also accompanied by

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

112 Chapter 13: Pediatric heart transplantation variability in physiological status that requires skilful Indications management by the anesthetic and intensive care spe- Improvements to every stage of transplantation have cialists. Further complexity arises from the significant greatly increased the success of the procedure, both differences between neonates, young children, and in terms of longevity and quality of life of the child- adolescents, again with regard to anatomy and physiol- hood recipients. Simultaneously, there have been great ogy,butalsoinregardtopsychologyandsocialback- strides made in the treatment of heart failure (HF) ground. In addition, patients requiring heart trans- and the surgical treatment of congenital heart dis- plantation in childhood often have coexisting condi- ease, allowing transplantation to be postponed or com- tions or syndromes that must be considered through- pletely avoided in some patients. With this in mind, out the transplant process. in 2007 the American Heart Association Council on With this heterogeneity in mind, pediatric trans- Cardiovascular Disease in the Young developed a doc- plant teams have been set up in many centers world- ument describing the current indications for pediatric wide as distinct entities to their adult counterparts, heart transplantation. with obvious benefits from the subspecialization. Predictably, the guidelines are largely based on However, the experience of these teams is generally non-randomized studies or on consensus clinical much less than that of their adult counterparts. The lat- expert opinion, rather than large, multi-center ran- est data from the International Society for Heart and domized controlled trials. In evidence-based medicine Lung Transplantation (ISHLT) reveal that only 11% terms, these are level of evidence B and C. The main of pediatric transplant centers performed more than indication for transplantation in children that is widely 10 per year since 2001, compared with 58% of adult agreed (class I indication) is severe HF associated with centers. Low volume of cases is known to be a risk fac- impaired function of the systemic ventricle (usually tor for mortality post-transplantation, and it is easy of left ventricular [LV] morphology but sometimes to understand how increased experience and famil- right ventricular [RV] morphology in some types of iarityofeveryaspectoftransplantationwouldlead congenital heart disease). However, it is also consid- to a more successful program. The small numbers of ered reasonable to list children with moderate HF if procedures per center also greatly limit both quantity they have severe limitation of activity/exercise or sig- and quality of research possibilities, yet this is coun- nificant growth failure as a result of the chronic HF. terbalanced by effective multi-center clinical audit and Exercise testing can aid listing such patients, although research, such as with the Pediatric Heart Transplant it is difficult in children (as discussed in the next StudyGroup.However,evenwiththesecaveats,com- section, Timing of listing). Intractable arrhythmias plications, and complexities, pediatric heart transplan- are considered an indication for listing, provided all tation has survival rates slightly better than that of treatment options including an implantable defibril- adults. lator have been considered. Heart transplantation for One further important difference between pedi- restrictive cardiomyopathy is considered appropriate if atric and adult heart transplantation is how to define thereisevidenceofreactivepulmonaryhypertension success for the pediatric transplant program and for within safe limits. That is, when the pulmonary vascu- the individual patient. If a 55-year-old man with coro- lar resistance (PVR) can be reduced using pulmonary nary artery disease survived 15 years after transplan- vasodilators to a transpulmonary gradient (TPG) of tation, it would be considered by many patients and under15mmHgandaPVRof≤6 Wood units·m2.The health care providers alike as a successful and worth- latter PVR rule applies to the other main indications while intervention with obvious benefit to the patient for transplant previously mentioned. and society as a whole. Although the success and Less strong indications without universal agree- worth of 15-year survival following infant transplan- ment but for which the weight of evidence is broadly tation could never be questioned, the prospect of re- in favor (class II) include children whose heart dis- transplantation or death in adolescence or early adult- ease is inoperable and may lead to development of hood invites further philosophical, ethical, and eco- irreversible pulmonary vascular disease; e.g., restric- nomic questions. Although difficult, these issues must tive cardiomyopathy with mildly elevated pulmonary be discussed with any prospective patient and family to pressures. There are also some other congenital prob- allow careful consideration before agreeing to be trans- lems for which the prognosis for conventional surgery plant listed.

113 Section 2: Heart

is so bad that transplantation listing can be considered, ciency, have been shown to have high prognostic value such as coronary atresia/stenoses or severe valve regur- in adults with HF. However, despite widespread use in gitation or impaired ventricular function. the adult population, information regarding the prac- Cardiac re-transplantation in children is mainly tical clinical value of exercise testing as a prognos- indicated when there is poor ventricular function and tic tool in pediatrics is very limited. Our own experi- significant coronary artery vasculopathy (class I indi- ence is that exercise testing is possible in children over cation). It is more controversial when ventricular func- 10 years of age and sometimes younger. Our data are tion is preserved (class II indication). There is gen- broadly supportive of a peak oxygen consumption of eral agreement that re-transplantation should not be less than 50% of predicted for age, although we have performed during an episode of acute rejection. Early also found peak blood pressure response to exercise to re-transplantation has a poor outcome, and there is be an important prognostic test. consensus that this should not be performed within 6 Exercise testing is complicated in congenital heart months of transplant. disease. Peak heart rate appears to be a useful prognos- ItisimportanttostressthatHTshouldnotbeused tic marker. Oxygen consumption and ventilatory effi- as the primary therapy for any infant with congen- ciency are also useful. However, the values at which ital heart disease without the presence of concomi- transplantationshouldbeconsideredvarywiththe tant coronary, valvular, or ventricular impairment. It type of congenital heart disease. Patients with a Fontan is,however,reasonabletoconsidertransplantationin circulation have much lower baseline levels of oxy- the setting of previously repaired congenital heart dis- gen consumption and ventilatory efficiency. A simple ease without ventricular dysfunction if there is risk of 6-minute flat walking test is still valuable in congenital developing fixed raised PVR that could preclude future heart disease. transplantation, severe aortic or systemic valve insuffi- For younger children, pre-school age and infants, ciency, severe cyanosis, persistent arrhythmia, or per- exercise testing is impossible. It is in this age group that sistent protein-losing enteropathy. biomarkers of HF are particularly useful. In patients with chronic HF, B-type natriuretic peptide level of Timing of listing for transplant of over 300 pg/ml may be predictive of poor outcome in pediatric cardiomyopathy. Other biomarkers are cur- ambulatory patients rently being assessed. Although increasing proportions of HT are performed in very ill patients on high-dose inotropes or mechan- Transplantation for congenital ical support, some patients undergo transplantation while still ambulatory and at home. Many pediatric heart disease cardiologists base timing on evidence of LV systolic Transplantation for congenital heart disease illustrates dysfunction and symptoms of HF. Severity of ven- best many of the peculiarities of heart transplant in tricular dysfunction has been found to be predic- the pediatric age group. As experience with pedi- tive of outcome in some studies but not in others. atric transplant has grown, the boundaries of what Symptoms appear to provide poor prognostic capa- anatomy is considered transplantable have extended to bility too, because even asymptomatic children with the point now that the only contraindications from a incidental discovery of cardiomyopathy can have a purely anatomical point of view are severe pulmonary poor prognosis. Finding ways to identify those patients artery hypoplasia and pulmonary vein stenosis. How- with the highest risk of clinical deterioration and/or ever, these may be correctable by combined heart–lung death would greatly assist in clinical decision making, transplantation or using extended donor pulmonary including the indication and prioritization for trans- arteries for pulmonary hypoplasia. plant. Concurrently, improved conventional surgical Extensiveevidencesupportstheuseofcardiopul- techniques have led to the treatment of more and monary exercise testing as a tool to select patients more congenital heart disease without the need for with increased short-term mortality who should be early transplantation. However, these, along with the offered transplantation. Besides peak oxygen uptake, success of non-curative surgical procedures such as several additional variables, such as ventilatory effi- the Fontan pathway, have only postponed the need

114 Chapter 13: Pediatric heart transplantation for transplant in many patients, reducing strain on ing pulmonary resistance, or inducing human leuko- the youngest, most scarce donors, but in many cases cyte antigen (HLA) sensitization by virtue of blood adding to the waiting list in later childhood. In reality, transfusions. Any of these complications of delaying these improvements may have even increased the transplant may reduce the success of future transplan- overall numbers of potential recipients, as many tationandhavetheeffectofanoveralldecreaseinlife infants would previously have died during neonatal expectancy. operations or on neonatal transplant waiting lists. Priortolistingfortransplantation,detailedscan- Many of these children are now surviving infancy, ning of the heart and pulmonary vasculature with but with circulatory systems that are destined to fail an appropriate combination of echocardiography, at some point, and by the time a heart transplant is angiography, computed tomography (CT), or mag- required, some families are finding transplantation a netic resonance imaging (MRI) must be undertaken to more attractive option, and palliative care strategies ensure that the anatomy is favorable. These scans must less so. be available to the surgeon before accepting an offer of In most registries, congenital heart disease remains a potential organ so he can request extended portions a risk factor for transplantation. Attrition is early and of the great vessels, which may be required to facili- related to the complexity of the reconstruction. Yet tate non-standard anastomoses. Because it is likely that it would be wrong to be uniformly negative about other organs may be being harvested from the same transplantation for congenital heart disease. Registry donor, it is important for the cardiac surgeon to speak data have shown that transplantation for previous Sen- to other transplant teams to enable the most judicious ning or Mustard procedures can be achieved with a use of donor vessels, or perhaps arrange for alternative low mortality rate. Previously palliated infants and tissue to be collected. For instance, if the heart trans- young children with a Glenn circulation also appear plant surgeon feels he is unable to harvest enough of to have relatively low risk. However, it is the Fontan the pulmonary arteries for a particular recipient, he patients who represent the highest risk, with a recent may be able to use donor pericardium or descending paper showing a relative risk of death of 8.6 com- aorta. pared with that of other congenital operations. The Fontan patients often have liver or renal problems Pulmonary vascular resistance and may suffer more infections by virtue of protein- The assessment of PVR is particularly crucial in order losing enteropathy and lymphopenia. Protein-losing to reduce the rate of right HF post-transplant, but it enteropathy increases the risk of transplant, perhaps can be technically difficult, particularly in congeni- because it may be a marker of increased pulmonary tal heart disease. In general the guidelines document vascular resistance. Resistance is hard to calculate referred to previously advises that transplantation is in the Fontan patient with multiple sources of pul- possible if PVR can be reduced using pulmonary monary blood supply, pulmonary arteriovenous mal- vasodilators to a TPG under 15 mmHg and a PVR of formations,collaterals,andfenestrations. ≤6 Wood units·m2. There is a tendency for pediatric For all congenital patients, it is obvious that risk centers to push the boundaries of transplantable PVR, can be reduced by having an experienced congenital and some units will take on children with resistance surgeon perform the operation with appropriate anes- higher than this, but it is clear that risk is increased. thetic and intensive care teams. This is particularly A long period of inotrope or vasodilator therapy may important in adult congenital heart disease, where few reduce PVR, as may a period of mechanical support by adult transplant teams have extensive congenital expe- reducing LV end-diastolic pressure. rience. Because the lifespan of a transplanted organ is lim- ited, no potential recipient should be transplanted too Mechanical support early. The aim of pediatric programs is often to delay Traditionally pediatric units offered extracorporeal transplantation for as long as possible by optimizing membrane oxygenation (ECMO) support as a bridge medical management. However, this delay must not be to transplant, as most ventricular assist devices (VADs) at the expense of other organ damage (most notably were not suitable for small children. This strategy renal failure or cirrhosis in Fontan patients), increas- allowed only a short time for support before serious

115 Section 2: Heart

complications ensued. The development of the pneu- ABO incompatibility matic assist device suitable for children, the Berlin Compared with adult transplantation, the prospec- Heart, has allowed long-term bridge to transplant tive donor pool of hearts suitable for infant recip- in young children. Successful support is possible for ients is small. Infants with blood group O have many months, although few children are discharged a further reduced donor pool. This imbalance has from hospital. Most experience has been obtained in prompted advances in transplanting across blood children with dilated cardiomyopathy who are older groups – the so-called ABO-incompatible transplant. than 1 month of age. Usually the device is implanted in In adults, heart transplants across ABO blood groups children with progressive cardiac failure and inotrope have occurred only by mistake, with high rates of both dependency. Few data exist on support for very young hyperacute rejection and mortality. children who weigh less than 4 kg. It seems likely In contrast to mature adult immune systems, how- that the neonatal group will be more difficult to man- ever, infants have very low levels of isohemagglutinins, age, with the small 10-ml pumps and anticoagula- allowing the possibility of transplanting an organ of tion complexities increasing their risk of thromboem- blood type B, say, into a recipient of type A. In 1996, bolic complications. Typically the device is used as a Lori West led the team that performed the first inten- bridge to transplant, although there are reports of use tional heart transplant across an ABO incompatibility. asbridgetorecovery,andithasbeenusedinmyocardi- The recipient was a 25-day-old infant with hypoplas- tis. In general, fulminant myocarditis is still treated ticleftheartsyndromeofbloodgroupOwhoreceived with ECMO in most centers, although this may change an organ of blood group AB. During cardiopulmonary as experience with the Berlin Heart increases. How- bypass, plasma exchange was performed until the lev- ever, ECMO does avoid an LV scar. Conversion from els of isohemagglutinins were non-detectable. In the ECMO to Berlin Heart is often performed, although paperreportingthefirst10ABO-incompatibletrans- primary implantation of Berlin Heart is preferred in plants, rates of mortality, rejection, and morbidity were most units. The use in single-ventricle circulations equivalent to those of ABO-compatible transplants, and particularly the Fontan circulation also appears with two deaths and one other graft loss unrelated more complex. The limited data available do suggest to ABO activation. It is thought that donor-specific a higher mortality in the single-ventricle group. How- B-cell elimination results in immunological tolerance ever, the recent huge increase in Berlin Heart inser- in recipients of ABO-incompatible hearts. In order to tion in the United States illustrates the success of this ensure the ongoing success of the graft, future transfu- device. sion of blood products must occur according to strict criteria (Table 13.1). The exact age and level of iso- Surgical modifications hemagglutinins that are acceptable for ABO mismatch transplant are unclear, but an isohemagglutinin dilu- The particular anatomy in each case of congenital tion of 1:16 is used at our center as a cut-off value, and heart disease will require adaptation by the surgeon. few such transplants have been performed over the age Previous sternotomy may have produced significant of 2 years. mediastinal adhesions, which can cause heavy intra- operative hemorrhage, as can collateral vessels that have resulted from years of cyanosis. At the same time, Lymphocytotoxic antibodies knowledge of substernal great vessels may lead some Another problem in pediatric transplantation is the surgeons to expose the femoral vessels prior to ster- presence of pre-existing HLA antibodies, which have notomy to facilitate emergency femoro-femoral bypass been linked to increased hyperacute, cellular, and should the need arise. humoral rejection and increased mortality post- Before implantation can occur, it is sometimes nec- transplant. There are several factors that cause the for- essary to normalize recipient anatomy, for instance, in mation of such antibodies. However, most antibod- thecaseofbilateralcavopulmonaryanastomoses.Any ies are formed in response to blood products (par- extra procedures that are anticipated must be allowed ticularly white cells and platelets) or homograft tis- for when planning the operation in order to synchro- sue. Thus prior HLA sensitization is much more com- nize the prepared recipient and the arrival of the organ, moninchildrenwhohaveundergonesurgeryfor thus avoiding long ischemic times. congenital heart disease, or those who have required

116 Chapter 13: Pediatric heart transplantation

Table 13.1 Required blood groups for transfusion after ABO-incompatible pediatric heart transplantation Immunosuppression It does appear that acute cellular rejection, in partic- Donor Recipient ular that associated with hemodynamic compromise, blood blood Indicated Indicated Indicated is becoming less common. This may be the result of group group plasma red cells platelets newer drugs and increased experience. Over the last AB O AB O AB decade or so, there has been a gradual shift toward B O AB or B O AB or B using induction therapy in pediatric heart transplan- A O AB or A O AB or A tation, with current levels at about 60%. Antithymo- cyte globulin remains the preferred agent in chil- AB B AB O or B AB dren, although the proportion of centers using an AB AB O or B AB interleukin-2 receptor antagonist such as basiliximab AB A AB O or A AB is increasing. Basiliximab may be more effective given BA AB O or A AB prior to bypass and organ implantation, and its more selective action is felt by some to cause less of the problems associated with “over-immunosuppression,” mechanicalsupport.Forchildrenonlongerterm such as post-transplant lymphoproliferative disorder mechanical support, the “prophylactic” use of supple- (PTLD) and cytomegalovirus (CMV). ments of iron, folic acid, and erythropoietin may avoid Maintenance therapy is most commonly a combi- blood transfusion. nation of a calcineurin inhibitor (CNI) and cell cycle When a patient is sensitized, a prospective cross- inhibitor. Tacrolimus (TAC) has become the CNI of match between donor and recipient serum prior to choice in 58%, taking over from cyclosporine (CyA) transplantation would be the perfect solution, but it at least in part due to the latter’s unwanted side-effect is time-consuming, which limits the geographical area profile of hirsutism and gingival hypertrophy, which from which a sensitized recipient can receive an organ. canleadtoproblemswithcomplianceinpediatric Alternatively, it is possible to test antibodies against patients. Cell cycle inhibitors are used by 80% in the specific HLA antigens, thereby facilitating a “virtual” first year, with MMF the most common, being used in cross-match once a potential donor has been found. 59% of children; approximately one third of all patients Althoughthiswillnotbeasspecificasaprospective are prescribed it in combination with TAC. This fig- cross-match, it has been used safely in both adult and ure is slightly less than the equivalent value in adults, pediatric transplantation. approximately half of whom are taking TAC and MMF It is also possible to reduce the level of HLA in combination. antibodies prior to transplantation. Strategies for Steroids still hold a place in pediatric practice, de-sensitization include intravenous immunoglobu- particularly in the first year post-transplant, and in lin, methotrexate, mycophenolate mofetil (MMF), the treatment of rejection. The proportion of patients cyclophosphamide, and rituximab, and newer mon- using steroids is slightly less than in adults, with 55% oclonal antibodies to plasma cells, to lower the level of children taking prednisolone at 1 year; by 5 years of panel reactive antibodies prior to transplantation. post-transplant, this figure has fallen to less than 40%. However, there has not been uniform success with these strategies, and there is often a limited window of low antibody levels in which transplantation can Renal disease be performed. This has led some North American Chronic use of CNI induces renal dysfunction. This centers to transplant across the HLA barrier using is clearly crucial in pediatrics, where patients are intra-operative plasma exchange and post-operative still young and hopeful of a second heart transplant plasmapheresis with other antibody-reducing strate- when the first one eventually fails. Nephrotoxicity can gies. Results are still at a preliminary stage, but it be reduced by using lower levels both early post- appears that primary graft failure can often be avoided, transplant and later in the recipient’s course. The main and although antibody-mediated rejection is common, pediatric use of sirolimus (SRL) and everolimus (EVL) it may resolve with possible accommodation to the has been in children with impaired renal function after antibody. The long-term consequences of this strategy transplant, along with reduced doses of CNI. Others are as yet unknown. have used CNI regimes. The latter may be associated

117 Section 2: Heart

Figure 13.1 Intravascular ultrasound images of the left anterior descending coronary artery of the same patient taken 1 and 2 years post-transplant. The figure on the left shows a relatively healthy artery, with good lumen size, and only a small amount of intimal thickening in the upper left quadrant. The figure on the right shows the effective lumen now decreased to only slightly larger than the ultrasound probe, with the rest of the artery replaced by grossly thickened intima. The patient had clinically significant CMV disease during the period between the images.

with an increased risk of rejection, and greater surveil- conventional chemotherapy, and even specific T cells lance with biopsy and non-invasive testing are needed. have been generated to destroy the EBV-associated B cells.

Lymphoma Cardiac allograft vasculopathy Unlike adult transplant medicine, pediatric transplant Like adults, the major obstacle to late patient sur- teams do not have to deal with a wide range of malig- vival in pediatric heart transplantation is cardiac allo- nancies after transplant. Even skin malignancies are graft vasculopathy (CAV), an accelerated form of uncommon in children, but this does not preclude the obliterative cardiovascular disease (Figure 13.1). CAV need for sun protection and surveillance. The main causes considerable morbidity and mortality, affecting malignancy in children after transplant is lymphoma. approximately 50% of patients 5 years post-transplant. Usually, this is a B-cell lymphoma and is triggered by It is the leading cause of death in children more than 3 Epstein-Barr virus (EBV). Surveillance using quantita- years post-transplant. As it is accompanied by remod- tive polymerase chain reaction for the virus may give eling and compensatory luminal dilatation in the early early warning of incipient lymphoma, and many teams phases, it may best be diagnosed with intravascu- will reduce immunosuppression when viral loads are lar ultrasound rather than conventional angiography. high.Theriskoflymphomaappearsgreaterinchildren However, because this technique is not widely avail- that seroconvert after transplant. The site of the lym- able in children, it is likely that the prevalence of phoma varies, but is often in the gut (typically associ- the disease has been underestimated. Moreover, once ated with a low albumin), lungs (where chronic chest a diagnosis of CAV has been made, graft survival infections may be wrongly diagnosed), or tonsils. Tis- islimitedto50%at2years.Itisthusthegreat- sue diagnosis is usually needed. Treatment is similar to est limitation on long-term outcome and remains the that of PTLD in the adult and may involve reduction most pressing problem for transplant programs world- of immunosuppression, with rituximab increasingly wide. The cause remains unknown, but conventional used in the initial stages. Treatment may escalate to risk factors such as smoking, hypertension, diabetes,

118 Chapter 13: Pediatric heart transplantation hypercholesterolemia, and obesity are less of a problem heart transplant recipients should look forward to in pediatrics than in adults. In children, chronic viral reintegrating fully into normal schooling and social infection such as CMV and recurrent acute or chronic activities. cellular rejection appear important. It does seem that Adherence to post-transplant medication regimes chronic inflammation is important in the development is not universally high among childhood recipients, of CAV. and it is commonly cited as a cause of acute rejec- Current treatment is severely limited. The use of tion and death, particularly among teenagers. Early statinshasbeenshowntoimproveoutcomeaftertrans- recognition of clues to non-compliance, for instance plantation in adults, and pravastatin has been safely wildly variable immunosuppressant levels and erratic used in children including infants with few prob- lifestyles (including school avoidance and irregular lems. Pravastatin has the advantage of not interfer- sleeping patterns) must be taken seriously, and causes ing with cytochrome p450 and thus CNI. Reversal of to the root problems must be addressed. Addressing CAVismuchharder,althoughSRListhoughttohave concerns over the cosmetic side effects from medi- useful anti-proliferative properties. Ultimately, how- cation and more convenient medicine regimes may ever, many patients with CAV face the prospect of re- help. transplantation relatively soon after diagnosis. In the Ultimately, psychological and behavioral difficul- pediatric age group, it is the most common indication ties must be considered on an individual basis. A coor- for re-transplantation, making up just over half of all dinated multi-disciplinary approach to any problems, re-transplants. involving medical and nursing teams both at the trans- plant center and locally, must be undertaken when Psychological, developmental, and problems arise and include the family doctor, school teachers and nurses, family members, and teams of behavioral issues specialistssuchassocialworkersandspecialeduca- Ultimately, the success of individual transplants and tional providers when necessary. the program as a whole depends on well-informed, psychologically stable and motivated patients. In pedi- atric transplant, these are even more vital, as the Outcomes advent of adolescence often magnifies any psycholog- The overall survival for children following trans- ical stressors. The prevalence of behavioral problems plantation has steadily increased throughout the his- in post-transplant children is over 25%, a frequency tory of the subspecialty. The ISHLT data show half- that exceeds that of children following conventional life for patients aged 11–17 years of 11.3 years and cardiac surgery. Moreover, the effects of such prob- thoseaged1–10yearsof15.5years.Despitetrans- lems, in particular non-compliance with immuno- plantation in infancy having a higher early mortal- suppression, are potentially more serious. It is there- ity, half-life is excellent at 18 years. The highest rate fore of paramount importance to make psychologi- of graft loss in all age groups was seen in the first 6 cal and social assessments of potential recipients prior months post-transplant. With improvements in sur- to transplant and try to address any issues that may gical expertise and immediate postoperative inten- arise. sive care, there has been a very marked reduction Cognitive development of children following car- in early deaths and significantly reduced deaths from diac transplant is obviously an important outcome acuterejectioninthefirst12months.Survivalafter variable. Many studies have tried to evaluate accu- the first 12 months, however, has largely plateaued, rately the impact of transplantation. In general, the creating roughly parallel survival curves from year transplant group has mean mental and psychomo- 2onwards. tor scores at the low end of the normal range. Many Our figures from Great Ormond Street (Lon- factors are postulated as causes for this discrep- don, United Kingdom) illustrate the improvements ancy between the transplant and healthy population, in outcome in more detail and largely mirror expe- including hospitalization, missed school, cardiopul- rience from the registry (Figure 13.2). Overall sur- monarybypass,andthesideeffectsofpost-transplant vival has increased over time, with patients under- drugs. Encouragingly, the vast majority of pediatric going transplant in the recent eras having a much

119 Section 2: Heart

Survival Functions Figure 13.2 Survival of pediatric patients after heart transplantation at 1.0 +++++++++++++++++++ Era of transplantation Great Ormond Street Hospital, London, +++++++++++++++++++++++++++++++++++++++++++++ 1st fifty transplants United Kingdom. +++++++++++++++++++++++++++++ 2nd fifty transplants 3rd fifty transplants 0.8 +++++ 4th fifty transplants +++++++++++++++++++++ 5th fifty transplants ++++++++++++++++ 6 0.6 +++++++++++++++++++++++ + 1st fifty transplants- censored ++ 2nd fifty +++ +++++++++++++++ + transplants- 0.4 + censored Cum Survival + 3rd fifty transplants- censored 0.2 + 4th fifty transplants- censored + 5th fifty transplants- censored 0.0 + 6-censored

0.00 5.00 10.00 15.00 20.00 Time from transplant to last known followup

PreTxDiag Figure 13.3 Comparison of the survival 1.0 + of patients with cardiomyopathy and + Cardiomyopathy + congenital heart disease. +++ Congenitial Heart ++ +++++ Disease ++++++++++++++++ +++ ++ Cardiomyopathy- 0.8 ++++ +++ + +++++ censored +++ +++ +++ ++++ +++++++ + + Congential Heart + + ++++ ++ +++++++++ + Disease- + ++ +++++++ censored 0.6 +++++ ++++++++ + +++ + ++ + +++++ +++ ++++

Survival 0.4

0.2

0.0

0.00 5.00 10.00 15.00 20.00 Time from transplant (years)

better survival, particularly in the first 6 months mately 20% in the CHD group, and that conditional post-transplant. survival of patients who have survived 1 year is similar Diagnosis has an important bearing on survival between the groups. post-transplant. Over the 20 years of the transplant program at Great Ormond Street, those patients Further reading undergoing transplant for congenital heart disease Canter CE, Shaddy RE, Bernstein D, et al. Indications for fared less well than children with cardiomyopathy heart transplantation in pediatric heart disease: a (Figure 13.3). However, on closer scrutiny, it is easy to scientific statement from the American Heart see that the difference in survival between these two Association Council on Cardiovascular Disease in the groups is largely due to an early attrition of approxi- Young; the Councils on Clinical Cardiology,

120 Chapter 13: Pediatric heart transplantation

Cardiovascular Nursing, and Cardiovascular Surgery Mangat J, Carter C, Riley G, Foo Y, Burch M. The and Anesthesia; and the Quality of Care and Outcomes clinical utility of brain natriuretic peptide in paediatric Research Interdisciplinary Working Group. Circulation left ventricular failure. Eur J Heart Fail 2009; 11: 2007; 115: 658–76. 48–52. Kirk R, Edwards LB, Aurora P, et al. Registry of the Pollock-BarZiv SM, den Hollander N, Ngan BY, et al. International Society for Heart and Lung Pediatric heart transplantation in human leukocyte Transplantation: Twelfth Official Pediatric Heart antigen sensitized patients: evolving management Transplantation Report-2009. JHeartLungTransplant and assessment of intermediate-term outcomes in a 2009; 28: 993–1006. high-risk population. Circulation 2007; 116: Lamour JM, Kanter KR, Naftel DC, et al. The effect of age, I172–8. diagnosis, and previous surgery in children and adults West LJ, Pollock-Barziv SM, Dipchand AI, et al. undergoing heart transplantation for congenital heart ABO-incompatible heart transplantation in infants. N disease. JAmCollCardiol2009; 54: 160–5. Engl J Med 2001; 344: 793–800.

121 Section 3 Lung Chapter Recipient selection

14 J.S. Parmar

Key points operative risk to the patient (5–10% mortality in first r Critical organ shortage makes careful patient 30 days), it is a clinical imperative to ensure that all selection essential. other therapeutic options are fully explored prior to r Patients should be New York Heart considering this as a treatment option. Association class III–IV with a median An acceptable risk profile will vary from patient to survival of Ͻ2years. patient and will depend on the overall clinical picture. However, in general, when considering who to refer, r The best outcomes are in recipients with the patient should have a chronic disorder causing single-organ failure. single-organ failure with a predicted 2-year survival r Complex patients should be discussed with from the organ failure of less than 50%. Most potential transplant centers early. recipients with this severely limited prognosis will be r Infections with hepatitis B virus, hepatitis C in at least New York Heart Association (NYHA) class virus, human immunodeficiency virus, and III or IV.Patients with this severity of disease are often highly resistant bacteria are relative very limited and suffer with a poor quality of life. This contraindications. r combination of poor prognosis and poor quality of life Pulmonary rehabilitation prior to listing for make the operative risks more acceptable. transplant is essential. Although the ideal LT candidate is a patient who has single-organ failure with no associated comorbidi- Thelownumberofdonatedlungsandheart–lung ties, very few potential recipients fall into this ideal blocks means that careful patient selection remains category, and most will have associated or indepen- critical to ensure the best utilization of a scarce dent medical issues. It is critical that these conditions, resource. This chapter focuses on how to select patients if present, should have an independent prognosis for who will gain maximum benefit from lung transplan- survival of at least 5 years, as the median survival for tation (LT). In the first part of this chapter, general lung transplant recipients is 5.7 years. When consider- considerations and exclusions pertaining to all poten- ing who to refer for consideration of LT, there are a few tial recipients are outlined, and in the second half the absolute contraindications, which would be impedi- focus is on disease specific guidance for the major ments to consideration of LT (Table 14.1). In addition recipient groups: chronic obstructive pulmonary dis- there are a number of comorbidities, which can impact ease (COPD), cystic fibrosis (CF), idiopathic pul- on recipient suitability. monary fibrosis (IPF), and idiopathic pulmonary arte- rial hypertension (IPAH). Absolute contraindications With the improving outcomes achieved over the last Background 20 years in the field of LT, there is an increasing Transplantation of cardiothoracic organs is a ther- desire to re-examine clinical scenarios that were ini- apeutic option for a large variety of end-stage car- tially considered to be absolute contraindications. An diorespiratory diseases. As LT carries a moderate early example of this is the transplantation of patients who

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

122 Chapter 14: Recipient selection

Table 14.1 Contraindications to consideration of lung Chronic infections, both pulmonary and extrapul- transplantation monary, are a cause of major morbidity and mor- Incurable chronic extrapulmonary infection tality in potential LT recipients and can be impedi- Malignancy mentstolisting.Thepresenceofachronicinfection Any other medical condition with a prognosis < 5 years affecting an extrapulmonary site, which is not curable prior to transplantation, is an absolute contraindica- Active substance addiction or abuse within the last 6 months tion, because immunosuppression will exacerbate the Untreatable psychiatric or psychological condition infection. History of non-compliance with medical therapy Pulmonary infections with highly resistant bac- Absence of a reliable or consistent social support system teria have been shown to have poorer outcomes in Significant chest wall or spinal deformity comparison with non-infected patients. For instance, Severely limited functional status with poor rehabilitation patients with CF who are colonized with Burkholderia potential cenocepacia have been shown to have a very poor out- Colonization with highly resistant or virulent bacteria, fungi, or come and thus are not generally currently considered mycobacteria candidates for LT. Similarly, patients who are recur- Obesity, defined as a BMI Ͼ 30 kg/m2, or severe cachexia, rently colonized with atypical mycobacteria who fail defined as BMI Ͻ 18 kg/m2 to clear on medical therapy have a high rate of re- Corticosteroid therapy Ͼ 15 mg/day infection and are thus not considered to be good can- Severe gastroesophageal reflux with dysmotility didates. This is particularly important in patients who are colonized with fast-dividing atypicals (Mycobacte- ria kansasii or Mycobacteria abscessus)whoareatrisk of systemic dissemination. are ventilator-dependent. Previously, this state had The presence of fungus in the native lungs can been associated with poor outcomes, but with increas- cause problems after LT and will need careful assess- ing experience, patients may undergo transplantation ment in each individual. The presence of large cavities, from either ventilator-based or extracorporeal mem- pleural involvement, and highly resistant fungi may be brane oxygenation (ECMO)–based respiratory sup- a contraindication. As this is an area in which experi- port with reasonable outcomes. However, even with ence is limited, it is prudent to discuss each patient on these improvements, there remain certain conditions a case-by-case basis with the local transplant team. that are still considered to be absolute contraindica- Overthelastfewyearstherehasbeenanincreas- tions. The presence of chest wall deformity that would ing recognition of the importance of gastroesophageal militate against successful surgery is a clear exam- reflux disease (GERD) as a cause of chronic rejection ple of this. Coexisting viral infections with hepatitis (obliterative bronchiolitis). For this reason, patients B virus (HBV), hepatitis C virus (HCV), and human who have severe reflux associated with esophageal dys- immunodeficiency virus (HIV) have been considered motility are not considered for LT. absolute contraindications to LT. However, experience The presence of a malignancy is still considered an with these particular infections is growing in other absolute contraindication, as most are exacerbated by solid organ transplants so that some centers have listed long-term immunosuppression. Although each case and performed LT on a small number of patients should be considered on its merits, a disease-free inter- with these infections. Evidence of severe, untreatable val of at least 5 years is recommended for most major second organ damage or failure is an absolute con- malignancies. The one exception to this guidance is the traindication, unless it is feasible to perform a com- presence of limited cutaneous malignancies that can be binedtransplant.Anexampleofthisisthepresence treated with local excision. In this particular situation, of severe concurrent liver disease in a CF patient with the disease-free period can be reduced to 2 years. end-stage bronchiectasis. In this situation, isolated Cigarette smoking remains the number one cause organ transplantation may result in decompensation of lung disease worldwide. Smoking-related lung in the other failing organ and thus a poor outcome. injury is significantly greater in transplanted lungs However, combined synchronous lung–liver trans- than in native lungs. For this reason, recipients are plantation from the same donor may provide a viable required to demonstrate a period of complete absti- alternative. nence of at least 6 months prior to listing for LT. A

123 Section 3: Lung

similar approach is also used for patients who have Corticosteroid usage a history of illicit drug abuse. Periodic testing of the Many respiratory patients are given high-dose steroids urine for toxicology will help to confirm abstinence. as part of their treatment regime. This is particularly true in IPF patients. High-dose steroids affect healing Comorbidities generally but in particular the healing of the bronchial anastomosis. Early experience in this area demon- As highlighted previously, the presence of comorbidi- strated a higher rate of dehiscence of the bronchial ties outside of the failing respiratory system are impor- anastomosis and concomitant mortality in patients tant considerations that can impact patient outcomes. who are on high doses. Thus the preference is for Each individual comorbidity may not be a significant patients to be on the lowest dose tolerable, ideally concern in isolation; however, if there are multiple below 15 mg per day of prednisolone. comorbidities, then the total impact on the survival of the patients will need to be assessed. Although the number of possible comorbidities is limitless and spe- Gastroesophageal reflux disease cific guidance on each and every situation is impos- Many of the potential recipients have respiratory dis- sible, the most frequently occurring are considered in eases that have been shown to be exacerbated by coex- detail next. istent GERD. The highest frequency of GERD has been described in CF and IPF patients. All recipients are screened with barium swallows and 24-hour pH stud- Compliance ies. If there is significant reflux, patients are advised The transplanted lung remains vulnerable to acute that this is a risk factor for early obliterative bronchi- rejection throughout the lifetime of the graft. The risk olitisandthatitwillneedcarefulfollow-upandprob- of rejection is highest in the first 6 months but remains able further surgery. The procedure of choice is a Nis- high if the recipient develops low levels of immuno- san’s fundoplication, which has been shown to improve suppression. In view of this, a previous history of outcomes. poor compliance with treatment is a significant con- cern for any patients being considered for LT. Care- Psychosocial assessment ful assessment and re-enforcement of the message of The whole transplant process from referral through absolute compliance with immunosuppression is a key to the long-term follow-up involves a series of emo- requirement. tional highs and lows. For both the referring and trans- plant team, this can present dilemmas, as there is a Rehabilitation delicate balance between offering the patient hope and the realistic expectation of LT. Many patients find the The general condition of the patient needs to be suffi- uncertainty that the transplant process encompasses cient robust to be able to cope with the severe stresses extremely difficult. These situations require alarge that the process of transplantation places on the recip- degree of mental strength and fortitude to cope with ients. Potential recipients are encouraged to maintain the fluxes. Although there is no solution to these diffi- as higher level of activity as possible and to undertake culties, patients who are well supported by friends and some form of regular exercise every day. Patients with families appear to fare better. When necessary, formal severe musculoskeletal deconditioning are poor candi- evaluation and ongoing support by psychiatric or psy- dates for LT and if possible should be optimized prior chological services may be invaluable. to referral. Pulmonary rehabilitation can be extremely useful in helping to train patients in how best to exer- cise within the very limited confines of their pul- Age monary reserve. It is useful for all patients who are Age is an independent risk factor for outcome post LT, being considered as potential candidates to have had with outcomes declining with rising age, in particular at least one course of pulmonary rehabilitation. Six- over the age of 60 years. Although there is no abso- minute walk tests can be a useful metric for determin- lute age cut-off and each patient is considered on his ingthelevelofdeconditioningandanyimprovement or her clinical merit, age associated with other comor- after pulmonary rehabilitation. bidities may be an absolute contraindication. For

124 Chapter 14: Recipient selection guidance, we will consider patients up to the age of may not be successful, and these patients may require 55 years for combined heart and lung transplant, up to more aggressive intervention. For patients who are 60 years for bilateral LT, and up to 65 years for a single potential recipients, this may make explantation of LT. However, robust individuals without comorbidities the native lungs difficult. Although it is impossible who are outside this age will also be considered in some to give categorical advice for these patients as gen- centers. eral principle, it is advisable to ensure that the least disruptive intervention, pleural abrasion rather than Osteoporosis pleurectomy, is performed that will provide adequate resolution of the pneumothorax but may allow the Osteoporosis (OP) in the general population is a risk potential explantation of the lungs at the time of factor for spontaneous bone fractures. Patients with transplantation. chronic respiratory disease appear to be more vulner- able to OP than the general population. This increased vulnerability is likely to be multi-factorial; however, significant contributing factors include use of long- Disease-specific guidance term oral corticosteroids, chronic immobility, and in COPD accounts for approximately 40% of LTs per- CF patients, malabsorption of vitamin D. The preva- formed, with CF and IPF accounting for 20% each. lence of OP in waiting list recipients has been esti- IPAH accounts for approximately 5% of recipients. As mated at between 29–43%, and a much higher high these indications form the major bulk of patients, spe- rate of fractures is seen after LT in comparison with cific clinical considerations in these groups are dis- other solid organ transplants. cussed in Table 14.2. Although there are a variety of methods to assess bone density, the best established is dual-energy X- ray absorptiometry (DEXA), and using this method, COPD osteoporosis is defined as a T-score of less than –2.5. COPD patients are the largest single group of patients Using the World Health Organization (WHO; FRAX) referred for consideration of LT. Deciding on which scale, it is possible to estimate the risk of fracture for patients to refer for assessment can be difficult. Some an individual. This tool can be very helpful in estab- of the useful clinical indicators to assess for are low lishing risk profiles. Most units will regard severe or baseline lung function (forced expiratory volume in 1 symptomatic osteoporosis with T-score of less than – second [FEV1] Ͻ25 % and transfer coefficient for car- 3.5 as a relative contraindication. Some patients ben- bon monoxide [TLCO] Ͻ20% of predicted), especially efit from treatment with intravenous bisphosphonates if this is associated with resting hypercapnia and or to try and improve their bone density. hypoxia (PaCO2 Ͼ 6.0 kPa, PaO2 Ͻ 8kPa),theclin- ical syndrome of Cor Pulmonale, or the presence of Body mass index secondary pulmonary hypertension on echocardiog- Ͼ Extremes of body weight are associated with poorer raphy (estimated pulmonary systolic artery pressure outcomes after LT. Lower body mass indices (BMI) of 40 mmHg). Frequent exacerbations associated with a less than 18 are associated with poor nutritional sta- rapid decline in lung function and/or the requirement tusandcanleadtopooroutcomesduetopoorpost- for intensive care admission are useful pointers for surgical rehabilitation. Higher BMIs of greater than referral for LT. 30 present surgical challenges that may translate into The BODE index (body mass index, airflow poorer outcomes and the development of late weight- obstruction, dyspnea, and exercise capacity), although associated comorbidities. not validated in this specific group, can be useful to assess the probability of prognostic benefit from LT (Table 14.3). A BODE score of Ͼ7mayindicatea Pneumothorax patient who would gain benefit from LT. Although Patients with advanced respiratory disease due to the these measures may help to identify some patients, structural damage to the lung are very vulnerable it is important to consider the quality of life of the to the development of pneumothoraces. As a result individual, as this is a potent consideration in COPD of the underlying disease, a conservative approach patients.

125 Section 3: Lung

Table 14.2 Disease-specific guidance

Disease

COPD CF IPF IPAH

Lung function FEV1 Ͻ 25% FEV1 Ͻ 30% VC Ͻ 60% Coexisting lung disease TLCO Ͻ 20% FVC Ͻ 40% TLCO Ͻ 50%

Oxygen PaCO2 Ͼ 6 kPa PaCO2 Ͼ 6.5 kPa Resting hypoxemia PaO2 Ͻ 8 kPa PaO2 Ͻ 8 kPa PaO2 Ͻ 7.5 kPa Early desaturation on Overnight desaturation exercise (mean Ͻ 90%) Other BODE Ͼ 7 Hemoptysis Ͼ 240 ml RAP Ͼ 15 mmHg RAP Ͼ 15 mmHg ITU admissions Pneumothorax Rapid decline Mixed venous sats Ͻ 60% Frequent exacerbations Rapid decline Secondary PH MPAP Ͼ 50 mmHg Secondary PH Weight loss CO Ͻ 2l/min Poor quality of life Frequent exacerbations Failure of medical therapy Nasal ventilation Small diabetic females ITU admissions

Table 14.3 Calculation of body mass index, degree of airflow Cystic fibrosis obstruction and dyspnea, and exercise capacity (BODE) index Potential recipients with CF benefit from being looked after in dedicated CF units where they are carefully Points on BODE index followed up. Patients who have low lung function 01 2 3 (FEV1 Ͻ 30% predicted or forced vital capacity [FVC] Ͼ Ͻ Ͻ 40% predicted) should be considered for trans- FEV1 (% of 65 50–64 36–49 35 predicted) plantation. In addition, any patients who have rapidly Ͼ Ͻ declining lung function with increased frequency of 6minutewalk(m) 350 250–349 150–249 150 infective exacerbations should have the possibility of Dyspnea score 0–1 2 3 4 (4 most severe) LT raised. These may be associated with changes in Ͼ Ͻ oxygen requirement and possibly the requirement for Body mass index 21 21 non-invasive ventilation (PaCO2 Ͼ 6.5 kPa, PaO2 Ͻ 7.5 kPa). Both of these should independently right ventricular (RV) failure. These are manifest as prompt consideration of referral for LT. The devel- progressive fall in exercise tolerance (NYHA class III– opment of severe hemoptysis (Ͼ200 ml) or recurrent IV, 6-minute walk test Ͻ 350 m), syncopal episodes, pneumothoraces is associated with a poor progno- increasing peripheral edema and ascites, hemoptysis, sis, and patients with these complications should be and chest pains. In addition, patients who are deterio- referred for consideration. A specific group of patients rating despite increasing pulmonary vasodilator ther- at high risk of rapid decline are young female diabetic apy should be referred for LT. Right heart catheter patients, and they should be referred early for consid- measurements can be very helpful in helping to deter- eration. mine the patients who should be referred. IPAH IPF Over the last 5 years, there has been a dramatic IPF is the commonest of all the diffuse parenchymal improvement in the range of therapies available for lung diseases. It is unfortunately associated with a very treatment of IPAH patients. This has resulted in an poor prognosis and is largely unresponsive to current improvement in the median survival for most patients. therapies. Although large strides have been made in A rare subset of patients with pulmonary venoocclu- understanding the pathophysiology of IPF, LT remains sive disease and pulmonary capillary hemangiomato- the only therapy that has been shown to offer prognos- sis do not respond to pulmonary vasodilator therapy tic benefit. Unfortunately, IPF patients also have the and should be referred for LT early. The clinical con- highest mortality on the transplant waiting list. Early sequences of progressive IPAH are all the features of referral of patients with IPF is imperative to ensure that

126 Chapter 14: Recipient selection patients have a reasonable prospect of transplantation. resistance Ͼ 5 Wood units, or TPG Ͼ 15 mmHg) Of particular concern are patients with rapidly chang- remain indications for combined heart–lung trans- ing lung function, e.g., a fall in FVC of more than 10% plantation, although survival is inferior to that of heart in6monthsandTLCOofϽ 39% predicted. Resting transplant alone. hypoxemia (saturations Ͻ88%) and early desaturation on 6-minute walk testing are adverse prognostic signs and merit referral for transplantation. The presence of Conclusions secondary pulmonary hypertension may allude to a Lung transplantation offers prognostic benefit and an different diagnosis but may reflect the severity ofthe improvement in the quality of life for carefully selected underlying lung disease. Its presence is an additional patients. In the current era there remains a criti- marker of disease severity and should merit referral. cal shortage of donor organs, and thus unfortunately, recipient selection remains extremely important. As expertise increases, previous contraindications may be Selection for heart–lung re-examined, and patients who were previously con- transplantation sidered unsuitable may now be given the opportunity The success of single and bilateral lung transplanta- of LT. The inherent complexity of these patients means tion has improved so that combined heart–lung trans- that proscriptive inclusion and exclusion criteria will plant has fewer indications. In addition, urgent sta- not cover every situation, and thus where there is any tus heart (and in some countries, lung) transplant doubt over a patient’s suitability, early discussion with recipients has limited the availability of heart–lung the transplant team is advised. blocks so that very few procedures are performed in the current era. In the 1990s, congenital heart dis- Further reading ease with secondary pulmonary hypertension (Eisen- Dolgos S, Hartmann A, Isaksen GA, et al.Osteoporosisisa menger syndrome) was the most frequent indica- prevalent finding in patients with solid organ failure tion (approximately 30%). Other complex congeni- awaiting transplantation – a population based study. tal heart defects have also been treated successfully Clin Transplant 2010; 24: E145–52. with heart–lung transplantation, including univentric- D’Ovidio F, Singer LG, Hadjiliadis D, et al. Prevalence of ular heart with pulmonary atresia, truncus arteriosus, gastroesophageal reflux in end-stage lung disease and hypoplastic left heart syndrome, although palli- candidates for lung transplant. Ann Thorac Surg 2005; ation or repair of complex congenital lesions is now 80: 1254–60. preferred due to improved survival. In patients with Fisher AJ, Verleden GM, Massard G. Lung Transplantation simple cardiac defects, repair of the defect combined European Respiratory Monograph 45. Sheffield, United with single or bilateral lung transplantation is a poten- Kingdom: European Respiratory Journals Ltd., tial option. Primary pulmonary hypertension with September 2009. right heart failure accounted for approximately 25% of Iacono A, Groves S, Garcia J, Griffith B. Lung combined heart–lung transplants, although now bilat- transplantation following 107 days of extracorporeal eral lung transplantation is the preferred option. CF membrane oxygenation. Eur J Cardiothorac Surg 2010; 37: 969–71. remains an indication for which right heart failure is thought unlikely to recover or domino transplanta- Orens JB, Estenne M, Arcasoy S, et al.International tion is feasible. Ischemic heart disease with end-stage guidelines for the selection of lung transplant candidates update–a consensus report from the Pulmonary lung disease and cardiac failure with fixed moderate to Scientific Council of the International Society for Heart severe pulmonary hypertension (systolic pulmonary and Lung Transplantation. JHeartLungTransplant2006; artery pressure Ͼ 75–80 mmHg, pulmonary vascular 25: 745–55.

127 Section 3 Lung Chapter Living donor lobar lung transplantation

15 Hiroshi Date

Key points tomy from a smaller donor. After induction of gen- r Bilateral living donor lung transplantation in eral anesthesia, donors are intubated with a left- which two healthy donors donate their right sided double-lumen endotracheal tube. The donors or left lower lobes is an alternative to are placed in the lateral decubitus position and a cadaveric transplantation. posterolateral thoracotomy is performed though the r The most common procedure involves a right fifth intercostal space. Fissures are developed using lower lobectomy from a larger donor and a linear stapling devices. The pericardium surround- left lower lobectomy from a smaller donor. ing the inferior pulmonary vein is opened circum- ferentially. Dissection in the fissure is carried out r Right-sided donors are more likely to have a to isolate the pulmonary artery to the lower lobe perioperative complication than left-sided and to define the anatomy of the pulmonary arter- donors, probably secondary to right lower ies to the middle lobe in the right side donor and and middle lobe anatomy. to the lingular segment in the left side donor. If the r Pulmonary arterial hypertension is not a branches of middle lobe artery and lingular artery contraindication to living donor lung are small, they are ligated and divided. Intravenous transplantation. prostaglandin E1 may be administered to decrease sys- r Although only two lobes are transplanted, tolic blood pressure by 10–20 mmHg. Five thousand living donor transplantation seems to be units of heparin and 500 mg of methylprednisolone associated with less frequent primary graft are administered intravenously. After placing vascu- failure than cadaveric transplantation. lar clamps in appropriate positions, the division of the pulmonary vein, the pulmonary artery, and bronchus Living donor lobar lung transplantation (LDLLT) was are carried out in this order. On the back table, the developed at the University of Southern California lobes are flushed with preservation solution (such as (USC) to offset the mismatch between supply and ET-Kyoto solution) both antegradely and retrogradely demand for those patients awaiting cadaveric lung from a bag approximately 50 cm above the table. transplantation. A single donor was used initially; Lobes are gently ventilated with room air during the however, the results of single lobe transplantation were flush. not satisfactory. Therefore, bilateral LDLLT in which Recipients are anesthetized and intubated with a two healthy donors donate their right or left lower single-lumen endotracheal tube in children and with lobes (Figure 15.1) was developed. As of 2010, LDLLT a left-sided double-lumen endotracheal tube in adults. has been performed in approximately 350 patients The “clamshell” incision is used, and both chest cav- worldwide. The survival of LDLLT appears to be better ities are entered through the fourth intercostal space. than that of cadaveric transplantation. Pleural and hilar dissections are performed as much aspossible.Theascendingaortaandtherightatrium Surgical technique are cannulated after heparinization, and patients are The most common procedure involves a right lower placed on standard cardiopulmonary bypass (CBP). lobectomy from a larger donor and a left lower lobec- After bilateral pneumonectomy, the right lower lobe

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

128 Chapter 15: Living donor lobar lung transplantation

Figure 15.1 Bilateral living donor lobar lung transplantation. Right and left lower lobes from two healthy donors are implantedinarecipientinplaceofwhole right and left lungs, respectively.

Figure 15.2 Right lower lobe implantation. The bronchus, the pulmonary vein, and the pulmonary artery are anastomosed consecutively. The venous anastomosis is conducted between the donor inferior pulmonary vein and the recipient superior pulmonary vein.

implantation is performed, followed by the left lower Donor selection, size matching, lobe implantation. The bronchus, the pulmonary vein, and the pulmonary artery are anastomosed and outcome consecutively. The venous anastomosis is conducted Although immediate family members (relatives within between the donor inferior pulmonary vein and the third degree or a spouse) have been the only donors the recipient superior pulmonary vein (Figure 15.2). in our institution, other institutions have accepted Just before completing the bilateral implantations, extended family members and unrelated individuals. 500 mg to 1 g of methylprednisolone is given intra- Eligibility criteria for living lobar lung donation at venously, and nitric oxide inhalation is initiated at Kyoto University are summarized in Table 15.1. 20 ppm. After both lungs are reperfused and venti- Potential donors should be competent, willing to lated, CPB is gradually weaned and then removed. donate free of coercion, medically and psychosocially Three surgical teams are required in LDLLT, and suitable,andfullyinformedofrisks,benefits,andalter- they communicate closely to minimize graft ischemic native treatment available to the recipient. In our insti- time. tution, potential donors are interviewed at least three

129 Section 3: Lung

Table 15.1 The eligibility criteria for living lung donation (Kyoto dimensional computed tomography (CT) volumetry University) has been recently introduced in LDLLT following wide Medical criteria use in living donor liver transplantation. Age 20–60 years The Vancouver Forum Lung Group has summa- rized the world experience of approximately 550 living ABO blood type compatible with recipient lung donors. Sixty percent of the live lung donors have Relatives within the third degree or a spouse been male, 76% have been related to the recipient, and No significant past medical history 24% were unrelated. There has been no reported peri- No recent viral infection operative mortality of a lung donor. Approximately No significant abnormalities on echocardiogram and 4% of live lung donors have experienced an intraop- electrocardiogram erative complication that included the necessity of a No significant ipsilateral pulmonary pathology on computed right middle lobe sacrifice. Approximately 5% of them tomography have experienced complications requiring surgical or Arterial oxygen tension ≥ 80 mmHg (room air) bronchoscopic intervention. Right-sided donors may Forced vital capacity, forced expiratory volume in 1 second ≥ 85% be more likely to have a perioperative complication of predicted than left-sided donors, probably secondary to right No previous ipsilateral thoracic surgery lower and middle lobe anatomy. No active tobacco smoking Social and ethical criteria No significant psychiatric disorders Recipient selection No ethical issues or concerns about donor motivation All recipients should fulfill the criteria for conven- tional cadaveric transplantation. Because of possible serious complications in the donor lobectomy, LDLLT times to provide them with multiple opportunities to should be reserved for critically ill patients who are ask questions, reconsider, or withdraw as a donor. unlikely to survive the long wait for cadaveric lungs. Because only two lobes are implanted in LDLLT, In our LDLLT experience (n = 60), 36 patients (60%) appropriate size-matching between donor and recip- were bed bound and seven (12%) were on a ventila- ient is important. We have previously proposed a for- tor. Controversy exists regarding whether LDLLT can mula to estimate the graft forced vital capacity (FVC) be applied to patients already on a ventilator or requir- basedonthedonor’smeasuredFVCandthenum- ing re-transplantation. LDLLT may lead to better sur- ber of pulmonary segments implanted. Given that the vival than conventional cadaveric lung transplantation right lower lobe consists of five segments, the left for re-transplantation. It has been reported that peri- lower lobe of four, and the whole lung of 19, total operative mortality of re-transplantation is only 7.7% FVC of the two grafts is estimated by the following in patients who had LDLLT versus 42.3% in the cadav- equation. eric group. Total FVC of the 2 grafts In the United States, because only two lobes are = Measured FVC of the right donor × 5/19 transplanted, cystic fibrosis represents the most com- + Measured FVC of the left donor × 4/19 mon indication for LDLLT, as these patients are usu- ally small in body size. The distribution of diagnoses When the total FVC of the two grafts is more than is different in Japan, where cystic fibrosis is a very rare 45–50% of the predicted FVC of the recipient (cal- disease. Bronchiolitis obliterans, interstitial pneumo- culatedfromaknowledgeofheight,age,andsex), nia, and idiopathic pulmonary arterial hypertension we accept size disparity regardless of the recipient’s are the three major indications in Japan (Table 15.2). diagnosis. LDLLT appears possible despite pulmonary hyperten- Ithasbeenreportedthatpatientsreceivinglobes sion; in fact, mean pulmonary artery pressure has whose combined resultant total lung capacity (TLC) been shown to decrease by discharge in transplanted wasanticipatedtobemorethan80%oftherecip- patients, validating the functional capacity of the two ient’s predicted TLC had a 5-year survival of 57% lobes to handle the entire cardiac output in these compared with 26% in those who did not. Three- recipients.

130 Chapter 15: Living donor lobar lung transplantation

Table 15.2 Recipient indication for LDLLT (n = 60) (%) Bronchiolitis obliterans 16 100 90 Interstitial pneumonia 16 80 Idiopathic pulmonary arterial hypertension 15 70 Bronchiectasis 5 60 Lymphangioleiomyomatosis 4 50

Cystic fibrosis 1 Survival 40 30 Eisenmenger syndrome 1 20 Emphysema 1 10 Eosinophilic granuloma 1 0 0123456789101112 Postoperative management of Ye a r Figure 15.3 Survival after living donor lobar lung transplantation the recipient (n = 60). The 5- and 10-year survivals were 91% and 84%, The patient is usually kept sedated and ventilated for respectively. at least 3 days to maintain optimal expansion of the implanted lobes. Fiberoptic bronchoscopy should be 30 LDLLT recipients. Five- and 10-year survival rates performed regularly to assess donor airway viability were 91% and 84% in 60 LDLLT recipients, respec- and suction any retained secretions. Bedside postop- tively (Figure 15.3). erative pulmonary rehabilitation is initiated as soon as The question of whether two pulmonary lobes can possible. provide sufficient long-term pulmonary function and Postoperative immunosuppression usually consists clinical outcome to recipients is debated. The USC of triple-drug therapy with cyclosporine (CyA) or group has reported that LDLLT provides compara- tacrolimus (TAC), azathioprine (AZA), or mycophe- ble intermediate and long-term pulmonary function nolate mofetil (MMF) and corticosteroids without and exercise capacity to bilateral cadaveric lung trans- induction. Acute rejection should be diagnosed on plantation in adult recipients surviving more than the basis of radiographic and clinical findings with- 3 months after transplantation. out transbronchial lung biopsy because the risk of pneumothorax and bleeding after transbronchial lung Comparison with cadaveric lung biopsy may be greater after LDLLT. Because two lobes transplantation are donated by different donors, acute rejection is usu- The current availability of cadaveric donor lungs has ally seen unilaterally. Early acute rejection episodes are not been able to meet the increasing demand of poten- characterized by dyspnea, low-grade fever, leukocy- tial recipients in most counties. The recent change by tosis, hypoxemia, and diffuse interstitial infiltrate on the Organ Procurement and Transplantation Network chest radiograph. A trial bolus dose of methylpred- in the United States to an urgency/benefit allocation nisolone 500 mg is administered, and clinical signs are system for cadaveric donor lungs in patients 12 years carefully observed. and older has most recently led to a marked decline in LDLLT. However, in countries where such an alloca- Recipient outcome tion system does not exist, LDLLT is still a relatively There are only three groups that have reported recipi- common procedure. For example, the average waiting ent outcome. The USC group recently published their time for a cadaveric lung is more than 3 years in Japan, 10-year experience of 123 LDLLT recipients, includ- hence the requirement for LDLLT. ing 39 children. In their series, re-transplantation and In general, the ischemic time for LDLLT is much mechanical ventilation were identified as risk factors shorter than cadaveric transplantation. Although only for mortality. One, 3- and 5-year survival rates were two lobes are transplanted, LDLLT seems to be asso- 70%, 54%, and 45%, respectively. The St. Louis group ciated with less frequent primary graft failure. We reported similar results in 38 pediatric LDLLT recip- believe that using a “small but perfect graft” is a great ients. We recently published institutional results in advantage in LDLLT.

131 Section 3: Lung

Experienced centers have recently reported the Further reading incidence of bronchial complications in cadaveric lung BarrML,BelghitiJ,VillamilFG,et al. Areportofthe transplantation to be about 5%. Contraindications to Vancouver Forum on the care of the live organ donor. cadaveric lung transplantation include current high- Lung, liver, pancreas, and intestine data and medical dose systemic corticosteroid therapy because it may guidelines. Transplantation 2006; 81:1372–87. increase airway complications, although low-dose pre- Date H, Aoe M, Sano Y, et al. Improved survival after transplantation corticosteroid therapy (≤20 mg/day living-donor lobar lung transplantation. JThorac of prednisone) is acceptable. Various factors, such Cardiovasc Surg 2004; 128: 933–40. as short donor bronchial length, high blood flow in Date H, Kusano KF, Matsubara H, et al. Living-donor lobar the small grafts implanted, and well-preserved lung lung transplantation for pulmonary arterial parenchyma with short ischemic time, may contribute hypertension after failure of epoprostenol therapy. JAm to better oxygen supply to the donor bronchus, result- Coll Cardiol 2007; 50: 523–7. ing in excellent bronchial healing in LDLLT. Shoji T, Bando T, Fujinaga T, Date H. Living-donor Bronchiolitis obliterans syndrome (BOS) has been single-lobe lung transplant in a 6-year-old girl after the major obstacle after cadaveric lung transplanta- 7-month mechanical ventilator support. JThorac tion. LDLLT may be associated with a lower incidence Cardiovasc Surg 2010; 139: e112–13. of BOS, especially in pediatric patients. Transplanting Starnes VA, Bowdish ME, Woo MS, et al. A decade of living two lobes obtained from two different donors appears lobar lung transplantation. Recipient outcomes. JThorac to be beneficial in the long term, because the contralat- Cardiovasc Surg 2004; 127: 114–22. eral unaffected lung may function as a reservoir in case Sweet SC. Pediatric living donor lobar lung transplantation. of unilateral BOS. Pediatr Transplantation 2006; 10: 861–8.

132 Section 3 Lung Chapter Surgical procedure

16 Faruk Ozalp,¨ Tanveer Butt, and Stephan V.B. Schueler

Key points for these changes is based on case series and informa- r tion gathered from large international registries rather Bilateral lung transplantation has evolved than on randomized controlled trials (RCTs), in par- into a routine procedure and is the most ticular, the registry of the International Society for frequently performed method. Combined Heart and Lung Transplantation (ISHLT). heart–lung transplant has been in decline Although single lung transplantation (SLT), bilat- internationally, with negligible overall eral lung transplantation (BLT), and combined heart– numbers now reported annually. r lung transplantation (HLT) have all been proven to be The surgical incision is not only successful for the appropriate recipient, based on the center-specific, but also surgeon-specific. information from the registries, some of these proce- Some centers advocate smaller incisions to dures have undergone considerable modification over minimize trauma and allow faster time. Overall, BLT has evolved into a routine proce- postoperative recovery; however, there is dure and is the most frequently performed method of limited evidence in this area of practice. r the three, whereas HLT has been in decline interna- Cardiopulmonary bypass is often avoided for tionally, with negligible overall numbers now reported both single and bilateral lung annually. transplantation; however, clamping of the pulmonary artery leading to significant increase in pulmonary artery pressure, right Single lung transplantation atrial distension, and desaturation or Traditionally SLT has been the procedure of choice hypercapnia should prompt its use. in patients with non-infective end-stage lung dis- r The technique of lung donation after cardiac ease such as chronic obstructive pulmonary disor- death has been established successfully in der (COPD); idiopathic pulmonary fibrosis (IPF); and ␣ recent years, thereby avoiding damage to rare conditions such as 1-antitrypsin deficiency, sar- donor organs by the effect of brainstem death coidosis, and pulmonary vascular disease. However, and expanding the donor pool. in recent years, there has been a clear shift toward BLT for these patients because the overall outcome has improved and the long-term beneficial impact of BLT on quality of life and exercise capacity has become Since the early series of successful lung transplanta- more apparent. tion in the 1980s, it has become clear that this treat- ment method offers the best outcome for patients with deteriorating end-stage lung disease and who are unre- Technique sponsive to conventional therapy. Over the past two Different incisions are used for SLT. Most common decades, some adjustments have been made regarding is the posterolateral thoracotomy; less frequent are recipient and donor selection, treatment of rejection, anterolateral thoracotomy and median sternotomy, maintenance immunosuppression, and, importantly, whichareusuallyusedifCPBhastobeemployed. the selection of the surgical procedure. The evidence The selection of the surgical incision is not only

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

133 Section 3: Lung

center-specific but also surgeon-specific. Furthermore, bronchial ischemia; however, a number of reports have some centers advocate smaller incisions to minimize demonstratedsimilarorbetterresultsusingcontinu- trauma and allow faster postoperative recovery; how- ous suture alone. ever, there is limited evidence since the numbers are ASatinskyclampisplacedacrosstheligatedpul- small and RCTs are not available. monary veins to facilitate the suture of the left atrium. After anaesthesia and tracheal intubation with a On the right side, it is beneficial to dissect into the double-lumen tracheal tube, the patient is secured on inter-atrial groove, which allows more secure place- the table in the lateral position. The tracheal tube ment of the clamp. The bridge between the upper position is re-checked, as it can be displaced by this and lower vein is cut, creating a single left atrial cuff. maneuver. Skin preparation is carried out on the chest, Both cuffs are carefully aligned and a continuous 4/0 abdomen, and ipsilateral groin. monofilament suture is used for anastomosis. The For posterolateral thoracotomy, a curved incision suture is left untied to allow de-airing via this route is made from the mid-axillary line around the infe- later on. The PA is anastomosed with a running 5/0 rior angle of the scapula, and the pleural space is monofilament suture and the first branch of the recip- entered through the fifth intercostal space. Any adhe- ient PA is used for appropriate alignment. This is cru- sions are dissected, avoiding trauma to phrenic and cial to avoid kinking or twisting of the anastomosis. vagal nerves, and the hilar structures are exposed. After meticulous de-airing via the left atrial suture line, The pulmonary artery and veins and the main the clamps are taken off and the donor lung is reper- bronchus are identified, dissected, and encircled with fused in a controlled fashion by gradual release of the tapes. Before dividing the artery and the veins, it is PA clamp over 10 minutes. Lungs are ventilated with important to decide whether cardiopulmonary bypass room air. (CPB) has to be used to allow safe implantation of IfCPBisused,aPAventcandecompresstheRV, the new lung. Therefore, a vascular clamp is placed allowing a bloodless field that facilitates an open PA on the main pulmonary artery (PA) for 10 minutes and anastomosis without the need for a clamp. This also the hemodynamic and respiratory stability is assessed. allows easier alignment of the vessels. If the left atrial Significant increase of PA pressures, right atrial disten- cuff is very short, CPB facilitates the heart to be fibril- sion, and desaturation or hypercapnia should prompt latedorevencardioplegedtofacilitateanopensurgi- the use of CPB. Cannulation for right SLT is either cal anastomosis. After completion of the anastomoses, throughtherightfemoralvesselsorviathethoraco- transesophageal echocardiography (TEE) examina- tomy using the ascending aorta and the right atrium. tion of the left atrial anastomosis allows exclusion ofa In left SLT, arterial cannulation is performed, either pressure gradient. Bronchoscopy can clear the airway via the left femoral artery or the descending aorta, and can help to identify any suture line problems. The and the main pulmonary artery is used as the venous chestisclosedinstandardfashionoverapicalandbasal return. drains. The extraction of the lung continues with clamp- ing and division of the PA followed by the pulmonary veins, which are isolated outside the pericardium, lig- Bilateral lung transplantation ated and divided. The main bronchus is then dissected The number of BLTs (otherwise known as bilateral justtothelevelabovetheupperlobebronchialtake- sequential single-lung transplant) is now on the rise off.ThePAisfurtherdissectedoffthesurrounding because of improved outcomes, exercise capacity, and tissue, ensuring a sufficient length for the anastomo- quality of life. Initially this procedure was considered sis. For the purpose of orientation, the first branch is to be more suitable for patients with infective lung marked for alignment with the donor PA. diseases such as cystic fibrosis (CF) or bronchiectasis, Using a running 4/0 monofilament suture, the making the new SLT at risk of cross-contamination. bronchus is anastomosed first, avoiding extensive dis- Nowadays, patients with COPD and IPF who tradi- section of adjacent recipient tissue that could jeopar- tionallywouldhavereceivedSLTareincreasinglycon- dize bronchial healing. In earlier series, a combina- sidered for BLT. tion of a running suture for the membranous part of The question of whether there is evidence for or the bronchus and interrupted figure-of-eight sutures againsttheuseofCPBforBLTseemstobeunan- for the cartilaginous part was used in order to prevent swered. Avoiding CPB has the well-known advantage

134 Chapter 16: Surgical procedure of evading complications such as increased inflam- CPB can be weaned off and the cannulae removed. matory response and coagulopathy. In our own expe- After hemostasis and placement of apical and basal rience, however, these complications seem to have chest drains on each side, the chest is closed using two no effect on outcomes. The use of CPB may have or three wires for the sternum and pericostal Vicryl significant advantages throughout the whole proce- sutures for the ribs. dure, including hemodynamic stability, better access in Echocardiographic assessment of the left atrial and patients with dense adhesions, better management in PA suture lines to exclude flow-limiting narrowing of technically challenging situations, and the possibility the lumen is important, as is flexible bronchoscopy of controlled reperfusion of the transplanted lung. to assess the bronchial anastomosis before leaving the The preferred surgical incision for BLT is a bilateral operating theatre. Broncho-alveolar lavage samples of transverse thoracotomy joint across the middle, best donor lung are analyzed for unknown donor bronchial known as a “clamshell” incision. Bilateral thoracotomy infection. through the fourth intercostal spaces is performed and the sternum divided after ligation both of the internal thoracic arteries. Combined heart–lung transplantation Alternatively, a median sternotomy is advocated, Depending on the recipient diagnosis, the removal of particularly in patients in whom adhesions are not the heart and both lungs can be quite challenging. In anticipated, e.g., emphysema patients. However, access patients with congenital heart disease and cyanosis, to hilar structures especially on the left side can be lim- there are large mediastinal collaterals requiring care- ited. The advantages of a median sternotomy are better ful dissection and ligation. In addition, the chance of woundhealingandpaincontrol. damage to the phrenic, recurrent laryngeal, and vagus After gaining access to the mediastinum and sys- nerves is significant. Through a median sternotomy temic heparinization, the pericardium is opened, the and wide opening of the pleural spaces, some of the ascending aorta and the right atrium are cannulated anterior part of the pericardium is removed. Preserv- for CPB, and the dissection of both hilar structures is ing the lateral walls of the pericardium together with commenced. Special care is taken to prevent recurrent the phrenic nerve prevents the heart from slipping laryngeal, phrenic, and vagal nerve injury. The inferior into the left chest. Incision of the pericardium poste- pulmonary ligaments are divided and the hilar struc- riortothephrenicnerveatthehilumisperformedto tures prepared for lung extraction as for SLT. Once allow the donor lungs to pass through later on. the lungs are mobilized and hilar dissection is com- Cannulation for CPB is achieved using the ascend- pleted, CPB can be commenced. A pulmonary vent ingaortaandboththeinferiorvenacava(IVC)and is placed for a bloodless field. Both lungs are resected superior vena cava (SVC) with tapes around for sealed with meticulous hemostasis, especially in the posterior occlusion. CPB is commenced and body temperature hilum, as access to this area is very restricted once the is lowered to 28◦C. The ascending aorta is dissected lung is implanted. In case of infected lung disease, a off the PA and cross-clamped. Routine cardiectomy is thorough washout of the thoracic cavity is carried out carried out by dividing the IVC and SVC, ascending to reduce the risk of contamination due to transection aorta, PA, and left atrium, leaving the left atrial cuff of the main bronchi. within the pericardium. The left-sided phrenic nerve The preparation of the donor lungs and implanta- is freed by a horizontal incision anterior to the pul- tion process are the same as for SLT. If no CPB is used, monary veins. Pulmonary ligaments are divided and the recipient lung with the worse function is implanted hilar structures dissected on both sides. Pulmonary first.ButifCPBisused,thesequenceseemsmore vessels are ligated and the main bronchus is stapled, related to surgeon preference. with care taken to avoid injury to the vagus and recur- Cold gauze packs are put on the posterior wall of rent laryngeal nerves. Intrapericardial remnants of the the thoracic cavity before the donor lung is placed left atrium and pulmonary artery are removed. Only inside. Once both lungs are implanted and de-airing a small portion of PA at the level of the ligamentum is completed, a period of 30 minutes of controlled arteriosum is left intact to save the recurrent laryngeal reperfusion is commenced in order to ameliorate lung nerve. The pericardium at the superior mediastinum injury. During this period of reperfusion, a mean PA is opened and the distal trachea dissected and stapled pressure of 15–20 mmHg should not be exceeded. justabovethecarina.Thetracheawillnotbeopened

135 Section 3: Lung

until the donor organs are prepared for implantation. evidence that the lung parenchyma can survive for up Before placing the donor organs inside the chest cav- to 2 hours without ventilation or circulation if the lung ity, meticulous hemostasis, especially of the posterior is inflated at the time of circulatory arrest on account mediastinum, is secured, and a thorough washout is of oxygen still being present in the alveoli. performed. Proper organ assessment of DCD lungs is diffi- The heart–lung block is trimmed and brought into cult due to a lack of information in such donors. the operative field and the lungs are passed into the For instance, the differential pulmonary vein gases, pleural spaces through gaps created below the phrenic which are often crucial in decision making about the nerves. Alternatively the organ block can be placed viability of donor lungs, is not available in this sce- anterior to the phrenic nerves. This minimizes the risk nario. Thus the assessment is subjective and done of nerve damage and allows easier hemostasis. The mainly by visual inspection and palpation of the chest is irrigated with cold saline solution. lungs. This clearly requires significant lung transplant Tracheal anastomosis is performed first using con- experience. tinuous 4/0 monofilament suture and after trimming After cardiac arrest and a “hands-off” period, bron- the donor trachea to just one cartilaginous ring above choscopy is performed. Rigid bronchoscopy is pre- the carina. The recipient trachea is transected just ferred for adequate removal of secretions. The trachea above the carina. The donor aorta is then anastomosed is then re-intubated and ventilation commenced with using a 4/0 monofilament suture. After meticulous de- inspired oxygen of 30%. This is all carried out while the airing, the aortic cross-clamp can be released and the abdominal surgeons are in the process of performing two caval anastomoses performed while the heart is laparotomy and inserting perfusion cannulae for kid- being re-perfused. Alternatively, the IVC anastomosis ney and liver harvesting. can be carried out with the cross-clamp in place and Median sternotomy is performed; pericardium and with suction catheter in the right atrium for a blood- pleural cavities are widely opened. The lungs are less field. Then the SVC anastomosis can be performed assessed visually for compliance, atelectasis, tumor, without a cross-clamp. edema, consolidation, or any other abnormality. The The lungs are re-perfused and ventilated gently PA is incised and any clot is removed by suctioning. with room air and reduced tidal volume for 20–30 Low-potassium dextran solution is infused antegrade minutes. The patient is re-warmed and weaned from into the PA, and the left atrial appendage is incised CPB, followed by de-cannulation. Two chest drains are to allow efflux of the preservation solution. To aug- placed in each chest cavity. After meticulous hemosta- ment hypothermia, topical cold saline is poured onto sis, the sternotomy is closed as usual. the lung surface. Uniform distribution of the preserva- tion solution is essential for adequate protection. After Lung transplantation from donation completion of perfusion, the lungs are removed as usual. Retrograde flush through the pulmonary veins after cardiac death enhances protection and helps to further remove clots After intensive animal research and clinical experience in the PA. If there is any doubt about the organ qual- gained from kidney and liver donation, the technique ity, very recent studies suggest that further assessment of lung donation after cardiac death (DCD) has been of the lungs on the ex-vivo lung-perfusion apparatus is established successfully in recent years. a valuable adjunct tool for assessing the suitability of Damage to donor organs is aggravated by the DCD lungs prior to implantation. effect of brainstem death in donation after brain death (DBD) by the inflammatory response, release of cat- Further reading echolamines, hemodynamic instability, and metabolic Aigner C, Jaksch P, Seebacher G, et al. Single running and neuroendocrine derangement, resulting in organ suture – the new standard technique for bronchial injury. Avoiding brainstem death and its detrimen- anastomoses in lung transplantation. Eur J Cardiothorac tal effects is an attractive alternative. However, this Surg 2003; 23: 488–93. requires a different logistical approach to lung preser- Gammie JS, Cheul Lee J, Pham SM, et al. Cardiopulmonary vation, as the organ will have to go through a period of bypass is associated with early allograft dysfunction but warm ischemia without significant antegrade circula- not death after double-lung transplantation. Thorac tion via the PA or the bronchial arteries. There is good Cardiovasc Surg 1998; 115: 990–4.

136 Chapter 16: Surgical procedure

Garfein ES, McGregor CC, Galantowicz ME, et al. Macchiarini P, Ladurie FL, Cerrina J, et al.Clamshellor Deleterious effects of telescoped bronchial anastomosis sternotomy for double lung or heart-lung in single and bilateral lung transplantation. Ann transplantation? Eur J Cardiothorac Surg 1999; 15: Transplant 2000; 5: 5–11. 333–9. Hlozek C, Smedira NG, Kirby TJ, et al. Cardiopulmonary Pochettino A, Bavaria JE. Anterior axillary muscle-sparing bypass for lung transplantation. Perfusion 1997; 12: thoracotomy for lung transplantation. Ann Thorac Surg 107–12. 1997; 64: 1846–8.

137 Section 3 Lung Chapter Management during surgery

17 DavidIpandPeterSlinger

Key points ing list. Specific information required includes patient r Thoracic epidural analgesia should be height and current weight; results from latest pul- considered in all cases, but may be most monary function test; trans-thoracic echocardiogra- safely sited postoperatively. phy; left heart catheterization if appropriate; lung per- r A left-sided double-lumen endotracheal tube fusion scan, which gives information on which lung is preferred in all cases except left-sided will better tolerate one-lung ventilation (OLV); pres- single lung transplant when a right-sided ence of antibodies to the donor; and current blood double-lumen tube may be better. tests. r Dynamic hyperinflation may occur in severe In addition to the usual anesthetic issues of obstructive lung disease, leading to decreased aspiration risk, airway assessment, comorbidities, venous return and circulatory collapse. medications, and adverse reactions, assessment on the day of surgery focuses on the current illness r Right ventricular failure may be encountered state and amount of deterioration since investiga- at induction of anesthesia, after commencing tions were performed, as the patient’s physical state one-lung ventilation, during manipulation of may be significantly worse than investigations may the hilum, at pulmonary artery clamping, suggest. after reperfusion, and with severe early graft dysfunction. r Inhaled aerosolized prostacyclin is an Premedication alternative to inhaled nitric oxide; it is less Normal medications that need to be continued expensive and does not require a bulky include bronchodilators, antibiotics, and pulmonary delivery system. vasodilators. If the patient is receiving intravenous prostaglandins, it is continued until cardiopulmonary bypass (CPB) is initiated. Immunosuppression Preparation regimes commence pretransplantation and vary between institutions. Broad-spectrum antibiotics are Patient assessment given for prophylaxis, but for patients with suppu- The broad categories of end-stage lung disease requir- rative lung disease or a current chest infection, the ing lung transplantation (LT) are suppurative, obstruc- choice of antibiotics will be determined by the actual tive, restrictive, and pulmonary vascular. They each or suspected microbiological burden. The microbio- lend themselves to different potential complications logical burden of the donor is also taken into account. during anesthesia and thus require slightly different Due to the lack of respiratory reserve, sedation outside management strategies. of the operating room is not recommended and The nature of transplantation surgery is that it is should only be given with extreme caution, as it may unpredictable and emergent. Hence the preoperative easily precipitate a cardiorespiratory arrest due to work-upoftransplantrecipientsmustbethoroughly hypoxemia, hypercarbia, or increased pulmonary performed in advance, with appropriate updating of vascular resistance (PVR) resulting in acute right clinical data and investigations while on the wait- ventricular (RV) failure.

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

138 Chapter 17: Management during surgery

Thoracic epidural analgesia instability. It may also identify intracardiac thrombus and other unexpected abnormalities. In the case of Adequate postoperative analgesia is important to facil- unexplained or refractory hypoxemia, it can detect the itate extubation. Thoracic epidural analgesia (TEA) presence of intracardiac shunting. provides superior analgesia compared with systemic opioids; however, special problems and risks with preoperative epidural insertion need to be consid- Induction ered in the LT population. First, in the event of a The induction of anesthesia is one of the most criti- bloody tap, the risk of an epidural hematoma is mag- cal periods, and the surgeon and perfusionist must be nified with subsequent full heparinization for CPB. present and prepared to urgently perform sternotomy Timely decompression may be delayed due to pro- and initiation of CPB in the event of severe cardiores- longed surgery and inability to elicit clinical signs in piratory instability. In the patient with little or no car- the sedated and intubated patient. Second, the ben- diorespiratory reserve, cardiovascular collapse can be efitofTEAisshortenedorlostiftrachealextuba- precipitated by many factors. These include hypoxia, tion is delayed due to other complications. Therefore, hypercarbia, the reduction of endogenous sympathetic patients should be selected carefully based on risk drive, drugs causing myocardial depression or vasodi- and benefit discussions. Alternatives are to have the latation, and the commencement of positive pressure epidural inserted postoperatively in an awake or lightly ventilation of the lungs causing reduction of systemic sedated patient prior to tracheal extubation or only if venous return and increased RV afterload. systemic multi-modal analgesics are inadequate. For Therefore, the hemodynamic goals of induction of single-sided surgery, paravertebral catheter insertion anesthesia are to preserve systemic vascular resistance is an alternative to TEA. (SVR), myocardial contractility, and avoidance of any increase in PVR. This can usually be achieved with a Monitoring and vascular access titrated narcotic-based induction consisting of 0.05– ␮ Vascular access comprises a large peripheral intra- 0.1 mg/kg of midazolam, 5–10 g/kg of fentanyl, and venous (IV) cannula, a pulmonary artery (PA) catheter judicious doses of propofol followed by a muscle relax- sheath with side port, a multi-lumen central line, and ant. Other induction regimes have also been described. an arterial cannula. Mandatory monitoring includes In patients with high aspiration risk, gastric acid lower- five-lead electrocardiography; pulse oximetry; inva- ing premedication and cricoid pressure during induc- sive measurement of arterial, central venous, and PA tion with suxamethonium or high-dose rocuronium pressures; urine output via an indwelling catheter; is used. A non-titrated rapid-sequence induction is temperature; capnography; spirometry; and anesthetic rarely justifiable due to the risk of hemodynamic col- agent gas analysis. lapse. Ketamine is an alternative and ideal choice of Other options include depth-of-anesthesia moni- induction agent in patients with severe pulmonary toring, cerebral oximetry using transcutaneous near hypertension. infrared spectroscopy, continuous arterial blood gas With the exception of a left-sided single lung trans- monitoring, continuous cardiac output, and mixed plant in which a right-sided double-lumen endotra- venous oximetry. cheal tube may be better, a left-sided double-lumen endotracheal tube is preferred for all other cases. Use of a bronchial blocker through a single-lumen Transesophageal echocardiography tube is an alternative, but a double-lumen tube allows There is no consensus about routine use of intraop- increased surgical flexibility, irrigation of the divided erative transesophageal echocardiography (TEE) dur- bronchus, differential ventilation after the graft is ing LT for assessing surgical anastomotic sites. It is reperfused, and faster lung isolation. The double- useful for evaluation of pulmonary hypertension, RV lumen tube is changed to a single-lumen tube at the dysfunction, and suspicion of a patent foramen ovale. end of surgery before leaving the operating room. It provides useful information about left and right Doing this under direct vision or preferably using a heart preload, left ventricular (LV) and RV function, tube exchange catheter is recommended, as airway regional wall motion abnormalities, and intracardiac edema and swelling can occur during the course of air, especially when used in situations of hemodynamic the operation. Loss of the airway in this situation

139 Section 3: Lung

Table 17.1 Disease-specific intraoperative anesthetic considerations Recipient pathology Intraoperative complications

Emphysema (COPD, ␣1-antitrypsin deficiency) Hypotension with positive-pressure ventilation due to auto-PEEP.

Cystic fibrosis Profuse thick secretions. Difficult to maintain baseline PaCO2 with positive pressure ventilation. Small stature, often difficult access in chest for surgeon. Idiopathic pulmonary fibrosis May have associated diseases (e.g., scleroderma). Often older with coronary artery disease. May have severe pulmonary hypertension. May not tolerate one-lung ventilation and may require CPB. Primary pulmonary hypertension Cardiovascular collapse secondary to hypotension on induction. Always require CPB. Bronchoalveolar carcinoma Profuse watery secretions

has occurred with disastrous outcomes. The anesthetic Restrictive lung disease considerations specific to recipient pulmonary pathol- This group includes patients with idiopathic pul- ogy are shown in Table 17.1. monary fibrosis, connective tissue disease, and drug- or radiation-induced parenchymal lung disease. Ven- Obstructive lung disease tilating these patients can be difficult. To obtain This group includes patients with chronic obstruc- adequate tidal volumes, peak inspiratory pressures ␣ tive pulmonary disease (COPD) and 1-antitrypsin exceeding 40 cm H2O may be required. This is toler- deficiency, and bronchiolitis obliterans syndrome pre- ated especially if bilateral lung transplantation (BLT) senting for re-transplantation. Patients with obstruc- is planned. Volume-controlled ventilation allows bet- tive lung disease may undergo a single or bilateral LT. ter control of tidal volumes, and some ventilators The majority of this group will have smoking-related may not allow pressure control ventilation mode to emphysema and so may have coexisting cardiovascular exceed 40 cm H2O. These patients benefit with some disease. PEEP. An intensive care unit–level ventilator may be A preinduction arterial blood gas is required for required if the ventilator on the anesthesia delivery baseline PaCO2. There is an increased risk of pneu- unit is not adequate. Pulmonary hypertension is com- mothorax with positive pressure ventilation and cen- mon and can be severe. Elderly patients with end-stage tral line insertion. Cor pulmonale may exist with restrictive lung disease are one subgroup in whom the increased PVR, elevated PA pressure, and right heart long-term outcomes from SLT are equivalent to those dysfunction. These patients already have a reduced from BLT. venous return to the left heart with reduced LV end- diastolic volume, reduced stroke volume, and reduced cardiac output compared with normal patients. A Suppurative lung disease longer expiratory time is required to prevent dynamic The majority of these patients have cystic fibrosis (CF), hyperinflation, further reduction in venous return, but may also include patients with non-CF bronchiec- and circulatory collapse. Disconnecting the endotra- tasis. At induction, a single-lumen tube is inserted to cheal tube (ETT) and allowing lung deflation will per- allow a fiberoptic bronchoscope with a large working mit venous return to the heart and restore blood pres- channel to perform adequate pulmonary toilet. The sure if this is the cause of hypotension. These patients single-lumen tube is then changed to a double-lumen have a high level of intrinsic positive end expiratory tube. Aggressive airway toileting is required during pressure (PEEP), and external PEEP is usually not surgery as lung manipulation can squeeze purulent required and may in fact cause hyperinflation, increase alveolar secretions into the large conducting airways. shunt, and reduce PaO2. A large alveolar dead space Intraoperative ventilation of CF patients can be makes the end tidal CO2 underestimate the PaCO2 by extremely difficult and may require high airway pres- an unpredictable amount. Frequent correlation to arte- sures and relatively large volumes to maintain the rialbloodgasesisrequired.Itisappropriatetoventilate patient’s preoperative PaCO2. For chronically hyper- the patients to their baseline PaCO2. capnic patients, if CPB is required, we try to maintain

140 Chapter 17: Management during surgery the PaCO2 during CPB at the patient’s normal prein- Table 17.2 Treating hypoxia during one-lung ventilation duction baseline. Due to multi-resistant organisms, Increase FiO2 to 1.0 some CF patients require consultation with infec- Resume two-lung ventilation tious diseases or microbiology specialists to determine CPAP or O insufflation to non-ventilated lung appropriate antibiotic cover. 2 Optimize PEEP to ventilated lung Expediting surgical ligation of ipsilateral pulmonary veins and Pulmonary hypertension arteries to reduce shunt fraction In the context of patients with pulmonary hyperten- sion presenting for lung transplantation, the cause may be (1) idiopathic, (2) associated with lung disease or OLV are listed in Table 17.2. Severe respiratory acidosis hypoxia (pulmonary fibrosis, COPD), or less com- (pH Ͻ 7.2) can be problematic, and strategies emplo- monly,(3)associatedwithconnectivetissuedisease yed to increase alveolar minute volume may ultimately and (4) intracardiac shunt with Eisenmenger physiol- be unsuccessful, therefore requiring CPB. ogy.Itmayalsobeseenpost-transplantationinpri- Temperature monitoring and active warming is mary graft dysfunction. required, as hypothermia worsens pulmonary hyper- In patients with severe pulmonary hypertension tension, coagulopathy, and risk of arrhythmias. We and right heart failure, it is advisable to have cen- routinely use both upper- and lower-body forced-air tral venous access and the pulmonary artery catheter heating blankets. Magnesium sulphate (2 g) is infused floated into position prior to induction because these to try to prevent arrhythmias. patients may become hemodynamically compromised Repeated hilar manipulation and cardiac compres- with induction of anesthesia and need inotropic or sion lead to reduced cardiac output and hypotension. vasopressor support. Preinduction surgical femoral Vasopressor support is invariably required to main- cannulation under local anesthesia for cardiopul- tain adequate perfusion pressure. Optimizing fluid sta- monary bypass should be considered for very high-risk tus is important, but excessive fluid therapy is to be patients. avoided as the graft is susceptible to pulmonary edema. Finding evidence to recommend one type of fluid over Maintenance of anesthesia anotherisdifficultduetomultipleconfoundingfac- Patients undergoing LT feature prominently in stud- tors and small series of patients; however, the volume ies of intraoperative awareness. Care must be taken to of intraoperative gelatin-based colloid given has been ensure that these high-risk patients receive adequate associated with worse postoperative oxygenation and amounts of anesthetic agents. Maintenance of anesthe- delayed extubation. sia by propofol infusion, inhalational anesthetic agent, or both have been described. Nitrous oxide is avoided Single versus bilateral LT due to its deleterious effect on PVR. The theoretical Most patients with end-stage parenchymal lung dis- advantage of a total IV technique is less inhibition ease can get symptomatic improvement with SLT. Sup- of hypoxic pulmonary vasoconstriction, although this purative lung disease is a contraindication for SLT. does not translate to a clinically significant increase in Although SLT allows more patients to undergo trans- PaO2 when less than 1.0 MAC of inhalational anes- plantation by splitting pair of lungs between two thetic is used. There is also less myocardial depres- patients, case-control studies have shown BLT to be sion and smoother transition to CPB if required. The an independent factor in improved long-term survival. advantage of an inhalational technique is bronchodi- Potential therapeutic benefits of BLT include a reduc- latation, especially in patients with reversible obstruc- tion in alveolar damage during reperfusion, improved tive airways disease. A disadvantage is the slower pulmonary compliance and mechanics, and the avoid- wash-in of the anesthetic due to the limited minute vol- ance of native lung pathology. ume encountered in some of these patients. OLV is better tolerated in the lung with greater per- fusion. The shunt fraction is greater compared with Cardiopulmonary bypass OLVinthedecubituspositionduetothelackofben- In institutions that do not use CPB routinely, planned efit of gravity. The strategies of managing hypoxia on CPBisusedwhenpatientshaveseverepulmonary

141 Section 3: Lung

hypertension, require a concomitant cardiac proce- ated with a longer period of postoperative mechanical dure such as repair of intracardiac shunt, or require ventilation, more pulmonary edema, and increased plasmapheresis due to human leukocyte antigen early mortality, although this is controversial. (HLA) mismatch and presence of HLA antibodies against the donor. In unplanned cases, the decision RV failure to use CPB is made if the patient (1) does not tol- RV failure can be precipitated by an acute increase in erate OLV (either hypoxia or refractory hypercarbia), afterload or ischemia due to either prolonged hypoten- (2) does not tolerate clamping of a PA (RV failure), or sion or air embolism. This is encountered at induc- (3) has other refractory hemodynamic instability. If tion, after commencing OLV, during manipulation of high PA pressures do not fall after the first side is reper- the hilum, at PA clamping, after reperfusion, and with fused, CPB may also be commenced to protect the graft severe early graft dysfunction. RV afterload reduc- from pulmonary hypertension, which can contribute tion is the primary treatment goal. Also important to primary graft dysfunction. CPB is required if the are preserving coronary perfusion and biventricular patient already is dependant on extracorporeal mem- inotropic support. Basic things such as correcting aci- brane oxygenation (ECMO) or the Novalung interven- dosis, hypoxia, hypercarbia, hypothermia, minimizing tional lung assist device, although some centers con- ventilation pressures, and undoing the precipitating sider this a contraindication to LT due to potentially event (if possible) should not be overlooked. Inotropes poorer outcomes. In our adult practice, two thirds of most commonly used are phosphodiesterase inhibitors cases are done without CPB. In the cases in which CPB (e.g., milrinone), norepinephrine, and epinephrine. is used, two thirds are elective, and one third is for spe- cific intraoperative problems, most commonly hemo- dynamic instability. Stages of surgery Warm (37 ◦C) beating heart CPB is routinely used unless a concomitant cardiac procedure is also per- Dissection and removal of native lung formed.Bypasstimesareoftenprolonged,soacen- This may be difficult and prolonged if extensive pleu- trifugal pump offers some advantage over a continu- ral adhesions are present. This is more likely in patients ous flow roller pump. Venous drainage may often be with previous thoracic surgery or restrictive or suppu- impaired by severe surgical manipulation of the heart rative lung disease. Significant blood loss may occur. and hilum; therefore, vigilance is required, and flows For BLT performed off pump, if there is a significant may often need to be temporarily reduced to prevent perfusion asymmetry, the lung with the lower perfu- complete emptying of the reservoir. Bicaval cannula- sion is transplanted first. Testing the hemodynamic tion provides less interruption of venous return than consequences of dividing the PA by pinching it off tem- single two-stage atrial cannulation. porarily is advised, as the RV can acutely fail. Reperfusion and ventilation of the first side com- menceswhilestillonCPBtolimitthewarmischemic Anastomosis of donor lung time of the graft. Major coagulopathy is invariably Surgical access to the hilum and atrium requires present after coming off prolonged CPB, so platelets retraction on the heart, causing intermittent periods and fresh-frozen plasma should be available if coagu- of hypotension and low cardiac output. Communica- lopathy after reversal of heparinization is present. An tion between anesthetist and surgeon is critical. The antifibrinolytic is routinely given. We use Tranexamic circulation usually requires support with vasopres- acid (30 mg/kg load and 16 mg/kg infusion until the sor agents. Some degree of fluid expansion may be end of CPB). required to optimize preload, although caution should The advantages of CPB are hemodynamic stability be exercised because low-pressure pulmonary edema and that it allows controlled reperfusion of grafts. in the grafts may occur if excessive fluid has been given. Disadvantages of CPB are hemolysis and activation of proinflammatory cascades that increase the risk of lung injury. It also increases the need for transfusion Reperfusion of graft of blood products due to hemodilution, coagulopathy, This is a busy period which that often requires an extra and platelet dysfunction. Use of CPB has been associ- set of hands to deal with differential lung ventilation

142 Chapter 17: Management during surgery and cardiovascular emergencies at the same time. up to 25% of LTs and increases both short- and long- Before tying the final stitch of the atrial anastomosis, term mortality. It may manifest immediately following the graft is de-aired through an opening in the atrial reperfusion during surgery and can occupy a spectrum anastomosis. This is done by inflating the lung to a sus- of mild self-limiting disease to fulminant respiratory tained pressure of 15–20 cm H2O and partially releas- failure. Other features encountered in the operating ing the PA clamp. The atrial clamp is then released to room are poor respiratory compliance, high PVR, and de-airtheatrialcuffbeforethefinalknotistied.Atthis pulmonary edema. point, hypotension may occur due to several causes. There are sometimes leaks in the vascular anastomosis Nitric oxide that need to be controlled by placing further sutures. Inhaled nitric oxide (iNO) in the range of 10–40 A significant amount of blood loss may rapidly occur parts per million (ppm) may be used in established in the interim, requiring prompt intravascular volume PGD to treat severe hypoxia or elevated PA pres- replacement. Temporary myocardial stunning occurs sures. NO increases cyclic guanosine monophosphate, because the initial venous return to the left atrium is which relaxes smooth muscle. It improves ventilation/ cold and contains ischemic metabolites, and treatment perfusion matching and decreases PA pressure by with a small bolus of epinephrine or calcium is often selectively vasodilating pulmonary vasculature in ven- required to improve myocardial contractility. Coro- tilated portions of the lung. The effect is transient, nary artery air embolism may also occur. This is usu- with rebound pulmonary hypertension seen during ally seen in the right coronary artery as it is uppermost weaning, and survival benefit has not been shown, in the supine position. ST depression or elevation may but its use may help stabilize the patient sufficiently be seen in the inferior leads and, if treated with vaso- during this turbulent period. Prophylactic use of iNO pressor agents, is usually transient. has not been shown to prevent PGD. Disadvantages The PA clamp is slowly released over a 10-minute of iNO include the potential for methemoglobinemia period, limiting initial flows to the vascular bed of the and cytotoxicity from free radical production. NO oxi- graft, which has been shown to reduce primary graft dizes to nitrogen dioxide and other higher oxides that dysfunction. To reduce lung injury, initial ventilation are pulmonary toxic. Therefore, specialized monitor- tothegraftshouldbewithalowinspiredoxygencon- ing and delivery systems need to be used, which con- centration (FiO ), low peak inspiratory pressures of 2 tributes to the very high cost. 15–20 cm H20, and a positive end expiratory pressure of 5 cm H20. The respiratory rate is initially set to 8– 10 breaths per minute. The other lung requires 100% Prostaglandins oxygen,mosteasilygivenviaaself-inflatingbag.Once Inhaled aerosolized prostacyclin (PGI2)isanalter- the PA clamp is fully released, ventilation settings need native to iNO. It is less expensive and does not to be adjusted to ensure adequate minute ventilation require a bulky delivery system. Commercial PGI2 and CO2 removal, and the FiO2 titrated to a safe level “Epoprostenol” is reconstituted in sterile glycine, of oxygen saturation. For the first side of an off-pump diluted with normal saline according to body weight, BLT, the new graft needs to support the ventilatory and aerosolized via a “low-flow” nebulizer that deliv- requirements of the patient while the contralateral side ers 8 ml/hr of solution with a driving flow rate of is operated on. For patients on CPB, ventilation is con- 2 L/min to give a typical dose of 50 ng/kg/min. Rel- tinued on the initial settings, and restriction of venous atively large doses are required due to the inefficiency cannula drainage by partial clamping allows some RV of aerosolized particles reaching the alveoli. A contin- ejection to perfuse the graft. This is titrated to aPA uous nebulizer is required due to the short half-life of pressure of 10–15 mmHg. epoprostenol. A small amount of literature in the use of PGI2 for primary graft dysfunction shows equivalence in lowering PA pressures, improving oxygenation, and Primary graft dysfunction lack of systemic side effects compared with iNO. There Primary graft dysfunction (PGD) is a devastating com- are no studies that show long-term outcome in this set- plication akin to acute lung injury due to the trans- ting. Prostaglandins administered IV are rarely used in plantation process. It has been reported to occur in the acute setting due to their systemic effects causing

143 Section 3: Lung

hypotension and non-selective pulmonary vasodi- Lee JC, Jason D, Christie JD. Primary graft dysfunction. latation that may worsen intrapulmonary shunt and Proc Am Thorac Soc 2009; 6: 39–46. oxygenation. McIlroy DR, Pilcher DV, Snell GI. Does anaesthetic management affect early outcomes after lung transplant? Further reading An exploratory analysis. Br J Anaesth 2009; 102: 506– Christie JD, Van Raemdonck D, de Perrot M, et al.Report 14. of the ISHLT Working Group on Primary Lung Graft Subramaniam K, Yared JP. Management of pulmonary Dysfunction Parts I to VI. JHeartLungTransplant2005; hypertension in the operating room. Semin Cardiothorac 24: 1451–500. Vasc Anesth 2007; 11: 119–36.

144 Section 3 Lung Chapter Postoperative care and early complications

18 Vlad Vinarsky and Leo C. Ginns

Key points plantation (BLT) over HLT, especially with the recog- r nition that LT alone can be performed even in the The major challenge of postoperative fluid setting of severe right heart failure. The postoperative management following lung transplantation management of HLT recipients is similar to that of SLT is to minimize pulmonary edema while and BLT patients. Furthermore, the level of immuno- maintaining optimal cardiac function. r suppressionandthemajorityoftheearlypostoperative Immunosuppression protocols are usually complications, including acute rejection and infection, divided into three broad categories: are related to the lung allograft rather than the car- induction, maintenance, and treatment of diac allograft. For these reasons, this chapter focuses rejection. r on postoperative management and early complications Primary graft dysfunction, the major cause of LT. for morbidity and mortality in the early postoperative period, is diagnosed when acute and hyperacute rejection, infection, Postoperative management venous anastomosis complication, and After transplantation surgery, patients are transferred cardiogenic pulmonary edema have been to the intensive care unit, usually with a single- excluded. lumen endotracheal tube in situ. Universal precau- r Lung transplant recipients’ increased tions, including hand washing, masks, and gloves, are susceptibility to infections frequently indicated. An isolation room with HEPA-filtration of requires prompt treatment with ambient air may limit the incidence of fungal infec- anti-microbials and lowering of the net level tions from extrinsic sources in transplant patients. of immunosuppression. Ventilation Over the last several decades, lung transplantation Ventilator management in most cases follows standard (LT) has become a viable option for a number of postoperative practice. The FiO2 is adjusted to main- end-stage lung diseases. Heart–lung transplantation tain a PaO2 greater than 65 mmHg. Significant baro- (HLT) remains one alternative for select patients with trauma due to increased airway pressure or excessive end-stage cardiopulmonary disease, although its use tidal volume is uncommon after LT, and higher air- is infrequent. In the annual report on heart and lung way pressures may have a beneficial effect in mini- transplantation in 2009, the International Society for mizing postoperative pulmonary edema. Weaning the Heart and Lung Transplantation (ISHLT) received patients from the ventilator requires appropriate man- reports of only 75 HLT recipients compared with 2708 agement of postoperative pain. A functional thoracic LTs performed in 2007. Similarly, only 29 HLTs were epidural catheter is useful (in the absence of coag- performed in the United States in 2009, in contrast ulopathy), although intravenous (IV) opiates via a to 1661 lung transplants performed during the same patient-controlled analgesia approach may be useful period. There has been a growing preference for single- after the patient achieves a conscious state, followed lung transplantation (SLT) and bilateral-lung trans- by transition to oral medications. Anti-inflammatory

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

145 Section 3: Lung

pain medications should be avoided because of added lation, as seen upon awakening from anesthesia or nephrotoxicity with the peri-operative antimicrobial weaning from a ventilator, the RV generates a peak sys- and immunosuppressive medications. Tracheal extu- tolic pressure similar to that required pre-operatively. bation is carried out when the patient has a normal The resultant abrupt rise in PA pressure, in combi- mental status and has achieved a reasonable rate of nation with increased vascular permeability and the ventilation and spontaneous tidal volume. Early tra- absence of lymphatic continuity, can cause rapid fluid cheal extubation in the operating room may be pos- accumulation in the donor lung and lead to labile oxy- sible, especially following SLT in chronic obstructive genation and hemodynamics following transplanta- pulmonary disease (COPD) patients, depending on tion. Preemptive treatment for this condition involves institutional experience and expertise. Good bron- maintaining a high degree of sedation, or even mus- chopulmonary hygiene with frequent endotracheal cular paralysis, in the first 3 to 5 days postoperatively. aspiration of secretions and physiotherapy is crucial This approach minimizes catecholamine stimulation in achieving and maintaining extubation in transplant during the time in which ischemia and reperfusion recipients. injury in the allograft is resolving. Minimizing other Patients with emphysema who undergo SLT provocations to RV output, such as hypercarbia or require special attention to airway pressures and the hypoxia, within the first few days of following trans- compliance difference between the allograft and the plantation is also beneficial. Following this period, native lung. Hyperinflation of the native lung will patients are cautiously awakened and standard ventila- progressively interfere with ventilation of the allograft tor weaning can proceed while closely monitoring car- and can result in compromise of cardiac filling. Efforts diac output, blood gases, and pulmonary artery (PA) to control hyperinflation of the native lung include pressures. In some cases, additional measures such as lower levels of positive end expiratory pressure (PEEP; low dose beta-blockade to diminish myocardial con- 1–3 cmH2O). Positioning of the patient with the tractility or IV anxiolytic therapy may be required in native lung down may further increase impedance of the presence of significant hemodynamic abnormali- that hemithorax and limit hyperinflation, although ties. Other strategies to control pulmonary edema and increased blood flow to the native lung induced by resultant hypoxia in the setting of pulmonary hyper- this maneuver may require adjustment of ventila- tension include fluid restriction and judicious use of tory parameters in order to maintain satisfactory higher PEEP. Inhaled nitric oxide (iNO) as well as gas exchange. In rare circumstances, when marked intravenous or inhaled prostacyclin can be employed edema has occurred in the allograft, independent lung in select cases. ventilation using a double-lumen endotracheal tube may be needed. In such cases, more rapid ventilation of the allograft with smaller tidal volumes, using PEEP Fluid management and hemodynamics and higher inspiratory flow rates, will provide carbon Regarding postoperative fluid management, there is dioxide clearance, whereas ventilation of the native often a tension between the need to maintain suffi- lung at very low rates, with standard volumes and no cient left atrial filling pressure for adequate cardiac PEEP, will maintain oxygenation but minimize hyper- output and the desire to minimize pulmonary edema. inflation. Early tracheal extubation after SLT in COPD As noted above, the effects of ischemia–reperfusion may decrease the risk of native lung hyperinflation. injury and lymphatic discontinuity all contribute to a tendency toward pulmonary edema in the lung graft. PA pressures and pulmonary wedge pressures need Pulmonary hypertension to be kept as low as possible postoperatively without Patients with severe pulmonary hypertension who compromising ventricular preload and cardiac output. undergo LT are managed differently because they have Achieving this goal requires continuous hemodynamic a unique postoperative physiology. The right ventricle monitoring and accurate intake-output measurements (RV) in these patients has been conditioned to gener- during this early postoperative period in order to ate peak systolic pressures against a markedly elevated make the necessary adjustments in rates of fluid PVR. Following LT, an abrupt near-normalization of administration, inotrope, pressors, and diuretic ther- the PVR occurs, accompanied by improved ventricular apy. Most patients have an increase in total body water hemodynamics. With minimal catecholamine stimu- immediately postoperatively but may be relatively

146 Chapter 18: Postoperative care and early complications intravascularly depleted. Although the optimal fluid The addition of inhaled antimicrobial therapy, either for volume replacement remains unknown, it is com- tobramycin or colistin, can have additive effects in the mon practice to use colloid over crystalloid and to management of Pseudomonas in these patients. Post- transfuse with red blood cells to a relatively higher operative antimicrobial coverage should be modified hemoglobingoal(upto10mg/dl)inLTrecipients. if pathogens are identified in the sputum of the donor For most patients, the effects of a reduction in pul- that are not already covered by the recipient-specific monary vascular resistance (PVR) almost immediately regimen. Duration of treatment is dictated, in part, following LT results in improved RV and secondarily by the clinical appearance of the bronchial anasto- left ventricle (LV) performance. Some inotropic sup- mosis and by the microbiology recovered from donor port may be warranted, however, in patients who have cultures or serial bronchoscopy examinations of the preexisting RV hypertrophy, particularly when pres- allograft. sure overload of the RV occurs during the implanta- Routine prophylaxis for fungal organisms is use- tion procedure or following CPB. These cases require ful when preoperative recipient sputum cultures have inotropic stimulation with a selective phosphodi- demonstrated the presence of Aspergillus at any time esterase III inhibitor milrinone, a pure beta-receptor preceding the transplant procedure, when there has agonist such as dobutamine, or a balanced alpha and been evidence of heavy overgrowth of yeast (e.g., Can- beta agent such as dopamine or norepinephrine in dida) in the donor sputum culture or when cytolytic order to optimize cardiac output in the face of a induction immunosuppression is used. Some centers reduced preload relative to the preoperative state. continue to use fungal prophylaxis for all recipients Preoperative renal dysfunction may further com- after LT. Aerosolized amphotericin B, fluconazole, and plicate fluid management postoperatively. A low car- itraconazole have the most published experience for diac output state, prolonged central venous hyperten- prophylactic use. If the recipient has severe structural sion, and the chronic use of nephrotoxic antibiotics lung disease such as bronchiectasis and Aspergillus or or diuretics prior to transplantation may have adverse heavy growth of Candida is identified in the sputum effect on glomerular filtration rate. The liberal use of preoperatively, the patient is treated with IV lipid for- diuretic therapy is helpful in the postoperative man- mulation of amphotericin B followed by oral voricona- agement of these patients. After stabilization and tra- zole for an extended period of time, even in the absence cheal extubation of the LT recipient, ongoing loss of of evidence of invasive or disseminated fungal infec- lean body mass occurs for patients in a postopera- tion. Duration of therapy is guided by surveillance tive stress state. At this point, daily recording of body bronchoscopy results. weight is useful in assessing fluid balance. Weight loss Pneumocystis jiroveci (carinii) pneumonia in LT of up to 10% of pre-operative lean body mass is typi- patients has been virtually eliminated by the routine cal in the first 4 weeks following LT. Continued use of use of trimethoprim-sulfamethoxazole beginning 1 diuretic therapy is usually required during this period week postoperatively. Alternative treatments recom- in order to maintain a non-edematous state. mended for patient with sulfa allergy include ato- vaquone, dapsone, and aerosolized pentamidine. Sulfa desensitization is frequently possible in such patients Anti-microbial therapy and is often recommended given the efficacy of sulfa Prophylaxis against Gram-positive organisms in com- prophylaxis against P. jirov ec i anditsbroadspectrum bination with broad-spectrum antibiotics to provide against a range of additional organisms (e.g., Toxo- appropriate coverage for organisms identified preop- plasma gondii, Listeria spp, Nocardia spp, Streptococcus eratively from the sputum of the recipient is required. pneumoniae). Recipients, who have been recently hospitalized and The incidence of herpes simplex infection, includ- exposed to drug-resistant bacterial pathogens or those ing mucosal ulceration and pneumonitis, has been with cystic fibrosis (CF) will require additional cover- eliminated by the routine use of acyclovir prophylaxis age for Pseudomonas.ForCFpatients,ongoingsurveil- after lung transplantation. Cytomegalovirus (CMV) lance of sputum flora and determination of antibiotic infection, however, remains an important problem fol- sensitivities is important in the period prior to trans- lowing lung transplantation. The incidence of CMV plantation in order to develop an appropriate multi- infection following lung transplantation is related to drug anti-microbial regimen for peri-operative use. the preoperative CMV status of both the donor (D) and

147 Section 3: Lung

the recipient (R). A discordant CMV status between the use of promotility agents and a cathartic bowel the donor and the recipient may result in either pri- regimen. mary infection of the donor lungs by the recipient (in the case of D–/R+) or in the more serious cir- Immunosuppression cumstance of primary systemic CMV infection (in the The induction of a state of relatively non-specific case of D+/R–). In either case, without CMV pro- immunosuppression by pharmacological means has phylaxis, the incidence of acute and chronic rejec- been the key to successful LT. Specific, permanent tol- tion and of mortality is higher in mismatched patients erance of the allograft without the need for chronic than among patients in whom CMV status is concor- medicationswouldbeideal,butthisispresentlynot dant. The use of ganciclovir and valganciclovir prophy- possible. As a result, although the current regimens laxis decreases the incidence of primary disease and lead to satisfactory control of most acute rejection improves the outcome of CMV-disparate lung trans- processes, the combined side-effects of these medica- plants. Prophylactic ganciclovir therapy should be tions and their incomplete ability to control chronic resumed whenever immunosuppression is augmented rejection in the lung account for the major long-term to treat episodes of acute rejection in CMV-discordant morbidity and mortality associated with LT. Immuno- patients. In some centers, CMV-negative recipients suppression regimens depend on experience and pref- who receive lung from CMV-positive donors also are erence of each institution, and protocols vary widely treated with CMV hyperimmune globulin, given their among LT programs. In general, the protocols are high risk of developing severe disease. divided into three broad categories: induction, main- tenance, and treatment of rejection. Nutrition Most patients receive induction immunosuppres- sion despite the lack of evidence for this treatment Maintaining optimal nutrition in the postoperative post-LT in clinical trials. The percentage of patients period is essential and may improve operative out- who received induction has increased steadily from comes. Most patients can resume oral intake follow- 24% in 1997 to 62% in 2008. Although polyclonal ing tracheal extubation, although dietary consultation anti-thymocyte globulin (ATG) or anti-lymphocyte should be obtained to assess the adequacy of caloric globulin (ALG) dominated cytolytic therapy in 1997, intake. Energy and protein needs specifically after LT over the last decade, the interleukin-2 receptor (IL-2R) have not been quantified; however, energy require- agents (basiliximab, daclizumab), CD52 agents (alem- ments in postsurgical patients and in other transplant tuzumab), and muromonab-CD3 (OKT3) are being patients are estimated at 1.35–1.75 times basal energy used more frequently as induction immunosuppres- expenditure, and protein requirements are predicted sion. to be 1.3–2.5 g of protein per kilogram of body weight. No standard protocol exists for the timing of Frequently, because of poor appetite and slow recov- initiation or the regimen of maintenance immuno- ery, the addition of liquid nutritional supplements will suppression therapy following LT. However, virtually be needed to meet these caloric requirements. If the all centers use a three-drug regimen for mainte- patient is unable to resume an oral diet due to pro- nance that includes an anti-proliferative agent, an anti- longed ventilatory support or any postoperative com- metabolite, and a steroid. Historically, this regimen plication that prevents eating, enteral alimentation via consisted of cyclosporine (CyA), azathioprine (AZA), a nasogastric feeding tube is preferred, although IV and a low-dose prednisone. Recently, tacrolimus total parenteral nutrition may be required. Specialty (TAC) has replaced CyA and mycophenolate mofetil diets may be recommended if hyperglycemia, hyper- (MMF) has replaced AZA in most regimens. Close tension, or hyperkalemia develops as a result of side drug level monitoring is necessary in the early period effects of immunosuppressive medications, including after transplantation and when any possible drug inter- high-dose steroids and calcineurin inhibitors (CNIs). action is likely. Patients with CF often have a malabsorptive syn- drome that will require resumption of preopera- tive pancreatic enzyme supplementation. CF patients Early complications of LT are also prone to both gastroparesis and meconium Early complications of LT can be classified into four ileus equivalent, both of which may be avoided by broad categories: complications of the surgery itself,

148 Chapter 18: Postoperative care and early complications re-implantation response and primary graft dysfunc- Reimplantation response and PGD tion (PGD), immunologic complications including The re-implantation response has been defined rejection, and organ-specific complications of the broadly as a triad of worsening gas exchange, immunosuppressive agents used to prevent rejection, decreased lung compliance, and alveolar and inter- including direct side effects of the medication and stitial infiltrates, typically most extensive in perihilar infection. regions. This syndrome develops in the immedi- ate post-transplant period, usually within the first 72 hours. In its mildest form, it remains extremely Complications of the operation itself common, occurring in the majority of patients to some degree. The most severe form of re-implantation In the first 2 weeks following LT, the greatest cause of response represents PGD. The syndrome of PGD is the morbidity and mortality are complications due to the leading cause of death early after lung transplantation, transplant operation itself. These early complications accounting for nearly 30% of mortality in the first include those related to the LT operation specifically, 30 days. PGD is characterized by diffuse alveolar such as airway anastomotic complications, and those damage on pathology, but the pathogenesis of the complications related to thoracic surgical operations disease and the risk factors that determine the degree generally. The latter include pain, phrenic nerve dys- of lung injury remain largely unknown. Importantly, function, retained bronchial secretions and mucous PGD remains a diagnosis of exclusion, and one plugging, atelectasis, pneumonia, persistent air leaks, must consider other diagnoses that could lead to the tension and non-tension pneumothoraces, hemor- same clinical presentation. The differential diagnosis rhage, respiratory failure, and cardiac and hemody- includes hyperacute and acute rejection, infection, namic complications. Therapies of certain of these gen- venous anastomosis complication, and cardiogenic eral thoracic surgical complications must be modified pulmonary edema. for the LT recipient. For example, concern about heal- Treatment of re-implantation response and PGD ing of the airway anastomosis limits endotracheal suc- is supportive with mechanical ventilation, diuretics, tioning to the airways proximal to the anastomosis. and fluid restriction as tolerated. The typical course Retained bronchial secretions and mucous plugging of mild re-implantation response is marked by pro- distal to the anastomosis are treated instead with fre- gressive clearing of the infiltrates to complete res- quent therapeutic flexible bronchoscopies in the early olution within several days to several weeks. How- postoperative period. ever, patients with PGD, marked by severe hypoxemia Major technical complications following LT have refractory to conservative management, require pro- become increasingly rare with improvements in longed mechanical ventilation, and they are treated operative technique and peri-operative management. like patients with acute respiratory distress syndrome PA obstruction can occur as a result of anastomotic (ARDS). General principles of supportive care in these stenosis, kinking, or extrinsic compression. Persistent patients include protective lung ventilation strategies pulmonary hypertension and unexplained hypoxemia with lower tidal volumes to target lower plateau pres- may be evident in these cases. Left atrial anastomotic sures, and fluid restriction with tolerance of some obstruction can also occur due to faulty anastomotic degree of renal dysfunction. Development of ARDS technique or due to extrinsic compression by clot unilaterally in an SLT recipient can make the ven- or pericardium. This problem results in more severe tilatory requirements of the transplanted lung differ abnormalities than PA obstruction, including marked dramatically from those of the remaining native lung. pulmonary hypertension and ipsilateral pulmonary When there is marked compliance difference between edema. Diagnostic methods for theses vascular the native lung and the lung graft, independent lung anastomotic complications include routine intra- ventilation using double-lumen endotracheal tube, as operative measurements of anastomotic gradients mentioned previously, should be considered. and transesophageal echocardiography, which is iNO has been used clinically to treat cases particularly helpful in assessing the left atrial anasto- of established PGD with refractory hypoxemia or mosis. Postoperatively, diagnostic measures include markedly elevated PA pressure. Extracorporeal mem- contrast angiography and ventilation/perfusion brane oxygenation (ECMO) can be instituted using scanning.

149 Section 3: Lung

arteriovenous (VA) routes in order to provide opti- Pleural effusions are common after LT, particu- mal gas exchange and allow decompression of the pul- larly when a large size discrepancy exists between monary circulation. Alternatively, venovenous (VV) donor lungs and the thorax. Continued chest tube ECMO avoids the complete diversion of pulmonary drainage following the primary procedure is not indi- arterial flow from the anastomosis and the graft, which cated as a preventive measure for these effusions may be critical in the setting of disrupted bronchial and may actually lead to secondary infection and artery circulation. The ECMO approach is guided by empyema. Management of these effusions is conser- the center’s experience, and treatment success of PGD vative, with diuretic therapy and dietary salt restric- with both VA and VV approaches have been reported. tion. Invasive measures, such as thoracentesis or tube Despite all of these maneuvers, PGD remains the most drainage, are indicated only for effusions complicated common serious early complication of LT with a high by a delayed pneumothorax, enlarging effusions, or rate of mortality. for large effusions that persist for more than 4 weeks postoperatively. Empyema is uncommon after LT. It is seen most Pleural complications frequently when a persistent pleural space due to a pro- Pleural space complications are quite common after longed air leak, a chronic effusion, or an incompletely LT. Pneumothorax may occur on either side of a expanded lung is secondarily infected, usually with lung graft or on the side of a native lung. Pneu- sputum flora. Management is initially conservative, mothoraces that arise from the lung graft are of with tube drainage and antibiotic therapy. However, greatest concern because of the possibility of airway when the etiologic agent has a limited antimicrobial dehiscence communicating with the pleural space. sensitivity (Burkholderia cepacia or Aspergillus), more Flexible bronchoscopy is indicated for diagnostic pur- aggressive measures such as transposition of muscle poses in patients presenting with this rare complica- flaps or limited thoracoplasty to obliterate the space tion. In most cases, placement of a chest tube with re- may be indicated. Chylothorax has been reported after expansion of the allograft will limit the process. More LT as well. commonly, pneumothorax is the result of rupture of a bullous lesion in an emphysematous native lung after SLT. Conservative management with intercostal tube Early airway complications drainage is indicated. Occasionally, pneumothoraces The door to successful LT was opened by the demon- will be noted after BLT when a significant size discrep- stration that wrapping the airway anastomosis with ancy exists between the donor lungs and the recipient a protective flap of well-vascularized tissues, such as thorax. In these cases, the space will resolve sponta- intestinal omentum, can provide an adequate blood neously over a short period of time, and specific inter- supply for healing. However, interruption of the nor- ventions are not required. mal bronchial blood supply to the anastomotic site still Hemothorax caused by traumatic lesions during makes it the most vulnerable site for airway compli- surgery is a potential complication of all thoracic cations. There is surgeon and center variation about surgical procedures. The underlying pulmonary dis- the anastomosis technique. Omentopexy and other tis- ease of the LT recipient frequently leads to the for- sue wraps are employed less frequently, whereas the mation of dense pleural adhesions and hypertrophy use of telescoping anastomosis and end-to-end anas- of the bronchial circulation. These changes may pro- tomosis with bronchial artery revascularization has mote hemorrhage following the removal of the recip- grown. The surgical technique influences the type of ient’s native lung(s). Any degree of hemorrhage that possible airway complications. For example, unlike an occurs can be exacerbated by the anticoagulation nec- end-to-end anastomosis, telescoping anastomosis that essary for cardiopulmonary bypass, which is required involves intussusception of the donor bronchus into during some transplant operations. The incidence of that of recipient may decrease the incidence of early major hemothorax may be reduced by the use of airway dehiscence but may increase the incidence of a transverse, or “clamshell,” incision rather than a later airway stenosis. median sternotomy in BLT recipients, which allows Early airway anastomotic complications include better visualization of the pleural surface and posterior necrosis, dehiscence, and anastomotic infection. mediastinum. Necrosis varies from mild and asymptomatic to severe

150 Chapter 18: Postoperative care and early complications and life-threatening. Severe necrosis and bronchial ies consistently reporting an incidence of 20–30%. dehiscence present with difficulty weaning from Approximately one fifth of VTE events occur in the the ventilator, pneumomediastinum, subcutaneous first 30 days after transplantation. Many of the clots emphysema, pneumothorax, and persistent air leak. are associated with IV central catheters in the upper Anastomotic necrosis is diagnosed by direct visu- extremities. Other risk factors for VTE after LT include alization with bronchoscopy, although computed older age and diabetes. tomography scanning can often identify dehiscence on the basis of bronchial wall defects and irregulari- ties. Relative devascularization of the anastomosis also Gastrointestinal complications makes it particularly more susceptible to infections. The incidence of early and late gastrointestinal Anastomotic granulation tissue can also interfere with (GI) complications in LT recipients substantially the clearance of secretions from the distal lung and exceeds that observed in patients undergoing non- can serve as a nidus for colonization by pathogens. For transplantation thoracic surgery. Delayed gastric these reasons, significant granulation tissue should emptying is extremely common after LT. Gastro- be excised by rigid bronchoscopy or laser excision, paresis may be related to vagus nerve injury during sometimes repeatedly until healing is complete. transplantation and can be exacerbated by medica- tions. This complication can lead to gastroesophageal Cardiovascular and hemodynamic reflux disease, malnutrition, malabsorption of anti- rejection medications, and recurrent aspiration of complications gastric contents into the allograft. Biliary pathology Both peri-operative myocardial infarction and cardiac such as cholelithiasis and choledocholithiasis, as well arrests have occurred in LT recipients, though rela- as pancreatitis are frequent later manifestations of GI tively infrequently. More common cardiac and hemo- complications of LT, often related to direct toxicities dynamic complications are atrial arrhythmias and of immunosuppressive medications. episodes of systemic hypotension. Postoperative atrial GI complications may present atypically. Trans- arrhythmias, including atrial fibrillation and flutter, plant patients may fail to manifest the usual symp- are not unexpected given that atrial clamps are applied toms and signs of intra-abdominal pathological pro- intraoperatively and that some handling of the heart cesses due to their immunosuppressive medications, is inevitable. They occur in up to 40% of patients, and which reduce their ability to mount inflammatory most convert spontaneously or with medical therapy responses. This can result in delays in diagnosis and limited to several weeks, making the need for DC car- lead to poorer outcomes. For this reason, a high index dioversion infrequent. of suspicion and liberal use of diagnostic studies has Systemic hypotension requiring vasopressor and been recommended in evaluating abdominal com- inotropic support is common in the first week fol- plaints post-transplant. Additionally, pretransplant GI lowing transplantation. The hemodynamic profile seen imaging to identify pre-existing abdominal condi- typically includes severely depressed systemic vas- tions can be extremely helpful in post-transplant cular resistance with a normal or diminished car- management. diac index. This complication has been associated A GI complication unique to patients with underly- with re-implantation response. Postoperative hemo- ing CF is the development of a meconium ileus equiv- dynamic instability also has been common in patients alent. In this syndrome, inspissated intestinal con- with underlying pulmonary hypertension. In the first tents in the distal ileum and proximal colon cause week post-transplant, minimal activity, suctioning, or small bowel obstruction, presenting with pain, dis- pain can cause wide swings in the PA pressures in tension, constipation, and bilious vomiting. The syn- these patients, which can result in episodes of sys- drome is likely related to malabsorption due to the temic hypotension or hypertension, as well as oxygen exocrine pancreatic insufficiency commonly occur- desaturation. ring in CF patients. Preventive strategies center around Venous thromboembolism (VTE), including deep the resumption of supplemental pancreatic enzymes venous thrombosis and pulmonary embolism, is a as soon as bowel function permits. Some centers also common diagnosis in all postoperative patients. VTE add N-acetylcysteine to the enteral feedings of these is a frequent complication of LT, with several stud- patients.

151 Section 3: Lung

Renal complications bility of organ donation is established by donors’ brain death, and hence donors are intubated and mechan- Nephrotoxicity remains one of the most common ically ventilated. This exposes donor lungs to aspira- long-term complications of anti-rejection medica- tion, airway colonization, and parenchymal infections. tions. However, the renal toxicity of post-transplant Second, LT recipients are commonly intubated for medications can become apparent within the first hos- longer periods of time after their operation than other pitalization after LT. Calcineurin inhibitor (CNI) tox- solid organ transplant recipients, further increasing icity in LT recipients may be exacerbated by multi- their risk of pneumonia. Third, the lung allografts, like ple other drugs and require CNI dose adjustment, e.g., the bowel, are unique among organ transplants in that furosemide, amphotericin B, ganciclovir, acyclovir, they are exposed continuously to the external envi- trimethoprim-sulfamethoxazole, and the aminoglyco- ronment, and consequently, to potential pathogens. sides. Similar to with other organ transplants, other Finally, two important pulmonary defense mecha- early renal complications after LT requiring treatment nisms, the cough reflex and mucociliary clearance, include hypertension, hypomagnesemia, and hyper- are impaired in transplanted lungs. All of these fac- kalemia. tors increase LT patients’ susceptibility to pulmonary infections and, in fact, the lung is the site of the major- Rejection following LT ity of their post-transplant infections. Hyperacute rejection has been documented following LT in a number of case reports. Acute cellular rejection after LT is characterized by lymphocyte-predominant Bacterial infections inflammatory infiltrate around blood vessels and/or The majority of infections in LT recipients are caused airways. This syndrome of acute rejection is rare in the by bacteria, and the most common bacterial infec- immediate post-transplantation period prior to hospi- tions in these patients are pneumonia. Bacterial pneu- tal discharge, likely because of the relatively high level monias occur most frequently in the early postop- of immunosuppression that normally follows lung erative period, although the incidence has decreased transplantation. Symptoms of acute rejection include with routine antibiotic prophylaxis against any poten- non-productive cough, dyspnea, and fatigue, although tial pathogens isolated during the first 7–10 days. a number of people with rejection can be asymp- Gram-negative bacilli are the most common causative tomatic. Signs of rejection include fever, rales and organisms, although Gram-positive bacteria including wheezes on exam, hypoxia, and a fall in forced expi- methicillin-resistant Staphylococcus aureus (MRSA) ratory volume in 1 second (FEV1). Radiographic find- have become important pathogens as well. Clinically, ings associated with acute rejection are non-specific, these infections present with fever, new or increased including septal line thickening, new pleural effu- radiographic infiltrates, and the presence of a new or sions, and new or increased air-space disease. When newly predominant organism by sputum Gram stain acute rejection is suspected early after transplantation, and culture. The diagnosis frequently is confirmed infection must be excluded, since infection and rejec- by bronchoscopy. Patients are treated with culture- tion can occur simultaneously and can be confused directed antibiotics for 2–4 weeks, often accompanied even histologically. The treatment of allograft rejec- by a modest reduction in immunosuppression. When tion depends on severity and clinical setting. Treat- a limited spectrum of antimicrobial sensitivity is avail- ment protocols also vary according to different centers able, the use of aerosolized antimicrobial therapy in but are similar to those described in other transplanted additional to intravenous therapy may be helpful. organs (Chapter 3). Potential host sources of contamination of the respiratory tract should be evaluated, particularly in patients with CF. Chronic sinus infection is common Early infections amongCFpatientsandmayactasasourceofcon- The rate of infectious complications is higher inLT tamination of the lower respiratory tract. Gastroe- patients than in recipients of other solid organs, sophageal reflux is common in both CF and COPD accounting for more than 20% of mortality in the first patients and can lead to recurrent aspiration pneumo- 30 days following LT. There are multiple factors con- nia in dependent regions of the lung. Elevation of head tributing to this higher infection rate. First, the possi- of bed and administration of pro-motility agents, in

152 Chapter 18: Postoperative care and early complications addition to acid suppression with proton pump and contribute to development of PTLD. Respi- inhibitors, will control the reflux in most cases. After ratory syncytial virus (RSV) has been implicated anSLT,thenativelungmaybecomethesiteofpneu- in a broad spectrum of respiratory illness in lung monia or a lung abscess. Standard therapy is recom- and other solid organ transplant recipients, ranging mended in the early post-transplant setting, although from isolated rhinitis to fatal ARDS. RSV hyper- a focal structural abnormality may require surgical immune globulin and aerosolized ribavirin have removal if it becomes the source of recurrent infec- been used to treat these infections, with successful tion. Transplant patients are susceptible to infections outcomes. with bacterial pathogens in sites other than the lungs. This includes catheter infections, wound infection, and Clostridium difficile colitis. Fungal infections Fungal infection in the LT recipient is relatively rare, especially early after surgery. Candidemia can occur CMV and other viral infections due to critical illness and recent surgery. Candida albi- CMV is a chief opportunistic infection in LT recipi- cans may also cause invasive pneumonia following LT. ents and the second most frequent infection in these The fungal source is usually the donor trachea. patients overall. CMV disease in the lung transplant Aspergillus species are the most common cause recipient can result from either (1) reactivation of a of fungal infections following LT, although they latent CMV infection acquired by the recipient prior occurusuallyseveralmonthsafterthetransplanta- to transplant, or (2) transmission of CMV to the recipi- tion. Aspergillus may present as part of the resident ent following transplant, from either the donor lung or flora of the recipient or it may cause an invasive infec- transfused blood products, resulting in a new primary tion, either an ulcerative tracheobronchitis involving infection. The frequency of CMV disease, therefore, the anastomosis and airways or a disseminated ill- depends on whether the recipient and donor carry ness with pneumonitis and fungemia. Aspergillus air- latent CMV, which is determined by the presence or way colonization is not uncommon after LT, and in absence of antibodies to CMV. Post-transplant infec- mostcasesistransient.InvasiveAspergillus disease, tions are most frequent in recipients positive for CMV in contrast, is less common but frequently fatal. An antibodies, indicative of prior CMV exposure, regard- ulcerative tracheobronchitis due to Aspergillus involv- less of CMV status of donor. CMV infections in these ing the anastomotic site and large airways also has been patients are felt to be due to re-activation of their described and appears to be highly specific to LT and ownlatentinfections,duetotheirimmunosuppressive HLT. Unfortunately, symptoms are often minimal or therapy. However, CMV infections are most severe in absent in these infections initially, which can lead to recipients negative for CMV antibodies who receive a delay in diagnosis. Consequently, fungal infections lungs from CMV-positive donors. CMV infections in are frequently disseminated at the time of recogni- these patients are felt to be due to primary infections tion and are associated with high mortality rates. Some with the virus transmitted from the allograft, which centers routinely use galactomannan antigenemia and has been documented to occur in renal transplant beta-D-glucan assays to screen for invasive fungal recipients. Primary infections have been associated infections such as aspergillosis, although the data with more severe manifestations of disease, and higher on the use of these tests are limited. Beta-D-glucan mortality rates, than re-activation of latent infections. assay, in particular, must be interpreted with caution CMV and other viral infection are extremely rare in the because of false-positive results due to a variety of first few weeks after transplantation, especially with antibiotics. the use of ganciclovir prophylaxis. Although sputum colonization may be treated with Herpes simplex virus (HSV) can cause necrotizing an azole or inhaled amphotericin B, such therapy tracheobronchitis or esophagitis, and less frequently, rarely is able to clear the airways of fungus. In the pneumonitis in the early weeks post-transplant, but absence of symptoms, a specific anatomical abnormal- these infections have become uncommon with the ity or radiographic changes suggesting pneumonitis, widespread use of prophylactic acyclovir or ganci- most patients with sputum colonization can be man- clovir. Epstein-Barr virus (EBV) may cause fever, aged without specific antifungal therapy. When inva- malaise, pharyngitis, and adenopathy in LT patients, sive Aspergillus infection is identified, therapy with

153 Section 3: Lung

treatment dose of voriconazole is required. Further- Bhorade SM, Stern E. Immunosuppression for lung more, during any infection in the LT recipient, includ- transplantation. Proc Am Thor Soc 2009; 6: 47–53. ing fungal infection, lowering the net level of immuno- Christie JD, Edwards LB, Aurora P, et al.TheRegistryof suppression should be considered. the International Society for Heart and Lung Transplantation: Twenty-sixth Official Adult Lung and Heart-Lung Transplantation Report – 2009. JHeartLung Further reading Transplant 2009; 28: 1031–49. Augoustides JG, Watcha SM, Pochettino A, Jobes DR. Early Lee JC, Christie JD. Primary graft dysfunction. Proc Am tracheal extubation in adults undergoing single-lung Thorac Soc 2009; 6: 39–46. transplantation for chronic obstructive pulmonary Paul S, Escareno CE, Clancy K, et al. Gastrointestinal disease: pilot evaluation of perioperative outcome. complications after lung transplantation. JHeartLung Interact Cardiovasc Thorac Surg 2008; 7: 755–8. Transplant 2009; 28: 475–9.

154 Section 3 Lung Chapter Long-term management and outcomes

19 Paul Corris

Key points itoring the status of the lung graft, long-term out- r Although the outcomes of lung comes are also directly dependent on trying to mini- transplantation lag behind those of other mize the side effects of immunosuppression and care- solid organs, clear evidence of a survival ful attention to the function of other organs. LT is benefit exists for this procedure for patients associated with higher rates of acute rejection than with advanced lung disease as a whole. other solid organ transplantation, necessitating higher r The long-term management of patients levels of immunosuppression. Moreover, the lung is following lung transplantation should focus also at greater risk of injury due to direct exposure on the general medical health of the patient to insults including pathogens and inorganic particles. as much as monitoring of the graft, with This injury may increase the immunogenicity culmi- particular emphasis on tight control of nating in allograft rejection. hypertension, diabetes, dyslipidemia, and The nephrotoxicity associated with calcineurin renal function. inhibitors (CNIs) is well recognized, and a balance must be struck in deciding on the target level of drug r Bronchiolitis obliterans syndrome remains between preventing episodes of acute rejection and the major barrier to improved longer term maintaining renal function. Virtually all grafts are sub- outcomes, though progress has been made in ject to a lifelong threat of rejection, and patients will the approach to therapy. A subset of such generally require maintenance triple immunosuppres- patients with evidence of neutrophilic sive therapy comprising CNI, cell-cycle inhibitor, and inflammation respond well to azithromycin. corticosteroid. r Viral, fungal, and bacterial infections remain Patients beyond 6 months after transplantation common after lung transplantation, with the will generally be seen at three monthly intervals. At majority being due to common community each clinic visit, blood tests, chest radiograph, and pathogens rather than opportunists. lung function (forced expiratory volume in 1 second r The incidence of malignancy is increased [FEV ], forced vital capacity [FVC], forced expira- following lung transplantation with skin 1 tory flow 25–75% [FEF25–75%]) are performed. Some cancers and post-transplant transplant centers perform surveillance bronchoscopy lymphoproliferative disorder most common with lavage and transbronchial biopsy to detect indo- lent asymptomatic infection/or rejection up to the first Successful lung transplantation (LT) requires a sub- year. stantial continuous commitment from the patient and lifelong adherence to immunosuppressive medication. Long-term surveillance Patients are required to attend post-transplant clin- ics and undergo frequent testing to monitor graft sta- Lungs tus, drug levels, renal/hepatic function, glucose, and History, clinical examination, sequential lung lipids. Although there remains a strong focus on mon- function, and imaging provide the cornerstone of

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

155 Section 3: Lung

surveillance of the lungs following transplantation. phoproliferative disorder (PTLD) are common and Monitoring lung function has been shown to be par- increase with time following LT. ticularly effective, and many units train recipients to perform home-based daily spirometry and self-report sustained falls in FEV1.Thedevelopmentofnewsigns Maintenance immunosuppression on physical examination such as crackles, squeaks, Tacrolimus (TAC) is now the most commonly used and a pleural effusion mandate further elucidation of calcineurin inhibitor, with mycophenolate mofetil cause. A detailed commentary on the lung-specific (MMF) the most used anti-metabolite initiated post- complications following transplantation follows in LT. Several small single-center studies have suggested thischapter,andthisshouldbeusedtodrivethe a trend toward reduced acute rejection episodes in LT investigation plan and management. Shared care patients treated with MMF and the potential for less protocols with effective communication should be long-term development of malignancy. Nearly all cen- organized in patients who live at a distance from ters continue to use corticosteroid at a low mainte- the transplant center to ensure that local follow- nancedose(≤10 mg per day), with fewer than 5% up includes monitoring of the lung function and of patients weaned from this therapy. There are no imaging. controlled studies of steroid withdrawal in lung trans- plantation. The exact role of mammalian target of rapamycin (mTOR) inhibitors is not fully evaluated following lung transplantation; however, several stud- Patient ies suggest that they can be used to replace CNI in It has been increasingly realized that improved out- patients with renal dysfunction. For further details, see comes depend on careful review of the whole patient Chapter 3. as well as monitoring of the lungs, and that care- Although it is possible to establish a maintenance ful management of general medical issues such as dose of immunosuppressive drugs using target drug detection and control of systemic hypertension is levels, in practice the intensity of immunosuppres- very important. As a consequence, it is important sion for any individual is often determined by clini- that the long-term management and follow-up of cal course. The optimal level of immunosuppression such patients includes an ordered review. Systemic after LT can be perceived as a compromise between an hypertension, renal dysfunction, dyslipidemia, and acceptable number of acute rejections, the number of hyperglycemia are common complications and need infections, and other complications of therapy, includ- careful monitoring and control via optimization of ing blood count and renal function. The two largest treatment. Hypertension is usually managed using biopsy studies have demonstrated that acute rejection angiotensin-converting enzyme (ACE) inhibitors or is most prevalent in the first 6 months. Overall the data calcium channel blockers initially, and it is impor- suggest that with target cyclosporine (CyA) levels of tant to check on renal function and potassium 200–400 ng/ml in the first 6 months after transplanta- levels shortly after commencement of the former. tion, approximately 50% of patients can expect to have Statin therapy is required in the vast majority of at least one episode of acute rejection. patients to control lipids, and new onset diabetes The monitoring of immunosuppression by using may occur in up to 30% of recipients. Glycemic con- infection as a surrogate marker is both intuitively and trol should follow usual guidelines to prevent com- practically more difficult than the monitoring of rejec- plications in the long term and commonly requires tion. Epstein-Barr virus (EBV) polymerase chain reac- insulin. tion (PCR) monitoring of viral load has been proposed Renal dysfunction that progresses with time is as a surrogate marker of immunosuppression and can commonly seen as a consequence of CNI-based be used to reduce the level of immunosuppression if an immunosuppression and mandates regular monitor- increase in the EBV copy load is measured. The major- ing of estimated glomerular filtration rat, proteinuria, ity of transplant physicians, however, simply monitor and drug levels at least at three monthly intervals. the frequency of viral, bacterial, and fungal infections, Finally physical examination of the patient is impor- and as in a patient with deteriorating renal function, tant, with specific emphasis on skin and lymph node a non-rejecter with frequent infections may have their regions, since skin cancers and post-transplant lym- target drug levels reduced.

156 Chapter 19: Long-term management and outcomes

Therapeutic drug monitoring ents for emphysema receiving MMF therapy, and this should be excluded in a patient with regular bacterial CyA infections. Therapeutic drug monitoring using trough levels has Bronchial infections predominate, and the devel- beenusedsincethestartoflungtransplantationpro- opment of a cough productive of sputum positive grams. Studies in LT recipients have shown that C2 for Pseudomonas aeruginosa may be the first sign monitoring (CyA level at 2 hours post-dose) is the of the development of chronic graft dysfunction, best single time point predictor of an abbreviated area predating the development of bronchiolitis obliter- under the concentration curve. Initial C2 levels should ans syndrome (BOS) by many months. Both com- be between 1000–1500 ng/ml; after 6 months, the tar- moncommunityviralandbacterialinfectionsoccur get level can be reduced to below 1000 ng/ml. seasonally, and frequent infection has been associ- ated with the development of BOS. Lower respira- TAC tory infections including pneumonia are also common A recent comprehensive review has suggested that tar- and usually caused by regular community-associated get trough levels should be between 10–25 ng/ml in viruses (e.g., influenza, parainfluenza, respiratory syn- the first 2 weeks, 10–20 ng/ml up to 3 months, and cytial virus, adenovirus) and bacteria (e.g., Strepto- 10–15 ng/ml thereafter. In practice, however, these coccus pneumoniae and Haemophilus influenza)rather levels may be associated with worsening renal func- than opportunist infections despite immunosuppres- tion, and moreover, full randomized trials of TAC have sion. Two notable exceptions to this rule comprise employed lower maintenance ranges. cytomegalovirus (CMV) and Pneumocystis jiroveci. LT recipients are more susceptible to infection with Cell-proliferation inhibitors (sirolimus intracellular bacteria such as Legionella spp and mycobacteria. and everolimus) Pseudomonal infection is particularly troublesome Although these agents are not nephrotoxic in their own in the context of chronic allograft dysfunction due to right, they do amplify the nephrotoxicity associated BOS. The onset of pseudomonal infection in the spu- with CNI therapy. As a rule of thumb, the dose of CyA tum or lavage should prompt the commencement of should be reduced by 50% if these drugs are added, nebulized therapy such as colimycin on a twice-daily with a target trough level of 50–80 ng/ml. Drug levels basis to reduce the colonizing load of bacteria in the of sirolimus and everolimus should be kept between 3– lung, and this may be required on a permanent basis if 8 ng/ml, as there is evidence of a lack of efficacy in lev- colonization occurs. els of less than 3 ng/ml and increased toxicity in levels Treatment of community-acquired lower airway above 8 ng/ml. respiratory tract infections or pneumonia should be as per guidelines for such infections in general; however, Longer term management issues one must avoid the use of macrolide antibiotics with- outareductioninCNIdosageandclosemonitoring post-transplantation of levels since there is a danger of precipitating acute renal failure. In practice, these antibiotics should be Infection (also see Chapter 4C) avoided as first-line agents. Infections are the commonest cause of morbidity and A patient presenting with a diagnosis of pneumo- mortality at any point after LT, and the incidence of nia should be treated urgently with antibiotic therapy infection is higher in LT recipients compared with the to cover both typical and atypical community bacte- recipients of any other solid organ. The combination of ria, with appropriate culturing of blood, sputum, and higher levels of immunosuppression and direct expo- pleural effusion if present. Urine should be sent for sure to the external environment provides a plausible antigenemia. The antibiotics can subsequently be tai- reason. Other predisposing conditions include dener- lored by the microbiological results. A patient who is vation with impaired mucociliary clearance, reduced failing on this approach after 48 hours and in whom cough, and injury to the bronchial mucosa. Studies no positive microbiological diagnosis is forthcoming have also demonstrated that hypogammaglobuline- should undergo bronchoscopy and bronchoalveolar mia may also occur, particularly in transplant recipi- lavage with transbronchial biopsy.

157 Section 3: Lung

Viral infection CNI doses and monitor levels because azole antifun- gal agents interact leading to an increase in CNI levels, Respiratory viral infections can occur at any time which can rapidly become toxic and lead to acute renal post-transplantation and, in common with the non- failure. transplant population, occur seasonally. They can Infection with Pneumocystis jiroveci was previously range from causing a mild coryzal illness to severe significant, but with the introduction of lifelong pro- fulminant infection and are often complicated by sec- phylaxis in the form of co-trimoxazole (Septrin), the ondary bacterial infections. Seasonal flu vaccination is incidence and morbidity has dramatically decreased. recommended for all LT recipients. CMV is the commonest viral pathogen isolated in the postoperative period and remains problematic Acute cellular rejection and post-transplantation despite surveillance monitoring surveillance biopsies and prophylaxis. Presentation with CMV disease can Acute allograft rejection is common, with most be variable with organ-specific features (e.g., lung, GI patients experiencing more than one episode in the tract, CNS, or retina) or a non-specific, insidious pre- first year. Though it is most commonly seen within sentation with symptoms of lethargy and malaise and the first 3 months, later presentation is also seen and low-grade temperature. Leukopenia may be present should alert one to the possibility of poor adherence to on blood tests. Disease typically occurs in the first immunosuppression. Acute rejection can be identified 3 months, but it may be delayed by prophylaxis. Diag- on lung biopsies obtained via transbronchial biopsy nosis is by CMV PCR detecting a significant eleva- (TBBx) at fiberoptic bronchoscopy. Many transplant tion in viral DNA copies or observation of charac- centers perform regular bronchoscopy and TBBx in teristic CMV viral inclusion bodies on tissue biopsy. addition to spirometry in the first year to enable early Treatmentiswithintravenous(IV)ganciclovirora diagnosis and treatment of asymptomatic rejection, 2-week course of valganciclovir (dose-titrated against with the aim of preserving graft function and protect- renal function) depending on the degree of viremia. ing against BOS. The role of surveillance TBBx remains Relapse rates of 60% in primary infection and 20% in controversial, with some centers advocating its useful- previously exposed recipients have been reported. ness and others declaring it an unjustified risk. Acute rejection can be asymptomatic or present Fungal infection with low-grade fever, lethargy, dyspnea, and/or small Candida and Aspergillus account for the majority of pleural effusion on chest radiograph associated with a fungal infections. Colonization is frequent, with a drop in pulmonary function. FEF 25–75% is a more combined incidence of up to 85%, though the preva- sensitive marker of lung function and may fall before lence will vary considerably from center to center and a decrease in other lung function values. Classification depend on local environment and changes in environ- is according to histological criteria found on TBBx and ment. Any perturbation of surrounding soil associ- ranges from none (A0) to severe (A4) (International ated with new buildings, for example, will lead to an Society for Heart and Lung Transplantation [ISHLT]) increase in environmental Aspergillus spores. (Table 19.1). Aspergillus is often asymptomatic and detected on RejectionsgradeA2oraboveareusuallytreated surveillance bronchoscopy and lavage. True invasive with IV methylprednisolone (10 mg/kg) followed by Aspergillus infection is estimated at approximately 5% tapering dose of augmented oral prednisolone (ini- of LT recipients, with only 3% of those colonized pro- tially 1 mg/kg). Recurrent episodes of rejection and gressing to invasive disease. However, Aspergillus colo- lymphocytic bronchiolitis are risk factors for chronic nization has been linked to airway complications such allograft rejection and dysfunction, with acute rejec- as bronchial stenosis. tion being the single most important risk factor for the The spectrum of presentation includes asymp- development of BOS. Treatment options for steroid- tomatic colonization, necrosis, ulceration at the anas- resistant cellular rejection include T-cell ablation with tomosis, invasive pulmonary disease, and systemic anti-thymocyte globulin or similar medications. Total fungal sepsis. Systemic aspergillosis carries a high lymphoid irradiation has been shown to be effective mortality of up to 75%. Voriconazole is the current at controlling the frequency of rejection when drug- treatment of choice, but care must be taken to reduce based immunosuppression has failed.

158 Chapter 19: Long-term management and outcomes

Table 19.1 Pathologic grading of lung rejection (ISHLT) Category Grade Meaning Appearance

A 0 None Normal lung parenchyma acute rejection 1 Minimal Inconspicuous small mononuclear, perivascular infiltrates 2 Mild More frequent, more obvious, perivascular infiltrates. Eosinophils may be present. 3 Moderate Dense perivascular infiltrates, extension into interstitial space. Can involve endotheliosis, eosinophils, and neutrophils. 4 Severe Diffuse perivascular, interstitial and air-space infiltrates with lung injury. Neutrophils may be present. B 0 None No evidence of bronchiolar inflammation airway inflammation 1R Low grade Infrequent, scattered, or single-layer mononuclear cells in (lymphocytic bronchiolitis) 2R High grade bronchiolar submucosa X Ungradeable Larger infiltrates of larger and activated lymphocytes in bronchiolar submucosa. Can involve eosinophils and plasmacytoid cells. No bronchiolar tissue available

Non-cellular antibody-mediated rejection associ- BOS is now the major limiting factor reducing ated with the deposition of complement-activating long-term survival and quality of life after LT and antibody in the lung should also be considered in remains a challenging problem despite improvements a patient failing to respond to steroid therapy and in other aspects of care. BOS affects up to 50–60% requires treatment directed toward B and plasma cells. of patients surviving more than 5 years and is ulti- The development of both donor-specific and non- mately responsible for more than 30% of all deaths HLA antibodies and subsequent development of BOS 3yearspost-transplantation.Thetimingofonsetis remains to be fully elucidated. Plasmapheresis, IV highly variable, ranging from months to years, with immunoglobulin G, and rituximab therapy has been a median time from transplantation to diagnosis of advocated as effective treatment for acute antibody- 16–20 months. It also results in significantly reduced mediated rejection, and patients who develop antibod- health-related quality of life, with more than 80% of ies should probably receive MMF. patients being functionally limited 2 years after diag- nosis. The histological lesion of chronic rejection is oblit- Chronic allograft dysfunction (BOS) erative bronchiolitis (OB), characterized by loss of epithelium, an increase in fibrosis, and eventual occlu- (see also Chapter 4A) sion of the bronchiole by intraluminal granulation tis- BOS is characterized by progressive airflow obstruc- sue. The mechanism is thought to be due to repeated tion and is a clinical diagnosis defined by a persistent, injury and inflammation that causes airway epithelial irreversible drop in pulmonary function (FEV1,FVC, damageandloss,resultinginanexaggeratedhealing FEF25–75%) supported by radiological findings on response and fibro-proliferation involving myofibrob- high resolution CT, such as bronchial dilatation, tree in lasts. BOS can be patchy in distribution and is not bud appearance, and mosaicism with air trapping, best invariably seen on TBBx, despite a typical functional seen in expiratory images. BOS is a diagnosis of exclu- decline. Positive histology is not required to make a sion and can be formulated only when other causes diagnosis (Table 19.2). of drop in FEV1 (i.e., acute rejection, infection, and Recently the importance of detecting early, sub- bronchial anastomotic stricture) have been excluded clinical BOS before irreversible fibroproliferative dis- by bronchoscopy, bronchoalveolar lavage (BAL), and ease has become established has been recognized, and TBBx.Clinicallytheremaybesqueaksfoundonaus- anewstageofBOS0-phasbeenadded.Thisuses cultation. If alternative causes of reduced lung func- FEF 25–75% in addition to FEV1 to diagnose loss of tionarepresent,thentheymustbetreated,andifthe function. In established disease, patients experience reduction in lung function persists, a final diagnosis of repeated symptomatic infection followed by persistent BOS can be made. colonization with organisms such as pseudomonas

159 Section 3: Lung

Table 19.2 Severity criteria for BOS (ISHLT) Table 19.3 Risk factors for development of BOS Stage Spirometry Alloimmune Non-alloimmune

BOS 0 FEV1 Ͼ 90% baseline, FEF25–75 Ͼ 75% baseline Recurrent acute Primary graft Ischemic-reperfusion rejection (A2 or dysfunction injury BOS 0-p FEV1 81–90% baseline, FEF25–75 < or equal to 75% above) baseline Lymphocytic Viral infections CMV pneumonitis BOS 1 FEV1 66–80% baseline bronchiolitis community on BAL respiratory virus BOS 2 FEV1 51–65% baseline HLA mismatching Bacterial infection Pseudomonas BOS 3 FEV1 < or equal to 50% baseline Anti-HLA Gastric reflux Baseline FEV1: average of the two highest FEV1 measurements obtained more than three weeks apart post transplantation with- antibodies Airway ischemia out preceding bronchodilator therapy. Medication non-compliance Older donor age, prolonged graft and aspergillus. It is unclear whether pseudomonas ischemic time infection is a risk factor for development of BOS or simply a representation of damaged airways that are prone to colonization. Gastroesophageal reflux disease The rate of progression of disease can also be vari- Gastroesophageal reflux disease (GERD) is common able, including (1) sudden onset and acute deteriora- after LT (65–70%) due to a combination of medication- tion followed by rapid decline of lung function; (2) ini- induced gastroparesis and damage to the vagal nerve tial acute deterioration in lung function followed by during surgery, causing delayed gastric emptying and a plateau and prolonged period of stability; and (3) altered lower esophageal sphincter function. Post LT, insidious onset with slow, progressive decline in lung many asymptomatic patients are found to have GERD, function. Acute onset occurring early after surgery is as measured by 24-hour pH studies. Denervation of associated with a worse prognosis. There are many risk the lungs reduces the cough reflex, and the likeli- factors both hypothesized and proven for developing hood of silent aspiration is increased. This continuous BOS (Table 19.3). micro-aspiration of gastric contents and bile acids may promote chronic inflammation, airway damage, and Alloimmune-dependent risk factors bacterial infection/colonization, all of which may pre- Alloimmune risk factors include number and sever- dispose to BOS. GERD has been demonstrated to be ity of episodes of acute rejection, lymphocytic bron- a reversible cause of allograft dysfunction, and fundo- chiolitis, and HLA mismatching. Acute rejection is plication has been shown to improve pulmonary func- the most significant risk factor for development of tion and improve survival in patients with graft dys- chronic rejection. In multivariate analysis, three or function and reflux in uncontrolled studies, but this more episodes of A2 grade or above acute rejection requires confirmation. was strongly associated with the development of BOS. Late episodes of acute rejection are also associated with Treatment of BOS subsequent development of BOS. Lymphocytic bron- Azithromycin is already used in the treatment chitis/bronchiolitis has been shown to be a risk factor of bronchiectasis, pan-bronchiolitis, and cystic for BOS independently of acute perivascular rejection. fibrosis (CF) due to its anti-inflammatory and immunomodulatory effects. Studies to date have Non–alloimmune-dependent risk factors been on small numbers but have shown promising CMV pneumonitis is a recognized risk factor for BOS. results using thrice weekly dosing at 250 mg or 500 Other viral and bacterial infections have also been mg once daily, with responders showing significant implicated, as well as gastroesophageal reflux, medica- improvement in FEV1.Responseappearstobe tion non-compliance leading to low immunosuppres- better in patients with higher levels of neutrophils sion levels, and airway ischemia and injury. Primary on BAL and those who started treatment earlier graft dysfunction has also been shown to increase the post-transplantation. Randomized controlled trials to risk. substantiate this effect are ongoing. Two phenotypes

160 Chapter 19: Long-term management and outcomes of allograft dysfunction have been proposed, given Table 19.4 World Health Organization (WHO) classification of the apparent reversibility of a previously designated PTLD irreversible condition: (1) classic BOS, which does Category Subtype not respond to azithromycin and is associated with a Early lesions Reactive plasmacytic hyperplasia BAL neutrophilia less than 15%; and (2) neutrophil- Polymorphic Polyclonal or monoclonal associated reversible airways disease, which is at least Monomorphic Diffuse large B-cell lymphomas reversible in part with azithromycin and associated Burkitt’s/ Burkitt’s-like lymphoma with a BAL neutrophilia greater than 15%. Plasma cell myeloma Other treatments for BOS with uncontrolled T-cell lymphomas Peripheral T-cell lymphoma studies suggesting benefit include cytolytic ther- Others Hodkgin’s disease-like apy, photophoresis, total lymphoid irradiation (TLI), Plasmacytoma-like Myeloma cyclophosphamide, and methotrexate. TLI has been used with benefit in very small numbers of patients and has been shown to significantly slow the rate of possible for histology, including immunohistology and decline of FEV1, with some patients gaining lung func- determination of EBV status. tion. Potentially TLI may have its role in rapidly dete- Survival is variable, with some centers quoting riorating patients not responding to azithromycin. 25–60% survival rates. Factors predicting a poor out- There is uncontrolled evidence suggesting that come are performance status 3 or 4 at the time switching patients to TAC from CyA slows the rate of diagnosis, late presentation following transplanta- of decline in patients with BOS. Increasing mainte- tion, monomorphic disease, and CNS and allograft nance immunosuppression is ineffective and generally involvement. leads to increasing problems with infection. Prevent- Initial treatment for polymorphic EBV-positive ing infection may be an important strategy, but has not disease is reduced immunosuppression, and approxi- been formally examined in a clinical trial. mately 25–50% of patients respond to reduction alone. No clear benefit has been shown for antiviral ther- apy. More than 90% of PTLD after solid organ trans- Post-transplant lymphoproliferative plant is CD20 positive, defining a role for anti-CD20 disorder antibodies (rituximab), which have been associated with complete remission. Response rates with ritux- LT has the highest reported incidence of post- imab are around 60%, which is similar to cyclophos- transplant lymphoproliferative disorder (PTLD; 4– phamide, doxorubicin, vincristine, and prednisone 10%) which is thought to be due to the higher (CHOP)chemotherapygiveninthecontextofmore degree and duration of immunosuppression needed aggressive, CNS disease or those failing to respond to to prevent rejection (further detail in Chapter 4A). rituximab. The incidence of PTLD is highest in the first year, when it is usually polyclonal and responsive to reduced immunosuppression. Later disease is com- monly monomorphic and has a clinical course resem- Nephrotoxicity bling that of diffuse large B-cell lymphoma. Patients CNIs are nephrotoxic and have a deleterious effect on acquiring EBV infection from the donor are at highest kidney function. This is in addition to the insults to risk of developing PTLD. World Health Organization renal function in the peri-operative period, such as classification of PTLD is shown in Table 19.4. hypotension, sepsis, and medications. It is not unusual Diagnosis can be difficult, and a high index of sus- for patients to need a period of ultrafiltration or picion is needed for diagnosis due to the varied clin- hemodialysis in the immediate postoperative period. ical and histological presentation. Symptoms include Ninety-one percent of patients exhibit a decrease malaise, sweats, and weight loss, or may be due to in renal function at 6 months post-transplantation. the site and extent of disease, e.g., lymphadenopa- The presence of hypertension post-transplantation has thy. The commonest extranodal sites involved after been linked to more severe renal dysfunction requiring LT are the allograft (69–89%) and the GI tract (20– aggressive blood pressure control with ACE inhibitors 34%). Excisional biopsies should be obtained wherever or angiotensin receptor blockers. Factors worsening

161 Section 3: Lung

100 Figure 19.1 Adult lung transplantation, survival 1988-1994 (N=4,307) by era (transplants: January 1988 – June 2007). 1995-1999 (N=6,553) From J Heart Lung Transplant 2009; 28: 989–1049. 2000-6/2007 (N=14,626) Reproduced with permission from the International 75 Society for Heart and Lung Transplantation.

50

Survival (%) Survival Survival comparisons by era 1988-84 vs. 1995-99 p = 0.0010 25 1988-84 vs. 2000-6/07 p = 0.0001 1988-89 vs. 2000-6/07 p = 0.0010

0 024681357910 Years

chronic renal failure include hypertension, hyperlipi- at 5 years. Dyslipidemia is virtually universal in demia, and diabetes mellitus. patients other than those with CF and statin therapy is required. Indeed, there is some evidence to support Skin and other cancers the role of statin therapy as an anti-inflammatory agent in LT leading to improved outcomes. The commonest malignancy after LT is skin cancer, with approximately 8% of patients developing this after 5 years and 18% at 10 years. Other non-PTLD solid Venous thromboembolism malignancies are seen in 6% and 14%, respectively. The incidence of venous thromboembolism is 8–29% Patients should be advised to undergo yearly skin after LT. The factors responsible are not well under- checks and always wear sun block outdoors. stood; however, concomitant infection, immobility, and immunosuppression all contribute. It is therefore Osteoporosis important to consider pulmonary embolism as a cause Osteoporosis is an important cause of morbidity post- of breathlessness in a patient presenting acutely with transplantation and has the potential to compromise this symptom post-transplantation. outcome after LT due to poor mobility and pain sec- ondary to fractures. The frequency of osteoporosis is increased after LT due to the use of long-term Outcomes immunosuppressive medication, particularly steroids, LT is now a proven treatment option for carefully occurring on top of already demineralized bone due selected patients with end-stage lung disease, such to chronic disease and prolonged immobility pre- as CF, interstitial lung diseases, pulmonary arterial transplantation. Studies have shown that patients lose hypertension, and emphysema. The procedure is per- approximately 5% of bone density within the first formed to provide survival benefit, alleviate symp- transplant year, with a post-transplantation osteoporo- toms, and to improve quality of life. Though outcomes sis prevalence of 78% and fracture rate of 18%. Patients continue to lag behind those of other solid organs, should have bone scans at 3–5 yearly intervals follow- the current mean actuarial 5-year survival rate is 50% ing transplantation, and osteopenia or osteoporosis according to the ISHLT database (Figure 19.1). Indi- should be treated with bisphosphonates and calcium vidual centers report 5-year survival rates of up to 75%. supplementation. The increased survival rate over the last few years is mainly due to a better operative and peri-operative Drug-induced diabetes mellitus and outcome, thanks to improved surgical techniques, peri-operative anesthetic, and intensive care manage- dyslipidemia (metabolic syndrome) ment. Careful longer term surveillance and manage- The reported incidence of diabetes mellitus after LT ment as described previously has also contributed. is 24.3% at 1 year post-transplantation and 33.5% Chronic allograft dysfunction or chronic rejection,

162 Chapter 19: Long-term management and outcomes

Lung Transplantation for Cystic Fibrosis (n=180) Figure 19.2 Actuarial survival after lung transplantation for cystic fibrosis, Newcastle Freeman Hospital. 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 024681 3 5 7 9 1011 12 Years histologically diagnosed as OB, remains the leading in general have reported significant improvements cause of morbidity and of late mortality after LT, which in all domains of health-related quality of life follow- accounts for up to 30% of all deaths between 3 and ing transplantation, and the majority of long-term 5 years after the transplantation procedure. Although survivors are living independent lives, with many the prevalence of OB/BOS after LT has not changed when it is feasible returning to full-time education or significantly, remaining at 40–50% 5 years after trans- employment. plantation, some patients now respond to therapy, which may explain why patients with this condition live longer. Better understanding and use of immuno- Further reading suppression together with an appreciation of the need Belperio JA, Weigt SS, Fishbein MC, Lynch JP. Chronic to monitor and control general health has improved lung allograft rejection: Mechanisms and therapy. Proc Am Thorac Soc 2009; 6: 108–21. longer term care of recipients. Survival rates do vary according to transplant pro- Christie JD, Edwards LB, Aurora P, et al. The Registry of the International Society for Heart and Lung cedure, era, patient age, and underlying diagnosis, and Transplantation: Twenty-sixth Official Adult Lung and young patients with CF in particular have excellent Heart-Lung Transplantation Report. JHeartLung outcomes (Figure 19.2). Several centers report 10-year Transplant 2009; 28: 989–1049. survival rates in excess of 50%. Pooled data suggest Iverson M, Corris PA. Immunosuppression. Eur Respir better long-term outcomes and an improved health- Mono 2009; 45: 147–68. related quality of life for younger patients with emphy- Kotloff RM, Ahya VN. Medical complications of lung sema undergoing BLT. transplantation. Eur Resp J 2009; 22: 334–42. A survival benefit following transplantation has Yates B, Murphy DM, Forrest IA, et al. Azithromycin been shown for advanced pulmonary disease as a reverses airflow obstruction in established bronchiolitis whole, but there is some debate in the literature over obliterans syndrome. Am J Resp Crit Care Med 2005; patients with advanced COPD. Studies following LT 172: 772–5.

163 Section 3 Lung Chapter Pediatric lung transplantation

20 Stuart C. Sweet and Samuel Goldfarb

Key points dropping from more than 80 in the late 1990s to below r 70 in the early 2000s, the number of pediatric LTs per Pediatric lung transplantation presents many year has risen steadily and exceeded 100 for the first unique challenges compared with adult lung time in 2008 (Figure 20.1). Interestingly, the steady transplant and is a viable therapy for infants, increase likely reflects increases outside the United children, and adolescents with end-stage States, as the number of transplants performed there pulmonary parenchymal and vascular has remained steady in recent years. disease. r Elsewhere in this volume, the details of all aspects Survival after pediatric lung transplantation of LT from an adult perspective, including patient has improved over the past decade; however, selection, surgical techniques, early and late compli- long-term survival rates remain well below cations, and outcomes, have been covered (Section 3, those of heart and other solid organ Chapters 14–19). This chapter addresses aspects of LT transplants. that are unique to infants, children, and adolescents. r Ongoing challenges remain regarding ensuring that patient selection decisions and Indications/contraindications allocation systems maximize survival and/or quality-of-life benefit. The primary diagnoses leading to LT in the pediatric r Increased competition for organs due to agegrouparecysticfibrosis(CF),makingupmore increased numbers of adults undergoing lung than 50% of the population, and pulmonary hyperten- transplantation will require innovative sion,eitheridiopathicorrelatedtocongenitalheart strategies to ensure that children have equal disease. opportunity for transplant. A detailed breakdown of diagnoses for which chil- dren are considered for LT is shown in Figure 20.2, grouped by age at time of transplantation. In children The first human lung transplantation (LT) was per- younger than 1 year, the most frequent diagnoses are formed by Hardy in 1963; however, it was not until those seen primarily in children (pulmonary hyper- the early 1980s that challenges related to rejection tension associated with congenital heart disease, pul- and healing of the airway anastomoses were success- monary vein stenosis, and rarely, alveolar capillary fully overcome. Interest in offering such potential life- dysplasia) or unique to children (pediatric interstitial saving interventions to children was kindled by suc- lung disease syndrome disorders, including diseases cess in adults; the first pediatric LT was performed at of surfactant metabolism such as surfactant protein B the University of Toronto in 1987. Now into a third and C mutations and ABCA3 transporter mutations). decade, LT and heart–lung transplantation (HLT) have CF becomes the most common indication in patients become accepted therapies for end-stage pulmonary 6–11 years of age. In the 12–17 year group, nearly 70% disease in children. As of June 2009, 1400 LT and of pediatric LTs are performed in patients with CF. In more than 500 HLT in pediatric recipients have been themostrecentdecade,therelativepercentageofchil- reported to the Registry for the International Society dren with primary pulmonary hypertension requir- for Heart and Lung Transplantation (ISHLT). After ing LT has diminished significantly, largely because of

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

164 Chapter 20: Pediatric lung transplantation

120 Figure 20.1 Number of pediatric lung and heart–lung transplants performed each year internationally (International Society for Heart and 100 Lung Transplantation [ISHLT]) and in the United States (Organ Procurement and Transplantation 80 Network [OPTN]).

60

40 ISHLT US/OPTN 20 Number of Transplants Number of

0

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Transplant Year

100% Pulmonary Hypertension 90% Congential Heart Disease 80% Interstitial Lung Disease 70% 60% Bronchiolitis Obliterans (non retransplant) 50% Surfactant Protein B 40% Deficiency 30% Percent of Patients Percent Retransplant 20% Other 10% 0% Cystic Fibrosis <1 (N = 83) 1-5 (N = 99) 1-3 yr 3-5 yr 6-11 (N = 136) 12-17 (N = 69) Age Time of Transplant

Figure 20.2 Indications for pediatric lung transplantation. the introduction of effective medical therapies includ- There are several pediatric-specific aspects to the ing prostaglandins (epoprostenol), phosphodiesterase list of relative contraindications. Children, particularly inhibitors (bosentan), and sildenafil. In contrast, in infants, are more likely than adults to require mechan- spite of a steady increase in the median survival for CF, ical ventilation at referral or prior to LT. Although therelativepercentageofCFasthediagnosisleading mechanical ventilation is a significant risk factor for to pediatric lung transplant has not changed apprecia- morbidity and mortality in adults and older children, bly in recent years. the impact on infants is less clear. Therefore, ventilator Contraindications for LT in children generally par- use is generally not considered a contraindication in allel those in adults and are described in detail in infants and older children unless associated with sys- Chapter 14. These include malignancy within the last temic infection. 2 years, sepsis, multi-organ failure, active tuberculo- Because the majority of children referred for sis, acquired immune deficiency syndrome, hepatitis B LT have CF as their underlying diagnosis, organ- or C, severe neuromuscular disease, and documented isms harbored in their airways may raise concerns. refractory non-adherence to a medical regime. In general, colonization with multi- or pan-resistant

165 Section 3: Lung

Table 20.1 Indications for and timing of pediatric lung transplantation Specific disease Timing of referral

Surfactant deficiencies Patients with SPB deficiency and ABCA3 deficiency with refractory respiratory failure should be referred immediately. Patients with SPC deficiency and less severe forms of ABCA3 deficiency may respond to medical therapy and should be referred when unrelenting progression of disease develops. Idiopathic pulmonary hypertension Patients who present in NYHA class III or IV or have evidence of right heart failure should be referred immediately. Patients who fail to respond adequately to vasodilator therapy should also be referred. Eisenmenger syndrome When the trajectory of PH appears to be worsening with impaired exercise tolerance and worsening quality of life Other pulmonary vascular disorders (pulmonary These patients should be referred immediately since they typically do not respond to vein stenosis, alveolar capillary dysplasia) medical management and are at risk for sudden death.

Cystic fibrosis Patients with percent predicted FEV1 values less than 30%, frequent hospitalizations, and refractory hypoxemia or hypercapnia should be referred for transplant. Bronchopulmonary dysplasia Patients with recurrent or severe episodes of respiratory failure or evidence for progressive pulmonary hypertension

Diffuse parenchymal lung disease Patients without evidence for systemic disease that could affect outcome should be referred early.

ABCA3: ATP-binding cassette subfamily A member 3 gene; FEV1: forced expiratory volume in 1 second; NYHA: New York Heart Association Classification; PH: pulmonary hypertension; SPB: pulmonary associated surfactant protein B. Adapted from Faro et al., 2007.

Pseudomonas aeruginosa is not felt to pose increased organs when they are at significant risk of dying with- risk. However, Burkholderia cepacia complex, particu- out an LT. Although in some cases, such as surfac- larly B. cenocepacia andarelatedorganism(B. glad- tant protein B deficiency or alveolar capillary dyspla- ioli), carries significant risk. Interestingly, infection sia, this is a straightforward decision, in other cases, with either one of the two so-called epidemic strains such as surfactant protein C deficiency, the unpre- common in the United States (strain PHDC and the dictable natural history of the disease process makes Midwest clone) are associated with better survival this decision difficult (Table 20.1). Even in the case when compared with infection with non-epidemic of CF, for which a significant body of modeling work strain. In general, B. cenocepacia colonization remains exists, controversy exists regarding whether pediatric an absolute contraindication to LT at a significant per- LT for CF confers a survival benefit. In this setting centage of pediatric centers. of limited predictive data, variable disease course, and Finally, non-adherence can be a particularly chal- unique diagnoses, most pediatric centers use multiple lenging in pediatrics, as often the child’s parents share factors, including diminished waiting list survival esti- significant responsibility for following the medical reg- mates (when available), worsening growth and nutri- imen. In this situation, care must be taken to minimize tion status, increased frequency of hospitalizations, the risk that denying a child the opportunity for trans- and potential for significant improvement in overall plant will be perceived to be a punishment for his or quality of life as components of the decision to accept her parents. In many centers, such psychosocial risks organs. become a significant contraindication only when other medical risk factors are present or the child and fam- ily demonstrate inability to meet a set of agreed upon Allocation expectations for care and follow-up. Decisions regarding the timing of transplantation Once listed, determining the appropriate time to are also influenced by the efficacy of the underly- accept organs becomes the next challenge. Ideally, ing allocation system (see Chapters 40 and 41). In based on the trajectory of illness in spite of optimal the United States, legislative guidance for organ allo- medical therapy, patients are deemed ready to accept cation systems include a directive to “recognize the

166 Chapter 20: Pediatric lung transplantation differences in health and in organ transplantation andtoalesserextentweight,withdonorswithin issues between children and adults throughout the sys- 25% above or below the recipient height considered tem and adopt criteria, polices, and procedures that acceptable. address the unique health care needs of children.” For The main difference in surgical technique relates to this reason, and because end-stage organ dysfunction theincreaseduseofbypass.Becausesomestudieshave can significantly impair growth and development, US suggested that cardiopulmonary bypass (CPB) is an transplant allocation systems give priority to children. independent or contributing factor for primary graft Also, due to steady increases in the number of adults dysfunction (PGD), most adult centers attempt to per- receiving LT, children face increased competition with form lung transplantation without CBP by using dual- adults for organs (this situation may explain the find- lumen endotracheal tubes to allow single-lung venti- ing that the ratio of transplants to waiting list deaths lation (see Chapter 17). Because dual-lumen tubes are in pediatric candidates remains higher than that in not widely available or utilized in pediatric settings, adults). For these reasons, preferential allocation of CPB is used in most pediatric LT. There is some evi- lungs from pediatric donors to pediatric recipients was dence that CBP does not carry additional risk of PGD included in the Lung Allocation System (LAS), intro- in pediatric recipients, based on a single-center com- duced in 2005, which allocates lungs to candidates over parison of the incidence of PGD in a large number of 12 years of age based on a combination of transplant pediatric and adult LT recipients. benefit and medical urgency. The LAS models include The vast majority of pediatric LT recipients receive several indicators of disease severity, including forced two lungs; for those with CF and other suppurative vitalcapacity(FVC),bodymassindex,serumcrea- diseases, the decision is based on the infection risk. tinine, presence of diabetes, 6-minute walk test dis- For the remainder, the decision is generally dictated tance, supplemental oxygen requirement, serum car- by a desire to provide as much healthy tissue as pos- bon dioxide levels, need for assisted ventilation, func- sibletobalancethepotentialthatthelungswillnot tional status, and presence of pulmonary hyperten- grow commensurate with the patient and/or develop sion. More recently, a mechanism to reduce wait- chronic allograft dysfunction. When HLT is a consid- ing list mortality for LT candidates younger than eration, the choice typically hinges on whether irre- 12yearsofagewasimplemented.Thenewsystem versible left ventricular failure is present. In congen- stratifies younger patients based on objective medical ital heart disease, where concomitant repair will be urgency criteria and distributes organs from donors required,LT(asopposedtoHLT)isoftenconsidered under 12 years of age over a much greater distance in order to limit the potential for cardiac allograft before offering them to older children or adults. complications. Taken together, these changes to lung allocation in the Because of the relative scarcity of donors under United States have significantly reduced the duration the age of 12 years, younger pediatric candidates are of waiting time required to receive deceased donor more likely to be considered for technical LT variants lungs and thus increased the frequency of “urgent” than adolescents or adults. Historically there have been referrals, particularly of children and adolescents two approaches used. In the first, initially reported with respiratory failure, some with unknown etiol- by Starnes and colleagues, a lower lobe from each of ogy. Because, apart from infants, LT from mechanical two living donors, serves as one lung for the recipient. ventilation poses increased risk, and the child’s level Living donor lung transplantation (LDLT) has been of illness usually precludes an effective psychosocial used most often in children and adults with CF with evaluation, most centers have been extremely selec- an unexpected and rapid progression of their disease. tive regarding accepting this group of patients for LT Few centers offer LDLT because the procedure requires evaluation. significant resources and the donor lobectomy carries significant risk (see Chapter 15). Another area of technical variation involves reduc- Surgical procedure and complications ing the size of lungs procured from larger (often The LT and HLT procedures performed in pedi- adolescent or adult) deceased donors. Lung size may atric recipients are essentially the same as those used bereducedbylobectomy(mostcommonlyinvolving in adults and described in Chapter 16. Size match- the right middle lobe or lingula), resection of a tissue ing of organs is based primarily on donor height, wedge by a linear stapling device, transplant of a single

167 Section 3: Lung

lobe, or split LT (where two smaller “lungs” are created Transbronchial biopsies (TBBx) are also more chal- from a single deceased donor lung). Case series sug- lenging in pediatrics, particularly in infants and tod- gest that outcomes in recipients of reduced-size trans- dlers. Due to limitations imposed by airway size, endo- plant are comparable to those of recipients of full-sized scopes used for bronchoscopy in young children have grafts in adults. a suction channel 1.2 mm in diameter (compared with Although the smaller airways and vasculature of 2.0 mm or greater for larger scopes). Although smaller children raise the possibility of a higher frequency of forceps that fit into these channels are available, in anastomotic complications, this has not been docu- practice, obtaining sufficient tissue for diagnosis of mented. A systematic review of airway complications rejection is often quite difficult. More importantly, ade- in a large pediatric center found similar frequencies of quate airway tissue is rarely present in such biopsies to complications in pediatric age group as compared with allow assessment of airway inflammation (“B” grade) adults. or BOS grade. For this reason, establishing a histolog- ical diagnosis of chronic lung allograft dysfunction in infants and young children often requires performing Management challenges an open lung biopsy. When sufficient pulmonary func- The post-transplant management of pediatric LT tion information is not available to establish a BOS recipients is guided by the strategies used in adults diagnosis and biopsies are non-diagnostic, many pedi- (see Chapter 18). A consortium of pediatric LT atric centers rely more heavily on imaging modalities programs has developed a consensus approach to such as ventilation/perfusion scanning and inspira- immunosuppression, including tacrolimus, mycophe- tory/expiratory high-resolution CT scanning to assess nolate mofetil, and prednisone. For this consortium, for the presence of airflow obstruction and guide deci- induction remains a center-specific decision with 60– sions regarding obtaining open lung biopsies before 80% of recipients receiving some form of induction, initiation of therapy for chronic allograft dysfunction. usually an interleukin-2 receptor antagonist. Most For children, therapeutic challenges also exist. pediatric centers perform surveillance biopsies. Many newer medications, including immunosuppres- However, several important challenges exist for sant and anti-infective agents, do not have a pedi- pediatric recipients. Perhaps the most important atric license. Most do not include the liquid forms relates to the diagnosis of bronchiolitis obliterans syn- required for young children. Liquid preparations must drome (BOS). Although spirometric measurement of be specially prepared by the pharmacist and may have forced expiratory volume in 1 second (FEV1)and a short shelf life. Thus for patients not residing near forced expiratory flow (FEF) 25–75% are necessary for a pediatric medical center, they may be difficult to the clinical diagnosis of BOS, children under 4 years of readily obtain. Moreover, absorption and pharmacoki- age are generally not able to perform spirometry, and netic data for infants and children may not exist, results are not considered reliable until 6 years of age. making dosing decisions for drugs with narrow ther- Using external compression vests, infant pulmonary apeutic indices more complex. Finally, generic prepa- function testing can provide information regarding the rations for many of the primary immunosuppression presence of airflow obstruction, but requires special- agents have recently become or will soon become avail- ized equipment and experience. Because they require able. The potential variations allowed by the generic anesthesia, however, such tests cannot be performed approval process add another dimension of complex- with the same frequency as conventional spirome- ity to the therapeutic decision making for pediatric LT try. Moreover, BOS criteria do not exist for parame- recipients. ters obtained using infant pulmonary function testing. Finally, infections pose greater risk in children, For older children, because spirometric parameters are particularly infants. Young children may not have proportional to height, the most recent BOS grading completed their primary immunization series by the scheme revision recommends that percent predicted time they undergo transplantation. The efficacy of values, rather than absolute measurements, be used for immunization in the setting of immunosuppression calculating the BOS score. This approach has yet to be is unclear. Although there is some evidence that validated. Thus, for infants and to lesser extent older live viral vaccines may be safely administered after children, diagnosis of chronic allograft dysfunction by transplant, most centers remain reluctant to give live pulmonary function criteria is problematic. viral vaccines. For other infectious agents, particularly

168 Chapter 20: Pediatric lung transplantation

(A) Figure 20.3 (A) Survival after pediatric 100 compared with adult lung transplant. (B) Survival after pediatric lung transplant, P = 0.9423 Adult (N = 27,851) stratified by era of transplant. Figures Pediatric (N = 1,174) modified, with permission, from Aurora 75 et al., 2010.

50 N at risk = 13

Survival (%) Survival N at risk = 9

25 N at risk = 14

HALF-LIFE Adult = 5.2 Years, Pediatric = 4.6 Years 0 0 123456789101112131415 Years Post Transplant

(B) 100 HALF-LIFE Unconditional 1988-1994; 2.8 Years; 1995-2001:4,3 Years; 2002-6/2008: 5.3 Conditional 1988-1994; 7.1 Years; 1995-2001:6,9 Years; 2002-6/2008: na 80 1988-1994 vs. 1995-2001: p = 0.0849 1988-1994 vs. 2002-6/2008: p = <0.0001 1995-2001 vs. 2002-6/2008: p = <0.0261 60 N at risk = 11

40 1988-1994 (N = 208) N at risk = 13 Survival (%) Survival 1995-2001 (N = 457) 2002-6/2008 (N = 519) N at risk = 15 20

0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Years Post Transplant respiratory pathogens, children are often immunolog- cause significant lung injury or mortality. Many cen- ically naive. Thus it is not surprising that respiratory ters treat these viruses aggressively with cidofovir and viralinfections(RVI)havebeenshowntoeffectout- ribavirin, respectively. Lastly, respiratory fungal infec- come in children. In an international, multi-center ret- tionswereidentifiedasariskfactorformortality rospective study of pediatric LT recipients, RVI was an in a retrospective, multi-center study of pediatric LT independent risk factor for 1-year mortality (hazard recipients. ratio = 2.6), with younger age identified as one of the In addition to direct effects of infection, the inci- risk factors for RVI. dence of post-transplant lymphoproliferative disorder Although the impact of cytomegalovirus (CMV) in (PTLD) in pediatric LT recipients is higher than that LT has been significantly reduced by the availability of in adults. Elevated quantitative polymerase chain reac- ganciclovir, children are more likely to be CMV nega- tion (PCR) for Epstein-Barr virus (EBV) has been tive and thus have higher risk for CMV complications. showntobeasensitiveandsomewhatspecificmarker Indeed, CMV was associated with increased mortal- for PTLD; most centers monitor this test on a regu- ity within the first post-transplant year in pediatric LT lar basis. When EBV PCR is elevated, positron emis- recipients. Adenovirus and paramyxoviruses includ- siontomographycanbeasensitiveandspecifictest ing parainfluenza and respiratory syncytial virus can to identify PTLD. Other morbidities associated with

169 Section 3: Lung

100% Figure 20.4 Causes of death after pediatric lung transplant. Data adapted 90% from Aurora et al., 2010.

80% Other 70% Technical 60% Cardiovascular

50% Graft failure Infection 40% Lymphoma 30% Acute rejection Percent of Patients Percent 20% Bronchiolitis obliterans 10%

0% 0-30 days 31 d to 1 yr 1 - 3 yr 3-5 yr > 5 yr (N = 67) (N = 118) (N = 136) (N = 69) (N = 51) Time after Transplant

LT, in particular end-stage renal disease and diabetes, quency of re-transplantation has increased recently, occur with similar frequency to that of adults. perhaps as a result of the new allocation system in the United States. However, outcomes are worse for re- transplant patients in both adults and children, with Outcomes only 37% 5-year survival in the most recent ISHLT In spite of the challenges outlined above, pediatric LT cohort. recipients have survival comparable to that of adults Causes of death after pediatric LT are depicted in (see Chapter 12 and Figure 20.3A), with median sur- Figure 20.4. Graft failure and infection are important vival in the most recent ISHLT registry report of 4.6 causes of death in the first year after transplant. Inthe years. Outcomes for pediatric recipients have demon- first 30 days after transplant, surgical complications are stratedimprovementinthemostrecentcohort(Figure an important cause of death, with a slightly higher rate 20.3B). This era effect is due primarily to reduction in than in adults (13.4% versus 8.2%). After the first 30 early mortality; the era effect disappears when compar- days and before a year post-transplant, infection from ison is conditional on survival of at least 1 year. any cause accounts for roughly 40% of deaths. Infec- One-year mortality risk, in addition to the era tion remains an important cause of death throughout effect noted above, was increased in patients on the the follow-up period, but by 1–3 years after transplant, ventilator at the time of transplant (relative risk BOS becomes the leading cause of death, accounting [RR] = 3.73) and lower in patients receiving bilateral for nearly 40% of the mortality. This is a somewhat lung transplantation (RR = 0.40). One-year mortality higher frequency than in adults, where BOS is respon- risk was also lower in centers performing more than sible for roughly 25% of deaths beyond the first year. five transplants per year. Five-year mortality risk again This difference can be accounted for by the larger per- includes an era effect as well as being on a ventilator centage of deaths in adults from non-PTLD malignan- prior to transplant (RR = 2.69). However, one large cies and “other” causes. single center found no mortality risk associated with Poor long-term outcomes in adolescents may be ventilator use in infants. Younger children had lower one reason why more pediatric LTs are not being per- 5-year mortality than adolescents, typically ascribed to formed, particularly in the United States. Most reports poor adherence in adolescence, but the fact that infants ascribe this to poor adherence in the teenage popu- appear to have a lower incidence of acute rejection and lation, as medical non-adherence appears to be more BOS may also play a role. common in pediatric LT recipients. Yet other mech- Re-transplant accounts for roughly 7% of pedi- anisms, including the impact of hormonal changes atric LT, compared with only 2.5% in adults. The fre- associated with puberty, have not been fully explored.

170 Chapter 20: Pediatric lung transplantation

Nonetheless, adherence is a significant area of focus Growth during the assessment of candidates pretransplant Growth is an important outcome measure unique to and is an important consideration post-transplant. pediatrics, as growth and developmental delays are Most pediatric transplant practitioners have encoun- a significant side effect of end-stage organ failure. tered adolescents who developed graft failure after Although growth rates improve following LT, patients choosing to stop taking their immunosuppressant may not attain normal height due to the suppressive medications. effect of immunosuppressant medications, primarily Further study of adherence in pediatric transplant corticosteroids. This is an additional reason (beyond is needed. Literature regarding objective measures of infection risk) that most transplant programs attempt adherence, identification of non-adherence risk fac- to reduce dosage of immunosuppressant medications, tors, and potential interventions is limited. One area of particularly corticosteroids, as soon as possible after particular interest is care transition. Transplant candi- transplant.Growthhormonehasbeenusedfollowing dates and recipients undergo several transitions dur- transplant, but there are theoretical concerns about the ing their experience during the process of diagnos- risk of triggering rejection and one report of negative ing the underlying disease leading to transplant, with impact. Steroid-free immunosuppression regimens are new care providers often introduced. Another care being explored with some success in kidney, liver, and team is encountered after referral to the transplant heart transplantation, but have not been reported in center, usually followed by relocation to the trans- pediatric LT. plant center to await transplant. After a successful Data regarding growth after pediatric LT is lim- transplant, patients return home a few months later ited. A single-center study found the overall rate of and often need follow-up with new pediatric subspe- somatic growth was roughly 64% of the predicted val- cialty care providers. Outside of the medical arena, ues. Some patients in that study who were doing well adolescence is a prolonged period of transition dur- more than 4 years post-transplant achieved growth ing which increased independence is attained, par- parameters in the normal range. Thus with careful ticularly when attending college or university. This reduction in immunosuppression, adequate growth process places increased responsibility for adherence can occur. Growth of the transplanted lungs is another on the adolescent/young adult patient. Unfortunately, unique outcome measure for pediatric LT. Although maturity and judgment don’t always develop commen- studies using immature animals showed that lung tis- surate with independence. Depression, anxiety, and suegrowthcanoccurafterLT,similarinvasivestudies adjustment disorders may also play a role in this pro- are not possible in humans, and non-invasive assess- cess; all are comorbidities associated with chronic dis- ment of lung growth is difficult. Spirometry volumes ease in the adolescent population. The CF patient reg- (i.e., FEV andFVC)followingLTmaybeinthe istry reports an incidence of these problems of 5–17%, 1 normal range in infants and older children. Diffus- with increasing frequency in older teenagers. Depres- ing capacity of carbon monoxide (DLCO) is a bet- sion rates in general in patients pretransplant can be termeasureoflunggrowth,asitassessessurfacearea as high as 20%. Thus access to mental health ser- for gas exchange. However, DLCO does not appear to vices for treatment of depression is a critical aspect increase in pediatric recipients of cadaveric and living of pretransplant care. Moreover, because transplanta- donor transplants. Unfortunately, DLCO is not read- tion does not guarantee cure for anxiety or adjust- ily measurable in the infant population. Finally, air- ment or depressive disorders, transplant care providers way growth, as measured by serial imaging studies, has must be cognizant of persistence or recurrence of these been demonstrated in a small retrospective study. Thus problems post-transplant. Finally, transition of care to further study is required to clarify whether growth of an “adult” transplant center occurs (usually between lung tissue occurs after pediatric transplant. the ages of 18 and 21 years), often with expectations for minimal parental involvement in the care process. The latter transition may be associated with reduced adherence. Thus research exploring risk factors for Transplant benefit/quality of life ineffective care transition and possible intervention Most transplant physicians and surgeons would agree to better prepare patients for these transitions is that the primary goal of LT is to prolong life. More- needed. over, allocation systems and clinical decision making

171 Section 3: Lung

regarding the timing of transplantation tend to focus atric LT recipients. Long-term side effects, particu- on providing lungs for transplant when death without larly renal dysfunction, also require further attention. transplant is likely. As noted above, recent changes to Understanding how to minimize the effects of trans- the system for allocation of lungs in the United States plantation and immunosuppression on pediatric LT has been modified to incorporate predicted survival recipients is critical. Moreover, identifying the etiolo- benefit in the prioritization process. This change may gies responsible for and providing interventions to be particularly important for pediatrics, as a recent improve poor outcomes in the adolescent population study suggested that children with CF who under- remains an important area for study. went transplantation in the United States under the Removing BOS as the primary barrier to long-term prior system did not achieve a survival benefit. The survival continues to be a priority in LT research. In study findings are contrary to previous findings from pediatrics, as few centers perform enough transplants the United Kingdom, where a different allocation sys- each year to adequately power outcome studies, uni- tem was used and was subject to other methodologic form treatment strategies and multi-center collabora- criticisms. Nonetheless, the study reinforced the con- tions will help to identify strategies for earlier diagno- cept that in decisions regarding benefit of transplan- sis and allow assessment of treatment efficacy. A key tation, quality of life (QoL) must be considered, ide- research opportunity may be exploring the reduced ally with objective measures. Unfortunately, literature incidence of rejection and BOS in the naive but devel- systematically evaluating QoL either before or after oping immune system of infants. pediatric LT are limited. Although the QoL and sur- In spite of these obstacles, nearly 2000 LTs and vival in adult LT recipients was better than candidates HLTs have been performed in children during the whoremainedonthewaitinglist,extrapolationtochil- last three decades, giving these patients a chance for dren and adolescents must be approached cautiously, long-term survival. Although recent improvements in and measures of childhood development must also be outcome for children with CF and pulmonary hyper- included. Although in the most recent ISHLT registry tension will hopefully reduce the number of children report, 80% of children surviving 7 years after LT had needing LT, those who do should continue to benefit. no activity limitation, published research investigat- ing QoL in pediatric LT recipients has largely been Further reading descriptive and contains limitations regarding study Aurora P, Edwards LB, Kucheryavaya AY, et al. The design and methodology. Nonetheless, one study sug- Registry of the International Society for Heart and Lung gests that children may experience psychological dif- Transplantation: thirteenth official pediatric lung and ficulties and an impaired QoL after transplant. Itis heart-lung transplantation report. JHeartLung also worth emphasizing that assessments of QoL may Transplant 2010; 29: 1129–41. be affected by the child’s baseline capabilities prior to Faro A, Mallory GB, Visner GA, et al. American Society of transplant and their expectations after transplant. Sys- Transplantation executive summary on pediatric lung tematic assessment of QoL and developmental impact transplantation. Am J Transplant 2007; 7: 285–92. of transplant on pediatric recipients remains an under- Elizur A, Faro A, Huddleston CB, et al. Lung explored area of pediatric LT. transplantation in infants and toddlers from 1990 to 2004 at St. Louis Children’s Hospital. Am J Transplant 2009; 9: 719–26. Future considerations Wells A, Faro A. Special considerations in pediatric lung Itishopedthatimprovedoutcomeswillleadto transplantation. Semin Resp Crit Care Med 2006; 27: increased focus on growth and development in pedi- 552–60.

172 Section 4 Liver Chapter Recipient selection

21 Alex Gimson

Key points evidence-based definitions of candidate disease sever- ity, transplant outcome, and organ quality. r The selection process in the liver transplant A stringent process of selection of appropriate can- context entails making a life and death didates for liver transplantation is necessary for a num- decision that has the potential of saving the ber of reasons. First, liver grafts are a finite, pre- life of an individual but, inevitably, doing so cious resource as a result of the wide discrepancy at the expense of denying a life-saving organ between the supply of and demand for all deceased to another. organs in general and absence of life-sustaining ther- r The practice of candidate selection and organ apy in patients with end-stage liver disease in partic- allocation is predicated on two fundamental ular. Thus the selection process in the liver transplant ethical principles: justice and utility. context entails making a life and death decision that r The most commonly used scoring system for has the potential of saving the life of an individual but, estimatingprognosisistheModelfor inevitably, doing so at the expense of denying a life- End-Stage Liver Disease, which is derived savingorgantoanother.Second,notallpatientswith from serum creatinine, bilirubin, and end-stage liver disease would necessarily derive a sur- international normalized ratio. vival benefit from the procedure, and those who would r The Milan criteria for selecting patients may not necessarily benefit equally. Third, a thorough with hepatocellular carcinoma for assessment of potential liver transplant candidates is transplantation require the presence of a required to disentangle liver-related extrahepatic man- single mass lesion ≤ 5cmindiameterorup ifestations, which may be considered indications for tofivelesionsthatareall≤ 3 cm, in the transplantation, from extrahepatic disease that would absence of evidence of extrahepatic disease contraindicate the procedure. Examples include the or radiological evidence of vascular invasion. need to exclude fixed chronic obstructive or restric- r Patients with alcoholic liver disease are tive pulmonary disease in patients with suspected hep- usually excluded from transplantation if they atopulmonary syndrome; dementia and other neu- show clinical or histological evidence of rodegenerative and psychiatric disease in patients with active alcoholic hepatitis or evidence of chronic encephalopathy; malignancy of the colon and return to drinking after full professional biliary tree in patients with primary sclerosing cholan- assessment and advice. gitis; surreptitious alcohol intake in patients with end- stage alcoholic liver disease; and underlying malig- nancy in patients with Budd-Chiari syndrome. The ultimate objective of candidate selection for liver In addition to the need to be stringent, the candi- transplantationistoofferittothosewhoaresuffi- date selection process for liver transplantation should ciently sick to justify the risks of the procedure but ideally be evidence-based, transparent, objective, not too sick to benefit from it. Important progress has time-efficient, standardized, and multi-disciplinary, been made in recent years toward devising selection as comprising hepatologists, liver transplant surgeons, well as allocation criteria based on more objective and anesthetists, transplant nurse coordinators, dieticians,

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

173 Section 4: Liver

dentists,socialworkers,and,whereappropriate,spe- of hospital and/or intensive care stay and blood trans- cialists from other disciplines, including psychiatrists. fusion requirements); and morbidity indicators (e.g., Although the principles of candidate selection and incidence of vascular and biliary complications, renal organ allocation are generic, important differences in dysfunction, sepsis, and immunosuppression-related their application exist between different health care malignant disease). systems as a result of differences in the degree of mis- match between the demand for and supply of donor liver organs, legislative framework, availability of alter- Ethical standpoints native forms of organ donation (e.g., live donor and The practice of candidate selection and organ alloca- splits), type of health economy, and prevalent cultural tion is predicated on two fundamental ethical princi- norms. In addition, the selection criteria should be ples: justice (or equity) and utility (or efficiency). The subject to regular review and modification as the rele- principle of justice stipulates that liver organs should vant evidence base continues to evolve. be prioritized according to the degree of an individ- This chapter is confined to deceased organ trans- ual’s need for transplant, thereby ensuring that those plantation in adults and does not discuss live donor with equal needs can have equity of access to this or pediatric selection and allocation. In liver trans- scarce resource. A number of scoring systems have plant practice, a distinction needs to be made between been utilized to predict the survival of patients with the process of selection of appropriate candidates for end-stage liver disease (Table 21.1), a surrogate for transplant, which is the main focus of this chapter, and need for transplantation. A need-based system would that of organ allocation for those candidates who have therefore prioritize patients according to their pro- been placed on the waiting list for the procedure. Both jected survival without transplant, irrespective of their of these processes are underpinned by similar consid- post-transplant outcome. erations with respect to the relevant clinical end points In contrast, the principle of utility mandates that and ethical standpoints, each of which will be briefly candidate selection and organ allocation prioritize discussed next. patients with the greatest survival prospects after undergoing transplant, thereby maximizing the over- all absolute (rather than net) life-saving utility of Clinical end points this finite resource. Thus a purely utilitarian system The outcome of liver transplantation can be defined would select and subsequently rank transplant can- by a number of end points. Although patient sur- didates according to their anticipated post-transplant vival, defined as the length of time between the dates outcome, regardless of their survival prospects in the of transplant and death or last known follow-up, has absence of this procedure. historically been the most important and commonly In liver transplantation practice, incorporating utilized end point, graft survival, defined as the time those two ethical principles is not always straightfor- interval between transplant and graft loss (as a result of ward because patients with the greatest need for trans- death or graft failure requiring re-transplant surgery), plantation (the sickest) are often also those most likely represents another key outcome measure. Another to succumb from it. However, a sensible compromise important end point that incorporates the “intention- can be achieved by utilizing the principle of transplant to-treat” principle is transplant benefit, which is a mea- benefit in both candidate selection and organ alloca- sure of life-years gained, taking account of survival tion decisions, thereby prioritizing organs according to while waiting for transplantation as well as thereafter. the difference in survival prospects with and without Arguably, assessment of transplant benefit to guide transplant or, in other words, the number of life-years decision making in this setting of candidate selec- gained by undergoing transplant. tion and organ allocation should ideally be based on Although the outcome of selection and allocation graft rather than patient survival. since it is the life- decision making based on the principles of medical savingpotentialofonelivergraftthatisusuallyunder need may intuitively be considered similar to that of consideration. Other important but less commonly a benefit-based system, since most of the variability used outcomes include improvement of quality of life in survival benefit is usually determined by differ- or functional status; quality-of-life–adjusted survival, ences in survival without, rather than with, transplant, cost-effectiveness, resource utilization (e.g., duration current evidence suggests that many of the candidates

174 Chapter 21: Recipient selection

Table 21.1 Clinically useful prognostic scores for selection of chronic liver disease candidates for liver transplantation Points assigned

Parameter 1 2 3 Ascites Absent Slight Moderate Bilirubin Ͻ2 mg/dl (Ͻ34.2 micromol/liter) 2–3 mg/dl (34.2 to 51.3 micromol/liter) Ͼ3 mg/dl (Ͼ51.3 micromol/liter) Albumin Ͼ3.5 g/dl (35 g/liter) 2.8–3.5 g/dl (28–35 g/liter) Ͻ2.8 g/dl (Ͻ28 g/liter) Prothrombin time Seconds prolonged Ͻ4 4–6 Ͼ6 INR Ͻ1.7 1.7–2.3 Ͼ2.3 Encephalopathy None Grade 1–2 Grade 3–4 A total score of 5–6 is considered grade A (well-compensated disease); 7–9 is grade B (significant functional compromise); and 10–15 is grade C (decompensated disease). 1. MELD score: MELD = 10 × [0.957 × loge (creatinine in mg/dl) + 0. 378 × loge(bilirubin mg/dl) + 1.120 × loge (INR) + 0.643] 2. UKELD score: UKELD = (5.395 × loge (UKELDINR) +1.485 × loge (creatinine in mmol/L) + 3.130 × loge (bilirubin in micromol/liter) – [81.565 × loge (sodium in mmol/L)] + 435 3. Child-Pugh score: Modified Child-Pugh classification of severity of liver disease

selected for transplantation by need-based systems multi-organ transplantation (most commonly simul- have in fact no demonstrable survival benefit from the taneous liver–kidney grafting). Each of those settings procedure. However, although the development of an will be discussed in the following sections. accurate model for predicting liver transplant benefit remains to be accomplished, indirect implementation of the principle of transplant benefit in the selection Candidate selection in chronic of liver transplant candidates is feasible by excluding liver disease those with survival prospects without transplantation, as judged by currently available need-based prognos- Disease severity scores tic scoring systems (e.g., Model for End-Stage Liver The allocation of donor livers to transplant candidates Disease [MELD], United Kingdom Model for End- has been the subject of considerable public health pol- Stage Liver Disease [UKELD], Child-Pugh score) that icy debate in much of the Western world in general and exceed the observed or predicted 1- or 3-year survival the United States in particular. after transplantation. For many years, the main tool for estimating sur- The role of public preferences in formulating liver vival without transplant has been the modified Child- transplant selection and allocation criteria is also con- Pugh score (Table 21.1), which was originally devel- troversial. On the one hand, it is the public that ulti- oped to predict the survival prospects of patients mately both funds transplantation and provides organ undergoing porto-systemic shunt surgery. In addition, donors, and any transplant selection and allocation disease-specific prognostic models for primary biliary system should therefore command public confidence. cirrhosis, primary sclerosing cholangitis, hepatitis C On the other hand, basing allocation decisions on virus (HCV) cirrhosis, and alcoholic cirrhosis have public preferences can be problematic because those beendevelopedbutareusedlesscommonly. preferences tend to adversely prejudice patients whose However, the use of the Child-Pugh score in the disease is perceived as being self-inflicted and are setting of candidate selection and organ allocation has therefore regarded by most philosophers as unethical. proved contentious for several reasons. First, the cat- Conceptually, the selection of liver transplant can- egorical nature of both the overall score (stratifying didates is best considered in four distinct, albeit patients in three prognostic groups) and the format not mutually exclusive, contexts: chronic liver dis- of some of its components (prothrombin time, albu- ease, variant syndromes (including hepatocellular car- min, bilirubin) effectively ascribes prognostic value to cinoma), acute liver failure, re-transplantation, and what are arbitrary thresholds and assigns patients with

175 Section 4: Liver

potentially diverse prognoses to the same risk group. Table 21.2 Commonly used selection criteria for liver transplantation for candidates with benign end-stage chronic Furthermore, the inclusion in the Child-Pugh score of liver disease two variables that cannot be objectively defined (sever- Ͼ ity of ascites and encephalopathy) has exposed it to MELD score 14 or criticism of subjectivity and vulnerability to manip- UKELD score Ͼ 49 or ulation. Lastly, the Child-Pugh score had never been Child-Pugh score ≥ 7 (i.e., Child’s class B or C) developed or validated in the setting of candidate and selection or liver allocation before its use for this Predicted 5-year post-transplant survival of Ͼ50% (subjectively purpose. judged on the basis of absence of absolute contraindications Increasing dissatisfaction with successive and/or combinations of relative contraindications or markers inequitable liver organ allocation policies, based of adverse post-transplant survival) on the Child-Pugh score, led to the adoption in 2002 of a new evidence-based continuous scoring system States, the evidence suggests that only those with that accurately predicts prognosis among patients with MELD scores greater than 14 derive a survival ben- end-stage liver disease as the basis for the US national efitat3yearsaftertransplantation,andthosewith allocation system of donor livers. This essentially low MELD scores (with a threshold that remains to medical need–based scheme is based on the Model for be determined) should not therefore be listed for the End-Stage Liver Disease (MELD, Table 21.1), which is procedure. amathematicalformuladerivedfromthreeobjective, In the pre-MELD era, minimal listing criteria for readily available, and reproducible laboratory param- liver transplant candidates in the United States for- eters (serum creatinine, bilirubin, and international mally stipulated the presence of Child-Pugh score normalized ratio [INR] of prothrombin time). The of ≥7 which corresponds to a projected 1-year mor- MELD score, which had originally been developed tality rate in the absence of transplantation of ≥ 10%. to predict the mortality of patients with end-stage In the United Kingdom, current listing guide- liver disease undergoing transjugular intrahepatic lines stipulate that chronic liver disease candidates portosystemic shunting and subsequently proved an should only be selected for transplant if their projected accurate predictor of 3-month mortality of wait-listed 1-year mortality in the absence of the procedure, as US liver transplant candidates, can be calculated using estimated by UKELD, exceeds the observed 1-year a hand-held computer and has values ranging between mortality post-transplant among such patients in the 6 and 40, with adjustment made for those receiving United Kingdom. This currently equates to a mortality renal support (serum creatinine set to 4 mg/dl) rate of 9% or more, which corresponds to a UKELD or warfarin anticoagulation (INR capped at 2.5). score Ͼ49. A number of modifications to the MELD score have sincebeenproposed(seebelow),butnoneiscurrently implemented in practice. Variant exceptional diagnoses Similarly, a continuous mortality risk score, the The use of generic mortality scoring models such United Kingdom Model for End-Stage Liver Disease as MELD and UKELD to guide candidate selection (UKELD) score, was developed and validated for end- and/or organ allocation decision making is not appro- stage chronic liver disease candidates awaiting liver priate for all patients with chronic liver disease. Such transplantation in the United Kingdom (Table 21.1). “variant syndromes,” also referred to as exceptional This score is calculated from the serum bilirubin, INR, diagnoses or MELD exceptions, are recognized in the serum creatinine, and serum sodium concentrations United Kingdom and United States selection and allo- and was shown to outperform MELD in predicting cation systems as valid indications for transplant that mortality on the UK liver transplant waiting list. may merit selection and/or additional priority for the procedure beyond that dictated by the UKELD Disease-severity selection criteria and MELD scores, respectively (Table 21.3). Some of those diagnoses are automatically granted such status, (Table 21.2) whereas others currently require formal approval by a Although no formal “minimal listing” criteria for liver peer-review system in the form of a National Appeals transplant candidates currently exist in the United Panel in the United Kingdom or a Regional Review

176 Chapter 21: Recipient selection

Table 21.3 Commonly used criteria for selection for liver transplantation for patients with variant syndromes Variant syndrome Selection criteria

1. Diuretic-resistant International Ascites Club grade 3 ascites (massive abdominal distension) with current imaging ascites documenting the presence and severity of ascites and UKELD/MELD scores are below the required criteria for listing and ≥2 of the following additional criteria: ≥3 therapeutic paracenteses (≥ 2leach)inlast60days. ≥2 episodes of spontaneous bacterial peritonitis with supporting documentation (ascitic polymorphonuclear granulocyte count ≥250 or positive culture) Previous transjugular intrahepatic portosystemic shunt Ascites unresponsive or intolerant to maximum doses of diuretics (i.e., spironolactone 400 mg/day and furosemide 160 mg or equivalent) ≥2 therapeutic thoracenteses Serum sodium ≤125 mEq/l 2. Hepatopulmonary The presence of intrapulmonary shunting as evidenced by echocardiographic appearance of microbubbles syndrome in the left heart ≥ cycles after venous injection of agitated saline and alveolar-arterial oxygen gradient of ≥ 2Kpa(or≥ 2.7 kPa if older than 64 years) in the setting of liver disease and/or portal hypertension. 3. Chronic hepatic Confirmed by EEG or trail-making tests with at least two admissions in 1 year or requirement for encephalopathy endotracheal intubation for airway protection due to exacerbations of encephalopathy that have not been manageable by standard therapy. 4. Persistent and Pruritus consequent on cholestatic liver disease that is intractable after therapeutic trials, which might intractable pruritus include cholestyramine, ursodeoxycholic acid, rifampicin, ondansetron, and naltrexone, and after exclusion of psychiatric comorbidity that might contribute to the itch. 5. Familial amyloidosis Biopsy-proven evidence of amyloid deposition and documentation of the most common TTR gene mutations in the absence of significant/debilitating cardiac involvement (on the basis of ventricular wall thickness, ejection fraction, arrhythmias, and New York Heart Association class), malnutrition or autonomic neuropathy (assessed by the polyneuropathy disability score). 6. Primary Homozygous familial hypercholesterolemia with absent LDL receptor expression and LDL receptor gene hyperlipidemias mutation. 7. Polycystic liver Patients should meet the following criteria: disease (PLD) Satisfy criteria for massive PLD (total cyst:parenchyma ratio Ͼ1) and have a complication of the PLD that is likely to resolve after liver transplantation Are not candidates for, or have disease that has failed to respond to, non-transplant interventions for relief of symptoms; malnutrition may be considered a primary contraindication to non-transplant surgery. Have clinically significant manifestations of liver disease that can be attributed to massive PLD, which may include cachexia, ascites, portal hypertension (variceal bleeding), hepatic venous outflow obstruction, biliary obstruction, cholestasis, or recurrent cyst infection. Have severe malnutrition (assessment made on the basis of hypoalbuminemia or decreased lean body mass). Have serum albumin Ͻ 2.2 mg/dl. Have lean body mass reflected by decreased midarm circumference, measured in the non-dominant arm midway between the acromion and the olecranon process: Ͻ23.1 cm in female patients and Ͻ23.8 cm in male patients. 8. Portopulmonary Defined as the occurrence of a combination of pulmonary hypertension (mean pulmonary artery pressure hypertension Ͼ25 mmHg as estimated by right heart catheter), raised pulmonary vascular resistance (Ͼ240 dynes/sec/cm−5) and normal (≤15 mmHg) pulmonary capillary wedge pressure and/or transpulmonary gradient of Ͼ10–12 mmHg in the setting of evidence of portal hypertension. Patients must have demonstrated a mean pulmonary artery pressure less than 50 mmHg, and those with a mean pulmonary artery pressure of Ͼ35 mmHg who have demonstrated evidence of significant response (mean pulmonary artery pressure Ͻ35 mmHg and PVR of Ͻ400 dynes/sec/cm−5 in the presence of a satisfactory RV function) to a 12-week trial of pulmonary arterial hypertension therapy (iloprost, sildenafil, or bosentan). 9. Uncommon hepatic Examples include carcinoid tumors that are limited to the liver, hepatic epithelioid hemangioendotheliomas tumors (even with extrahepatic spread), and hepatic adenomas in the setting of glycogen-storage disease. 10. Small for size Defined as meeting 4 of the following 6 criteria: Ͼ5 days after living donor liver transplantation; serum syndrome bilirubin concentration ≤10 mg/dl in the absence of rejection or common duct obstruction; bile duct ischemia (leak); INR ≤1.5; significant ascites as noted on clinical exam; biopsy with centrilobular ballooning, necrosis, and cholestasis. (cont.)

177 Section 4: Liver

Table 21.3 (cont.) Variant syndrome Selection criteria

11. Recurrent cholangitis Patients who have ≥2 culture-proven bacteremias within a 6-month period or who have septic complications of bacterial cholangitis and no underlying correctable dominant lesion. Bacteremia should be non-iatrogenic (unrelated to a procedure such as recent endoscopic retrograde cholangiogram or transhepatic cholangiogram) and should occur in a patient who does not have a biliary tube or stent; in addition, these bacterial cholangitic episodes should occur in patients who have been treated with antibiotic therapy that has failed to suppress these septic episodes. 12. Cholangiocarcinoma As part of a research study protocol including the administration of neoadjuvant therapy before transplantation, and operative staging to exclude patients with regional hepatic lymph node metastases, intrahepatic metastases, and/or extrahepatic disease and the following criteria: Candidates must satisfy diagnostic criteria for hilar CCA: malignant-appearing stricture on cholangiography and biopsy or cytology results demonstrating malignancy, carbohydrate antigen 19–9 Ͼ100 U/ml, or aneuploidy. The tumor should be considered unresectable on the basis of technical considerations or underlying liver disease (e.g., PSC). If cross-sectional imaging studies (CT scan, ultrasound, MRI) demonstrate a mass, the mass should be Ͻ3cm. Intra- and extrahepatic metastases should be excluded by cross-sectional imaging studies of the chest and abdomen at the time of initial exception and every 3 months before score increases. Regional hepatic lymph node involvement and peritoneal metastases should be assessed by operative staging after completion of neoadjuvant therapy and before LT. Endoscopic ultrasound-guided aspiration of regional hepatic lymph nodes may be advisable to exclude patients with obvious metastases before neoadjuvant therapy is initiated. Transperitoneal aspiration or biopsy of the primary tumor (either by endoscopic ultrasound, operative, or percutaneous approaches) should be avoided because of the high risk of tumor seeding associated with these procedures. 13. Heriditary Complicated by intractable high-output heart failure, biliary sepsis, and intrahepatic hemorrhage. hemorrhagic telangiectasia 14. Portal hypertensive Refractory variceal bleeding can be defined as acute severe variceal bleed requiring Ͼ6 units in 24 hours or bleeding Ͼ2 units per day over 3 days, airway intubation and the insertion of a Minnesota or Blakemore tube, in spite of optimal endoscopic treatment, coagulation support, and pharmacologic management and where is either ineffective or contraindicated. Chronic recurrent variceal or portal hypertensive gastropathy bleeding can be defined as the requirement of more than 2 units of blood transfusion per week for more than 6 weeks in a patient with a TIPS or if TIPS is contraindicated and lack of response to endoscopic and pharmacological treatment. EEG: electroencephalogram; LDL: low-density lipoprotein; TIPS: transjugular intrahepatic porto-systemic shunt; UKELD: United Kingdom Model for End-Stage Liver.

BoardintheUnitedStates.Giventherelativelysmall diagnoses include hepatocellular carcinoma (see volume of such cases and often subjective nature of below), familial amyloid polyneuropathy, primary non-mortality end points, the selection and alloca- oxalosis, hepatopulmonary syndrome, and tion criteria for patients with variant syndromes (par- portopulmonary hypertension. ticularly those without hepatocellular carcinoma) are 2. Diseases for which the pretransplant mortality inevitably less well-developed, less objective, and less risk may not be reliably estimated by evidence-based than those used for other patients with MELD/UKELD alone. Examples include chronic liver disease. HCV/human immunodeficiency virus coinfection Patients with exceptional diagnoses can be and total intestinal failure requiring concurrent grouped into at least three broad categories: small bowel transplantation. 1. Those for whom there are good reasons to believe 3. Diseases or syndromes that may merit selection, that longer waiting time for a transplant would but not necessarily prioritization, for transplant not necessarily compromise their transplant-free on the grounds of poor liver disease-related short-term survival prospects but would have a quality of life in the absence of high mortality risk significantly deleterious impact on their survival without transplant. Examples include patients prospects after the procedure. Examples of such with chronic liver disease and refractory pruritus,

178 Chapter 21: Recipient selection

chronicorrelapsinghepaticencephalopathy,and excluding from transplant many patients with poten- recurrent cholangitis. tially curable disease. Expansion of the HCC selec- tion criteria by adoption of those of the University of Candidate selection for hepatocellular California, San Francisco (a single-tumor nodule ≤ 6.5 cm; or three or fewer tumors, the largest being carcinoma Ͻ 4.5 cm with the sum of the total tumor diameters Most liver transplant programs have adopted the equaling Ͻ 8 cm) or the more recently proposed “up- Milan criteria for selecting patients with hepatocellular to-seven criteria” (a single nodule ≤ 7cm;twonodules carcinoma (HCC) for transplantation. These require with each ≤ 5 cm; three nodules with each ≤ 4cm; the presence of a single mass lesion ≤ 5cmindiameter four nodules with each ≤ 3cm;fivenoduleswitheach or up to five lesions that are all ≤ 3cmintheabsence ≤ 2 cm) has been shown to be associated with a 5-year of evidence of extrahepatic disease or radiological evi- graft survival exceeding 50%. dence of vascular invasion. Lesions between 5 and 7 cm in diameter may be considered after demonstra- Candidate selection in patients with tion of favorable tumor biology. In addition, the pres- ence of clinical, radiological, or histological evidence acute liver failure of cirrhosis and/or portal hypertension is an addi- The process of selection for patients with acute liver tional prerequisite for transplantation in most selec- failure presents clinicians with the unique challenge of tion schemes since resection or ablation is usually con- the need for prompt assessment and decision making sidered more appropriate for non-cirrhotic, or even while carefully balancing the risks of death and pro- some compensated cirrhotic, patients who otherwise gression to non-transplantable disease with the poten- satisfy the above criteria. tial for recovery in the absence of a transplant. Acute Although histological proof is only occasionally liver failure patients are given the highest priority for required, radiological assessment by means of ultra- donor organs in almost all allocation systems. The sonography, computed tomography, magnetic reso- selection criteria in patients with acute liver failure in nance imaging, and (in borderline cases) hepatic the United States, United Kingdom, and much of the angiography is usually used to establish the diagno- Western world are predicated on those developed by sis, size, and number of focal liver lesions. The radio- King’s College Hospital (Table 21.4). These criteria dif- logical diagnosis of HCC requires evidence of lesion fer according to the underlying etiology (depending on congruity and both arterial enhancement and por- whether or not it is related to acetaminophen inges- tal venous phase washout on at least two radiologi- tion) and use easily obtainable parameters. The MELD cal imaging modalities with the size being assessed by score may also be of prognostic utility in these cases. the widest dimensions on either scan. Biopsy evidence The Clichy criteria, which are commonly used in of HCC and/or markedly elevated alpha-fetoprotein France and a number of Northern European countries, level are additional selection criteria in patients with stipulate that a transplant is necessary in patients with Ͻ inconclusive radiological evidence of HCC. In addi- grade 3 or 4 encephalopathy and a factor V level 20% Ͻ tion, chronic liver disease candidates who have a ris- in those younger than 30 years of age or 30% if ing alpha-fetoprotein level ≥ 500 ng in the absence of older than 30 years of age. These criteria are rarely imaging evidence of HCC may be listed for liver trans- used elsewhere as a result of the limited availability plantation in the United States. of factor V assay, uncertain applicability to patients Notwithstanding these criteria, however, 20% of with non–hepatitis B virus (HBV) acute liver disease, patients undergoing liver transplantation for HCC in and scarce evidence of reproducibility in validation the United States in the MELD era exhibit no explant studies. evidence of HCC. The fibrolamellar variant of HCC is not constrained by these size and volume criteria Candidate selection for for transplant, although, given the absence of cirrho- re-transplantation and simultaneous sis, resection is considered more appropriate in most such cases. liver and kidney transplantation There has been increasing recognition in recent Both candidate selection and organ allocation in years that the Milan criteria may be too restrictive, the setting of re-transplantation are compounded by

179 Section 4: Liver

Table 21.4 Selection criteria for patients with acute liver failure Category 1: Etiology: Acetaminophen poisoning – pH Ͻ 7.25 more than 24 hours after overdose and after fluid resuscitation Category 2: Etiology: Acetaminophen poisoning – coexisting prothrombin time Ͼ 100 seconds or INR Ͼ 6.5, serum creatinine Ͼ 300 mmol/l or anuria, and grade 3 or 4 encephalopathy Category 3: Etiology: Acetaminophen poisoning – serum lactate more than 24 hours after overdose, Ͼ 3.5 mmol/l on admission or Ͼ 3.0 mmol/l after fluid resuscitation Category 4: Etiology: Acetaminophen poisoning – any two of the criteria from category 2 with clinical evidence of deterioration (e.g., increased ICP, FiO2 ≥ 50%, increasing inotrope requirements) in the absence of clinical sepsis Category 5: Etiology: Seronegative hepatitis, hepatitis A, hepatitis B, or an idiosyncratic drug reaction; prothrombin time Ͼ 100 seconds or INR Ͼ 6.5, and any grade of encephalopathy Category 6: Etiology: Seronegative hepatitis, hepatitis A, hepatitis B, or an idiosyncratic drug reaction; any grade of encephalopathy, and any three from the following: unfavorable etiology (idiosyncratic drug reaction, seronegative hepatitis), age Ͼ 40 years, jaundice-to-encephalopathy time Ͼ 7 days, serum bilirubin Ͼ300 mmol/L, prothrombin time Ͼ 50 seconds, or INR Ͼ 3.5 Category 7: Etiology: Acute presentation of Wilson’s disease or Budd–Chiari syndrome; combination of coagulopathy and any grade of encephalopathy Category 8: Hepatic artery thrombosis on days 0–21 after liver transplant Category 9: Early graft dysfunction on days 0–7 after liver transplant with at least two of the following: AST Ͼ 10|000 IU/l, INR Ͼ 3.0, serum lactate Ͼ 33 mmol/l, absence of bile production Category 10: Any patient who has been a live liver donor who develops severe liver failure within 4 weeks of the donor operation AST: aspartate aminotransferase; Fio2: inspired oxygen concentration; ICP: intracranial pressure; INR: international normalized ratio.

unique clinical and ethical dilemmas. In practice, ity of post-transplant recovery of native renal function, patient beneficence and the consequences of subop- high risk of post-transplant mortality in the setting of timal organ utilization or technical reasons for graft increased recipient disease severity, and high quality of failure are often more pressing considerations in re- kidney grafts utilized in liver–kidney transplants. On transplant decision making than mitigating concerns the other hand, published data showed that impaired about compromising graft utility as a result of infe- renal function has an adverse impact on both pre- rior survival and risk of disease recurrence. In general, and post-liver transplant survival; that dual transplan- patients with irreversible graft failure should ideally be tation is associated with superior rejection-free kid- listedforre-transplantationatalowerdegreeofdisease ney allograft but not patient survival compared with severity than that warranting primary transplantation, a sequential kidney-after-liver procedure; that com- given the poor outcome of re-transplanting sick recip- bined transplantation confers patient and liver allo- ients, risk of accelerated disease progression in the graft survival benefit over solitary liver transplantation absence of re-transplantation, and underestimation of only among those requiring long-term (Ͼ3months) mortalitybyscoringsystemssuchasMELDamongre- dialysis; and that equivalent kidney allograft survival listed candidates. It has therefore been argued that re- to that after solitary kidney transplantation can only transplant candidates in the United States are under- be achieved if dual transplantation is restricted to served by the MELD-based allocation system and that patients on long-term dialysis with MELD scores less such candidates need to be given additional priority than 23. points, as is the case currently with HCC, to maximize A recent consensus conference in the United States graft utility. has concluded that Regional Review Boards should Similarly, judging the appropriateness of liver determine listing for simultaneous liver and kidney transplantation concurrently with another organ, usu- transplantation, as with other MELD exceptions, with ally a kidney, presents transplant professionals with a automatic approval for (1) end-stage renal disease difficult challenge. with cirrhosis and symptomatic portal hypertension On the one hand, for example, offering simultane- or hepatic vein wedge pressure gradient ≥ 10 mmHg; ous liver and kidney transplantation to patients with (2) liver failure and chronic kidney disease with a hepatorenal syndrome could constitute an inappropri- measured glomerular filtration rate ≤ 30 ml/min; ate use of a scarce resource given the high probabil- (3) acute kidney injury or hepatorenal syndrome with

180 Chapter 21: Recipient selection creatinine ≥ 2.0 mg/dl and dialysis ≥ 8weeks;and Most centers’ selection guidelines require at least 6 (4) liver failure and chronic kidney disease with renal months of recorded abstinence before offering a liver histology demonstrating Ͼ 30% glomerulosclerosis transplant to patients with alcoholic liver disease. Nev- or 30% fibrosis. Similar criteria are now used in the ertheless, there is no robust evidence base to support United Kingdom. this practice. Allowing sufficient time for any poten- Rarely, patients with heart and liver failure will be tial improvement in liver function with abstinence considered for combined heart–liver or heart–lung– that might allow transplantation to be avoided is very liver transplant. Indications include heart failure with important. Thus assessment of the prognosis for sobri- cirrhosis, hemochromatosis, and cystic fibrosis. The ety should also be based on the patient’s insight, social short-term mortality is higher than that of each trans- support, and the presence of psychological comor- plant procedure alone, but the long-term outcome is bidity. Active cigarette smoking is also a contraindi- similar to that of heart or liver transplant alone. Selec- cation in some transplant centers because of recog- tion should be done in conjunction with Regional nized increased transplant morbidity and mortality. Review Boards or similar organizations. Methadone usage is not necessarily a contraindication but requires careful analgesic management both before and after transplantation. Contraindications and delisting criteria Few transplant systems have formulated standard- The relative sparseness of published evidence in this ized de-selection criteria for patients already listed area makes it difficult to define the degree of dis- for transplant. Indeed, the evidence suggests a lack of ease severity that renders liver transplantation a futile agreement among US transplant programs regarding undertaking. In addition, several other recognized which cases should be removed from the waiting list recipient risk factors for post-transplant mortality due to a change in clinical status. In practice, however, needtobetakenintoaccount,and,althoughnonepro- patients are de-selected for one of the following rea- scribes transplantation, accumulation of these risk fac- sons: (1) if their clinical status sufficiently improves tors or extremes thereof is, in practice, deemed a bar or, in the case of those with HCC or other exceptional to the procedure, particularly if they are considered diagnoses, progresses to a point whereby they cease to predictive of less than 50% 5-year survival after meet the criteria on which their original selection for transplantation. the procedure was based; (2) if they develop one of the These factors include advancing age (although no above absolute contraindications while on the waiting specific age limitation to successful liver transplan- list; (3) if their clinical status deteriorates to an extent tation has been identified), diabetes, ischemic heart whereby they would be deemed unlikely to survive the disease, renal dysfunction, requirement for renal sup- procedure. Determination of this point is heavily sub- port, requirement for mechanical ventilation, malnu- ject to clinical judgment but, in practice, is guided by trition, morbid obesity, extensive portal venous sys- the accumulation of the above-mentioned adverse risk tem thrombosis, advanced encephalopathy, previous factors. upper abdominal surgery, prior transplant, poor func- tional status, hyponatremia, and higher serum potas- Further reading sium. Importantly, there is no good evidence that liver Neuberger J, Gimson A, Davies M, et al. Selection of disease severity (e.g., assessed by MELD) per se, how- patients for liver transplantation and allocation of ever extreme, predicts transplant futility. donated livers in the UK. Gut 2008; 57: 252–7. Current guidelines for selection of patients with Steinman T, Becker B, Frost A, et al. Guidelines for the alcoholic liver disease usually exclude those with clin- referral and management of patient’s eligible for solid ical or histological evidence of active alcoholic hepati- organ transplantation. Transplantation 2001; 71: 1189. tis, history of repetitive episodes of non-adherence to United Network for Organ Sharing (UNOS). Allocation of medical care, evidence of return to drinking after full livers. Available at: www.unos.org. professional assessment and advice, and history of cur- WiesnerR,LakeJR,FreemanRB,GishRG.ModelforEnd rent or consecutive illicit drug misuse (except occa- Stage Liver Disease (MELD) exception guidelines. Liver sional cannabis use). Transpl 2006; 12(Suppl 3): S128–36.

181 Figure 4B.1 Chronic, active TA can be identified by the inflammatory infiltrate that is typically under the endothelium and accompanied by intimal fibrosis. This artery from a renal allograft shows T cells within an arterial wall and focally lifting the endothelium of an affected vessel stained for CD3 by immunohistochemistry. Antibody also may play a role in the pathogenesis of TA, as shown in experimental studies.

(A) (B)

Figure 4B.2 Contrasting appearances of TA and non-immunologic arteriosclerosis can be appreciated in sections stained for elastic fibers. (A) An artery from a renal allograft with TA shows intimal fibrosis without an increase in elastic fibers and with infiltrating inflammatory cells. (B) An artery from a native kidney with arteriosclerosis shows duplication of the elastic lamina, termed fibroelastosis, and few inflammatory cells. A and B, Weigert elastic stain. Figure 4B.3 In chronic antibody-mediated rejection CHR, C4d is Figure 4B.4 In CHR, multi-lamination of peritubular capillary walls deposited in peritubular capillary walls, as shown in this occurs in renal allografts, best shown by electron microscopy. This is immunofluorescence micrograph of an allograft kidney cryostat a manifestation of repeated episodes of endothelial injury and section stained with anti-C4d monoclonal antibody. CHR is well repair mediated by antibodies. Similar changes occur in glomerular documented in the kidney but not yet in other organs, although it capillaries, where it is termed transplant glomerulopathy. probably exists.

Figure 4B.5 Transplant glomerulopathy (TG) is defined by duplication of the glomerular basement membrane, easily appreciated when severe by light microscopy in sections stained with PAS. This is highly associated with circulating anti-donor antibodies and deposition of C4d. However, the pattern of injury can be due to other causes, such as thrombotic microangiopathy. (A) (B)

Figure 4B.6 Chronic rejection in the liver affects bile ducts and vessels. (A) In vanishing bile duct syndrome, a manifestation of chronic rejection in the liver, there is loss of bile ducts in portal tracts with associated cholestasis. (B) Chronic allograft vasculopathy in an hepatic artery shows prominent foam cells in the intima. Foam cells, which are macrophages filled with lipid, are characteristic of chronic rejection and are probably related to hyperlipidemia. Photomicrographs courtesy of Dr. V.A. Marcus, McGill University (A) and A.J. Demetris, University of Pittsburgh (B). Figure 4B.7 Chronic rejection in heart allografts is generally best appreciated in angiographic studies, such as intravascular ultrasound, because the diagnostic lesion is in the small-to-large Figure 4B.8 In addition to TA, chronic rejection in the lung results arteries not typically sampled in endomyocardial biopsies. TA is in edematous myofibroblastic obliteration of a bronchiolar lumen, manifested in the heart by the same features as in other organs, termed obliterative bronchiolitis, which can be identified in biopsies namely, concentric intimal fibrosis leading to luminal stenosis and and has characteristic effects on pulmonary function (Courtesy of eventual occlusion with intramural chronic inflammation, as shown Dr. E.J. Mark, Massachusetts General Hospital). in this explant sample (Courtesy of Dr. J.R. Stone, Massachusetts General Hospital). Section 4 Liver Chapter Living donor liver transplantation

22 Koji Hashimoto, Cristiano Quintini, and Charles Miller

Key points Theoretically, LDLT has the potential of supply- r ing an unlimited number of liver grafts, but its prac- Living donor liver transplantation has been a tical application is mitigated by the ethical princi- major area of development in the field for the ple of primum non nocere –firstdonoharm.Living last two decades. r organ donation is the only field in medicine in which Major technical and physiological advances a healthy person undergoes a major surgical proce- have made this technology the standard of dure without presenting a pathological condition and care in parts of the world where deceased in which the only aspiration of the patient is to ben- donor options are rare or non-existent. r efit another human being. Therefore, the initial era of In most Western countries where the LDLT involved only adults donating a small portion majority of liver transplantation is performed of their liver to small pediatric recipients, thus maxi- with deceased donor grafts, the decision to mizing donor safety. But with ever-increasing success usealivingdonorratherthanwaitfora and experience, living donor technology was gradually deceased donor graft is a complicated ethical applied in the adult setting. The first successful applica- and surgical conundrum. tion of adult-to-adult LDLT was performed in 1993 in r The combination of concern for donor safety Japan by Makuuchi using the left lobe rather than the and the possible availability of a deceased right lobe, again due to concerns about donor safety donor graft has limited the expansion of and the known morbidity and mortality associated adult-to-adult living donor liver with right hepatic lobectomy. This was successful, but transplantation in the West. its application was limited by the theoretical and actual amount of liver tissue needed by the recipient. As the demand for this life-saving therapy has increased, sur- The idea of donating part of the liver from a liv- geons have been forced to utilize the larger right lobe, ing donor was conceived and described in the late especially when the donor is smaller than the recipi- 1960s, but it took more than 20 years to imple- ent. However, donor morbidity and mortality remains ment clinically. In December 1988, Raia and col- a major issue, and the significant risk of donor harm or leagues attempted the first living donor liver trans- death must always be borne in mind by both the trans- plantation (LDLT) on a 4-year-old boy who died plant team and the donor and their relatives. 6 days after the transplant. In July 1989, the first suc- cessful LDLT was performed by Russell Strong in Aus- tralia; a pediatric patient received a left lateral segment Indications (segments II and III) from his mother. This was fol- The indications for LDLT are the same as that for lowed by the first successful LDLT of a child in the deceased donor transplantation (see Chapter 21). United States by Cristoph Broelsch at the University of Careful selection of both donor and recipient is cru- Chicago; their team performed 20 cases in the ensuing cial in preventing donor complications and optimiz- 12 months. ing recipient outcomes. Poor survival rates have been

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

182 Chapter 22: Living donor liver transplantation reported in LDLT involving recipients with very high plantation, thus negating one of the most beneficial Model for End-Stage Liver Disease (MELD) scores. aspects of LDLT. In Western countries or where there are significant cadaveric donor programs, the candidate for a poten- tial LDLT is listed for donation after brain death Donor evaluation (DBD) liver transplantation. Because a patient with The aim of the donor evaluation is to assess whether a high MELD score is more likely to receive a DBD the donor is medically and psychologically suitable organinatimelyfashion,LDLTintheUnitedStates for living donation. Equally important is to identify is now mainly applied to adult patients with lower anatomical conditions that could increase donor risks MELD scores who, for a variety of reasons such as and jeopardize either donor or recipient recovery. encephalopathy, large tumor, or other unusual com- Donor selection criteria vary slightly among dif- plications, are disadvantaged and do not fair well with ferentprograms.IntheUnitedStates,toavoidacon- the MELD system. In addition, living donor options flict of interest, the evaluation of the donor is con- should not be used for indications that result in only ducted by a designated donor advocacy team, which short-term gains for the recipient. independently assesses donor candidacy. During the evaluation, the donor is educated regarding the risks of Special recipient indications the procedure. This includes discussion of the morbid- ity and mortality rates reported in the medical litera- Hepatocellular carcinoma ture,aswellasoutcomesofthesurgeonswhowillper- form the operation. The donor has the right to with- Hepatocellular carcinoma (HCC) patients usually have draw his willingness to donate right up to the time less portal hypertension and lower chemical MELD of surgery. Normally the donor should be completely scores. Furthermore, the shorter waiting time can healthy, between 18 and 55 years of age, and have a reduce the risk of drop-out from transplant waiting list clearandestablishedrelationshipwiththerecipient. due to tumor progression. The Adult-to-Adult Living The donor and recipient should be blood group iden- Donor Liver Transplantation Cohort Study (A2ALL) tical or compatible. group revealed that LDLT recipients had a shorter The first step of the evaluation begins with a thor- waiting time compared with DBD liver transplanta- ough medical history and physical examination. Par- tion (LT) recipients (160 versus 469 days), but a higher ticular emphasis is placed in the psychosocial evalu- rate of HCC recurrence within 3 years (29% versus 0%) ation of the potential donor. The donor should make than classical transplant recipients. However, it is still the decision voluntarily, without any coercion and controversial whether LDLT accelerates the recurrence any direct or indirect financial gain deriving from the of HCC. donation. An extensive lab profile and serologic tests Although HCC was once the major indication complete the first step of the evaluation. If the donor for LDLT in the United States, with the introduction is deemed to be a good candidate from a medical and oftheMELDsystemandtheprioritygiventopatients psychosocial perspective, then the next step is to assess with HCC, most patients can receive a DBD graft the anatomical and surgical aspect. Recent advances within 3 months, thus making the use of LDLT far less in the three-dimensional reconstruction of the liver important. using multi-phase computed tomography (CT) scans has contributed to a precise non-invasive mapping Hepatitis C of the most important vascular structures, allowing An early study from Spain suggested that hepatitis C for a preoperative simulation of the graft procure- virus (HCV) recurrence is more severe after LDLT. ment (Figure 22.1). The type of graft (left lateral seg- One possible explanation is that the regenerating liver ment, left lobe, right lobe) is determined by the donor is more susceptible to HCV infection; a hypothesis liveranatomyandbytherecipientsizeandsever- based on early in vitro studies. However, more recent ity of disease. Biliary imaging can be performed with studies suggest that there is no difference in HCV contrast-enhanced CT imaging or by endoscopy. At recurrence. Because of the high chance for recurrent the end of the evaluation process, only 30–40% of disease after any type of transplant, patients with HCV aspiring donors will be considered good candidates for cirrhosis are best managed by avoiding early trans- donation.

183 Section 4: Liver

a GRWR of at least 1% to give a margin of safety to the recipient in case of certain specific technical complex- ities. A GRWR below 0.6–0.8% increases the chance of developing postoperative liver insufficiency known as small-for-size syndrome (SFSS). Donor safety is the primary concern; therefore, the ideal graft is the one that leaves a donor an FLR above 35% and at the same timeprovidesagraftwithanadequatesizewithrespect to the recipient. Familiarity with the liver anatomy is essential for safe performance of LDLT. The liver can be divided into two lobes (right and left). Each lobe can be fur- ther divided into four segments. The left lobe con- sists of segments I–IV; the right lobe consists of seg- ments V–VIII. Each segment is independent from a functional stand point and relies on one arterial and portal venous inflow (segmental branch of the hepatic artery and portal vein). The venous outflow is differ- ent, as both lobes share drainage via the middle hep- atic vein (MHV); this anatomical detail has created an extensive literature regarding the appropriate par- tition of the MHV to either the donor or recipient. A clear understanding of the biliary duct anatomy is critical in preventing complications in both donor and recipient. There are essentially three types of grafts that canbeconsideredinLDLT(Figure 22.2); the left lat- eral segment (segments II and III), the left lobe (seg- ments I–IV), and the right lobe (segments V–VIII). The smallest graft is represented by the left lateral seg- ment, which usually represents 20–25% of the total liver volume. This graft is reserved for pediatric recipi- Figure 22.1 Three-dimensional reconstruction of the liver using ents. The left lobe, which usually represents 30–40% of multi-phase CT scan. the total liver volume, is usually offered to teenagers or small adults. Finally, the right lobe, which represents Donor operation and type of about 60–70% of the total liver volume, is reserved fortheremainderoftheadultpopulation.Thisisthe liver grafts largestgraft,andalthoughitoffersthemostconsistent In order to understand the type of grafts used in LDLT, results in the recipient, it is also the one that is asso- it is important to define two concepts: the future liver ciated with the highest morbidity and mortality in the remnant (FLR) and the graft-to-recipient body weight donor. ratio (GRWR). The FLR is the proportion of the whole donor liver that is estimated to remain after the dona- tion. An FLR of 30–35% is considered a safe and Donor complications acceptable lower limit under which donation should Despite donor safety being of paramount importance notbeattemptedduetothehigherriskofdevelop- in LDLT, finite morbidity and mortality rates have ing postoperative liver insufficiency or failure. The been reported worldwide. The mean rate of compli- GRWR is the ratio between the donor graft weight cations for left lateral segment, left lobe, and right and the recipient body weight. The lower limit of graft lobe living donor hepatectomy in the most experi- acceptability is considered to be approximately 0.6– enced centers are, respectively, approximately 15%, 0.8%; however, many transplant programs like to have 25%, and 35%. These remain very high despite a

184 Chapter 22: Living donor liver transplantation

Right lobe Left lobe Figure 22.2 Diagram of the liver anatomy; the liver can be divided into two lobes (right and left). Each lobe can be further divided in four segments. The Left Lateral Segment left lobe consists of segments I–IV; the right lobe consists of segments V–VIII.

number of improvements in surgical technique and Table 22.1 Classification of complications according to the patient care, which must always be borne in mind Clavien system when considering LDLT as a treatment option. It is Grade 1 Any deviation from the normal postoperative important to report these complications according to course without the need for pharmacological treatment or surgical, endoscopic, and the Clavien system, which scores them according to radiological interventions. five categories of severity (Table 22.1). Despite this, Allowed therapeutic regimens are drugs as reported complications vary significantly among dif- antiemetics, antipyretics, analgesics, diuretics, electrolytes, and physiotherapy. ferent programs, possibly due to different experience This grade also includes wound infections opened or reporting methodology. Biliary complications rep- at the bedside. resent the most frequent source of morbidity occur- Grade 2 Complications requiring pharmacological ring in 5–25% of these patients. Vascular complica- treatment with drugs other than such allowed for grade 1 complications. tions are rare, but accompanied by significant mor- Blood transfusions and total parenteral nutrition bidity and mortality (5–7%). The complication rate are also included. is directly correlated to the amount of parenchyma Grade 3 Complications requiring surgical, endoscopic, or removed from the donor, although complications radiological intervention. associated with anesthesia and postoperative recov- Grade 3a Intervention not under general anesthesia. ery (deep vein thrombosis, pulmonary embolism, etc.) Grade 3b Intervention under general anesthesia. have been reported. The overall donor mortality rate Grade 4 Life-threatening complications (including central is approximately 0.1% for left lobe donation and 0.5% nervous system complications) requiring for right lobe donor hepatectomy. This remains high intensive care unit stay. and must be discussed closely with donors and their Grade 4a Single-organ dysfunction (including dialysis). family when considering LDLT and during the consent Grade 4b Multi-organ dysfunction. process. Grade 5 Death of the patient.

Recipient operation portal hypertension may in fact be responsible for graft The hepatectomy is performed preserving the retro- congestion and dysfunction associated with SFSS. To hepatic vena cava. Intraoperative hemodynamic stud- combat this problem, various forms of inflow mod- ies are emerging in recent years as a tool to guide ulation such as portosystemic shunts, splenectomy, implantation technique and inflow modulation. Severe splenic artery ligation, and infusion of vasoactive

185 Section 4: Liver

agents have been described as promising tools to avoid high urgency and have the best chance of receiving an SFSS in marginal-sized grafts. organ from a DBD in a timely fashion. Even consider- One of the most important technical aspects of ing this disadvantage, the long-term survival in adult LDLT is optimization of venous outflow. The graft LDLT is satisfactory. is placed in an orthotopic position. The hepatic vein anastomosis should take into consideration the final Left lobe versus right lobe position of the graft and the anatomical adjustments in Information regarding the comparison between left the first month post-transplantation, when the partial lobe and right lobe grafting is very limited. Although graft usually doubles its size. Venous outflow recon- favorable outcome of LDLT using left lobe graft has structionsareverycommonwhenusingrightlobesas been reported in adult recipients, many transplant cen- opposed to left lateral segments and left lobes, which ters still routinely use right lobe grafts. This practice most commonly present with a single common out- best ameliorates the issue of graft size inherent in the flow. fact that right lobe represents 60–70% of whole liver After completing the hepatic vein anastomosis, the volume, whereas the left lobe provides only 30–40%. portal vein anastomosis is performed and the graft The larger graft is more likely to provide at least 40% is reperfused. The arterial anastomosis is often per- of the recipients’ standard liver volume, be able to meet formed using loop magnification or the microscope the patient’s metabolic demands, and withstand the due to the small caliber of the vessel. The duct-to-duct hyperdynamic splanchnic flow seen in adult cirrhotics. biliary reconstruction is performed whenever possi- In determining whether a donor liver can pro- ble using the recipient bile duct. This decreases the vide sufficient hepatocyte function, it is important biliary complication rate, is easier to perform, and to estimate the functional capacity of the graft. This provides endoscopic access to the duct in case of estimate is known as “functional graft size” and is a complications. When multiple ducts are presents, the composite function of actual graft size modified by biliary reconstruction is usually achieved by mean of a severity of the recipients’ condition, the degree of por- hepato-jejunostomy with a Roux-en-Y limb. tal hypertension, and the degree to which a graft’s out- flow might be impaired. In using left lobe grafts, actual Recipient outcomes graft size may not always exceed the 40% of standard liver volume threshold. Even in such cases, left lobe Graft and patient survivals grafts have excellent venous outflow and can provide adequate functional mass in patients with low MELD For children, graft and patient survivals are compara- score and/or little portal hypertension. However, if bleorbetterforLDLTthandeceaseddonorgrafting.In small left lobe grafts are used for patients with high a large series from Kyoto University, 5-year graft and MELD score or with severe portal hypertension, the patient survivals were 81% and 82%, respectively. Sim- risk of graft failure is very high. Thus careful donor and ilarsurvivalrateshavebeenobservedinotherAsian, recipient selection clearly affects the outcome. In addi- European, and US centers. tion, inflow modifications to reduce portal flow have In general, adult patients undergoing LDLT have also been beneficial when using left lobe grafts. lower MELD scores than those undergoing DBD LT, so direct comparisons of post-transplant survival may be misleading. On the other hand, when analyzed on Recipient complications an intent-to-treat basis from the time of evaluation, the A2ALL group found a significant advantage for Small-for-size syndrome those recipients receiving living donor grafts. Most In adult-to-adult LDLT, recipients have a risk of early of the advantage was due to the avoidance of death on postoperative graft failure that is separate and distinct the waiting list. from primary non-function seen in deceased donor In Asia, where the number of DBD is extremely grafting. This graft dysfunction is known as small- limited, LDLT has been performed even in patients for-size syndrome (SFSS) and is characterized by pro- with very advanced liver failure. In contrast, those gressive cholestasis, intractable ascites, coagulopathy, patients are rarely considered as candidates for LDLT and renal failure. SFSS typically results in a reduc- in Western countries because such patients are listed as tion of graft survival rate and may increase recipient

186 Chapter 22: Living donor liver transplantation mortality. The pathogenesis of SFSS is multifactorial varieties: early bile leaks and late biliary strictures. and has not been completely elucidated. Small graft The incidence of biliary complications has decreased size has been shown to be related to SFSS; however, over time, but it is still higher than that of DBD actual graft size does not always reflect functional liver liver transplantation. Because right lobe grafts often mass as described previously. In general, when GRWR have multiple bile ducts, they are associated with a is less than 0.8% or graft volume is less than 40% of higher rate of biliary complications than the left lobe a recipient’s standard liver volume, the risk of SFSS graft, which almost always has only a single duct becomes higher. to reconstruct. In recent years, most centers prefer The severity of liver disease and recipient sta- duct-to-duct reconstruction to hepato-jejunostomy tus along with severe portal hypertension also affects because it is less time-consuming, is associated with the risk of SFSS. Portal hyperperfusion to the small a lower incidence of early bile leaks, and allows graft induces shear stress and sinusoidal injury as well easy postoperative access to examine the bile duct as vasospasm in the hepatic artery. Although MELD endoscopically. score does not accurately reflect the severity of liver Bile leaks typically originate from either the anas- disease for all liver transplant candidates, patients with tomoticsiteorcutsurfaceofthelivergraft.They alowerMELDscoretendtohavealowerriskofdevel- are diagnosed when biliary drainage is seen from the oping SFSS. Thus patient selection plays a crucial role abdominal drain or the patient develops fever, abdom- in regard to graft and patient outcomes. inal pain, and abnormal liver function tests. In most In the case of portal hyperperfusion in the small cases, this complication can be managed by percuta- graft, inflow modification of the portal vein is essen- neous drainage and biliary stenting. If left untreated, tial. To reduce portal vein flow, a pharmacologi- bile leaks can lead to sepsis and graft dysfunction. If cal approach may be beneficial. Beta-blockers and infection occurs, antibiotic treatment in addition to somatostatin have been used in an attempt to attenu- drainage will be necessary. ateportalhyperperfusion.Surgicalapproachesinclude Biliary anastomotic strictures are usually caused splenic artery ligation, splenectomy, and portocaval by local ischemia at the anastomotic site. Stricture shunt. The application of these approaches should be causes elevated liver function tests with or without bil- based on intraoperative measurement of portal vein iary dilatation. It usually can be managed with a bil- flow and pressure. Ideally, portal vein flow should be iary stent placed endoscopically or percutaneously and less than 2 ml/min/g (liver weight). According to a rarely requires surgical revision. Biliary complications recent report from Kyoto University, portal vein pres- secondary to hepatic artery complications are resistant sure less than 15 mmHg is related to better outcome. to these interventions, and re-transplantation may be The major concern with portocaval shunt is the steal required. phenomenon of the portal flow to the systemic circu- lation, which may jeopardize graft regeneration and function. When this occurs, the shunt must be closed. Vascular complications Assuring perfect venous outflow is equally as impor- The concept of post-transplant vascular complica- tant as appropriate portal inflow in avoiding SFSS. In tions in LDLT is not different from DBD liver trans- the left lobe graft, venoplasty between left and mid- plantation. Hepatic artery thrombosis (HAT) is a dle hepatic veins is useful to increase the diameter of catastrophic complication with high risk of biliary graft venous orifice. In the right lobe graft, drainage necrosis and graft loss. The incidence of HAT has sig- of middle hepatic vein tributaries is crucial to prevent nificantly decreased from 25% in the 1990s to less than congestion of the anterior segment. When the infe- 10% in the last 10 years. This improvement has been rior right hepatic vein is greater than 5 mm in diame- achieved by the introduction of microsurgical tech- ter, reconstruction of this vein to vena cava is strongly niques in hepatic artery anastomosis. Frequent moni- recommended. toring with duplex ultrasonography is very useful for early detection of HAT. Urgent thrombectomy and revascularization are sometimes effective to pre- Biliary complications vent devastating biliary complications. However, most Biliary complications are the most common cause recipients with early HAT experience intrahepatic bil- of significant recipient morbidity and come in two iarystricturesandbilomas,whichcancausebiliary

187 Section 4: Liver

sepsis. Despite efforts at graft salvage, these patients (Ͼ70% of total liver volume). In these cases, the FLR haveahighriskofgraftfailureandmortality. after right lobectomy will be less that 30%, which leads The incidence of portal vein thrombosis ranges to an unacceptably high risk of donor morbidity and from 2–6%; it is also a devastating complication after mortality. To reduce the donor risk and obtain suf- LDLT. Risk factors include small graft size, presence of ficient liver volume for an adult recipient, two small portal vein thrombus at the time of LDLT, and use of grafts from two different living donors can be trans- vein grafts for portal vein reconstruction and the pres- planted into the single recipient. There may be many ence of large porto-systemic shunts. Early diagnosis different graft combinations used, such as two left lat- with duplex ultrasonography is key to reduce graft loss eral segments, two left lobes, a left lobe and small right and mortality. Thrombectomy and revascularization lobe, or a small right lobe and a left lateral segment is the gold standard for portal vein thrombosis; how- graft. The purpose is to keep the donor risk index ever, new interventional techniques including percu- as small as possible for each donor while providing taneous thrombolysis and suction thrombectomy have adequate functional liver mass to the recipient. Com- been reported. mon complications in dual-graft recipients are biliary Hepatic outflow obstruction is a serious complica- anastomotic stricture and outflow obstruction of the tion after LDLT. The main causes of the obstruction hepatic vein. Although dual-graft LDLT is accepted in are a twist of the hepatic vein anastomosis or rotation Asian countries, it has not become widely used in the of the graft and compression of the anastomosis after West. graft regeneration. Appearance of monophasic wave form and disappearance of triphasic wave form is diag- nostic in duplex ultrasonography. Venoplasty and stent ABO-incompatible LDLT placement are effective to avoid graft failure. The long- Liver transplantation across the ABO blood type bar- term efficacy and patency of these intravascular stents rier is usually not done except in emergent situa- need to be evaluated. tions when an ABO-compatible donor is not avail- able. ABO-incompatible LT is associated with a high Rejection (LDLT versus DBD LT) risk of antibody-mediated rejection, infectious com- plications, and vascular thrombosis resulting in poor Kidney transplant recipients from living donors expe- graft and patient survivals. Immunomodulation is key rience a lower incidence of acute and chronic rejec- to minimizing complications. In Japan, the utilization tion compared with those from deceased donors. This of ABO-incompatible donors is not just a rescue ther- might be a consequence of shorter ischemic time and apy; it is an extended routine application if no com- better graft quality, as well as an immunological advan- patible donors are available. According to the Japan tage due to human leukocyte antigen (HLA) match- Study Group for ABO-Incompatible Transplantation, ing between biologically related individuals. This find- this technique was started for pediatric recipients in ing is less obvious in LDLT. The overall rates of acute the early 1990s and was recently extended to adult rejection in LDLT are 47–68% in children and 11– patients. 33% in adults. According to a retrospective study from Interestingly, recipient age is a major determinant the A2ALL group, however, biopsy-proven acute rejec- of graft and patient survival in ABO-incompatible tion occurred in 27% of LDLT for adult recipients, LDLT. The 5-year survival rate of infants (Ͻ1year which was comparable to recipients from deceased old) is 85%, which is comparable to ABO-compatible donors (27%). Comparative rates of acute rejection LDLT. In contrast, the 5-year survival rate of adults have also been reported in children. However, the rates (Ͼ16 years old) is only 52%. ABO-incompatible LDLT of steroid-resistant rejection and chronic rejection are can be considered as a standard treatment for pediatric lower in LDLT for children. patients when no other identical or compatible donor is available. For adult patients, ABO-incompatible Recent topics in LDLT LDLT is still a challenge. The recent improvement in survival rates justifies continuing ABO-incompatible Dual graft LDLT for adults in particular areas where living Up to 25% of living donors are not suitable for right donorsarerealisticallytheonlysourceofliver lobe donation due to a proportionately large right lobe grafts.

188 Chapter 22: Living donor liver transplantation

Paired liver donor exchange program Further reading Another approach to avoid ABO incompatibility Akamatsu N, Sugawara Y, Tamura S, Imamura H, Kokudo between donor and recipient is paired donor exchange. N, Makuuchi M. Regeneration and function of hemiliver Paired kidney donor exchange programs have success- graft: right versus left. Surgery 2006; 139: 765– 72. fully increased organ availability in many countries since 1986 and are currently functioning as a valu- BarrML,BelghitiJ,VillamilFG,et al. Areportofthe Vancouver Forum on the care of the live organ donor: able tool for patients with ABO-incompatible donors. lung, liver, pancreas, and intestine data and medical A liver donor exchange program was implemented in guidelines. Transplantation 2006; 81: 1373– Korea in 2003 and Hong Kong in 2009. At the Asan 85. Medical Center in Seoul, 16 donor–recipient pairs Fan ST (eds). Living Donor Liver Transplantation.Hong (eight pair sets) were involved in an exchange program Kong: Takungpao Publishing, 2007. from 2003 to 2009. Operations were performed on an Fisher RA, Kulik LM, Freise CE, et al. A2ALL Study Group. elective basis in 12 and on an emergency basis in 4. Hepatocellular carcinoma recurrence and death After exchange, all pairs were ABO-identical or ABO- following living and deceased donor liver compatible.The5-yeargraftandpatientsurvivalrates transplantation. Am J Transplant 2007; 7: 1601–8. were 93.8%. Although there are logistical, ethical, reli- Gruessner RWG, Benedetti (eds). Living Donor Organ gious, cultural, and mathematical issues limiting avail- Transplantation. New York: McGraw-Hill, 2008. ability of paired donor programs, recipients with no HwangS,LeeSG,MoonDB,et al. Exchange living donor suitable donors can benefit from this modality, and the liver transplantation to overcome ABO incompatibility results are far better than those achieved with ABO- in adult patients. Liver Transpl 2010; 16: incompatible grafts. 482–90.

189 Section 4 Liver Chapter Surgical procedure

23 Simon J.F. Harper and Neville V. Jamieson

Key points mortality, was not achieved until 1967, such was the r formidable challenge it presented. Now long estab- A successful liver transplant requires a lished as a standard treatment for fulminant liver dis- number of procedures, including donor ease, the liver transplant operation is still a major hepatectomy, preparation of the donor liver, undertaking for both patient and surgeon alike. How- recipient hepatectomy, and implantation of ever, continuous refinement of all aspects of the proce- the liver graft. r dure over nearly 50 years has contributed greatly to the The cold ischemic time of the liver (period excellent outcomes now available to increasing num- between cold perfusion and reperfusion with bers of patients with liver disease. A successful liver recipient blood) should not be more than transplant relies on the effective execution of several 16 hours and ideally less than 12 hours. r component procedures, frequently involving several The recipient hepatectomy is frequently the surgeons; fundamentally these are the donor hepate- most difficult part of the transplant ctomy, preparation of the donor liver, recipient hepa- procedure, described by Starzl as “the tectomy, and implantation of the liver graft. bloodiest and most stressful experience in a surgeon’s life.” Donor procedure r If there is hemodynamic instability during the anhepatic phase, veno-venous bypass is The donor procedure represents the first crucial step instigated via the femoral vein (usually via in a liver transplant and should be performed by an the saphenous vein) to an internal jugular or experienced retrieval surgeon, aiming to preserve the subclavian line in the neck, with or without a donor liver in optimal condition (discussed further in combined portal bypass limb. Chapter 5). r Reperfusion of the liver is often the most dangerous part of the transplant procedure, Donor hepatectomy in donors after and close communication between surgeon brainstem death and anesthetist is crucial. In procurement from donation after brain death (DBD) donors, unlike donation after cardiac death Liver transplantation is life-saving but remains one of (DCD) donors, maintenance of cardiac output after the most challenging surgical procedures undertaken. confirmation of death allows a significant proportion The unique anatomical and physiological properties of dissection to be performed prior to cold perfu- of the liver and the profound pathological changes sion with preservation solution (“warm dissection”). associated with liver disease were long thought to This has the distinct advantage that important struc- render transplantation impossible. The first human tures are much easier to identify and can be prepared liver transplant was performed by Starzl in 1963 on a with greater precision, minimizing organ damage and 3-year-old girl with biliary atresia; the patient unfor- shortening dissection after perfusion (“cold dissec- tunately succumbed to massive intra-operative hem- tion”), during which suboptimal cooling of the organs orrhage. Successful liver transplantation, without early may occur. It is suggested by some surgeons that only

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

190 Chapter 23: Surgical procedure

Figure 23.1 Anatomy of accessory hepatic arteries. The left accessory or replaced hepatic artery arises from the left gastric and traverses the lesser omentum to enter the left lobe of the liver. The right accessory or replaced hepatic artery arises from the SMA near its origin and runs behind the pancreas and portal triad before entering the posterolateral aspect of the hilum of the liver.

minimal warm dissection should ever be undertaken, cled at this early stage in preparation for cannula- however, arguing that a prolonged laparotomy in the tion to allow rapid perfusion in the case of sud- context of major pathophysiological changes associ- den donor instability or cardiopulmonary arrest. The ated with brainstem death may cause organ injury. inferior mesenteric vein (IMV) is then prepared if With this later approach, an en bloc technique, as com- it is to be used for perfusion of the portal system. monly used in DCD donation (see next section), is In combined liver–pancreas procurement, IMV can- usually employed for organ extraction, reserving deli- nulation is usually avoided, as the increased portal cate hilar dissection to the back table, thus minimiz- pressure can cause pancreatic edema and compro- ing risk of injury. There is little robust evidence to mise perfusion. In this scenario, early division and support either approach, and surgeons frequently vary direct cannulation of the portal vein during cold dis- the extent of warm dissection undertaken depending section is performed instead, allowing both portal on factors such as donor instability, aberrant donor perfusion of the liver and adequate drainage of the anatomy, and donor obesity. pancreas. A midline laparotomy and sternotomy is per- The liver is mobilized by dividing the falciform formed and can be extended using transverse abdom- and left triangular ligaments. The lesser omentum is inal incisions to maximize surgical access. A full then divided, taking care to exclude a left replaced exploratory laparotomy is undertaken to exclude intra- or accessory hepatic artery, present in approximately abdominal sepsis or malignancy, and any suspicious 10% of donors. If identified, the left accessory artery lesions are biopsied and sent for urgent histological is preserved by dissecting the left gastric artery, from analysis. Early assessment of liver quality is performed, which it arises, from the lesser curve of the stom- in particular the degree of steatosis and, in cases of ach, leaving it and the common hepatic artery (if donor death secondary to trauma, exclusion of liver present) with a common inflow at the celiac trunk injury. (Figure 23.1). Dissection commences with mobilization of the The liver hilum is dissected to identify struc- right colon, duodenum, and small bowel leftward and tures of the portal triad and exclude or preserve a superiorly (the Cattell-Braasch maneuver) to expose replaced or accessory right hepatic artery. The com- aorta and inferior vena cava (IVC). The aorta is encir- mon hepatic artery is traced from the origin of the

191 Section 4: Liver

gastroduodenal artery (GDA) to the celiac axis, where Donor hepatectomy in donors after the splenic artery is identified. The common bile duct (CBD) is found on the right edge of the portal triad, cardiac death ligated distally, and divided. Hilar dissection above the In DCD retrieval, a rapid technique after immediate level of the GDA and extensive peri-ductal dissection perfusion and cooling is employed. After confirma- is avoided to protect the fragile blood supply to the tion of death and a stand-off period, usually 5 min- CBD. utes, rapid laparotomy and iliac artery or aortic can- The portal vein is then encircled and adjacent tis- nulation and perfusion is performed. Instillation of sue carefully inspected for the presence of a right crushed ice, thoracotomy for atrial venting and aor- accessory/replaced hepatic artery arising from the tic cross clamp, division and direct cannulation of origin of the superior mesenteric artery (SMA; the portal vein, and division and saline flushing of Figure 23.1). If present, the vessel must be traced along CBD must follow in rapid succession. After adequate its course behind or within the pancreas in preparation perfusion, donor hepatectomy is most rapidly and for careful preservation during hepatectomy. Extensive safelyperformedenblocwiththepancreas,assum- dissection at this stage, particularly intra-pancreatic, ing the presence of accessory hepatic arteries and is often best avoided by excising a “cuff” of pancreatic avoidingtheneedforanyfurtherinsituhilardissec- parenchyma during the cold dissection (to be removed tion. The major attachments of the liver are divided on the bench) or in the case of combined pancreatic as described for DBD donors, ensuring the left gas- retrieval, removal of liver and pancreas en bloc for back tric artery is dissected away from stomach and pre- table separation. This approach can also be used in the served with the organ block. The anterior surface of rare anatomical variant in which the common hepatic thepancreasisexposedbydividingthegreateromen- artery arises from the SMA. tum, short gastric vessels, the transverse mesocolon, Preparation for cold-phase hepatectomy is comple- and the root of the small bowel mesentery. The head ted by slinging the supraceliac aorta behind the divided of the pancreas is mobilized by full Kocherization right crus, systemic heparinization of the donor, and and stapled transection of the pylorus and proximal cannulation of aorta and IMV. The supraceliac aorta is jejunum. The tail of the pancreas is elevated medially then clamped, perfusion with preservation fluid com- using the spleen as a “handle” or can be transected if menced, venting performed via atrium or IVC, and not for transplantation. The resection is completed by crushed ice instilled into the peritoneal cavity. A total dividing the SMA flush with the aorta and continu- volume of 5–6 liters of preservation solution is infused ing superiorly to create an aortic patch at the celiac via portal and systemic cannulae, aiming to observe ostium. clearing of the vented effluent. It is very important that the cold-phase hepatectomy is timely to minimize the Benchwork preparation of the liver interval between core cooling with preservation fluid infusion and effective external cooling in the organ for transplantation transport carrier. Mobilization should therefore be Preparation of the liver for transplantation is usu- started during perfusion by dividing the suprahepatic ally performed following a period of efficient cool- IVCandthediaphragmwellclearoftheligamentous ing in an ice box and transportation (if required) to attachments of the liver to release the right lobe. On the recipient center. The cold ischemic time (period completion of perfusion, the portal vein is transected between cold perfusion and reperfusion with recipi- and the hepatic artery dissected back to the aorta leav- ent blood) should not be more than 16 hours and ide- ing a patch around the celiac trunk, preserving the left ally less than 12 hours. When possible, the recipient gastric in the case of a left accessory artery. A right hepatectomy should be commenced simultaneously accessory artery is preserved as described above, usu- with liver procurement and transportation to allow ally maintaining a stump of SMA, although a common much shorter cold ischemic times. This is particu- patch with the celiac trunk can also be used. The donor larly pertinent in the context of DCD donation, where hepatectomy is completed by dividing the infrahepatic detrimentaleffectsofwarmandcoldischemiaare IVC above the renal veins and transecting the adherent potentially compounding. Prior to implantation, the right adrenal gland, thereby avoiding capsular injury IVC is dissected clear of surrounding tissue ensuring to the liver. that both atrial and diaphragmatic muscle is removed,

192 Chapter 23: Surgical procedure as fibrotic changes within this may contribute to keeping peri-ductal dissection to a minimum. The post-implantation caval stenosis. Care must also be portal vein is then encircled and remaining peri- taken to avoid injury to the main hepatic veins, which portal tissue (predominantly lymphatics) divided, can have a significant extra-parenchymal component, ensuring that an accessory right hepatic artery is not and to ligate phrenic veins, which are a troublesome missed.Theinfrahepaticvenacavaisthendissected source of bleeding on reperfusion. The portal vein and and encircled, often requiring division of the right hepatic artery are dissected free of surrounding tissue adrenal vein. and reconstructed if required (discussed below); the Mobilization of the liver continues with division bile duct is left undisturbed. At this stage, reduction of the right triangular ligament and separation of the or splitting of the graft for two recipients can be per- bare area of the liver from the retroperitoneum, expos- formed. The technical details of split liver transplanta- ing the retrohepatic IVC on the right side. The ongo- tion for both adult and pediatric grafts are discussed in ing dissection around the IVC then depends on the Chapters 22 and 26. elected management of graft caval outflow, which can be approached with three different techniques: “classi- Recipient procedure cal” caval replacement, “piggyback,” or cavocavoplasty (Figure 23.2). In a “classical” liver transplant, com- Recipient hepatectomy and caval plete retrocaval dissection is performed and the supra and infra-hepatic vena cava cross-clamped, allowing anastomosis resection of the liver and intrahepatic vena cava en The preoperative preparation of potential recipients bloc (Figure 23.2A).Themaindisadvantageofthis for liver transplantation is discussed in detail in Chap- technique is the marked decrease in venous return ter 24. Once under anesthesia, central and periph- during the anhepatic phase, which can lead to pro- eral venous access is obtained and arterial cannu- found hemodynamic instability. It is therefore essen- lation and catheterization performed. The patient is tial that a “test” cross-clamp be performed prior to draped, leaving access to chest, abdomen, and groins, hepatectomy. If instability is observed or expected, and a cell-saver system is prepared. An inverted “L,” venovenous bypass is instigated via open cannula- Mercedes Benz, or rooftop incision is made in the tion of the femoral vein (usually via the saphenous right upper quadrant, and an exploratory laparotomy vein)inthegroinconnectedbyarollerpump–driven is performed. The recipient hepatectomy is frequently circuit to an internal jugular or subclavian line in the most difficult part of the transplant procedure, the neck, with or without a combined portal bypass described by Starzl as “the bloodiest and most stress- limb. In early canine and human liver transplanta- ful experience in a surgeon’s life.” Intra-abdominal tion, with the absence of experienced anesthetic care, varices can be extremely large, friable, and associ- the introduction of bypass massively reduced oper- ated with torrential hemorrhage, which can start with ative mortality. However, it is now recognized that cavernous cutaneous tributaries at the initial skin the majority of patients can tolerate caval cross-clamp incision. The surgeon must often strike a difficult without bypass, avoiding the associated risks of vas- balance between acceptable control of hemorrhage cular injury, pulmonary thromboembolism, and air and adequate rate of progression through the pro- embolus. cedure, in the knowledge that both variceal dilata- The “piggyback” and cavocavoplasty techniques tion and coagulopathy will rapidly improve following almost completely obviate the need for bypass, as implantation. the liver is dissected from the vena cava, leaving the The hepatectomy begins with division of the left tri- latter in situ without the need for cross-clamping angular ligament, falciform ligament, and lesser omen- and interruption of venous return. This technically tum before moving to the hilum. In direct contrast challenging maneuver, first performed by Calne in to the donor hepatectomy, the dissection is focused Cambridge, United Kingdom, involves careful sepa- high up in the hilum to obtain maximal length of ration of liver and vena cava starting caudally with vessels for anastomosis. The common hepatic artery the division of the numerous tributaries draining the is ligated and divided at its bifurcation, as this can caudate lobe until only the three main hepatic veins often be split to form a patch for the arterial anas- remain. In the “piggyback” technique, the three veins tomosis. The CBD is divided at a similar level, again are cross-clamped and, after excision of the liver,

193 Section 4: Liver

(A) (B)

Figure 23.2 Techniques for caval anastomosis. (A) The classical caval replacement involves complete excision of recipient intrahepatic vena cava en bloc with the explanted liver, which is then replaced by donor cava using upper and lower end-to-end anastomoses. (B) The “piggy back” technique requires preservation of recipient vena cava and cross-clamping of the main hepatic veins, which are opened into a single orifice for anastomosis to the top end of the donor vena cava. (C) The cavocavoplasty is performed by joining donor and preserved recipient vena cavae via longitudinal venotomies following closure of both ends of the donor vena cava.

(C)

opened into a single orifice for anastomosis to the date lobe hypertrophy, the risks associated with caval top end of the donor vena cava (the bottom end preservation may outweigh the potential benefits. being closed off; Figure 23.2B). In a cavocavoplasty, the recipient hepatic veins are stapled or sutured Portal venous anastomosis closed at hepatectomy, as are the ends of the donor On completion of the caval anastomosis, the por- vena cava, and donor and recipient vena cavae are tal venous and hepatic arterial anastomoses are per- anastomosed together via symmetrical longitudinal formed. It is common practice to complete the por- venotomies (Figure 23.2C). Despite the advantages of tal anastomosis first and reperfuse the liver prior to the cava-preserving techniques, the classical approach the arterial anastomosis in order to minimize sec- remains very useful, particularly in offering maximal ondary warm ischemic time (the time in which the tumor clearance in transplantation for hepatocellular liver is removed from ice but not yet reperfused). How- carcinoma and rapidity of hepatectomy in the unstable ever, these anastomoses can be performed in reverse patient. Furthermore, in severely distorted cirrhotic or both completed prior to reperfusion, the argument livers, particularly with significant compensatory cau- being that portal perfusion alone may allow organ

194 Chapter 23: Surgical procedure

(A) (B)

(C) (D)

Figure 23.3 Techniques for portal inflow to the transplanted liver. (A) The simple portal vein anastomosis is performed with a continuous suture, and it is crucial that a “growth factor” is left to allow for significant expansion at the anastomosis on reperfusion. (B) In cases of portal vein thrombosis with minimal SMV involvement, an anti-pancreatic jump graft between SMV and donor portal vein using donor iliac vessels can be created. In cases of more extensive SMV involvement, a portocaval hemitransposition can be performed by end-to-side (C), or by end-to-end (D) anastomosis of recipient vena cava to donor portal vein.

warming in the context of suboptimal oxygenation, following reperfusion. If the vein is too short, exces- particularly of the biliary tree. The portal anastomo- sive tension causes tearing of the thin-walled vein, sis usually involves simple end-to-end apposition of and an overlong portal vein is prone to kinking and donor and recipient portal veins, although this must be thrombosis. It is also very important that on com- done with technical precision. The correct final length pletion of the continuous suturing for the anasto- of the portal vein is essential, and accurate estima- mosis, the knot is tied several centimeters from the tion is facilitated by placing packs between the unper- vein to allow a “growth factor” on reperfusion and fused liver and diaphragm during the anastomosis, prevent a dumbbell-shaped anastomotic narrowing thus simulating the increased size of the graft observed (Figure 23.3A). Alternatively, tying of the final knot

195 Section 4: Liver

can be delayed until after release of clamps and com- GDA. In the recipient, branch patches using the hep- plete filling of the vein. aticartery bifurcationoratoriginoftheGDAare most The greatest challenge when performing the por- frequently employed. In the situation in which poor tal venous anastomosis is the presence of portal vein inflow or vascular disease prevents use of a GDA patch thrombosis (PVT), which was originally an absolute in the recipient hepatic artery, more proximal explo- contraindication to liver transplantation, but is now ration along the common hepatic artery is usually not part of standard practice. Despite advances in surgi- helpful. The best option in such cases is formation of cal techniques, PVT at transplantation is still associ- an aortic conduit running from an anastomosis on the atedwithahighriskofre-thrombosis,graftloss,and front of the infra-renal aorta, via a carefully created mortality. The difficulty in creating adequate portal retropancreatic tunnel, to the graft hepatic arteryFig- ( inflow is compounded by severe portal hypertension ure 23.4A). The ideal conduit is usually both donor and variceal formation, marked peri-portal fibrosis, iliac vessels joined end-to-end or a segment of donor and previous porto-systemic bypass procedures that thoracic aorta, avoiding prosthetic grafts as far as frequently require reversal. The extent of PVT governs possible. the management and has been classified by Yerdel into In the presence of variations in donor arterial fourgrades.Grade1and2PVTdescribesthrombus anatomy, a right accessory artery is most commonly in the portal vein with no more than minimal involve- prepared for implantation by anastomosing a short ment of the superior mesenteric vein (SMV), causing trunk of SMA, with right accessory artery attached, less than or greater than 50% occlusion, respectively. In between the celiac trunk and the recipient hepatic the majority of these cases, thrombectomy and/or exci- artery (Figure 23.4B). Alternatively, the accessory sion of the affected vein with primary anastomosis to artery can be anastomosed to a branch on the com- distal portal vein or proximal SMV can be performed. mon hepatic trunk (usually the GDA or splenic artery), A similar approach can on occasion be applied to grade again creating a single anastomosis to the recipient 3 PVT (complete PV occlusion extending to the prox- artery (Figure 23.4C). On occasion, the celiac trunk imal SMV), although exploration and thrombectomy and SMA (with right accessory artery) can be kept behindthepancreasshouldbeavoidedduetoriskof on a single aortic patch and anastomosed together, significant bleeding, pancreatic injury, and pancreati- particularly when employing an aortic conduit. A left tis. In such cases, drainage of the SMV proximal to accessoryhepaticarteryisusuallyeasilymanaged,pro- the occlusion is best achieved by using a jump graft viding the liver has been retrieved correctly. In most to the donor portal vein fashioned using donor iliac cases, anastomosis using the donor celiac patch is per- vessels (Figure 23.3B). In grade 4 PVT (extensive SMV formed allowing inflow into both the main hepatic thrombosis) and grade 3 PVT in which a bypass is artery and left hepatic artery via the left gastric with- not possible, portal inflow can sometimes be derived out need for reconstruction (Figure 23.4D). In uncom- from large collateral vessels and donor portal anasto- mon cases when the celiac trunk is unusable or the mosis to left gastric, right gastroepiploic, splenic, right left accessory has been damaged beyond simple repair, or middle colic, pericholedochal, and renal veins have reconstruction is performed by joining the end of the been described. In the absence of adequate collaterals, accessoryoradjoiningleftgastrictoanotherceliac the options remaining are cavoportal hemitransposi- artery branch (e.g., splenic artery or GDA). tion, in which the lower end of the transected recipient IVC is anastomosed to the graft portal vein instead of the lower vena cava, and multi-visceral transplantation (Figures 23.3Cand3D). Reperfusion and bile duct anastomosis Reperfusion of the liver is often the most danger- ous part of the transplant procedure, and close com- Hepatic artery anastomosis munication between surgeon and anesthetist is cru- The dominant recipient hepatic artery, whether com- cial. Prior to reperfusion, it is essential that the liver mon or accessory, is prepared and used for the arterial is flushed of potassium-rich preservation fluid to anastomosis. In the donor artery, the common anas- avoid significant risk of arrhythmia or cardiac arrest. tomotic sites are the origin of the celiac trunk with This is performed using a cold colloid infusion via aortic patch or branch patches using the left gastric or the portal vein just before completion of the caval

196 Chapter 23: Surgical procedure

(A) (B)

(C) (D)

Figure 23.4 Techniques for hepatic arterial anastomosis. (A) In the absence of adequate inflow from the recipient coeliac axis, a conduit from the infrarenal aorta to the donor hepatic artery can be created, most commonly using both donor iliac arteries anastomosed together. In the presence of a right accessory artery an attached segment of SMA can be anastomosed to the coeliac aortic patch (B), or the accessory vessel can be anastomosed to a branch of the celiac trunk (the GDA shown here) (C). The left accessory artery is perfused by preserving the left gastric artery from which it arises. (D)

197 Section 4: Liver

anastomosis and/or using a blood flush on reperfu- duct size discrepancy, and concern over recipient duct sion, allowing loss of 300–500 ml of blood via the caval viability. anastomosis prior to closure and release of caval The transplant is completed by thorough hemosta- clamps. Despite this precaution, profound hypoten- sis, confirmation of adequate graft perfusion, place- sion may occur on reperfusion due to the combination ment of drains to the hilum and suprahepatic space, of a marked systemic ischemia–reperfusion response and closure. The patient is then transferred to inten- and significant blood loss. sive care or high-dependency setting for close postop- Once stability of the patient and meticulous erative monitoring. hemostasis has been established, occasionally tak- ing several hours, the bile duct anastomosis is per- formed. The standard technique involves apposition Further reading of donor and recipient ducts with minimal prepara- Busuttil R, Klintmalm G. Transplantation of the Liver. tory dissection to preserve blood supply. The supply Philadelphia, PA: Saunders, 2005. tothecutendscomesfromthreefinevesselsrunning Calne R. Liver Transplantation: the Cambridge-King’s longitudinally along the duct surface, and so inter- College Hospital Experience. (2nd Edition). Orlando, FL: rupted suture placement is used to minimize its dis- Grune and Stratton, 1987. ruption, aiming to avoid local ischemia and anasto- Humar A, Matas A, Payne W. Atlas of Organ moticdehiscenceorstenosis.Alternatively,thedonor Transplantation. London: Springer, 2006. bile duct can be anastomosed to a Roux-en-Y loop Molmenti E, Klintmalm G. Atlas of Liver Transplantation. of jejunum via a small enterotomy, again using inter- Philadelphia, PA: Saunders, 2002. rupted suturing. The most common indications for the Starzl T. The Puzzle People: Memoirs of a Transplant latter technique are transplantation for primary scle- Surgeon. Pittsburgh, PA: University of Pittsburgh Press, rosing cholangitis, re-transplantation, donor/recipient 1992.

198 Section 4 Liver Chapter Peri-operative care and early complications

24 John Klinck and Andrew J. Butler

Key points peritonitis or pneumonia. Common pathophysiolog- ical disturbances and their management in the peri- r Severe liver disease affects all organ systems, operative period are summarized in Table 24.1. and comorbidity is common. r The transplant procedure is marked by cardiovascular instability, major hemorrhage, Cardiovascular and important electrolyte, acid–base, and Cardiovascular function in liver failure is char- hemostatic disturbances. acterized by a disturbance of microcirculatory r Marginal grafts are increasingly used, function causing arteriovenous shunting, increased increasing the risk of postoperative cardiac output, and abnormal blood volume distribu- complications. tion, in proportion to the severity of the underlying r Bleeding is usually due to transections in the hepatic disease. Central blood volume is reduced, complex mesh of portosystemic collateral inducing increased sympathetic and hormonal vaso- veins; therefore, fluid management, portal constrictor activity. Despite this, splanchnic blood hyperemia, and blood loss may be linked. volume and flow are increased. Subtle functional r The most important early complications are andstructuralchangesoccurintheheart,nowdes- primary non-function, hepatic artery cribed as cirrhotic cardiomyopathy. These include thrombosis, and bleeding. impaired responses to increased preload and afterload and conduction abnormalities. Left atrial enlarge- ment, mild left ventricular (LV) hypertrophy, and The peri-operative and early postoperative manage- diastolic dysfunction are common echocardiographic ment of the liver transplant recipient presents a findings. formidable challenge. This chapter outlines the patho- Patients with alcoholic cirrhosis, amyloidosis, or physiology of liver disease as it affects patient selection Wilson’s disease may have overt cardiomyopathy. and management in the peri-operative period and key Any evidence of significant cardiac disease must aspects of anesthetic, surgical, and early postoperative be taken seriously in view of the major insults care. imposed during and after surgery. Patients with known ischemic heart disease (IHD) or peripheral or cerebrovascular disease should be considered for Preoperative assessment coronary angiography. Significant coronary disease and/or global LV dysfunction probably contraindicate Pathophysiology and relevant comorbidity transplant. Chronic liver disease is associated with portal hyper- The approach to patients without known IHD but tension, impaired hepatic synthetic function, mal- with risk factors (age Ͼ 55 years, diabetes, hyperten- nutrition, peripheral and pulmonary microvascu- sion, smoking, obesity) is more controversial. Non- lar shunting, renal impairment, and encephalopathy. invasive stress imaging (dobutamine stress echo or Life-threatening complications are common, includ- myocardial perfusion scan) followed by coronary ing variceal hemorrhage and sepsis from bacterial angiography is costly and will identify occult disease

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

199 Section 4: Liver

Table 24.1 Preoperative disorders and peri-operative management in the liver transplant recipient System Disorder Peri-operative management

Cardiovascular: systemic Increased cardiac index Balance preload and vasoconstrictors (see text) and splanchnic Reduced central and increased splanchnic If renal or cardiac dysfunction consider piggyback or blood volumes veno-venous bypass Impaired diastolic function TEE Cardiomyopathy (esp. alcohol, amyloid, Pacing wire if amyloid polyneuropathy Wilson’s, hemochromatosis) Autonomic neuropathy (mild in cirrhosis, marked in amyloid) Cardiovascular: pulmonary Portopulmonary hypertension (PA mean Ͼ 25, Preop right heart catheter if Doppler PA sys Ͼ 40 (to PVR Ͼ 250) differentiate from high-flow state and overload) Defer transplant and treat if PA mean Ͼ 35 or RV dilated or impaired Intra-operative PA catheter (± TEE) essential if pulmonary hypertension suspected

Respiratory Restrictive defect (ascites) FiO2 ≥ 0.5 Pleural effusion Drain large effusion early intra-operatively (beware Flow-related or anatomical intrapulmonary re-expansion pulmonary edema, especially at shunting (hepatopulmonary syndrome) reperfusion) Non-cardiogenic pulmonary edema (fulminant Cautious use of PEEP hepatic failure) Obstructive airways disease (esp. cystic fibrosis, ␣1-antitrypsin deficiency) Interstitial lung disease (primary biliary cirrhosis). Renal Hepatorenal syndrome (prerenal failure from Preoperative hemodiafiltration if K+ Ͼ 5.5 splanchnic “steal”) Maintain arterial pressure Acute tubular necrosis from sepsis Consider intraoperative pressor (norepinephrine or Tacrolimus/cyclosporine-related renal vasopressin) impairment Anhepatic, mannitol Renal tubular acidosis Electrolytes/metabolic Hyponatremia, hypomagnesemia, Defer transplant if high surgical risk and Na Ͻ 122; hyperkalemia, metabolic acidosis, and treat hyperkalemia if preanhepatic Ͼ 5.0 or rapid hypoglycemia in fulminant liver failure anhepatic rise; – MgSO4 if any arrhythmia; consider tris-buffer, hemodiafiltration if acidosis severe; close monitoring and treatment of hypo/hyperglycemia Hematological/coagulation Reduced/defective synthesis of Vitamin Consider prophylactic tranexamic acid if high risk and K–dependent clotting factors no prothrombotic history Low-grade DIC ± hyperfibrinolysis Assess coagulation clinically before treatment Anemia, thrombocytopenia (hypersplenism and (cannulation sites, surgical field): treat clinical marrow depression) coagulopathy according to TEG and laboratory data Platelet dysfunction (blood products, antifibrinolytic, protamine). Note impaired synthesis of anticoagulant Maintain normothermia. factors may preserve hemostasis If intractable coagulopathy consult hematologist and consider FVIIa by local protocol. Central nervous system Encephalopathy Avoid/minimize benzodiazepines; in fulminant liver Cerebral edema failure with grade III/IV encephalopathy consider ICP monitoring and maintain cerebral perfusion pressure Ͼ60 mmHg (norepinephrine), ± mannitol/ hypertonic saline/thiopental to control ICP. DIC = disseminated intravascular coagulation; ICP = intracranial pressure.

in many (5–27%). However, few patients are suitable Like IHD, valvular aortic stenosis is increasingly for revascularization, and there is no evidence of better seen in older patients referred for liver transplanta- outcomes. Screening on the basis of clinical assessment tion and remains an important cause of peri-operative (including exercise tolerance) and resting echocardio- death. Mild functional mitral, tricuspid, and aortic graphy alone is more cost-effective but will occasion- regurgitation are frequently found on echocardiogra- ally miss occult disease that (rarely) might have been phy but are usually attributable to the hyperdynamic treated. state.

200 Chapter 24: Peri-operative care and early complications

Pulmonary adults presenting for liver transplant. Recurrent chest infections may be a feature, but history and clini- Pulmonaryhypertensionisseeninupto20%ofadult cal signs are often minimal. Flow-volume measure- liver transplant candidates and is usually identified ments show an obstructive pattern, often mixed with by transthoracic echocardiography. In most cases it the restriction of vital capacity seen in massive ascites. is caused by LV overload (high cardiac output in the Therearenodataonitseffectonlivertransplantout- presence of diastolic dysfunction) and is not true por- come. Forced expiratory volumes in 1 second (FEV ) topulmonary hypertension. This condition is patho- 1 as low as 30% of predicted are associated with good logically indistinguishable from primary pulmonary outcomes in young recipients with cystic fibrosis (CF), hypertensionandoccursinlessthan4%ofadultcan- but in older patients, values below 40–50% in the didates. Peak echo-Doppler pulmonary artery (PA) absence of pleural effusion or ascites require a cautious pressure greater than 40 mmHg should be investi- approach. gated by PA catheterization to measure mean PA pres- Other causes of respiratory impairment include sure, pulmonary capillary wedge pressure (PCWP), ascites and hydrothorax, poorly controlled asthma, and pulmonary vascular resistance (PVR). Mean PA aspiration, and pulmonary infection. Primary biliary pressure greater than 35 mmHg with a high PVR cirrhosis is occasionally associated with interstitial (Ͼ240 dyn s/cm5) has a reported 50% peri-operative lung disease, which causes disproportionate dyspnea mortality, approaching 100% if greater than 50 mmHg and a restrictive defect on pulmonary function testing, or if associated with right ventricular (RV) dys- Coagulation defects are often identified in lab- function. Effective treatments are available (silde- oratory screening tests, and reduced concentrations nafil, bosentan, iloprost) and appear to reduce peri- of Vitamin K–dependent clotting proteins are well- operative risk. In most patients, pulmonary hyperten- documented. However, liver disease has complex sion resolves after transplant. effects on hemostasis and on the balance between Hepatopulmonary syndrome is characterized by pro- and anti-hemostatic processes, and these mea- marked pulmonary shunting and orthostatic arterial surements may not predict bleeding. Prothrombin desaturation (orthodeoxia) in the absence of intrinsic time (PT) or international normalized ratio, activated pulmonary disease. It is associated with portal hyper- partial thromboplastin time, and platelet count all tension and presents a spectrum of severity, affecting highlight procoagulant functions, whereas the effects up to 33% of transplant candidates, about 4% severely. of anti-coagulant proteins, such as protein C and The most common pathophysiological feature is the anti-thrombin, are not measured. Similarly, although “perfusion–diffusion” defect of high flow through dif- platelet numbers and function may be reduced, high fusely dilated pulmonary capillaries. Arteriovenous levels of factor VIII and von Willebrand factor in cir- malformations identifiable on computed tomography rhosis may compensate. The fibrinolytic system is also (CT) pulmonary angiogram are an uncommon cause. affected, as both pro- and anti-fibrinolytic factors may Contrast echocardiography is a sensitive screening be decreased in cirrhosis. The net effect of these imbal- tool and also serves to exclude right-to-left shunt at the ancesisunpredictable,andsensitivityofthecoagula- atrial level. A preoperative PaO less than 50 mmHg or 2 tion system to factors promoting both bleeding and radiolabeled macroaggregated albumin uptake in the thrombosis is probably increased. Routine preopera- brain greater than 30% are associated with increased tive administration of fresh-frozen plasma, platelets, 90-day mortality, although this is not prohibitive. and other products is no longer commonly practiced, In our unit, oxygen dependency, age greater than and treatment may be better based on clinical findings, 50 years, and the presence of other major comorbidity such as bleeding from puncture sites, oozing in the would probably contraindicate transplant. Although operative field, or failure of shed blood to form clots. intra-operative problems with oxygenation are rarely observed, weaning from mechanical ventilation may be prolonged, and these patients are more vulnerable Hematological to sepsis and other post-transplant complications. Anemia is common, from impaired hematopoiesis, Obstructive airways disease, usually related to gastrointestinal bleeding, or hypersplenism. Varices smoking but occasionally to familial emphysema or shouldbetreatedandovertirondeficiencycorrected, alpha-1 antitrypsin deficiency, is seen in up to 18% of but transfusion is only carried out for active bleeding

201 Section 4: Liver

or in the presence of symptoms clearly attributable to ated catabolism during infection and surgery. Poorer low hemoglobin. outcomes are seen in low body mass index (BMI) patients, and most centers would not list at BMI less Metabolic than 15. BMIs above this value but with low mea- sured nutritional indices (e.g., grip strength, triceps Hyponatremia may be caused by diuretic therapy, sec- skin-fold thickness, mid-upper arm circumference ondary hyperaldosteronism, and other poorly under- Ͻ fifth percentile) are common but are not regarded stood renal abnormalities. It predicts worse outcome, as contraindications to transplant. BMI less than 18.5 with or without transplant. Rapid increases in plasma associated with other major comorbidities would be sodium in the peri-operative period, associated with regarded as potentially prohibitive. Calorie, protein, transfusion, carry a significant risk of central pon- and vitamin supplementation is essential. tine myelinolysis and long-term neurological disabil- Morbid obesity (BMI Ͼ 35) is increasingly com- ity. Many centers suspend patients from the transplant moninpatientspresentingforlivertransplantation. waiting list when values fall below 122–125 mmol/l Itisknowntoaffectintensivecareunitlengthof Ͻ and attempt correction. Patients with Na 122 mmol/l stay and long-term outcomes but is not regarded as are considered on an individual basis according to prohibitiveinmostcenters.Anabdominal(“apple- the risks of major operative blood loss. Factors such shaped”) pattern of obesity affects surgical access, peri- as anemia, re-transplant, portal vein thrombosis, and operative respiratory parameters, and ventilator wean- extended criteria donor may favor deferral in the ing much more than a pelvic (“pear-shaped”) pat- patient with plasma sodium below this threshold. tern, although this has not been formally investigated. Preoperative hyperkalemia is uncommon but Marked abdominal obesity (BMI Ͼ40) in the presence should be treated aggressively, since fatal intra- of one or more major comorbidities would contraindi- operative hyperkalemic arrest, usually at reperfusion cate transplant in many units. of the liver, still occurs. It may be associated with renal failure, treatment with spironolactone, or trans- fusion. Preoperative values above 5.5 mmol/l should Acute liver failure with multi-organ failure be treated with hemofiltration, continued into the Special problems are encountered in patients with intra-operative phase if it does not fall to less than multi-system failure caused by fulminant hepatic fail- 5.0 mmol/l. ure, primary non-function of a liver graft, or terminal decompensation in chronic liver failure. These patients Renal usually have severe coagulopathy and encephalopa- Renal impairment develops easily because of underly- thy. Non-cardiogenic pulmonary edema, renal insuf- ing circulatory and hormonal disturbances. Overtreat- ficiency, and/or sepsis may also be present. Raised ment with diuretics is a common cause, and acute intracranial pressure is commonly seen in fulminant tubular necrosis may be seen in patients with acute hepatic failure with encephalopathy and may rapidly hepatic failure or sepsis associated with spontaneous progress to fatal brainstem compression if untreated. bacterial peritonitis or chest infection. Hepatorenal Intracranial pressure monitoring is advocated in this syndrome, a form of prerenal failure occurring in the setting, although hemorrhagic complications, occa- absence of clinical hypovolemia or intrinsic renal dis- sionally fatal, occur, and no consensus on the bal- ease, is also common. This may respond to vasopressor ance of risks has emerged. Paralysis, ventilation, and treatment combined with volume loading and resolves 30-degree head-up tilt are essential, supplemented after transplantation. Renal impairment is a strong by mannitol and pentobarbital if needed. Moderate predictor of increased mortality. hypothermia is now used in many units to control severe intracranial hypertension. Continuous veno- venous hemofiltration may be needed to correct hyper- Nutritional kalemia and control metabolic acidosis. Nutritional impairment is severe in up to 30% Sepsis may be difficult to diagnose in this context, of patients, resulting from anorexia, malabsorption, as many of its clinical manifestations are mimicked impaired protein synthesis, and chronic catabolism. in terminal hepatic decompensation. However, fever, Nutritional reserve is further depleted by acceler- hypotension, and dependence on vasopressors clearly

202 Chapter 24: Peri-operative care and early complications suggest its presence, and the risks of proceeding with venous bypass circuit (see below). The femoral route transplantation at this stage may be prohibitive. is usually avoided because the cava is clamped intra- operatively, and these vessels may be needed for ven- Intra-operative care ovenous bypass. Subclavian cannulation has a higher risk of arterial hematoma and pneumothorax and in Immediate preoperative preparation our unit is now performed only under ultrasound and fluoroscopic guidance in an angiography suite. Full multi-system assessment should be performed before listing for transplantation, and the patient reviewed when a donor liver becomes available. It Monitoring is important to exclude untreated infection (e.g., Radial or femoral arterial and central venous pres- incipient bacterial peritonitis or bronchopneumonia), sure monitoring are essential, whereas PA catheters hyperkalemia, and severe hyponatremia. If unusual are used routinely in most centers. Radial artery pres- risks are identified, a discussion between the attending sure monitoring may underestimate aortic pressure anesthesiologist, hepatologist, and surgeon is essen- in hypotensive states, especially when vasopressors tial. Although cold ischemia times should be kept be are used, and should be interpreted with caution. PA as short as possible, retrieval times and liver preser- pressure and thermodilution cardiac output measure- vation techniques typically allow adequate time for ments help in the assessment and management of assessment and intervention. The patient’s usual pro- hypotension, which may arise unpredictably because ton pump inhibitor should be given, along with antibi- of changes in venous return, altered systemic vascu- otic prophylaxis according to local protocol. lar resistance (SVR), and cardiac or embolic events. A Ten units of blood are routinely cross-matched, PA catheter is essential if pulmonary hypertension is and at least 20 additional group-specific units should present or suspected. be available if needed. Predictors of transfusion Cardiovascular monitoring is further enhanced by requirement include preoperative anemia, renal the use of transesophageal echocardiography (TEE), impairment, and previous transplant. Fresh-frozen which gives continuous if subjective information on plasma and platelets must be available because of the ventricular function and an immediate diagnosis of possibility of coagulopathy during surgery. embolization of air or thrombus. A history of variceal bleeding is a contraindication for TEE. Anesthetic technique Measurement of arterial blood gases, sodium, Careful preoxygenation and a rapid-sequence induc- potassium, glucose, lactate, ionized calcium, and tion (or modified) are recommended. A nasogastric hematocritshouldbeperformedatfrequentintervals, tube and urinary catheter are placed. Maintenance at least hourly during the initial and closing phases with fentanyl, remifentanil, or sufentanil given with of the operation and more often during the anhepatic air/oxygen/desflurane or isoflurane are common tech- period. niques. The choice of muscle relaxant is arbitrary, with Coagulation monitoring practices vary widely. In most units using atracurium or cis-atracurium. Short- most centers, routine coagulation screening tests, acting agents are advocated as many patients can be including PT, partial thromboplastin time, fibrinogen, awakened soon after surgery. and platelet count, are supplemented by thromboelas- tography (TEG). Despite poor agreement with con- ventional laboratory coagulation tests, TEG provides Vascular access a prompt global assessment of coagulation function, Large-bore intravenous access is vital. Two periph- and targeted sample treatment also allows demonstra- eral lines are dedicated to transfusion. A triple-lumen tion of heparin effect and platelet function. However, catheter and PAcatheter introducer can both be placed as outlined above, abnormal results in both conven- in the right internal jugular vein under ultrasound tional and TEG testing are poorly correlated with visu- guidance. It is our practice to place two 5F cannulae ally apparent coagulopathy (oozing in the surgical field alongside these in the right internal jugular vein. These and/or lack of visible clot), and their main value may may be used as reserve volume-infusion lines or re- be to indicate the appropriate treatment when coagu- wired to 10F size for use in the return limb of a veno- lopathy is a clinical problem. Bedside (point-of-care)

203 Section 4: Liver

monitors are also available but, again, often reflect When surgical bleeding occurs, many now accept a clinical coagulation poorly in the setting of liver hemoglobin transfusion threshold of 7.0 g/dl (hemat- disease. ocrit 20–22). Transfusion at a higher value may be pru- dent in renal impairment, given the sensitivity of renal oxygen consumption to hemoglobin concentration in Blood replacement and fluid management experimental models. Blood product use during liver transplantation has declinedoverthelast20years.Thecurrentmedian red cell requirement is to 2–5 units, with a substan- Management of coagulopathy tial proportion of recipients avoiding blood products New concepts in the interpretation of coagulation tests altogether. Refinement in surgical techniques, better in liver disease, outlined above, should be consid- understanding of hemostasis, and improved anesthetic ered in intra-operative management. Further changes management have all contributed. imposed by the procedure include dilution, pathologi- Because bleeding during liver transplantation is cal fibrinolysis, effects of synthetic colloids, and release not usually caused by problems with the major anas- of heparinoids and inflammatory mediators from the tomoses, but more often by transections in the com- graft. Prophylactic antifibrinolytic (tranexamic acid) is plex mesh of portosystemic collateral veins, fluid man- used selectively in many units. agement, portal hyperemia, and blood loss may be Treatment of established coagulopathy is guided linked. Patients with cirrhosis and portal hypertension by monitoring (see below). In the presence of fibri- have splanchnic hypervolemia, which volume load- nolysis or heparin effect, an anti-fibrinolytic or pro- ing increases. They also have smaller increases in car- tamine can be given. Otherwise, platelets, fresh-frozen diac output with fluid infusion, so the larger volumes plasma, and cryoprecipitate remain the mainstay of needed to improve systemic perfusion may have a dis- treatment, with frequent assessment of the surgical proportionate effect on portal venous pressure and field to minimize the number of units given. Although flow. Aggressive administration of crystalloid and col- data in liver recipients are limited, factor concentrates, loid may also have a more detrimental effect on clot- including fibrinogen and prothrombin preparations, ting, since factor levels are low. appear safe and effective and are increasingly used to Thus the conventional approach of administer- treat established coagulopathy. Normothermia and a ing blood products preemptively and optimizing car- pause in the procedure (after full reperfusion) to allow diac output by generous fluid loading has been ques- the liver to recover from surgical handling may also tioned. Indeed, fluid restriction with vasopressor infu- help. sion during the dissection and anhepatic phases has Recombinant factor VIIa (eptacog alpha activated) been associated with very low blood product use. On has controlled intractable bleeding in patients with the other hand, volume restriction requires liberal use complex acquired coagulation defects, but two ran- of vasopressors and risks systemic and especially renal domized trials have failed to show efficacy when the hypoperfusion. The optimal approach to volume man- drug is given prophylactically. There is no consen- agement remains to be determined. sus on the role of this costly treatment, and most The choice of fluid is controversial. Albumin centers use it on specialist advice according to local appears to be safe and has no effects on coagulation protocol. apart from dilution, but is costly. Gelatin solutions are A growing number of case reports indicate that inexpensive and have been used in high volumes in hypercoagulability and thromboembolism may cause liver recipients in Europe for many years, but are more serious or fatal complications during liver transplant. allergenic. Hydroxyethyl starch solution (HES 130/0.4) The incidence of pulmonary thromboembolism or has been reported to be safe in terms of effects on renal intracardiac thrombus formation has been estimated function and coagulation. This was well tolerated in at 1–1.5%. Hypercoagulability is demonstrable on a recent series of liver recipients, but effects of high- TEG and other tests in a significant number of liver infusion volumes are unknown. A balanced electrolyte recipients, particularly those with primary sclerosing preparation of this is under development, reduc- cholangitis and HCC, but its importance as a cause ing chloride ion load, which may reduce metabolic of intra-operative thrombosis is unclear. Thromboem- acidosis. bolism may occur at any stage of the procedure, and

204 Chapter 24: Peri-operative care and early complications mortality is high (up to 68% in one series). Aggressive ECG changes associated with hyperkalemia should be therapy with thrombectomy or thrombolysis has been treated with calcium chloride. successful in some cases. TEE should permit immedi- Trisodium citrate in transfused blood and plasma ate diagnosis. depresses ionized calcium, which should be main- Autotransfusion techniques are widely used and tained to preserve myocardial function. Calcium chlo- safe, but effects on coagulation and even on total use ride is given if hypotension occurs in the pres- of bank blood are still unproven. Cell-salvage may be ence of a depressed ionized calcium value. A less contraindicated in patients with hepatic malignancy marked effect is seen on magnesium concentrations, and when enteric contamination of the peritoneum although supplementation in the absence of either has occurred. arrhythmia or availability of rapid measurement is not routine. Metabolic acidosis is usually absent or minimal ini- Electrolyte and acid–base changes tially, unless there is renal impairment, but becomes The infusion of large volumes of blood products and increasingly apparent during the procedure. Trans- reperfusion of the donor liver cause marked changes in fused blood introduces a substantial quantity of exoge- plasma biochemistry. Plasma sodium is subnormal in nous lactic acid into the circulation, whereas liver many recipients and tends to increase during surgery. lactate and urea metabolism are impaired. Acid Risesofmorethan12mmol/lin24hourshavebeen metabolitesassociatedwithvenousstasisintheportal associated with pontine myelinolysis and neurological and lower body circulations, as well as those that accu- injury. In susceptible patients this may happen with mulate in the new liver during storage, are released smaller increases, and this has been well-described in into the general circulation on reperfusion, causing a liver recipients. Avoiding sodium bicarbonate, mini- further increase in acidosis. A poorly functioning liver mizing use of citrated blood products, and washing of fails to metabolize lactate, and worsening lactic acido- bank blood may attenuate this increase. sis is a typical sign of graft failure. Plasma potassium increases on reperfusion of the Acidosis is usually treated with modest hyperven- liver. Many anesthetists give 5–10 ml of calcium chlo- tilation. The place of sodium bicarbonate in the man- ride at reperfusion to reduce the risk of hyperkalemia- agement of acidosis remains controversial. Evidence induced ventricular tachycardia or fibrillation. In most that global circulatory function is impaired by mod- patients redistribution follows within seconds, and a erate metabolic acidosis is slight, whereas detrimen- progressive decrease is subsequently seen. However, taleffectsonoxygendeliveryandintracellularpH pre-existing renal failure, residual beta-blockade, and and plasma lactate associated with bicarbonate ther- a relatively large, fatty, or ischemic donor liver may be apy have been described. Alternative buffers produc- associated with prolonged and life-threatening hyper- ing less or no carbon dioxide, including dichloroac- kalemia. etate and tris-buffer, may be of value but remain to be Hyperkalemia may also complicate rapid transfu- fully assessed. sion, especially in the presence of renal impairment. If liver and cardiovascular function are adequate Plasma potassium concentration in stored blood after reperfusion, there is a tendency for metabolic aci- increases with storage time and may be greater than dosis to clear. Potassium may be needed in the early 20 mmol/l. Potassium concentration should be postoperative period when reuptake by the grafted checked frequently when transfusion is rapid and liver can cause hypokalemia donor units washed in the cell saver whenever hyper- kalemia is anticipated. An increase in potassium concentration of greater than 0.5 mmol/l between suc- Glucose control cessivemeasurementsoravalueabove5.0mmol/lat In most patients blood glucose is normal before anytimeshouldpromptwashingofallfurtherdonor operation but increases during the procedure units. Values above 5.0 mmol/l during the dissection because of administration of acid–citrate–dextrose or anhepatic phases should be treated aggressively blood and stress-related insulin resistance. Hypo- with furosemide and glucose-insulin. Nebulized or glycemia, although seen preoperatively in patients intravenous salbutamol and sodium bicarbonate with fulminant hepatic failure, is rarely observed should also be considered. As at reperfusion, any intra-operatively, even when normal hepatic glucose

205 Section 4: Liver

release is interrupted during the anhepatic and early are needed to maintain cardiac output, may result in reperfusion phases. Recent literature has highlighted high filling pressures following unclamping. This may the association between poor glucose control and have adverse effects on RV function, gas exchange, and adverse outcomes in the peri-operative and critical hepatic blood flow. care settings, but a clear causal relationship has not When a piggyback technique or venovenous been established, and hypoglycemia is a significant bypass is used, the decrease in cardiac output is usually hazard when tight control is attempted, especially less (20–30%). Arterial pressure is well maintained, in the unconscious patient. Insulin infusions have although this depends on the flows obtained through been advocated to moderate blood glucose values and the bypass circuit. Nonetheless, a progressive decline prevent hyperkalemia, ranging from 2–10 units per in cardiac output occurs during the anhepatic phase, hour. Blood glucose and electrolyte checks, at least and both bypass and the piggyback technique fall short hourly, are essential. of maintaining a normal circulatory state, reflected in worsening metabolic acidosis. Reperfusion of the transplanted liver is usually Cardiovascular changes associated with reduced heart rate, contractility, and Surgical bleeding presents the greatest threat, and it is peripheral vascular tone. Portal unclamping releases essential that blood lost can be replaced rapidly at any desaturated blood from the obstructed portal circu- point during the procedure (see Chapter 23). Cardiac lation, which mixes with cold, potassium-rich preser- filling may be impaired during the dissection phase vation fluid in the new liver and enters the systemic by intermittent obstruction of venous return during circulation. Slowing of the heart and hyperkalemia surgical manipulation of the liver or by direct com- are common, and asystole, tachyarrhythmia, and ven- pression of the diaphragmatic surface of the heart. tricular fibrillation are sometimes seen. Blood pres- Anesthetic agents and hypocalcemia will amplify these sure decreases in almost all patients because of arte- effects. Interpretation of filling pressures may be riolar vasodilatation and transient myocardial depres- difficult during the dissection phase, since caval sion. These changes may be caused by inflamma- obstruction is variable and may coexist with true hypo- tory mediators from the ischemic liver, by peptides volemia. Observation of the surgical field and commu- released during splanchnic stasis, or by reflex vasodi- nication with the surgeon are vital when there is uncer- latation.Inmostpatients,bloodpressureandcar- tainty about the cause of reduced venous return. The diac output are restored within minutes. In some, response to clamping of the inferior vena cava (IVC) typically those with a marginal graft or pre-existing forhepatectomyandtounclampingwhenthegrafted vasopressor dependency, recovery takes longer, and liver is reperfused depends on several factors. The most sustained vasopressor support is needed. This has important of these is whether the cava is side-clamped been described as the “post-reperfusion syndrome.” for a piggyback implantation or cross-clamped for the In patients undergoing urgent re-transplantation for conventional method. graft non-function or infarction, or in whom unrec- Cross-clamping of the cava at hepatectomy pro- ognized sepsis has supervened, hypotension and duces a marked (40–50%) decrease in central venous acidosis after unclamping may be progressive and pressure and cardiac output. SVR increases, but a irreversible. decrease in blood pressure is expected. Provided car- Transient pulmonary hypertension may be seen at diac filling pressures and contractility are maintained, reperfusion, as central blood volume increases. PCWP frank hypotension (systolic Ͻ 80 mmHg) is unusual. is raised and the TPG is normal (Ͻ15 mmHg). Pre- Although the need for veno-venous bypass is usu- existing pulmonary hypertension is usually diagnosed ally predetermined by other factors (discussed below), by echocardiography at the time of referral, but is occa- when its use is not planned, many teams perform a sionally found only on placement of the PA catheter trial clamping of the IVC to assess the patient’s abil- for transplant. If it is true portopulmonary hyperten- ity to maintain an adequate systemic blood pressure. sion (PVR Ͼ 240 dyn s/cm5,TPGϾ 15 mmHg), the Once the liver is removed, lower filling pressures are risks of proceeding need to be balanced against those accepted as long as the arterial blood pressure is sat- of deferring transplant in favor of treatment. If severe isfactory. Over-transfusion at this stage, more likely (mean PA pressure Ͼ 50 mmHg, or RV dysfunction), when bypass is not used since higher filling pressures peri-operative mortality approaches 100%, and the

206 Chapter 24: Peri-operative care and early complications operation should not proceed. If moderate (35– Respiratory function 50 mmHg), a trial of treatment with prostacyclin Changes in respiratory function caused by liver disease and/or inhaled nitric oxide is advocated, ideally mon- have been described above. Further respiratory prob- itored by TEE. It may be reasonable to proceed if lems arising during surgery are not common. Rapid right heart function is well maintained, especially if the arterial oxygen desaturation occurs easily after induc- patient responds to vasodilator therapy. tion of anesthesia, even after pre-oxygenation, since functional residual capacity is often reduced by ascites or hydrothorax. Pulmonary edema presents a more Veno-venous bypass important hazard. Over-transfusion can occur because of the highly variable rate of blood loss. Plasma oncotic Veno-venous bypass is a pumped extracorporeal shunt pressure is reduced and some patients appear to have taking blood from the femoral and (sometimes) portal abnormal vascular permeability, especially after reper- vein to the axillary or internal jugular vein, intended fusion and if graft function is poor. Intra-operative to decompress the portal system and maintain venous drainage of a large pleural effusion (hepatic hydrotho- return during the anhepatic phase. Heparin-bonded rax) is routine but can cause re-expansion pulmonary tubing and non-occlusive pumps allow its use without edema, which may only present after reperfusion. systemic heparinization. Use has declined as use of the Atelectasis, tension pneumothorax, and pulmonary piggyback technique has become more widespread, embolism may occur intra-operatively but are rare. and because many centers have demonstrated good Patients with the hepatopulmonary syndrome and results without it. It is now used selectively in CF present few problems intra-operatively, although most liver transplant programs and in some units in those with severe pulmonary shunting, postop- never. erative ventilation and oxygen dependency may be Advocates of bypass cite physiological advantages prolonged. during the final stages of dissection and during the anhepatic period, at least when the implantation techniqueinvolvesfullclampingofthecava.These Renal function include decompression of the portal, lower caval, Intra-operative changes in renal function are related to and renal venous systems and maintenance of car- marked alterations in cardiac output and renal blood diac output, thus preserving splanchnic and renal flow. Urine flow is diminished and markers of renal blood flow. Hemodynamic stability and systemic aci- injury are raised during caval clamping, when cardiac dosis are improved, and control of surgical bleed- output is reduced and renal venous pressure acutely ing is easier because venous pressure in portosys- raised. This response may be attenuated when veno- temic collaterals is reduced. However, evidence that venous bypass or a piggyback implantation is used. bypass improves results is lacking, and a compre- Pre-existing renal impairment, prolonged hypoten- hensive evaluation of its complications has yet to be sion, and high transfusion volumes are associated with published. a high risk of post-operative renal failure. No mea- The most serious hazards of bypass are perforation sures have yet been shown to prevent intra-operative of central vessels during insertion of large-bore percu- renal injury, although a rationale for the use of low- taneous catheters, and embolization of air or throm- dose vasopressors including vasopressin and nore- bus, all of which have caused fatalities. Body temper- pinephrine exists, given the pathophysiology of the ature decreases during bypass unless a heat exchanger hepatorenal syndrome. However, concerns about the is used, which may carry an added risk of thromboem- effects of these on perfusion of the newly implanted bolism. Local complications at access sites also occur, liver remain. including nerve injury, hematoma, lymphocele, and infection. Most units now reserve the technique for Early postoperative care and patients most likely to gain from its use. Indications may include severe portal hypertension, renal or car- complications diac impairment, marked metabolic acidosis or vaso- Management of liver transplant recipients between pressor dependency, and hypotension on trial clamp- transplantation and discharge is usually under- ing of the cava. taken by a multi-disciplinary team that includes

207 Section 4: Liver

Table 24.2 Diagnosis and management of early postoperative complications Complication Risk factors Diagnosis Management

Bleeding Previous surgery ↑Drainage Transfuse, correct coagulation, Severe portal hypertension ↓ Blood pressure re-explore Poor graft function ↓ Hemoglobin Primary non-function Marginal or NHB donor, steatosis ↑ Lactate MOF support, urgent Long cold or warm ischemic ↑ K+ re-transplant (consider times Acidosis hepatectomy once donor Revascularization problems Coagulopathy identified) Oliguria MOF Hepatic artery thrombosis Primary sclerosing cholangitis Doppler US, CT, or direct Early thrombectomy or revision; Small artery or complex arterial angiography to confirm urgent re-transplant if graft anatomy, Liver enzymes may not rise failure Bile leak ↑ warm ischaemia: delayed Bile in drains (drain bilirubin Ͼ Repair (Roux loop drainage) arterialization serum); perihilar collection on NHB donor US or MRI Donor duct too long Rejection Ͼ 5–7 days post-transplant ↑ ALT and bilirubin Pulsed prednisolone Pyrexia Biopsy Sepsis Ͼ 5–7 days post-transplant ↑ WBC and CRP HDU/ICU monitoring, volume Debilitated patient Tachycardia resuscitation, cultures (blood, Atelectasis/pneumonia Hypotension drains, intravascular catheters, Recent pulsed steroid Delirium wound, sputum, urine); Oliguria antibiotics; USS/MRI for Acidosis collections Deterioration of graft Rejection Identify cause as above (imaging Treat cause as above function (after 5–7 days) Sepsis ± biopsy, septic screen) Bile leak Compromised arterial inflow ALT: alanine aminotransferase; CRP: C-reactive protein; MOF: multi-organ failure; NHB: non–heart-beating; US: ultrasound; WBC: white blood cell count.

intensivists, hepatologists, and transplant surgeons. hoursofadmissiontoICUisusuallypossible,butthis Recipients (and their grafts) are heterogeneous, and depends on recipient comorbidity, complexity of the the postoperative course is highly variable. Frequent surgery, and especially adequate initial graft function. adjustment of immunosuppression, anticoagulation, To be extubated, the patient must be awake, hemo- antimicrobial treatment, and prophylaxis are all dynamically stable, and normothermic, with good gas required, and much of this can be protocol-driven. exchange and no bleeding or major acidosis. Mea- The formulation of a set of protocols that are regularly sures such as 30-degree head up tilt while ventilated reviewed and modified is essential to high-quality, (semi-sitting once extubated) and regular tracheal co-coordinated care. Complications are summarized toilet/physiotherapy may reduce pulmonary atelecta- in Table 24.2.Themostimportantearlycomplications sis and subsequent infection. Adequate analgesia and are primary non-function, hepatic artery thrombosis, exercises to encourage deep breathing and coughing and bleeding. Biliary leaks, rejection, and sepsis tend are also important. The use of modest positive end- to occur after the first 5–7 days. expiratory pressure (PEEP) while ventilated and non- invasive positive pressure ventilation after tracheal extubation may be considered and do not appear to Intensive care unit care compromise hepatic venous flow. Most liver recipients are transferred to the inten- Non-cardiogenic pulmonary edema (acute respi- sive care unit (ICU) for postoperative care. Tracheal ratory distress syndrome) is an infrequent complica- extubation in the operating room or within a few tion, but occurs in the setting of massive transfusion,

208 Chapter 24: Peri-operative care and early complications graft non-function or infarction, and fulminant sepsis Hemorrhage or rejection. It is managed with mechanical ventilation Hemorrhage requiring re-exploration is now unusual, with small tidal volumes, PEEP,and increased inspired butevenpatientsclosedwithadrysurgicalfieldmay oxygen sufficient to prevent systemic hypoxemia. Sig- bleed in the immediate postoperative period. Recog- nificant hepatic hydrothorax can be drained transdi- nition and timely intervention are essential. Bleed- aphragmatically at the time of surgery, and recurrence ing may not be revealed by abdominal drainage and large enough to need a pleural drain is uncommon. should be suspected in the setting of a rising heart rate, Prophylactic antibiotics for 24–48 hours are routine in falling blood pressure, reduced skin temperature, and most units. oliguria. A falling hemoglobin concentration should confirm the diagnosis. Although an abnormal PT is not routinely corrected post-transplant, because it is Primary non-function usefulasamarkerofimprovingsyntheticfunction, signs of active bleeding should prompt volume resus- Primary non-function is often first apparent in the citation and administration of fresh-frozen plasma, operating room, manifest by coagulopathy, persistent platelets, and red blood cells as soon as possible. If the hyperkalemia, acidosis, hyperlactatemia, hypotension, patient fails to stabilize after initial resuscitation in the and oliguria. These worsen in the early postopera- ICU, then he/she should be returned to the operating tive period, accompanied by hypoglycemia, anuria, room immediately. Failure to do so may compromise deteriorating gas exchange, and circulatory collapse. the newly implanted liver. Urgent re-transplantation may be life-saving, but Although early biliary leak is uncommon, it will these patients deteriorate rapidly and may become be readily detected by careful inspection of drain fluid too sick to withstand the procedure. Some groups for volume and especially color. The diagnosis is con- perform total hepatectomy once a suitable replace- firmed by measuring the concentration of bilirubin: if ment organ has been identified, which may stabilize greater than that in the patient’s serum, surgical explo- the patient, but clear indications and evidence are ration is usually required. Typically, the standard duct- lacking. to-duct anastomosis is converted to a Roux loop. Post-ICU care Hepatic artery thrombosis Management of the patients on the ward (initially The patient should have a Doppler ultrasound study in a high-dependency area) is best undertaken by a within 24 hours and at least daily afterward to con- multi-disciplinary team that includes hepatologists, firm hepatic arterial and portal venous flow. Absence transplant surgeons, dietitians, and physiotherapists. of either should prompt urgent angiography and Progress is highly variable between patients, but the thrombectomy or revision. Vascular thrombosis may dominant concerns and their management are out- not result in immediate biochemical changes and may lined below. otherwise be missed when early intervention could Immunosuppression in our institution is usu- save the graft. The highly variable vascular anatomy ally tacrolimus (TAC)-based, with dosing adjusted to of the liver and the variety of vascular reconstruc- obtain serum trough levels of 10–15 ng/l. TAC is well- tion techniques may make imaging difficult to inter- absorbed in the upper GI tract and the intravenous pret, so a diagram for the radiologist is essential. Com- preparation is rarely used, even in the immediate post- plex arterial or venous reconstruction may increase operative period. The TAC level must be adjusted to the likelihood of thrombosis, as may the postopera- take albumin binding into account, since hypoalbu- tive use of vasopressors. Loss of arterial inflow, if not minemia is common in the immediate postoperative causing early graft infarction and sepsis, leads even- period, and target values based on whole-blood mea- tually to ischemic cholangitis, biliary leak, and stric- surements may be nephro- or neurotoxic. tures, and usually precipitates re-transplantation. If Patients who are intolerant of TAC are given urgent re-transplantation is needed, immunosuppres- cyclosporine. In some patients, sirolimus may be sion is reduced and antibacterial and antifungal agents more suitable because of nephro-protective and anti- given. neoplastic effects, but it is not used in the immediate

209 Section 4: Liver

postoperative period because of its inhibitory effect on and discussed with the microbiology team. Unusual wound healing. The use of prednisolone and azathio- causes of infection should also be considered, includ- prine (AZA) is variable. Most patients are also given ing organisms potentially transmitted with the donor prednisolone, but those who undergo transplantation organ. Many organs are retrieved from donors who for hepatitis C virus are, if possible, maintained on a themselves have had sepsis. Culture results from the steroid-free regimen. It is our practice also to com- donor and targeted antimicrobial treatment should be mence AZA once lymphocyte counts are greater than considered in recipients with unusual presentations of 0.5 × 109 /l. See Chapter 3 for more details. sepsis. Deterioration of graft function after initial stabi- lization is common, indicated by rising alanine amino- transferase and bilirubin. The most likely causes in Acute cellular rejection the early postoperative period are vascular or biliary This usually first presents in the first week after trans- complications, sepsis, and rejection. Imaging of the plant, signaled by rising bilirubin and hepatic enzyme transplanted organ is the first step, initially by ultra- levels, sometimes accompanied by eosinophilia, an sound. Absence of arterial flow requires confirmation unexplained rise in C-reactive protein, or pyrexia. by direct or CT angiography and can occur with lit- Percutaneous liver biopsy is performed to confirm tle or no biochemical change. It usually results in the the diagnosis. When persistent ascites, thrombocy- need for re-transplantation. Once confirmed, appro- topenia, or other coagulopathy make this difficult priate anti-fungal and antibiotic therapy should be or hazardous, a transjugular approach may be used, commenced and immunosuppression reduced. At re- although this may be challenging if cavocavoplasty was transplantation, a conduit of donor iliac artery is anas- used to implant the liver. Biopsy-proven acute cellu- tomosed to the aorta, since the recipient celiac axis is lar rejection is treated with pulsed intravenous methyl- often no longer usable. prednisolone. This is usually very effective, although Ultrasound should be followed by triple-phase CT an attenuated response, or steroid-resistant rejection, if a biliary leak (usually heralded by a collection of is occasionally encountered and is often diagnosed fluid at the hilum) or infected collection is suspected, only after multiple pulses. It may then be necessary remembering that liver recipients may not complain of to use anti-thymocyte globulin. This involves a 2- abdominal pain in the presence of biliary peritonitis. week course with doses adjusted against lymphocyte Tesla magnetic resonance imaging can be helpful, but count. interpretation in the immediate postoperative period Antibody-mediated rejection in liver recipients is is difficult. If there is graft dysfunction, any collec- very unusual but should be considered in cases of per- tion should be percutaneously sampled or drained and sistent rejection with vascular changes on histology the fluid sent for bilirubin and culture. If a percuta- and rising levels of donor-specific antibody. Treatment neous approach is not possible, endoscopic ultrasound regimes vary but usually involve plasmapheresis and or laparotomy may be needed. anti-lymphocyte agents. If a biliary leak is confirmed, the treatment of Ongoing ascitic losses occur in patients who have choice is formal reconstruction with a Roux-en-Y loop had large volumes of ascites before transplant. This of jejunum. In most cases the leak is from the biliary is usually self-limiting but may persist for several anastomosis, but it may be from accessory ducts or weeks. However, new-onset ascites should prompt even biopsy sites. Complete evacuation of the bile col- investigation of portal venous and caval anastomoses. lection is essential as the hepatic artery typically tra- In particular, narrowing of the top caval anastomo- versesitandisoftenfoundtobeinspasm. sis is associated with ascites, peripheral edema, and Sepsis is common after transplant, especially renal dysfunction. Imaging to confirm caval steno- after 5–7 days, and is frequently associated with sis usually involves cavography with pressure stud- liver dysfunction. The usual sites of postsurgical ies and portal venous wedge pressure measurements. infection must be assessed (chest, wound, urinary Treatment is difficult and in the early postoperative tract, cannulation sites) and cannulae, catheters, and period may be best achieved with surgical revision. drains removed if possible. Intra-abdominal collec- Later presentations are surgically hazardous and may tions require drainage, either radiologically guided or be more safely managed by percutaneous stenting or surgical. Antimicrobial therapy must be initiated early dilatation.

210 Chapter 24: Peri-operative care and early complications

Further reading measures to reduce them. Liver Transpl 2003; 9: 1320–7. Fonouni H, Mehrabi A, Soleimani A, Muller¨ SA, Buchler¨ MW, Schmidt J. The need for venovenous bypass in liver Safadi A, Homsi M, Maskoun, et al. Perioperative risk transplantation. HPB (Oxford) 2008; 10: predictors of cardiac outcomes in patients undergoing 196–203. liver transplantation surgery. Circulation 2009; 120: 1189–94. Ramos E, Dalmau A, Sabate A, et al. Intraoperative red blood cell transfusion in liver transplantation: Influence Steadman R. Anesthesia for liver transplant surgery. on patient outcome, prediction of requirements, and Anesthesiol Clin North Am 2004; 22: 687–711.

211 Section 4 Liver Chapter Long-term management and outcomes

25 William Gelson and Graeme J.M. Alexander

Key points > 1988 100 r The most common causes of death after the Patients: 74534 first year following transplantation are Grafts: 82961 75 recurrent and de novo malignancy, return of the original liver disease in the graft, sepsis, cardiovascular disease, and chronic rejection. 50 r Immune suppression should be tailored according to the disease causing the primary Survival (%) 1968 to 1988 25 liver disorder, with varied doses or different Patients: 2014 agents. Grafts: 2214 r The most common cause of renal 0 impairment after liver transplantation is 0462 810 calcineurin-inhibitor nephrotoxicity; Ye a r s substitution with mycophenolate mofetil or Figure 25.1 Patient and graft survival following liver sirolimus may be required. transplantation in Europe. Engraftment was between May 1968 and December 2008. Graft and patient survival curves are shown for transplants before and after 1988. The curves demonstrate that short-term survival has markedly improved, whereas long-term Early patient and graft survival, defined as the imme- survival has not. Data are taken from the European Liver Transplant diate postoperative period and the first year after Registry (www.eltr.org). liver transplantation, have both improved substantially since the procedure became adopted widely in the There are clear and consistent causes of late mor- 1980s. In contrast however, survival for those patients tality (Table 25.1) that are common to all transplant who are alive after the first year has not improved in programs and to each era, and one explanation for the the longer term. This has been demonstrated in large lack of improvement in long-term survival is subop- cohorts from both the United Kingdom and the United timal management at an earlier stage for those condi- States. Survival curves from the European Transplant tions that lead to late deaths. Registry (Figure 25.1)showasimilarpatternfromyear Late graft failure is much less common now than 3 onwards, where patients undergoing liver transplan- before, a change that can be attributed to technical tation have a comparable survival when the operation improvements, experience, and a much lower likeli- was performed between 1968 and 1988 to those under- hood of chronic rejection, so that recurrence of the taken after 1988. original disease now represents the greatest challenge. The comparison of long-term survival during each It is not an inconsiderable challenge, since it is hard to era is not “like for like,” since the indications for trans- thinkofmanyadultliverdisordersthathavenotbeen plantation and the underlying causes of liver injury reported to recur after liver transplantation. have changed substantially with time, but the absence Few studies examine those factors that predict late of a demonstrable improvement in long-term survival mortality. A preliminary analysis of the UK Trans- remains a cause for concern. plant database using Cox regression analysis with

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

212 Chapter 25: Long-term management and outcomes

Table 25.1 Causes of death after liver transplantation (e.g., nephrotoxicity, infection, cardiovascular disease, beyond year-1 in Europe or cancer). Hepatocellular carcinoma 22% Long-term corticosteroid use should be limited De novo malignancy 16% to those patients with a history of rejection that has Return of the original liver disease in the graft 14% been difficult to manage and dosing minimized by concomitant use of other forms of immune suppres- Sepsis 8% sion. The biopsy appearances of late acute rejection Cardiovascular disease 7% may resemble “autoimmune hepatitis,” and patients Chronic rejection 5% with this picture may respond to corticosteroids read- ily but relapse after withdrawal or dose reduction. The use of budesonide to minimize systemic side effects death after year-1 as the outcome variable found in this context is supported by anecdote and expe- that certain underlying liver disorders at the time of rience but not yet by randomized controlled trials transplant carried an increased risk (these included (RCTs). alcohol-related, autoimmune hepatitis, cryptogenic or Azathioprine (AZA) or mycophenolate mofetil hepatitis C virus [HCV] related, and the presence of (MMF) are often used as long-term maintenance hepatocellular carcinoma [HCC] at the time of trans- immunosuppression. There are no large head-to- plantation) and further that the choice of immune sup- head trials of these two agents in liver transplan- pression was also influential, whereby prednisolone tation, and choice is center-specific. A minority of useatlastfollow-upwasassociatedwithanincreased patients are intolerant of AZA, whereas a small risk of death. The latter, of course, may be cause or con- number develop centrilobular injury that resolves sequence and requires further evaluation in prospec- after withdrawal. Calcineurin inhibitors (CNIs) are tive studies. oftenthecoretherapyforimmunesuppression, The etiological associations with survival after liver again with no RCT evidence of benefit of a single transplantation were corroborated by a recent study regime. that examined “life years lost” after transplantation. The use of sirolimus (SRL) after liver transplanta- When the UK liver transplant cohort survival was tion is controversial. It is not endorsed by the US Food standardized and compared with that of the UK pop- and Drug Administration because of two studies that ulation at large, the etiologies of alcohol, autoimmune, demonstrated adverse outcome. The first showed an HCV, and HCC were associated with loss of 14 years, association of SRL with early hepatic artery throm- 5 years, 17 years, and 24 years, respectively. Conversely, bosis. The second revealed an excess mortality after the etiologies of cryptogenic and primary biliary cir- switching from CNI to SRL in a group of patients with rhosis were associated with a gain in life years at 5 and established grafts. The excess mortality did not reach 7 respectively. The cause of life years gained remains a statistical significance, and subsequent studies from matter of debate and may be in part due to the selection the US Transplant Registry have demonstrated a sur- process of candidates for liver transplantation. What- vival benefit. ever the risk factors for patient loss, optimizing man- agement and prevention of late complications is cen- tral to improving long-term outcome. Management of the complications of immunosuppression Immunosuppression Common complications seen in long-term survivors In common with other organ transplants, immune of liver transplantation that are unrelated to the graft suppression in long-term survivors of liver trans- are shown in Table 25.2. plantation involves a balance between effective con- trol of rejection and complications related to ther- apy (discussed further in Chapter 3). Complications Metabolic syndrome of immune suppression may be related to the origi- Up to 45% of liver transplant recipients have metabolic nal etiology (e.g., accelerated disease progression in syndrome, defined as three or more of excessive weight viral hepatitis) or unrelated and similar to other organs gain, hypertension, diabetes, and hyperlipidemia.

213 Section 4: Liver

Table 25.2 Common complications in long-term survivors of graft and patient survival. In one study, those who liver transplantation developed sustained post-transplant diabetes melli- Complication Risk tus had a 10-year graft survival rate of 59% com- pared with 73% in non-diabetics and patient survival Cardiovascular disease Cardiac event RR approximately 3 of 69% versus 78%. Diabetes is also a major risk fac- Heart disease Prevalence up to 18% tor for coronary heart disease, cerebrovascular dis- Metabolic syndrome ease, and peripheral vascular disease in liver transplant Hypercholesterolemia Prevalence 15–62% recipients. Hyperlipidemia Prevalence up to 33% Patients should be screened at each clinic visit with Hypertension Prevalence 19–82% Type 2 DM SPR approximately 3 a random glucose, and hemoglobin A1c should be Obesity Prevalence 7–30% measured at least annually. Suspected diabetes melli- SPR approximately 6 tus should be confirmed with a fasting glucose and, Prevalence 29–47% where indicated, a formal glucose tolerance test. Man- Malignancy agement involves a combination of lifestyle advice, oral Solid organ SIR approximately 2.5 to 4 Lymphoma SIR approximately 30 hypoglycemic agents, insulin, and reduction or dis- Other continuation of corticosteroid therapy. Switching from Osteoporosis Prevalence 26% CNI has not proved beneficial. Fractures in women SIR approximately 2 Renal impairment Prevalence 35–83% SIR: standardized incidence ratio; SPR: standardized prevalence Hyperlipidemia ratio; RR: relative risk. A fasting lipid profile should be measured annually and patients treated according to current guidelines Hypertension with a statin in the first instance and a fibrate as sec- ond line. Dietary changes alone are ineffective. Simvas- The diagnosis of hypertension relies on a combination tatin, atorvastatin, or pravastatin are appropriate first- of clinic measurements, home measurements, and 24- line agents. The Framingham Cardiac Risk Score has hour ambulatory blood pressure monitoring. Given not been validated for liver transplant recipients, who current anti-hypertensive guidelines, the target blood have an additional risk that in our view merits a lower pressure should be 140/90 mmHg and lower (130/ threshold for treatment. 80 mmHg) in those with renal disease and/or dia- betes mellitus. In the presence of diabetes mellitus or proteinuria, a reasonable first-line anti-hypertensive Obesity is an angiotensin-converting enzyme (ACE) inhibitor, Obesity is a risk factor for hypertension, type 2 dia- with a calcium channel blocker as second line. In the betes, hyperlipidemia, and cardiovascular disease as absence of proteinuria, the converse applies with a cal- well as surgical complications and graft dysfunction. It cium channel blocker as first-line therapy and com- is common, with a rising incidence. CNIs and corticos- bination with an ACE inhibitor as second line. Beta- teroids undoubtedly exacerbate the problem. Dietary blockade is less effective in liver transplant recipients adviceandexercisemaynotbeaseffectiveasin treated with CNIs and may paradoxically increase cen- clinical practice outside the transplant field. Thera- tral blood pressure. peutic manipulation of immunosuppression (particu- larly corticosteroid avoidance) may be important, and Diabetes pharmacological therapies such as Orlistat should be Risk factors for type 2 diabetes include ethnicity, a considered; in some cases, surgical referral should be positive family history, age greater than 45 years, glu- considered. cose intolerance pretransplant, use of prednisolone and/or TAC, and the etiologies of HCV infection and cryptogenic (which probably reflects previous Renal impairment non–alcohol-related fatty liver disease in this set- Renalfailurehasasignificantimpactonsurvival ting). Effective management of diabetes is impor- post-transplant and may double mortality 13 years tant,asitisassociatedwithimpairedlong-term after transplantation. It is also an additional, major

214 Chapter 25: Long-term management and outcomes risk factor for cardiovascular disease. The most com- dence of prostate and breast cancer does not appear to mon cause is CNI nephrotoxicity; other causes include be increased. Preventative strategies include smoking diabetic nephropathy, hypertensive nephropathy, and avoidance, protection from sun exposure, and mini- HCV-associated glomerulonephritis. mizing immune suppression so as not to suppress the In the case of CNI nephrotoxicity, successful anti-viral and anti-tumor effects of the host immune approaches have included substitution with MMF or response. SRL. It is critical to recognize renal impairment at an early stage and take appropriate action before the changes become irreversible. In our view, any deterio- Graft-related complications ration in renal function at any stage merits immediate review of pharmacotherapy. Dose of CNI may need to Allograft rejection be reduced and plasma levels carefully monitored. Late acute cellular rejection (ACR) is similar both his- tologicallyandbiochemicallytoearlyACR,butmaybe difficult to distinguish from autoimmune (or perhaps Osteoporosis more correctly alloimmune) hepatitis. A variety of Bone mineral density is often low pretransplant clinical signs and symptoms may be observed, includ- despite prophylaxis and therapy, and falls further, ing fever, malaise, abdominal pain, hepatomegaly, and often substantially, in the first 3 months following increasing ascites. Laboratory abnormalities include liver transplantation, rising to pretransplant or even abnormal liver biochemistry, elevated inflammatory healthy/normal age-matched values within 2 years. markers, and an eosinophilia. ACR is a pathological During this time fracture risk is highest. diagnosis with three major histological hallmarks: a Treatment with calcium, vitamin D, and bisphos- mixed portal inflammatory infiltrate, non-suppurative phonates should be considered. Vitamin D deficiency cholangitis, and endothelialitis. Endothelialitis later is common before and after liver transplantation and can affect any transplanted vessel, including those often overlooked. Long-acting bisphosphonates are inaccessible to biopsy, such as the cava. now available, which may have better compliance. The incidence of late ACR is between 7% and 23% CNIs are a probable risk factor for osteoporosis, but and may coincide with a period of suboptimal immune there is insufficient evidence to advise withdrawal in suppression. Despite being a risk factor for chronic the presence of osteoporosis. ductopenic rejection, late ACR usually responds either to therapeutic manipulation of immune suppression or a short course of high-dose intravenous methyl- Neoplasia prednisolone. Rarely, anti-thymocyte globulin will be Liver transplant recipients have an increased inci- required. In our center, patients with late ACR are dence of both cancers that are related to chronic viral maintained on at least dual-agent immune-suppressive infection and cancers related to the underlying dis- therapy long-term. ease. Examples of virus-related cancer include basal Chronic ductopenic rejection (CDR) usually cell carcinoma; squamous cell carcinoma of the skin, occurs in the first 2 years after transplantation. Its cervix, and oropharynx; and lymphoma. Examples incidence is decreasing, probably as a consequence of etiology-related cancers include carcinoma of the of improved immune suppression regimens and colon in patients with primary sclerosing cholangi- perhaps the introduction of more effective therapy for tis (PSC) and esophageal carcinoma in alcohol-related cytomegalovirus (CMV). The clinical onset is more liver disease. insidious than ACR, with cholestasis being a promi- Like other organ transplants, cancer is one of the nent feature. Histology shows progressive ductopenia most important causes of death in liver transplant with inflammation, whereas features of a superim- recipients (discussed further in Chapter 4A). Skin posed ACR may also be present. CDR may reverse cancers and lymphoma have an incidence of more with optimization of immunosuppression, but more than 10 times that of the background population, often necessitates consideration of re-transplantation whereas for solid organ cancers, the risk is increased ultimately. The presence of bile ducts within more approximately four-fold. Most but not all solid organ than 50% of portal triads is a good prognostic sign malignancies are more common. Notably, the inci- in terms of treatment response. Risk factors for CDR

215 Section 4: Liver

include recurrent episodes of ACR, late-onset ACR, of detectable immunity. The role of vaccination against CDR in a previous graft, untreated CMV infection, HBV post-transplantation remains to be resolved, but inadequate immune suppression, and primary biliary there is insufficient evidence at present to support its cirrhosis or primary sclerosing cholangitis as the use. Vaccination pretransplant for those naive to HBV underlying disease etiology. is essential but not always effective.

Disease recurrence Hepatitis C virus Graft infection with HCV is almost invariable and Hepatitis B virus probably occurs within hours of implantation. Hepatitis B virus (HBV) graft infection led to death Chronic inflammation is inevitable and fibrosis from liver failure in the majority of patients prior to the progression more rapid, probably as a side effect of introduction of effective antiviral agents in the 1990s immunosuppression. and remains a risk in (1) recipients positive for HBsAg Between 10% and 30% of recipients with hepatitis pretransplant whether or not they are HBV DNA pos- C develop cirrhosis by year 5. This number increases itive in serum, although the risk of graft infection is to more than 40% by year-10. A rapidly progressive greater with a higher titre of HBV DNA; (2) HBV-naive fibrosing cholestatic hepatitis affects a minority (2– recipients of grafts from anti-HBc positive donors; 5%). Protocol biopsies help to characterize disease (3) recipients with suboptimal HBV prophylaxis post- progression. In the early post-transplant period, it transplant (either the lack of prophylaxis, a poor pro- may be difficult to distinguish between acute rejection phylactic regime allowing HBV mutants to arise, or andgraftinfectionwithHCVonhistologicalgrounds poor patient compliance). alone, but the clinical scenario is often informative; Prophylaxis with a combination of HBV immune hepatitis in the first 28 days is most often rejection globulin and at least two antiviral agents from lamivu- andthatbeyond28daysismostlikelytobeHCV- dine, adefovir, tenofovir, and entecavir is effective for related. Some centers employ additional histological those positive for HBsAg pretransplant, whereas a assessment; the presence of Mcm-2 in the portal infil- combination of two of these agents is effective usu- trate is indicative of ACR, whereas the absence of lym- ally for naive recipients of an anti-HBc positive donor. phocyte Mcm-2 expression suggests HCV-related graft The choice of regime differs, often markedly, between injury. centers, and the use of HBV immunoglobulin (Ig) Advanced donor age is the most important risk may be determined by cost, but certain principles are factor for rapid progression of HCV-related fibrosis, common. Thus pretransplant therapy for HBV may but other factors include high levels of corticosteroid select treatment-resistant mutants, so resistance profil- use or rapid withdrawal of immunosuppression. Ther- ing pretransplant is helpful in those with HBV DNA in apy with pegylated interferon-␣ and ribavirin may serum at the time of liver transplant assessment. Some be effective, with sustained virological response (viral patients will have been treated before, sometimes with- clearance) achieved in approximately 20%. However, out the knowledge or support of transplant physi- treatment side effects may be significant, and rates of cians, which should inform post-transplant manage- acute rejection of up to 45% are reported. Although ment. The frequency of administration and the dose of arecentCochranereviewfoundnomortalityben- HBV Ig (if used) must be sufficient to maintain anti- efit with treatment (duration of follow-up was lim- body levels above 100 IU/L at all times. Detection of ited), many hepatologists are more confident now HBsAg or HBV DNA post-transplant must be investi- with antiviral therapy in the transplant setting than gated urgently. The optimum duration of prophylaxis previously. after transplantation (or perhaps more accurately, pre- ventative therapy) is uncertain, and breakthrough after withdrawaloftreatmentaslateas5yearsaftertrans- Autoimmune diseases plantation has been reported. It is certainly possible to ThesecarryanincreasedriskofACRintheearlystages reduce the number of agents used after a period of 3–5 and late ACR if immune suppression is decreased. In years of effective prophylaxis/therapy, but our view is addition, all three conditions can be manifest on the that some form of treatment is essential in the absence grafted liver.

216 Chapter 25: Long-term management and outcomes

Primary biliary cirrhosis Structural complications At 10 years from transplantation, biochemical and his- Biliary stricture and incisional hernia are the most tological changes characteristic of primary biliary cir- common late surgical complications after liver trans- rhosis are found in 60% of those recipients whose plantation. Incisional hernias are managed by a com- original disease was primary biliary cirrhosis. Initial bination of weight loss, supports, and surgical inter- immune suppression with cyclosporine rather than vention. Anastomotic problems may present with a TAC may be beneficial in preventing the onset of combination of abnormal liver tests, cholangitis, and these changes in the grafted liver but appears to jaundice. Management varies from antibiotic therapy have no benefit if used after these changes are estab- and observation, through stenting to biliary recon- lished. Treatment with ursodeoxycholic acid may also struction by means of a Roux-en-Y anastomosis. be of benefit for established disease. Some patients Non-anastomotic strictures are usually the result of with recurrent disease progress rapidly to cirrhosis ischemia, recurrent PSC, or cholangitis. If strictures with portal hypertension. Diagnosis relies on liver are isolated, they may be managed by stenting and/or biopsy, as anti-mitochondrial antibody status is not biliary reconstruction. If, as is often the case, stric- informative. Graft failure secondary to recurrent pri- tures are multiple, management is with antibiotics and mary biliary cirrhosis necessitates consideration of re- ursodeoxycholic acid. Progressive biliary disease or transplantation, but recurrent disease per se does not the development of secondary biliary cirrhosis should seemtoaffectgraftorpatientsurvival. prompt consideration of re-transplantation. Late hepatic artery thrombosis usually presents Autoimmune hepatitis with liver abscess or biliary complications. It may, however, be an incidental finding at ultrasound, and The risk of recurrent disease is such that long-term liver function tests may be normal. Attempts at stent- immunosuppression with two agents is recommended. ing and reconstruction are usually futile, and a further Recurrentordenovoautoimmunehepatitisoccursin transplant may be required. approximately 25% of grafts during the first 5 years Portal vein thrombosis and stenosis are rare late after transplantation and more than 50% after 10 years complications. They may present with features of por- of follow-up. Diagnosis is made by a combination of tal hypertension or be an incidental finding. Treat- abnormal liver biochemistry, elevated serum IgG, and ment of portal hypertension (beta-blockade or band histological features. ligation) and anti-coagulation is often required; alter- natively, surgical or radiological portosystemic shunt Primary sclerosing cholangitis should be considered. PSC carries an increased risk of hepatic artery throm- bosis and thus graft loss. Recurrence of PSC occurs in about 30% of allografts after 5 years and appears more Long-term follow-up of liver aggressive in post-primary grafts, with rapid progres- transplant recipients sion to liver failure in many cases. There are no strate- Review frequency varies between centers and depends gies to prevent disease recurrence. Diagnosis involves partly on patient morbidity. The aim of follow-up a combination of cholangiography and liver biopsy. is to screen for graft dysfunction and the late com- Treatment is of disease complications such as biliary plications of liver transplantation outlined previ- stricture and re-transplantation in the face of graft ously in this chapter. Biannual review is a reasonable failure. minimum requirement, with a consultation, blood pressure measurement, skin examination, urinalysis, Metabolic diseases and blood tests (full blood count, liver biochemistry, Iron overload may occur in patients who undergo renal function tests, fasting glucose, HBA1c measure- transplantation for hemochromatosis and is prevented ment, lipid profile, and trough immune suppression by maintaining iron stores at physiological levels with level for patients on CNI or SRL). Patients should recurrent venesection. It is known that recidivism in undergo population-based screening (mammography, those who undergo transplantation for alcohol-related colon cancer surveillance, dermatology review, and liver disease is associated with a reduced survival. cervical smear testing) as per local protocol. Dual-

217 Section 4: Liver

Figure 25.2 An approach to the management of abnormal liver tests in late survivors of liver transplantation. MRCP: magnetic resonance cholangiopancreatography.

energy x-ray absorptiometry (DEXA) scanning to this and reflect the fact that the selection process for screen for osteoporosis should be performed 2 years re-transplantation needs to improve. after transplantation and at least every 5 years there- Although selection for re-transplantation is cur- after. rently center-dependent, there is evidence that nation- Certain patient groups require additional inves- ally applied scoring systems may be informative. In tigations. These include 6-monthly ultrasound and the United States, the Model for End-Stage Liver Dis- ␣-fetoprotein surveillance for patients who undergo ease (MELD) score is used to select patients for pri- transplantation following HCC, endoscopy follow- mary grafts. With a score of greater than 25 imme- ing primary sclerosing cholangitis, and protocol liver diately prior to transplantation, survival at 1 year is biopsies to identify disease progression following HCV less than 60% for post-primary grafts, compared with infection. less than 80% for first grafts. MELD score also predicts Abnormal liver biochemistry is found in approxi- mortality in recipients with graft failure, with a score of mately 60% of patients, and minor abnormalities are 25 equating to a survival of 50% at 1 year, compared usually managed expectantly. Significant elevations with 60% in patients with cirrhosis and liver failure should prompt intervention and/or investigation. A before primary engraftment, i.e., patients with liver management scheme for deranged liver tests late after graft failure are more likely to die than patients with liver transplantation is given in Figure 25.2. failure of their native liver.

Re-transplantation Quality of life after liver Survival after liver re-transplantation is worse than transplantation after primary transplantation and decreases progres- One of the major aims of liver transplantation is to sively with each new graft. In a period of donor improve quality of life. During the first 6 months shortage, second and subsequent liver transplants may after transplantation, physical and mental quality-of- be difficult to justify. Primary liver graft survival in life scores improve. Unfortunately, this trend is not Europe is presently 70% at 3 years, 65% at 5 years, sustained, and after year 1, emotional and mental and 55% at 10 years. The corresponding values for a health–related quality-of-life scores begin to decrease. second graft are 50%, 46%, and 37% and for a third Employment is often used as a surrogate marker for 43%, 39%, and 30%, respectively. In the United King- quality of life. Approximately 50% of liver transplant dom,oneofthelistingstandardsforlivertransplanta- recipients are employed, and studies have confirmed tion is that projected 5-year mortality be greater than that those in work have a better quality of life than 50%. These data for post-primary grafts fall short of those out of work. Sexual dysfunction improves after

218 Chapter 25: Long-term management and outcomes liver transplantation. Although 30% of men report sex- After pregnancy, care should be taken with dosing ual inactivity before transplantation, only 15% report to prevent drug toxicity, as trough levels may increase. this afterwards. Fewer women seem to experience sex- Breast feeding is usually discouraged due to theoretical ual dysfunction before transplantation (35%). This toxicity in breast milk. increases modestly to 40% after transplantation. Over- all, few publications exist in the field of quality of Potential future strategies life after liver transplantation. Given that quality of The “holy grail” of solid organ transplantation is life is a major post-transplantation end point, fur- often regarded as achieving immunological tolerance. ther studies are required. In the mean time, psy- Although this may be achieved in the future with chosocialhealthshouldbeconsideredasanimportant new immunosuppressive strategies, it is apparent that facet in the long-term management of liver transplant some liver transplant recipients are already “opera- recipients. tionallytolerant”inthattheydonotrequirelong-term immunosuppression. Conception and pregnancy A handful of studies (the largest with 93 patients) The first successful pregnancy after liver transplan- have examined the withdrawal of immunosuppression tation was reported in 1978. As of December 2009 from patients with established grafts. Approximately and since 1991, there have been 266 successful births 20% of recipients seem to tolerate this in the medium in 152 liver transplant recipients recorded by the term (maintain normal liver biochemistry) and 10% in National Transplant Pregnancy Registry in the United the long term. Twenty percent develop clinically signif- States. icant rejection, but graft loss is rare. Liver-related complications are thought to include Studies of peripheral lymphocyte phenotypes and an increased risk of acute cellular rejection during liver biopsy characteristics appear to be the most likely pregnancy, affecting approximately 6%. This may be avenue in which rejection risk will be quantified. Bio- due to altered pharmacokinetics, which leads to lower logical agents that target peripheral lymphocytes des- trough levels of immunosuppressant, necessitating an tined for the liver appear to be the most likely candi- increasedfrequencyofmonitoringandoftenincreased dates to provide targeted, liver-specific immune sup- dosing. Liver biopsy should be avoided whenever pos- pression. sible. More focus is required on late complications Maternal complications include increased risk of so that long-term outcomes improve. For every re- hypertension (34%), pre-eclampsia (25%), and dia- transplantation due to graft failure, a patient dies on betes (7%). Fetal complications include increased risk the waiting list. Thus long-term graft preservation is of low birth weight (35%), prematurity (40%), and fetal notonlypertinenttocurrentrecipients,buttofuture loss (26%). Pregnancies should be regarded as high risk recipients also. and be managed in a center where support is available from a liver transplant team. There is no definitive evidence that liver transplant Further reading immunosuppression is teratogenic. Prednisolone is Bacon BR, O’Grady JG, di Bisceglie AM, Lake JR. Comprehensive Clinical Hepatology. Philadelphia, PA: safe, CNIs are probably safe, and the effects of SRL Elsevier, 2005. are unknown (but given the anti-proliferative effects BentenD,StauferK,SterneckM.Orthotopicliver of SRL, it should be avoided). There are a few case transplantation and what to do during follow-up: reports of teratogenicity in association with AZA with- recommendations for the practitioner. Nat Clin Pract out transplantation and a similar number of reports of Gastroenterol Hepatol 2009; 6: 23–36. teratogenicity in association with MMF use in renal ILTS/AASLD Transplant Course: Long-Term Outcomes of transplant recipients. Overall they are probably both Adult and Pediatric Liver Transplantation. Liver Transpl safe. 2009; 15 (S2), 1–94.

219 Section 4 Liver Chapter Pediatric liver transplantation

26 Hector Vilca-Melendez and Giorgina Mieli-Vergani

Key points New and more varied immunosuppressive agents, r better organ preservation, and improvement in peri- Liver transplantation is the accepted and postoperative care have greatly contributed to treatment for a wide variety of liver diseases the success of LT. Currently, the 1-year survival in in children. r pediatric LT is approaching 95% for low-risk elec- Over the past 10 years, a number of tive cases and is close to 80% even for acute liver innovative surgical techniques have been failure. developed to overcome the shortage of Long-term management of post-transplant chil- size-matched donors, particularly in children dren entering into “normal” adulthood has become the younger than 5 years of age. r new challenge for the transplant community. Factors Graft and patient survival at 1 year has intrinsic to this patient population, such as adherence continued to improve, being currently to immunosuppression, drug therapy complications, greater than 85%. psychosocial development and education, transition to r Complications are relatively common, but adulthood, employment, and future parenthood, are provided graft function is satisfactory, now assuming great importance. long-term survival is expected. r Causes of early mortality include graft dysfunction and sepsis. Late mortality is Indications for liver transplantation most often due to sepsis, post-transplant LT should be considered for any child with end-stage lymphoproliferative disease, and liver disease with a predicted survival of less than non-adherence to immunosuppressive 1 year or with a severe impact on quality of life. The medication. pretransplant assessment and peri-operative manage- ment of children with liver disease require input from As more and more children have undergone liver a large and experienced multi-disciplinary team. The transplantation(LT),theshortageofsize-matched decision on when to list for transplant is critical in donors has become more of a concern and has led order to minimize the risk of dying while on the wait- to the development of new surgical techniques based ing list and to ensure that the child is in optimal con- on the segmental anatomy of the liver. The use of dition to survive transplantation. There are a num- reduced liver, as originally described by Bismuth and ber of factors that influence the timing of transplanta- colleagues, provides a suitable size graft for a child, but tion, including age, etiology of the underlying liver dis- wastes the rest of the liver mass. The so called “split” LT ease, past medical and surgical history, and quality of then evolved to provide two viable partial grafts that life. couldbesharedbetweenachildandanadult.Living At Kings College Hospital, two thirds of children related LT has also played an important role in reduc- are less than 5 years of age at the time of LT. Young age ing the waiting time for a cadaveric organ and has now does not play a significant role in outcome, except for become a routine procedure in pediatric LT (discussed children less than 3 months of age, who tend to do less further in Chapter 22). well.

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

220 Chapter 26: Pediatric liver transplantation

Clinical indications for LT include: and hepatic artery resistance index (HARI) ≥ 1.0 on Doppler ultrasound are all indications of progressive r Chronic liver disease with progressive liver failure liver disease and of the need for LT. The differenti- and impaired synthetic function (prolonged ation between isolated biliary atresia (90% of cases) international normalized ratio [INR] and low and biliary atresia splenic malformation (BASM) syn- serum albumin) drome (10% of cases) is very important for surgical r Severe jaundice with loss of muscle mass, failure planning before LT, owing to the anatomical varia- to thrive, and severe osteoporosis tions commonly found in BASM, including cardio- r Portal hypertension manifested by variceal vascular defects, polysplenia, situs inversus, absence bleeding and intractable ascites r of inferior vena cava (IVC), and preduodenal portal Encephalopathy, profound lethargy vein. r Spontaneous bacterial peritonitis, recurrent cholangitis ␣ r Intractable debilitating pruritus 1-antitrypsin deficiency Thisisoneofthemostfrequentmetabolicliverdisor- Some children develop hepatorenal or hepatopul- ders that results in cirrhosis and requires LT in chil- monary syndrome, which often reverses after LT. Pul- dren. Clinical indications for LT are the same as for monary hypertension may be present in children with other cirrhotic disorders, but progression of disease is long-standing chronic liver disease: if unrecognized particularly rapid after the first signs of decompensa- and severe (systolic Ͼ50 mmHg), this may increase the tion requiring early listing. risk of death intra-operatively or in the early postop- erative period. Developmental and/or growth retarda- tionarecommoninchildrenwithchronicliverdisease Non-cirrhotic metabolic liver disorders and should be addressed before and after LT. Non-cirrhotic metabolic liver disorders that cause no Quality-of-life issues must be considered when structural damage to the liver but have deleterious sys- deciding the timing of transplantation as they can have temic effects because of toxic metabolites or enzyme a profound effect on the child’s future. These issues deficiencies should be considered for LT (Table 26.1). include the psychosocial aspects of living with chronic In these conditions, the purpose of LT is to improve liver disease and educational prospects. the quality of life by correcting the metabolic defect, rather than being a life-saving procedure. Two other alternative strategies can be used to achieve this: hepa- Chronic liver disease tocyte transplantation and auxiliary partial orthotopic Etiologies causing chronic liver disease in children and LT (APOLT). In both procedures, the aim is to pro- potentially leading to LT are summarized in Table 26.1. vide the defective metabolite or enzyme preserving Some of the most important are described below. the whole or part of the native liver, thus providing a “safety net” in case of graft failure. The overall patient Biliary atresia survival for LT for metabolic disorders is excellent, without significant difference in survival for cirrhotic Thisisaprogressiveobliterativecholangiopathyof and non-cirrhotic variants. unknown etiology that accounts for nearly 50% of children referred for LT. Kasai portoenterostomy is the best first treatment option if performed within Alagille syndrome 2 months of life, leading to clearance of jaundice in This is an autosomal-dominant disorder that affects 80% of the cases. This success has been also linked liver (intrahepatic bile duct paucity), heart (peripheral to the experience of the surgical center performing pulmonary stenosis, tetralogy of Fallot), eyes (poste- the procedure. If complete correction of serum biliru- rior embryotoxon), and skeleton (butterfly vertebra, bin is achieved, 50% of these children will keep their short stature) with characteristic dysmorphic facial native livers up to their teenage years, but eventu- features. A third of the patients with this condition ally 70–80% will require LT. Progressive cholestasis, require LT, and a thorough multi-organ assessment recurrent episodes of cholangitis, cirrhosis with syn- is necessary before LT, as cardiac problems may con- thetic function failure, severe portal hypertension, traindicate the procedure. Renal tubular acidosis can

221 Section 4: Liver

Table 26.1 Chronic liver diseases in children that may require treatment with liver transplantation Extrahepatic biliary atresia Non-syndromic Isolated biliary atresia Progressive familial intrahepatic cholestasis (PFIC) Biliary atresia splenic malformation (BASM) syndrome Normal gamma glutamyl transferase (GGT) Metabolic liver disease PFIC-1 (Byler disease) Cirrhotic PFIC-2 (Bile salt export pump [BSEP] deficiency) Elevated GGT ␣1-antitrypsin deficiency Wilson’s disease MDR3 deficiency Tyrosinemia type 1 Idiopathic neonatal hepatitis Cystic fibrosis Parenteral nutrition related Respiratory chain disorders Chronic hepatitis Glycogen storage disease type IV Autoimmune Bile acid synthetic defect Autoimmune hepatitis (AIH) Non-cirrhotic AIH type 1 (smooth muscle and/or antinuclear antibodies) Crigler-Najjar syndrome type 1 AIH type 2 (anti liver kidney microsome type 1 antibodies) Glycogen storage disease type Ia and Ib Autoimmune sclerosing cholangitis Propionic acidemia Infectious origin Urea cycle deficiency Postviral hepatitis (B, C, others) Ornithine transcarbamylase (OTC) deficiency Liver tumors Arginosuccinate lyase (ALS) deficiency Benign Carbamoylphosphate synthetase (CPS1) deficiency Hemangioendothelioma (if large and non-resectable) Citrullinemia Malignant Primary hyperoxaluria Hepatoblastoma (if non-resectable) Familial hypercholesterolemia Hepatocellular carcinoma (if small with no metastasis) Factor VII deficiency Fibrolamellar carcinoma (if small with no metastasis) Protein C deficiency Others Intrahepatic cholestasis Polycystic liver disease Syndromic Budd-Chiari syndrome Alagille syndrome Cryptogenic cirrhosis

cause renal impairment, whereas intracranial bleed- autoimmune disorders include inflammatory bowel ing might be associated with a generalized vascu- disease, thyroiditis, nephrotic syndrome, and diabetes. lopathy caused by the disease. In our experience, Immunosuppressive therapy is the cornerstone dynamic cardiac testing with cardiac catheterization of treatment for AIH, with initial high doses of and dobutamine stress resulting in an increase in car- prednisolone (with or without AZA) followed by diac output greater than 40% indicates that the candi- progressive tapering over a period of 4–8 weeks. dates may safely undergo LT. Profound cholestasis and Second-line therapy for difficult-to-treat patients intractable pruritus are common indications for trans- includes mycophenolate mofetil (MMF) and at times plantation in these patients; cirrhotic decompensation tacrolimus (TAC) or cyclosporine (CyA). Normal- and development of HCC accelerate the need. Despite ization of serum transaminases is seen after 6–9 satisfactory 1-year survival rates after transplantation, months in 75–90%, but relapse occurs in up to 40% of the benefit seen in this systemic disease is limited, and patients and long-term treatment is usually necessary. catch-up growth is generally not satisfactory. LT is eventually required, usually several years after diagnosis in 10–20% of patients. A high incidence of chronic graft hepatitis is seen after LT, suggesting Autoimmune hepatitis disease recurrence (10–35% of cases between 15–30 There are two types of autoimmune hepatitis (AIH) months post-transplant). In patients who undergo in childhood: type 1, characterized by smooth muscle transplantation for AIH, it is prudent to continue with antibody (SMA) and antinuclear antibodies, and type steroid-based immunosuppression after LT to prevent 2, characterized by anti-liver kidney microsomal type recurrence. 1 antibody. Children with AIH frequently present acutely, particularly those with type 2, who can even develop fulminant hepatic failure, whereas a Hepatoblastoma third have an insidious onset, often with flares and This is the most common malignant liver tumor in remissions, compatible with chronic liver disease children. Its prognosis has been drastically altered and cirrhosis, particularly in type 1 AIH. Associated by effective chemotherapy. Treatment schedules of

222 Chapter 26: Pediatric liver transplantation the International Childhood Liver Tumour Strat- Table 26.2 Causes of acute liver failure in children that may egy Group of the International Society of Pediatric require treatment with liver transplantation Oncology (SIOPEL) have led to excellent response Viral to chemotherapy, allowing previously unresectable Viralhepatitis,AtoE Seronegative hepatitis (non A–E) tumors to be downsized to the point of resectability Adenovirus, parvovirus, herpes simplex in more than 60% of cases, with very good survival Metabolic rates. Those tumors that remain unresectable can be Neonatal hemochromatosis treated by LT (with grafts from cadaveric or living Tyrosinemia type 1 Wilson’s disease donors). Tumor recurrence is still the most common Fatty acid oxidation defects cause of death (20–25%) and in general is related to Galactosemia the presence of previous lung metastases. One or two Drug- or toxin-induced courses of post-LT chemotherapy may be given, but Acetaminophen overdose this remains controversial. Mushroom poisoning Autoimmune Autoimmune hepatitis (type 2 and type 1) Acute liver failure Giant cell hepatitis with Coombs-positive hemolytic anemia Acute liver failure (ALF) is rare in children, but is asso- (it may recur post LT) ciated with significant mortality (up to 70% without LT). The presence of encephalopathy is a clear indica- Attempts to support liver function until transplanta- tion for LT in adults with ALF, but in children it is more tion by using liver assist devices and hepatocyte trans- difficult to identify. Severe impairment of liver func- plantation are still in the experimental phase. The use tion requiring transplantation is defined by an INR of APOLT for ALF in children has produced patient Ͼ4,severemetabolicacidosis,cardiovascularinsta- survival rates (85%) comparable to those of whole liver bility, rapidly shrinking liver on ultrasound, and the replacement, with the added advantage of withdrawal presence of renal failure. Cerebral edema, with raised of immunosuppression if the native liver regenerates. intracranial pressure and potential brainstem herni- Hepatic regeneration, however, is slower in patients ation, is the most serious complication, and multi- with seronegative hepatitis when compared with other disciplinary medical management is necessary to sup- etiologies. portthepatientuntilasuitableorganisfound,asthisis the only effective treatment option. Contraindications Neonatal hemochromatosis for LT in ALF include: r Neonatal hemochromatosis is a rare disorder associ- Irreversible neurological damage (persistent fixed ated with massive intrahepatic and extrahepatic iron and dilated pupils) r deposition, and it is the most common cause of hepatic Severe respiratory failure failure in the neonatal period. There is a high recur- r Active uncontrolled sepsis rence rate within families, but the pattern of inheri- r Underlying systemic disease not correctable by tanceisstillunknown.Itpresentswithsevereliverfail- transplantation (e.g., mitochondrial cytopathies, ure at birth or immediately after, and the diagnosis is hemophagocytic lymphohistiocytosis) made when associated with at least two of the follow- ing findings: positive family history, elevated serum Causes of pediatric ALF treatable by LT are sum- ferritin levels, confirmed extrahepatic iron deposi- marized in Table 26.2, and some of the most important tion (salivary gland punch biopsy), and iron overload etiologies are described below. detected by magnetic resonance imaging. Initial medi- cal treatment is attempted with anti-oxidant and iron- Acute viral hepatitis chelating agents; however, their efficacy has been ques- Acute viral hepatitis is the most common cause of tioned, and LT is recommended early, usually when ALF in children. It is predominantly seronegative in severe coagulopathy or encephalopathy is present. the United States and Europe, being characterized The small size of these newborns represents a chal- by severe impairment of liver synthetic function and lenge in obtaining a size-match organ; therefore, while bone marrow failure in 10% of cases. Hepatitis A waiting for transplantation, exchange transfusions are and E are the most frequent cause of ALF in Asia. recommended if INR is Ͼ5. Extensive neurological

223 Section 4: Liver

assessment must be performed before the surgical pro- dren. Although there is no lower age donor limit, the cedure. Despite all these precautions, patient survival use of livers from donors less than 6 months of age rates after LT range from 50–60%; however, survivors has been associated with a higher incidence of hepatic have a good long-term outcome. artery thrombosis. Another increasingly important source of liver Wilson’s disease grafts is from donation after cardiac death (DCD) donors. It is possible to use liver grafts from DCD Wilson’s disease is an autosomal-recessive disorder of donors for pediatric recipients with satisfactory copper metabolism that affects children after 3 years of medium-term outcome. Only the best DCD donors, age and may present as decompensated chronic liver which fulfill strict quality criteria, should be selected diseaseorasacuteliverfailure.Diagnosisismade forreducedorsplitLTforchildren. when the liver disease is associated with two of the following criteria: positive family history, low serum ceruloplasmin, high liver copper, presence of Kayser- Liver transplant procedure Fleischer rings on eye examination, elevated 24-hour Thetransplantprocedureinachildcommenceswith urinary copper excretion (before and/or after peni- a bilateral subcostal incision, careful mobilization of cillamine challenge), and Coombs negative hemolytic the anatomical structures around the liver, particularly anemia. The insidious chronic presentation can be when the recipient has had previous operations result- successfully treated with long-term chelating agents, ingindenseadhesions(e.g.,Kasaiportoenterostomy), including penicillamine (or trientine) alone or in com- andexcisionofthediseasedliverwithorwithoutthe bination with zinc acetate or sulphate, but acute pre- retro-hepatic vena cava. The implantation technique sentation with encephalopathy is generally fatal, and will depend on the type of graft used. LTistheonlyeffectiveoption.Forthosecaseswith ALF but no encephalopathy, a prognostic scoring sys- Whole liver graft tem has been devised to predict mortality and the need Orthotopic liver replacement using a whole-size pedi- for LT based on readily available blood tests, including atric graft is accomplished by anastomosing the vena bilirubin, INR, aspartate aminotransferase, albumin, cava, portal vein, and hepatic artery to the corre- and white blood cell count. In our experience, children sponding structures of the recipient and allowing with a score of greater than 11 die without transplan- blood reperfusion. Biliary reconstruction is generally tation, but larger studies are necessary to validate this achieved by end-to-end common bile duct anastomo- scoring system. sis or hepatico-jejunostomy. Surgical issues Reduced size liver grafts Adult donor livers can be cut down to provide smaller Donor liver grafts grafts, based on the segmental anatomy of the liver. The Donor liver grafts for children are most commonly graft can be reduced to a left lateral segment (segments obtained from donation after brain death (DBD) II and III), the use of which can overcome a donor donors. Both donor and recipient should be ABO to recipient size discrepancy of 10:1. Use of the left or blood group compatible, but preferably an identical right lobe overcomes lesser degrees of size discrepancy. ABO-matched graft is used; however, under the age of Further reduction of the left lateral segment is possible 1 year, grafts from incompatible blood group donors in order to provide a single segment graft (monoseg- can be used safely. ment) to implant in very small babies. In our center, Pediatric donors who could provide size-matched thepartofthelivernotusedfortherecipientaftergraft organs are not common, and surgical techniques that reduction is not discarded but rather is used for hepa- allow the use of partial liver grafts for children from tocyte transplantation. adult donors (reduced or split or living donor LT) have had a great impact on reducing the number of chil- Split liver grafts dren dying on the waiting list. There is no upper age Split LT provides two grafts from a single donor, the limit for liver donation, but livers from donors over left lateral segment for a child and the right lobe for an 40 years of age tend to be used less commonly in chil- adult. Excellent patient and graft 1-year survival rates

224 Chapter 26: Pediatric liver transplantation

Figure 26.1 Split liver grafts. The splitting technique allows two patients to receive transplants from a single donor. The left lateral segment usually goes to a child and the right lobe to an adult. Figure 26.2 Living-related liver transplant. The left lateral segment from a living donor is generally used in pediatric liver transplantation. have been achieved (90% and 87%, respectively; Fig- ure 26.1). The liver is divided just to the right of the falciform ligament providing a left lateral segment for technique includes the partial resection of the native a child and an extended right lobe (segments IV–VIII) liver (generally the left lateral segment) followed by for an adult. The left hepatic artery and left portal vein the implantation of a size-matched partial graft. In are allocated to the pediatric graft if they are of a suit- ALF, this operation is technically more demanding able size for a safe anastomosis. In most cases, the divi- than replacing the whole native liver, but allows sion at the hilum will result in one bile duct for the withdrawal of immunosuppression when the native left lateral segment; but rarely, it might be necessary to liver has recovered from the acute injury. In metabolic perform two hepatico-jejuno anastomoses for biliary diseases, such as Crigler-Najjar syndrome, urea cycle reconstruction. defects and propionic acidemia, and in coagulation factor deficiencies, auxiliary grafting provides enough Live donor grafts (see Chapter 22) hepatocyte mass to correct the enzymatic/synthetic defect and at the same time preserves otherwise nor- This surgical technique utilizes the left lateral segment mal native liver to serve as a “safety net” in case of graft fromalivingrelateddonor(Figure 26.2). It was first failure. performed in Brazil, but successful cases were subse- quently carried out in Australia and the United States. Although it is performed worldwide with remarkable Immunosuppression success, the actual donor morbidity/mortality remains TAC is now the preferred agent for maintenance unknown due to the lack of accurate reporting. In our immunosuppression in pediatric LT. In the first center, live donation for children has become a rou- 3 months, immunosuppression generally requires the tine operation, but a strict protocol to select the donors use of steroids, which are rapidly weaned or with- is essential. The advantages of this surgical modality drawn in the majority of children. In our center, include reduction of the time on the waiting list; sched- steroids are rapidly weaned to a very low long- uled, planned procedure; and shorter cold preserva- term maintenance dose (1–2.5 mg/day) together with tion times. Recipient survival is over 90% at 1 year, low TAC levels (∼5 mg/l) starting from 3 months and for countries lacking cadaveric donation, living after surgery. With the introduction of calcineurin donation LT represents the only source of grafts for the inhibitors (CNIs), the rate of acute cellular rejection pediatric population. (ACR) has steadily decreased, but the better long-term outcome has led to an increase in CNI-related adverse Auxiliary liver graft effects. CNI-induced renal impairment can be as high Some pediatric patients with ALF or with liver-based as 60% in the pediatric transplant population, eventu- metabolic disorders can be treated by APOLT. This ally leading to end-stage renal disease requiring renal

225 Section 4: Liver

transplantation. CNI-sparing regimes have been advo- retrieval). However, medical factors can also con- catedinvolvingeitherreductionindrugexposure tribute to this complication, such as underlying pro- and/or introduction of other immunosuppressive thrombotic disorders (anticardiolipin antibody), high drugs. Dose reduction of CNIs can be achieved early by hematocrit and raised blood viscosity, poor fluid man- using induction with interleukin-2 receptor antibod- agement and arterial hypotension, and severe ACR ies or anti-thymocyte globulin (ATG), or early after causing liver swelling and increased resistance to arte- LT (around 3 months) with the addition of adjuvant rial flow. Early recognition (scheduled Doppler ultra- drugs such as MMF, sirolimus (SRL), or everolimus. In sound and urgent CT angiography) and immediate cases of serious CNI-related side effects, it is possible surgical revascularization may salvage the graft in up to replace the CNI with MMF, but corticosteroid dose to one third of patients. should be increased. Late HAT (several years after transplantation) can occur, and the intrahepatic arterial supply invari- Post-transplant complications ably is revascularized by collateral vessels. Presen- tation is often subtle and the complications include The majority of complications and deaths occur within mild liver dysfunction, late biliary stricture, recurrent the first 3 months after LT. Hence early recognition low-grade cholangitis, intraparenchymal abscesses, and correction of these complications is essential to and bacteremia. If hepatic collateralization is ade- improve graft and patient survival. quate, management is mainly conservative, but if the hepatic injury is severe, re-transplantation is Primary graft dysfunction required. Primary graft dysfunction (PGD) is relatively rare (2– 5%), and risk factors include donor characteristics, Portal vein thrombosis or stenosis organ preservation, and technical or immunological Portal vein thrombosis (PVT) occurs in approximately complications in the recipient. Signs of poor graft 2–15% of liver transplants, particularly in small chil- function include hemodynamic instability, the need dren with a hypoplastic portal vein that is mainly for inotrope support, metabolic acidosis, and coagu- associated with extrahepatic biliary atresia (EHBA). lopathy. Urgent re-transplantation is the only real ther- Surgical technical problems (e.g., a portal vein too apeutic option, but it carries significantly increased short causing stretching or too long causing kink- risk compared with primary LT. ing) and previous PVT are recognized risk factors. Early PVT presents with graft dysfunction or gastroin- Postoperative bleeding testinal bleeding, and the diagnosis is confirmed by Risk factors for post-transplant bleeding include poor Doppler ultrasound and aortoportography. Sequential graft function, renal failure, hemodialysis, and large measurement of the spleen size is often a good guide intra-operative blood loss. Bleeding from the cut regarding the continuing presence or development of surface of a partial graft may reflect venous out- portal hypertension, which is generally associated with flow obstruction. Exploratory laparotomy and surgical a low platelet count and slight prolongation of the INR. hemostasis is needed if the correction of a low platelet Early surgical intervention to restore portal venous count or low fibrinogen does not achieve hemostasis. flow will usually rescue the graft. However, in up to 50% of cases, it is not possible to Late portal vein complications present with signs identify an active bleeding site during surgery. of portal hypertension such as variceal hemorrhage or splenomegaly. The underlying cause may be tech- Hepatic artery thrombosis nical, but on occasions remodeling of a partial liver graft causes stretching of the portal vein and possible Hepatic artery thrombosis (HAT) is one of the most thrombosis. serious complications, which often leads to graft loss. The incidence in children varies from 5–8% and is less common in partial than full size liver grafts. Early Venous outflow obstruction HAT is generally linked to surgical risk factors includ- This is uncommon and represents technical failure in ing poor surgical technique, kinking of the artery, the anastomosis of the graft to the hepatic veins or IVC. and previous endothelial injury (caused during organ It may cause serious difficulty in controlling bleeding

226 Chapter 26: Pediatric liver transplantation from the cut surface of the graft or in the longer term Other surgical complications a clinical picture similar to Budd-Chiari syndrome. Perforationofthesmallorlargebowelisuncommon Suprahepatic caval stenosis is also rare and usually (6%) except in children with EHBA who have under- due to technical shortcomings. It presents with bilat- gone previous surgery, in whom the incidence is over eral leg and lower trunk edema, renal impairment, and 10%. A low threshold for emergency laparotomy is at times ascites. Doppler ultrasound, cavography, and needed to correct this complication and avoid a seri- pressure measurements will confirm the presence of ous septic episode. Re-perforation is also a common a significant gradient across the stenosis. Dilatation event. using interventional radiological techniques may solve the problem. Immunological complications ACR occurs in 40–60% of children within the first month, usually between 5 and l5 days after LT. ACR is Biliary complications suspected when there is a rise in the serum transam- In children, biliary drainage at LT is re-established inase levels, but the diagnosis is only confirmed by by bile duct–to–bile duct anastomosis (if the graft liver biopsy. The typical histological features are ofa is size-matched) or more commonly by hepatico- dense periportal cellular lymphocytic and eosinophilic jejunostomy (if a partial graft from an adult). Biliary infiltration with endotheliitis and bile duct damage. leaks are the most common technical complications, The treatment of ACR is based on increasing immuno- occurring in 5–30% of children, and they might arise suppression, generally with daily boluses of methyl- from the biliary anastomosis, the cut surface of par- prednisolone (10 mg/kg/day for 3 consecutive days). tial liver grafts, the T-tube insertion site (if used), and Steroid-resistant rejection occurs in approximately 6– unrecognized segmental bile duct left opened (gen- 30% of these patients, and rescue therapy with MMF erally, segment IV bile duct). Diagnosis can easily seems to be effective in achieving good long-term graft be made when bile appears in the abdominal drains, function. The use of ATG or muromonab-CD3 for butitcanpresentinsidiouslywithfeverormildgraft ACR carries numerous adverse effects, including a sig- dysfunction, and only ultrasound-guided drainage of nificant risk of over-immunosuppression. Grafts that a collection will confirm the problem. Endoscopic do not respond to rescue therapy have a high chance or percutaneous cholangiography and stenting of the of progressing to chronic rejection. common bile duct will usually lead to resolution. Late cellular rejection (LCR) can happen at Anastomotic biliary strictures occur in 10–35% of 6 months or more post-transplantation, and it is not cases and generally within the first year after trans- uncommon in children (8% of cases). It seems to be plantation. They might be related to a previous bile related to inadequate immunosuppression due to nec- leak episode and present with cholestasis, cholangi- essary dose reductions to avoid for example sepsis or tis, or features of biliary obstruction on liver function post-transplant lymphoproliferative disorder (PTLD), tests (mainly elevated gamma glutamyl transferase) or but also to abnormal intestinal absorption or patients’ histology. Dilatation of the biliary tree may be seen poor adherence to medication. LCR can progress to on liver ultrasound. Endoscopic or percutaneous tran- chronic rejection and can also be complicated by the shepatic balloon dilatation and, if indicated, place- onset of de novo AIH. ment of a biliary stent leads to the resolution of the Chronic rejection used to be a significant cause of majority of early strictures. Surgical reconstruction graft loss, but its incidence is decreasing (5–10%). It (e.g., hepatico-jejunostomy) is necessary if the stric- generally happens within the first year post-transplant, ture does not resolve with repeated dilatations. with jaundice, pruritus, and elevated cholestatic liver Intrahepatic diffuse cholangiopathy is less fre- function tests (alkaline phosphatase and gamma glu- quent and tends to present late (often Ͼ12 months tamyl transferase), with only mild to moderate eleva- after transplantation) and is associated with hepatic tion of the serum transaminases and preserved hep- artery thrombosis, preservation injury, and use of atic synthetic function. The diagnosis is histologi- ABO-incompatible grafts. Very few patients can be cal and characterized by loss of bile ducts and arte- managed conservatively, and the majority require re- riopathy. The treatment is to increase immunosup- transplantation. pression with TAC and/or SRL. This may reverse the

227 Section 4: Liver

changes in some children, but the majority progress to with Aspergillus, Cryptococcus, and mucormycosis and re-transplantation. coccidioidomycosis, although less common, may carry a high mortality risk. Because of nephrotoxicity associ- ated with the use of conventional amphotericin B, the Infection liposomal form of the drug is preferred. Other options Bacterial, viral, or fungal infections are frequent in the such as itraconazole or voriconazole require careful postoperative recovery period. Risk factors pretrans- monitoring of graft function and immunosuppressive plantation include younger age (increased susceptibil- drug levels. ity to pathogens like coagulase–negative staphylococci Herpes viruses are the most important viral or respiratory syncytial virus [RSV]), etiology (trans- pathogens post-transplant. CMV infection generally plant for CF, biliary atresia and cholangitis, chronic presents between 1 and 3 months after LT in chil- liver diseases with spontaneous bacterial peritonitis), dren, and it can be caused by primary infection and donor infections (cytomegalovirus [CMV] and (e.g., from donor graft or blood products), reactiva- Epstein-Barr virus [EBV]). Post-transplant risk factors tion of latent infection, or superinfection with dif- include poor graft function, graft ischemia (e.g., HAT), ferent CMV strain. Prophylaxis with ganciclovir has prolonged ICU stay, ventilator dependence, indwelling decreased the rate and severity of CMV infection and catheters, gut perforation, over-immunosuppression, should be used in high-risk cases (CMV-seropositive and re-transplantation. Bacterial and fungal infec- donor/seronegative recipient); however, prophylaxis tions are a common early problem after LT (within may simply delay the onset of CMV disease. Monitor- the first month), whereas viral infection is more fre- ingofantigenemia(pp65assayorCMVDNApoly- quent later (CMV, EBV). Gram-negative septicemia merase chain reaction [PCR]) has had a significant and systemic fungal infection have a high mortality impact on recipient management. Treatment contin- risk. ues until the CMV viral load is non-detectable. Bacterial pneumonia and IV line infections occur Young children who are EBV negative before LT are earlyafterLT,andtheyshouldbetreatedaggressively. particularly at risk of developing EBV-related PTLD Gram-negative sepsis is often associated with a biliary (incidence of 4%). Determination of the EBV load by leak, ascending cholangitis, bowel perforation, or PCR predicts the risk of PTLD, but diagnosis must be graft ischemia (HAT). It is important to note that confirmed histologically in enlarged lymph nodes or immunosuppressive therapy may minimize the other affected tissues. Management of PTLD includes clinical signs of sepsis. Prophylactic anti-microbial the withdrawal of immunosuppression alone or in therapy is used for 5–7 days and is adjusted or stopped combination with antiviral agents, with a success rate after the results of perioperative cultures. Most of 67%. Failure to respond or recurrence should be established infections can be successfully treated; treated with monoclonal antibodies against CD-20 however, colonization with multiple antibiotic- B cells (rituximab) or chemotherapy, depending on resistant bacteria is becoming a serious problem that severity. requires the implementation of strict infection control Other herpes viruses, such as herpes simplex virus measures. and varicella zoster virus, potentially reactivate after The risk of fungal infections seems to be associated transplantation, but treatment with acyclovir is gener- with pre-transplant fungal colonization and pyrexia ally highly effective. and also with post-transplant immunosuppression, The incidence of adenovirus infection after LT has bacterial infections, and EBV infection. Fungal sep- decreased to 4% since the use of TAC-based immuno- sis after LT is more common in patients with previ- suppression. However, when present, it may cause ful- ous acute liver failure, liver re-transplantation, graft minant hepatitis or necrotizing pneumonitis, often dysfunction, HAT, and bowel perforation. It should due to over-immunosuppression. Other common viral be suspected in any post-transplant patient with fever pathogens associated with upper and lower respira- and high white blood count while receiving broad- tory tract infections are influenza and RSV. Vaccina- spectrum antibiotics. Candida species are the most tion and/or antiviral treatment, when available, may common pathogens; these generally respond well to help avoid significant and potentially life-threatening fluconazole, caspofungin, and micafungin. Infections complications.

228 Chapter 26: Pediatric liver transplantation

Re-transplantation family-related variables may cause significant bias in the results. The incidence of re-transplantation for LT in chil- There is a strong correlation between quality of life, dren ranges from 9–29%. Common causes for re- graft function, and complications requiring repeated transplantation are HAT (72%), PGD (19%), chronic hospital admission. After LT, the liver function tests rejection (15%), and biliary complications (10%). One- normalize in the majority of patients, physical symp- year patient survival rates of 67% have been reported tomsofend-stageliverdiseaseimproveordisappear in a multi-center study (including cases from 1996– (ascites, jaundice, pruritus, and fatigue), metabolic 2004); however, in other studies there was no survival bone disease recovers, and those who were not able difference between re-transplantation (91%) and pri- to walk before LT because of rickets or hypertrophic mary LT (92%). Emergency re-transplantation (within pulmonary osteoarthropathy are able to do so within 1 month after LT) appears to have a poorer prognosis. a year. In general, transplant patients feel they have a Allocation of a second liver because of failure of the normal level of physical function; however, HRQOL first graft may be controversial if outcome is perceived scores for physical function in this group of patients as inferior, but re-transplantation is the only ethical are lower than those in healthy children. option for the survival of those children needing this Psychosocial functioning also appears lower than procedure. in healthy children, especially in the area of educa- tion. The most common reason for delay in educa- Outcome and quality of life tional progress is missed school days before and after Short-term survival after LT in children is excellent LT, but it may be also a consequence of developmental (1-year patient and graft survival Ͼ90% for elective delay or learning disabilities (8%). However, in a study cases) with no difference according to age, indicat- assessing children who had survived 20 years after LT, ing the significant impact of improved surgical tech- 90% completed high school and 50% attended college, niques and innovations in care for smaller recipients. indicating that we can expect improvement in their LT for ALF in children still produces inferior outcomes cognitive function. (1-year survival of 75%), with early death related to Family factors play an important role in HRQOL. multi-organ failure, sepsis, and neurological complica- After LT, parents tend to view themselves as more tions. Children who survive 1 year after LT generally relaxed and able to apply disciplinary boundaries. continue to do well, with 5-, 10-, and 20-year patient However, many families have difficulties in adjusting survival rates of 86%, 82%, and 66%, respectively. to the new situation. They miss the need to nurse their Causes of late deaths include recurrence of malig- child through deteriorating health and they have anx- nancy, PTLD, and steroid-resistant rejection. iety about the child’s long-term future, as it is not The initial obstacles to survival, particularly organ possible to give an absolute assurance that the child preservation, surgical technique, and immunosup- will remain well. Parents often report that their child’s pression, have been addressed, but the problems fol- behavior is not entirely normal, and behavioral imma- lowing successful transplantation are only now begin- turity often persists in the form of acts of defiance or ning to be recognized. Major interest has now turned aggression, particularly in adolescents. towards the quality of life of long-term survivors. Adolescent transplant recipients are at increased Twenty-year LT survivors will be seen more frequently risk of medication non-adherence, with incidence over the next decade, and we should be attempting to ranging from 17–53%. For the adolescent patient, this address their needs now. is a difficult period, with significant physical and emo- Health-related quality of life (HRQOL) assesses tional changes taking place, interaction with peers, markers of overall well-being and functional out- thedesireforautonomyfromthefamily,andthe comes, including physical, psychological, and social imminent transfer of their medical care from pedi- functions. However, HRQOL assessment is very dif- atric to adult services. Furthermore, the develop- ficult in the pediatric population due to intrinsic fac- ment of secondary chronic illness due to immunosup- tors such as the age of patients (children or ado- pressive medication (e.g., renal dysfunction, weight lescents), neuro-developmental considerations, and gain, hypertension) carries a negative impact on source of reporting (parent or self-reporting), and their HRQOL. Other factors increasing the risk of

229 Section 4: Liver

non-adherence include inadequate knowledge of the Further reading medication regime, depression and anxiety, substance Bartlett A, Rela M. Progress in surgical techniques in abuse, life stress, and inadequate relationship with the pediatric liver transplantation. Pediatr Transplant 2010; medical team. Transitional care programs for pediatric 14: 33–40. LT patients have been advocated to deal with these Bucuvalas J. Long-term outcomes in pediatric liver problems and to help in the gradual shifting of man- transplantation. Liver Transpl 2009; 15: S6–11. agement skills from the medical team and parents to Dhawan A. Etiology and prognosis of acute liver failure in the adolescents themselves. children. Liver Transpl 2008; 14: S80–4. As the life expectancy of children undergoing DuffyJ,KaoK,KoC,et al. Long-term patient outcome and transplantation continues to increase, we must be pre- quality of life after liver transplantation: analysis of pared for their future needs. These needs cannot be 20-year survivors. Ann Surg 2010; 252: 652–61. measured just as clinical outcomes after LT, but they Hartley J, Davenport M, Kelly D. Biliary atresia. Lancet should also consider improvement in quality of life. 2009; 374: 1704–13. It is essential to ensure that important socioeconomic Mieli-Vergani G, Vergani D. Autoimmune paediatric liver milestones are achieved by these patients: completing disease. World J Gastroenterol 2008; 14: 3360–7. education, being in employment, and having their own Muiesan P, Vergani D, Mieli-Vergani G. Liver families. It is likely that in the future, these factors will transplantation in children. JHepatol2007; 46: influence the timing of transplantation. 340–8.

230 Section 5 Kidney Chapter Recipient selection

27 Ernest I. Mandel and Nina E. Tolkoff-Rubin

Key points Timing of referral Observational studies suggest that survival after trans- r Thorough pretransplant evaluation is plant is worse the longer a patient is on dialysis. essential for maximization of both patient Furthermore, patient and graft survival are improved and graft survival. in patients transplanted preemptively, that is, prior r Cardiovascular evaluation and screening to the initiation of dialysis. Except for a zero anti- should be performed in nearly all patients. genmismatch(forwhichpatientscanbeeligibleatan r Screening for other chronic conditions, estimated glomerular filtration rate [eGFR] less than including infections and malignancy, 20 ml/min/1.73 m2), patients in the United States can- should be performed, and they should be not undergo transplantation with a deceased donor managed appropriately prior to kidney preemptively. Nonetheless, patients can be transplantation. 2 r listed once their eGFR is below 20 ml/min/1.73 m , Morbidity and mortality, as well as and, if they can identify a donor, preemptive living quality-of-life considerations, should be donor kidney transplantation should be pursued. Fur- taken into account when considering older thermore, the pretransplant evaluation and testing can candidates. r take several months to complete, potentially delaying A multi-disciplinary approach considering addition to the waiting list, and, therefore, especially in cognitive and other psychosocial factors is diabetics and others likely to progress to ESRD quickly, necessary to ensure successful early referral is essential. For all these reasons, referral transplantation. toatransplantcentershouldlikelybemadeonceeGFR is found to be less than 30 ml/min/1.73 m2.

Kidney transplantation is the renal replacement ther- The pretransplant evaluation apy of choice for suitable patients with advanced The preoperative evaluation of transplant candidates chronic kidney disease (CKD) or end-stage renal dis- is a multi-disciplinary process that involves transplant ease (ESRD), conferring both survival advantages and surgeons,nephrologists,mentalhealthprofessionals, quality-of-life improvements over peritoneal dialysis social workers, dieticians, financial coordinators, and and hemodialysis. With more than 80 000 patients transplant coordinators. The goal of the evaluation is currently awaiting transplantation in the United States, to identify any contraindications to transplantation, to optimizing the selection of candidates is essential determine immunologic factors impacting donor kid- to ensuring the equitable allocation of this scarce ney options, to screen for comorbid conditions that resource. In addition, for those awaiting deceased need to be identified and managed prior to transplan- donor transplantation, or for those fortunate to iden- tation, and to assess psychosocial factors that could tify a potential living donor, a thorough pretrans- affect the success of the transplant. In the pages that plant evaluation is needed to maximize both graft and follow, each of these components of the pretransplan- patient survival. tation evaluation will be addressed.

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

231 Section 5: Kidney

Table 27.1 Contraindications to transplantation and human leukocyte antigen (HLA) typing (A, B, DR, Active malignancy and at some centers, DQ), is performed to help iden- Untreated infection tify both potential matches and to preclude recipient- donor pairings that would have poor outcomes. Addi- Severe comorbid conditions: cardiac, pulmonary, or vascular disease tionally, patients with advanced kidney disease but not yet requiring dialysis can be listed preemptively for a Unable to manage: uncontrolled psychiatric illness or psychotic behavior, persistent substance abuse, or significant history of zero antigen mismatch kidney, highlighting the impor- medical non-compliance tance of timely referral. Primary oxalosis Recipient serum is also tested against a panel of Irreversibly limited potential for rehabilitation cells representing a range of antigens in a local pop- ulation. Known as the panel-reactive antibody (PRA), it is expressed as a percentage and is understood as the Contraindications to transplantation percentage of potential donor antigens against which There are several absolute and relative contraindica- the recipient’s serum may react. Higher PRA suggests tions to transplantation (Table 27.1). The greatest mor- more difficulty finding a match, and as such, patients bidity and mortality from transplantation occurs in in the United States with PRA greater than 80% receive thefirstyearaftertransplantandconsistsofboth priority in organ allocation. The panel can vary from cardiovascular and infectious complications. As such, center to center and within center between different patients with unmanaged cardiovascular disease and time periods. Because of this variation by location and untreated or indolent infections should not undergo time, the peak, or highest, PRA is used in determin- transplantation until these are addressed. In addi- ing priority in organ allocation when a donor organ is tion, to maximize the benefit of transplantation and identified. Single-antigen bead technology has allowed the allocation thereof, patients with high near-term the exact determination of HLA sensitization, so for (≤2 years) mortality, such as those with advanced or many patients, virtual cross-matching confirms suit- uncontrolled malignancy or severe non-renal disease, ability of a particular donor organ (discussed further should likely not undergo transplantation. Patients in Chapter 28). with histories of medication non-adherence or psychi- atric illness or social circumstances, including recre- ABO incompatibility and desensitization ational drug abuse, that may limit their ability to adhere to post-transplantation care need to prop- protocols erly address these concerns prior to transplantation. Advances in immunobiology have led to the develop- Finally, the candidacy of patients with primary oxalo- ment of desensitization protocols to allow for trans- sis should be considered carefully, with simultane- plantation between ABO-incompatible or sensitized ous liver and kidney transplantation being the option individuals. These protocols include some combina- of choice. Several of these contraindications can be tion of plasmapheresis, intravenous immunoglobulin modifiedorintervenedonpriortotransplantation, G, and preoperative immunosuppressants along with as will be discussed. Identification of these abso- appropriate antibiotics for cytomegalovirus (CMV) lute contraindications or modifiable, relative con- and Pneumocystis prophylaxis in a now immunosup- traindications is the purpose of the pretransplantation pressed host. Some protocols also include rituximab. evaluation. At the pretransplantation visit, emphasis should be placed on identifying all potential living donors so that Immunology the latest protocols can be employed when possible. Tissue typing and histocompatibility Paired exchange and kidney list donation Immunologic evaluation begins with a thorough his- In recent years, severe organ shortage has led to inno- tory of potential antigen exposure, including prior vative solutions. One such solution is paired organ transplantation of any kind, blood product transfu- exchange, whereby recipients who have identified a sion, and, in female candidates, prior pregnancy. His- willing donor who is otherwise not able to donate to tocompatibility testing, including blood typing (ABO) them due to immunologic factors are paired with a

232 Chapter 27: Recipient selection similarly situated individual when the potential donors impact transplant outcome. Moreover, severe heart match each others’ intended recipients. An alternative failure and severe valvular disease may impact candi- when a suitable living donor exchange is not possi- dacy for transplantation surgery. Calcification associ- ble involves pairing a recipient and their willing but ated with CKD places these patients at particular risk incompatible donor with a recipient on the waiting list. for developing valvular disease and subsequent conse- The recipient at the top of the waiting list in the donor’s quences. As such, it is reasonable to assess function and blood group receives the living donor kidney, while the valvular disease prior to transplantation via transtho- donor’s recipient is given priority for a deceased donor racic echocardiography. kidney. These exchanges can be fraught with anxiety and disappointment, such that the setting of realistic Peripheral vascular disease expectations is essential at the first, and at subsequent, pretransplant visits. Peripheral vascular disease is important both as a risk factorforischemicheartdiseaseaswellasanimpor- tant consideration in operative planning. Signs and Cardiovascular disease symptoms of peripheral vascular disease, such as clau- Cardiovascular disease is the leading cause of death, dication; rest pain; past vascular insufficiency, includ- and therefore graft loss, in the first year post- ing history of ulcerations, infections, or amputations; transplant. As such, preoperative assessment of cardio- and in male patients, erectile dysfunction, should be vascular disease is essential to improving patient and elicited. Physical examination should include a thor- graft survival. ough assessment of peripheral arterial pulses and lower extremities for signs of vascular insufficiency Ischemic heart disease such as hair loss or ulcers. Symptoms or signs of peripheral arterial disease should be followed up with All patients presenting for transplantation evaluation non-invasive vascular studies, and further evaluation are at increased risk for ischemic heart disease by and management initiated based on these results. virtue of their chronic kidney disease. For many, Finally, in consultation with local transplant surgeons, the underlying cause of kidney disease, particularly CT of the iliac vessels without contrast may be indi- diabetes and hypertension, is also associated with cated for assessment of vascular calcification and oper- increased risk. The preoperative history must focus on ative planning. risk factors for and manifestations of ischemic heart disease, including Framingham risk factors, prior history of ischemic heart disease, and current signs Tobacco use and symptoms of ischemic heart disease, including Because of its strong association with cardiovascu- assessment of exercise tolerance. Physical exam should lar disease in general, active tobacco use should be include a thorough cardiovascular exam, and an elec- explored and cessation encouraged prior to trans- trocardiogram should be obtained. Nearly all patients plantation. This should be re-addressed at each sub- should undergo non-invasive testing, with many sequent visit and medical management initiated or centers choosing dobutamine stress echocardiography referral made to available cessation resources where or nuclear sestamibi (Thallium stress test, myocardial applicable. perfusion scintigraphy). At least one study of pre- transplantation elective revascularization in diabetics Hypercoagulable states has been shown to confer a survival advantage. Results of non-invasive testing that are suggestive of ischemic Patients with thrombophilic states are at risk for heart disease should therefore be followed up with early graft loss. Many patients on hemodialysis via cardiac catheterization and intervention where fistulae or grafts have experienced access clotting. appropriate. Patients with systemic lupus erythematosus are at risk for anti-phospholipid antibody syndrome. Patients with a history of thrombotic events, whether clas- Structural heart disease sic venous thromboembolism, access clots, or more Systolic and diastolic heart failure, as well as signifi- subtle suggestions such as recurrent miscarriages in cant valvular disease that can lead to heart failure, may women, should be screened for hypercoagulable states

233 Section 5: Kidney

including factor V Leiden, prothrombin gene muta- donor kidneys from donors with known HCV of a less tion, protein C and S deficiencies, anti-thrombin III virulent genotype, although this remains an area of deficiency, hyperhomocysteinemia, and antiphospho- controversy. At the very least, knowledge of the geno- lipid antibody syndrome (anti-cardiolipin and lupus type would be important in making this decision, and anti-coagulant). If a thrombophilic state is identified, so the potential recipient’s genotyping should be deter- anti-coagulant therapy should be initiated and contin- mined as part of the pretransplant evaluation. Finally, ued as indicated. asyet,thereisnoavailablepost-transplantprophylaxis against HCV. Infection Human immunodeficiency virus Pretransplant assessment of known infections and screening for other as yet unknown ones is necessary, Successful management of human immunodeficiency both to ensure the safety of planned immunosuppres- virus (HIV) with anti-retroviral therapy has made sion, as well as to determine, and possibly expand, it possible to offer transplantation to HIV-infected the pool of donor candidates from which the potential patients while minimizing their infection risk in the recipient can draw. setting of postoperative immunosuppression. Patients with HIV may therefore be viable transplant candi- dates, although many units would still consider HIV a Hepatitis B virus contraindication. Preoperative infectious disease eval- Screening for hepatitis B virus (HBV) should be per- uation should be obtained to ensure optimal disease formed pretransplant, including both surface and core management, including achieving an undetectable antibody testing. Viral load should be tested in all sur- viral load, and to review antiretroviral therapy for pos- face antigen–positive patients, as well as in surface sible interactions with post-transplant immunosup- antigen–negative patients who are core antibody pos- pressive agents. itive to rule out acute infection. HBV e antigen test- ing should be performed in screen-positive patients Tuberculosis as well. Patients with HBV should be referred for All transplant candidates, particularly those who are possible liver biopsy to assess the extent of liver dis- from or have traveled to endemic countries, or have ease, which would inform both transplant candidacy other risk factors for mycobacterial exposure or dis- and the possible need for simultaneous liver trans- ease, should have a tuberculin skin test as part of plant. In considering transplantation for patients with the transplant evaluation. Alternatively, blood antigen HBV, patients who are HBV surface antibody positive testing for tuberculosis (TB) can be performed, espe- may receive a kidney transplant from an HBV core cially in patients at high risk for exposure or those antibody–positive donor. Finally, post-transplant pro- likely to be anergic (lack of immunologic reaction to phylactic lamivudine may be indicated for HBV sup- foreign substances), yielding a false-negative skin test. pression. A positive test of either type, in the absence of a history of past treatment, should trigger further imaging and Hepatitis C virus evaluation and would necessitate therapy for latent TB before transplantation due to the risk of reactivation of As in the case of HBV, screening for hepatitis C virus latent disease with immunosuppression. Once treated, (HCV) should be performed prior to transplantation. post-transplant prophylaxis for treated latent TB is not All antibody-positive patients should have viral load indicated. checked to rule out false positives. Patients with HCV should be referred for liver biopsy and considera- tion of antiviral therapy for the same reasons men- Other infections tioned above. If antiviral therapy is indicated, it should Chronic bacterial infections, such as cellulitis, be administered pretransplant, as interferon and rib- osteomyelitis, or upper or lower urinary tract infec- avirin can precipitate acute rejection. In considering tions, should be treated and clearance documented transplantation for patients with HCV, patients with prior to transplantation. Such patients may ben- HCV genotype 1 may be considered for deceased efit from consultation with an infectious disease

234 Chapter 27: Recipient selection specialist, especially one focusing on transplant appropriate cancer screening prior to transplantation infectious disease where available. Lesions at risk for these same reasons. At the very least, this should for becoming infected, such as active diabetic foot include cervical smear and mammogram for female ulcers, should be assessed and managed prior to candidates, and colonoscopy for all candidates over transplantation. Finally, all patients should be seen by age 50 years. Candidates with HBV or HCV should a dentist to rule out any occult oral infections. undergo appropriate screening for hepatocellular car- CMV antibody should be obtained to guide post- cinoma, including serum ␣-fetoprotein and imaging. transplant prophylaxis needs coupled with donor Despite the controversial nature of prostate cancer characteristics. Screening for Epstein-Barr virus, her- screening, given the importance of an intact lower pes simplex virus, syphilis (rapid plasma reagin and urinary tract for the success of the transplant and fluorescent treponemal antibody), and human T- the risk of outlet obstruction and allograft obstructive lymphotrophic virus-1 is recommended as well. uropathy,aswellasrecipientmortalityfromaggres- Screening for chronic infections not already dis- sive prostate cancer, screening with prostate-specific cussed previously, such as schistosomiasis, strongy- antigen is advisable. Finally, especially with the risk of loides, histoplasmosis, coccidiomycosis, and Chagas acquired cystic disease and possibly increased preva- disease, should be performed on a case-by-case basis lence of renal cell carcinoma in ESRD patients, screen- for those from endemic areas and treatment initi- ing ultrasonography of the kidneys should be per- ated pretransplant as indicated. Prophylactic treat- formed as part of the transplant work-up. Patients with ment may be needed long term after transplant suspicious lesions should undergo further evaluation. as well. If the cancerous lesion is less than 3 cm, the patient can proceed with transplantation after undergoing Prophylactic vaccination nephrectomy without a waiting period. For lesions greater than 3 cm, transplantation should be deferred Vaccination history should be obtained and any fortherequisite2-to5-yearperiodtoensureremission needed vaccines administered in advance of any prior to transplantation. planned living donor transplant or as soon as possible for those awaiting a deceased donor. Relevant vaccines include Haemophilus influenzae b; hepatitis A virus Urologic disease and HBV; human papillomavirus; measles, mumps, Because the transplanted kidney usually drains into and rubella (MMR); tetanus, diphtheria-acellular, and the native lower urinary tract, underlying urologic pertussis; polio; meningococcus; Streptococcus pneu- disease can affect the transplant outcome. Although moniae; varicella zoster virus (VZV); and influenza. there is no evidence to suggest that routine screen- Immune status should be confirmed for VZV, MMR, ing for urologic disease improves graft or patient sur- and HBV, and vaccine administered as indicated. vival, those with known urologic disease, including Finally, potential transplant recipients should be given history of congenital obstructive uropathy, bladder seasonal influenza, and any other relevant influenza dysfunction, bladder or prostate cancer, nephrolithia- vaccines, on an annual basis. sis with episodes of obstruction, or recurrent urinary tract infections, or those with symptoms suggestive Malignancy of any of the above, should be evaluated by a urolo- By general expert consensus, a prior history of malig- gist prior to transplantation. Results of such an evalu- nancy precludes transplantation until a durable remis- ation should guide decision making regarding the need sion has been achieved. This is suggested to maxi- for preoperative medical management of any urologic mize both patient survival, given the risk of disease condition, as well as the need for alternative allograft acceleration or recurrence on immunosuppressants, drainage (e.g., bladder augmentation or urinary diver- as well as graft survival, given the increased risk of sion), or in the case of recurrent infection or retained host death with untreated or uncontrolled malignancy. nidus, the need for pretransplant native nephrectomy. Dependingonthemalignancy,adisease-freeperiod Finally, it is reasonable to perform urinalysis and cul- of between 2 and 5 years is generally accepted as ade- ture on all patients immediately prior to transplanta- quate. All transplant candidates should undergo age- tion to ensure a sterile urinary tract.

235 Section 5: Kidney

Primary renal disease and risk of ered for simultaneous kidney and pancreas transplant to minimize future complications. Even when con- recurrent disease in the allograft sidering such options, a living donor kidney without Riskofrecurrentdiseaseintheallograftvariesbythe pancreas transplantation would be preferred due to etiologyoftheprimaryrenaldisease,asdoestherisk graft survival, which can then be followed by deceased of graft failure from recurrent disease. For example, donor pancreas transplant. If a living donor cannot histologically, nearly all diabetics will manifest recur- be identified, then simultaneous deceased donor pan- rent disease in the allograft, though allograft dysfunc- creas and kidney transplant would be the best option. tion from recurrent disease is unlikely. On the other In the future, diabetes management via islet cell trans- hand, a small minority of patients with idiopathic plantation, coupled with kidney transplantation, may focal segmental glomerulosclerosis (FSGS) or mem- be considered. branous nephropathy and a slightly higher percent- age of those with atypical or familial hemolytic-uremic Obesity syndrome will manifest recurrent disease that will lead Studies have shown that obesity, defined as body mass to graft failure. However, even these cases are con- 2 founded by the possible emergence of de novo dis- index (BMI) over 30 kg/m ,confersasmallbutstatisti- ease (e.g., membranous nephropathy or thrombotic cally significant increased risk of both graft failure and microangiopathy associated with certain immunosup- patient death. Obese patients also have higher peri- pressants).Becausetherateofrecurrenceaswellasthe operative risk, including risk of cardiovascular events clinical significance of recurrence (i.e., the rate of allo- and postoperative wound healing deficiencies lead- graft dysfunction resulting from recurrent disease) is ing to increased wound dehiscence and wound infec- varied, the etiology of the primary renal disease in gen- tion. Moreover, the long-term survival benefit of kid- eral should not affect the candidacy for transplanta- ney transplantation over dialysis has been shown to tion.However,oncerecurrentdiseasehasnecessitated diminish in patients with BMI greater than 40. As such, repeat transplantation, as can be the case in recurrent obese kidney transplant candidates should be strongly idiopathic FSGS, candidacy for further transplantation encouraged to lose weight either on their own or, in should be considered carefully. Given recent develop- the case of those with BMI greater than 40, possibly ment of protocols for treating recurrent FSGS in the in conjunction with weight loss surgery. Population- allograft, pretransplantation protein excretion should based studies suggest that obese patients can lose up be assessed, especially when preemptive transplanta- to 10% of their body weight in a 6-month period with tion is being performed, preferably with a 24-hour maximal effort, and in most cases this a reasonable urine collection or at least a spot protein/creatinine goal to target. The actual cutoff for BMI above which ratio. This will permit post-transplant screening for weight loss should be considered mandatory prior to recurrent disease. transplantation remains a center-by-center decision, although it would be reasonable to defer patients with BMI greater than 40 until significant weight loss has Diabetes mellitus been achieved. Patients with diabetes, whether or not diabetes is the primary cause of renal disease, comprise a large Age minority of kidney transplant candidates. Because dia- At a minimum, patients considered for transplantation betes confers increased risk for cardiovascular mor- must have completed the immunizations described bidity and mortality as well as infectious complica- above and responded appropriately. As such, although tions, management of diabetes should be optimized there is no absolute minimum, candidates for trans- prior to transplantation, from glycemic and other plantation are usually at least 1 year of age. On the metabolic control to podiatric care. Although, as men- other end of the spectrum, with the rapid growth tioned above, allograft failure from recurrent diabetic of the ESRD population over 65 years of age, older disease is rare, because diabetics are at high risk for patients account for an increasingly greater percentage cardiovascular disease and infectious complications, of transplant candidates. Morbidity and mortality out- type I, or juvenile onset, diabetics should be consid- comes in older patients undergoing transplantation are

236 Chapter 27: Recipient selection

Table 27.2 Extended criteria kidney donor and therefore graft loss, should prompt an evaluation definition and referral for counseling or treatment as appropriate Age Ͼ60 or 50–59 years and two of the following: prior to transplantation. Finally, assessment of candi- CVA as cause of death dates’ social situation including social support struc- Hypertension ture should be conducted. Terminal creatinine Ͼ1.5 mg/dl CVA: cerebrovascular accident. Managing the list Following pre-transplantation evaluation and satis- favorable, including peri-operative morbidity and factory completion of any needed testing, patients mortality as well as long-term patient and graft sur- are added to the evaluating center’s waiting list. vival. However, the rate of infectious complications Frequent re-evaluation of candidates should be per- is markedly higher with advanced age. Therefore, the formed to ensure ongoing suitability for transplan- expected survival benefit from transplantation, con- tation, although how frequently is unclear. Immuno- sidering the patient’s life expectancy with and without logical status should be automatically reassessed transplantation, as well as quality of life after trans- frequently with repeat tissue typing and antibody test- plantation as compared with dialysis, must be taken ing via updated PRA status. As for other factors, at the into account when deciding about the candidacy of very least, regular reassessment of cardiovascular and older patients. Furthermore, because older patients infectiousdiseasestatusisadvisable,possiblyasoften maynotsurvivetotransplantationduetolongwait as every 2 years. times, extended criteria donor (ECD; Table 27.2)kid- To ensure ability to accept organ offers in a timely neys should be considered for older transplant candi- fashion, in addition to regular reassessment of candi- datesasameansofshorteningwaitinglisttime. dacy, some centers maintain a second or parallel lists of candidates, such as those eligible for or willing to Psychiatric, cognitive, and consider ECD kidneys. psychosocial evaluation Further reading Neither psychiatric illness nor cognitive impairment Bunnapradist S, Danovitch GM. Evaluation of adult kidney should be considered a contraindication to trans- transplant candidates. Am J Kidney Dis 2007; 50: 890–8. plantation. However, complicated postoperative med- Kasiske BL, Cangro CB, Hariharan S, et al. The evaluation ication regimes and the need for strict medication of renal transplant candidates. Clinical practice adherence means that a thorough assessment of the guidelines. Am J Transplant 2001; 2(Suppl 1), 3–95. transplant candidate’s psychiatric health and cognitive Kasiske BL, Snyder JJ, Matas AJ, Ellison MD, Gill JS, Kausz ability should be performed by an appropriate profes- AT. Preemptive kidney transplantation: The advantage sional. A history of non-compliance may mean that and the advantaged. JAmSocNephrol2002; 13: 1358–64. candidatesshouldbegivenanopportunitytoprove Montgomery RA, Zachary AA, Ratner LE, et al. Clinical abilitytoadheretotherapy,possiblybyinsistingon results from transplanting incompatible live kidney a time-limited trial whereby the patient can demon- donor/recipient pairs using kidney paired donation. strate ability and intention to comply. A history of sub- JAMA 2005; 294: 1655–63. stance abuse, including alcohol or illicit drugs, inso- Pilmore H. Cardiac assessment for renal transplantation. far as it increases risk of non-compliance with therapy Am J Transplant 2006; 6: 659–65.

237 Section 5 Kidney Chapter Sensitization of kidney transplant recipients 28 Nick Pritchard

Key points transplantation in a highly sensitized population and outcomes in this particular group. r The most important alloantibodies are directed against human leukocyte antigen (HLA) and are formed in response to Sensitization exposure to foreign HLA molecules, usually at the time of blood product administration, Definitions of sensitization previous transplant, or during pregnancy. Sensitization is defined as the presence of preformed r The role played by non-HLA antibodies in alloantibody in the serum of prospective transplant graft injury is increasingly recognized. recipients. The most important alloantibodies are r Virtual cross-matching is the process of directed against HLA. They are formed in response predicting the likelihood of a significant to exposure to foreign HLA molecules, usually at antibody-mediated immune event between the time of blood product administration, previous recipientandgraft;itmaybeusedin transplant, or during pregnancy. More recently non- pretransplant risk assessment and can also HLA antibodies directed against endothelial cell tar- assist with the process of organ allocation. r gets have been shown to be associated with antibody Desensitization may allow transplantation of mediated rejection (AMR) and poor long-term graft presensitized patients with donor-reactive outcomes. HLA or ABO antibodies by reducing antibody levels to such a degree to make transplantation safe. Quantization of sensitization Anti-HLA antibodies are also called panel-reactive For many years, the importance of HLA antibody– antibodies (PRA) because they have historically been antigen interactions to transplant outcome has been quantified by testing the potential recipient’s serum appreciated. Matching donor and recipient HLA anti- against a “panel” of lymphocytes harvested from a gens has led to better transplant outcomes, and there is number of individuals with a wide range of HLA types. good evidence showing that responses to mismatched Using a complement-dependent cytotoxicity cross- HLA antigens are the principal element determining match (CDC) to detect the presence of antibodies in immunological outcomes after solid organ transplan- the recipient serum against lymphocyte surface anti- tation. This chapter considers the importance of recip- gens, the extent of sensitization is determined by calcu- ient sensitization with particular reference to renal latingthepercentageofdonorsinthepanelwherecell transplantation. It considers the clinical relevance of lysis occurs. To define levels of sensitization, patients both HLA and non-HLA antibodies in graft outcomes. with a PRA of 0–10% are considered to be non- As part of this, it discusses strategies for antibody test- sensitized, whereas highly sensitized patients are con- ing and application of those tests to clinical practice sidered to be those patients with a PRA of greater than and the assessment of immunological risk. Finally, it 85%. These criteria are arbitrarily determined to define looks at the strategies to improve the likelihood of subgroups of patients, but the level of sensitization

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

238 Chapter 28: Sensitization of kidney transplant recipients appears to have a graded effect on outcomes: the higher 50% of patients and non-donor HLA in the other the PRA, the poorer the outcome. 50%. Development of de novo donor-specific antibody Measurement of PRA by CDC has been largely (DSA) does not necessarily induce acute AMR, which replaced by more sensitive and less cumbersome occurs in less than 5% of renal transplants. Despite solid-phase assays, which report a calculated PRA or the absence of acute AMR, the risk of graft loss is population-reactive antibodies. The antibody speci- three times higher following the development of HLA ficities of all HLA antibodies present in the recipient antibody, and outcomes are equally poor regardless of serum are determined. Then from knowledge of the whether or not this is donor-specific. However, DSA is HLA antigen frequency in the donor population, the associated with the development of chronic AMR and proportion of donors against which the serum will transplant glomerulopathy. react can be calculated. The calculated PRA for any individual will be higher than the CDC-PRA because of the increased assay sensitivity, but it does provide a Pre-formed HLA antibody present at more reliable and standardized measure of sensitiza- the time of transplant tion. ThepresenceofHLAantibodyatthetimeoftransplant implies previous exposure to and subsequent devel- Causes of sensitization opment of persistent B-cell memory responses to the Sensitization occurs through the development of long- relevant HLA antigen. Depending on the level of anti- lived B-cell memory responses in response to expo- body present, hyperacute rejection (HAR) or acceler- sure to foreign antigen and most importantly HLA ated AMR may occur at the time of transplantation. molecules. The three primary sources of such exposure The level of antibody present is defined by laboratory are pregnancy, blood transfusion, and previous trans- testing, and this provides an assessment of immuno- plants. These sensitizing events have an additive and logical risk. interacting effect on the PRA. Beyond these three clear In addition to this humoral or B-cell sensitiza- causes for sensitization, there may be other factors tion, it is increasingly being realized that a compo- that contribute to sensitization, including, for exam- nent of preformed allo-specific T-cell memory may ple, infection. Interestingly, a significant minority of also be present in potential recipients. A number patients (approximately 10% of males and 20% of nul- of studies have shown a high frequency of donor- liparous women) receiving their first transplant with reactive memory T cells. These can occur indepen- no previous history of blood transfusion appear to be dentlyofthepresenceofhighlevelsofalloantibody sensitized (PRA Ͼ 10%). The cause for this is as yet and B-cell sensitization; T-cell sensitization is less unidentified. This may be due to failure to identify clearly related to alloantigen exposure because cross- previous transfusions and unrecognized pregnancy in reactivity within the T-cell repertoire is common, and women, but there may be some cross-reactivity or it is possible for recipient T cells to recognize peptide heterologous immunity from other previous immune associated with donor major histocompatibility com- interactions. plex molecules. The presence of such a memory T-cell response increases the risk of T-cell–mediated rejec- tion (TCMR). However, as yet it is not routine in most Clinical relevance of HLA and centers to screen for T-cell sensitization, and there is non-HLA antibodies no clear evidence that high levels of T-cell sensitization has significant impact on long-term graft and patient De novo development of HLA survival. antibody post-transplant The majority of patients receiving a first renal trans- Non-HLA antigens planthavenodetectableHLAantibodyatthetimeof With better identification and understanding of the transplant. Following transplant, approximately 25% interaction between the presence of HLA alloreactiv- of recipients will develop de novo HLA antibody, ity and cross-match results, the role played by non- and most will develop this within the first year. The HLA antibodies in graft injury is increasingly recog- HLA antibodies that develop are donor-specific in only nized. The origin of non-HLA antibodies is less clear

239 Section 5: Kidney

Table 28.1 Common molecular targets for non-HLA antibody Defining the role of non-HLA antibody–antigen responses in solid organ transplants interactions in transplant outcome is still in its infancy. Specificity Allograft type Type of rejection Clarification of the pathogenetic mechanisms by MICA Kidney, heart, pancreas HAR, AR, CR which non-HLA antibodies arise, how they contribute to graft injury, and the impact they have on long-term Vimentin Heart AR, CR outcomes is crucial and to date remains uncertain. The AT1RKidneyAR importance of non-HLA immune responses is increas- HMA-1 & -2 Kidney HAR, AR ingly being recognized, and they are likely to become ICAM-1 Heart Uncertain more important as problems related to HLA response MICA: major histocompatibility complex Class I-related chain A; are better understood and managed. AT1R: angiotensin II type 1 receptor; HMA-1 & -2: hepatic mem- brane associated antibodies-1 & -2; ICAM-1: intracellular adhe- sion molecule-1; HAR: hyper-acute rejection; AR: acute rejection; Pretransplant risk assessment CR: chronic rejection. based on cross-match and antibody screening results than HLA alloreactive antibodies. Non-HLA antibod- In most UK centers, a CDC and flow cytometry cross- ies that have been identified as having a deleteri- match (FCXM) against T and B cells is routine prac- ous effect in transplantation are heterogeneous, with tice prior to transplantation, with interpretation in a wide range of different specificities. Relevant anti- the context of up-to-date solid-phase antibody screen- bodies have been identified with reactivity against ing results. A schema incorporating results from all epithelial cells, endothelial cells, and monocytes. They oftheseassaysisshowninTable 28.2. Alongside the can occur through alloimmunization through the laboratory data defining the presence and level of any same mechanisms as HLA sensitization, and the pres- DSA, the past sensitization history must be taken into ence of non-HLA antibodies is strongly associated consideration. Relevant factors that increase the risk with HLA pre-sensitization. However, autoreactive of immune-related problems are shown in Table 28.3, non-HLA antibody may occur without sensitizing and this further refines the immunological risk of the events and are found in patients with autoimmune donor–recipient pair. disease. In general terms, the vast majority of transplants It is estimated that 10% of C4d-positive acute rejec- undertaken would be from the low-risk category, with tion episodes in renal transplants are due to non- most centers preferring to avoid higher risk proce- HLA antibodies. In HLA-identical renal grafts, non- dures. However, the availability of polyclonal anti- HLA antibodies may be responsible for up to 80% bodies such as anti-thymocyte globulin and mono- of irreversible rejection episodes. Non-HLA immune clonal agents against CD25, CD20, and CD52 for use responses contribute substantially to kidney failure, in induction protocols to reduce the risk of rejection with transplants from HLA-identical siblings failing or as treatment for refractory AMR allows success- more frequently in recipients with higher PRAs. Of ful transplantation of higher risk patients. In this way, the different factors contributing to graft loss, 38% although some categories of antigen may be absolute of graft failures are due to non-HLA antibody, 18% barriers to transplantation, some higher risk donor– are due to HLA antibody, and 43% are due to non- recipient combinations may, in certain situations, be immunological causes. seen as carrying acceptable levels of risk. Factors relat- The full range of clinically relevant specificities ing to donor, recipient, and locally available resources that non-HLA antibodies may interact with is still will play a role in the final decision regarding what con- not elucidated. However, there are an increasing num- stitutes an acceptable level of risk and how it is man- ber of antibodies against a wider range of non-HLA aged. specificities that have been identified. These have been implicated in mediating a wide range of clin- ically relevant immune injuries. A discussion of the Virtual cross-matching details of these different antibodies is beyond the With the introduction of solid-phase assays for detec- scope of this chapter, but they are summarized in tion of anti-HLA antibody, it has become possible to Table 28.1. determine the presence of HLA-DSA “virtually” by

240 Table 28.2 Immunological pretransplant risk assessment based on donor cross-match and antibody screening Cross-match method CDC-XM FCXM Current or Antibody screening T cell B cell T cell B cell historical results Interpretation

++++CurrentIgG Very high riska. Probable HAR. HLA Class I DSA –+–+CurrentIgG Very high riska. HAR unlikely (reported only in HLA high titre HLA-DR) but accelerated/refractory Class II AMR and poor long-term prognosis. DSA – + – + Current Weak High riskb. B cells express higher levels of HLA IgG Class I than T cells, thus B+/T– XM may indicate HLA low levels of HLA Class I DSA. Class I DSA – – + + Current IgG High riskb. Lower titre DSA detected by more HLA sensitive FCXM. HAR unlikely but high risk of Class I severe accelerated AMR. DSA –––+CurrentIgG Intermediate riskc. Significant risk of acute and HLA chronic AMRd. Class II DSA ++++HistoricalIgG High riske. Risk of anamnestic 2◦ T & B cell. HLA Class I DSA –+–+HistoricalIgG High riske. Risk of anamnestic 2◦ T & B cell. HLA Class II DSA – + – + Historical Weak Intermediate riskc. IgG HLA Class I DSA – – + + Historical IgG Intermediate riskc. HLA Class I DSA –––+HistoricalIgG Intermediate riskc. HLA Class II DSA ––––CorHIgG Low risk. DSA detected by solid-phase assay HLA alone. Significance remains unclear. Class I or II DSA + + + + C or H Negative Low risk. Autoreactive IgM/IgG non-HLA Ab. – – + + C or H Positive not donor HLA specific Low risk. a Absolute veto to transplantation. b HAR unlikely but risk of severe early AMR and considered a contraindication to transplant in most centers. c Intermediate-risk transplants should be avoided if reasonably possible (i.e., short waiting time, easy to avoid unacceptable mismatches) but may be undertaken with appropriate clinical caution; consideration for enhanced immunosuppression and antibody removal and close post-transplant antibody monitoring. d In the past +B cell FCXM even in the presence of DSA was considered low risk. Retrospective analysis of B+/T– XM transplants has subsequently shown the presence of HLA Class II DSA to be a risk for acute and chronic AMR. e Risk of anamnestic secondary T- and/or B-cell response; need to consider high-risk immunosuppression strategy, the duration, titre, and priming source of antibody and repeat mismatches (pregnancy or re-graft). Historical positive cross-matches caused by cross-reactive alloantibodies (avoiding the main specificity and priming stimulus) constitute intermediate immunological risk and are less likely to be

associated with refractory T- or B-cell response. Section 5: Kidney

Table 28.3 Factors increasing immunological risk for ence of unacceptable antigens without the need for transplant. formal cross-match testing. A positive virtual cross- DSA titres Persistently high DSA titres match forms the basis for identifying patients who Rising DSA titres wouldbesuitableforaugmentedimmunosuppression Past positive, current negative titres or desensitization. The ability to accurately predict High PRA Broad sensitization (even if non–donor-specific) positive cross-matches is used in identifying compat- Presence of Ͼ 1 DSA (esp. If Class I ible donor-recipient pairs in paired-donor exchange and Class II) programs. Transplant history History of previous early rejection and graft loss Number of previous transplants HLA match Increased number of mismatches Correlation of virtual cross-matching Type of mismatch (e.g., DR mismatch) Repeat mismatch with cell-based assays Recipient–donor Husband → wife or child → mother risk The key factors when considering the utility of vir- relationship previous exposure tual cross-matching are how well it correlates with Sensitizing events Number of previous pregnancies or the results from CDC and FCXM testing and ulti- blood transfusions mately how the virtual cross-match predicts clinical outcomes. The correlation of the virtual cross-match with CDC has been shown to be low, but this probably comparing recipient HLA antibody specificities with reflects the significantly higher sensitivity of the solid- HLA typing of the donor. Virtual cross-matching is the phase assays to detect HLA antibody. The correlation, process of predicting the likelihood of a significant as might be expected, was much higher compared with antibody-mediated immune event between recipient FCXM and was greater than 85% in most studies but and graft. Virtual cross-match has applications in pre- is slightly lower in highly sensitized individuals. The transplant risk assessment and can also assist with the predictive value of a negative virtual cross-match in a process of organ allocation. clinical setting is good; it is associated with a very low risk for early AMR. This has been shown in a number Uses of virtual cross-matching of studies in which the risk of early rejection has been The virtual cross-match is at least as sensitive as FCXM as low as 4% and allograft survival is good even among for the detection of HLA-DSA but gets around the sensitized patients. problems of non-specific binding of irrelevant anti- The predictive value of a positive virtual cross- bodies, which produce a false-positive result. A neg- match depends very much on the immunosuppressive ative virtual cross-match in a non-sensitized patient, regimen that is utilized. It is increasingly clear that provided a full sensitization history is available, and transplanting in the presence of DSA is possible with knowledge of previous screening may allow a cross- little impact on short-term outcomes as plasmaphere- match to be omitted pretransplant. This reduces cold sis and antibody therapies become more widely used. ischemic time and improves graft outcome. Unfortu- In the absence of these treatments, when the cross- nately,insensitizedpatients,theerrorrateforthe match assay is negative and the presence of a DSA is virtualcross-matchistoohigh(15%)toallowfor detectable by solid-phase assay alone, then the pres- omission of the cross-match. To clarify differences in ence of antibody does predict poorer outcomes. In terminology between UK and US readers, this spe- CDC negative transplants with DSA detected by solid- cific use of solid-phase assays to avoid a pretransplant phase assay, the incidence of AMR is higher, and graft cross-match is in the United Kingdom what is recog- survival is also lower in recipients in whom DSAs are nized as a virtual cross-match; in the United States, present compared with those in whom they are not. a wider definition incorporating all the elements dis- When FCXM is negative and HLA-DSA is detected by cussed in this section is considered as virtual cross- solid-phase assay, there is an increased risk of AMR, matching. but short-term outcomes appear good and the impact The converse position of a positive virtual cross- on long-term outcomes is less certain. Thus the pres- match can eliminate unnecessary work and expe- ence of HLA-DSA detectable only by solid-phase assay dite deceased organ allocation by predicting the pres- does represent a significant risk for AMR and possible

242 Chapter 28: Sensitization of kidney transplant recipients

Table 28.4 Causes for inaccuracies in virtual cross-matches Possible causes for false-negative virtual cross-match AMR caused by non-HLA Non-HLA antibodies that will not be detected as the virtual cross-match is HLA restricted. The incidence of antibody early AMR related to non-HLA antibody is very low, and these events prove problematic to detect with currently available assays. Incomplete donor HLA Current solid-phase assays cover the most frequent allelic variants of HLA-A, -B, -Cw, -DRb1, -DRb3–5, -DQb, typing and –DP loci. Unfortunately, donor HLA typing often does not include HLA-Cw and -DP and does not have allelic resolution. DSA against these antigens has been shown to lead to AMR, but the exact relevance of preformed Cw- and DP-DSA is not fully understood. However, until full HLA typing of donors is undertaken, the virtual cross-match will not identify all potential DSAs. Failure to detect relevant DSA directed against an epitope that is not represented in the assay panel or the presence of interfering HLA antibody substance preventing detection of relevant antibodies are rare but recognized problems. Possible causes for a false-positive virtual cross-match In vitro artefact HLA molecules used for testing may be denatured by the production process, producing novel epitopes or exposing epitopes that are not accessible in vivo, allowing antibody binding which in the clinical setting is not relevant. Variable pathogenicity of DSA may bind to donor HLA but does not induce graft injury. To date, no routinely available characteristic of the DSA HLA-DSAs (e.g., HLA Class, sensitizing event, DSA level) has been found to be unequivocally predictive of clinical significance. A number of other putative pathogenic factors that may affect the clinical impact of HLA-DSA have been proposed, but as yet these remain unproven. worse long-term outcomes but not is not in most cen- Transplantation options for ters a contraindication to transplantation. highly sensitized individuals There are multiple problems faced by patients who Problems with virtual cross-matching are sensitized, and these form a considerable part of There is good correlation between virtual cross- those waiting for transplantation. Approximately one matches, FCXM, and clinical outcomes, but the third of patients on the UK transplant register have results are not perfect. Some of these differences detectable HLA antibody, and more than half of these Ͼ can be explained by the differences in sensitivity are highly sensitized (PRA 85%). Although these of cell-based and solid-phase assays, but there are highly sensitized individuals are prioritized for well- technical and procedural problems that may also matched deceased donor kidneys, they may well wait contribute. Common causes for false-positive and for many years before receiving a transplant. This in false-negative virtual cross-matches are outlined in itself is a very significant problem, as time waiting for Table 28.4. It is clear that there are a number of aspects a transplant is the most important modifiable risk fac- of virtual cross-matches that require improvement. tor that determines long-term patient and graft sur- On the technical side, expansion of the solid-phase vival. However, as the accuracy of immunological risk assay panels to include more potential epitopes and assessment has improved, the options available for the exact and complete donor HLA typing will improve successful transplantation of highly sensitized recipi- accuracy. In addition, the improvement of the assay ents has increased. itself to eliminate technical false positives is required. The options include paired-donation programs, From an immunological viewpoint, a better under- acceptable mismatch programs, and desensitization. standing of the important factors that contribute to The first two options do require relatively large organ pathogenicity of antibodies in a transplant setting exchange organizations in order to be effective but are are required to improve the final risk assessment. increasingly being introduced. Desensitization is not However, the virtual cross-match can be seen as a reliant on large donor pools, but the outcomes seen are useful tool in many areas of transplantation, and the still currently not well defined. In order for optimal addition of solid-phase assays to the immunological treatment to be offered to highly sensitized patients, assessment pretransplant does improve understanding all three in combination probably provide the optimal of the overall risk (see Table 28.2). solution.

243 Section 5: Kidney

Paired-kidney donation the rate of transplantation is increased, with approx- imately 60% of highly sensitized patients undergoing Paired-donor schemes involve the recruitment of liv- transplantation within 2 years (compared with 20% ing donor–recipient pairs who are unable to progress using the standard allocation program). The second to transplantation because of ABO blood group or benefit is that in contrast to sensitized patients receiv- HLA incompatibility. Full ABO and HLA details are ing a graft through the standard program, both short- then collated, and the pool of donors and recipi- term and long-term graft survival appears to be iden- ents are entered into a computer algorithm to pro- tical to that of non-sensitized patients. Although this duce the maximum number of acceptable donor and offers a huge benefit for highly sensitized patients, it is recipient pairs. Those pairs identified are then subse- not an answer for all. About 40% of patients will not quently cross-matched to confirm the results of the find an acceptable donor from the donor population, positive cross-match before proceeding to transplant. and for these patients, desensitization remains the only The details of the algorithms used differ to some degree option. between different programs but are optimized to allow for the maximum utilization of available live donor organs. Unfortunately, a large proportion of the recip- Desensitization ients within paired donor matching schemes are sen- Desensitization is the process that allows transplanta- sitized, and only a small percentage of patients who tion of presensitized patients with donor-reactive HLA are highly sensitized receive transplants through this or ABO antibodies. Complete removal of antibody is route; in the UK paired donor scheme, only 18% of impossible, so the purpose of these protocols is to patients with PRAs of 85–94% and 2% of patients with reduce antibody levels to such a degree to make trans- PRAs greater than 95% have received a transplant. plantation safe. However, for those who are successful in getting a transplant, the outcomes from within these programs ABO incompatible transplantation are excellent and comparable to those of standard live Transplantation across ABO incompatibility carries donor transplantation. ahighriskforHAR,andsoalllivingdonortrans- plants are performed with ABO compatibility. Unfor- Acceptable mismatch programs tunately, up to 35% of potential living donor trans- Acceptable mismatch programs have developed as a plants are ABO incompatible (ABOi). More than 1000 meanstoincreasetheprobabilitythathighlysensitized ABOi transplants have been preformed worldwide patients receive a suitable cross-match negative organ between all possible combinations of ABO incompat- from a deceased donor. The organ allocation algorithm ibility. ABO antibody testing does provide some dif- of most centers and organizations includes the HLA ficulties. There is very wide variation in ABO titers specificities against which the patient is sensitized. between individuals, but in general, levels are higher This then avoids the allocation of unacceptable mis- in blood group O individuals. Measurement of ABO matches, which would lead to a positive cross-match. immunoglobulin (Ig) G and IgM antibody titers is For highly sensitized patients, identifying all HLA performed by agglutination assay using either tube antibody specificities is extremely difficult. Therefore, dilution or solid-phase card techniques with antibody these programs identify specificities against which the titers assayed against reference red blood cells (RBCs) recipient has never developed antibodies to generate and in some cases donor RBCs. Recipient serum is a number of acceptable mismatches, which, along- serially diluted, and the last dilution where positive side the patient’s HLA phenotype, is used for alloca- agglutination occurs determines the antibody titer. tion. Antibodies against HLA-A, -B, -C, -DR, and - Unfortunately, these assays are subject to wide inter- DQ are identified by solid-phase assay and confirmed and intra-institutional variability related to differences using cell-based techniques. Organs are then allocated in protocols and observer reporting. This has lead at the highest priority on the basis of blood group to differences in reported starting titers that can be compatibility. These programs have provided a num- desensitized (1:256–1:1024) and safe antibody titer for ber of advantages over the standard practice of priori- transplant (1:4–1:16) between different centers. The tizing highly sensitized patients through the allocation ABO antibody titer at assessment also informs deci- of additional points in the allocation algorithm. First, sions over the number of antibody treatments that are

244 Chapter 28: Sensitization of kidney transplant recipients

Table 28.5 Comparison of different antibody removal techniques Plasma exchange DFPP Immunoabsorption Plasma volume treated 1 2–3 3 Coagulopathy ++++ ++ +/– Use of clotting products +++ + – Hypocalcemia ++++ ++ + Length of treatment 2 hours 3–4 hours 6–8 hours Cost + ++ ++++ DFPP: double-filtration plasmapheresis. likely to be required and also the likelihood of AMR, some level of positive CDC when additional treatment which is higher in those with titers greater than 1:128 post-transplant is possible. before antibody removal. Desensitization treatment protocols HLA incompatible transplantation The immunosuppressive protocols that have been Assessment for HLA incompatible (HLAi) transplant developed have three elements: DSA (either HLA should include analysis of the CDC, FCXM, and solid- or ABO) is eliminated or reduced to safe levels, phase assay plus a detailed review of the sensitiza- new antibody production is inhibited, and augmented tion history. As previously discussed (see earlier note immunosuppressive regimes to prevent cellular rejec- and Table 28.3), the antibody specificity, Ig isotype, tion are generally employed. antibody strength, and variation over time (current versus historical) should be determined. Antibody Antibody removal strength is determined by serial dilution of recipient Antibodyremovalisperformedbyeitherplasma serum in CDC and FCXM. As with ABOi, the strength exchange, double-filtration plasmapheresis, or of antibody is defined by the dilution at which the immunoabsorption. There are pros and cons to each cross-match becomes negative. Antibody strength may of these methods (Table 28.5), but all are capable also be assessed by fluorescence intensity in FCXM. of adequate antibody removal for desensitization. Although cell-based assays remain the mainstay of Following antibody removal by whatever method, a this process, solid-phase assays, which have increased rebound increase in plasma Ig levels is seen. This is sensitivity and specificity and the capability of par- partly through redistribution of Ig from the extravas- tialorfullautomation,areincreasinglyused.Some cular space and partly through re-synthesis, and the centers have correlated solid-phase assay results with precise contribution of these two factors is unclear. cross-match results to establish acceptable DSA lev- Redistribution is overcome by further antibody els for desensitization and transplantation. As with removal to deplete total-body Ig levels. ABOi assays, it is important to note that these cor- relations are center-specific due to variations in lab- oratory protocols and equipment. Consideration of a Reduction in antibody production number of factors in the sensitization history of any A number of agents have been employed to reduce recipient must be taken into account in decisions about antibody production. Splenectomy was widely used whetherornottoproceedtodesensitization(SeeTable in the past, particularly in Japan. However, it has 28.3). These in conjunction with the laboratory data been largely replaced by rituximab, which appears define acceptable levels of DSA that could success- equally effective. Splenectomy is now almost exclu- fully be desensitized, the amount of plasma removal sively reserved for the treatment of AMR unrespon- treatment that may be required, and a target level sive to standard treatment in certain centers. Intra- acceptable for transplantation. These thresholds vary venous Ig has pleiotropic immunosuppressive effects between centers; some centers would insist on a neg- on both humoral and cellular immune responses but ative CDC before progressing, whereas others accept isthoughttoinhibitantibodyproductionthrough

245 Section 5: Kidney

Rituximab®1g Induction antibody Figure 28.1 Treatment protocol for desensitization. IVIG: intravenous immunoglobulin.

Antibody removal +/- IVIG

-28 -14 -7 0 7 Days Transplant

Tacrolimus

Mycophenolate

Prednisolone

binding inhibitory Fc receptors on plasma cells and oped world-wide incorporating these elements, but through induction of apoptosis in both plasma cells most follow a similar outline, which is described in and B cells. It has been employed in three dif- Figure 28.1. ferent ways in desensitization. It has been used after each antibody removal treatment at low dose Postoperative protocols (100 mg/kg) to inhibit antibody re-synthesis. In some Following ABOi and HLAi transplant, antibody protocols it is given at intermediate dose (500 mg/kg) removal protocols vary between centers; most use immediately before transplantation to reduce antibody three to five elective antibody removal treatments, but production. High-dose intravenous Ig (2 g/kg) has some centers adjust treatment according to antibody been used for both live donor and deceased donor titers (Table 28.5). ABO antibody titers are measured transplantation administered as four doses 1 month daily for the first 2 weeks and then two to three times apart prior to transplantation. This protocol does not per week for a further 2–4 weeks. A rapidly rising titer employ antibody removal but appears to produce sim- is seen in approximately 10% and most centers would ilar results. Rituximab is probably the most widely treat this with further antibody removal, as it may pre- employed method for inhibition of antibody pro- dict AMR. Subsequent antibody monitoring is infre- duction. Most centers employ a single-dose regimen quent, and later rises in antibody titer are not uncom- (1000 mg or 375 mg/m2), which has been shown to mon but in the absence of graft dysfunction do not eliminate all circulating B cells. appear to be significant for long-term outcomes. At any stage that graft dysfunction is detected, ABO antibody Augmented immunosuppressive regimens titers should be checked in case of AMR. However, it In most situations, augmented immunosuppres- should be noted that C4d deposition is seen in 75% of sion employing tacrolimus, mycophenolate mofetil ABOi transplants and is not significant in the absence (MMF), and steroid are employed. This is principally of other features of AMR. employed to inhibit cell-mediated responses, but in In HLAi transplants, DSA levels are monitored addition,MMFmayhaveanimpactonantibody using solid-phase assays. This is partly through the dif- production. These are usually commenced 7–14 days ficulties of access to donor blood samples and partly before transplant. The use of induction antibody treat- because of the technical complications provided by ment is universal. In ABOi transplants, anti-CD25 the use of rituximab and Alemtuzumab interfering antibodies have been employed. However, because with cell-based assays. Antibody screening is similar of the significantly higher risk of TCMR in HLAi to ABOi transplantation with monitoring daily or on transplants, it is usual to employ T-cell depleting alternate days for the first 2 weeks and then two to antibodies such as Thymoglobulin or Alemtuzumab. three times per week for a further 2–4 weeks, with most A number of different protocols have been devel- episodes of rejection occurring within this period.

246 Chapter 28: Sensitization of kidney transplant recipients

Ongoing infrequent monitoring of DSA levels beyond survival is approximately 85% (c.f. 95% for HLA- this period is usually performed. Rapidly rising DSA compatible transplants), which is similar to that seen levels or graft dysfunction requires transplant biopsy for deceased donor transplants. Longer term data are to exclude AMR, which is common. The significance insufficient to draw clear conclusions on patient and of rising DSA levels in the absence of rejection and the graft survival compared with HLA-compatible trans- finding of C4d deposition on biopsy without other evi- plantation. However, some reports have shown early dence of rejection is unclear. If these occur in the first graft loss due to arteriopathy and glomerulopathy, 4–6 weeks after transplant, many centers would treat particularly in those with HLA Class II DSA. Addi- with further antibody removal. Beyond that period, tional concerns remain over the likely increased rates management is usually expectant with increased mon- of opportunistic infections and malignancy that may itoring for the development of graft dysfunction. occur given the significantly elevated immunosuppres- sive load that these patients receive. Outcome following desensitization The incidence of TCMR in ABOi transplantation is Further reading similar to that in ABO-compatible transplants (10– BSHI/BTS Joint Guidelines for the Detection and 20%), as might be expected. The incidence of AMR Characterisation of Clinically Relevant Antibodies in is approximately 10% and is more common in those Allotransplantation. Available at www.bts.org.uk and patients with higher initial antibody titers (Ͼ1:128) www.bshi.org.uk, May 2010. andthosewithearlyandrapidreturnofABOanti- Doxiadis II, Claas FH. Transplantation of highly sensitized bodies after transplant. World-wide data suggest that patients via the acceptable mismatch program or both short-term (1 year) and long-term (Ͼ5years) desensitization? We need both. Curr Opin Organ Transplant 2009; 14: 410–3. outcomes are excellent, with both patient and graft survival comparable to that of ABO-compatible trans- Fuggle SV, Martin S. Tools for human leukocyte antigen plants. antibody detection and their application to transplanting sensitized patients. Transplantation 2008; 86: 384–90. Unlike ABOi transplant, HLAi transplants have inferior outcomes to those in HLA-compatible live Jordan SC, Peng A, Vo AA. Therapeutic strategies in management of the highly HLA-sensitized and donor transplants. There is an increased incidence of ABO-incompatible transplant recipients. Contrib DGF. The incidence of TCMR is significantly higher Nephrol 2009; 162: 13–26. than seen in HLA-compatible transplants, with a 20– Tambur AR, Ramon DS, Kaufman DB, et al. Perception 60% risk of severe rejection. The risk of AMR is also versus reality? Virtual crossmatch – how to overcome significantly higher compared with ABOi transplant, some of the technical and logistic limitations. Am J with an incidence of between 30–40%. One-year graft Transplant 2009; 9: 1886–93.

247 Section 5 Kidney Chapter Live donor kidney donation

29 Arthur J. Matas and Hassan N. Ibrahim

Key points nificantly better for recipients who undergo transplant r For a patient with end-stage renal disease, before initiating dialysis (preemptive transplant) or the best option is a living (related or after a short course of dialysis versus recipients who unrelated) donor transplant. undergo transplant after a more prolonged dialysis r A transplant before initiation of dialysis course. This was clearly shown in a study of “paired- (preemptive transplant) is associated with kidney” deceased donor transplants, in which one kid- better post-transplant outcomes than a ney from the donor was transplanted to a recipient transplant after initiation of dialysis. who had been on dialysis less than 6 months and the other kidney (from the same donor) was transplanted r Mortality associated with donor to a recipient who had been on dialysis more than nephrectomy is about 0.03%, and major 24 months (Figure 29.1). At 60 months post-trans- morbidity is less than 1%. plant, recipients with less than 6 months of pretrans- r Long-term outcome (survival and quality of plant dialysis had 20% better graft survival. life) after donor nephrectomy is similar to With regard to choosing between living donation that of the age-matched population. or deceased donor transplantation, there is no doubt that a living donor transplant is the best option. For the recipient, the surgery for living and deceased donor Patients with chronic kidney disease (CKD) need to transplants are almost identical, and the immunosup- make two important decisions, with each requiring pression is similar. Yet, patient and graft survival are careful consideration. The first is whether to proceed significantly better after a living donor transplant. In with transplantation or dialysis as primary therapy. addition, having a living donor transplant is far more Ideally, this decision should be made early in the likely to permit preemptive transplantation. Impor- course of CKD. Patients need to understand the risks tantly, when considering living donor transplants, the and benefits of each treatment modality. The second is results of related (human leukocyte antigen [HLA] whether to proceed with living donation or to go on non-identical) and unrelated transplants are identi- the waiting list for a deceased donor transplant. cal. Figure 29.2 shows the 5-year graft survival out- When compared with maintenance dialysis, a come for first transplants by donor source. HLA- successful transplant is associated with significantly identical living donor transplants have the best out- longer life and significantly better quality of life. come, and all subgroups of living donor transplants However, a transplant is associated with the risks of do better than deceased donor transplants. Note, how- surgery and of immunosuppression-related complica- ever, that living unrelated transplants have the same tions. Whereas in practice, patients transition from outcome as non-identical living related transplants. dialysis to transplant (e.g., after being on the wait- The fact that unrelated donor kidney transplants are ing list) or from transplant to dialysis (e.g., after graft associated with excellent long-term results has allowed failure), a decision early in the course of CKD that marked expansion of the donor pool and permit- transplantation is to be the primary therapy may facil- ted development of non-directed donation and paired itate early transplantation. Transplant results are sig- exchanges.

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

248 Chapter 29: Live donor kidney donation

100 Figure 29.1 Unadjusted graft survival in 2405 recipients of paired kidneys with short compared with long end-stage 90 renal disease time. From Meier-Kriesche HU, Kaplan B, Waiting time on dialysis as the strongest modifiable risk factor for 80 0-6 months78% on dialysis renal transplant outcomes: A paired donor kidney analysis, Transplantation 2002; 74:1377–81, with permission from 70 Wolters Kluwer Health. >24 months on dialysis 63% 60 58%

50 % event-free survival % event-free 40

29% 30

20 0 12 24 36 48 60 72 84 96 108 120 Months post-transplant

100 Figure 29.2 Actuarial graft survival and half life (t 1/ ): first transplant by donor source. 90 2 Based on = Organ Procurement and 80 Transplantation Network/United Network for 79 70 Organ Sharing data as of July 2009. First transplants of kidney only, done between 60 59 January 1996–December 2005. Patient death 59 was considered a graft loss. MM: mismatched; 50 48 n T1/2 LRD: living related donor; LUD: living unrelated 40 HLA-ID 4,134 26.3 donor; SCD: standard criteria deceased donor. Mike Cecka, personal communication. 30 MM LRD 27,120 14.7 Graft Survival (%) Graft Survival 20 LUD 12,431 15.7 10 SCD 52,309 10.7 0 0 12345678910 Years post-transplant

The major concern about living donor transplan- time of initial contact by a potential donor, followed tation is the risk to the donor. The donor operation by sending the donor candidate a package containing is a major procedure that is associated with morbid- both written information and a video about donor risk. ity, mortality, and the potential for adverse long-term If the donor contacts us again after reviewing the mate- consequences, secondary to living with a single kidney. rial, a clinic visit for evaluation is scheduled. At that A major concern regarding the use of living donors time, the candidate meets with the surgeon, nephrolo- has been whether unilateral nephrectomy predisposes gist, coordinator, and social worker/advocate. The goal to the development of kidney disease and/or prema- of the evaluation is to determine that the prospective ture death. A transplant program considering living donor is healthy and will tolerate the operation, has donation should have well-defined protocols to fully two normally functioning kidneys, and has no diseases inform donor candidates about the potential risks of that would affect renal function or could be transmit- the procedure and to make sure that consent is volun- ted with the kidney graft. A critical aspect is making tary. Our protocol includes a screening interview at the sure that the candidate has complete information on

249 Section 5: Kidney

the risks and benefits of donation so as to be able to Experimental data have suggested that renal abla- make an informed decision. As they decide whether tion is associated with compensatory changes that lead or not to proceed with their evaluation and donation, to subsequent progressive renal dysfunction. Of note, prospective donors must be informed of the risks. They most of these studies have been in the rat model. In must also recognize that time will be lost from work this model, renal dysfunction is preceded by increas- and family. Although the cost of the donor’s evalua- ing proteinuria. Thus it has been a concern that some tion and surgery is paid for by the recipient’s insur- living donors have developed proteinuria shortly after ance (in the United States), the donor is away from donation. In spite of these concerns, every study of work for several weeks, often without compensation long-term donor follow-up, to date, has shown donor for lost wages or traveling expenses. Recipient surgery survivaltobesimilartoorbetterthanthatofthe and immunosuppression are discussed in Chapters 30 age-matched general population. Similar observations and 3, respectively. In the following sections, we review have been made after uninephrectomy (when there is data on donor risks and outcomes. a normal contralateral kidney) in the non-donor pop- ulation. There have been no data to suggest increased Donor surgical risks cardiovascular disease after donor nephrectomy. Per- haps donor nephrectomy is not associated with suf- Laparoscopic donor nephrectomy has been associated ficient renal dysfunction to increase the risk for car- with less pain and a faster recovery for the donor diovascular disease; alternatively, we may not have fol- than conventional open nephrectomy; as a result, in lowed donors long enough to be able to discern any the last decade, laparoscopic nephrectomy has become increase in the risk. Additional long-term studies are the procedure of choice at many transplant centers. needed to resolve this issue. Importantly, the surgical risks for laparoscopic and Stability of long-term renal function of donors is open nephrectomy are similar. Peri-operative mortal- also of concern after kidney donation. As described ity has been reported in a number of studies to be above, in the rat model, renal ablation has led to sub- 0.3–0.4%. The most common causes of death have sequent progressive renal dysfunction, and the damage been pulmonary embolism, bleeding, and infection. is proportional to the amount of renal tissue removed. Major morbidity has occurred in less than 1% of donor The importance of this experimental observation is nephrectomies, and minor complications in less than that humans have an age-related loss of renal function. 10%. There are minimal differences between open A possibility is that the remaining kidney will slowly and laparoscopic nephrectomy. Given that laparo- lose functional mass (with aging) and that this will scopic nephrectomy is an intra-abdominal procedure, reach a critical threshold where progressive renal dis- one long-term concern is the potential for bowel ease is inevitable. However, the injury seen in the rat obstruction. model appears to be species-specific; similar progres- sion to renal failure has not been seen in other animal Long-term donor outcome models. In humans who are not kidney donors, evi- A major concern is whether donor uninephrectomy dencethatareductioninrenalmassmayleadtopro- leads to adverse long-term outcomes. This concern has gressive renal failure comes from studies of children been driven by both clinical and experimental stud- born with abnormal kidneys and from reports of pro- ies. Donor uninephrectomy leads to the immediate gressive damage developing in patients born with only loss of about 50% of renal function. Compensatory one kidney. In such situations, it has not always been changes take place in the remaining kidney, and within clear that the patient had one normal kidney. In con- 6 weeks, measured renal function is about 75–80% of trast, other long-term (Ͼ40 years) follow-up studies of pre-nephrectomy. Studies in the general population non-donors who underwent nephrectomy for unilat- (and a number of subgroups) have noted that mild eral disease have not shown progressive deterioration decreases in renal function correlate with increased in renal function. long-term mortality and increased cardiovascular risk. Prospective living donors are screened to deter- One such subgroup is kidney transplant recipients: mine that they have two normal kidneys at the time of mild renal dysfunction after kidney transplantation is nephrectomy. To date, numerous studies have exam- associated with increased risk for cardiovascular dis- ined renal function, proteinuria, and hypertension. ease, cardiac death, and major cardiac events. Isolated cases of renal failure after donor nephrectomy

250 Chapter 29: Live donor kidney donation have been reported, and the United Network for Organ wassimilartothatamongcontrolsfromtheNational Sharing (UNOS) database has noted that 172 former Health and Nutrition Examination Survey who were donors have been listed themselves for a deceased matched for age, sex, race or ethnic group, and BMI. donor kidney. Of these, some had donated one kid- When donors with more than 20 years of follow-up ney before the establishment of the UNOS database, were compared with matched controls, again there making it difficult to determine a denominator and were no differences. calculate the incidence of end-stage renal disease (ESRD). However, no large, single-, or multi-center Donor quality of life series (with numerator and denominator well-defined) Numerous studies, using a variety of standardized has demonstrated any evidence of progressive deteri- instruments (most commonly the SF-36, a short-form oration of renal function in a significant proportion 36-question survey) have studied living donor quality of living donors. In recognition of this benign course, of life. In general, living donors report a similar or bet- most insurance companies do not increase premiums ter quality of life, as compared with the general popula- for kidney donors. A limiting factor in most of these tion. When concerns were raised, they were related to studies is that the average follow-up time has been less possible negative effects on recovery and future health, than 20 years. Given that most living donors have a life the amount of time to return to routine daily activi- expectancy of more than 20 years, longer follow-up is ties and commitment, the financial consequences and necessary. implications, and the potential penalization by life or In the largest study to date, we have recently health insurance companies. In addition, many living reviewed long-term outcome in our donor popula- donors report feeling abandoned after surgery by the tion. Between 1963 and 2007, 3698 patients underwent transplant program and are disappointed by the lack of donor uninephrectomy at our institution. We were any follow-up after their hospital discharge. These con- able to collect follow-up information on 99%. Donors cerns need to be taken into consideration by every liv- were matched with the general population for age, sex, ing donor program. In general, the hospital culture is and race or ethnic group and outcome compared. We centered on “the sick.” Because donors are not viewed found that the long-term survival of donors was sim- as being “sick,” they may not get the attention they ilar to the matched controls. ESRD developed in 11 deserve. donors, a rate of 180 cases per million persons per year, Risk factors for less positive quality of life after as compared with a rate of 268 per million per year in donation have also been identified, including poor the general population. donor or recipient physical outcome, a negative In addition, in our study, a subset of donors personal donor–recipient relationship, and financial returned to our institution for detailed studies: mea- hardship. In addition, most studies are of living related surement of the glomerular filtration rate (GFR) and donors. It needs to be determined whether unrelated urinary albumin excretion and assessment including donors have similar outcomes, since their relationship the prevalence of hypertension, general health status, dynamic and motivation to donate is complex. Most and quality of life. Of the subgroup of donors hav- of the aforementioned studies were done in donors ing detailed studies (12.2 ± 9.2 years after donation), who underwent open nephrectomy. It will be impor- 85.5% had a GFR of 60 ml/min/1.73m2 or higher, tant to learn whether the same issues develop after 32.1% had hypertension, and 12.7% had albuminuria. laparoscopic nephrectomy. Such information can help The risk of proteinuria, contrary to other studies, was transplant programs design protocols to increase liv- similar to age, sex, ethnicity, and body mass index ing donor satisfaction. (BMI) matched controls; a finding that clearly chal- lenges commonly held beliefs regarding donor risk. Older age and higher BMI, but not a longer time since Pregnancy after donation donation, were associated with both a GFR that was Historically, some centers would not accept women lower than 60 ml/min/1.73m2 and hypertension. A of child-bearing age as donors because of concern longer time since donation, however, was indepen- that there would be increased risk of pre-eclampsia, dently associated with albuminuria. Most donors had eclampsia, and renal damage in the context of preg- quality-of-life scores that were better than population nancy after uninephrectomy. As the shortage of organs norms, and the prevalence of coexisting conditions has become more severe, and wait times for a deceased

251 Section 5: Kidney

donor kidney have increased, most centers now accept need 50–60-year follow-up after donation to be women of child-bearing age as donors. Although the able to inform future candidates about the true numbers are limited, studies to date have not shown long-term risk. increased maternal or fetal risks for pregnancies after 2Todate,reportsoflong-termfollow-up(Ͼ20 donation when compared with the general population. years) are limited to “ideal” donors selected over 2 However, and albeit with small numbers, there is a sug- decades ago. Due to the tremendous organ gestion that post-donation pregnancies may be associ- shortage, there has been loosening of the criteria ated with increased risk of pre-eclampsia. for acceptance for donors. As discussed above, some centers now accept donors with single-drug hypertension and/or obesity. It behooves those Future concerns centers using “expanded criteria living donors” to Population studies have shown that smoking, obesity, define the short- and long-term outcome for such hypertension, and elevated blood glucose levels are donors. If risks are increased (versus “ideal” associated with an increased risk of proteinuria and donors), then future candidates need to be kidney disease. In the United States today there is informed. an epidemic of type 2 diabetes and of hypertension; 3 Is proteinuria or mild renal dysfunction (seen in therateofbothdiseasesincreasesasthepopulation somelivingdonors)associatedwithanincreased ages. Therefore, many donors may end up with one or risk of cardiovascular disease or mortality? both of these diseases. Both are well-described causes 4 If living donors develop native kidney disease or of ESRD. In addition, many centers currently accept another disease that might affect their remaining obese donor candidates. Studies are necessary to deter- kidney (particularly type 2 diabetes), will they mine whether there is increased long-term risk for suffer an accelerated course to renal failure? such donors. A critical question is whether living donors who subsequently develop any form of kidney disease, Acknowledgments even years after nephrectomy, will have an accelerated We thank Stephanie Daily for her help in the prepara- course to renal failure. We have recently studied the tion of the manuscript of this chapter. outcome of living donors who develop type 2 diabetes after donation. In our entire experience (n = 3777), Further reading 154 donors developed type 2 diabetes at a mean of 17.7 AbecassisM,BartlettST,CollinsAJ,et al. Kidney ± 9.0 years after donation. Mean follow-up after devel- transplantation as primary therapy for end-stage renal opment of diabetes has been 7 years. To date, estimated disease: a National Kidney Foundation/Kidney Disease GFR is no different between donors with diabetes and Outcomes Quality Initiative (NKF/KDOQITM) Conference. Clin J Am Soc Nephrol 2008; 3: 471–80. those without during similar follow-up duration after donation; diabetic donors, however, were more likely The Authors for the Live Organ Donor Consensus Group. to be hypertensive and proteinuric. Consensus statement on the live organ donor. JAMA 2006; 284: 2919–26. Evans RW, Manninen DL, Garrison LP, Jr, et al. The quality Limitations of current data of life of patients with end-stage renal disease. NEnglJ Med 1985; 312: 53–9. Numerous questions remain regarding the surgical and peri-operative risks, long-term outcome, and Ibrahim HN, Foley R, Tan L, et al. Long-term consequences quality of life for living kidney donors: of kidney donation. NEnglJMed2009; 360: 459–69. 1 What is the very long-term outcome of donor Meier-Kriesche HU, Kaplan B. Waiting time on dialysis as the strongest modifiable risk factor for renal transplant nephrectomy? Most published studies have a outcomes: a paired donor kidney analysis. meanfollow-upoflessthan15years.Thosewith Transplantation 2002; 74: 1377–81. longer follow-up have not demonstrated any Wolfe RA, Ashby VB, Milford EL, et al. Comparison of increased risk. However, larger numbers and mortality in all patients on dialysis, patients on dialysis longer follow-up are necessary. Many donors are awaiting transplantation, and recipients of a first in their 20s and 30s at the time of donation. We cadaveric transplant. NEnglJMed1999; 341: 1725–30.

252 Section 5 Kidney Chapter Surgical procedure

30 Paul Gibbs

the patient) or the size of the kidney can lead to con- Key points siderable surgical difficulties. r Each hour of cold ischemia will affect the long-term outcome of the kidney transplant, and cold ischemic times should be kept to a Cadaveric donor nephrectomy minimum. The most common surgical approach used for cadav- r Identification of the bladder during surgery eric donor nephrectomy is the en bloc technique may be improved by filling the bladder through a large abdominal incision. This technique through the catheter with a dilute methylene permits rapid removal of the kidneys, therefore reduc- blue solution. ing the risk of renal vascular spasm and irreversible r If there is significant atherosclerosis, with anoxic injury due to prolonged warm ischemia. The calcification resulting in non-compressible, heart and liver teams often begin the dissection simul- solid arteries that cannot be clamped, it may taneously. The kidneys can be dissected free and easily be necessary to use the common iliac artery removed after in situ flushing with preservation solu- or even on occasion the aorta to perform the tion of the graft and the liver and pancreas have been arterial anastomosis. removed. If the heart is not harvested, the descending r If the transplanted kidney fails and aorta can be cross-clamped in the left pleural cavity transplant nephrectomy is performed within without prior dissection. An aortic cannula is placed a short time, mobilization and removal of the below the renal artery and fixed. The supraceliac aorta kidney is usually straightforward. is clamped, and cold perfusate is instilled into the r If using older donor kidneys, reduced renal kidneysthroughtheaorticcannula,using,forexam- function may necessitate transplantation of ple, University of Wisconsin (UW) solution. A vent in both kidneys to a single recipient; these can the inferior vena cava (IVC) prevents venous engorge- usually be placed together into the same iliac ment and drains the perfusate. After the hepatectomy fossa. and pancreatectomy are complete, dissection is carried out in a bloodless field. The right colon and duode- num are mobilized and reflected up into the left upper When discussing renal transplantation with prospec- abdomen. The lesser splanchnic nerves and left meso- tive recipients, the procedure is usually described as colon are divided to expose the left kidney. Deep dis- a moderate-sized operation, and although this is true section mobilizes both kidneys medially with Gerota’s in most cases, it is not one to be taken lightly. A first fascia. Both ureters are identified and transected close transplant in a “virgin” iliac fossa is indeed relatively to the bladder and dissected up to the level of the lower straightforward,butitisalsotruethattheoperation renal pole. The IVC and aorta are divided just below can be very technically challenging. A previous trans- the level of the aortic cannula. The en bloc dissection plant, severe atherosclerosis as is the case in many renal is completed by incising the prevertebral fascia while failure patients, or limited access due to either patient the aorta and IVC are retracted upward, together with size(boththedepthoftheiliacfossaandadiposityof the kidneys and ureters. The kidneys are separated on

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

253 Section 5: Kidney

ice following removal from the donor: the left renal maticcordmobilizedandretractedmediallyinmales. vein is divided to include a cuff of IVC, whereas the The inferior epigastric vessels are also usually divided, IVC remains with the right kidney. From the posterior although very occasionally, if there is any doubt about aspect, the aorta is opened and the renal arteries iden- thebloodsupplyofthelowerabdominalwalltheycan tified,andthenthekidneysareseparated.Atotalof be preserved. The external iliac artery and vein are 250–300 ml of cold perfusate is flushed through each dissected free, taking care not to damage lymphatics kidney using a cannula in each renal artery. The kid- unnecessarily. The renal vessels are anastomosed end neysarebaggedinperfusatesolutionandstoredonice to side to the external iliac vessels using standard vas- for transportation to the recipient hospital. All remain- cular clamps to control the iliac vessels and usually 5/0 ing back table preparation is left for the recipient sur- monofilament suture such as Prolene. Individual sur- geon. gical preference dictates which is anastomosed first, although most commonly the vein precedes the artery. Cold ischemia Once each anastomosis is complete, the renal ves- sel is clamped and blood flow restored to the leg. This Transplantation using organs from cadaveric donors is allows the anastomosis to be tested and even on occa- always performed with the overriding need to mini- sion revised without compromising the kidney’s circu- mize the cold ischemic time of the organ. The length lation by further interruption after reperfusion. Once of time acceptable to store an organ is variable and for both anastomoses are satisfactory, the kidney is reper- kidneys can be extended out to 48 hours. However, it fused with blood. is now well established that each hour of cold ischemia There are a number of techniques for anasto- will affect the long-term outcome of the kidney trans- mosing the ureter to the bladder. These include the plant, and cold ischemic times should be kept to a min- Leadbetter–Politano or a direct vesico-ureteric anas- imum and certainly below 24 hours. tomosis. In the latter technique, a small anterior cys- tostomy is made, dividing the muscle layers of the The standard cadaveric transplant bladder and making a small opening in the mucosa of the bladder. Direct mucosa to mucosa anastomosis is Technique performed using interrupted absorbable sutures. The Prior to the patient being anesthetized, the kidney bladder muscle layer is then closed over the distal end should be inspected and back table dissection per- of the ureter as an anti-reflux measure. A short dou- formed to ensure that it is not damaged. The renal ble J stent can be used to protect this anastomosis; this artery and vein are freed from any retroperitoneal fat, is usually removed at 6 weeks. It is important to trim and any polar arteries are identified and if necessary the ureter back as far as possible to ensure an adequate reconstructed by anastomosis to the main artery if they blood supply to the ureteric side of the anastomosis have not been included on the aortic patch. Where and minimize either anastomotic leaks or stenosis. there are multiple veins, smaller ones may be tied, Identification of the bladder can be surprisingly although it may be prudent to anastomose both veins difficult, especially in patients on long-term peritoneal where they are of equal size. Care must be taken to dialysis in whom the peritoneum has thickened or with avoid denuding the ureter of its peri-ureteric tissue, as a small shrunken bladder due to many years of anuria. this may render it ischemic. Identification may be improved by filling the blad- Following induction of anesthesia, the patient der through the catheter with a dilute methylene blue is catheterized and the operative site prepared and solution. draped. A curvilinear incision is made in the iliac An alternative to using the patients’ bladder is fossa, extending from the midline supra pubic area to to perform a direct anastomosis with the recipients’ the level of the anterior superior iliac crest, although native ureter, spatulating both ureteric ends. The native in patients with access problems, this can be extended ureter is usually divided, and even if there is resid- further. The oblique muscles are divided, but in most ual renal function in the native kidneys, this rarely casesitispossibletoleavetherectusabdominisintact. results in troublesome hydronephrosis. This uretero- The peritoneum is mobilized medially to expose the ureterostomy is usually protected with a double J stent, iliac vessels; during this mobilization, the round lig- although passing the stent antegrade down the native ament is usually divided in females and the sper- ureter is often not possible and a small cystostomy may

254 Chapter 30: Surgical procedure need to be made to allow retrograde passage of the ever, tracheal intubation and muscle relaxation are stent. generally employed to allow optimum surgical access. The Leadbetter–Politano procedure involves the Central venous access is generally obtained (often creation of a submucosal tunnel along which the ureter using ultrasound guidance as multiple lines may have is tunneled before being brought through the mucosal been placed in the past), but invasive arterial mon- layer into the bladder. The distal ureter is spatulated, itoring is not often required. The patient should be everted, and secured to the bladder mucosa, creating a catheterized, usually once anesthetized. Immediately ureteric nipple. The technique requires an anterior cys- prior to reperfusion of the donor kidney, most cen- tostomy. Finally, in patients with bladder dysfunction ters administer a cocktail of drugs comprising methyl or absence, it is sometimes necessary to drain the urine prednisolone as immunosuppressant, mannitol, and into an ileal conduit. This is an isolated segment of vas- sometimes frusemide. Adequate intravascular filling is cularized small bowel, which is widely used by urolo- required to maximize organ perfusion; some clinicians gistsasareplacementbladderfollowingtotalcystec- target a central venous pressure (right atrial pressure) tomy. Ideally it should be created prior to the patient of at least 10 mmHg at this point. It is also important to being placed on the waiting list. In this instance, the maintain blood pressure, especially as the donor may ureter is anastomosed end to side to the bowel wall, have had significant hypertension. ensuring mucosa-to-mucosa apposition. The anasto- After waking (at the end of the procedure), the mosis can be stented in the same way as the standard patient is usually looked after in a high-dependency technique and brought out through the stoma on the area or step-down unit. Hourly urine measurements skin surface. and careful fluid balance is imperative. A Doppler There are a number of variations to this standard ultrasound may be performed within a few hours of the technique, which can be utilized depending on cir- operation to achieve a baseline reading of renal arterial cumstances. If healthy, the internal iliac artery can flow, even when there is good immediate function. In be divided and swung up to allow end-to-end anas- patients with no function, this is obviously crucial to tomosiswiththerenalartery.Thisisusefulincases ensure that the vessels are patent. where there is no aortic patch on the renal artery and is therefore widely used in live donor transplants. The internal iliac vein can be divided to allow bet- Living donor renal transplant ter access to the external vein where the right kid- (see Chapter 29) ney is being transplanted. This is to make the venous A live patient donating a kidney is a challenging situ- anastomosis more accessible, as the right renal vein ation in which any complication must be strenuously is short. Various techniques for lengthening the vein prevented. The preferred donor procedure is a laparo- have also been described, usually utilizing a segment of scopic nephrectomy, with mobilization of the kidney the donor vena cava. In the majority of cases, this is not assisted by the use of a hand port, usually through a required. small infra-umbilical midline incision through which Many patients with renal failure have significant the kidney is removed. Choice of which kidney is atherosclerosis, with calcification resulting in non- removed is dependent mainly on the arterial anatomy, compressible solid arteries that cannot be clamped. In avoidingifpossibletheneedformultiplerenalartery these cases and also in re-transplants into the same anastomoses in the recipient, although often the left iliacfossa,itmaybenecessarytousethecommoniliac kidney may be a less challenging procedure. The split artery or even on occasion the aorta to perform the in function between the kidneys should also be taken arterial anastomosis. Careful preoperative assessment into account. by computed tomography scanning should allow iden- tification of calcified arteries before listing for trans- plantation. Recipient procedure This is essentially the same as for a cadaveric renal transplant, with the main difference being the arte- Peri-operative care rial anastomosis. Because there will not be an aortic General anesthesia is the norm for kidney transplant patch, use of the internal iliac artery (when disease- recipients. Most will be fasted prior to surgery; how- free) for the arterial anastomosis is common, although

255 Section 5: Kidney

the external iliac artery may need to be used in cases iliac vessels. The disadvantage of the latter technique is where there is significant disease. The other techni- that if graft nephrectomy is required in the early post- cal challenge is that both the vein and artery are often operative period, the donor vessels will also need to be shorter in a live donor kidney, making performance of removed, resulting in the need to reconstruct the iliac the anastomosis more technically challenging. There is vessels. also a high incidence of two renal arteries due to early bifurcation of the main renal artery and the length of artery lost due to the stapling required in laparo- Donation after cardiac death scopic retrieval. One method of dealing with multi- cadaveric donors ple renal arteries is to excise a length of the internal The number of kidney transplants from donation after iliac artery including its first branch. This can then cardiac death (DCD) donors has increased substan- be used as a Y graft and anastomosed to the two tially over the last 10 years. These donors may be quite renal arteries on the back table, eliminating the need elderly (age Ͼ 70 years) and as a consequence have a to perform difficult arterial anastomoses at depth in higher incidence of pre-existing renal disease. It is pru- the iliac fossa. End-to-end reconstruction of the inter- dent to biopsy kidneys from elderly donors or those nal iliac is then performed, allowing an easier, safer with significant past medical history including dia- anastomosis. betes or hypertension, and recent practice has seen the successful transplantation of both kidneys into Pediatric transplantation (see one patient where the biopsy has revealed significant glomerulosclerosis. Both kidneys unless very large can Chapter 33) be transplanted into the same iliac fossa, with the prox- The number of children who become donors is thank- imal kidney being anastomosed onto the common iliac fully very small, which means that the many children artery and vena cava. It is important to remember requiring renal transplantation will receive an adult to implant the proximal kidney first so that it can organ, whether from a cadaveric or live donor. This has bereperfusedwhilethesecondkidneyisimplanted. implications with regard to the size match of kidney Although DCD kidneys have a lower immediate func- to recipient, and it is often necessary in smaller chil- tion rate compared with donor kidneys after brain dren to place the kidney within the peritoneal cavity death, the medium- and long-term outcome seems to and anastomose the renal vessels to the aorta and the be comparable, and they are a valuable extra source of vena cava. The initial incision needs to be longer then kidneys for transplantation. in adults, extending as far as the tip of the 12th rib. The kidney is usually placed retroperitoneally behind the cecum, ensuring access to enable percutaneous biop- Complications sies to be performed in the postoperative period. Oth- Early vascular complications include both bleeding erwise the procedure is essentially the same as for an and thrombosis. Hemorrhage sometimes occurs up to adult transplant. a week later if there is ongoing wound infection. It can be arterial or venous, and the subsequent hematoma can be large enough to obstruct urine flow and cause En bloc renal transplantation hydronephrosis. All significant hematomas should be In very small child donors (age Ͻ 2years),therisk evacuated, even if there is no ureteric obstruction to of thrombosis of the kidney if used as a single organ reduce the risk of ongoing infection. is very high; these kidneys are therefore often trans- Arterial thrombosis is uncommon and is usually a planted as a pair, en bloc, still attached to the aorta and technical issue. When identified, it requires immediate vena cava into both children and adults. This allows re-exploration if there is a chance of saving the kidney. the aorta and cava to be used to perform the vascular Unfortunately, the kidney has often already infarcted anastomoses, considerably reducing the risk of throm- and has to be removed. Renal vein thrombosis is more bosis. The proximal end of both great vessels is over- common (around 6% in some series) although still sewn and the distal end anastomosed end to side in unusual and can result in the sudden loss of the kid- the usual manner. Alternatively, the aorta and cava can ney within the first week after transplantation. Its cause be anastomosed as interposition grafts into transected is usually not technical; all patients should have a

256 Chapter 30: Surgical procedure thrombophilia screen performed, although an abnor- and artery usually requires a small vein patch to allow mality is rarely found. Immunological factors may also closure without stenosing the vessel. The long saphe- play a role, although the exact mechanism remains nous vein is the usual source for this. uncertain. Late nephrectomy of failed grafts is not routinely Ureteric complications are the commonest spe- performed. It is indicated for a number of clinical rea- cific complication. The incidence is in the region of sons, including persistent bleeding from a failed kid- 10–15%. The majority of cases are due to ischemia ney due to ongoing rejection following withdrawal of and usually require revision surgery to re-implant the immunosuppression, chronic infection, or the need to ureter into the bladder. This can be a difficult oper- create space for another transplant if both iliac fos- ation, especially if there is a long-standing infection sae have previously been used for transplantation. This resulting in dense fibrosis. Options include direct re- needs to be done prior to re-listing. The kidney is implantation into the bladder of the ureter if there is usually densely adherent to the surrounding tissue, enough length, the formation of a Boari flap from the and mobilization can be slow and needs to be intra- bladderwallwheretheureterisshort,oruseofthe capsular. It is important for dissection to remain close native ureter, which can be anastomosed directly to to the kidney because damage to the iliac vessels is easy the renal pelvis if necessary. Ureteric complications are at this stage as they are often lying close to the kidney often serious due to the ongoing infection and result in as a result of fibrosis. The kidney is resected, dividing the loss of a kidney. all vessels high in the hilum, leaving the donor ves- Late vascular complications are usually stenosis sel anastomoses to the iliac vessels intact. At this stage of the arterial anastomosis or a stricture in the renal after a transplant, rupture and bleeding is not a signif- artery. There is a possibility that this may be related to icant concern. clamping the artery during implantation. Radiological stenting may sometimes be possible; however, surgical Further reading bypass of the stenosis is otherwise performed using a Mangus RS, Haag BW. Stented versus nonstented saphenous vein graft. extravesical ureteroneocystostomy in renal transplantation: a meta-analysis. Am J Transplant 2004; 4: 1889–96. Transplant nephrectomy Odland MD. Surgical technique/post-transplant surgical Unfortunately, a transplant nephrectomy is sometimes complications. Surg Clin North Am 1998; 78: 55–60. required. If performed within a short time after trans- Steffens J, Stark E, Haben B, Treiyer A. Politano-Leadbetter plant, mobilization of the kidney is usually straightfor- ureteric reimplantation. BJU Int 2006; 98: 695–712. ward, as is its removal. It is usually prudent to remove Wilson CH, Bhatti AA, Rix DA, Manas DM. Routine the donor vessels, as if left in situ with ongoing rejec- intraoperative ureteric stenting for kidney transplant tion, they have been known to rupture, causing life- recipients. Cochrane Database Syst Rev 2005; 4: threatening hemorrhage. The defect in the iliac vein CD004925.

257 Section 5 Kidney Chapter Peri-operative care and early complications

31 Lorna Marson and John Forsythe

Key points History r Careful assessment and preoperative Any change in the patient’s condition since the time preparation of the renal transplant recipient of transplant assessment should be sought. In partic- is essential, paying particular attention to ular, history of recent infection will be important in hydration status, cytomegalovirus the decision of whether or not the transplant should prophylaxis, and immunosuppressive regime. proceed. Review of any urinary problems is useful at r Dialysis immediately before transplant this stage. Other salient points to note in the history surgery is associated with delayed graft include details of dialysis, including time of last treat- function. ment and any access or other related problems. Vol- r Close postoperative monitoring is required, ume of urine output (if any) per day should be deter- especially of urine output and fluid balance. mined and a history of past or present urinary tract problems documented. Evidence of recent or current r A high level of suspicion must be maintained infection should be sought, including access site (vas- for postoperative hemorrhage, as this may cularorabdominalwall,includingperitonitis).Other compress the transplanted kidney, causing important factors include previous surgery and evi- vascular thrombosis and permanent damage. dence of ischemic heart disease and peripheral vas- r Complications following renal cular disease. Recipient blood group, tissue typing, transplantation include delayed graft and virology (cytomegalovirus [CMV], Epstein-Barr function, hemorrhage, acute rejection and virus, human immunodeficiency virus [HIV], hepati- unwanted effects of immunosuppressive tis B virus [HBV], and hepatitis C virus [HCV]) must therapy. be recorded in the notes. Donor details should also be included in recip- This chapter aims to describe the management of ient clerking, including age, cause of death, blood the renal transplant recipient from the time of their group, and tissue typing. All donor details in the notes admission to hospital to discharge, including imme- should be made anonymous. In many centers across diate preoperative preparation, postoperative manage- the world, the recipient has access to their own clinical ment, and early post-transplant complications. Surgi- notes, and therefore significant donor details recorded cal aspects of the transplant procedure are covered in in the notes would threaten donor anonymity. Impor- Chapter 30. tant clinical factors that may have some relevance to graftoutcomeshouldberecorded. Preoperative evaluation All patients undergoing renal transplantation will have Examination beenassessedpriortolistingfortransplantation,but A full physical examination of the patient should they still require a thorough review on admission for be performed, including observation of fluid status, transplant, as some time may have elapsed since their peripheral pulses, and abdominal scars/hernias. In assessment visit. addition, it is essential to examine for signs of active

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

258 Chapter 31: Peri-operative care and early complications infection. Careful fluid replacement may be required vascular event or who have had a drug-eluting coro- prior to the transplant operation. nary stent placed – usually for a defined period such as 12 months. Some centers may suspend such patients Investigations from the transplant waiting list until completion of dual therapy when they can be reactivated on aspirin Essential investigations prior to surgery include full alone. blood count, blood urea nitrogen, creatinine and elec- trolytes, baseline calcium/liver function tests, blood glucose, clotting screen/INR (if on warfarin), and Pretransplant management of oral anti-coagulants blood group and save. Viral serology status should also In living donor transplant recipients, warfarin should be checked, in particular, CMV, HIV, HBV, and HCV. be discontinued 5 days prior to the transplant, and the A negative donor–recipient human leukocyte anti- international normalized ratio (INR) should be moni- gen (HLA) cross-match is required in order to proceed tored. An INR less than 1.5 is acceptable for the trans- with transplantation. This subject is covered in detail plant to take place. If preoperative anti-coagulation in Chapter 28. is required, this can be administered via IV unfrac- tionated heparin once the INR is subtherapeutic. This Preoperative management should be stopped 6 hours before transplant. Alter- natively, subcutaneous low-molecular-weight heparin Dialysis may be used, but its effect is unpredictable in patients Hemodialysis is indicated before transplantation if withchronickidneydiseaseandsoshouldbeusedwith serum potassium is greater than 5.5 mmol/l or sig- caution. nificant fluid overload is present. It is worth noting In deceased donor transplant recipients, the INR that preoperative hemodialysis is associated with an should be checked urgently on admission. If the INR increased risk of delayed graft function, probably due is greater than 1.4, reversal of warfarin should be con- to hypovolemia. The majority of patients will require sidered using either prothrombin complex concentrate only 2–3 hours of dialysis. This should be performed or 1–2 mg of vitamin K. The disadvantage of vitamin K with minimal anticoagulation and with the aim of leav- is a relative lag time until the reversal of anticoagula- ing the patient at 1–2 kg above their dry weight. tion and a very variable impact on coagulation there- Patientsmaybe“nilbymouth”forsometimeprior after. Fresh-frozen plasma may also be administered to surgery. In these circumstances, fasting has the effect (to replace clotting factors) when there is no availabil- of increasing the risk of hyperkalemia, and an intra- ity or no experience in using prothrombin complex venous (IV) infusion of glucose-containing solution concentrate. will help control the rise in potassium. Drug therapy Immunosuppressive therapy Antihypertensive medications Most centers use induction therapy with two doses of an anti-CD25 monoclonal antibody such as basilix- These should all be reviewed during the preopera- imab at induction of anesthesia and on postoperative tive evaluation. In general, beta-blockers should be day 4. continued; however, angiotensin-converting enzyme inhibitors and related drugs should be stopped peri-operatively. Other anti-hypertensive medication Prophylaxis should be withheld in the early postoperative period and kept under regular review. Deep venous thrombosis prophylaxis Although the risk of deep venous thrombosis (DVT) Anti-platelet and anti-coagulation therapy post-transplant is relatively low (in the region of 2%), Aspirin therapy can be continued peri-operatively, as the use of DVT prophylaxis is recommended, with it has not been shown to increase the risk of postoper- compression stockings and administration of subcuta- ative hemorrhage. A combination of aspirin and clopi- neous heparin (5000 units) at anesthetic induction and dogrel is used in patients who are at high risk of a daily thereafter until the patient is mobile.

259 Section 5: Kidney

Table 31.1 Donor and recipient CMV status and risk of disease particularly if there are factors making the kidney Donor (D)/recipient (R) status Risk of CMV disease higher risk. A brief description of the procedure should be given, including the length of time for D+/R+ 10–30% the operation, the presence of drains, and a urinary D+/R– 50–80% catheter postoperatively. In addition, it is wise to men- D–/R+ 0–30% tion the presence of a ureteric stent and the need to D–/R– 0% remove this sometime after the transplant. It is also importanttowarnthepatientthatthekidneymaynot Antibiotic prophylaxis work straight away after transplantation. This is partic- A single dose of prophylactic antibiotic at the time ularly important if the donor history makes this more of surgery is indicated. The choice of drug will vary likely, e.g., donation after circulatory death (DCD) according to the individual unit’s protocol. donor. In situations in which there is an increased risk of complication postoperatively, e.g., transmission of CMV prophylaxis infection or malignancy, this should also be discussed with the patient. This should include the balance of risk The risk of developing CMV disease postoperatively between accepting such an organ and remaining on the varies depending on the donor and recipient CMV waiting list for future transplant. status (Table 31.1).Undoubtedlythereisarolefor prophylaxis: to reduce the risk of CMV disease dur- ing the early months following transplantation when The early postoperative phase the patients are most heavily immunosuppressed, The presence or absence of primary function in the although it may just delay the onset of CMV disease. earlypostoperativephaseisveryimportant.Primary Differentprophylaxisstrategiesmaybeadopted, function is expected in all living donor transplants including universal prophylaxis given to all patients for and most deceased donor procedures, particularly if whom the donor is CMV positive. The length of pro- the donor was donation after brain death (DBD) with phylaxis is currently under debate, but a recent study a short cold ischemic time. Prolonged cold ischemic in renal transplantation has demonstrated better con- time is associated with an increased risk of delayed trol of disease when prophylaxis is given for 200 days, graft function, which in turn affects acute rejection rather than 100 days. Some centers do not administer rates and long-term outcome. prophylaxis at all, but rather regularly monitor CMV levels by polymerase chain reaction assay for the first 6 Fluid management months following transplantation. Treatment with val- Urine output may vary between none and several liters ganciclovir is then commenced when a threshold num- over 24 hours. The patient must therefore be closely ber of viral copies have been reached. monitored postoperatively. Fluid balance assessment Prophylaxis against pneumocystis should be made using vital signs, central venous pres- sure, urine output, and daily weights. Early observa- Trimethoprim-sulfamethoxazole is usually prescribed tion of trends in vital signs and urine output allows for a period of 3–12 months post-transplant as prophy- the rapid diagnosis of problems such as hemorrhage, laxis against pneumocystis. or more rarely, thrombosis. Administration of IV fluids is required to replace Consent fluid losses and to maintain circulating volume. Fluid In most cases a patient information booklet will have loss after transplant comprises urine output, insensible been given to the patient well in advance. This allows losses (approximately 500 ml/24 hours), and loss from the patient to take in the necessary details of the oper- surgical drains. This fluid should be replaced with crys- ation and the balance of risk between transplantation talloid at a rate of approximately urine output (for the and remaining on dialysis. Ideally, consent should be last hour) plus 40 ml/hr for maintaining fluid status. takenwellinadvanceofthedayofsurgery. This should be carefully monitored on an hourly basis Once the results of all the investigations are for the first 24 hours following transplant. obtained and a decision has been made to proceed with It is noted above that different combinations of the transplant, informed consent should be obtained, donor/recipient will give rise to differing risk of

260 Chapter 31: Peri-operative care and early complications delayed graft function. This should be borne in mind Risk factors for the development of DGF include when assessing the patient in the first hours follow- donor factors such as donor age older than 60 years, ing transplant. For instance, in a live donor trans- hypertension, retrieval factors such as requirement of plant (which has been completely uncomplicated), the inotropic support, increased warm ischemic time (e.g., absence of urine or the presence of very small urine in DCD), prolonged cold ischemia, and recipient fac- output should be a serious finding demanding urgent tors such as obesity, HLA sensitization, hemodialy- investigation and action (see next section). sis within 24 hours prior to surgery, and long second warm ischemic time. Long-term consequences of DGF Evaluation of the oliguric/anuric patient include increased acute rejection risk, higher serum creatinine at 1 year, and, if DGF is combined with acute If an early postoperative check reveals that the patient rejection, reduced long-term graft survival. is passing little urine, a thorough clinical assess- DGFismanagedbycarefulattentiontofluidbal- ment must be undertaken. The patency of the urinary ance and avoidance of drug toxicity. The patient should catheter should be first checked. Hematuria may con- undergo a protocol biopsy on day 5 to ensure that there tribute with the formation of small clots, and gentle is no concomitant rejection process. If DGF persists, bladder washouts are required. Fluid status should be the biopsy should be repeated every 7–10 days, until determined, and if hypovolemia is suspected, careful the onset of graft function. Reduction of calcineurin fluid challenges (such as a 250-ml saline bolus) should inhibitor (CNI) levels is also appropriate once the betriedandtheresponsenoted.Thereisasignificant biopsy has demonstrated the absence of rejection, with incidence of vascular complications after renal trans- a target trough level for tacrolimus (TAC) of approxi- plant (approximately 2%). If other causes of low urine mately 5 ng/ml, using other agents to augment overall output have been excluded, then a Doppler ultrasound immunosuppression. scan should be performed, which should detect a vas- cular abnormality. Technical complications Postoperative analgesia Wound Risk factors for wound complications following renal Most patients will be managed with patient-controlled transplantation are obesity, long duration of dialy- analgesia (PCA), with intravenous morphine or fen- sis, elderly patients, high doses of immunosuppressive tanyl. In addition, they should be prescribed regular drugs such as prednisolone or sirolimus, and diabetes. paracetamol. Following removal of the PCA, regular Wound infection may be superficial or deep. If superfi- oral opioids will be required, but care should be taken cial, thorough cleaning and consideration of antibiotic to avoid unwanted side effects, including nausea, vom- therapy is required. Deep wound infections, involv- iting, itching, and decreased consciousness, since these ing fascial or muscular layers, may require surgical agents are metabolized partly through renal handling. drainage and laying open. Antibiotic treatment is indi- cated if there are systemic signs of sepsis. A less com- Early postoperative complications mon complication is wound dehiscence, which also may be superficial or deep, requiring primary or sec- Delayed graft function ondary closure as appropriate. If the patient is euvolemic, with a well-draining uri- nary catheter and a normal Doppler ultrasound scan, Postoperative hemorrhage the likely cause of a low urine output is delayed graft Usually occurring within the first 24 hours follow- function (DGF) secondary to acute tubular necrosis. ingtransplantation,thisshouldbeidentifiedrapidly DGFisdefinedastheneedfordialysiswithinthe through careful monitoring of the patient’s fluid sta- first week following transplant. DGF is unusual in liv- tus, and persistent signs of hypovolemia, even if they ing donor transplants (2–5%). It complicates 10–30% are subtle, should be assumed to be due to bleeding of all deceased donor grafts and is most common in until proven otherwise. Other signs of bleeding are DCD kidneys (50–80%). Renal function recovers in excessive amounts of drain fluid and increased anal- the majority of patients, often within the first few days, gesia requirement. If the patient is stable, a Doppler butitmaytakeseveralweeks. ultrasound scan may be useful, as it may reveal a

261 Section 5: Kidney

peri-nephric hematoma, but this may also just delay firmed if the fluid contains high levels of creatinine and definitive management. There should be a high level of potassium. suspicion and the patient should return to the operat- Management includes replacement of the urethral ing theater if there is any doubt. Ongoing bleeding into catheter. If a stent is in place, it is likely that the patient a relatively small anatomical space will cause pressure will require surgical treatment. If, however, there is on the renal vessels and may lead to vascular thrombo- no stent in place at the time that the leak is detected, sis. When the patient returns to surgery, the most com- some time may be bought by placing a percutaneous mon site of hemorrhage is around the renal hilum, not nephrostomy and performing a nephrostogram, which the anastomosis. will help delineate the site of the leak. Some centers will attempt to treat the problem by antegrade placement Graft thrombosis ofastentacrossthesiteoftheleak.However,thebest This predominantly occurs within the first week treatment is usually surgical re-exploration and repair, post-transplant. Venous thrombosis is more common, often by performing a new ureteric to bladder anasto- occurring in between 2–4%. Renal artery thrombosis mosis. Ureteric obstruction is usually a late complica- is rare, with an incidence of around 1%. Risk factors tion following renal transplantation and can be largely for graft thrombosis include retrieval injury to donor avoided in the early postoperative phase by the routine vessels, atherosclerosis of donor or recipient vessels, use of a ureteric stent. technical error with the anastomoses, angulation of the vessels, or external compression from a hematoma Lymphocele formation or lymphocele. In addition, persistent postoperative This complication commonly occurs between 2 weeks hypotension may lead to thrombosis as a result of the and months following transplantation and is due to low flow state. Graft thrombosis should be excluded in leakage from recipient lymphatics dissected at the time cases where there is acute graft dysfunction, particu- of surgery. The fluid can build up to a large quantity larly if there was initial function and abrupt anuria or and exert significant compression on the transplanted oliguria ensues, despite adequate fluid management. kidney, giving rise to ureteric obstruction. In some sit- Inaddition,inasituationwherethekidneywouldbe uations the lymphocele may press on the iliac veins, expected to have primary function, such as a living with increased swelling and discomfort in the leg on donor transplant, and the patient is oliguric or anuric, the side of the transplant. Aspiration may be required this should be excluded as a matter of urgency. Venous as a matter of urgency to reduce the pressure. Defini- thrombosis will cause severe pain as the graft becomes tive treatment is surgical, with fenestration of the peri- swollen. Urgent Doppler ultrasound will confirm the toneum overlying the lymphocele, allowing drainage diagnosis, with absent venous flow and reversed arte- of the lymphatics into the peritoneal cavity. This can rial diastolic flow. The patient should be returned to usually be performed laparoscopically and be aided by surgery urgently if there is to be any chance of sal- the use of laparoscopic ultrasound to ensure no dam- vaging the graft. Graft nephrectomy is by far the most age to surrounding structures. likely outcome, and the patient must be warned of this prior to return to the operating theater. Early immunological complications Urological complications Acute rejection Ureteric leak or obstruction may occur due to a sur- Acute rejection occurs in 10–25% of renal transplants gical problem, distal ureteric ischemia, or immuno- and is most common during the first 3 months follow- logical damage to the ureter. Ureteric leak tends to ing transplantation. Risk factors for the development occur in the early postoperative phase and presents of acute rejection include sensitization (through pre- with increased wound pain and/or increasing fluid dis- vious transplant, pregnancy, and blood transfusion), charge from the wound or drain, commonly occurring HLA mismatches, and DGF. In addition, the immuno- following removal of the urethral catheter. Ultrasound suppressive regimen adopted will have an impact; may detect fluid around the kidney, which, if aspirated, induction therapy with anti-CD25 antibodies reduces should be sent for biochemistry. A urine leak is con- acute rejection rates, mycophenolate mofetil (MMF) is

262 Chapter 31: Peri-operative care and early complications a more effective agent than azathioprine (AZA), and Table 31.2 Banff 1997 classification of T-cell–mediated TAC is probably associated with lower rates of acute rejection (TCMR) rejectionthancyclosporineinrenaltransplantation. Borderline changes Patients present with acute graft dysfunction, with Suspicious for TCMR increasing serum creatinine and oliguria. This is usu- No arteritis ally detected by medical staff as the patient is often asymptomatic. Doppler ultrasound scan is required Tubulitis without significant interstitial inflammation or to exclude other causes of renal dysfunction, such as Interstitial inflammation without significant tubulitis ureteric obstruction or vascular complications, usually Acute TCMR followed by renal biopsy. In addition, serum should be Acute tubulo-interstitial rejection sent for antibody screening, particularly if antibody- 1A: Significant interstitial inflammation and moderate tubulitis mediated rejection (AMR) is suspected. 1B: Significant interstitial inflammation and severe tubulitis Treatment should be commenced prior to biopsy Acute vascular rejection results if there is a high index of suspicion, in the form of pulsed IV methylprednisolone (250–500 mg). 2A: mild or moderate arteritis with or without interstitial inflammation or tubulitis Immunosuppression should be optimized, with care- 2B: Severe arteritis with or without interstitial inflammation or ful monitoring of CNI levels. The majority of rejection tubulitis episodes are treated effectively with pulsed steroids, 3: Transmural arteritis ± fibrinoid necrosis but when the response is poor or there is a rapid return of rejection following treatment, alternative therapies can be used, such as polyclonal antibodies, e.g., anti- rejection episodes are confined to the tubulointersti- thymocyte globulin (ATG). tium, with vascular involvement associated with worse response to therapy and reduced graft survival. Early antibody-mediated rejection Acute AMR accounts for approximately 10% of rejec- Early complications of immunosuppressive tion episodes in the early post-transplant period. The therapy diagnosis can be difficult to make and is based on biopsy findings of acute tissue injury, with endothe- CNIs lial cells staining positive for C4d (discussed further Monitoring of levels is essential to ensure adequate in Chapter 2). In addition, AMR is associated with the drug exposure while minimizing exposure to toxic presence of anti-donor HLA donor-specific antibod- side effects such as nephrotoxicity. Initially, trough lev- ies (DSAs) on serological testing. AMR is caused by elsshouldbetakenregularlyandthetreatmentdose the binding of DSAs to graft endothelial cells, which should be amended accordingly. Optimal levels will leads to complement activation and endothelial cell depend on various factors such as HLA mismatch, injury. whether the kidney is DCD or DBD, and the pres- The aim of treatment of AMR is to remove DSAs, ence of previous sensitizing events. Patients with dete- which may reduce the likelihood of further injury, riorating renal function in the context of high TAC and to reduce vascular inflammation. Daily plasma levelscanbemanagedwithdosereductioninthe exchange removes circulating DSAs, and pulsed doses first instance, progressing to biopsy if this does not of methylprednisolone reduce vascular inflammation, result in improved function. Histologically, TAC tox- with the added benefit of treatment of any associated icity is associated with arteriopathy, with vascular hya- T-cell–mediated rejection. line deposits. Another less common but devastating side effect T-cell–mediated rejection of CNIs is the development of thrombotic microan- This accounts for the majority of rejection episodes giopathy, with deterioration in renal function in and is characterized by lymphocytic infiltration of the association with biopsy findings of endothelial cell interstitium, tubules, and vessels. Such rejection is swelling and capillary thrombi with fibrinoid necro- classified according to the Banff 1997 working clas- sis in arterioles. It may or may not be associated with sification, as outlined on Table 31.2.Themajorityof hematological abnormalities, such as falling platelets

263 Section 5: Kidney

and microangiopathic hemolytic anemia. CNIs should glycemic control may be an early side effect, in both be stopped and daily plasma exchange may be required diabetic and non-diabetic patients. until a response is seen, continuing treatment on alter- nate days thereafter for 2–4 weeks. This strategy can Further reading be accompanied by treatment with ATG to cover the LuanFL,StuckeyLJ,ParkJM,KaulD,CibrikD,OjoA. CNI-free period before considering options for long- Six-month prophylaxis is cost effective in transplant term immunosuppression. patients at high risk for cytomegalovirus infection. JAm Soc Nephrol 2009; 20: 2449–58. Anti-proliferative agents Racusen LC, Colvin RB, Solez K, et al. Antibody-mediated rejection criteria – an addition to the Banff 97 Early complications include myelosuppression, with classification of renal allograft rejection. Am J Transplant a fall in hemoglobin, white blood cells, and platelet 2003; 3: 708–14. count, requiring cessation of therapy if severe. MMF RacusenLC,SolezK,ColvinRB,et al. The Banff 97 is commonly associated with diarrhea, which can be working classification of renal allograft pathology. managed by altering the dosing regimen from 1 g twice Kidney Int 1999; 55: 713–23. daily to 500 mg four times a day. An alternative prepa- Taylor CJ, Kosmoliaptsis V, Sharples LD, et al. ration, Myfortic, is available, which, in some patients, Ten-year experience of selective omission of the has fewer gastrointestinal side effects, and so this could pretransplant crossmatch test in deceased donor be tried, or the patient could be switched to AZA if the kidney transplantation. Transplantation 2010; 89: diarrhea is severe. 185–93. Yarlagadda SG, Coca SG, Formica RN Jr, Poggio ED, Parikh Steroids CR. Association between delayed graft function and allograft and patient survival: a systematic review and The majority of side effects of steroids occur in the meta-analysis. Nephrol Dial Transplant 2009; 24: later period following transplantation. However, poor 1039–47.

264 Section 5 Kidney Chapter Long-term management and outcomes

32 Sharon Mulroy and John D. Firth

Key points each year. Chronic graft dysfunction, the progres- r sive loss of glomerular filtration rate (GFR) occurring Advanced donor age is the strongest months and years after transplantation such that most predictor of poor long-term graft survival. r patients eventually lose their grafts, remains a major Transplant glomerulopathy is an challenge for transplant clinicians. This chapter con- alloimmune-mediated lesion that is strongly centratesonissuesthatarisefrom6monthsaftertrans- linked to anti–human leukocyte antigen plant onwards and considers issues in the early post- (HLA) class II antibodies. r transplant period only insofar as they affect long-term Proteinuria is common after renal management and outcome. transplantation (45% of cases at 1 year) and is an important marker of graft injury. Graft and patient survival r Renal transplant recipients are at a 3–5-fold higher risk of cancer than the general The Organ Procurement and Transplantation population. Network/US Scientific Registry of Transplant Recip- ients report shows an increasing number of renal r Urinary tract infection is the most common transplants performed in the United States between infection in kidney transplant recipients and 1998 and 2007, with deceased donors increasing from usually occurs in the first year following 7898 to 10 083 and live donors increasing from 4409 transplantation. to 6033. One-year graft survival has also increased r Chronicallograftnephropathy(CAN) during the same time from 88.8 to 91.4% for patients remains the most common cause of chronic receiving grafts from deceased donors and 94.6 to graft dysfunction, where the mainstay of 96.5% for those receiving grafts from live donors. treatment is calcineurin inhibitor reduction One-year patient survival is also excellent at 95.9% for or withdrawal. recipients of deceased donor kidneys and 98.7% for recipients of live donor kidneys in 2007. Long-term Kidney transplantation is a common and routine treat- survival is also good, with 10-year graft survival ment. Over the last 20–30 years, there have been great of 43.3% and 59.3% for recipients of deceased and improvements in short-term graft survival. This is live donor grafts, respectively, and patient survival explained by a number of factors, including advances of 61.2% and 77.1%, respectively. Extended criteria in human leukocyte antigen (HLA) typing and cross- donor graft survival is inferior, with 80.6% 1-year and match techniques, newer immunosuppressive agents, 27% 10-year survival in 1998, the first year for which and better methods for the prevention and treat- 10 year data is available. The largest report looking ment of early infection. Very good short-term results at the difference between donation after cardiac now mean that management emphasis is increasingly death (DCD) and donation after brain death (DBD) on medium- to long-term outcomes. However, the donors (1998–2004) showed that approximately 3% rate of attrition of “established” graftsgraft ( half- of kidneys were from DCD donors but that there was life) has not improved substantially over many years, no difference in 5-year patient (81%) and graft (67%) with approximately 4% of grafts continuing to fail survival rates. However, there was a difference in the

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

265 Section 5: Kidney

Table 32.1 Risk factors associated with poor graft survival deathwithafunctioninggraftisthesinglemostimpor- Donor factors Advanced age tant cause of graft loss, accounting for up to 50% of Hypertension all graft failures occurring after the first year after Vascular disease transplant. The most common causes of death in renal Type (living better than cadaveric) transplant recipients are premature cardiovascular dis- Peri-transplant Type of cadaveric donor (cardiac death less factors good than brain death) ease, infection, and malignancy, all of which are exac- Long cold ischemia time erbatedbyorcausedbyimmunosuppression. Immune High panel-reactive antibody titres factors Many HLA mismatches Acute rejection (some types) Chronic graft dysfunction Non-immune Donor/recipient size/gender mismatch Chronic graft dysfunction is the progressive loss of factors Recipient hypertension GFR beginning months or years after transplantation. Poor drug compliance Drug toxicity It is extremely common and in most patients ulti- Recurrent disease mately leads to graft failure (also see Chapter 4B). Spe- Infection – CMV, BKV cific causes of graft injury are listed in Table 32.2 and are discussed in detail next, but it is important to rec- incidence of delayed graft function: 41% and 24% for ognize that graft injury is often multi-factorial. DCD and DBD, respectively. Injury begins prior to transplantation. The peri- transplant process results in variable acute injury Causes of graft loss dependant on the mode of patient death and degree of ischemia–reperfusion injury, which occurs on a vari- A number of factors are correlated with poor graft out- able background of pre-existing donor kidney dis- come (Table 32.1 and Figure 32.1). With the advent of ease (age, hypertension, vascular disease). The impor- more powerful immunosuppressive medications and tance of such injury is emphasized by studies of living the expansion of the donor pool to include older spousaldonorkidneysthatrevealenhancedsurvival donors with cardiovascular comorbidity, the impact of compared with cadaveric transplants, despite poor HLA matching on outcomes in deceased donor trans- HLA matching. plantation has diminished considerably. Advanced Graft injury continues after transplant in response donor age is now the strongest predictor of poor long- to both immune and non–immune-mediated mech- term graft survival. anisms and leads to chronic graft dysfunction, the terminology of which is confused. The most com- Death with a functioning graft mon cause is chronic allograft nephropathy (CAN), a Although patient survival after transplantation is supe- term that should be applied to grafts with damage rior to that of patients who remain on dialysis, it is characterized by interstitial fibrous and tubular atro- still inferior to that of the general population. Indeed, phy (IF/TA) identified on biopsy. The term “chronic

12.50% Figure 32.1 Causes of graft loss (censored for mortality). Other 7.50% Recurrent glomerulonephritis

Acute rejection 45% 5% Transplant glomerulopathy

15% De novo glomerular disease

Chronic allograft nephropathy (IF/TA) 15%

266 Chapter 32: Long-term management and outcomes

Table 32.2 Causes of graft injury clinical course variable, reflecting the fact that CAN Pretransplant factors (IF/TA) is the common end point of many different causes of graft injury. The most important risk factors Acute peri-transplant Ischemia–reperfusion injury injury Brain death for CAN (IF/TA) are acute rejection, donor age, pre- existing donor disease, and exposure to calcineurin Pre-existing donor Cold ischaemia time disease Age-related GFR loss inhibitors (CNIs). Hypertension Atheromatous vascular disease Transplant glomerulopathy Post-transplant factors Transplant glomerulopathy is a specific histological Immune-dependent Acute T-cell or antibody-mediated rejection lesion characterized by thickened glomerular capillary Subclinical acute cellular rejection walls that have double contours, with reduplication or Chronic antibody-mediated lamination of the glomerular basement membrane on rejection Transplant glomerulopathy electron microscopy. The condition can be subclinical Recurrent glomerulonephritis as an isolated histological lesion, but over time is asso- Immune-independent Recipient factors ciated with progressive worsening of histopathologi- Hypertension cal change and the development of CAN (IF/TA) and Diabetes mellitus chronic vasculopathy. Incidence increases with time, Renovascular disease Transplant artery stenosis beginning in the first year after transplant (4% of all Atheromatous disease biopsies at 1 year, increasing to 20% at 5 years). Clin- Calcineurin inhibitor toxicity ical presentation is with proteinuria (often nephrotic Urinary obstruction Urinary sepsis range), hypertension, and allograft dysfunction. CMV (associated with There is strong evidence from animal models and “over-immunosuppression”) clinical studies that transplant glomerulopathy is an BKV (associated with “over-immunosuppression”) alloimmune-mediated lesion that is strongly linked to anti-HLA class II antibodies. Donor-specific antibod- ies (DSAs; most commonly class II) are present in up rejection” has often been used incorrectly in this con- to 75% of cases, and staining with C4d is usually (but text but should be reserved for the situation in which not always) positive. Risk factors include prior sensiti- biopsy has shown an immune-mediated cause (cell or zation, the presence of HLA antibodies (class II Ͼclass antibody mediated) of graft injury. To further dissuade I) prior to transplantation, and acute rejection. There from imprecise usage of the term CAN to describe any is a high incidence in patients undergoing desensitiza- form of chronic graft dysfunction, the 2007 Banff clas- tion protocols (see Chapter 28). sification recommended that it be replaced in descrip- Even in the absence of significant proteinuria or tion of renal transplant biopsy findings with the phrase graft dysfunction at diagnosis, transplant glomeru- “interstitial fibrosis and tubular atrophy without evi- lopathy is a progressive condition associated with poor dence of specific aetiology” (IF/TA), with an emphasis outcome. In one study of patients undergoing protocol on differential diagnosis from specific causes of graft biopsy at 1 year, more than 50% of those with the con- injury. dition reached a combined end point of graft loss or loss of greater than 50% GFR over the next 36 months. Chronic allograft nephropathy Chronic allograft nephropathy (IF/TA without evi- Chronic antibody-mediated rejection dence of specific etiology) is not a diagnosis but Chronic antibody-mediated rejection is diagnosed describes a non-specific response to graft injury that based on the triad of presence of circulating DSAs, is the most common reported cause of death-censored positive C4d staining, and morphologic evidence renal allograft loss (Figure 32.1). Moderate to severe of chronic tissue injury on allograft biopsy includ- CAN (IF/TA) is present in 25% of renal transplants at 1 ing transplant glomerulopathy (glomerular double year and 90% by 10 years. Clinically it is characterized contours), peritubular capillary basement membrane by progressive renal dysfunction, hypertension, and multi-layering, IF/TA, and arteriolar fibrous intimal variable proteinuria. Its onset is unpredictable and the thickening. The condition is present in 10–30% of

267 Section 5: Kidney

biopsies done to evaluate chronic graft dysfunction pare favorably with those of cadaveric grafts. Aggres- andisassociatedwithpoorgraftoutcome.Riskfactors sive recurrence is rare, and as with native IgA disease, include prior sensitization, DSAs (class II Ͼ class I), there is no proven treatment, although corticosteroids and the degree of HLA mismatch. The contribution of and cyclophosphamide have been tried. non-HLA antibodies is poorly understood. Atypical hemolytic uremic syndrome Recurrent glomerulonephritis Recurrence in patients who develop end-stage renal Recurrent glomerulonephritis is common but does not disease following diarrheal associated (D+)HUSis always lead to allograft loss, with recurrent disease very uncommon. Atypical (no diarrhea; D–) HUS often following an indolent course. Overall, 10-year (aHUS) is a rare disorder caused by dysregulation of graft survival is similar in patients with biopsy-proven complement pathways that results in the development glomerulonephritis as their primary renal diagno- of thrombotic microangiopathy within the kidney. sis as for those with other causes of renal failure, Recurrence occurs after transplantation in approxi- and hence glomerulonephritis is not a contraindica- mately 50% of patients with aHUS, and no interven- tion to transplantation in most cases. Biopsy-proven tion has been shown to be effective, with graft loss glomerulonephritis is the primary cause of graft loss almost inevitable. in around 20% of kidney transplant recipients, with the 10-year incidence of allograft loss due to recur- Renovascular disease rent glomerulonephritis being 8.4% in one large series. Recurrence rates vary between the different types Transplantrenalarterystenosis(RAS)usuallypresents of glomerulonephritis, with primary focal segmental between 3 months and 2 years after transplant, but can glomerulosclerosis (FSGS), immunoglobulin A (IgA) occuratanytime.Itisarelativelycommonandpoten- nephropathy, and atypical hemolytic uremic syndrome tially curable cause of refractory hypertension and (HUS) being most likely to recur. graft dysfunction, usually presenting with an asymp- tomatic rise in creatinine (possibly in relation to com- mencing an angiotensin-converting enzyme [ACE] Focal segmental glomerulosclerosis inhibitor or an angiotensin receptor blocker [ARB]), Primary FSGS recurs in 30–40% of renal transplants, hypertension that is difficult to treat, and/or fluid over- perhaps more commonly in grafts from living donors. load.Thepresenceofabruitisneithersensitivenor Recurrence usually (Ͼ80%) occurs within the first year specific. and may happen within hours of transplantation. Risk RAS usually arises close to the surgical anastomo- factors include rapid progression of the primary dis- sis, but may occur more distally. It is the result of inti- ease to end-stage renal failure (Ͻ3years),pediatric mal scarring and hyperplasia in response to injury to recipients, and recurrence in a previous renal allo- the donor or recipient vessels during harvesting or graft. Indeed, recurrence in a second graft is almost transplantation. Stenoses may also occur several years inevitable (Ͼ80%) in the event that a first renal allo- after transplant as a result of development of athero- graft is lost to recurrent FSGS. Several studies have matous disease in the renal transplant artery or more reported a variable response to plasmapheresis (up to proximally in the iliac vessels. 50%), but relapse occurs in more than half of these patients when the treatment is stopped. Others have Polyomavirus associated nephropathy reported success with rituximab or cyclophosphamide, but this is anecdotal (case reports and small series). Polyomaviruses (BK virus [BKV], JC virus [JCV], and SV40) are small, non-enveloped dsDNA viruses that can infect, and remain latent in, a large number of IgA nephropathy target tissues, including urothelium, becoming reac- IgA nephropathy recurs in approximately 30% of tivated during states of immunosuppression. In more patients by 10 years after transplant, with graft loss than 95% of cases, polyomavirus-associated nephropa- resulting in up to 10%. The risk of recurrence is higher thy is caused by BKV,which is an emerging problem in with a transplant from a living related donor, but this renal transplantation, with reported incidence varying is not a strong contraindication as outcomes still com- from 1–10%.

268 Chapter 32: Long-term management and outcomes

There are no known clinical risk factors that Table 32.3 Clinical history and investigation of graft clearlyidentifythekidneytransplantrecipientwho dysfunction will develop BKV nephropathy. Pretransplantation Clinical history – points of particular relevance seroprevalence for BKV varies from 60–80%. BKV Original diagnosis and clinical course of primary renal disease nephropathy is more frequent in seronegative recipi- Number and severity of previous rejection episodes ents of a seropositive donor kidney (BKV D+/R–), but it is not confined to this group, and BKV reactivation Urological symptoms may come from the donor or recipient. A high overall Compliance with medication burden of immunosuppression is a risk factor for BKV Drugs nephropathy, but there is no compelling evidence that Nephrotoxic agents, e.g., NSAIDs, ACE inhibitors/ARBs Drugs that raise plasma creatinine but not by reducing GFR, any particular immunosuppressive agent is worse (or e.g., trimethoprim better) than any other. Investigations BKV reactivation is common after renal transplan- HLA antibody screen to look for de novo donor-specific tation, with viruria detected in up to 30% of patients. antibody (DSA) BKV infection can be detected by polymerase chain Current and historical calcineurin inhibitor levels reaction (PCR) for BKV DNA in plasma or urine and BK virus screening by examination of the urine cytologically for BKV inclusion-bearing epithelial cells (decoy cells) or by Urine Microscopy electron microscopy for viral particles. These tests are Culture sensitive for detecting active viral replication but are Urinary protein quantification not specific for nephropathy. A negative BKV urine or Cytology plasma PCR excludes the diagnosis of BKV nephropa- Ultrasound renal transplant thy. Quantitative PCR for BKV in plasma may be pre- Colour Doppler renal transplant artery dictive for BKV nephropathy, with a threshold value Assess bladder emptying ≥ ( 10 000 copies/ml) having been suggested, below +/– Imaging native kidneys which the condition is unlikely. Biopsy remains the gold standard for diagnosis of BKV nephropathy, which causes interstitial inflamma- effective in preserving graft function. Most patients tion and tubular cytopathic changes, often with visible with progressive loss of GFR are asymptomatic and viral inclusions. However, these changes may be patchy present with a rise in serum creatinine, by which time and isolated to the medulla and may be missed in up to significant graft damage (IF/TA) may already have one third of patients if a single core is taken; hence at occurred. Some centers therefore advocate protocol least two cores should be examined. Diagnosis should biopsy to facilitate the early detection of IF/TA and be confirmed by immunohistochemical staining spe- its specific causes, but biopsy is not without hazard, cific for BKV or cross-reacting SV40 large T antigen. and there is no evidence that this approach results in The histological differentiation of BKV nephropathy improved long-term graft outcomes. Serial measure- from acute cellular rejection can be difficult, but the ment of GFR may make earlier detection of graft dys- absence of definitive features of acute cellular rejection function possible. such as endothelialitis is helpful. Any deterioration in graft function requires a thor- Late diagnosis of BKV nephropathy, or misdiagno- ough clinical history, examination, and investigation sis (and treatment) as acute rejection, results in rapid to exclude potentially reversible causes such as dehy- loss of graft function in more than 50% of patients. dration, obstruction, urinary tract infection, trans- Early diagnosis and timely reduction in immunosup- plant renal artery stenosis, or BKV nephropathy (Table pression has improved outcomes significantly. 32.3). If initial investigations do not reveal a cause for the loss of GFR, then renal transplant biopsy is nec- essary in most patients to make an accurate diagno- Strategies to improve graft function sis, with it being particularly important to distinguish The multiple mechanisms of graft injury that are asso- between (1) ongoing immune-mediated graft injury, ciated with or lead to long-term attrition (Table 32.2) which will usually require enhancement of immuno- mean that no single intervention is, or is likely to be, suppression; (2) CAN (IF/TA), for which the mainstay

269 Section 5: Kidney

of treatment is CNI reduction or withdrawal; and Box 32.1 Recommendations for CNI withdrawal (3) BKV nephropathy, for which reduction in overall Patient Ͼ 6–12 months post-transplant immunosuppressive burden is generally required. No evidence of current immune-mediated acute or chronic graft injury on renal transplant biopsy Management of chronic allograft Able to tolerate a therapeutic dose of anti-metabolite Mycophenolate mofetil 2 g daily or mycophenolate nephropathy (IF/TA) sodium 1440 mg daily There is no consensus regarding the best management Azathioprine (1.5–2 mg/kg per day) of CAN (IF/TA). The differential diagnosis between Start prednisolone 10 mg once daily if not on CNI-related nephrotoxicity and other causes of renal corticosteroids graft injury remains very difficult. Histological lesions Reduce cyclosporine dose by 30–50% every considered relatively specific for CNI toxicity (see 2–4 weeks above) can occur through other processes, and there Increased graft surveillance until CNI withdrawal is is considerable inter- and intra-observer variability in complete the histological scoring of renal transplant biopsies. Despite these difficulties, in the absence of any evi- dence of ongoing immune-mediated graft injury, CNI MMF has replaced AZA in routine protocols in toxicity remains the major modifiable factor for pro- many centers over the last decade. However, despite gressive CAN (IF/TA). the widely held prejudice that it is a more effective Immunosuppressive regimens that avoid CNI use immunosuppressant than AZA, CyA withdrawal in completely are generally regarded as being associ- patients receiving CyA, MMF, and corticosteroids is ated with an unacceptable risk of acute rejection. also associated with an increase in acute rejection, up Recent efforts have therefore focused on devising regi- to 10% of cases in some studies. This is balanced by mens that minimize CNI exposure, either by substitu- stabilization of, or improvement in, short-term graft tion with a mammalian target of rapamycin (mTOR) function and a trend toward a sustained improve- inhibitor or maintenance on prednisolone and an mentingraftfunctionat5years.Therearelittle anti-metabolite alone (usually mycophenolate mofetil data on TAC withdrawal, but similar principals prob- [MMF] if tolerated at a therapeutic dose). Although ably apply. Due to these concerns regarding the safety there are many uncontrolled studies in this area, there of CNI-free immunosuppression, recent studies have are few randomized controlled trials, and none have looked at minimizing CNI exposure in combination focused on longer-term graft outcomes. Some data with interleukin-2 (IL-2) antibody induction, MMF, suggest that tacrolimus (TAC) may be less nephro- and steroids. This strategy seems safe and effective toxic than cyclophosphamide (CyA), but long-term in the short term, but long-term outcome data are outcome data are lacking. needed. Practical details of how to withdraw CNIs are given in Box 32.1. Such withdrawal has an additional CNI reduction and withdrawal benefit in that it may also reduce hypertension and The main risk of CNI withdrawal is acute rejection. dyslipidemia. A meta-analysis of 13 studies of CNI withdrawal and Several early studies reported that replacement subsequent maintenance on an anti-metabolite and of CNI with sirolimus (SRL) in patients with CAN prednisolone showed that the excess risk of acute rejec- resulted in short-term improvement or stabilization of tion is around 10% when CyA is withdrawn from graft function in most cases, although many could not an azathioprine (AZA)-based regimen. This caused tolerate the new drug, and discontinuation rates were great concern, but subsequent long-term data revealed high.However,itisnowclearthatSRLisnotthehoped a trend toward improved graft survival in patients for panacea. In many studies it has been shown that weaned from CyA (relative risk [RR] for graft failure = switching to SRL produces poor outcomes in patients 0.92). Nevertheless, concern regarding acute rejection with significant proteinuria. Recent data from a large has meant that maintenance therapy with AZA and randomized controlled trial (the CONVERT Trial) did steroids is not routine practice, although it may be a notshowasignificantimprovementinGFRafterSRL good option in some cases of CAN (IF/TA). conversion. A subset of patients with well-preserved

270 Chapter 32: Long-term management and outcomes

Box 32.2 Recommendations for CNI conversion to Screening for and management an mTOR inhibitor No evidence of current immune-mediated acute or of polyoma virus–associated chronic graft injury on renal transplant biopsy nephropathy No significant proteinuria Many centers now advocate screening, by urine cytol- Malignancy ogy or BKV PCR (plasma or urine), for the first 2 Start prednisolone 10 mg once daily if not on years after transplantation. This allows the early iden- corticosteroids tification of patients with BKV reactivation who are at risk of developing BKV nephropathy and timely inter- vention by preemptive reduction in immunosuppres- graft function (GFR Ͼ 40 ml/min) and normal uri- sion before irreversible parenchymal damage (IF/TA) nary protein excretion did show a modest benefit, but occurs. Various strategies have been employed, with patients with GFR less than 40 ml/min tolerated con- one approach being to withdraw the anti-metabolite version to SRL poorly, with an unacceptable number and to minimize steroid dose in recipients on a stan- of treatment-related adverse events. There is now con- dard CNI-based triple therapy regimen. sensus that waiting until there is advanced permanent There are observational data that cidofovir and graft damage is associated with poor outcomes after the fluoroquinolones may improve viral clearance, and conversion to SRL, but it seems reasonable to consider both have been shown to have anti-BKV activity in conversion in patients with early CAN (IF/TA) and vitro, but there is no strong evidence that they offer without significant proteinuria (Box 32.2). The occur- any advantage above reduction in immunosuppres- rence of post-transplant malignancy provides a further sion alone. Intravenous IgG has also been used to rationale for switching. treat BKV nephropathy. Given its immunomodulatory effects, as well as anti-BKV properties, this might offer benefit in patients with concurrent acute rejection, but Use of novel immunosuppressive agents because of its cost and unproven efficacy, intravenous It is hoped that novel non-nephrotoxic immunosup- IgG should not be used routinely. pressive agents may offer better options for CNI-free The main risk of reducing immunosuppression is induction and maintenance regimens in the future, acute rejection, but failure to clear virus leads to poor butmanyinitiallypromisingpossibilitieshavenot graft function and outcome. Close monitoring of viral delivered. Recent studies using costimulatory block- titers and graft function is essential. It may take sev- ade with belatacept in combination with MMF showed eral weeks for BK-virus specific immunity to recover, results that have encouraged some, but a significant during which time viremia persists and renal injury excess of post-transplant lymphoproliferative disorder may continue to accumulate. Following reduction in (PTLD) indicates that a substantial note of caution is immunosuppression, if BKV titres do not fall, further in order. reduction in immunosuppression may be required, and further biopsy is indicated to exclude rejection (or other pathology) if graft function worsens. Repeat Management of chronic imaging to exclude obstruction is also sensible in this antibody-mediated rejection and/or context because BKV has been associated with the transplant glomerulopathy development of ureteric stricture. There are no data from randomized controlled tri- als (RCTs) that any specific therapy improves long- Other long-term issues term outcome when renal transplant biopsy shows chronic antibody mediated rejection and/or trans- Adherence to treatment plant glomerulopathy. Some studies have reported Non-adherence is common and is often overlooked in success with conversion to TAC- and MMF-based the busy transplant clinic. Studies have estimated up regimens. Other strategies involve antibody removal to 50% of transplant patients may be non-adherent (plasmapheresis), intravenous immunoglobulin, and at some time; the spectrum of behaviors ranges from rituximab. the occasional missed dose to consistently missing

271 Section 5: Kidney

100 Figure 32.2 Kidney transplant survival from 3–10 years in relation to systolic blood pressure at 1 and 3 years post-transplant. The respective 1- and 3-year pressures 90 are indicated to the right of each curve, together with the number of patients (n) studied. 80 1 yr 3 yr ≤140 ≤140 n = 11450 70 >140 ≤140 n = 4118 60 ≤140 >140 n = 4058 >140 >140 n = 4778 50 % Grafts Surviving

40

0 0 246810 Years

all medication doses. Decreased adherence is asso- tedthattherearenofirmdatatosaythatsuchtreat- ciated with acute rejection and a seven-fold overall ment alters allograft survival. risk of allograft loss and is thought to precede up to one third of allograft failures, particularly late allo- graft losses. Non-adherence is often non-intentional. Prevention of cardiovascular morbidity Common barriers include simple forgetfulness, the and mortality complexity of immunosuppressive regimens, prob- The incidence of cardiovascular disease is very high lems with filling prescriptions (including cost in many after kidney transplantation, with the annual rate of countries), and lifestyle factors. Improved communi- fatal or non-fatal cardiovascular events being 3.5– cation and a clear understanding of the patient’s per- 5.0%, much higher than that of the general popula- spective are important in identifying a solution. Inten- tion. Strategies to reduce cardiovascular risk focus on tional non-adherence, where the patient does not want minimizing time on dialysis, prevention and aggres- to take their medication, is more difficult to address. sive treatment of traditional cardiovascular risk fac- Again, it is very important to gain the patient’s per- tors (hypertension, diabetes mellitus, dyslipidemia, spective. Drug-related side effects, the patient’s own discouragement of smoking), and preservation of graft beliefs regarding their medications, and peer pressure function. (particularly among adolescents) are all important factors. Management of hypertension Anti-proteinuric measures Hypertension is extremely common in renal transplant Proteinuria is common after renal transplantation recipients, with prevalence 50–90% in various studies. (45% of cases at 1 year) and is typically low grade It remains an independent risk factor for cardiovascu- (≤500 mg/d in two thirds of cases). It is an impor- lar events after transplantation and is associated with tant marker of graft injury: there is a strong associa- reduced graft survivalFigure ( 32.2). However, there tion between proteinuria and reduced graft survival, have been no RCTs in renal transplant recipients to with the level of proteinuria further stratifying this determine whether blood pressure lowering improves risk – 3.9% of recipients with proteinuria less than graft survival, and if so, what blood pressure should be 150 mg/d 1 year post-transplant will have lost their targeted. Nevertheless, in view of the potential benefits graft by 5 years, compared with 41.2% with proteinuria to both patient and graft survival, the general consen- greater than 3000 mg/d. Given this, it seems reasonable sus is that hypertension should be treated aggressively, that anti-proteinuric measures are applied rigorously. and most transplant physicians target a blood pressure These include the use of the maximum tolerable dose of less than 130/80 mmHg as recommended for other of an ACE inhibitor and/or ARB, but it must be admit- populations of high-risk patients.

272 Chapter 32: Long-term management and outcomes

Table 32.4 Causes of post-transplant hypertension complications include groin hematoma and pseudo Drugs – CNI, corticosteroids aneurysm formation. There is a risk of contrast- Kidney allograft dysfunction induced nephrotoxicity, and cholesterol embolic disease has been reported. Renovascular Transplant renal artery stenosis Percutaneous transluminal angioplasty is the treat- Atheromatous arterial disease proximal to the allograft artery ment of choice for most patients with a hemodynam- anastomosis ically significant transplant RAS. Re-stenosis is com- Factors relating to native kidneys mon (30%) and may be treated by repeat angioplasty with or without stenting. Drug-eluting stents may offer There are many reasons why renal transplant recip- an advantage in the prevention of re-stenosis, but their ients have hypertension (Table 32.4). When the blood use in peripheral vascular disease has been disap- pressure is difficult to control, and particularly when pointing, and there are very little data on their use it is associated with unexplained graft dysfunction, or in the treatment of transplant RAS. Surgery may be there is more than 20% decline in renal function after required for patients with stenoses who are unsuit- introduction of an ACE inhibitor or ARB, patients able for angioplasty, or in whom angioplasty has been should be screened for transplant RAS. unsuccessful. Color Doppler ultrasonography is a non-invasive Aside from the specific matter of transplant RAS, method of diagnosing transplant RAS (87–94% sen- all patients should be given advice on lifestyle mod- sitivity, 86–100% specificity), but is very dependent ifications (relating to low-salt diet, exercise, alcohol ontheskillandexperienceoftheoperator.Magnetic intake, and weight loss), which can produce small but resonance angiography is non-invasive and very beneficial effects on blood pressure. With regard to sensitive, but gadolinium cannot be used in the many antihypertensive drugs, there is insufficient evidence patients with a GFR less than 30 ml/min because of to strongly recommend any particular class of anti– the risk of nephrogenic sclerosing fibrosis. Angiog- hypertensive. Our practice is to follow guidelines for raphy remains the gold standard, but is invasive and the general adult population (e.g., British Hyperten- should be reserved for patients with positive screening sion Society Guidelines), taking into account post- investigations or in whom there is a strong clinical transplant complications and relevant comorbid fac- suspicion despite negative ultrasound Doppler. Local tors (Table 32.5).

Table 32.5 Use of common anti-hypertensive agents in kidney transplant recipients Advantages/indications common in Disadvantages/contraindications Agent/class kidney transplant recipients common in kidney transplant recipients

Angiotensin-converting enzyme Proteinuria Hyperkalemia inhibitors (ACEi)/angiotensin Heart failure with systolic dysfunction Anaemia receptor blockers (ARBs) Post-myocardial infarction Transplant renal artery stenosis Transplant erythrocytosis Calcium channel blockers Chronic stable angina Edema Increased CNI levels (allow reduction in Gum hypertrophy (some agents) dose/cost) Increase CNI levels (if not anticipated) Thiazide diuretics Heart failure with systolic dysfunction Hyperuricemia – gout Hyperkalemia Dyslipidemia Edema Impaired glucose tolerance Hyponatremia Beta-blockers Heart failure with systolic dysfunction∗ Hyperkalemia Chronic stable angina Dyslipidemia Post myocardial infarction Impaired glucose tolerance Alpha-blockers Bladder outflow obstruction Edema Postural hypotension CNI, calcineurin inhibitor (e.g. cyclosporine, tacrolimus) ∗Carvedilol, bisoprolol, metoprolol.

273 Section 5: Kidney

Table 32.6 Risk factors for the development of new-onset tion aimed at increasing exercise and avoiding post- diabetes after transplantation (NODAT) transplant weight gain can be helpful, but immunosup- Pretransplant pression is the major and obviously modifiable issue. It Obesity has been hard to quantify the relative risks of different immunosuppressants on the development of NODAT; Ethnicity (African American or Hispanic) RCTs have used varying definitions of the condition, Ͼ Age ( 60 years) with widely different immunosuppressive regimens Family history of type 2 diabetes and doses of particular immunosuppressants. Hepatitis C virus and CMV disease One strategy to reduce the incidence of NODAT Impaired fasting glucose is to transplant patients without using steroids. Post-transplant However, as discussed previously in the context of management of CNI toxicity and BKV nephropathy, Immunosuppression (in order of diabetic potential) Corticosteroids the clear risk of reducing overall immunosuppression Tacrolimus is acute rejection. Some (but not all) studies of early Cyclosporine (within 3 months) or late steroid withdrawal from Sirolimus CyA-based regimens have reported an unaccept- Post-transplant weight gain able risk of this condition. By contrast, steroid-free CMV: cytomegalovirus. immunosuppression (or early steroid withdrawal) in conjunction with TAC, MMF, and antibody induction (both IL-2 blocking and T-cell depleting) seems safe Pretransplant diabetes in low immunological risk recipients. If NODAT is Even though the survival benefits seen after trans- diagnosed, then in the absence of rejection, it may plantation are greater among patients with diabetes be reasonable to modify immunosuppression. The than those without, both patient and graft survival are strategy most often used is rapid steroid reduction worse in diabetic than non-diabetic renal transplant and conversion from TAC to CyA. recipients (5-year post-transplant patient survival of Regarding targets for glycemic control, standard approximately 70% versus 90–95%). This is a major practice is to aim for the same as in the non-transplant Ͻ issue given that diabetes is the cause of end-stage population (HbA1C 7%) using a standard approach. renal disease in 15% of renal transplant recipients All of the standard dietary, oral hypoglycemic, and and that type 2 diabetes is increasingly present as a insulin treatments can be employed, excepting that as comorbid condition as the average age of patients on for other patients with chronic kidney disease, met- the transplant waiting list rises. The excess mortality formin (or other biguanides) should not be given in diabetic patients is almost entirely as a result of to renal transplant recipients with estimated GFR cardiovascular deaths, although infective causes are (eGFR)less than 30 ml/min. also more common. Management of dyslipidemia New-onset diabetes after transplantation Many renal transplant recipients have dyslipidemia, New onset diabetes after transplantation (NODAT) is often caused or exacerbated by impairment of renal common, with incidence in the first year after trans- function and immunosuppressive medications. Based plant reported in the range 5–20% and US registry data on the fact that such patients are at very high car- recently declaring a prevalence of 41% at 3 years post- diovascular risk, it is generally accepted that mod- transplantation. This is an important matter because, ifiable risk factors such as dyslipidaemia should be despite its short duration, NODAT is associated with treated aggressively, although there is no RCT evi- reduced long-term graft survival (RR graft failure = dence of benefit in hard clinical end points in the 1.63) and increased cardiovascular mortality (RR renal transplant population. Aside from dietary inter- death = 1.87). vention and lifestyle modifications, the drugs most NODAT is caused by insufficient insulin release commonly employed (as in other patient groups) in response to an increase in insulin resistance. Risk for elevated low-density lipoprotein cholesterol levels factors are listed in Table 32.6. Lifestyle modifica- (Ͼ2.6 mmol/l) are the statins, but there is concern

274 Chapter 32: Long-term management and outcomes

18 000 10 Patients 9 16 000 n = 198 046 8 14 000 7 6 12 000 5 10000 4 3 8000 Relative Risk Patients 2 6000 expected 1 0 ry ey a 4000 Skin tate Mouth Ova Kidn Pros 2000 Melanoma Esophogus Leukaemia

Kaposos sarcoma Hodgkin’s disease

Cumulative incidence (per 1Cumulative 00 000) 0 0 24 6810 Non-Hodgkin lymphom Years Type of cancer

Figure 32.3 Incidence of all types of malignant tumors in Figure 32.4 Relative risk of cancer in renal transplant recipients cadaveric renal transplant recipients compared with the expected compared with patients on the waiting list. incidence in the general population. at 3 years, with very high rates (Ͼ50%) reported after regarding the high incidence of clinically significant long-term follow-up of patients in sunny parts of the myopathy when these are used concurrently with CyA, world. Treatment is by local excision. which increases the blood levels of all statins regardless of their pathway of metabolism. PTLD Significant elevation of serum triglycerides (Ͼ5.65 mmol/l) that persists despite lifestyle modifications PTLD is driven by Epstein-Barr virus present in latent and treatment of secondary causes is usually treated form in B lymphocytes. It occurs in 1–2% of renal with ezetimibe. Nicotinic acid can be used as an alter- transplant recipients, most often presenting with lym- native, but fibrates are generally avoided due to risk of phadenopathy, graft infiltration, and central nervous myositis/rhabdomyolysis. system or gut involvement. The usual initial manage- ment is by step-wise reduction of immunosuppres- sion, which requires close monitoring but is effective Malignancies (see also Chapter 4A) in more than 50% of cases, with standard chemother- Renal transplant recipients are at 3–5 fold higher risk apy (often combined with rituximab) used if this does of cancer than the general population (Figure 32.3), not lead to the desired response. See Chapter 4A for and the cancers that they develop are often more further details. aggressive, with poor survival. However, the situation varies greatly for different types of cancer: some com- Management of immunosuppression in the mon cancers (including breast and prostate) do not occur more frequently; non-melanoma skin cancer patient with malignancy and PTLD are particularly common (Figure 32.4). Malignancy arising in a renal transplant recipient Women with renal transplants should be offered shouldgenerallybetreatedinthesamemannerasit and encouraged to accept regular breast and cervi- wouldbeinanyotherpatientwithcomparablerenal calscreening.Allrenaltransplantrecipientswillhave function. However, given that therapeutic (regarding urinalysis performed as part of regular clinic mon- the transplant) immunosuppression may have played itoring: the development of new hematuria should some role in inducing the malignancy, the issue of trigger investigation of the urinary tract because of whether or not the patient’s transplant immunosup- the increased risk of renal cell cancer in the native pressiveregimenshouldbealteredshouldbeconsid- kidneys (often with acquired cystic disease) and of ered. There is a difficult balance to be struck here: from bladder cancer. Screening for skin malignancy is also the point of view of treating the malignancy, complete required. withdrawal of immunosuppression would be best, yet The cumulative risk of non-melanoma skin cancer this would inevitably lead to rejection and loss of the after renal transplantation is 2.3% at 1 year and 7.4% renal transplant, which is something that few patients

275 Section 5: Kidney

wouldwishtoconsiderinanycircumstances.Data with relapsing (recurrent) UTI. Asymptomatic bac- on which to base rational decisions is sparse. Most teriuria is common, but there is no consensus regard- patients and most transplant physicians will attempt ing whether this should be treated, and in one study, to find a compromise where immunosuppression is treatment did not prevent symptomatic UTI. reduced but not withdrawn, with close monitoring of transplant function. In tumors that are clearly caused or exacerbated by immunosuppressive drugs, there is a Other infections very strong rationale for doing this, and as previously Acute surgical issues have resolved by 6 months after discussed, this strategy is well established for the man- transplant. Most patients are well and have returned agement of post-transplant lymphoproliferative disor- to normal activities, and their immunosuppression has der.Forsolidorgantumourswithoutaclearviralaeti- been minimized to that required for long-term main- ology, the rationale is less clear and any changes need tenance. Most infections occurring beyond this point to balance the risks of the tumor versus the risk of graft are typical community-acquired infections and, aside loss and quality of life on dialysis. from UTIs, include upper and lower respiratory tract Aside from simply reducing immunosuppression, infection (viral or bacterial) and gastroenteritis (usu- there are theoretical grounds for thinking that switch- ally viral). However, atypical or opportunistic infec- ing to an mTORi (e.g., sirolimus, as discussed in the tions are more common than in the general population management of CAN) may be helpful in a patient and should be sought if the clinical picture is not obvi- with malignancy because these agents have both anti- ous and/or initial treatment is unsuccessful. These are proliferative and immunosuppressive properties. Data discussed in Chapter 4C. from registries and randomized trials has shown the incidence of de novo post-transplantation malignancy is lower in sirolimus treated patients, but their role Transplant bone disease whenmalignancyhasoccurredislessclear.Datafrom Bone disease in renal transplant patients is com- small trials and case reports suggests that early con- plex: many patients have pre-existing chronic kidney versiontomTORiisbeneficialinnon-melanomaskin disease–mineral bone disease (CKD-MBD; which may cancer, Kaposi’s sarcoma, and renal cell carcinoma, but persist after transplantation), osteoporosis, or previ- the difficulty of switching to a drug that many patients ous parathyroidectomy. Renal transplant patients are find difficult to tolerate because of side effects should at high risk of fracture, but it is not proven that low not be underestimated. bone mineral density (BMD) or a loss of BMD pre- dictsthisasitdoesinthegeneralpopulation;indeed, it is known it does not predict fracture risk in patients Infectious diseases with stage 4–5 CKD. Unfortunately, there are no good data looking at the effect of bone-specific therapy on Urinary tract infection patient-level outcomes (mortality or fracture risk) in Urinary tract infection (UTI) is the most common renal transplant patients. infection in kidney transplant recipients. Most occur In patients with relatively good renal function Ͼ in the first year following transplantation. Analysis (eGFR 30 ml/min), it is reasonable to check BMD of the US Renal Data System database has shown regularly and offer appropriate treatment to those in that late UTI (Ͼ6 months after transplantation) is whom this is low. This might be with vitamin D, associated with poor renal allograft survival and active analogues of vitamin D, and/or a bisphospho- increased mortality. nate depending on the presence of CKD-MBD as indi- Escherichia coli is the most common uropathogen cated by levels of parathyroid hormone, calcium, phos- in renal transplant patients; Enterococcus species, Pseu- phate, alkaline phosphatase, and vitamin D. domonas, coagulase-negative staphylococci, Enter- obacter, and other organisms (group B streptococci Specific side effects of particular and Gardnerella vaginalis) also occur. Antibiotics should be given for 7–14 days, and an anatomical cause immunosuppressants (vesico-ureteral reflux, neurogenic bladder, bladder Aside from the notable general side effects of outflow obstruction) should be excluded in any patient immunosuppression, most importantly including

276 Chapter 32: Long-term management and outcomes susceptibility to some malignancies and some infec- Factors affecting pregnancy outcome tions (as discussed above), long-term transplant Standard advice is that women wishing to conceive recipients frequently suffer side effects specific to should have stable and satisfactory renal function particular immunosuppressants. andwaitatleast1yearaftertransplantationbefore Steroids are especially problematic: most of their attempting to conceive, which takes them beyond the complications arise and become clinically apparent period of greatest risk of acute rejection and viral infec- within the first 6 months, but those that commonly tion. Regarding women with CKD of any sort, the declare themselves clinically later than this include chances of a successful outcome to pregnancy reduce osteoporosis, avascular necrosis of bone, tendon rup- with increasing impairment of renal function. The tures, and cataracts. transplant kidney is not affected by nor is it a con- Aside from nephrotoxicity and diabetes (as dis- traindication to a vaginal delivery. Deliveries are more cussed above), complications of CNIs that often arise likelytobebyCesareansection,butthisisduetomed- or progress after 6 months include hyperuricemia/ ical indications. gout, distal limb pain, neurotoxicity, and (with CyA) coarsening of facial features, gum hypertrophy, and hypertrichosis. Longer term side effects of AZA and Medical management before and MMF include marrow suppression, and (with AZA) during pregnancy alopecia. Aside from general measures appropriate to all preg- nant women (e.g., encouragement to stop smok- Chronic kidney disease ing, folic acid supplementation), the most important requirement is for careful review of the patient’s med- Mostrenaltransplantrecipientshavestage3CKD, ications. Prednisolone (Ͻ15 mg/d), AZA, CyA, and meaning an eGFR of between 30 and 60 ml/min, and TAC are thought to be safe in pregnancy, but MMF most, for reasons discussed above, can expect this to (early pregnancy loss; severe structural abnormalities decline to stage 4 CKD (eGFR 15–30 ml/min) and Ͻ including hydrocephaly, cleft lip and palate, microtia, stage5CKD(eGFR 15 ml/min) over years. They and absence of external auditory canals) and SRL (ter- will therefore require diagnosis and management of atogenic in animal studies; very little clinical experi- the complications of CKD, in particular of renal min- ence) are not. eral and bone disorder and of renal anemia. Discus- sion of these is outside the scope of this chapter, and the reader seeking information should consult a gen- Further reading eral nephrology resource. Augustine JJ, Hricik DE. Steroid sparing in kidney transplantation: changing paradigms, improving outcomes, and remaining questions. Clin J Am Soc Pregnancy Nephrol 2006; 1: 1080–9. El-Amm JM, Haririan A, Crook ED. The effects of blood Women with advanced chronic renal failure and/or pressure and lipid control on kidney allograft outcome. receiving dialysis are rarely able to conceive and pro- Am J Cardiovasc Drugs 2006; 6: 1–7. duce a live birth, hence for many such women a sig- Golshayan D, Pascual M. Minimization of calcineurin nificant reason for wanting a transplant is the desire to inhibitors to improve long-term outcomes in kidney have children. The risks of pregnancy and childbirth to transplantation. Transpl Immunol 2008; 20: 21–8. both mother and child are higher in renal transplant Yarlagadda SG, Klein CL, Jani A. Long-term renal outcomes recipients than in women without medical problems, after delayed graft function. Adv Chronic Kidney Dis but pregnancy is usually successful in women with sta- 2008; 15: 248–56. ble, satisfactory renal transplant function, and consid- Zachariah MS, Tornatore KM, Venuto RC. Kidney erations of pregnancy are a routine part of long-term transplantation and pregnancy. Curr Opin Organ management. Transplant 2009; 14: 386–9.

277 Section 5 Kidney Chapter Pediatric kidney transplantation

33 Khashayar Vakili and Heung Bae Kim

Key points Pediatric kidney transplantation requires a dedi- r cated team of specialists including pediatric nephrol- Kidney transplantation has become the ogists, transplant surgeons, pediatric urologists, trans- treatment of choice for pediatric patients plant coordinators, dialysis personnel, nutritionists, with end-stage renal disease. r pharmacists, social workers, psychologists, infectious Transplantationhasbeenshowntoprovide disease specialists, and others. This multi-disciplinary improved long-term survival compared with approach to the care of these patients is critical dialysis as well as improved growth and in achieving long-term graft and patient survival. development in children. r Although there are some similarities between adult Advances in immunosuppression and pediatric kidney transplantation, there are many medications, surgical technique, and aspects of the medical and surgical care of the pediatric postoperative care continue to improve graft transplant patients that are somewhat unique and are and patient survival rates. r the focus of this chapter. Disease recurrence, chronic allograft Much of the data presented in this chapter are nephropathy, and complications of based on the North American Pediatric Renal Tri- immunosuppression remain significant als and Collaborative Studies (NAPRTCS), which has causes of morbidity and mortality in children collected data since 1987 on about 17 000 children following renal transplantation. with ESRD and about 10 000 children with renal transplants. Pediatric end-stage renal disease (ESRD) requires renal replacement therapy by either dialysis or kidney Indications and contraindications for transplantation. Most patients will transition between these two modalities at least once during their life- renal transplantation times. The decision to institute renal replacement ther- Chronickidneydisease(CKD)isdefinedaskidney apy is based on the presence of complications that damage based on pathological abnormalities or mark- include fluid overload, electrolyte disturbances, symp- ers of dysfunction in urine or blood and is the same tomatic uremia, failure to thrive, metabolic bone dis- as adult staging of CKD. Renal replacement therapy is ease, and psychomotor developmental delay. In the indicated for patients who progress to stage 5 CKD. majority of instances, the initiation of hemodialy- Diseases leading to ESRD in children differ sig- sisorperitonealdialysisprecedesrenaltransplanta- nificantly from those in adults. The primary diag- tion.However,intheabsenceofcontraindications, noses and demographics for pediatric patients under- kidney transplantation is the treatment of choice for going renal transplantation in 2008 are shown in Table children with ESRD, as it provides numerous ben- 33.1. The majority of ESRD cases in infants and young efits when compared with long-term dialysis. These children are a result of congenital or inherited disor- benefits include decreased morbidity and mortality, ders, whereas acquired renal disease is more preva- improvements in quality of life, and improvements in lent in older children and young adults. The etiology growth and development. of ESRD has an important impact on the preparation

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

278 Chapter 33: Pediatric kidney transplantation

Table 33.1 Diagnoses and demographics of pediatric patients of the patient for transplantation, postoperative man- undergoing renal transplantation agement, and overall graft survival. For example, Primary diagnosis % patients with obstructive uropathy may require pre- Aplasia/hypoplasia/dysplasia 15.9 transplant bladder reconstruction to allow for ade- quate graft drainage following transplant. Failure to Obstructive uropathy 15.6 adequately prepare the bladder can result in early graft Focal segmental glomerulosclerosis 11.7 loss. Diagnoses with a significant recurrence rate such Reflux nephropathy 5.2 as focal segmental glomerulosclerosis (FSGS) or atyp- Chronic glomerulonephritis 3.3 ical hemolytic uremic syndrome (HUS) may require Polycystic disease 2.9 special protocols to decrease the risk of disease recur- Medullary cystic disease 2.8 rence. In extreme cases such as type 1 primary hyper- oxaluria, simultaneous liver transplantation may be Hemolytic uremic syndrome 2.6 required to prevent recurrence of the primary disease. Prune belly 2.6 Some contraindications to transplantation include Congenital nephrotic syndrome 2.6 presence of metastatic malignancy, active infection Familial nephritis 2.3 with hepatitis B, severe multi-organ failure, positive Cystinosis 2.0 direct cross-match, or irreversible debilitating brain Pyelo/interstitial nephritis 1.8 injury. In approaching a patient following complete resection for malignancy, the long-term probability Membranoproliferative glomerulonephritis type I 1.7 of tumor-free survival is the main determining factor Idiopathic crescentic glomerulonephritis 1.7 in proceeding with kidney transplantation. For exam- Systemic lupus erythematosus nephritis 1.5 ple, many centers allow for a 1-year recurrence-free Renal infarct 1.4 period following resection of a bilateral Wilms’ tumor Berger’s (IgA) nephritis 1.3 before planning kidney transplantation. Relative con- Henoch-Schonlein¨ nephritis 1.1 traindications include ABO incompatibility, previous positive direct cross-match, active autoimmune dis- Membranoproliferative glomerulonephritis type II 0.8 ease (lupus erythematosus, anti-glomerular basement Wegener’s granulomatosis 0.6 membrane), human immunodeficiency virus (HIV) Wilms’ tumor 0.5 infection, psychomotor or psychiatric illness necessi- Drash syndrome 0.5 tating custodial care, chronic hepatitis C virus (HCV) Oxalosis 0.5 infection, significant history of non-compliance, or Membranous nephropathy 0.4 lack of family support or adequate home supervision. Other systemic immunologic disease 0.3 Sickle cell nephropathy 0.2 Preoperative management and Diabetic glomerulonephritis 0.1 evaluation Other 9.8 Kidney transplantation should be performed after a Unknown 6.2 thoroughwork-upandpreparationofthepatientas Gender well as the family. Allowing adequate time for cor- rection of nutritional deficits and metabolic derange- Male 59.4 ments are vital to minimizing post-transplant compli- Female 40.6 cations. In addition, because the majority of pediatric Race patients receive adult-size kidneys, the size of the child White 59.8 has to be suitable to accept an adult-size kidney. This Black 16.9 generally is of concern in infants weighing less than Hispanic 16.7 6.5 kg or with a length of less than 65 cm. As a result, renal transplantation in infants less than 1 year of age Other 6.6 is not common. Whenever possible, infants should be Based on data from North American Pediatric Renal Trials and Collaborative Studies (NAPRTCS). supported with dialysis until they grow to an appropri- ate size for transplantation.

279 Section 5: Kidney

In the majority of cases, pediatric kidney trans- lower urinary tract anomalies include vesicoureteral plantation is performed once a patient has developed reflux, posterior urethral valves, neurogenic bladder, ESRD and is on maintenance dialysis. However, about Prune belly syndrome, and bladder or cloacal exstro- 25% of children undergo preemptive transplantation, phy. Based on the NAPRTCS report, about 25% of with the highest rate in the 6–12-year-old age group. pediatric kidney transplant recipients have lower uri- Preemptive transplantation is thought to have many nary tract anomalies. The goal of the pre-transplant potential benefits, including avoidance of dialysis and urologic evaluation is to ensure the establishment of improvement in growth and development. a suitable urinary conduit with the aim of minimizing A thorough multi-disciplinary evaluation is nec- thefutureriskofinjurytothekidneyallograft.Pre- essary prior to renal transplantation. The recipient is transplant work-up may include determination of uri- generally evaluated by a nephrologist, transplant sur- nary flow rate, post-void residual volume, cystoscopy, geon, urologist, transplant nurse coordinator, social urodynamics, or voiding cystourethrogram. worker, psychologist, dietician, and pharmacist. A his- Bladder pressures over 40 cmH2Ohavebeen tory and physical examination should be performed, shown to result in allograft damage following kidney with special emphasis on identification of associated transplantation. Pretransplant strategies to decrease anomalies for children with congenital renal anoma- bladder pressure may include the use of anticholin- lies. The size and shape of the recipient’s abdomen ergics and intermittent catheterization, which may and prior abdominal operations are used as a guide to be used in patients with high post-void residuals. select the most suitable position for the kidney allo- These strategies will also need to be continued after graft. From a surgical standpoint, particular attention transplantation. In some cases, the bladder cannot should be paid to the quality of the femoral pulses. be adequately assessed due to oliguria or diversion. A history of femoral venous catheters, abnormali- Reconstruction of the urinary tract in patients with ties on examination, or pelvic malformations should functional kidneys prior to transplantation will allow prompt a radiographic evaluation (ultrasound, com- cycling of the bladder. If the urinary tract is not in puted tomography [CT] or magnetic resonance imag- continuity with the bladder, intermittent catheteriza- ing[MRI])oftheabdominalarterialandvenousvas- tion may be used for bladder cycling. Despite the above culature. Standard laboratory tests include complete strategies, sometimes the bladder will continue to be blood count, electrolytes, urea, creatinine, coagula- inadequate in size and function and the patient will tion parameters, liver function tests, and panel reac- therefore require urinary reconstruction. Patients with tive antibodies. Urine should be analyzed and cultured. inadequate bladder capacity will benefit from augmen- Cardiac and pulmonary evaluation should include tation with small intestine, colon, or stomach. Gastric chest X-ray, electrocardiogram, pulmonary function remnants can result in loss of acid in urine and result in tests, and echocardiogram when indicated. Serologic metabolic alkalosis. In most cases, the bladder recon- tests should be performed, including cytomegalovirus struction is performed prior to transplantation. (CMV); Epstein-Barr virus (EBV); HIV; toxoplasmo- Overall, about 60% of the kidney grafts are from sis; hepatitis B virus (HBV); HCV; measles, mumps, living donors with the parents comprising the major- rubella (MMR); varicella zoster virus (VZV); and her- ity (81%) of the living donors. The remainder of chil- pes simplex virus (HSV). Patients should have up-to- dren receive kidneys from either donation after brain date vaccinations, including HBV,tetanus, hemophilus death (DBD) or donation after cardiac death (DCD) influenza type b, polio, MMR, VCV, pneumococcal donors. DBD donors yield higher quality grafts and vaccine, influenza, and hepatitis A virus prior to trans- therefore constitute the majority of pediatric deceased plantation if at all possible. The use of immunosup- kidney donors. The use of DCD donors for pedi- pression medications after transplantation may blunt atric patients has become more controversial given the the appropriate response to killed vaccines and pre- recent changes in the allocation system, which allow clude the use of live-virus vaccines. Any live-virus for pediatric patients to be preferentially allocated kid- vaccines should be given at least 4 weeks prior to neys from younger deceased donors (under age 35). transplantation. Grafts from living donors tend to have shorter In contrast to the adult population, lower uri- ischemictimesandahigher5-yeargraftsurvival(85%) nary tract anomalies in children with renal failure are compared with deceased donor grafts (80%). One quite common. Common diagnoses associated with obvious disadvantage of living donors is the risk to the

280 Chapter 33: Pediatric kidney transplantation

linemiaaswellasincreasedriskofthrombosisdue to loss of coagulation inhibitors. In order to min- imize the risk of thrombosis after transplantation, native nephrectomies are performed prior to trans- plantation in patients with severe proteinuria such as infants with congenital nephrotic syndrome. However, in most other cases, native nephrectomy can be per- formedatthetimeoftransplant.Ifthekidneygraftisto be placed via a retroperitoneal approach into the iliac fossa, then native nephrectomies can be performed laparoscopically prior to transplantation. In certain congenital malformations, vascular anatomic aberrations can be encountered, such as left- sided vena cava, intraperitoneal aorta, or predominant abdominal azygous system. However, with careful pre- operative vascular mapping (CT or MRI) and plan- ning, these anatomic variations should not necessarily preclude renal transplantation.

Postoperative management Maintenance of adequate circulating volume and per- fusion pressures are important during and follow- Figure 33.1 Intraabdominal kidney transplant with anastomoses to the distal aorta and vena cava following bilateral native ing renal transplantation. Excessive volume overload nephrectomies. in the operating room or in the early postopera- tive period, especially in small recipients, can result donor. Therefore, potential donors undergo a rigorous in pulmonary edema and the requirement for pro- work-up prior to being selected as a donor. Grafts from longed ventilatory support. IV dopamine infusion living donors that will be at high risk of early failure can be used in the early postoperative period to secondary to recipient immune status or recurrent dis- maintain adequate perfusion pressures (mean arterial ease should generally be avoided. pressure Ͼ 60–65 mmHg). In addition to replacement of insensible losses, urine output should be replaced Surgical technique in a 1:1 ratio in the first few days following trans- In the pediatric population, the size of the patient dic- plantation to prevent hypovolemia. Output is gen- tates the position of the renal allograft (discussed fur- erally replaced with 1/2 normal saline, with adjust- ther in Chapter 30). The preferred location for the ments made to electrolyte balance based on urine and placement of the renal allograft is in either the right blood studies. Glucose should be included in the base- or left iliac fossa via a retroperitoneal approach with line insensible fluid rate until the patient is able to vascular anastomoses to the iliac vessels. In children tolerate enteral fluids. Particular attention should be weighing less than 20 kg, an adult-sized allograft usu- paid to the patient requiring significant fluid replace- ally requires an intra-abdominal approach, which pro- ment due to early renal allograft diuresis, as these vides more space for the kidney (Figure 33.1). In this patients can develop significant electrolyte abnormali- case, the distal aorta and vena cava are used for inflow ties. Hyperphosphatemia may be observed until nor- and outflow, respectively, due to the small size of the mal graft function is achieved, requiring treatment iliac vessels. The intra-abdominal approach also allows with phosphate binders. However, once the graft starts for native nephrectomies at the time of transplantation functioning well, hypophosphatemia and hypomagne- when indicated. Indications for native nephrectomy semia may result, necessitating initiation of oral sup- include hypertension, recurrent urinary tract infec- plements. Increase in blood creatinine concentration tions, significant proteinuria, or high urine output. may be secondary to acute tubular necrosis, hypov- Excess proteinuria may result in hypoimmunoglobu- olemia, vascular compromise, ureteral obstruction, or

281 Section 5: Kidney

rejection. Based on the overall clinical scenario, appro- Steroid dose reduction is of particular interest priate diagnostic work-up should be carried out. in the pediatric population since steroids can have Most centers perform a baseline imaging study many profound negative effects, including growth (ultrasound or mercaptoacetyltriglycine [MAG3] retardation, glucose intolerance, hypertension, aseptic scan) within the first 24 hours following transplan- bone necrosis, and impaired wound healing. Currently tation. Further imaging is usually dictated by the the maintenance regimen of prednisone, tacrolimus subsequent clinical course. If vascular compromise is (TAC), and mycophenolate mofetil (MMF) is used suspected postoperatively due to a decrease in urine for the majority (about 30%) of transplants. With the output or increasing creatinine, a Doppler ultrasound introduction of MMF and TAC, the use of azathio- should be used to evaluate blood flow through the prine and cyclosporine has declined significantly over renal artery and vein. A nuclear medicine scan the past 10 years. In addition, the use of prednisone (MAG3) can be used to assess perfusion, excretory has decreased from 95% in 1996 to 61% in 2007 per function, or the presence of urine leak. The radioiso- the NAPRTCS 2008 report. Many centers have devel- tope is concentrated in the urine, and a slow rate of oped strategies to wean the patient off steroids within excretion can be seen in acute tubular necrosis and the first 1–2 years following transplantation. rejection. Visualization of radioisotope outside the urinary tract is suggestive of a urine leak. Postoperative complications The urine reservoir (bladder or a urinary conduit) is decompressed with an indwelling catheter for the Delayed graft function first 5–6 days following surgery to allow for adequate Generally, a well-functioning kidney graft will have healing of the ureteral anastomosis. A surgical drain excellent function within the first 1–3 days, result- placedatthetimeofsurgeryismonitoredforquality ing in normalization of serum creatinine concentra- and quantity of output. If a urine leak is suspected, test- tion. However, in some cases, reaching normal renal ing the creatinine concentration in the drain fluid can graft function takes longer than expected. In most aid in diagnosis. cases, acute tubular necrosis (ATN) is the main cause Hypotension and severe hypertension should be of this delayed function. Based on NAPRTCS data, avoided in the early postoperative period. Low blood delayed graft function can be seen in 5% of living pressure may lead to hypoperfusion in an adult- donor and about 16% of deceased donor transplants. size kidney in a pediatric recipient, which may in Risk factors for ATN in the living donor transplants turn lead to a delay in function. Adequate circulat- include greater than five prior blood transfusions, his- ing volume should be ascertained and vasoconstric- tory of prior transplants, age less than 24 months, tors should generally be avoided due to the increased black race, native nephrectomy, and prior dialysis. Risk risk of renal vasoconstriction. Hypertension can be factors for ATN following deceased donor transplant treated with calcium channel blockers (nifedipine and include more than five blood transfusions, prior trans- amlodipine), hydralazine, beta-blockers, and alpha- plant, cold ischemia time greater than 24 hours, native blockers. Angiotensin-converting enyzme inhibitors nephrectomy, black race, donor age less than 2 years and angiotensin receptor blockers are avoided in the or greater than 50 years, and prior dialysis. Grafts with first several weeks after transplant due to the risk of early ATN have a lower 5-year survival rate (65%) inducing renal insufficiency. compared with grafts without ATN (85%). Immunosuppression Graft thrombosis Based on the NAPRTCS 2008 annual report, Vascular thrombosis of the renal graft is a devastating interleukin-2 antibodies are the most commonly event seen in about 2–3% of pediatric renal transplants used agents for induction therapy for both living and is a major cause of graft failure in the first year fol- donor (51%) and deceased donor (50%) pediatric lowing transplant. Vascular thrombosis usually occurs renal transplants. Antibody induction therapy, defined within the first several days following surgery; how- by administration of antibody at transplant or on day 1 ever, it can also occur after several weeks. The clinician post-transplant, was initiated in about 54% of recipi- should be alerted to the possibility of vascular com- ents in 2007. promise if there is a sudden decrease in renal function

282 Chapter 33: Pediatric kidney transplantation that results in oliguria or anuria. The diagnosis can be complications. Prior urologic surgery, pretransplant established by obtaining a Doppler ultrasound of the obstructive uropathy, or vesicoureteral reflux may graft to evaluate blood flow into the kidney. In addi- increase the risk of urologic complications following tion, an MAG3 radionuclide scan can be used to estab- transplantation. The effect of urologic complications lish the diagnosis. In nearly all cases, vascular throm- on long-term graft survival is not very clear. In the case bosis results in the loss of the graft, requiring removal of ureteral obstruction, prompt relief of obstruction of the graft. The only chance at salvaging the graft isif will likely not have long-term detrimental effects on re-operation and thrombectomy is carried out within the graft. Some studies have shown that vesicoureteral the first few hours following thrombosis. reflux may result in higher rates of graft dysfunction NAPRTCS has demonstrated increased risk of graft and failure. This is likely secondary to recurrent thrombosis in living donor recipients with prior trans- urinary tract infections, which have been correlated plantation. Cold ischemia time longer than 24 hours with increased risk of long-term graft failure. has been associated with increased risk of thrombosis. Recipients with previous bladder augmentation Donors older than 5 years of age and the use of anti- procedures are at higher risk of developing UTIs as body induction therapy are associated with decreased well as metabolic acidosis. The intestinal segment risk of thrombosis. absorbs urinary ammonia and ammonium chloride, which along with secretion of bicarbonate leads to Hemorrhage metabolic acidosis. Due to the high incidence of lower urinary tract Postoperative hemorrhage may result in the develop- dysfunction in the pediatric population with ESRD mentofahematomaaroundthegraft,therebycaus- undergoing renal transplantation, careful preoperative ing compression of the renal parenchyma and renal planning and meticulous operative technique are nec- vasculature. A compartment syndrome may develop, essary to minimize the risk of post-transplant uro- which will place the graft at risk of ischemia, infarc- logic complications. The involvement of a pediatric tion, and eventual failure. This is particularly true in urologist is crucial in obtaining optimum outcome in thecaseofgraftsplacedintheretroperitonealspace. patients with complex urologic problems. Clinically significant bleeding will require return to the operating room for control of bleeding and evacuation of hematoma. Prompt recognition of clinically signif- Graft rejection icant postoperative bleeding is important in order to Graft rejection occurs at least once in about 40–50% of minimize the risk of injury to the graft and the patient. kidney grafts. Over the past 25 years, the probability of rejection within the first year following transplant Urologic complications has decreased significantly, owing to improvements Urologic complications following transplantation in immunosuppressive regimens. Based on NAPRTCS occur in about 5–10% of cases. The most common data, the probability of first rejection within 12 months urologic complications following transplantation after living donor and deceased donor transplants is include ureteral leak, stricture, kinking, extrinsic 8.7% and 17.7%, respectively. In the late 1980s, the compression, and clinically significant vesicoureteral probability of rejection within the first year was more reflux. Obstruction of the urinary flow results in than 50%. Some of the risk factors for rejection fol- decreased urine output and hydronephrosis. Obstruc- lowing deceased donor transplantation include black tion may be secondary to edema at the ureterovesical race, two HLA-DR mismatches compared with no mis- anastomosis, kinking of the ureter, or extrinsic matches, and no induction therapy. In living donor compression from a lymphocele. An ultrasound or transplantation, recipient age, HLA-DR mismatches, MAG3 scan can be used for definitive diagnosis. and ATN are risk factors for rejection. Depending on the cause of obstruction, insertion Definitive diagnosis of graft rejection requires a of a ureteral stent or percutaneous drainage of the graft biopsy for histologic and immunohistochemical lymphocele may resolve the obstruction. In rare evaluation in addition to other appropriate investiga- instances, revision of the ureteral anastomosis may be tions to exclude other causes of rise in serum crea- necessary. Pre-existing bladder or ureteral pathology tinine. Biopsy is generally performed percutaneously may increase the chance of post-transplant urologic and under ultrasound guidance.

283 Section 5: Kidney

Chronic allograft nephropathy and disease leads to graft failure in 6.8% of cases, which makes it one of the top four causes of graft failure. chronic rejection FSGS is one of the most common causes of ESRD in Chronic allograft nephropathy (CAN) is the most children. FSGS recurrence in the allograft leads to graft common cause of eventual graft failure in the pedi- loss in 50% of patients. Recurrence may occur imme- atric transplant population, comprising about 35% of diately after transplantation, resulting in severe pro- all causes of graft failure. It is clinically manifested by teinuria, ATN, and small vessel thrombosis. There is a gradual decline in renal function over a course of no established treatment or prevention regimen for months to years. CAN has replaced what was previ- FSGS recurrence; however, plasmapheresis and ritux- ously referred to as chronic rejection. The exact patho- imab have been used with some success. physiology of CAN has not been elucidated; however, The most common cause of typical HUS in chil- immunological as well as non-immunological mecha- dren is enteropathic bacteria, and this does not usu- nisms have been proposed. Some studies have demon- ally result in ESRD or recurrence in the kidney strated a correlation between acute rejection and graft. However, atypical HUS or “non–Shiga toxin- eventual development of CAN. Non-immunological associated” HUS can lead to ESRD as well as recur- risk factors such as hyperfiltration secondary to rela- rence after transplantation. Atypical HUS may occur tively low nephron mass, large recipient size, African- in the setting of deficient von Willebrand factor– American race, older or female donors, and long cleaving protease (ADAMTS-13), metabolic disorders, ischemic times have also been suggested. or dysfunction in complement regulation. Comple- Unfortunately, there is no treatment for CAN, and ment dysfunction may result from factor H, I, B, or once it is diagnosed, renal function usually contin- C3 mutations or the presence of anti-factor H anti- ues to decline. Medical therapy should be focused bodies. Due to a high rate of recurrence and graft on reducing associated problems including hyperten- failure, factor H and I mutations are considered by sion and proteinuria. Because there appears to be an some to be contraindications to renal transplantation. association between acute rejection episodes and the However, treatment with plasmapheresis and fresh- development of CAN, aggressive and adequate treat- frozen plasma has produced some positive results. ment of acute rejection is crucial. Prevention of CAN Because factor H is produced in the liver, some advo- is especially important in the pediatric population as cate combined liver–kidney transplantation in this poor renal function has a significant effect on growth. setting. Recently, avoidance of calcineurin inhibitors is being Primary hyperoxaluria type 1 is an autosomal- evaluated as a potential strategy to decrease the inci- recessive disease caused by deficiency in peroxisomal dence of CAN following renal transplantation. alanine-glyoxylate aminotransferase (AGT), which leads to an overproduction of oxalate. High levels Disease recurrence after renal of oxalate lead to urolithiasis, nephrocalcinosis, and eventuallychronickidneydisease.Becausetheenzyme transplantation AGT is produced in the liver, liver transplantation The original disease process leading to renal failure is necessary to restore normal enzyme function. The may recur in the transplanted kidney. Some disease chance of disease recurrence in the transplanted kid- processes and their recurrence rates include FSGS, ney without a liver transplant is 90–100%; therefore, 14–50%; atypical HUS, 20–80%; typical HUS, 0–1%; most patients undergo a combined liver and kidney membranoproliferative glomerulosclerosis type 1, 30– transplant. Dialysis is used in the pretransplant setting 77%; membranoproliferative glomerulosclerosis type to maintain normal serum oxalate levels and is con- 2, 66–100%; systemic lupus erythematosus, 0–30%; tinued in the post-transplant setting to prevent early immunoglobulin A (IgA) nephritis, 35–60%; Henoch- oxalatedepositionanddamagetothegraft.Oxalateis Schonlein¨ nephritis, 31–100%; and primary hyperox- deposited and stored in a wide variety of tissues, and aluria type 1, 90–100%. therefore it is slowly released into the blood following The presentation of disease recurrence can vary transplantation. Native nephrectomies are required widely, ranging from subclinical elements of the orig- since the kidneys are a major site of oxalate deposition. inaldiseasetofulldiseaserecurrenceandeventual Autoimmune diseases such as IgA nephropathy, renal failure. Based on NAPRTCS 2008 data, recurrent Henoch-Schonlein¨ purpura, lupus nephritis, and

284 Chapter 33: Pediatric kidney transplantation anti-neutrophil cytoplasmic antibodies (ANCA)- disorder. Risk factors associated with mortality associated vasculitis may recur after transplantation include young recipient age, ATN within 30 days but do not lead to graft loss in the majority of cases. post-transplant, and specific underlying renal diseases including congenital nephrotic syndrome, oxalosis, and Drash syndrome. Patient and graft survival Based on the NAPRTCS 2008 report, the 1- and Further reading 5-year graft survival rates for pediatric living donor Cochat P, Fargue S, Mestrallet G, et al. Disease recurrence kidney transplantation are about 95% and 85%, respec- in paediatric renal transplantation. Pediatr Nephrol 2009; tively. Graft survival rates following deceased donor 24: 2097–108. kidney transplantation are 93% and 75%, respectively. Fine RN, Martz K, Stablein D. What have 20 years of data Some of the leading causes of graft failure in decreas- from the North American Pediatric Renal Transplant ing order of frequency are chronic rejection (or CAN), Cooperative Study taught us about growth following acute rejection, vascular thrombosis, death with func- renal transplantation in infants, children, and adolescents with end-stage renal disease? Pediatr tioning graft, recurrence of original kidney disease, Nephrol 2010; 25: 739–46. and patient discontinuation of medication. Harmon WE. Pediatric kidney transplantation. In Avner Patient survival rates at 1 and 5 years following ED, Harmon WE, Niaudet P, Yoshikawa N (eds). transplantation are 98% and 95%, respectively. Infants Pediatric Nephrology, 6th ed. Berlin: Springer, 2009. younger than 24 months of age receiving a deceased Humar A, Matas AJ. Kidney transplantation. In Humar A, donor organ have had lower survival rates in the past, Matas AJ, Payne WD (eds). Atlas of Organ but this has improved over the past 20 years. The Transplantation. Berlin: Springer, 2009. leading causes of death (percent of all causes) are North American Pediatric Renal Trials and Collaborative cardiopulmonary (15.4%), bacterial infection (12.6%), Studies (NAPRTCS) Annual Report (2008). Available malignancy (10.6%), viral infection (8.1%), and at: https://web.emmes.com/study/ped/annlrept/ non-specified infection (7.9%). The overall rate of annlrept.html (accessed March 1, 2010). malignancy is about 2.4% and about 80% of malignan- Sheldon C, Shumyle A. Urological issues in pediatric renal cies are related to post-transplant lymphoproliferative transplantation. Curr Opin Urol 2008; 18: 413–8.

285 Section 6 Other abdominal organs Chapter Pancreatic transplantation

34 Dixon B. Kaufman

Key points tamic acid decarboxylase (GAD 65), a protein tyro- r The results for all types of pancreas sine phosphatase-like molecule (IA-2), and insulin transplantation have improved over the past are seen, but not all patients with auto-antibodies go decade, in large part due to advances in on to develop diabetes, and auto-antibodies some- immunosuppression. times disappear many years after the onset of dia- r A successful pancreas transplant produces a betes. Increased susceptibility has been linked to par- normoglycemic and insulin-independent ticular human leukocyte antigen (HLA) combinations state virtually immediately after and immunoregulatory gene polymorphisms, as well revascularization. as certain insulin gene alleles. Exogenous insulin saves patients from rapid r The benefits of adding a pancreas transplant demise, but even the combination of newer formu- to ameliorate diabetes are profound – it saves lations of insulin and sophisticated administration lives. regimens does not prevent DM from being the leading r As with other organ transplants, episodes of cause of blindness and renal failure in adults and acute rejection appear to predispose to late the seventh leading cause of death in the United allograft loss. States. Death is sometimes due to massive, acute r Surgical complications are more common hyperglycemia and ketoacidosis, but is more often after pancreas transplantation compared due to secondary diseases (especially cerebral and with kidney transplantation and occur in coronary vasculopathy) arising from less severe 5–10% of cases. These are usually seen within hyperglycemia occurring over time. The develop- 6 months and are an important etiology of ment of secondary comorbidities is tightly linked to pancreas graft loss. chronic glucose control, as measured by the percent- age of non-enzymatically glycosylated hemoglobin (HbA1c). HbA1c levels in diabetics may be as high Rationale of pancreas transplantation as 20%, whereas normal values range from 4–5.9%. Any decrease in HbA1c is considered beneficial, for patients with diabetes mellitus but efforts to achieve normal HbA1C by increasing There are nearly 3 million people in the United doses of insulin are limited by the incidence of States with type 1 diabetes mellitus (DM1) and 35 000 hypoglycemia. new cases diagnosed each year that result from Iatrogenic hypoglycemia causes 2–4% of deaths in autoimmune destruction of the insulin-producing people with DM1, with less severe episodes manifest- cells in the pancreas. CD4+ and CD8+ Tcellsand ing as behavior change, cognitive impairment, seizure, macrophages infiltrate the islets, but clinical symp- and coma as the glucose-dependent brain cells decline toms may not appear until as much as 80% of the beta and then fail. The risk of death is highest in patients cells are destroyed. Antibodies directed against glu- with asymptomatic hypoglycemia, in whom the early

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

286 Chapter 34: Pancreatic transplantation signs of hypoglycemia (sweating, hunger, anxiety, sion are less morbid than the current state of ill palpitations, and tremor) are absent because of auto- health. nomicfailure,eitherasaresultofdiabetes-associated autonomic neuropathy or a blunted epinephrine response following repeated hypoglycemic episodes. Indications and contraindications to Glucagon is the primary insulin counter-regulatory hormone, but the injected insulin cannot respond to pancreas transplantation hypoglycemia and continues to suppress the release Approximately 1300 pancreas transplants are per- of glucagon. Because the autonomic pathway nor- formed annually in the United States. Two thirds mally serves as a back-up glucose-raising mechanism, involve a simultaneous pancreas and kidney transplant these patients lack the trigger both to take in exter- (SPK) for patients with diabetes and chronic or end- nal sources of glucose and to mobilize internal glucose stage renal failure. There are approximately 2200 can- stores. didates registered on the United Network for Organ The only treatments that have been demonstrated Sharing (UNOS) waiting list. Ninety percent of SPK to normalize HbA1c levels, influence the progression transplants are in recipients with DM1, and only about of secondary complications, and avoid hypoglycemia 10% (Ͻ90 cases annually) are select recipients with involve beta-cell replacement therapy with pancreas or DM type 2 (DM2). The benefits of adding a pancreas islet transplantation. Pancreas transplantation is supe- transplant to ameliorate DM are profound: it saves rior to islet transplantation with regard to the effi- lives. ciency and durability of achieving glycemic control Approximately 1500 candidates are registered on and its beneficial effects on diabetic secondary com- the UNOS waiting list for solitary pancreas trans- plications. A successful pancreas transplant produces plants. Solitary pancreas transplant falls into two cat- a normoglycemic and insulin-independent state virtu- egories. The first group are patients with DM who allyimmediatelyafterrevascularization.Itreversesthe have received a previous kidney transplant from diabetic changes in the native kidneys of patients with either a living or cadaveric donor. The pancreas-after- very early diabetic nephropathy, prevents recurrent kidney (PAK) group accounts for approximately 20% diabetic nephropathy in patients undergoing a simul- of patients receiving pancreas transplants. The main taneous pancreas–kidney transplant, reverses periph- consideration is that of surgical risk, since the risk of eral sensory neuropathy, stabilizes advanced diabetic immunosuppression has already been assumed. Most retinopathy, and significantly improves the quality of PAK recipients have received a prior living donor life. kidney allograft. The second group are non-uremic, However, there are important considerations of non-kidney transplant patients with DM1. In this sit- pancreas transplantation that currently precludes it as uation, one assesses the risk of immunosuppression therapy for all patients with DM1. First, it is unreal- to be less than the current clinical condition with isticthatallpatientswithDMcouldreceiveapan- conventional exogenous insulin administration. These creas transplant. There are too many patients with patients with DM have extremely labile disease, such DM1 and too few organs for transplantation. Second, that there is difficulty with day-to-day living, associ- pancreas transplantation involves significant surgery. ated with frequent emergency room visits and hos- Third, lifelong immunosuppression is required to pre- pitalizations for hypoglycemia or diabetic ketoaci- vent graft rejection. Therefore, the indications for dosis. Other patients may have significant difficulty pancreas transplantation are very specific and nar- with hypoglycemic unawareness that results in uncon- row. There are three circumstances where consid- sciousness without warning. This can be a devastat- eration for pancreas transplantation is reasonable ing condition for these select patients, affecting their for patients with DM1: (1) select patients who are employment and their ability to drive, with concern also excellent candidates for kidney transplantation; about suffering lethal hypoglycemia while asleep. The (2) patients with a well-functioning kidney trans- indications for a pancreas transplant alone (PTA) are plant receiving immunosuppression; (3) select patients essentially identical to those patients being considered who are extremely brittle or have significant fre- for an islet transplant. However, in the former situa- quency and severity of hypoglycemic unawareness tion, there are fewer contraindications with respect to such that the risks of surgery and immunosuppres- body mass index and insulin requirements.

287 Section 6: Other abdominal organs

Evaluation of candidates for clopramide, domperidone, or erythromycin may be useful. pancreas transplantation Mental or emotional illnesses including neuroses There are significant pre-existing comorbidities of and depression are common. Diagnosis and appropri- pancreas transplant candidates with advanced renal ate treatment of these illnesses is an important pre- disease, and it should be assumed that coincident transplant consideration, with important implications extra-renal disease is present. Advanced vascular dis- for ensuring a high degree of medical compliance. ease (in particular, coronary artery disease) is the most important comorbidity to consider in patients with DM1 with diabetic nephropathy. It has been esti- Pancreas organ allocation mated that DM1 uremic patients carry a near 50-fold UNOS is now in the process of redesigning the pan- greater risk of cardiovascular events then the gen- creas allocation for pancreas transplantation to bet- eral population. Because of the neuropathy associated ter address the needs of patients with DM with and with diabetes, patients are often asymptomatic because without concurrent renal failure since currently, wait- ischemia-induced angina is not perceived. Virtually ing time for SPK transplant varies widely across the all DM1 uremic patients should undergo coronary United States and there is a lack of specific listing crite- angiography because non-invasive tests are relatively ria. Under the new system, if a uremic DM1 candidate insensitive; however, care should be given to minimize on the list for an SPK transplant is allocated a pancreas contrast agents. This patient group also experiences an from a local deceased donor, then the kidney would increased rate of stroke and transient ischemic attacks. also be offered from the same deceased donor. SPK and Deathsrelatedtocerebralvasculardiseaseareapproxi- PTA candidates would also be combined onto a single mately twice as common and occur at a younger age in list. Allocating the cadaveric pancreas (with kidney for patients with DM and end-stage renal disease. Periph- SPK transplant) prior to procurement of the organs has eral vascular disease is significant, and DM1 uremic distinct advantages: (1) the transplant center perform- patients are at risk for lower limb amputation sec- ing the pancreas transplant could also procure the pan- ondary to ulcers associated with advanced somatosen- creas; (2) the patient can be admitted to the hospital for sory neuropathy. re-evaluation simultaneously, rather than sequential to Diabetic retinopathy is nearly ubiquitous, and sig- the procurement of the organ; (3) cold ischemia time nificant vision loss may have occurred. A patient with will be shorter. Pancreas allografts do not tolerate cold significant vision loss must have adequate support sys- ischemia as well as kidney and require revasculariza- tems to ensure they are able to manage the post- tion within 24 hours from procurement. transplant regime. Autonomicneuropathyisprevalent,commonly Transplant surgery and underestimated, and may manifest as gastropathy, cystopathy, and orthostatic hypotension. Neurogenic surgical complications bladder dysfunction is an important consideration in patients receiving a kidney transplant either simul- Deceased donor pancreas selection taneously or prior to subsequent pancreas transplan- Identification of suitable cadaveric donors for pancreas tation. Inability to sense bladder fullness and empty transplantation is one of the most important determi- the bladder predisposes to urine reflux and high post- nants of outcome. void residuals. This may adversely affect renal allo- Severalanatomicalandphysiologicalfactorshave graft function and increase the incidence of blad- been identified that affect the results of pancreas der infections and pyelonephritis. The combination of transplantation. In general, the criteria that deter- orthostatic hypotension and recumbent hypertension mine an appropriate donor for pancreas transplanta- results from dysregulation of vascular tone. There- tion are more stringent than for kidney or liver donors. fore, careful re-assessment of post-transplant anti- Important considerations include weight, age, hyper- hypertensive medication requirement is important. glycemia, hyperamylasemia, adiposity, and vascular Patients with severe gastroparesis may have difficulty anomalies. tolerating oral immunosuppressive medications and Obese donors weighing more than 100 kg are fre- prophylactic agents. Motility agents such as meto- quently not found to be suitable pancreas donors

288 Chapter 34: Pancreatic transplantation

Table 34.1 Contraindications of pancreas procurement for transplantation 1. History of type 1 diabetes mellitus 2. History of type 2 diabetes mellitus 3. History of previous pancreatic surgery 4. Intra-abdominal trauma to the pancreas 5. Donor age < 10 years and Ͼ 55 years (relative contraindication) 6. Donor weight < 30 kg and Ͼ 100 kg (taken in consideration with height) 7. Intraoperative assessment A. Vascular supply B. Severe edema, significant adipose infiltration, significant fibrosis or mass C. Pancreatic hemorrhage or trauma becauseofDM2orahighdegreeofadiposeinfil- tration of the pancreas. Importantly, pancreata from relatively older donors (age 55–65 years) and obese Figure 34.1 Pancreaticoduodenal allograft with exocrine bladder organ donors are associated with very successful islet drainage and systemic venous drainage. Illustration by Simon Kimm. isolation recovery required for an islet transplant. Therefore, all deceased organ donors should be con- curement teams will be able to successfully separate the sidered for procurement of pancreas and islets for liverandthepancreaseitherinsituorontheback- transplantation. bench without sacrificing quality of either organ for Hyperglycemia and hyperamylasemia are very fre- transplantation. quently observed in deceased organ donors. Hyper- The use of donation after cardiac death donors for glycemia is not a contraindication to pancreas pro- pancreas transplantation has been reported, although curement for patients who are known not to have DM1 with a higher rejection rate at the time of procurement. or 2. Deceased donor assessment may also include If the pancreas is deemed suitable, there is the added HbA1c levels. Hyperamylasemia is concerning, but consideration of the effect of delayed kidney graft func- reports have indicated that it has no meaningful influ- tion in a uremic SPK candidate. ence on pancreas graft function post-transplant. Pan- The use of living related and unrelated pancreas creatitis or pancreatic injury in the case of a donor donors has also been described. A distal pancreate- with trauma should be ruled out at the time of ctomy is performed for a segmental pancreas trans- procurement. plant. Anecdotal cases of combined live donor par- Perhaps the most important determinant of the tial pancreatectomy and nephrectomy have also been suitability of the pancreas for transplantation is by reported. These procedures are not widely performed direct examination of the organ during surgical pro- and are confined to one or two pancreas transplant curement. The degree of fibrosis, adipose tissue, and programs. specific vascular anomalies can be accurately assessed. Pancreata with heavy infiltration of adipose tissue are believed to be relatively intolerant of cold preservation Pancreas transplant surgery andhavethepotentialforahighdegreeofsaponifica- The surgical techniques for pancreas transplantation tion due to reperfusion pancreatitis that follows revas- are diverse, and there is no standard methodology used cularization. These organs may be more suitable for by all programs. islet isolation. The principles include providing adequate arte- The important vascular anomaly that must be eval- rial blood flow to the pancreas and duodenal seg- uated during procurement is the occurrence of a ment, adequate venous outflow of the pancreas via replaced or accessory right hepatic artery originating the portal vein, and management of the pancre- from the superior mesenteric artery. Experienced pro- atic exocrine secretions. The native pancreas is not

289 Section 6: Other abdominal organs

Figure 34.2 Pancreaticoduodenal allograft with exocrine enteric Figure 34.3 Pancreaticoduodenal allograft with exocrine enteric drainage and venous systemic drainage. Illustration by Simon Kimm. drainage and portal venous drainage. Illustration by Simon Kimm.

removed. Pancreas graft arterial revascularization is or anastomosis to the small intestine. Enteric drainage typically accomplished utilizing the recipient right ofthepancreasallograftisphysiologicalwithrespect common or external iliac artery. The Y-graft of the pan- to the delivery of pancreatic enzymes and bicarbon- creas is anastomosed end to side. There are two choices ate into the intestines for reabsorption and can be forvenousrevascularization:systemicandportal.Sys- constructed with or without a Roux-en-Y. The enteric temic venous revascularization commonly involves the anastomosis can be made side to side or end to side right common iliac vein or right external iliac vein. with the duodenal segment of the pancreas. The risk If portal venous drainage is utilized, it is necessary of intra-abdominal abscesses is extremely low, and the to dissect out the superior mesenteric vein (SMV) at avoidance of the bladder-drained pancreas has signif- the root of the mesentery. The pancreas portal vein icant implications with respect to the potential com- is anastomosed end to side to a branch of the SMV. plications that include bladder infection and the fre- This may influence the methodology of arterial revas- quent requirement for enteric conversion. Currently, cularization using a long Y-graft placed through a win- more than 80% of pancreas transplants are performed dow in the mesentery to reach the right common iliac with enteric drainage and the remainder with bladder artery. Portal venous drainage of the pancreas is more drainage. physiological with respect to immediate delivery of insulin to the recipient liver. This results in dimin- ished circulating insulin levels relative to that in sys- Complications of pancreas transplantation temic venous-drained pancreas grafts. There has not Surgical complications are more common after pan- been documented any clinically relevant difference in creas transplantation compared with kidney trans- glycemic control. plantation. Non-immunological complications of pan- Handling the exocrine drainage of the pancreas is creas transplantation account for graft losses in 5–10% the most challenging aspect of the transplant proce- of cases. These occur commonly within 6 months of dure. Pancreatic exocrine drainage may be handled via transplant and are an important etiology of pancreas anastomosis of the duodenal segment to the bladder graft loss in SPK transplantation.

290 Chapter 34: Pancreatic transplantation

Thrombosis tive intervention are essential if hemodynamic insta- Vascular thrombosis is a very early complication typ- bility develops. The hematoma should be removed, ically occurring within 48 hours. This is generally due exploration performed to identify any source of bleed- to venous thrombosis of the pancreas portal vein. The ing, and the viability of the pancreas (and kidney) etiology is not entirely defined but is believed to be confirmed. Gastrointestinal bleeding may occur in associated with reperfusion pancreatitis and the rela- enteric-drained pancreas transplant from a combina- tively low-flow state of the pancreas graft. To minimize tion of peri-operative anticoagulation and bleeding graft thrombosis, prudent selection of donor pancreas fromthesuturelineoftheduodeno-entericanasto- grafts, short cold ischemia times, and meticulous sur- mosis. It may be recognized by a fall in hemoglobin, gical technique are necessary. Approximately 5% of melena, or positive fecal occult blood. This is usu- pancreas grafts will need to be removed because of ally self-limiting; therefore, conservative management portal venous thrombosis. Arterial thrombosis is less is usually sufficient. Occasionally, interventional radi- common and is usually associated with anastomosis to ological embolization of a bleeding vessel or re- atherosclerotic vessels. operative exploration is required. Acute venous thrombosis results in a sudden rise in serum glucose and is occasionally associated with Transplant pancreatitis pain directly over the pancreatic graft and occasion- Transplant pancreatitis occurs to some degree in all ally ipsilateral lower extremity swelling from extension patients. A temporary elevation in serum amylase and of the thrombus into the common iliac vein. Confir- lipase levels usually occurs for 48–96 hours after trans- matory non-invasive perfusion imaging of the graft is plant. These episodes are transient and mild without performed using technetium-99m hexamethyl propy- significant clinical consequence. lene amine oxime. Ultrasonography can also be used to determine vascular flow in the allograft. Manage- Complications of the bladder-drained ment requires urgent operative intervention, either thrombectomy with vascular revision or graft excision. pancreas transplant The findings at surgery are an ischemic, dusky, non- Bladder-drained pancreas transplantation is associ- viable pancreas and duodenal segment with fresh clot ated with many mild to moderately severe complica- in the graft portal vein. Salvage of the thrombosed tions that arise because of the unusual physiology of graft is not to be expected but has occasionally been pancreatic exocrine secretions draining into the blad- described. der. The pancreas transplant will eliminate approxi- To reduce the incidence of pancreatic graft throm- mately 500 ml of bicarbonate-rich fluid with pancre- bosis, anti-coagulation is routinely used. Most cen- atic enzymes into the bladder each day. Change in pH ters employ a combination approach involving hep- of the bladder accounts, in part, for a greater increase arin and an anti-platelet agent early postoperatively, in urinary tract infections. although this increases the risk of hemorrhage. The Sterile cystitis, urethritis, and balanitis may occur management of mild postoperative bleeding is more after bladder-drained pancreas transplantation. This acceptable than the irreversible consequence of allo- is due to the effect of the pancreatic enzymes on the graft thrombosis. urinary tract mucosa. This is more commonly expe- rienced in male recipients. Urethritis can progress to urethral perforation and perineal pain. Treatment Hemorrhage options include conservative treatment or operative Bleeding from the vascular anastomotic site or cut enteric conversion. surfaces of the pancreatic graft will result in an Careful monitoring of serum electrolytes and acid– intra-abdominal hematoma. Clinical suspicion, the base balance is necessary because metabolic acidosis physical examination, serial blood counts, and atten- routinely develops as a consequence of bladder bicar- tion to abdominal drain output suggest postopera- bonate excretion and often requires oral bicarbonate tive hemorrhage. Discontinuation of anti-coagulants/ supplementation. Patients are also at risk of dehydra- anti-platelet agents and correction of coagulation tion, orthostatic hypotension, and electrolyte distur- abnormalities using conventional therapies is usually bance due to large volume losses that may require fluid effective. Aggressive resuscitative efforts and opera- replacement.

291 Section 6: Other abdominal organs

Reflux pancreatitis can result in acute inflamma- prophylaxis, including anti-fungal agents; minimizing tion of the pancreas allograft, mimicking acute rejec- anti-rejection immunotherapy and reduction in tion with symptoms of pain with hyperamylasemia. It acute rejection. There is no convincing evidence is believed to be secondary to reflux of urine through that a Roux-en-Y intestinal reconstruction decreases the ampulla into the pancreatic ducts and may occur incidence. even in the presence of mild prostatic hypertrophy. This frequently occurs in patients with neurogenic bladder dysfunction and necessitates investigation of Immunological aspects of pancreas the cause by pressure flow studies and cystourethrog- transplantation raphy. Recurrent graft pancreatitis requires enteric conversion. Immunosuppression for pancreas Urine leak from breakdown of the duodenal seg- ment is the most serious postoperative complication transplantation of the bladder-drained pancreas, and a high index of The outcome of pancreas transplantation with respect suspicion is required. It is usually encountered within to graft survival and rejection rates is dependent on the the first 2–3 months post-transplant but can occur choice of immunosuppression agents used. The risk many years later. Typical presentation is abdominal of pancreas allograft rejection is much greater than pain with elevated serum amylase, which can mimic that observed with kidney transplantation. The precise reflux pancreatitis or acute rejection. Supporting imag- reasons are not well defined but likely involve greater ing studies with CT and cystography confirm the immunogenicity of the pancreaticoduodenal graft. diagnosis. Operative intervention is usually required The majority of pancreas transplant programs use when the degree of leakage determines whether direct lymphocyte-depleting induction therapy combined repair, enteric conversion or graft pancreatectomy is with tacrolimus, mycophenolate mofetil (MMF), or indicated. sirolimus and prednisolone. Induction therapy is used with greater frequency in pancreas transplant recipi- Complications of the enteric-drained ents than for any other solid-organ recipients. One rea- son is the relatively higher risk of rejection observed pancreas transplant forpancreastransplantrecipientsascomparedwith The most serious complication of the enteric- other solid organ transplants. This combination has drained pancreas transplant is that of a leak and significantly improved graft survival rates. There intra-abdominal abscess. Patients present with are steroid avoidance protocols described for pan- fever, abdominal discomfort and leukocytosis, creas transplantation and reports of successful steroid commonly 1–6 months post-transplant. Computed withdrawal. tomography imaging and percutaneous drainage of intra-abdominal fluid for Gram stain and culture is essential. Bacteria and often fungi are commonly Pancreas allograft rejection found, and broad-spectrum antimicrobial therapy An understanding of the kinetics of the tissue injury is essential. Surgical exploration and repair of the during acute rejection is essential to making a timely enteric leak is necessary. A decision must be made diagnosis. Destruction of beta cells occurs relatively regarding whether the infection can be eradicated late following initial injury of the acinar tissue. There- without removing the pancreas allograft. Incomplete fore, the diagnosis of pancreas graft rejection by hyper- eradication of the infection will result in progression glycemia is a late and often irreversible situation. to sepsis and multi-organ failure. Peri-pancreatic Detection of changes in acinar cell function is the basis infections can result in development of a mycotic for early suspicion of pancreas graft rejection. The graft aneurysm at the arterial anastomosis, which may rup- is usually inflamed and patients may experience pain ture, and requires allograft pancreatectomy. A number and discomfort around the graft due to peritoneal irri- of factors have contributed to the reduction in inci- tation. This, coupled with elevation in the serum amy- dence of intra-abdominal abscess: greater recognition lase or lipase and, if bladder-drained, reduction in uri- of the criteria for procuring suitable cadaveric pan- nary amylase, may be the initial presentation of acute creas allografts; improved peri-operative antibiotic rejection.

292 Chapter 34: Pancreatic transplantation

The gold standard for confirming the diagnosis respectively. Kidney graft functional survival rates for of pancreas graft rejection is pancreas graft biopsy. the same post-transplant time periods were 91.8%, The biopsy may be performed by several methods, 84.4%, and 78.7%, respectively. Single-center reports including percutaneous, transcystoscopic in a bladder- from UNOS show wide variability of kidney and pan- drained pancreas, or open surgical biopsy. The useful- creas graft survival rates. ness of pancreas graft biopsy to confirm the clinical According to the IPTR, there was no clinically suspicion of rejection is so important that the surgical significant difference in pancreas graft outcome in procedure of pancreas transplantation should include bladder-drained pancreases versus the enteric-drained consideration of the intra-abdominal location of the pancreas. There is also no clinically significant differ- pancreas to make it accessible for percutaneous biopsy. ence in outcome in systemic venous drainage versus ThisisespeciallyimportantinPTAandPAKtrans- portal drainage. The immunologic risk for graft loss for plant procedures. technically successful cases of SPK transplantation has In SPK transplant, it is the kidney allograft that is decreased over time. The current rate of immunolog- the best indicator of a rejection episode and will man- ical loss is only 2% at 1 year. Relative risk factors for ifest as a rise in serum creatinine. This will prompt pancreas graft loss in SPK recipients have been deter- ultrasound and biopsy of the kidney allograft. Con- mined and include increasing recipient age and pro- current pancreas graft rejection would be treated by longed preservation time (Ͼ24 hours), and positive the renal anti-rejection therapy. It is extremely uncom- effects were shown for the use of MMF. mon for isolated pancreas allograft rejection to occur in SPK transplantation (1–2%), and the diagnosis requires both kidney and pancreas transplant biop- Effect of pancreas transplantation on sies. Treatment is guided by severity and may involve pulsed steroids or anti-lymphocyte immunotherapy secondary complications of diabetes with successful treatment in excess of 90%, if diag- Recipients of a successful pancreas transplant main- nosed promptly. The incidence of pancreas rejection tain normal plasma glucose levels without the need for has been reduced from approximately 80% to less exogenous insulin therapy. This results in normaliza- than 20% in the modern era of transplant immuno- tion of HbA1c levels and a beneficial effect on many suppression. secondary complications of diabetes, including blood pressure, diabetic neuropathy, autonomic neuropathy- Results of pancreas transplantation associated sudden death, and diabetic nephropathy. The durability of the transplanted endocrine pancreas The results of pancreas transplantation are typically has been established with the demonstration that nor- described in terms of both patient and pancreas graft malization of HbA1c is maintained for as long as the survival.Pancreasgraftlossisdefinedaspatientdeath allograft functions. The potential lifespan of the trans- with a functioning graft or loss of insulin indepen- planted pancreas is not precisely known since there dence irrespective of whether the pancreas allograft is are surviving pancreas allografts greater than 20 years in place or removed. The most valuable and complete post-transplant still functioning. information on the results of pancreas transplantation The quality of life of pancreas transplant recipi- comes from the UNOS Registry and the International ents has been studied widely. The functioning pan- Pancreas Transplant Registry (IPTR). creas allograft leads to even better quality of life when compared with that of recipients of a kidney trans- Outcomes of patient and graft survival plant alone. Virtually all patients of a successful pan- The results of SPK transplantation in terms of patient creastransplantreportthatmanagingtheirlife,includ- and graft survival have shown steady improvement ing immunosuppression, is much easier than prior to over time. According to results from the UNOS Reg- transplantation. Successful pancreas transplantation istry,SPKrecipientswhounderwenttransplantation will not elevate all patients with diabetes to the level in 2006 achieved 1-, 3-, and 5-year patient survival of health and function of that of the general popula- rates of 94.8%, 90.8%, and 86.6%, respectively. Pan- tion, but transplant recipients consistently report a sig- creas graft functional survival rates for the same post- nificantly better quality of life than do patients who transplant time periods were 84%, 77%, and 73.2%, remain diabetic.

293 Section 6: Other abdominal organs

Further reading long-term patient survival. Transplantation 2001; 71: 82. American Diabetes Association. Economic Costs of Diabetes in the U.S. in 2002. Diabetes Care 2003; OPTN/UNOS Pancreas Transplantation Committee Report 26: 917. to the Board of Directors, Juen 21–22, 2010, Richmond, VA. Available at: http://optn.transplant.hrsa.gov/ AxelrodD,SungR,MeyerKH,WolfeRA,KaufmanDB. CommitteeReports/board main PancreasTransplan- Systematic evaluation of pancreas allograft quality, tationCommittee 6 24 2010 9 59.pdf outcomes, and geographic variation in utilization. Am J Transplant 2010; 10: 837. Salvalaggio PR, Davies DB, Fernandez LA, Kaufman DB. Outcomes of pancreas transplantation in the United Ojo AO, Meier-Kriesche HU, Hanson JA, et al.Impactof States using cardiac-death donors. Am J Transplant 2006; simultaneous pancreas-kidney transplantation on 6: 1059.

294 Section 6 Other abdominal organs Chapter Pancreatic islet transplantation

35 Heidi Yeh and James F. Markmann

Key points pression. The exclusion of exocrine tissue removes r Successful islet transplantation would result the risk of post-transplant pancreatitis with its atten- in larger numbers of insulin-independent dant problems, and the infusion of a cell prepara- recipients compared with whole organ tion obviates the requirement for disrupting major pancreas transplantation, both because of a blood vessels as would be required for whole organ larger donor pool and a larger transplant implantation. Even in patients who could tolerate candidate pool. the invasive surgery required for the latter, recovery r Improving isolation techniques and novel after islet transplantation would be much shorter and immunosuppression are resulting in islet easier. auto-transplantation outcomes approaching Furthermore, whole organ pancreas transplanta- those of whole organ pancreas tion is limited by the insufficient number suitable transplantation. for transplant: only 16% of deceased donors yield a transplantable pancreas. Good-quality islets, on the r Islet auto-transplantation is available as a other hand, can be obtained in sufficient numbers method of avoiding diabetes in from older and higher body mass index (BMI) donors pancreatectomy patients. with less stringent hemodynamic requirements than whole organ pancreata, so that potential islet donors from the current donor pool represent more than 1000 Endogenous insulin production can be restored by additional recipients a year who could achieve insulin beta (␤) cell replacement, most commonly achieved independence. through whole organ pancreas transplantation as Islet isolation does involve an additional 20 hours described in Chapter 34. This involves a major vascu- of processing time, USD 40 000 of expense (not includ- lar procedure, which may be difficult for long-standing ing the cost of processing pancreata that yield islets diabetics who have developed severe peripheral vas- of insufficient number or of setting up an islet pro- culardiseaseorarepoorcandidatesformajorproce- cessing facility), and as many as six staff members dures because of coronary artery disease. Although the beyond that needed for whole organ transplantation. presence of exocrine tissue is useful for early detec- More importantly, islet transplantation is historically tion of inflammation in the transplanted organ, severe neither as efficient nor as durable as whole organ pan- transplant pancreatitis due to ischemia or rejection creas transplantation and is therefore not yet a satisfac- can result in serious morbidity, and even mortality. tory replacement for pancreas transplantation. How- For many patients, these short-term risks outweigh ever, emerging data suggest that newer protocols in any survival advantage offered by improved glycemic islet transplantation may yield comparable results to control. those of pancreas transplantation, and if this trend In contrast, the main risk associated with isolated continues, islet transplantation will no longer be con- islet transplantation is that of chronic immunosup- sidered an experimental procedure.

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

295 Section 6: Other abdominal organs

Deceased donor allotransplant ing and replacement of the islet-toxic components of their immunosuppression prior to islet transplanta- Recipient selection tion, many remain on their kidney immunosuppres- sionanddoequallywell,sothismaybelessofa The aforementioned complications limit islet trans- concern. Too few simultaneous islet and kidney trans- plantation to patients with type 1 diabetes mellitus plants have been performed to ascertain the risk to a (DM1; confirmed by stimulated C-peptide Ͻ 0.3 ng/ newly transplanted kidney done under islet transplant ml, in addition to clinical history) requiring insulin immunosuppression. for at least 5 years and one of the following complica- In addition to the usual screening done for solid tions despite appropriate medical care and good com- organ transplant, weight and insulin requirement are pliance: hypoglycemic unawareness or metabolic labil- important factors in considering candidacy for islet ity characterized by either frequent episodes (two or transplantation. Specific numbers vary by center and more per year) of severe hypoglycemia in which the trial, but body weight greater than 75–90 kg make it patient is unable to treat him- or herself or has had unlikely that an adequate number of islets will be pro- hospital admission for ketoacidosis. These complica- cured to achieve insulin independence, even with islets tions represent a higher risk of mortality from diabetic recovered from two or three donors. Patients with BMI disease, therefore increasing the benefit of immuno- greater than 26–30 kg/m2 whorequiremorethan50– suppression over exogenous insulin treatment, even if 60 IU/d of insulin are likely to have peripheral insulin insulin independence may not be permanent. resistance in addition to DM1 and are also unlikely to Poor quality of life because of diabetic retinopa- become insulin independent with islet transplantation. thy, autonomic neuropathy (gastroparesis, postural hypotension, neuropathic bowel or bladder), or per- sistent, severe neuropathic pain has been considered Processing as an indication for improved glycemic control by In the 1970s, the first attempts to use islet transplanta- islet transplantation. Although it is well-documented tion clinically were stymied by the inability to extract that progression of microvascular disease is slowed or and purify islets. It was not until 1986 that the devel- halted by lowering HbA1c, it is not as clear that estab- opment of automated islet-cell processing made it pos- lished disease can be reversed, so secondary comor- sible to collect sufficient numbers of islets to reverse bidities are not commonly used as justification for islet diabetes, albeit only for a few days. Current technolo- transplantation. However, DM1 patients with a well- gies are all based on this method described by Camillo functioning kidney transplant may be candidates for Ricordi. islet transplantation after kidney transplant (without Islet donors undergo the same screening as solid one of the above complications) if they show evidence organ donors, although acceptable age and BMI for of poor glycemic control on insulin, e.g., HbA1c ≥ islets are much higher than for whole pancreas. Gen- 7.5%, since they are already on immunosuppression. erally, the pancreas is harvested as if for whole organ Islet transplant has not yet been associated with renal transplant, then stored and transported, on ice, either graft loss and in fact may prolong renal graft function in standard preservation solution or at the interface and survival. There have also been reports that islets between a preservation solution and a second layer transplanted after successful kidney transplant show of oxygenated perfluorocarbon. If the latter method is better function and survival than those transplanted used, the duodenum and spleen are usually removed alone. prior to storage. This second layer is meant to pro- Simultaneous islet and kidney transplantation is vide enough oxygen to the organ by diffusion to sup- an option in DM1 patients who qualify for a kid- port whatever low rate of metabolism continues during ney transplant, but have not yet received one. This cold storage and prevent anoxic injury. It has shown has not yet been widely adopted in clinical trials promise in animal models, especially with prolonged because the immunosuppression regimen for kidney storagetimes,buthasnotyetshownaclearadvan- transplantation is well established and quite effec- tage in clinical use where the cold storage time is kept tive and includes components that had been thought to a minimum. The pancreas should accumulate less to preclude successful islet transplantation. Although than 12 hours and preferably less than 8 hours of cold some islet after kidney candidates undergo trial wean- ischemia time.

296 Chapter 35: Pancreatic islet transplantation

Processing begins with dissecting the duodenum clinical use, the islets should be Gram stain and endo- and spleen off, if not already done, and clamping off toxin negative. The latter is important not only for the main and accessory ducts before dividing the pan- prevention of systemic inflammatory response syn- creas at the neck. Cannulas are inserted into the main drome (SIRS) in the recipient but also to prevent a pancreatic ducts of each piece at the cut surface. An local inflammatory response that will prevent subse- enzyme blend containing collagenase is then infused quent engraftment or spur rejection. These islets may through the ducts until the two halves are well dis- be cultured for up to 72 hours to “rest” the islets tended. For about a decade, Liberase HI (manufac- and allow them to recover from any ischemic or trau- tured by Roche), which contained both collagenase matic injury. This can also be helpful if extra time and a neutral protease, thermolysin, had been used is needed to prepare the patient with pre-transplant as the standard reagent. Because production involved immunosuppression. contact with material from bovine brain, there was Islets are quantified by packed tissue volume concern about the risk of transmitting bovine spongi- (5–10 ml per infusion) and by islet equivalent (IE). An form encephalopathy with the islets, so this was dis- IE is defined by both insulin content, generally deter- continued in 2007. The standard switched to a collage- mined with an insulin granule binding dye such as nase only preparation (manufactured by Serva), which diphenylthiocarbazone and size, based on a spherical required addition of a protease. This switch was associ- islet of mean diameter 150 ␮M. It is rare to get more ated with reports of decreased islet yields, but this has than 600 000–700 000 islets per pancreas, with higher improved with alterations in processing techniques, BMI donors tending to yield more islets, and the num- and formal studies have not affirmed such observa- ber is usually much lower. Only 30–60% of islet isola- tions. Roche now makes a mammalian tissue-free ver- tions result in material adequate for transplantation. sion of Liberase HI, which otherwise appears, in ini- tial studies, to produce acceptable islets. However, vari- ability in the potency of enzyme blends from one lot to Infusion another continues to be an issue. The islets (Figure 35.1) are resuspended into 50– The pancreas is then cut into small pieces and 200 ml of solution supplemented with human serum placed into a digestion chamber, which is warmed, in albumin and heparin. More than 90% of islets are order to activate the enzymes. The digestion chamber transplanted into the liver by infusion through the continuously shakes, adding mechanical disruption to portalsystem.Thiscanbedonepercutaneouslyor enzymatic digestion. Continuous flow of fluid through through a mini-laparotomy and are allowed to infuse the chamber removes the smallest fragments of tis- under gravity. Percutaneous infusion requires only sue, which are sampled and visually inspected. When local anesthesia and is performed by interventional intact islets are visualized, the fragments are diluted radiology via transhepatic catheterization of the por- with cold culture media or other physiological solu- tal venous system. Portal vein pressure is recorded tion and kept cool to prevent further digestion. The before, during, and after the infusion; if pressures fibrous network of ducts and vessels remains in the exceed 22 mmHg, the infusion is temporarily halted digestion chamber. Further purification can be done to allow pressure to drop back below 18 mmHg. Care- by density-gradient separation, as highly purified islets ful monitoring of portal pressures and cessation of are less dense than islets still attached to exocrine tis- infusion as necessary have largely eradicated compli- sue. The goal is greater than 30% purity and greater cationssuchasportalhypertensionandsplenichem- than 70% viability (as determined by fluorescent dye orrhage. Because there is no reversal of heparin, coils exclusion or inclusion), and it is more difficult to get and gelatin-sponge pledgets are left in the puncture high-purity islets from younger donors, although cell tract during catheter removal to decrease the risk of clusters from younger donors are speculated to con- bleeding. This probably decreases the very low risk of tain pancreatic duct stem cells that may develop into biliary leakage and arteriovenous fistula as well. Alter- insulin-producing cells post-transplant. natively, access to the portal vein can be gained by open Asampleofthefluidmaybetakenforlong-term placementofaninfusioncatheterintoanappropriately microbiological culture in order to identify any rare sized omental vein. This is done under general anesthe- pathogens that may be infused into the patient, and sia through a small abdominal incision in the operat- antibiotics are added to the final tissue collection. For ing room. The vein is ligated upon withdrawal of the

297 Section 6: Other abdominal organs

planted islets offers some physiological advantages, allowing insulin to be released through its usual route of delivery through the liver, exposing hepatocytes to “portal” concentrations of insulin that stimulate glu- cose storage, and permitting the liver to play its normal role in regulating systemic insulin levels. Portal circu- lation contains the highest concentrations of immuno- suppression, which is advantageous in an islet alone patient, but this may prove toxic in a patient with a coexisting kidney allograft, which requires mainte- nance of adequate immunosuppressive levels in the systemic circulation. There are additional reasons that intraportal infu- sion may be deleterious to islets. Although the liver Figure 35.1 Islets isolated from deceased donor pancreas, ready hasanarterialaswellasportalvenousbloodsupply, for infusion. Photograph courtesy of J. Lei, MD, MBA, Massachusetts parenchymal oxygen levels are lower than in the pan- General Hospital. creas, probably particularly so for islets lodged in the small vessels of the portal system. Islets in the liver infusion catheter. The incidence of branch portal vein have greater exposure to any ingested toxins or those thrombosis (no complete portal vein thrombosis has released by gut bacteria, and the high glucose and lipid been reported) is similar with both procedures and has levels in post-prandial mesenteric blood may stress ␤ decreased with newer purification techniques, lower cells as well. Scattered throughout the liver, it is diffi- islet cell volumes, and the use of heparin, both 70 cult to biopsy islet allografts for monitoring purposes, IU/kg of heparin in the infusate and a continuous infu- and in the unlikely chance that the islets were contam- sion post-procedure. Hepatic macro-infarction has inated by microbes or malignant cells, they could not notbeenreportedinhumans,butmousemodelsshow be removed without damaging or removing the liver. peri-islet areas of hepatic necrosis by reflected light Auto-islets transplanted into the livers of pancreatec- confocal microscopy. This probably corresponds to the tomized dogs eventually lost function in nearly 80% small and transient rise in serum aminotransferases of animals within 15 months, suggesting that the liver seen after intraportal islet infusion. However, it does does not support long-term survival of islets, even in not appear to have any clinical sequelae. the absence of auto- and allo-immunity. An average of 8000–12 000 IE/kg recipient body Furthermore, placement of islets in a vascular weight is needed to achieve insulin independence, space exposes them to immediate blood-mediated although insulin independence has been reported with inflammatory reaction (IBMIR). IBMIR occurs when less than 6000 IE/kg. Less than 5000 IE/kg body proinflammatory factors such as tissue factor and weight as an initial infusion is not generally considered monocyte chemoattractant protein 1 are upregulated worth exposing the patient to foreign material and by islet cells during brain death and organ pro- immunosuppression, and two to four infusions may curement. These engage the coagulation and com- be required for insulin independence. Islets from mul- plement systems, platelets, and polymorphonuclear tiple donors may be combined into a single infusion, neutrophils (PMNs) when transplanted islets come in if the donors are simultaneously available, or infu- contact with blood in the portal vein. In addition, col- sions may be given serially, either when the initial infu- lagen and other extra-cellular matrix proteins present sion results in decreased insulin requirements without in islet clusters are prothrombotic and trigger inflam- complete insulin independence, or when an initially mation. Some speculate that as much as 75% of the successful graft function decreases to the point of the transplantedisletmassislosttoIBMIR,compounded recipient requiring insulin again. by inflammation induced locally by hypoxic hepato- The liver became the preferred site of transplanta- cytes surrounding islet emboli in the portal vessels. tion because animal studies showed that fewer islets Many other sites have been used in the laboratory, were required to reverse diabetes than in any other including the spleen, pancreas, peritoneum, stomach easily accessible sites. Intrahepatic placement of trans- wall, bone marrow, testis, brain, and systemic venous

298 Chapter 35: Pancreatic islet transplantation circulation (lodging in the lung), but only a few have as well as participating in IBMIR. Its use at induc- been used clinically. Placement of islets under the kid- tion improves marginal mass allograft outcomes in ney capsule is frequently used in mice, but was used rodents and has been associated with better likelihood only briefly in the early days of clinical islet trans- of insulin independence in marginal mass studies in plantation; surgical access to the kidney capsule is humans. highly invasive and the space has lower oxygen ten- One simple method of decreasing ␤-cell toxicity sion than the liver parenchyma. Intramuscular islet has been to maintain patients on gradually decreasing transplantation has been done in humans (at subther- amounts of insulin during and after islet transplanta- apeutic doses) and is of special interest because of easy tion. This allows the islets to recover and engraft before accessibility for biopsy, but in animals, it has required exposing them to hyperglycemia and increasing their coadministration of a growth factor or a prevascular- metabolic demands. Insulin dosing obviously needs ized device to achieve metabolic function and seems to be carefully monitored to avoid hypoglycemia, and to induce vigorous leukocyte infiltrates. One patient glucose may need to be administered. who received autologous islets into the brachioradialis Agents such as exenatide/Byetta aim to improve muscle continued to require insulin after pancreate- the function of existing islets. Exenatide is a synthetic ctomy for hereditary pancreatitis, but was C-peptide version of a hormone first isolated from Gila mon- positive for at least 2 years. One attempt was made ster saliva, which enhances glucose-dependent insulin at a combined intrathymic and intraportal transplan- secretion and suppresses glucagon secretion. It is in tation, but the patient failed to become insulin inde- clinical use to improve glycemic control in type 2 dia- pendent in spite of receiving pooled islets from seven betes mellitus in conjunction with oral hypoglycemics donors, and nothing is known of the contribution and also appears to increase the likelihood of achiev- that thymic injection made to long-term C-peptide ing insulin independence and prolong insulin inde- positivity. Pilot studies have been done and larger pendence in uncontrolled studies. trials are underway for subcutaneous placement of both allo- and xeno-islets in macrocapsules or within vascularized prostheses, but few data are available Immunosuppression (also see Chapter 3) now. Classically, glucocorticoids have played a major role in post-transplant immunosuppression. However, steroids are known to inhibit insulin secretion by ␤ Engraftment and function cells and may exert a direct toxic effect on them, as Many strategies are used to promote better engraft- well. At the same time, steroids increase the demand ment.Heparinisgivenatthetimeofinfusiontopre- for insulin by simultaneously stimulating liver glu- vent portal vein branch thrombosis, but heparin also coneogenesis and inhibiting peripheral uptake of has important anti-inflammatory activities. In addi- glucose. It was not until 1990 that a steroid-free tion to its anti-platelet and anti-thrombin properties, immunosuppression regimen including tacrolimus heparin inhibits complement activation and impairs was reported to result in long-term insulin indepen- leukocyte function by disrupting leukocyte adhesion dence of one patient. In that series, however, patients molecules’ interaction with their ligands and inhibit- did not have autoimmune diabetes and had become ing PMN elastases and proteases. Low-molecular- diabetic as a result of upper abdominal exenteration weight dextran sulfate can also be given with islet infu- (requiring liver transplant at the same time). Fur- sion. It is associated with a lower risk of bleeding than thermore, only five of nine patients were insulin-free heparin, and in addition to its anti-coagulant prop- 6 months after transplant. Real enthusiasm for islet erties, it also inhibits complement activation. Other transplantation as treatment for DM1 arose from ini- drugs are primarily anti-inflammatory agents, but do tial reports of the results of the “Edmonton protocol,” carry some increased risk of bleeding. Etanercept is which combined steroid-free immunosuppression a fusion protein consisting of the extracellular bind- with infusions from one to four donors, as necessary to ing portion of human tumor necrosis factor receptor achieve insulin independence. Seven of seven patients linked to the Fc portion of human immunoglobulin in a single-center study became insulin independent (Ig) G1. It binds and deactivates tumor necrosis fac- and remained insulin independent 5.5–15 months tor alpha (TNF␣), which is directly toxic to ␤ cells, lateratthetimeofpublication.

299 Section 6: Other abdominal organs

The international trial of the Edmonton proto- latory T cells (Tregs) and reduced allospecific T-cell col included 36 patients at nine international sites activity. who received induction with daclizumab and main- Other protocols still using TAC and/or SRL have tenance with sirolimus (SRL) and tacrolimus (TAC). also been promising. Teplizumab is a humanized Fifty-eight percent were insulin independent at any mouse anti-CD3 antibody that is also in clinical tri- point during the study, 44% were insulin independent als for newly diagnosed DM1 as an attempt to preserve 1 year after the final transplant, and only 14% were still the remaining ␤ cells. In an islet transplantation pilot insulin independent 2 years after transplant. An addi- study using teplizumab for induction with SRL and tional 28% required insulin but no longer had hypo- low-dose TAC maintenance, four of six patients were glycemic episodes at 1 year, and a total of 70% were insulin independent at 1 year and three of six were still still C-peptide positive at 2 years. There was a great insulin independent at 3 years. At this time, there are deal of center variability, with some centers achiev- still ongoing trials with teplizumab. ing 1-year insulin independence in 100% of their Some current islet transplantation trials use alem- patients, whereas other centers had no patients who tuzumab induction and TAC/MMF maintenance. A were insulin independent at 1 year. A 5-year follow- pilot study reports 90% insulin independence at up from Edmonton, as a single-center study, reported 2 years. An unpublished study has 10 of 12 insulin 10% who were insulin independent and 80% who were independent patients with maximum follow-up of 4 C-peptide positive. years (the remaining two patients stopped immuno- Part of the issue may be that TAC (perhaps a gen- suppression because of meningitis and a brain abscess, eralized calcineurin inhibitor [CNI] effect) can cause both of which have resolved with treatment and deficiencies in insulin secretion, reducing insulin immunosuppression withdrawal). These are small mRNA and protein production, as well as decreasing numbers of patients, but these results parallel long- ␤-cell proliferation. Animal studies suggest that CNIs term outcomes with whole organ pancreas transplan- also reduce peripheral tissue sensitivity to insulin. tation. Alemtuzumab has also been used in China SRL has also been shown to prevent ␤-cell division, for simultaneous islet and kidney transplants with inhibiting replacement of ␤ cellslosttoattrition.The promising results at 1-year follow-up. The fact that effects of TAC and SRL on ␤-cell proliferation are CD52 is present on natural killer cells, monocytes, additive. and macrophages raised the possibility that alem- Some new immunosuppression protocols attempt tuzumab might reduce IBMIR, and quantitative func- to address this problem by minimizing exposure to tion tests at 3 months do suggest improved engraft- TACand/orSRL.Clinicaltrialswithdaclizumabor ment. There is also evidence that islet transplant basiliximab induction and maintenance with belata- patients on alemtuzumab have donor-specific Tregs cept and mycophenolate mofetil (MMF) are under- that produce interleukin-10 and decreased allo- and way, with early, unpublished results of five of five islet-specific T-cell reactivity. patients insulin independent after a single infusion for The 2009 annual report of the Collaborative Islet 70–305 days. Transplant Registry suggests that T-cell–depleting Another TAC-free protocol used anti-thymocyte induction is associated with a 0.3–0.8-fold lower globulin for induction and efalizumab together with hazard ratio of losing insulin independence than either SRL or MMF for maintenance. Efalizumab is interleukin-2 receptor blockade induction. Unpub- of particular interest in transplant for DM1 because lished subanalysis of data collected by the Collabora- it has the potential to inhibit T cells that do not tive Islet Transplant Registry suggests that the use of T- require costimulation for activation and proliferation cell–depleting induction is associated with higher rates (especially in the setting of lymphopenia inducing of initial insulin independence and longer persistence, immunosuppression), such as memory T cells that approaching that of whole organ transplantation. would be present in autoimmune diseases such as DM1. There are unpublished reports that four of five patients remain insulin independent 150–240 days Post-transplant monitoring after withdrawl of efalizumab for safety reasons (some In the International Trial of the Edmonton Proto- patients remain on single-agent maintenance with SRL col, insulin independence was defined as HbA1 less or MMF only) and have increased number of regu- than 6.5%, fasting glucose ≤ 140 mg/dl three or fewer

300 Chapter 35: Pancreatic islet transplantation times a week, and post-prandial glucose ≤ 180 mg/dl patient and to document their progression or regres- four or fewer times a week, in the absence of exoge- sion following islet transplantation. Although markers nous insulin. C-peptide, particularly valuable in mon- are being developed for visualization of intrahepatic itoring graft function in patients who have developed islets by positron emission tomography and magnetic an insulin requirement after being insulin indepen- resonance imaging, there is not yet a standard imaging dent, helps determine whether immunosuppression technique. should be continued in order to maintain some level of endogenous insulin production. Auto-transplant and living donor Many patients who require supplemental insulin still experience freedom from hypoglycemic episodes allo-transplant and metabolic lability. It is likely that the residual The remaining cases of insulin-deficient diabetes fol- islet mass is responsible for enough insulin production low surgical removal of the pancreas for chronic that if the islets stop releasing insulin in response to pancreatitis or for pancreatic cancer and occasion- hypoglycemia, the exogenous insulin is not sufficient ally, non-specific inflammatory destruction of the to completely inhibit glucagon release from ␣ cells. endocrine tissue by chronic pancreatitis. There is ␣ cells may also become more responsive because of a fairly large experience with islet auto-transplant the intervening avoidance of hypoglycemic episodes for chronic pancreatitis, but because of the risk of while the patient was insulin independent. Prevent- malignant cells contaminating the islet preparation, ing severe hypoglycemic episodes and metabolic labil- thereisonlyonereportfromGermanyofisletauto- ity not only avoids hospital admissions and risk of transplant done for pancreatic adenocarcinoma. The death, but also provides a significant improvement in patient never achieved insulin independence and died the patient’s quality of life, even if insulin adminis- 2.5 years later from tumor recurrence, although not in tration is required. Patients with partial graft func- the liver. tion often have lower HbA1c levels when compared Insulin independence occurs less often with islet with pretransplant or to those with negative C-peptide, auto-transplantation: 32% at 1 year in the Univer- and this has been associated with fewer cardiovascular sity of Minnesota series of 173 patients, with the first deaths and better atherosclerotic profiles. These bene- transplanted in February 1977. However, it must be fits are generally thought to be worth the risk of con- remembered that islet auto-transplants are all single- tinued immunosuppression. donor infusions, the donor has a diseased pancreas Provocative testing such as measuring glucose and and may be non-ideal for other reasons, and that many insulin responses to mixed meals and intravenous glu- transplants in the series were done before islet isola- cose can give more quantitative information regard- tion techniques had been optimized (75% of the auto- ing declining islet function before the development transplant recipients received 5000 or fewer IE/kg). of insulin requirements, but the indications for treat- However, of those who do achieve insulin indepen- ment are not yet clear. Similarly, anti-islet antibodies dence, 74% remained so at 2 years, 69% at 5 years, are monitored with the rationale that re-appearance and 28% at 10 years. One of the patients continues after transplant may signal impending rejection, but to be insulin independent over 13 years later. In con- their significance is still ambiguous and not defini- trast, among islet allo-transplant patients who achieve tivelyworththerisksofincreasingimmunosuppres- insulin independence, only 45% remained so at 2 years sion. Anti-HLA antibodies have been associated with and less than 10% at 5 years. rejection in solid organ transplants and declining islet The experience in islet auto-transplant suggests function,aswell.Therehasbeenonecasereportofan that late attrition in allo-transplants is not due solely insulin-independent patient who developed anti-HLA to using the liver as the site of transplant, as sug- antibodies 6 months prior to developing poor glycemic gested by previously mentioned dog studies. There are control eventually requiring treatment with insulin. A a small number of autopsy specimens from islet allo- course of rituximab and intravenous IgG resulted in transplant patients who died from other causes, but recovery of insulin independence and loss of anti-HLA who had increasing insulin requirements with par- antibody expression. tial graft function. Although those show loss of islet Standard surveillance for diabetic complications mass without evidence of immune damage or infil- should continue as for usual clinical care of a diabetic tration, the difference between autograft and allograft

301 Section 6: Other abdominal organs

persistence suggests that much of the gradual loss of Minimizing IBMIR, which may begin with donor islet mass is likely immune-mediated. The fact that management, is crucial to improving engraftment 32% of patients can achieve islet independence when rates and the effectiveness of islet transplantation. only 25% received more than 5000 IE/kg supports the The risk of bleeding limits the dosage and duration thoughtthatpoorengraftmentinvolvesfactorspecu- of heparin for the purposes of IBMIR, but alternate liar to deceased donors, which are exacerbated by pro- forms of heparin such as 2-O,3-O-desulfated heparin, longed cold storage. which retain their anti-inflammatory properties and There has also been one report of successful liv- have much lower anticoagulation activity, may be an ing donor islet transplant in Japan. The recipient was option. In the laboratory, one group has coated islets diabetic because of pancreatitis and suffered from with heparin to hide the proinflammatory components hypoglycemic unawareness. Her mother, the donor, from their receptors in the bloodstream, minimizing donated her distal pancreas, which yielded more than systemic exposure to anti-coagulation. Other agents 400 000 islets. The immunosuppression regimen fol- with combined anti-coagulant and anti-inflammatory lowed the Edmonton protocol, supplemented with properties, such as activated protein C and CD39, are infliximab, an anti-TNF␣ monoclonal antibody. At the under investigation. 2-month follow-up, both donor and recipient were To prolong graft survival, more effective and safer insulin independent. forms of immunosuppression are constantly being sought. Besides developing new pharmaceuticals and Future of islet transplantation bioagents, tolerance induction and encapsulation tech- It is likely that islet allo-transplantation will soon be nology are also possible methods to protect allo- equivalent to whole organ pancreas transplantation in islets from the recipient immune system. Probably terms of graft function and survival, while retaining the most poorly defined factor in islet transplanta- the advantage of being a much lower risk procedure. tion is the long-term, non–immune-related loss of islet mass that is observed in animals and humans. This Recent trials suggest that as many as 80% of patients ␤ achieve insulin independence. Graft survival has also will require a better understanding of -cell homeo- improved significantly. In the Edmonton international stasis. trial, insulin independence persistence rates were 24% at 2 years, but this had improved to approximately Further reading 40% at 3 years for transplants done between 1999– Clinical Islet Transplantation Consortium. Available at: 2004, and then to approximately 60% at 3 years for www.citisletstudy.org. transplants done after 2004. For islet transplantation to Collaborative Islet Transplant Registry. Available at: www become standard treatment for DM1, further improve- .citregistry.org. ments are needed in immunosuppression and in dura- Fiorina P, Shapiro AMJ, Ricordi C, Secchi A. The clinical bility and effectiveness. Furthermore, more sources of impact of islet transplantation. Am J Transplant 2008; 8: islets are necessary. This last issue is being approached 1990–7. by efforts at developing animal sources of islets and International Islet Transplant Registry. Available at: www engineering ␤ cells in vitro, either from stem cells or .med.uni-giessen.de/itr. re-programming differentiated cells. Once islet allo- Islet Cell Resource Center. Available at: icr.coh.org. transplant becomes more reliable, living donors may Nath DS, Hering BJ. Islet cell replacement therapy. Clin Lab be considered. Med 2005; 25: 541–56.

302 Section 6 Other abdominal organs Chapter Intestinal transplantation

36 Stephen J. Middleton, Simon M. Gabe, Neville V. Jamieson, and Andrew J. Butler

Key points 250 r Parenteral nutrition remains the first-line treatment for irreversible intestinal failure. 200 r Intestinal transplantation now offers most 150 patients an improvement in quality of life as compared with complicated parenteral 100 nutrition. r Survival is now better than that of lung and 50 approaching that of liver transplantation, and Number of procedures systems are being developed to more 0 accurately predict postoperative survival of individual patients. 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 r Intestinal transplantation is a cost-effective Year alternative to parenteral nutrition as the costs Figure 36.1 Intestinal transplantations undertaken per year of the procedure equate to the combined (adults and children). Adapted from the International Registry, with costsof2yearsonparenteralnutrition,after permission: www.lhsc.on.ca/itr. which a cost saving is made. r Acute rejection of a functioning graft is partly accounted for by the presence of an initial reser- invariably associated with a change in its voir of patients who required the surgery leading to a function; the stomal output is usually rapid rise in its use and then by the steady increase in increased and more watery in consistency. transplant centers offering the procedure and gradual broadening of indications which maintain the rate of use at the current level. As results improve, the tech- The technical ability to transplant the intestine was niqueisofferedtopatientsearlier,andifthistrendcon- recorded over a century ago; however, many attempts tinues, it may be used as an alternative to parenteral at this procedure over the intervening years did not nutrition(PN)overthenext5to10yearsratherthan achieve even short-term graft or patient survivals. Suc- being an option only when PN cannot continue due to cess, in terms of medium-term survival with fully complications. functioning grafts, was finally reported in the late Current survival chances are probably best rep- 1980s. The key factor in the progression of the tech- resented by the data from the international registry. nique was the development of suitable immunosup- One-, 5-, and 10-year survivals are approximately 65%, pressive regimes. 40%, and 30%, respectively, but if the last 10-year expe- The exponential increase in the use of the pro- rience is considered, survival figures have improved cedure in the 1990s has been superseded by a less to 75%, 50%, and 40%, respectively, and in adults dramaticriseoverthelast10years,whichnowhas 5-year survival is approaching 60%. There are slight almost reached a plateau at about 200 procedures differences in survival between types of procedures worldwide per year (Figure 36.1). This trend can be (isolated intestine, intestine and liver, or multivisceral)

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

303 Section 6: Other abdominal organs

undertaken. These tend to fluctuate, and currently intestinal failure associated liver disease and visceral combined small intestine and liver grafts are associ- myopathy or neuropathy who develop liver disease and ated with a better survival, at least if the comparison central venous catheter sepsis. is made between those surviving the first postopera- tive year. The experience of the center is also an inde- pendent risk factor, with those centers with less vol- Timing of intestinal transplantation ume having an inferior survival record. Interestingly, Much of the improvement in postoperative survival is most of the improved survival has been seen in the first due to the realization that patients need to be in good year, after which the survival curves have not changed physical and psychological condition to survive the over the last two decades. This suggests that the main procedure. There is evidence that survival is inversely focus of attention should now turn to outpatient care associated with comorbidity, and clinical scoring sys- and follow-up and the understanding and prevention tems that will semi-quantify this risk are being devel- of chronic rejection. oped. Therefore, patients who have active reversible comorbidity are not usually offered transplantation. Management of patients with Those with sudden loss of intestine due to conditions such as volvulus or superior mesenteric artery throm- irreversible intestinal failure bosis should be established on PN and all efforts made PN remains the primary treatment for patients with to improve and stabilize their physical and psychologi- irreversible intestinal failure. This is currently associ- calcondition.Manysuchpatientswillremainwelland ated with a better survival chance than intestinal trans- content on PN and currently would not be considered plantation, with 10-year survival in the literature of as candidates for transplantation. 43–71%, and when well tolerated can lead to excellent Patients who cannot be established on stable PN duration and quality of life. However, occasionally due are usually not candidates for transplantation as they to complications of PN or to progression of an underly- areunlikelytosurvivetheprocedure.Thoseinitially ing disease process, PN cannot be safely continued and established on PN who subsequently develop com- transplantation is then considered (Table 36.1). Such plications or experience progression of an underly- transplantation candidates can be divided into three ingdiseaseprocessshouldbeevaluatedfortheirsuit- categories: ability for transplantation. Certain groups of patients 1. Patients with life-threatening complications of on PN can be considered as high risk for PN com- PN, such as (a) failing vascular access, (b) plications and therefore potential candidates for ear- PN-related liver disease, or (c) severe recurrent lier transplantation. The main PN-related indications line-associated sepsis. in adults and children are loss of vascular access and 2. Patients with indications for extensive PN-related liver disease (Table 36.1). Patients who evisceration of abdominal organs requiring lose two or more major venous access sites are can- concurrent transplantation. didates for assessment as it is important to evaluate 3. Patients requiring transplantation of other them while venous access remains sufficient for trans- abdominal organs. plantation. The risks associated with the transplanta- tion procedure are increased when venous access falls A fourth category is emerging that considers below two or three major veins, depending on the patients with very poor quality of life on PN as poten- extent of the planned surgery. Those who develop liver tial candidates. diseaseneedcarefulevaluationasitmaybepossi- When caring for this group of patients it is essen- ble to avoid liver transplantation if the small intestine tial to appreciate that they may benefit from a trans- is transplanted before progression to severe liver dis- plant in the future and should not be allowed to drift ease. This progression can often be insidious and isa through a downward spiral of PN complication or dis- leading cause of death on the transplantation waiting ease progression without considering transplantation list. as an option. Certain subgroups of patients on PN Occasionally, patients find their quality of life con- have a worse prognosis and should be monitored more siderably impaired by PN and explore the possibil- closely to allow timely referral. These include those ity of intestinal transplantation as a way of improv- with ultra-short bowel syndrome who tend to develop ing it. A cautious approach should be taken in these

304 Chapter 36: Intestinal transplantation

Table 36.1 Summary of indications for intestinal transplantation 1. Life-threatening complications of PN a) Progressive liver disease • Severe liver disease or progressive disease despite all remedial actions b) Recurrent septic episodes • IF patients who have severe septic complications (i.e., life-threatening line infection needing admission to ITU, or recurrent fungal or yeast or candidal infections) c) Lack of central venous access (including femoral veins) • For isolated intestine: venous access limited to three major sites • For intestine as part of a cluster graft: venous access limited to four major sites 2. Very poor quality of life thought to be correctable by transplantation. 3. Patients with indications for extensive surgery involving partial or complete evisceration: Adults a) Surgery to remove a large proportion of the abdominal viscera, which is considered untenable without associated multi-visceral transplantation (e.g., extensive desmoid disease, extensive severe mesenteric arterial disease requiring intervention) b) Localized malignancy considered to be amenable to curative resection, which would necessitate extensive evisceration (e.g., localized neuroendocrine tumors and cholangiocarcinoma – particular caution should be exercised with this group) Children a) Surgery that will lead to: • Terminal gastrostomy • Terminal duodenostomy • Ultra short bowel: In children Ͻ10 cm, with or without ileocecal valve • Recurrent fluid and electrolyte management problems needing frequent hospital admission 4. Patients requiring transplantation of other abdominal organs a) When the transplantation procedure is expected to preclude the possibility of future intestinal transplantation (loss of vulnerable and limited venous access, further HLA sensitization) b) When the risk is increased by excluding the intestine from the graft (predictable problems related to administering immunosuppression [e.g., line sepsis], ongoing severe intestinal disease such as diabetic visceral neuropathy, or ultra short bowel syndrome, causing fluid, electrolyte, and acid–base balance problems [damage to renal graft]), and when the need for subsequent intestinal transplantation is considered likely

situationsastherewilllikelybeatrade-offbetween be explained that they will have an ileostomy after gain in quality of life and reduced longevity. This can the procedure. There may be the possibility of joining be difficult for patients to appreciate, and all efforts the transplanted small bowel onto the patients native must be made to ensure that they have full psychi- colon (if still present) after a number of years, but this atric and psychological evaluation and support. It is is currently an area of development. also important to appreciate that quality of life may not improve after transplantation and they will not The process and requirements for return to a medication-free and morbidity-free life that they often seek. Most patients will be vulnerable to intestinal transplantation infections and drug side effects, although these fac- tors are usually well tolerated. It is also very impor- The transplant team tant for patients to appreciate that intestinal transplan- The establishment of an effective and comprehensive tation does not currently guarantee a life without a team of health professionals is one of the key factors stoma. If an overriding reason for a patient to have for the success of a program. Core members should a transplant is to remove their stoma, then it should include a lead gastroenterologist, a lead surgeon, and at

305 Section 6: Other abdominal organs

least one additional named senior colleague in each of abdominal organs. This is often required to resolve dis- these disciplines with adequate experience and knowl- ease or restore function of vital organs and occasion- edge to provide support and cover. There should be ally for technical reasons to reduce the risks associated an established team of transplant surgeons with suf- with the surgery. ficient numbers to prevent overcommitment of any Where irreversible intestinal failure is the only sig- individual during very active periods. Specialist nurse nificant dysfunction, “isolated” intestinal grafts are coordinators provide the platform of organization and used. This has the theoretical advantage over other nursing expertise. In addition to the core team, there optionsasitispotentiallyreversibleinthecaseof must be many other interested parties who are invited catastrophic rejection. Associated pancreatic disease as named opinions in their specialty. As the program such as in type 1 diabetes mellitus or gastropare- expands, this list of allied professionals becomes wide- sis as found in patients with visceral neuropathy ranging. or myopathy may warrant transplantation of stom- ach, duodenum, pancreas, and small intestine. This is Referral and assessment of adult patients often termed a “modified multi-visceral” graft because it excludes the liver. When undertaken in conjunc- Patients thought to be candidates are usually assessed tion with the liver, it is termed multi-visceral”. When during a short programmed admission to the trans- disease is limited to the small intestine and liver, plant center when a comprehensive evaluation of their such as in intestinal failure related liver cirrhosis, a gut function, anatomy, and comorbidity is undertaken. combined liver/intestine graft is offered to patients. In the United Kingdom, all adult patients are then The international registry report that isolated grafts discussed at the National Forum for Adult Intestinal are the commonest (59%), multivisceral account for Transplantation, at which recommendations regarding 24%, and combined intestine/liver for the remai- the most appropriate management plan are made. In ning 17%. each case, this process is a prerequisite to receiving National Commissioning Group financial support to undertake the procedure. It is important to involve external peer review Listing and problems on the before listing patients for transplantation. This encour- waiting list ages uniformity of practice and improves the selection In the United Kingdom, adult patients are considered process. A detailed preoperative assessment of patients by the National Forum for Adult Intestinal Transplan- allows their preoperative status to be semi-quantified tation for their suitability as candidates for transplan- and included when considering center performance tation and are also assessed locally for the feasibility of statistics. the transplant in each individual case. There may also Particular attention should be given to those who be local concerns about comorbidity and complexity of arelikelytofallthroughthewindowofopportunity. peri- and postoperative care that must be considered. These patients often have progressive disease, which Once listed, the final preoperative preparations are may advance to a point that contraindicates transplan- made and they are registered with the central govern- tation or results in death while on the waiting list. ing organization (UK Transplantation). At this stage it Examples of this include patients who are rapidly los- isimportanttoinformallthoseinvolvedinthecurrent ing venous access points and those bleeding from por- or future care of the patient, confirm that tissue typ- tal hypertension. Special consideration should also be ing have appropriate samples, a precise surgical plan given to PN-dependent patients who require trans- has been made, and fully informed written consent for plantation of other organs and may benefit from a clus- this plan and other possible outcomes obtained. The tergraftincludingintestineratherthanreceiveasubse- typical waiting time for a suitable donor in the United quent intestinal graft in the setting of an existing graft Kingdom is 9 months, and during this period patients and consequent immunosuppression. are subject to deterioration. Patients on the waiting list need to be followed carefully to ensure that they are Selection of type of grafts in a position to undergo transplantation if called. This Intestinal transplantation is frequently undertaken usually involves reviewing them twice a month until as part of a composite graft containing other intra- transplantation.

306 Chapter 36: Intestinal transplantation

Conditions leading to the need for intestinal challenging, whereas adults tend to find employment transplantation amajorproblem. The majority of patients have irreversible intestinal The immediate preoperative period failure and are not able to continue PN because of com- plications. Some have conditions that require extensive On admission for the transplantation procedure, base- evisceration that cannot be undertaken without trans- line tests should be repeated, and it is important to plantation. A short bowel is the commonest under- check that written informed consent for the proposed lying condition in both pediatric and adult patients. surgery is in the notes and confirm that the patient is In the pediatric transplanted population, it accounts aware of the likely postoperative situation in the inten- for 68% of patients and is predominantly the result sive treatment unit (ITU). The psychological status of of gastroschisis (24%), necrotizing enterocolitis (16%), the patient has a major effect on their postoperative volvulus (10%), and intestinal atresia (9%). In adult recovery;itisimportanttomakeitcleartothemthat transplanted patients it accounts for 58% and is pre- they will experience frequent setbacks postoperatively dominantly the result of volvulus (10%), Crohn’s dis- and that this is to be expected and should not be taken ease (8%), and trauma (7%). as an indication that the procedure is failing. Dysmotility including visceral myopathies and neuropathies represent another large group of patients The surgery (approxmately 14% overall). Adult patients with this The assessment summary and protocol should be condition tend to have a worse prognosis on PN and available wherever the patient is being managed. It is are consequently considered to be a high-risk group. common for the surgery to take 12–16 hours, after They tend to get central venous catheter sepsis and which patients often remain on ITU for 3–4 days, then lose venous access as well as develop liver disease high-dependency unit for 2–4 weeks before transfer and should therefore be watched more closely. Refer- tothewardfor4–6weeks.Thistimelineisobviously ral for consideration of transplantation should occur subject to large variations. Peri-operative immunosup- promptly if these complications occur. pression is summarized in Table 36.2. The majority of patients with tumors (7% over- all) have desmoids. These benign but locally inva- sive tumors also need very careful monitoring. Early Splenectomy referral may allow a less extensive transplant proce- Many patients require a splenectomy. The need for this dure. The growth of desmoids can suddenly acceler- cannot always be predicted; therefore, it is our prac- ate, and this should trigger an immediate discussion tice to prepare patients for this eventuality. Appropri- with a transplant center. Patients transplanted because ate preoperative vaccinations are given and postoper- of other tumors have usually had a relatively local- ative prophylaxis with penicillin V. ized neuroendocrine tumor or cholangiocarcinoma. The survival of some of these patients has been very Other considerations good, extending over 5 years, although a detailed study If renal transplantation is also undertaken, mannitol of the outcome of this group has not been published. 20% 100 ml intravenously (IV) and furosemide 20 mg IV should be given before revascularization. Transplantation in pediatrics Pediatrictransplantationisnotthespecificsubjectof Postoperative care this chapter. However, the general principles of how and when to use intestinal transplantation for adults Immunosuppression can also be applied in most cases to pediatrics. Per- Tacrolimus (TAC) should be started on the second haps the greatest difference from adults arises from the postoperative day. It can be given orally, but its absorp- underlying primary diseases and the interaction with tion should be confirmed by an absorption profile. family members. Certain aspects of the postoperative Blood levels taken at intervals over 12 hours pro- social and psychological care also differ from those vide a guide to the efficiency of absorption. and 12 of adults. Schooling and adolescence are particularly hour trough levels of 8–12 ng/ml are satisfactory

307 Section 6: Other abdominal organs

Table 36.2 Peri-operative immunosuppression Agent Dose Start Stop Comments Methylprednisolone 500 mg IV 1 hour before lymphocyte- Single dose Important to cover possible depleting agent cytokine release from lymphocyte depletion Methylprednisolone 20 mg IV BID Day 1 Day 7 Taper steroids depending on clinical picture Methylprednisolone 10 mg IV BID Day 7 Day 14 Stop if no signs of rejection Chlorpheniramine 10 mg IV 1 hour before lymphocyte- Single dose Important to cover possible depleting agent cytokine release from lymphocyte depletion Anti-lymphocyte 30 mg SC Intraoperatively and day 1 Day 0 and 1 only SC route minimizes effects from antibody (Campath) cytokine release Tacrolimus Aim for trough levels Day 2 Continue long Well absorbed from stomach; of 8–12 ng/ml term watch for renal impairment Note: The choice of the agents used may vary between centers. BID: twice a day; IV: intravenously; SC: subcutaneously.

(Table 36.2). If the IV route is required, then an infu- Salt and water balance sion delivering 0.05 mg/kg over 24 hours is com- Large shifts of fluid and electrolytes are often encoun- menced, aiming for levels at 24 hours of approximately tered postoperatively. The reasons for this have yet to 123–150 ng/ml. If the oral dose fails to achieve a sat- be fully elucidated. In the first 2 weeks, sodium losses isfactory level, then IV TAC should be administered. can occur by as much as 1 Mole per day, predomi- When converting from IV to oral dosing, the dose ratio nantly via the kidneys and also from high stomal out- is usually considered to be 1:3. The target level should put. The latter usually occurs after the third postopera- be reduced in the face of hypoalbuminemia. Steroids tive day and continues for 2–3 weeks. Occasionally this should be commenced on day 1, with IV methylpred- is protracted and requires large doses of loperamide. nisolone 20 mg administered twice a day (BID) for Codeine is less well tolerated and should be avoided if 7 days, then reduced to 10 mg BID for 7 days, and possible. In resistant cases it is worth a trial of sepa- thenstoppedifnoevidenceofrejection.Mycopheno- rating dietary fluids from solids. Measuring fluid and late mofetil (1g BD Po) can be used as an adjunct to electrolyte losses accurately is vital, and appropriate tacrolimus when tolerated. replacement is one of the most challenging aspects of the procedure. Antibiotics Theseshouldbebroadspectrum,andthepatient’s Renal function microbiological history should be considered. Viro- Special consideration must be given to renal function. logical monitoring should include cytomegalovirus Even in the absence of a renal graft, it is common to (CMV) polymerase chain reaction/immunoglobulin encounter renal impairment. This tends to occur in Mweeklyorateachoutpatientappointmentfrom the second and third week, probably due to progres- month 2–9 or at suggestion of CMV symptoms. sive TAC nephrotoxicity. Caution should be exercised as patients often have very low serum creatinine lev- els secondary to low lean body mass, and the relative Prophylaxis against deep venous and change in creatinine should be noted. graft venous thrombosis Prophylaxisisgivenagainstdeepvenousthrombo- Nutrition sis and graft venous thrombosis with low-molecular- It is common practice to commence PN, through a weight heparin. In addition, aspirin is started on dis- dedicated line when possible, 24–48 hours postoper- charge or if platelet count rises above 500 × 109/L. atively. We find that a 24-hour period free from PN is Long-term anti-coagulation should be considered if useful to focus on regaining fluid and electrolyte bal- thrombophilia is found or suspected. ance after the operation. A Jejunal feeding tube should

308 Chapter 36: Intestinal transplantation be placed at the time of surgery. As soon as there are patient is discharged, in which case once a week reduc- signsofintestinalmotility(usuallyafterafewdays), ing to twice a month in months 2 and 3 is usually more jejunal feeding is started and built up gradually over realistic, but this should be tempered by the patient’s 10–14 days to full requirement if tolerated. This often ability to inform the center in the event of an alter- takes longer, and the transition from PN to jejunal ation in stomal output or other sign of deterioration. feeding must be tailored for each case. Likewise, the From months 4–6 inclusive, once per month is usu- introduction of oral feeding varies widely; sips can be ally appropriate and after this every 2 months for the started as soon as nasogastric drainage tails off, but first year and when appropriate according to clinical food is usually not commenced for 10–14 days to allow progress. Biopsies are taken from the distal small intes- theanastomosistoheal.Weoftenfindthatpatients tine via the stoma (at least 15 cm proximal to the stoma areslowtotakeoralfood;someofthisarisesfrom to avoid artefact, using small cup non-spiked forceps), the commonly encountered nausea, but it also seems and if present, regular colonic graft biopsies should be to be related to their preoperative feeding state in that taken in addition. In situations when there is clinical those who took very little or no food preoperatively suspicion of rejection but routine biopsies are normal, have a tendency to find eating more difficult postoper- additional biopsies from the proximal intestine should atively. As oral feeding is established, the jejunal feed- be studied if access to this region is possible. ing can be tailed off. This usually takes 6–8 weeks, and there is often need for a small amount of residual jeju- nal feeding. On occasion, stopping jejunal feeding is Diagnosis of intestinal graft rejection required to stimulate patients to take their nutritional Acute rejection of a functioning graft is invariably requirement orally. IV glutamine may be included associated with a change in its function. The stomal in the postoperative regime in order to facilitate output is usually increased and more watery in con- enterocyte recovery. sistency. In addition, there are often signs of an inflammatory response with fever and elevated C- Physiotherapy reactive protein, and serum albumin may fall. Imme- diate ileoscopy is required, and mucosal biopsies Early mobilization and chest physiotherapy are of key should be taken for urgent analysis by an experienced importance. Attention to the musculoskeletal system histopathologist. If there are ongoing signs despite a so that the complications associated with the pro- negative histological assessment for rejection, a clini- longed surgery and ITU care can be minimized should caldecisionshouldbemade,asitispossibleforrejec- be another focus. tiontoaffectonlyasegmentofsmallbowel.Other investigations can be helpful, such as white blood cell Postoperative complications scintigraphy, positron emission tomography scanning, and capsule endoscopy. Nevertheless, it is not uncom- Rejection of intestinal graft mon to encounter a situation where a clinical decision Between 50% and 75% of patients experience rejection must be made on the basis of probability and risk– in the first few postoperative months. This figure varies benefit assessment. In the majority of cases, a pulse widely partly because of the lack of consensus regard- of IV methylprednisolone 500 mg daily for 3 days will ing the definition of acute rejection and also because not cause harm and will often settle mild to moderate of the heterogeneity of the anti-rejection regimes used, rejection. The clinical response to this can be evaluated andthisiscommonestintheearlypostoperative and assists in formulating a view concerning the diag- period.Themediantimeforthefirstepisodeofacute nosis of rejection. Escalation to anti-rejection agents rejection to occur is about 2.5 weeks after surgery, and with more adverse side effect profiles such as anti- this underscores the need to take frequent biopsies to thymocyte globulin (ATG) and alemtuzumab, how- monitor rejection in the first month. Most acute rejec- ever, should only be considered when the evidence tion episodes occur in the first year, after which the for rejection is strong. Patients at high risk of CMV level of monitoring can be considerably reduced and infection are likely to develop severe symptomatic dis- eventually stopped. It is our practice to take surveil- ease if they receive antibody treatment for rejection. lancebiopsiesthreetimesaweekforthefirstmonth Pneumocystis carinii pneumonia prophylaxis should and twice a week during months 2 and 3, unless the always be provided, and if patients are given ATG, they

309 Table 36.3 Common infections following intestinal transplantation in the United Kingdom Likely Infection Stage Location Frequency pathogens Clinical features Diagnosis Treatment

Bacterial Immediate/ Central line–related, 29% Bacterial infections Staphylococcus Brisk deterioration/septic Blood cultures/ Initially broad-spectrum Early Ͼ mid Ͼ late Superficial surgical site, account for 69% of aureus (including shock/and organ-specific pneumococcal/ antibiotics, then adjust to 17% all infections, 60% of MRSA) features (respiratory, legionella urinary include sensitivities of Pneumonia, 14% these are caused by Escherichia coli urinary, intra-abdominal) antigens/ known organisms. Abdominal gram positive Klebsiella Lower grade sepsis with Organ-specific: BAL; Vancomycin if known collection/peritonitis, organisms. Pseudomonas coagulase-negative sputum; urine culture MRSA-positive or if 14% Coagulase-negative staphylococci Intra-abdominal scans, cross-infection occurring. staphylococci: IV radiologically guided (Use to cover possible line infections only percutaneous aspiration MRSA if infection severe) for accessible Take special note of previous intra-abdominal infections, particularly collections antibiotic-resistant organisms and current hospital-acquired pathogens. Remove infected IV lines; drain collections. Fungal Early < mid Aspergillosis (30%) Fungal and candidal Aspergillus fumigatus Antibiotic-resistant Chest CT can be Aspergillosis with Median time to first Wound, pulmonary, infections account pneumonia diagnostic, respiratory amphotericin B/AmBisome, fungal infection 181 disseminated, cerebral for 14% of all Aspergillosis serious, samples other than BAL voriconazole, or days infections. particularly disseminated, and trans-bronchial caspofungin and intra-cerebral is biopsy unlikely to be usually fatal. helpful. PCR investigation. Candidal Immediate/early Candidiasis (Ͼ60%) Candida albicans Antibiotic-resistant sepsis Blood and urine culture, AmBisome/Abelcet oropharyngeal, line tip culture Caspofungin genitourinary, Fluconazole if mild/known to wound-related, line be sensitive infections Line removal if infected Viral Early < mid Ͼ late CMV, look for colitis, Viral infections, 17% Influenza, RSV, or CMV Flu-like illness, pneumonia Nose and throat swab, Antivirals, depending on Median time to first viral hepatitis, and retinitis. 60–85% of donor parainfluenza-3 (Influenza, RSV, CMV, nasopharyngeal circulating strains infection 91 days EBV–PTLDlate population are CMV CMV parainfluenza) aspirate, BAL, PCR (parainfluenza-3) Late episodes usually positive. VZV Organ disease (CMV) Nebulized ribavirin recurrence Activity in recipient HSV-1/2 Severe Chicken pox or (parainfluenza-3) usually implies EBV zoster, pneumonitis (VZV) Ganciclovir (CMV) reactivation. Human herpes virus 6 Systemic infection, organ Acyclovir (VZV, Herpes simplex Adenovirus disease (HSV-1/2) 1/2) From glandular fever to Discuss with virologist and PTLD (EBV) hematologist (EBV) Systemic infection, fever Discuss with duty virologist (Human herpes virus 6) (Herpes virus-6) Systemic infection, organ Cidofovir (discuss with duty disease (adenovirus) virologist) (Adenovirus). Immediate: first postoperative week; early: 2 weeks to 2 months postoperative; mid term: 2 months to 12 months postoperative; late: 12 months onwards. BAL: bronchoalveolar lavage; CMV: cytomegalovirus; CT: computed tomography; EBV: Epstein-Barr virus; HSV: herpes simplex virus; MRSA: methicillin-resistant Staphylococcus aureus; PCR: polymerase chain reaction; PTLD: post-transplant lymphoproliferative disorder. Chapter 36: Intestinal transplantation should also receive full-dose intravenous ganciclovir 80% for the duration of that treatment. 60% ⇐1 yr >1 yr The histological changes of rejection are often 40% 20% subtle and can be misleading. Several particular 0% histopathological features of acute rejection have been % of Patients describedinpatientspretreatedwithlymphocyte- MOF Other Cardiac depleting antibodies, such as sparse neutrophil RejectionTechnical Cerebral ThrombosisLymphoma infiltration in the lamina propria leading to more Not Specified prominent eosinophilic infiltration and sometimes Figure 36.2 Causes of death following intestinal transplantation eosinophilic cryptitis. The presence of apoptotic (adults and children). MOF: multi-organ failure. Adapted from the bodies in the crypt bases is perhaps the most specific International Registry, with permission: www.lhsc.on.ca/itr. feature but again must be semi-quantified and judged in the clinical context. taken into consideration. When IV lines appear clin- Acute vascular rejection is less common. When it ically infected, they must be removed and catheter occurs it can be rapid and overwhelming. It seems tips sent for culture. In the absence of an obvious to occur independently from acute cellular rejection source of infection, IV lines need to be removed when and is correlated with human leukocyte antigen (HLA) possible (Table 36.3). Broad spectrum antimicrobial panel-reactive antibodies (PRAs) and positive T/B-cell agents should be commenced according to microbiol- cross-matches. ogist advice and protocol. Tissue typing The postoperative period following HLA matching does not seem to reduce the incidence discharge from hospital of acute rejection. The presence of donor-specific anti- bodies increases the risk of acute rejection, and PRA The postoperative mortality rate remains relatively correlates with the incidence of acute rejection of the steep for about 5 years and then begins to plateau. intestinal graft. Therefore, recipients are screened for This emphasizes the importance of close monitoring PRAs in the preoperative assessment and during their of patients for at least 5 years and the need for rapid time on the waiting list. It is also often useful to check intervention for any deterioration in health while they PRAs in the postoperative period to assist in the diag- remain in this relatively high-risk period. It has been nosis of rejection. our experience that patients continue to have prob- lems after 5 years, requiring occasional hospitalization for short periods about once a year, usually for tran- Treatment of systemic sepsis sient sepsis or dehydration. It is essential to have a well- Patients are heavily immunosuppressed such that organized follow-up schedule that allows detection of onsetofsystemicsepsisisamedicalemergency.This problemsatanearlystage.Forpatientswhocannotget may be the first sign of rejection as a consequence of back to the center quickly, a shared care policy with translocation of bacteria. Consultation with a micro- local primary and secondary care providers is impor- biologist who is familiar with the clinical scenario is tant. Psychological issues tend to come to the forefront essential. Cultures should include blood, urine, and as the physical problems occur less frequently. stomal effluent; swabs of throat, wounds, or IV sites; and drain fluid. Viral studies for Epstein-Barr virus, CMV, herpes simplex virus, and adenovirus should Causes of postoperative death be undertaken and the advice of a virologist sought. In the early days of intestinal transplantation, the main A small intestinal biopsy should also be performed cause of death was acute cellular rejection followed incasesofsystemicsepsisofpossiblegraftorigin, rapidly by sepsis and multi-organ failure (Figure 36.2). such as concomitant increased ileostomy output, but Since the advent of lymphocyte-depleting induction thereshouldbealowthresholdforthisinanycase. immunosuppression, this has largely been replaced by Patients with chest signs should be considered for sepsis and multi-organ failure without acute cellular bronchoalveolar lavage. Organisms previously identi- rejection as the initiating event. The use of single- fied from patient specimens should be reviewed and or double-agent maintenance immunosuppression has

311 Section 6: Other abdominal organs

reduced the incidence of severe postoperative sepsis, out from transplantation the better, given that the and this has partly accounted for the improvement influences of lymphocyte depletion often last for a year seen in survival over the last 10 years. or more. If patients are taking SRL, they may need conversion to TAC prior to the procedure and remain Special considerations on this until healing is complete. It is also of critical importance to fully brief the surgeon and staff under- Pregnancy taking any intercurrent surgery and if possible have the patient regularly reviewed by a transplant clinician. Younger patients frequently want to consider repro- duction soon after the transplant procedure. There is some experience in the literature of intestinal and Employment multi-visceral patients successfully conceiving and Quality of life following transplantation is of great delivering healthy children. One of the main consid- importance. In addition to being a primary aim of the erations is the potential teratogenicity of immuno- transplantation, we consider it to be a key factor in suppression as discussed in Chapter 31. Mycopheno- the long-term survival of patients. Many patients feel late mofetil and sirolimus (SRL) should probably be a lack of self-worth and purpose if they cannot gain avoided. Again, it is important to involve a high-risk employment. This is compounded by the tendency for pregnancy service well in advance of this scenario so employers to avoid them because of the perception that that preparations can be made to appropriately man- they are unreliable due to their health status. age these patients who may have reproduction as their ultimate aim. Further reading Grant D. Current results of intestinal transplantation. The Travel International Intestinal Transplant Registry. Lancet 1996; 347: 1801–3. Patients who are vaccinated prior to transplantation Lloyd DA, Vega R, Bassett P, Forbes A, Gabe SM. Survival have a better chance of remaining immune than those and dependence on home parenteral nutrition: vaccinated after the procedure while on immunosup- experience over a 25-year period in a UK referral centre. pression. In general it is wise to encourage patients Aliment Pharmacol Ther 2006; 24: 1231–40. nottotraveltoregionswherethemedicalcareis Middleton SJ, Nishida S, Tzakis A, et al. Cambridge-Miami limitedand/orriskofinfectionhigh.Intercurrent score for intestinal transplantation preoperative risk infection may provoke rejection, and differentiating assessment: initial development and validation. between gastrointestinal infection and rejection can be Transplant Proc 2010; 42: 19–21. difficult. Middleton SJ, Pollard S, Friend PJ, et al.Adultsmall intestinal transplantation in England and Wales. Intercurrent surgery Br J Surg 2003; 90: 723–7. Todo S, Reyes J, Furukawa H, et al.Outcomeanalysisof71 The main consideration here is the increased risk of clinical intestinal transplantations. Ann Surg 1995; 222: infection because of immunosuppression. The further 270–80.

312 Section 7 Other Chapter Composite tissue allotransplantation Face transplantation 37 Maria Siemionow and Can Ozturk

Key points have an ideal color and texture match with the orig- r In recent years, face transplantation has inal missing tissue; however, their application applies become a clinical reality and in the future to only small or moderate defects. Prelamination and may become a standard procedure. prefabrication of flaps is used in reconstructing more complex facial tissue defects such as nasal lining and r Face transplantation is still classified as eye-socket defects. Expanded skin flaps are frequently “experimental” since there are not yet used in facial reconstructive surgery to provide identi- sufficient scientific data regarding long-term cal color and texture. Functional and aesthetic results outcome. are better when compared with those of other recon- r To date there are two scalp and 14 facial structive options. They are ideal for covering moder- allograft cases reported. ate skin and soft tissue defects if the adjacent tissues are not compromised. Microsurgical free tissue trans- The face and scalp are aesthetically and functionally fers are the technique of choice not only for recon- very important parts of the human body. Burn injuries, struction of extensive composite tissue defects, but also gunshot wounds, animal attacks, and extensive tumor for repairing small but technically challenging parts of surgery are the primary causes of traumatic facial the face such as oral and nasal lining. Walton showed deformities. These deformities may present with sin- successful results of functional restoration of nasal air- gle or composite tissue defects involving skin, mus- way and nasal aesthetic units by using various artis- cle, and/or bone, as well as different functional and tic combinations of cartilage grafts, forehead flaps, and aesthetic units. The face consists of six major aes- prefabricated radial forearm free flaps. The scapular/ thetic units, comprised of forehead, eye/eyebrow, nose, parascapular, serratus, latissimus dorsi, anterolateral lips, chin, and cheek. Primary closure, skin grafting, thigh, rectus abdominis, and fibula flaps are the other local flaps, tissue prelamination and prefabrication, free flap options commonly used in facial and scalp distant flaps, tissue expansion, and free tissue trans- reconstruction. There are various cases of scalp and fers are traditional reconstructive choices for covering partial face replantation published in the literature. facial defects. There are currently few reports of total Replantation provides the best functional and aes- face reconstruction using a full-thickness skin grafting thetic outcome because the defect is reconstructed technique. with original missing tissue. However, this is only pos- sible in selected cases. Conventional surgery Skin grafting is not an ideal reconstruction method for Face transplantation extensive facial defects because of the relatively poor Face allotransplantation is an alternative option for color and texture match, poor functional outcome, and complex facial defects that cannot be reconstructed secondary deformities such as ectropion and microsto- with the above procedures. Composite tissue allotrans- mia due to graft contraction. There are many different plantation (CTA) is a new developing field of mod- kinds of local flaps used in covering specialized parts ern plastic and reconstructive surgery. The “compos- of the face, such as nose, cheek, and lips. These flaps ite tissue” is a combination of different types of tissues

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

313 Section 7: Other

such as skin, fat, muscle, nerve, and bone that origi- Table 37.1 Sequence of procedures in mock facial nate from ectoderm and mesoderm. The complexity transplantation of these grafts may challenge immune responses in the 1. Transfer of the donor facial flap into the recipient’s facial donor’s body, making graft acceptance and tolerance defect more difficult to achieve when compared with solid 2. Coaptation of the supraorbital, infraorbital, and mental organ transplants. nerves Experimental studies on CTA began with the rat 3. Anchoring of the flap at the region of the mandibular and zygomatic ligaments hind-limb allograft models in the 1960s. Following development of successful solid organ transplanta- 4. Anchoring of the flap to the preauricular region, mastoid fascia, and temporal fascia tion in clinical practice, the first clinical CTA was a hand transplant, which was performed in Ecuador 5. Anchoring of the flap to the frontal bones in 1964. Despite immunosuppressive therapy includ- 6. Closure of the upper and lower gingivobuccal incisions ing steroids and azathioprine, the hand was ampu- 7. Closure of the upper and lower conjunctival incisions tated after 2 weeks due to rejection. In Septem- 8. Anastomoses of the external carotid arteries ber 1998, Dubernard and colleagues performed the 9. Anastomoses of the facial veins world’s first successful hand allotransplant in Lyon, 10. Coaptation of the great auricular nerves France. The immunosuppression protocol included anti-thymocyte globulin (ATG), tacrolimus (TAC), 11. Anastomoses of the external jugular veins mycophenolate mofetil (MMF), and prednisolone. To 12. Closure of the skin incisions date more than 50 forearm and hand transplants have been performed worldwide. In 2008, Machens and Hohnke¨ in Munich, Germany, performed the first whole arm allograft transplant. Cavadas and col- Preparation for face transplantation leagues in Valencia, Spain, have also proposed leg allo- graft transplantation, although to date only syngeneic Patient selection process transplantation has occurred. The successful results of The patient selection process must comply with the hand and limb transplantation made face transplanta- institutional review board–approved protocol. The tion feasible. optimal candidates for face transplantation should be patients between 18 and 60 years of age with mini- mal coexisting medical illness who agree to comply Cadaver studies with lifelong immunosuppression therapy and show a A series of cadaver dissections were performed in strong desire to proceed with face transplantation. preparation for face transplantation. The sequence of Potential candidates should be evaluated by a procedures in mock facial transplantation is described multi-disciplinary face transplant team, which may in Table 37.1. include plastic surgeons; transplant surgeons;, ear, nose and throat (ENT) surgeons; immunologist; psychiatrist; infectious disease specialist; social Computer modeling and identity worker; and bioethics specialist. Physical evalu- ation includes recording of facial tissue defects, issues current functional status of the facial muscles, and Biomodeling has come into clinical use with the sensory deficits. Screening for epstein-barr virus, stereolithography apparatus, a computer-controlled cytomegalovirus, human immunodeficiency virus, technique that builds anatomically accurate skeletal andhepatitisvirusesisalsoperformed.Tissue models from radiological data. Using computer-based typing and detection of human leukocyte antigen models, the face looks neither like the donor nor the (HLA) antibodies is determined as for solid organ recipient prior to injury, but carries more of the char- transplantation. acteristics of the recipient skeleton than of the donor Imaging is required to analyze the details of the soft tissues. However, an unrealistic presentation of the facial defect and determine necessary structures for final appearance of the face is the main disadvantage of allotransplantation. This includes computed tomogra- these programs. phy (CT; head, neck, chest, and abdomen), magnetic

314 Chapter 37: Composite tissue allotransplantation resonance imaging (MRI; to determine the soft tissue ofimmunologicalresponseofthegrafts,aimingto defect and abnormalities) and CT angiography (ves- diminish damage to the facial allotransplant by skin sel mapping) of the head and neck, carotid Doppler biopsies. studies, electromyogram, and nerve conduction stud- A description follows of the fourth facial allotrans- ies. After clinical review of the physical and psycho- plant in Cleveland, Ohio, in 2008. The patient was a logical condition of potential candidates, patients who 45-year-old woman who had a midface defect after a are accepted for transplant give informed consent. A shotgun injury in 2004. Two months after the trauma, thorough explanation of potential risks and possible the patient was transferred to The Cleveland Clinic and outcomes of the procedure should be given, includ- had 23 autologous reconstructive operations prior to ing potential complications of the surgery (i.e., graft the facial allotransplantation procedure. None of these failure, infection, death), lifelong immunosuppression autologous reconstructions provided adequate func- and related complications, acute/chronic graft failure/ tional or aesthetic results. Her facial defect included rejection, graft-versus-host disease, possible further absence of a nose with underlying maxillary bone, loss salvage or revision surgeries, and the importance of of orbicularis oculi and orbicularis oris muscle func- compliance and rehabilitation. tion, scarred lower eyelids with ectropion, right-eye enucleation, and facial nerve deficit (Figure 37.1).She could not eat solid food or drink from a cup and her Clinical cases of face transplantation speech was slurred due to palatal damage. She required Todatetherehavebeentwoscalptransplantsand14 a tracheostomy for ventilation and a percutaneous gas- facial allotransplantation cases reported in the litera- trostomy tube for nutrition. The donor was a woman ture and in media (Table 37.2). In 1982, Buncke and matching the patient in age, race, and skin complex- colleague performed the first scalp isotransplantation ion. There was mismatch of both blood group (donor between identical twins (same blood group and HLA group A, recipient group AB) and HLA tissue type tissue type). The patient, who had lost 60% of her scalp (two of six matches). After identifying recipient ves- tissue due to avulsion trauma, was treated with split- sels, the soft tissues, bones, and hardware from pre- thickness skin grafts taken from her identical twin. vious surgeries were removed. The template of the Later, the patient was treated with two free scalp flaps recipient’s facial defect was outlined based on three- from her identical twin. No immunosuppressive agent dimensional graft architecture. The CTA was included was given. The second scalp transplantation case in the over 535 cm2 (80% of donor face) of facial skin literature was performed by Jiang and colleagues and included Le Fort III midfacial skeleton, infraor- in 2005. This 72-year-old male patient had stage bital floor, bilateral zygoma, lower eyelids, nose, ante- IIIC recurrent cutaneous malignant melanoma of rior maxilla with teeth, anterior hard palate, alveo- the scalp vertex. After wide excision of the tumor, lus, upper lip, parotid glands, and intraoral mucosa. the defect was reconstructed with a CTA including After fixation of bone components, microvascular end- cephalo-cervical skin flap and bilateral ears that were to-end anastomosis of the bilateral facial arteries and taken from a brain-dead male donor. The patient veins was accomplished with an ischemic time of 2 received the combination of MMF, prednisone, and hoursand40minutesandtotaloperationtimeof daclizumab as immunosuppressive therapy. Accord- 22 hours. After 2 months she was discharged from ing to the authors, there were no signs of rejec- hospital (Figure 37.2). She had revision surgery to tion and no tumor recurrence during a 4-month remove excess tissue 18 months after transplant. Reha- follow-up. bilitation consisted of passive and active facial mus- In November 2005, Dubernard and colleagues per- cle exercises, speech therapy, sensory re-education, formed the first partial facial allotransplant in Amiens, and facial acceptance re-education, which began 48 France. The recipient was a 38-year-old woman with hours after transplantation. The sensory recovery of amyocutaneousdefectofthecentralfaceareadueto the graft was confirmed by presence of 7-mm two- a dog bite. The donor was a 46-year-old with identi- point discrimination at 8 months; however, motor cal blood group and five of six HLA antigen matches recovery improved over 1 year post-transplant. Finally, with the recipient. A radial forearm flap from the the patient gained missing facial functions such donor was transferred with facial CTA to the left as nasal breathing, smelling, drinking, eating, and submammary fold of the recipient for monitoring speaking intelligibly. She is satisfied with her facial

315 Section 7: Other

Table 37.2 Anatomical and surgical details of facial transplant patients reported to date Sensory and Patient CTA design Vessel and nerve repair functional recovery

Patient 1, Upper and lower lips, perioral Bilateral facial a. and v. (end to end). Light touch (10 weeks) 38 y, F – 2005 muscles. Bilateral infraorbital and mental Temperature (6 months) Amiens, France Partial nose with alar and triangular nerve. Eat and drink (1 week) cartilages, anterior septum. Cheek, Left mandibular branch (end to end). Labial contact (6 months) partial oral, and nasal mucosa. Chin and nose pyramidal muscle motion (12 months) Smile (14–18 months) Patient 2, Upper lip, nose, the right anterior Right external maxillary a. and Light touch (3 months) 30 y, M – 2006 maxilla, sinus. anterior facial v. (end to end). Temperature (8 months) Xian, China Right zygoma with lateral orbital Right facial nerve not well coapted. Unable to smile completely. wall, right parotid gland, and Facial nerve not functional. partial masseter. Eat, drink, and talk normally. Patient 3, Lower two-thirds of face, including Bilateral external carotid Light touch (3 months) 29 y, M – 2007 skin, soft tissue, lips, chin, cheeks, arteries and thyrolingofacial Temperature (3 months) Paris, France nose, bilateral parotid glands, trunks (end to end). Bilateral facial Eat and speak (10 days) parotid ducts, and intraoral and infraorbital nerves. Orbicularis oris and oculi voluntarily mucosa. contract (6 months) Spontaneous mimicry (9 months) Trigeminal and facial motor function (12 months) Patient 4, Le Fort III midfacial skeleton, total Bilateral common facial arteries, left Smell (2 days) 45 y, F – 2008 infraorbital floor, bilateral zygoma, external jugular, posterior facial, Light touch (5 months) Cleveland, OH lower eyelids, nose, anterior and right facial veins (end to end). Temperature (5 months) maxilla with teeth, anterior hard Bilateral facial nerve repair Upper lip occlusion palate, alveolus, upper lip, parotid interpositional nerve grafts. Facial mimicry glands, and intaroral mucosa. Eats and drinks from a cup. Speaks more clearly. Patient 5, Cheek, nose, parotid glands, and Bilateral external carotid a., external None reported 28 y, M – 2009 ducts. juguler vein and thyrolinguofacial Paris, France Upperandlowerlips. trunk. Premaxilla, chin and intraoral Bilateral facial nerves. mucosa. Patient 6, Upper two-thirds of face and entire Bilateral external carotid arteries and None reported 37 y, M – 2009 scalp with bilateral ears, forehead, thyrolinguofacial trunks (end to Paris, France lower eyelids, nose, bilateral end). cheeks, bilateral parotid glands Bilateral facial, supraorbital, and and parotid ducts, intraoral infraorbital nerves. mucosa. Patient 7, Midfacial tissue including skin, soft Left external carotid, right facial a. None reported 59 y, M – 2009 tissue, nose, upper lip, bilateral (end to end), bilateral facial vein. Boston, MA cheek, bilateral parotid ducts, Bilateral facial and infraorbital nerves. maxilla with teeth and palate. CTA = composite tissue allotransplantation; F: female; M: male; y: years (age).

appearance, and has regained self-confidence and tion to Doppler monitoring and tissue oxygenation social reintegration. measurements. Standard blood tests for monitoring immunosuppression including renal function should Care after face transplantation also be performed. Monitoring Immunosuppression Inspection for facial allograft viability and observation Currently, there is no standard immunosuppressive for signs of rejection such as erythema, edema, and protocol. A number of different induction and main- blistering should be done on a daily basis, in addi- tenance therapy protocols and different viral and

316 Chapter 37: Composite tissue allotransplantation

Figure 37.1 Patient prior to face transplant performed at Cleveland Clinic. Reproduced with permission from Siemionow MZ, Papay F, Djohan R, et al., First U.S. near-total human face transplantation: A paradigm shift for massive complex injuries. Plast Reconstr Surg 2010; 125: 111–22.

anti-microbial prophylaxis regimens have been 72 hours for the first 2 weeks, then weekly for the first reported. The most common regimen for face 3 months, twice a month during the first 6 months, transplantation is based on kidney transplantation and once a month thereafter during the first post- protocols and includes standard induction therapy transplant year. The oral mucosa of the facial allo- withATGfollowedbytriplemaintenancetherapy. graft may present with higher grades of graft rejections This maintenance therapy consists of TAC, MMF, when compared with the skin component. Another and prednisolone. Additionally, the donor-derived method for monitoring graft rejection episodes is bone marrow cell infusions aiming for chimerism and application of sentinel allografts such as radial fore- tolerance induction have been attempted. arm flap or skin graft from the face allograft donor. The presence of edema, skin blistering, scaling, erythema, hair weakness or loss, dermatitis, and papules of the Biopsy protocol allograft should alert the clinician for possible graft The protocol for monitoring facial allograft rejection rejection. In these circumstances, the patient should be includes punch biopsies of skin and oral mucosa every admitted and biopsies taken from the skin and mucosa

317 Section 7: Other

Figure 37.2 Patient after face transplant performed at Cleveland Clinic. Reproduced with permission from Siemionow MZ, Papay F, Djohan R, et al., First U.S. near-total human face transplantation: A paradigm shift for massive complex injuries. Plast Reconstr Surg 2010; 125: 111–22.

of the graft. The biopsy specimens must be reviewed detection and reversal of acute graft rejection by an experienced pathologist. A well-defined scale episodes. such as the Banff classification, developed for compos- ite tissue allografts, should be used for grading of acute rejection. Functional evaluation and physical Treatment for acute rejection depends on the clinical condition of each particular patient. The rehabilitation dose of immunosuppressants may be altered, a The comparison of functional outcomes for face trans- bolus of steroid given, or alternatively extracorpo- plant patients is difficult because each particular case real photo chemotherapy may be tried. Chronic presents with different facial defects and functional rejection of the facial allotransplant has not yet deficits. In addition, in contrast to aesthetic outcome, been reported. This may be due to be relatively which can be seen immediately after transplantation, short follow-up of face transplant patients or early thefinalfunctionaloutcomedependsonadequate

318 Chapter 37: Composite tissue allotransplantation motor recovery of facial nerves and muscles, which rently carries a 14% death rate, as two of 14 patients may take months or even years. have died (one due to non-compliance with immuno- Professional rehabilitation teams should start suppression and one due to infection). Because of facial re-education therapy and speech therapy in the complexity of this procedure and the uncertain early post-transplant phase. Progression is mea- benefits and risks, it is important to carefully screen sured by evaluation of facial muscle function (e.g., potential candidates. Further evaluation of outcomes mimicry, smile, blinking, eyelid closing, nasal flaring, is required, along with cooperation between different puckering), smell, swallowing, mastication, and institutions performing these novel procedures. speech. Functional MRI, electromyography studies, and volumetric analysis are objective measures of motor recovery of facial units, whereas tem- Further reading perature testing, two-point discrimination test, Cendales LC, Kanitakis J, Schneeberger S, et al. The Banff pressure-specified sensory device testing, and 2007 working classification of skin-containing Semmes-Weinstein monofilament tests are used to composite tissue allograft pathology. Am J Transplant monitorthesensoryrecoveryofthefacialallograft. 2008;8: 1396–400. According to the current literature, the motor recov- Dubernard JM, Lengele B, Morelon E, et al. Outcomes ery of facial nerves has been observed between 9 and 18 months after the first human partial face 12 months post-transplant. In contrast, recovery of transplantation. NEnglJMed2007; 357: 2451–60. sensory nerves has been seen within 3–6 months. Guo S, Han Y, Zhang X, et al. Human facial Additionally, serial photographic and video documen- allotransplantation: a 2-year follow-up study. Lancet tation is important for the comparative assessment of 2008; 372: 631–8. post-transplant function. Siemionow M, Agaoglu G, Unal S. Mock facial allotransplantation: a cadaver study in facial allograft transplantation in humans – part II. Plast Reconstr Surg Ethical considerations 2006; 117: 876–85. Due to the lack of long-term follow-up and out- Siemionow M, Papay F, Alam D, et al. Near-total human comes, facial allotransplantation is still considered face transplantation for a severely disfigured patient in as an “experimental” procedure. This procedure cur- the USA. Lancet 2009; 374: 177–8.

319 Section 7 Other Chapter Hematopoietic stem cell transplantation

38 Charles Crawley and Thomas R. Spitzer

Key points As a rescue therapy, the infusion of HSCs can circumvent the myeloablative effects of many r Hematopoietic progenitor/stem cell conditioning regimens, thereby permitting steep transplantation remains an important escalation of doses of chemotherapy and radiation treatment strategy for patients with marrow therapy and overcoming the resistance that tumor failure syndromes, inherited hematological cells may have developed to conventional doses of disorders, and advanced hematological chemo/radiotherapy. As described next, multiple malignancies. myeloma and non-Hodgkin’s lymphoma are examples r Improvements in supportive care, the use of of hematologic malignancies for which myeloablative reduced-intensity conditioning regimens, conditioning and autologous stem cell transplantation and advances in human leukocyte antigen have improved survival outcomes compared with typing methodology have led to better conventional-dose chemotherapy. The replacement outcomes. aspect of stem cell transplantation relies on the r Cell sources have changed with an increasing engraftment of normal multi-potential progenitor utiliation of donors other than HLA matched cells in order to replace a diseased marrow or immune siblings. system. Many diseases, ranging from hematopoi- r It is likely that allogeneic transplantation and etic malignancies to acquired non-malignant blood particularly cord blood as a stem cell source dyscrasias such as severe aplastic anemia, to congeni- for transplantation will play an increasing taldisordersofthemarroworimmunesystemsuchas role in the next 10 years. sickle cell disease and severe combined immunodefi- ciency syndrome, have been cured by stem allogeneic cell transplants with the restoration of normal Hematopoietic stem cell hematopoiesis and successful immune reconstitu- tion. More recently, reduced-intensity conditioning transplantation: principles and regimens, based on elegant preclinical models that rationale demonstrated that full hematopoietic chimerism can Hematopoietic stem cell (HSC) transplantation has be reliably established as an immunological platform expanded dramatically in its scope and range of for adoptive cellular immunotherapy using donor indications over the past four decades. With the estab- lymphocyte infusions (DLI), have shown considerable lishment of modern histocompatibility typing and potential as a strategy for reducing transplant-related improvements in graft-versus-host disease (GvHD) complications and capturing potent graft versus prophylaxis and supportive care, many patients malignancy effects of the transplant. The graft versus without a human leukocyte antigen (HLA)-matched malignancy effect of this strategy is usually dependent related donor are now able to receive a transplant, and on the the establishment of full donor hematopoiesis, less intensive preparative strategies have allowed older either spontaneously or after DLI. Mixed chimerism patients and patients with significant comorbidity to alone may be sufficient for specific tolerance induction undergo transplantation. for organ transplantation and for the correction of

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

320 Chapter 38: Hematopoietic stem cell transplantation certain non-malignant disorders of the marrow and kit, and CD133. They represent approximately 0.1% of immune system. The reduced intensity of the condi- marrow cells, but the population is heterogeneous and tioning regimen has greatly expanded the application may include c-kit and CD133-negative cells. Despite of stem cell transplantation by permitting transplants this, for practical purposes, CD34+ and CD38– in patients who, by virtue of age or comorbidity, would define a cell population of human progenitor cells not have previously been considered candidates for (HPCs) that will re-establish hematopoiesis in the sub- stem cell transplantation. The induction of mixed lethally irradiated non-obese diabetic (NOD)/severe chimerism has recently been shown to be an impor- combined immunodeficiency (SCID) mouse and in tant strategy for the induction of specific tolerance humans. for solid organ transplantation. Sustained tolerance and anti-tumor responses have been observed in patients with multiple myeloma and end-stage renal Donors disease who received an HLA-matched combined The majority of HPC transplants undertaken world- bone marrow and kidney transplant. This strategy has wide use autologous cells, meaning the patient acts as also been successfully performed in patients without his or her own donor. These cells are readily avail- an underlying malignancy who received an HLA- able and do not carry the same risks of GvHD or the haploidentical bone marrow and kidney transplant. risks of transmission of infection as allogeneic cells. The demonstration of successful tolerance induction Autologous transplantation allows for the delivery of has great potential for avoiding the complications of high-dose chemotherapy where hematological toxicity long-term immunosuppression and for reconstituting would be dose-limiting. The use of autologous HPCs normal immunity after transplant. These successful allows for marrow recovery and allows chemotherapy advances in stem cell transplantation have paved the doses to be escalated to the next dose-limiting toxicity, way for future cell-based therapies involving either which is usually gastrointestinal or lung. The benefit engineering of the stem cell graft or modulation of the in terms of disease control relies on a beneficial dose cellular environment post-transplant to optimize the response curve. The lack of a GvHD effect may also be anti-tumor potential of the transplant while avoiding a drawback. In addition, there is a risk of tumor con- the deleterious effects of GvHD. tamination in the graft with an increased risk of dis- ease relapse. This effect is seen most clearly in the lower Hematopoietic stem cells: definition relapse rate when syngeneic (identical twin) donors are used. The opportunity to use a syngeneic donor is rare, and sources but these donors are analogous to the use of autolo- Stem cells are defined as a population of cells that gous cells with the certainty that the graft is tumor- retain the capacity for indefinite self-renewal and to free. Allogeneic HPCs are cells donated from a third generate progeny with the potential to differentiate party. Allogeneic donors (with the exception of syn- into mature cells with a variety of functions. The ther- geneic donors) carry the risk of causing GvHD, and apeutic use of stem cells is frequently viewed as a very this is strongly linked to HLA match; hence the pre- recent development. The use of HSCs, however, is not ferred choice is usually an HLA-matched sibling. The new, as these cells have been in regular and increasing majority of patients, however, do not have an HLA- clinicaluseforthelast50years;indeedtheyremainthe matched sibling, and there is an increasing reliance only adult somatic stem cell in routine use in 2011. on volunteer-unrelated donors. Bone Marrow Donors In adults, the only source of HSCs is the bone mar- Worldwide is the central database to which most donor row. This is not true of the fetus, where HSCs may be registries submit their data. It provides a searchable found in the liver, spleen, and, more importantly, cord database of over 14 million donors, including over blood. Identification of HSCs in mouse and subse- 400 000 cord blood units. Despite this, there remain quently humans was initially through assays for clono- patients for whom a donor cannot be found. One alter- genic precursors in the form of long-term culture- native in this situation is the use of haplo-identical initiating cells and colony-forming units. Subsequent donors. For almost all patients, a parent/child or sib- phenotypic markers have been defined. Human HSCs ling who has a 50% tissue type match is available. This arecharacterizedbytheabsenceoflineagemarkers requires the use of extensive T-cell depletion or post- (Lin–) and are negative for CD38 but express CD34, c- transplant strategies to deplete alloreactive T cells to

321 Section 7: Other

avoid GvHD. Recently, the importance of natural killer Table 38.1 Cell dose (NK) cell activation has become clear. Particularly in Nucleated haploidentical transplants, NK cell activation is regu- cells CD34+ cells T cells lated by a balance of activating inhibitory and killer Cell source (×108/kg) (×106/kg) (×107/kg) cell immunoglobulin-like receptors. The outcomes of transplantation with donors whose NK cells are allore- Bone marrow 2.0 2.8 2.2 active in a donor versus host direction are associated Peripheral blood 9.0 7.0 27.0 with substantially better outcomes. Cord blood 0.3 0.2 0.4

Sources of cells uous sterile closed circuit, an anticoagulant added, During embryogenesis, hematopoiesis originates with and mononuclear cells separated by centrifugation. primitive erythroblasts developing from mesodermal G-CSF mobilization of stem cells will typically take cells in the yolk sac. By 12 weeks gestation, the liver is 5–6 days from start of G-CSF to completion of aphere- the major hematopoietic organ and subsequently the sis. If chemotherapy is used, then apheresis is usually spleen. Hematopoiesis starts within the bone marrow started as the blood count is recovering to a level of at 11 weeks of gestation and has become the major site approximately 1.0–3.0 × 109/l, and the whole mobi- ofhematopoiesisfrom20weeks;bybirthitisessen- lization procedure is approximately 10–14 days. In tially the sole site. Under normal conditions, HPCs are addition to ease of collection, peripheral blood col- relatively rare cells in the peripheral blood (approx- lections result in higher cell doses and more rapid imately 100 per ml); however, much higher concen- recovery of blood counts compared with marrow col- trations are found in the umbilical cord or in adults lection; consequently, for autologous transplantation, associated with factors driving proliferation with the peripheral blood is the overwhelming source of cells. marrow, such as recovery from chemotherapy or after In Europe in 2008, bone marrow was used in less than exposure to cytokines. 0.01% of autologous transplants. In practical terms, there are three principal sources Cord blood is an extremely rich source of stem of HPCs for therapeutic use. Marrow may be har- cells, which, until recently, was simply discarded with vested under general anesthesia in 5–10 ml aliquots the placenta after birth. Umbilical cord blood is either from the posterior iliac crests. In general 15–20 ml/kg collected with the placenta still in utero by an obstetri- are collected. The advantages are that cytokines or cian or midwife or after delivery of the placenta. The chemotherapy are avoided, which is a consideration former results in higher cell doses and a lower risk of for allogeneic donors; however, collection volumes bacterial contamination but is clearly more disruptive are limited and back pain and anemia are com- to delivery. Cord blood HPCs have a high proliferate mon. Peripheral blood collection is easier and does potential, low immunogenicity, and low risk of trans- not require a general anesthetic. HPCs are mobilized missionofinfectionandthistoadegreeoffsetsthedis- from the marrow with cytokines, usually granulo- advantage of low cell doses (Table 38.1). Cord blood cyte colony-stimulating factor (G-CSF) with or with- by its nature is used exclusively in allogeneic trans- out myelosuppressive chemotherapy. In general, the plants, and with the establishment of cord blood banks addition of chemotherapy increases the HPC yield, (currently 31 registered with Netcord), these provide although it also increases toxicity. One additional ben- a source of cryopreserved stem cells that are rapidly efitofchemotherapyisthatitmayalsoreducetumor available, avoiding the need for prolonged unrelated burden further; consequently, the choice of agent will donor searches. be to a degree disease-specific. Typical drugs used The first cord blood transplant was performed in include cyclophosphamide (CY) at doses of 1.5–4 1988 for a child with Fanconi’s anemia and resulted g/m2 or etoposide. Cells can also be mobilized with in full hematopoietic and immune reconstitution; the CXCR4 antagonists such as plerixafor, which trig- patient remains well some 15 years later. However, cell gers proliferation of CD34 stem cells and their release dose remains important and has delayed the use of from their niche, the marrow microenvironment, into cord blood in adults. A minimum cell dose of 3 × theperipheralblood.Cellsarecollectedbyleuka- 107/kg is generally recommended, and frequently two pheresis: essentially blood is collected on a contin- cord units are combined in adults.

322 Chapter 38: Hematopoietic stem cell transplantation

Principles of conditioning important factor in preventing disease relapse after allogeneic transplant. Indeed, this is probably the Conditioning serves two key purposes during stem most important mechanism by which allogeneic trans- cell transplant. First, it is important in eradicating plant can cure hematological malignancies that are the underlying disease, which in most cases is a incurable with conventional chemotherapy. Given that hematological malignancy. Indeed, for autologous much of the toxicity and non-relapse mortality is (and syngeneic) transplants, it is the sole mechanism of linked to intensive conditioning regimens, the focus disease control. For allogeneic transplants, there is the of studies in the 1990s has been to develop condition- additional requirement to allow for effective engraft- ing regimens that allow for cell engraftment with min- ment of the donor cells. The initial concept, although imal drug or radiation doses. The majority of these probably incorrect, was that host HSCs needed to be reduced-intensity conditioning (RIC) regimens focus eradicated from the marrow stroma to “create space” on immunosuppressive rather than myelosuppressive for the donor cells to engraft. A third function of drugs. Some of the seminal work was conducted by pretransplant conditioning regimens, which is more Storb and colleagues in Seattle. Using a dog model, important, is to suppress the host immune system suf- they demonstrated that full donor engraftment could ficiently to prevent a host versus graft rejection of the be obtained using a TBI dose of 2.0 Gy rather than the donorcells.Thedegreetowhichthehostimmunesys- conventional dose of 14.4 Gy. A wide range of RIC reg- tem needs to be suppressed depends on factors such as imens are now in clinical use, with an ever-increasing the degree of HLA mismatch, prior sensitization, stem proportion of transplants being undertaken using this cell dose, and donor T-cell dose. type of conditioning (Figure 38.1). Although in many Historical conditioning regimes developed from cases there is no clear evidence of improved sur- the observation that radiation could eradicate vival, there is a clear advantage in terms of reduction leukemic cell populations at doses that did not result in non-relapse mortality (NRM). The increase in NRM in irreversible organ toxicity. Hence the use of con- seen with age had precluded the use of conventional trolled total body irradiation (TBI) was developed. transplants beyond the age of 50 years, whereas RIC Subsequently CY was added, with reduction in has allowed the use of allogeneic transplants to be leukemia or disease relapse, and this combination extended into patients in their seventh decade. remains in routine use. Throughout the 1970s and 1980s, studies focused on attempts to find alterna- tives to radiotherapy and dose escalation; studies Early complications determined maximum-tolerated doses for drugs such The most common early complications occur as a as busulfan, melphalan, etoposide, and cytarabine. consequence of the aplasia due to the conditioning Studies addressing strategies to increase dose intensity regimen and gastrointestinal toxicity. Aplasia devel- further, with the aim of reducing the risk of leukemia ops from approximately 5–7 days after the onset of relapse, demonstrate a fundamental problem, whereby the conditioning regimen and therefore often coin- increases in chemo/radiotherapy result in increasing cides at or after the time of cell infusion, customarily organ toxicity. As a result, in all these studies, reduc- referred to as day 0. Engraftment occurs around day tions in relapse were offset by increases in non-relapse 12–16 for autologous transplants and a little later for mortality and did not translate into improvements allogeneic transplants. The timing of white cell count in survival. There have been very limited phase III recovery is also dependent on cell source, being fastest studies of these traditional conditioning regimens, with peripheral blood, followed by marrow, and slow- with the exception of a number of studies comparing est with cord blood. The gastrointestinal tract is also outcomes of CY and TBI with CY and busulfan. These sensitive to chemotherapy. Nausea is an early com- show differences in toxicity but no real differences in plication, although it is relatively well controlled with survival, with the exception of acute lymphoblastic 5HT3 antagonists or some of the newer anti-emetics. leukemia, where CY/TBI appears superior. Mucositis develops from day 1–7 after transplant, usu- ally reaching a maximum by day 10–12. Risk factors Reduced-intensity conditioning include poor dental hygiene, allogeneic transplanta- It has been increasingly recognized that immune tion, and particularly the use of methotrexate after mechanisms, the graft versus malignancy effect, is an transplant as GvHD prophylaxis. Despite widespread

323 Section 7: Other

1800 Figure 38.1 Transplant activity in the United Kingdom, 2000–2009. Allo: allogeneic; Auto: 1600 autologous; RIC: reduced-intensity conditioning. Provided by Keiren Kirkland, British Society of 1400 Blood and Marrow Transplantation. 1200 1000 ALLO AUTO 800 RIC 600 400 200 0

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

use of interventions such as chlorhexidine or benzy- the mortality rate has fallen considerably. Other viral damine mouth washes, there is little evidence of ben- infections including adenovirus, Epstein Barr, and efit over bland mouth washes such as saline or bicar- respiratory viruses are increasingly recognized during bonate. Management is then focussed on analgesia and this period. From approximately 4 months, the pres- treatment of infection. Severe mucositis can result in a ence and severity of GvHD has a major influence on marked reduction in oral intake and even airway com- immune reconstitution. Immunoglobulin deficiency promise. Nasogastric tubes are often poorly tolerated, is common, with susceptibility to pneumococcus and and parenteral hydration and nutrition is often indi- other encapsulated bacteria. Other complications are cated. Lower gastrointestinal (GI) damage can result in listed in Tables 38.2 and 38.4. dyspepsiaandmorecommonlydiarrhea.Itisimpor- tant to exclude infective causes, but in their absence, Hemorrhagic cystitis management is with anti-diarrheals. Hemorrhagic cystitis is seen more frequently after allo- Infections geneic transplantation. BK virus also causes hemor- rhagic cystitis. It is dormant in urothelium but can Improvements in supportive care and particularly the reactivate after allogeneic transplant, particularly with management of infection have been an important fac- TBI-based conditioning regimens. Management of tor in improving outcomes over the last 20 years; how- this can be problematic but involves intensive platelet ever, infections remain a major cause of mortality. The supportandbladderirrigation.Theroleofantiviral types of infections vary with time after transplant (Fig- drugs such as cidofovir or leflunomide is unproven. ure 38.2). During the aplastic phase (up to week 4), Occasionally arterial embolization or even cystectomy bacterial infections, fungal infections, and pneumo- is necessary. Anti-fibrinolytic drugs are contraindi- nias are typical. Presentation is with neutropenic fever cated as they increase the risk of bladder and ureteric and the management is as for chemotherapy-induced clot formation. Acrolein is a metabolite of CY and ifos- neutropenic fever with the empiric use of broad- famide which is excreted in the urine and is toxic to spectrum antibiotics. After engraftment for autol- urothelium. At high doses it causes hemorrhagic cys- ogous transplants, the infection risk reduces very titis in up to 25% of patients. Mesna is an effective pre- rapidly; however, for allogeneic transplants, months vention, which binds and neutralizes acrolein in the 1–4 are associated with a very marked impairment bladder. in cell-mediated immunity. In the 1980s and 1990s, cytomegalovirus (CMV) disease was a major cause of mortality. CMV reactivation and infection remains a Veno-occlusive disease problem; however, with the routine practice of screen- Veno-occlusive disease (VOD) is a syndrome of hep- ing by polymerase chain reaction or p65 antigenemia atotoxicity due to the conditioning regimen. It usu- and the use of preemptive (or prophylactic) therapy, ally occurs within the first 3 weeks after transplant

324 Chapter 38: Hematopoietic stem cell transplantation

Table 38.2 Other early complications Complication Time of onset Risk factors

Capillary leak syndrome Weeks 0–3 Extensive prior chemotherapy; use of growth factors Engraftment syndrome Within 72 hours of Growth factors neutrophil engraftment Diffuse alveolar hemorrhage Week 0–3 Age Radiation Conditioning intensity Allogeneic donor Idiopathic pulmonary syndrome Week 2–4 Conditioning intensity Age Radiation GvHD Microangiopathic hemolytic anemia Around 2 months CNI Radiation Mismatched donors Developing GvHD infection

Phase I, Preengraftment, Phase II, Postengraftment, Phase III, Late phase, <30 days 30–100 days >100 days Impaired celluiar and Host immune Neutropenia, mucositis Impaired cellular immunity humoral immunity and system defect and acute graft-versus- and acute and chronic graft- chronic graft-versus- host disease versus-host disease host disease Device risk Central line

Allogeneic Respiratory and enteric viruses patients Herpes simplex virus*† Cytomegalovirus*† Varicella-zoster virus† Epstein-Barr virus lymphoproliferative disease† Facultative Gram-negative bacilli

Staphylococcus epidermidis Encapsulated bacteria Gastrointestinal tract Streptococl species (e.g., pneumococcus) All Candida species Aspergillus species Aspergillus species

Pneumocystis carinii Toxoplasma gondii Strongyloides stercoralis

0 30 100 360 Days after transplant *Without standard prophylaxis ≥ †Primarity among persons who are seropositive High incidence ( 0%) before transplant Low incidenoe (<10%) Episodic and endemic Continuous risk

Figure 38.2 Patterns of infection after hemopoietic stem cell transplantation. With permission from Centers for Disease Control and Prevention guidelines. and is characterized by jaundice, weight gain, ten- ferase/aspartate aminotransferase pretransplant) or der hepatomegaly, and refractory thrombocytopenia. the use of hepatotoxic drugs. Specifically implicated Risk factors include allogeneic transplantation, HLA drugs include gemtuzumab, busulfan, and CY. The match, stem cell source, and disease status. The inten- incidence of severe disease varies but is approximately sity of conditioning is also important, as is previ- 8% and 3% in allogeneic and autologous transplants, ous hepatic damage (elevated alanine aminotrans- respectively. Although there are diagnostic criteria

325 Section 7: Other

Table 38.3 Acute GvHD Organ Grading System – overall grade rejection a rare occurrence. GvHD, however, is one of Stage Skin/rash Bilirubin Diarrhea the commonest and most challenging complications arising after allogeneic stem cell transplant. It arises 1 Ͻ25% body surface 34–50 ␮mol/l 500–1000 ml due to immune recognition between host antigen- 2 25–50% body surface 51–102 ␮mol/ 1000–1500 ml presenting cells and mature donor T cells. It is subdi- 3 Generalized erythema 103–255 ␮mol/ Ͼ1500 ml vided into acute GvHD, which generally occurs within 4 Generalized Ͼ255 ␮mol/ Pain ± ileus the first 3 months, and chronic GvHD, which occurs erythema with bullae after 3 months, although it is categorized by the clinical and or desquamation features rather than the time of onset. The overall inci- Adapted from Przepiorka D, Weisdorf D, Martin P, et al.Consensus dence varies and depends on factors such as the HLA Conference on Acute GVHD Grading. Bone Marrow Transplant 1995; 15:825–8. match, sex mismatch (female donors for male recip- ients carrying a higher risk), prior alloimmunization of the donor, and source of stem cells. Recipient fac- (Seattle and Baltimore), the differential is wide. Pro- torsarealsoimportant,includingage,typeofcondi- phylactic ursodeoxycholic acid reduces the risk, but tioning, the use of T-cell depletion, and the prophylaxis the management of established VOD is difficult. Opti- strategy. mizing supportive care is crucial, and there are some data that the use of defibrotide reduces mortality even in critically ill patients. Acute GvHD Acute GvHD presents with skin rashes, typically mac- ulopapular and often starting on palms and soles, then Graft-versus-host disease developing into generalized erythema and desquama- The host hematopoietic system is sufficiently sup- tion (Table 38.3). GI symptoms start with anorexia pressed by pretransplant conditioning to make graft and nausea leading to diarrhea and abdominal pain.

Table 38.4 Late complications Complication Time of onset Risk factors Intervention

Cataracts 3–10 years TBI/steroids Surgery Keratoconjunctivitis sicca Upto20%at15years Radiation Topical lubricants Chronic GvHD Lacrimal plugs Female Topical steroid Chronic obstructive pulmonary Up to 20% Chronic GvHD Treatment of infection + GvHD disease/obliterative OB 2–14% Infection bronchiolitis (OB) Smoking Restrictive lung disease 3–6 months Radiation Fractionation of radiotherapy Chemotherapy Lung shielding Steroids Avascular necrosis 4–10% median time to inset Steroid Analgesia 18 months Radiation Surgery Osteoporosis 10% at 12–18 months Steroid Calcineurin exposure Hypothyroidism months–years TBI Levothyroxine replacement Gonadal failure/infertility TBI/conditioning intensity Sex hormone replacement Atherosclerosis GvHD Radiation CNI exposure Second malignancy Median 5–6 years TBI Chronic GvHD Autograft Myelodysplasia Psychological Chronic GvHD

326 Chapter 38: Hematopoietic stem cell transplantation

Liver dysfunction is typically an obstructive picture requires the input of a multi-disciplinary team that with elevation of alkaline phosphatase and bilirubin includes specialist medical services, physiotherapy, earlier than transaminases. Other organ involvement rehabilitation services, and psychological support. and fever may be present but tends not to dominate Immunosuppression is a combination of CyA and the clinical picture. prednisolone (initial doses 1.0–1.5 mg/kg/d), and Prevention of GvHD is problematic, as it is diffi- this may be continued for a number of years. As cult to separate the detrimental effects of GvHD from with acute GvHD, there is no consensus regarding the desirable graft versus leukemia/tumor effects. The second-line therapy, but agents used include TAC, standard prophylaxis is cyclosporine (CyA) and short- MMF, SRL, psoralen and UVA therapy (PUVA), and course methotrexate (10–15 mg/m2 days 1, 3, 6, and extracorporeal photophoresis. 11), but other regimens using drugs such as tacrolimus Both acute and chronic GvHD are major causes of (TAC), sirolimus (SRL), and mycophenolate mofetil mortality and morbidity. Mortality is in part related (MMF) are used. The use of T-cell depletion, either to organ damage but is also due to the profound ex vivo or in vivo with drugs such as alemtuzumab immune deficiency associated with GvHD and inten- or anti-thymocyte globulin (ATG), is controversial. sive immunosuppressive medication, leading to life- This is associated with a significant reduction of both threatening opportunistic infections. Supportive care acute and chronic GvHD, but this may be offset by an is important, with prophylaxis for fungal and viral increase in relapse rate. Management of acute GvHD is infections as well as encapsulated bacteria and Pneu- with steroid (methylprednisolone 2mg/kg) plus a cal- mocystis carinii pneumonia in chronic GvHD. cineurin inhibitor, reducing after 2 weeks if there is a complete response. Failure to control is usually defined as progression after 3 days, no change after 7 days, or Donor lymphocyte infusions incomplete response after 14 days. There is no con- The management of disease relapse is a difficult prob- sensus regarding best second-line therapy; a variety of lem; however, in some circumstances, it may be possi- agents show response rates of 30–70%. The prognosis ble to control disease by shifting the donor/host bal- of steroid-resistant acute GvHD is poor. ance in favor of donor hematopoiesis. This can be achieved with an infusion of donor CD3-positive T cells. Further lymphapheresis from the donor is usu- Chronic GvHD ally required, and cells given in escalating aliquots Chronic GvHD usually develops within the first year based on T-cell dose (e.g., 5 × 106 − 1 × 108 CD3/kg). after transplant. It may develop as a progression from Donor lymphocyte infusions are also used when there or after resolution of acute GvHD. It may also develop isafallindonorhematopoiesisintheabsenceofdis- de novo without prior acute GvHD. Risk factors ease relapse (falling donor chimerism). The major risk include acute GvHD, the use of mismatched donors, of this approach is the risk of GvHD or aplasia, espe- patient age, the use of ATG or alemtuzumab in the cially if donor hematopoiesis is low. A major compo- conditioning phase, and the use of peripheral blood nent of the graft versus leukemia (GvL) effect is prob- cells. Chronic GvHD typically affects the skin, liver, ably GvHD targeting the host hematopoietic system, and gut. Skin features include depigmentation, lichen and to date there is no reliable method for promoting planus, and erythema and a maculopapular rash. Mus- GvL without promoting GvHD. cle cramps, myositis, and fasciitis can lead to joint con- tractures. In the GI tract, chronic diarrhea, pancreatic insufficiency, and esophageal strictures can develop. Indications and outcomes Oral ulceration is a common problem. Chronic GvHD The major indication for hematopoietic stem cell is either classified as limited or extensive (revised transplantation is for management of high risk or Seattle classification) or graded as mild, moderate, or advanced hematological malignancy (Figures 38.3 and severe according to the National Institutes of Health 38.4). For autologous transplantation, myeloma is the Global Score, which combines severity of dysfunction single most common indication. A meta-analysis of with the number of organs involved. nine randomized studies has shown a clear ben- Management of chronic GVHD is a major efit in terms of improvement in progression-free challenge,whichaswellasimmunosuppression survival, and two studies have shown a benefit in

327 Section 7: Other

600 Figure 38.3 Indications for allogeneic hemopoietic stem cell transplant in the United Kingdom in 2009. 500 MDS: myelodysplastic syndrome; MPS: 400 myeloproliferative syndrome. Courtesy of Keiren Kirkland and British Society for Blood and Marrow 300 Transplantation. 200 100 0

kemia Other

Lymphoma MDS/MPS Solid Tumour Acute Leukemia Chronic Leu Plasma Cell Disease Bone MarrowHemoglobinopathies Failure Autoimmune Diseases

Primary Immune Deficiency

Inherited Disorders of Metabolism

1000 Figure 38.4 Indications for autologous 900 hemopoietic stem cell transplant in the United 800 Kingdom in 2009. MDS: myelodysplastic syndrome; 700 MPS: myeloproliferative syndrome. Courtesy of 600 Keiren Kirkland and British Society for Blood and 500 Marrow Transplantation. 400 300 200 100 0

Lymphoma MDS/MPS Solid Tumour Acute Leukemia Chronic Leukemia Plasma Cell Disease Autoimmune Diseases

Inherited Disorders of Metabolism

overall survival. As a consequence, an autologous ifthereissomeevidenceofchemosensitivity.Forthe transplant is considered a standard of care in first acute leukemias, the role of autologous transplantation remission for myeloma, with a median survival of is limited. The role of allogeneic transplantation for approximately 5 years. Lymphoma is the other major first remission acute myeloid leukemia is controversial. indication for autologous transplant; with the excep- The risks of NRM outweigh the reduction in relapse tion of some peripheral T-cell lymphomas and man- risk for good-risk disease as determined by cytogenet- tle cell lymphoma, it is generally used beyond remis- ics. Chronic myeloid leukemia is the disease in which sion. The PARMA study of relapse diffuse large B-cell the benefits of allogeneic transplantation were first and lymphoma showed an improvement in 5-year survival most clearly demonstrated, with survival in good-risk from 32% to 53%. In Hodgkin’s lymphoma, autolo- patients approaching 90%. Over the last 10 years, how- gous transplant is widely recommended in relapsed ever,transplantationhasalmostceasedasafirst-line disease, with the exception of late relapses. For refrac- treatment as superior results can be obtained with tory Hodgkin’s, the outlook is very poor. Most series less toxicity with the tyrosine kinase inhibitors. Sim- show a consistent long-term survival of 20–30% and ilarly, the goal posts are shifting for other diseases, autologous transplantation is widely offered, especially with improvements in outcomes with unrelated and

328 Chapter 38: Hematopoietic stem cell transplantation cord blood transplant as well as refinements in risk Further reading assessment, both largely driven by better molecular Filipovich AH, Weisdorf D, Pavletic S, et al. National data. Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: I. Diagnosis and staging working group report. Hematopoietic stem cell Biol Blood Marrow Transplant 2005; 11: 945–56. transplantation organizations Hughes WT, Armstrong D, Bodey GP, et al. Guidelines for r the use of antimicrobial agents in neutropenic patients British Society for Blood and Marrow with cancer. Clin Infect Dis 2002; 34: 730–51. Transplantation (bsbmt.org) r Koreth J, Cutler CS, Djulbegovic B, et al. High-dose therapy European Group for Blood and Marrow with single autologous transplantation versus Transplantation (ebmt.org) chemotherapy for newly diagnosed multiple myeloma: A r CIMBTR (cibmtr.org) systematic review and meta-analysis of randomized r JACIE (www.jacie.org) controlled trials. Biol Blood Marrow Transplant 2007; 13: 183–96. r FACT (www.factwebsite.org) Lee SJ, Vogelsang G, Flowers ME. Chronic graft-versus-host r Netcord (www.netcord.org) disease. Biol Blood Marrow Transplant 2003; 9: r ASBMT (www.asbmt.org) r 215–33. International Society for Cellular therapy (www. Vogelsang GB, Pavletic SZ (eds). Chronic Graft Verus Host celltherapy.org) Disease: Interdisciplinary Management. Cambridge: r World Marrow Donor program (www.bmdw.org) Cambridge University Press, 2009.

329 Section 7 Other Chapter Corneal transplantation

39 Yvonne H. Luo and D. Frank P. Larkin

Key points r In the United Kingdom and Europe, the availability of donor corneas for transplantation is almost sufficient for requirements. r Most corneal transplants are performed as penetrating transplants in which a 7–8-mm diameter circle of full-thickness cornea is excised and replaced by donor cornea of similar diameter. r Around one sixth of corneas undergo one or more rejection episodes; the first episode of acute onset rejection typically occurs between6and12monthspost-transplant. r The key to successful treatment of corneal graft rejection is early recognition of the Figure 39.1 Diagrammatic cross section of the cornea. Reproduced with permission from American Society of Transplant rejection episode by the patient and clinician. Surgeons and American Society of Transplantation. r Corneal transplant survival in large published series is approximately 75% at partial-thickness, corneal transplant in 1888, almost 5 years, similar to that of vascularized organ all transplants were xenografts. grafts.

Cornea is by far the most commonly transplanted Corneal anatomy tissue worldwide, with nearly 40 000 procedures per- Viewed from the exterior, adult human cornea is formed in the United States and 4000 in the United almost circular and approximately 11 mm in hori- Kingdom annually. Corneal transplantation, or ker- zontal diameter, with an average central thickness of atoplasty, is the surgical procedure most commonly about 530 microns. It consists of three main layers: used in the management of blinding opacification of (1) epithelium, which regenerates rapidly; (2) stroma, the normally transparent cornea. In addition, it can which constitutes 90% of corneal thickness and is itself be used in uncontrolled infection, pain, or perfora- perfectly transparent on account of the parallel and tion of the cornea. The first recorded corneal trans- evenly spaced ultrastructural arrangement of its col- plant in a human was in 1838, preceding by many lagen fibrils; and (3) endothelium. The monolayer of years the invention even of anesthesia. The donor endothelial cells is critical for maintenance of stromal cornea in this procedure was from a pig, and this deturgescence by action of its ATPase pump to remove xenograft remained transparent for only a couple of aqueous from the stroma. Endothelial cells in man do weeks. Until the first reported successful lamellar, or not have the capability to replace lost cells by mitosis,

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

330 Chapter 39: Corneal transplantation which is one reason why disease restricted to this Table 39.1 Examples of indications for corneal transplantation monolayer is the indication for a high proportion of with age at surgery corneal transplants. Age at surgery Transplant Diagnosis (years) technique Processing of donor cornea Keratoconus 18–40 ALK or PK Fuchs endothelial Ͼ60 PK or PLK prior to transplantation disease In the United Kingdom and Europe, the availability of Post-cataract surgery Ͼ60 PK or PLK donor corneas for transplantation is almost sufficient endothelial for requirements. Satisfactory endothelial cell density decompensation is the quality parameter widely used in eye banks. Stromal scarring 20–35 ALK or PK Corneas of satisfactory quality can be obtained from Post-trauma/infection postmortemdonorsofallages,andinmostcountries Rheumatoid disease Ͼ40 PK there is no upper age restriction for corneal donation. Corneal melting Although some donor eyes are harvested at the time of Peter’s anomaly 2–5 PK multi-organ donation, most are removed postmortem. Transplant replacement Any PK Endothelial cell density remains satisfactory up to Keratoconus, Fuchs corneal endothelial disease, and corneal 24 hours after death. When donor eyes are transported decompensation following cataract surgery are the common- to eye banks, cornea is removed for storage at either est indications for transplantation in European centers. Trans- ◦ plant replacement is indicated in some patients following fail- 4 C for up to 10 days in chondroitin sulphate–based ure of a previous transplant. Case mix varies according to center. medium or 34◦C in serum-based medium. This poten- ALK: anterior lamellar keratoplasty; PLK: posterior lamellar kerato- tial storage interval is a major difference between pro- plasty; PK: penetrating keratoplasty. cessing of donor cornea and other tissues. The feasi- bility of donor corneal storage allows quality assess- ment of donor corneas, testing for potential infective pathogens (including hepatitis, human immunodefi- ciency virus, and syphilis) and scheduling of transplant surgery. On account of current uncertainty about pos- sible risk of prion disease transmission, neurological disorders of uncertain etiology such as dementia and Parkinson’s disease are contraindications to corneal donation. Techniques of corneal transplantation Other than in children and young adults, most corneal transplants can be undertaken under local anesthesia. The majority are performed as penetrating transplants in which a 7–8-mm diameter circle of full-thickness Figure 39.2 Full-thickness corneal transplant sutured to recipient cornea is excised and replaced by donor cornea of eye with continuous 10/0 nylon suture. The edge of the donor similar diameter. The proportion of partial thickness cornea is indicated by the arrow. (lamellar) transplants has increased in recent years and involves replacement of either anterior (epithelium Immune privilege in corneal andmostorallstroma)orposterior(endotheliumand the underlying Descemet’s membrane) cornea with transplantation donor tissue. Anterior lamellar surgery can be suf- Both the (1) anterior chamber together with the ficient to restore transparency in those corneas with peripheral recipient corneal bed and (2) allogeneic stromal opacity but healthy endothelium; conversely, donor cornea itself enjoy relative immune privilege. posterior lamellar replacement may suffice in those Factors contributing to this privilege include the fol- with healthy stroma. lowing:

331 Section 7: Other

1. Absenceofbloodandlymphaticvesselsinnormal Immunology of corneal cornea significantly reduces trafficking of immune cells and inflammatory mediators to/from the allograft rejection donor. Some features of the immunobiology of corneal rejec- 2. Expression of Fas ligand (FasL) on corneal cells. tion differ from allogeneic rejection of other trans- Many T-lymphocytes possess the surface molecule plantedtissues.Alsoincontrasttoothertransplanted Fas. Binding of Fas+ alloreactive T lymphocytes tissues, most information on mechanisms of rejection by FasL on corneal surface cells induces of cornea has come from experimental models rather lymphocyte apoptosis. than clinical biopsies. This is on account of the possi- 3. Low levels of major histocompatibility complex bility of diagnosis of rejection by direct visualization of (MHC) class I and II molecule expression on the the transplant and the fact that tissue diagnosis is not surface of corneal cells. Because graft proteins can required. As in all adaptive immune responses, affer- only be recognized by host T lymphocytes after ent and efferent components can be identified and are they are processed by antigen-presenting cells described separately. (APCs) and recombined with APC surface MHC molecules, low transplantation antigen expression Afferent phase by donor corneal cells reduces the chance of T-cell Under circumstances that the immune-privileged fea- activation. tures of the cornea are bypassed by the immune 4. Paucity of professional APCs (particularly system, the first stage in rejection is recognition of dendritic cells) in the cornea and anterior thepresenceofnon-selftissue.Ofthetworoutesof chamber. Lack of donor APCs within transplanted allorecognition, the indirect pathway predominates cornea implies that antigen presentation to host in corneal rejection. Recipient-origin APCs enter the T lymphocytes by the direct pathway is negligible. donor cornea to capture and process donor antigens Donor corneal allorecognition and rejection is and migrate to the neck lymph glands to present the therefore believed to occur in most patients via antigen in context with self MHC class II molecules to the indirect pathway, initiated by APCs of T cells. From macrophage-depleting studies, evidence recipient origin that migrate into the has been provided that these cells play a crucial role transplant. in the afferent phase of graft rejection. On account of 5. Anterior chamber–associated immune deviation the paucity of transplanted passenger donor APCs, it (ACAID). In experimental studies it has been is assumed that corneal allorecognition by the direct shown that prior introduction of alloantigens into pathway is insignificant in most patients, and this isan the anterior chamber causes downregulation of area of conceptual difference between the allorecogni- systemic delayed-type hypersensitivity responses, tion of cornea and other tissues. Furthermore, the pre- a phenomenon known as ACAID. This dominance of indirect allorecognition is believed to be downregulation of immune responses may the reason for the lack of benefit of human leukocyte contribute to graft acceptance in humans. antigen (HLA) class II matching in corneal transplan- 6. Presence of immunosuppressive proteins in tation: similarity between the MHC class II proteins aqueous humor. Proteins such as transforming expressed or upregulated on donor cells and the host growth factor ␤ dampen the immune responses in would facilitate more effective presentation of graft the anterior chamber. alloantigens to host T lymphocytes by the host APCs. Corneal grafts at high risk of rejection are identi- fied by several risk factors, most of which reflect Efferent phase breakdown of facets of immune privilege. Prospec- When T-helper cells have identified the presented anti- tive clinical outcome studies identify the most signif- gen as non-self, effector mechanisms are generated icant of these to be recipient corneal vascularization, against donor tissue. Cytokines, including particu- corneal inflammation at the time of transplantation, larly tumor necrosis factor, have been clearly iden- which induces APC infiltration in the recipient cornea tified in aqueous humor and the cornea prior to prior to surgery, and a previously rejected ipsilateral observed endothelial rejection onset. After corneal graft. transplantation, it has been shown that alloantibody,

332 Chapter 39: Corneal transplantation

Table 39.2 Corneal transplant survival rates and causes of failure Corneal transplant survival rates 1year 86% 5years 73% 10 years 62% 15 years 55% Leading primary causes of transplant failure 1 Allograft rejection 31% 2 Late endothelial failure 21% 3 Infection 13% 4Glaucoma8% Data adapted from Australian Corneal Graft Register Figure 39.3 Endothelial corneal transplant rejection. Loss of 2007 Report. transparency in the inferior cornea is indicated by poor visibility of the pupil margin. tory aggregates on the donor endothelium and signs of macrophages, cytotoxic T lymphocytes, and delayed- inflammation in the anterior chamber. type hypersensitivity responses are components of the Treatment is commenced as soon as possible with effector response. very frequent topical steroid (e.g., dexamethasone hourly), reducing in frequency with resolution of inflammatory signs. Addition of systemic to local Clinical features and treatment of steroid has no effect on outcomes of endothelial rejec- tion treatment. In patients who do not present late, corneal allograft rejection most episodes can be reversed. The key to successful Approximately one sixth of corneas undergo one or treatment of corneal graft rejection is early recognition more rejection episodes. The first episode of acute of the rejection episode by the patient and clinician. onset rejection typically occurs between 6 and 12 months after transplant, but may occur much later. Signs of rejection can be directly visualized, may Outcomes of corneal transplantation involveanyofthethreelayersofcornea,andmay Corneal transplant survival in large published series is progress from stroma to endothelium. Destruction approximately 75% at 5 years (Table 39.2), similar to of the epithelium is unimportant from a functional that of vascularized organ grafts. The commonest rea- standpoint, as repopulation of the epithelium by recip- son for corneal transplant failure is allogeneic rejec- ient origin cells readily occurs. Stromal rejection is tion, contrary to the standard belief in transplantation more frequently seen, typified by nummular transplant immunology that rejection of corneas does not occur. opacities in a patient presenting with moderate ocu- In transplants that are not identified prior to surgery lar discomfort. On account of the vital physiological to be at high risk of rejection, topical steroid alone importance of endothelial cells to donor corneal func- forapproximately6monthsisusedasrejectionpro- tion and the incapability of repopulation by mitosis, phylaxis, management that is very different from that endothelial rejection is of greatest importance. Rejec- of the cadaveric vascularized organ transplant recipi- tion episodes irreversibly damage the endothelium to ent. For those with bilateral blinding corneal disease a degree, reducing cell density and the pumping func- and also at high rejection risk (see above), systemic tion capability of the endothelium. Once the number immunosuppression is used in some centers as pro- of endothelial cells/mm2 falls below the threshold level phylaxis. Published case series are the only evidence required to maintain cornea clarity, irreversible stro- for drug selection. The only randomized trials in high mal swelling occurs and the cornea becomes cloudy. rejection risk corneal transplants have been under- During an endothelial rejection episode, visual distur- taken to examine the influence of HLA matching on bance is caused by stromal edema; there are inflamma- outcome, which in summary show a weak benefit from

333 Section 7: Other

matching class I antigens only. HLA matching is not GoreSM,VailA,BradleyBA,RogersCA,EastyDL, routinelyundertakeninpatientswhoarenotathigh Armitage WJ. HLA-DR matching in corneal rejection risk. transplantation. Systematic review of published evidence. Corneal Transplant Follow-up Study Collaborators. Transplantation 1995; 60: 1033–9. Further reading Niederkorn JY, Larkin DFP. Immune privilege of corneal Chatel MA, Larkin DFP. Sirolimus and mycophenolate as allografts. Ocul Immunol Inflamm 2010; 18: 162–71. combination prophylaxis in corneal transplant recipients Williams KA, Coster DJ. The immunobiology of corneal at high rejection risk. Am J Ophthalmol 2010; 150: transplantation. Transplantation 2007; 84: 806–13. 179–84. Williams KA, Lowe MT, Bartlett CM, Kelly L, Coster DJ George AJT, Larkin DFP. Corneal transplantation – the (eds). The Australian Corneal Graft Registry 2007 Report. forgotten graft. Am J Transplant 2004; 4: 678–85. Adelaide, Australia: Flinders University Press, 2007.

334 Section 8 The transplant service Chapter UK and European service – legal and operational framework 40 ChrisJ.RudgeandAxelO.Rahmel

Key points Europe while waiting for the transplant that never comes. Regulation must accept that not all human r The diagnosis and confirmation of death organs donated for transplantation are ideal and that mustbeseparatedfromanythingtodowith it is not possible to eliminate all risks. the issues surrounding organ donation and In a single chapter it is not possible to document transplantation. r in detail the arrangements in every European coun- In the UK, explicit consent/authorization try, but the authors have attempted to give a thorough must be given before organs are removed for review of the UK position and to describe the impor- transplantation. r tant similarities and differences across other European Living donors must be acting freely and countries. Throughout, the terms donation after brain voluntarily and no payment should be death (DBD) and donation after circulatory death offered for the donation. r (DCD) have been used rather than heart-beating and A “usual not unusual” part of the care of non–heart-beating donation. dying patients should be to ensure that the option of donation is available to all suitable patients and their families. Organ donation r The organ allocation system should be open, transparent, fair, and equitable. Legal structures Definition of death This chapter describes the legal and operational frame- United Kingdom works that are necessary at a national or supra-national Traditionally in the United Kingdom, the law has not level in order to support and regulate organ dona- been used to define how the diagnosis or confirma- tion and transplantation. Although there are differ- tion of death is to be performed, relying instead on ences in detail between countries, all those with a well- the medical profession to produce a code of practice. established donation program have in place a legal The courts have, however, endorsed such professional framework, a national donation system, and processes guidance. Following the acceptance of the concept of to ensure quality and safety of organs. They also have a brain death, the first such guidance was published in degree of regulation and oversight. Organ transplan- 1976. Several documents have been published since, tation requires equitable organ allocation and mea- the most recent and significant of which is “A Code sures to prevent organ or people trafficking, again with of Practice for the Diagnosis and Confirmation of appropriate regulation and oversight. The role of a Death,” produced under the auspices of the Academy national or international transplant registry is critical. of Medical Royal Colleges and endorsed by the Depart- Regulation and oversight are clearly essential, but ment of Health. This expands the basis on which death thereneedstobeaconstantefforttoensurethatreal- of the brainstem equates to death of the individual istic risk–benefit decisions are made. There is a world- and clarifies the clinical steps that must be followed wide shortage of donated organs, with thousands of before death can be diagnosed. Moreover, it also, and patients dying each year in the United Kingdom and for the first time, lays down clear clinical criteria for the

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

335 Section 8: The transplant service

diagnosis of death following irreversible cardiac and Europe. Four different categories of DCD donors are respiratory arrest. Of critical importance is the state- currently distinguished: ment that the working party believes that it is impor- A. Uncontrolled tant to separate completely the diagnosis and confir- mation of death from anything to do with the issues I. Dead on arrival at hospital surrounding organ donation and transplantation. This II. Unsuccessful resuscitation document deals solely with the diagnosis and confir- B. Controlled mation of death, whatever the cause, allowing this to be carried out in a variety of circumstances where fur- III. Awaiting cardiac arrest ther intervention aimed at sustaining life can be of no IV. Cardiac arrest while brain dead further benefit to patients. The UK Code of Practice is based on the essential Category IV DCDs with a completed brain death diag- prerequisite that the cause of the patient’s deep coma nosis, if medically suitable, can obviously be used as must be known. Thereafter, in the majority of patients, organ donors in all countries, with the exception that a series of well-defined clinical bedside tests are used DCD is legally not permitted at all in some countries todemonstratecompletelackofanybrainstemfunc- (Croatia,Germany,Hungary,andPoland)becausea tion, with the tests to be carried out twice by two med- completed brain death diagnosis according to the pre- ically qualified practitioners, of whom one must be viously mentioned national regulations is required in a consultant. If the tests are unequivocal and can all all cases of deceased donation. be carried out, then no further confirmatory tests are In other countries, DCD is not prohibited, but no required. These may, however, be helpful in certain cir- respective programs are in place yet. cumstances such as severe facial trauma that preclude Active DCD transplant programs currently exist the ocular and corneal reflex testing. in nine European countries only (Austria, Belgium, Czech Republic, France, Italy, the Netherlands, Spain, Europe Switzerland, United Kingdom). In some of these coun- In all European countries, organ transplantation is tries, only DCD of category I and II donors are legally guided by the overarching ethical requirement known approved (France, Spain), whereas in the other coun- as the “dead donor rule,” which states that patients tries, all types of donors are permitted. In the latter, the must be declared dead before the removal of any category III donors typically represent the majority of vital organ for transplantation. Deceased donation can all DCD donors. either take place after determination of brain death or Finally, there are substantial differences with of cardiac death. The determination of brain death is regard to the DCD protocols used in the different legally binding in all European countries prior to post- countries and sometimes even between the different mortem DBD. However, concerning the definition of centers within one country. As an example, the “no- brain death, two groups of countries can be distin- touch period” after cardiac arrest until the start of guished: in the majority of European countries, brain organ procurement varies between 2 and 20 minutes. death is defined as the complete and irreversible ces- Therefore, initiatives are underway aiming at harmo- sation of all cerebral and brainstem function. In Por- nization of the DCD protocols. tugal and Switzerland, as in the United Kingdom, the verification of the complete and irreversible cessation Consent of brainstem function is sufficient. United Kingdom Although the need for brain (stem) death diagno- A legal framework for organ donation after death sis as a prerequisite for organ donation is regulated by is an absolutely essential component of any trans- a parliamentary act in all countries, the medical and plant system. The Human Organ Transplant Act 1961 procedural details of the brain (stem) death diagnosis was replaced in England, Wales, and Northern Ire- are typically laid down in guidelines by national medi- land by the Human Tissue Act 2004 (and in Scot- cal associations or in specific decrees and thereby vary land by the Human Tissue Act 2006). The principal slightly from country to country. activities covered by these acts are the removal from The current situation with regard to regulations a deceased person of organs for the purpose of trans- concerning DCD is much more heterogeneous in plantation and, most importantly, the requirements for

336 Chapter 40: UK and European service – legal and operational framework appropriate consent (or authorization in Scotland) being Spain. But, as is the case in countries with before this can lawfully be carried out. Although there informed consent systems, in countries with presumed are small but relevant differences of detail between the consent, relatives or other close persons of the donor acts, they are considered together, and the general term areinvolvedwhenpossiblepriortotransplantation: of “consent/authorization” will be used. in all cases where there is no documented will of the The fundamental import of the acts is that they donor, any objection from the family is respected. But require that explicit consent/authorization must be even when the donor had consented but the fam- given before organs are removed for transplantation. ily objects to donation, although legally a donation This may be given by the donor themselves in life, for would be possible, in practice the wish of the family example, through registration on the National Health is respected and organ donation is not pursued. Service (NHS) Organ Donor Register. If the donor has So in summary, in spite of the formally different not given consent/authorization, then it may be given legal solutions to consent in the different European by others on his or her behalf, and the two acts describe countries, family consent does in fact play a major a hierarchy of individuals who may do so. The details role in every donation procedure. Therefore, it has differ slightly, but both place at the top of the order a been claimed that the varying donation rates that can nominated representative (if there is one) or the clos- be observed between European countries is only to a est relatives of the donor and end with a friend of long minor degree due to differences in legislation. standing. The issue of presumed consent, or opting- out legislation, was considered in detail in the United Living donation Kingdom, and a report in November 2008 concluded United Kingdom that although the decision was finely balanced, the rec- The two Human Tissue Acts mentioned above also set ommendation was that there should be no change to the legal framework for living donation. In principle the existing, explicit consent (or opting-in) legislation. they make several main provisions. First, that living donors must be acting freely and voluntarily, and that Europe no payment is offered for the donation. The acts are Consent to donation is the second prerequisite to post- clear that a criminal offense has been committed if any mortem organ donation, and details are laid down in reward is given or is to be given to the donor. Sec- national legislation in all European countries. In some ond, the acts established the Human Tissue Author- countries, explicit consent by the donor is necessary ity (HTA), which is required to publish codes of prac- (informed consent, opting-in). This consent can either tice that allow clinicians to work within the law. Most be expressed by the donor already during lifetime in specifically, the HTA is required to give prior approval writing, for example, by a donor card or by registra- for all living donor transplants in the United King- tion in a respective national donor registry. If such a dom, regardless of the nature of the genetic or other written statement does not exist, the relatives or other relationship between the donor and recipient. In prac- closepersonstothedonorareaskedinstead.Therel- tice, the overwhelming majority of such transplants atives are not asked about their opinion, but rather are between genetically related pairs or those with a about the (presumed) opinion of the deceased per- long-standing and well-defined emotional relationship son concerning organ donation, but of course, espe- such as spouse or partner, and in these circumstances cially in the absence of an explicit statement, transition HTAapprovalisgivenonthebasisofawrittenreport between the two is fluid. Informed consent systems from a trained and designated “Independent Asses- exist for example in Germany, the Netherlands, Slove- sor.” However, if the circumstances are more complex, nia, Switzerland, and as already explained in detail in an HTA Panel is convened to consider the application the previous section, in the United Kingdom. and, if satisfied, to give approval for the transplant. In the majority of European countries, a presumed This latter procedure is used, for example, in the case consent or opting-out system for achieving consent of the non-directed, altruistic, stranger donation when to donation is in place. Under this regulation, every an individual wishes to donate a kidney anonymously deceased person is considered a donor if no objection to whichever patient on the national waiting list is the was expressed during their lifetime. In most countries most appropriate recipient, and also in cases where that have an opting-out system in place, a non-donor there are increased risks to the donor and/or recipient, registry is established, the most prominent exception such as living donor liver transplantation.

337 Section 8: The transplant service

Despite this apparently cumbersome requirement only allowed if no deceased donor can be identi- for HTA approval for all living donor transplants, the fied. In practice this rule has only minor implica- system has worked extremely well since its introduc- tions, because in general the time between the decision tion in 2006, with approval being given within 2–3 days to perform a living donor transplant and the trans- in the majority of occasions. Clinicians are supported plantation is so short that a suitable deceased donor in knowing that their judgment and decisions have organ for the recipient does not become available in been endorsed by an independent assessor, and the time. system provides a robust process to minimize the like- In all European countries, living donation has to be lihood of organ trafficking from living donors. The leg- unpaid, organ trafficking is prohibited, and violations islation has also allowed the development of a paired- will be prosecuted, focusing on vendors and any med- exchange program for incompatible living donor pairs, ical personnel involved. which is now well established, with a growing num- ber of transplants, and “stranger” donations are also Organ donation organization increasing in number. Donor coordination Europe United Kingdom In most European countries, living donation is regu- In January 2008, the UK Health Departments pub- lated by law because the protection of the living donor lished the report from the Organ Donation Task has to be guaranteed, and any form of organ trade or Force “Organs for Transplants.” This report identified trafficking is to be prevented. In order to achieve this, the obstacles to donation after death in the United several restrictions to living donation are in place in Kingdom and made 14 recommendations. One over- the different European countries. In most countries, arching principle runs throughout the report: the either a defined genetic relationship (typically first- or ambition to make donation “usual, not unusual.” Cen- second-degree relatives) or another form of close rela- tral to this is the aim to transform the approach of tionship between donor and recipient is required. In those working in the care of potential organ donors several countries, an exception to this rule is allowed and to develop a collaborative of clinicians and donor in case of cross-over (paired exchange) living dona- coordinators within every hospital. This has involved a tion, as long as the relation of each donor/recipient complete reorganization of the donor transplant coor- couple fulfills the above-mentioned prerequisites. In dinator (DTC) network in the United Kingdom. Prior a few countries altruistic donation is allowed, typi- to 2008, DTCs were employed in a variety of ways cally as non-directed donation to the recipient pool, around the country, with the majority, but not all, because in the case of directed, altruistic donation, it is based within transplant units. Funding arrangements considered difficult to prevent abuse leading to organ also varied. A key task force recommendation was trading. In all countries, informed consent to dona- that NHS Blood and Transplant (NHSBT) should take tion is required. As a consequence, donation by minors over the direct employment of DTCs, that their num- is in most countries not possible. In only a few coun- bers should increase from approximately 100 to well tries, e.g., Slovenia, Switzerland, and the United King- over 200, and that their working arrangements should dom, are exceptions to this general rule possible under change significantly. Most of these changes have taken clearly defined, narrow preconditions. place over the past 18 months and are expected to be The majority of European countries have estab- complete early in 2011. There are currently over 185 lished specific ethics committees that check whether DTCs, and the majority of them are nurses from an living donation is in fact voluntary by the donor and, intensive care background. Their focus is entirely on in case of genetically unrelated donor–recipient pairs, organ donation; they have no responsibility for trans- whether in fact a close relationship between donor and plant recipients. The intention is that in due course recipient exists to prevent exploitation of the donor they will all be “embedded” within the individual hos- and possible organ trafficking. These committees are pital in which they are based, as part of the inten- also involved in all cases of altruistic donation. sive care and end-of-life teams. Their role is to raise In some countries, living donor transplantation awareness, ensure that local training needs are met, is considered subsidiary to deceased donor trans- and to encourage the identification of all potential plantation, so that living donor transplantation is donors. Typically they will work in their hospital for

338 Chapter 40: UK and European service – legal and operational framework

3–4 days/week, but (on a regional basis) they will all The role of critical care and emergency medicine participate in a 24/7 on-call rota to facilitate actual United Kingdom donors. The task force recognized that one of the key elements in a successful donation program is the need for those Europe working in critical care (intensive care and emergency Historically, donor coordination was taken care of in medicine) to believe that a “usual not unusual” part of most countries by local coordinators associated with or their care of dying patients should be to ensure that the even employed by the transplant centers. Such a model option of donation is available to all suitable patients still exists, for example, in Belgium and until recently and their families. This approach, which is building in Switzerland. In most countries, organ donation is considerable momentum in the United Kingdom, can now centrally organized, and national coordination only be sustained when all clinicians are supported offices have been set up. These are responsible for the and are able to work within a clear and unambigu- training, accreditation, and often also the employment ous framework of good practice based on legal, ethical, of the donor coordinators. As a result of this cen- and professional guidelines. To this end, the Depart- tralization process, in some countries such as Ger- ment of Health published guidance on the legal frame- many and the Netherlands, local coordinators are no work for DCD in November 2009, which is based on longer directly linked to the donor hospital but are an understanding of the Mental Capacity Act 2005 in located outside the hospitals. The Spanish experience England and Wales (and similar guidance was pub- with central management of in-hospital coordinators lished in May 2010 on The Adults with Incapacity Act has resulted in substantial increase in organ dona- 2000 in Scotland). The main legal concerns relate to tion and is considered to be a central element in suc- the management of a patient who is dying but not dead cessful transplantation. Therefore, most countries are and who lacks capacity to give consent for treatment. strengthening in-house coordination, of course taking The thrust of the advice is that if a person’s wishes into account organizational and structural differences were to be a donor, then certain actions that facili- in their national health care systems. tate donation may be considered to be in his or her The tasks of the transplant coordinators vary from best interests (and thus lawful) if they do not cause country to country. In some countries they are already the person harm or distress. Furthermore, an indepen- involved in the identification of possible donors, dent, high-profile UK Donation Ethics Committee has whereas in other countries their tasks start with the been established under the auspices of the Academy of assessment of identified potential donors. In all coun- Medical Royal Colleges, which met for the first time in tries they are responsible for the organization of the January 2010. whole donation process. If necessary they take care The second component of the “collaborative” men- that brain death diagnosis can be performed by trained tioned above, which is to be established in every acute experts, they support the local hospital staff in main- hospital, is a Clinical Lead for organ donation. More taining the donor on the intensive care unit (ICU), and than 95% of hospitals have now made an appointment, they cooperate with the retrieval teams prior to and and the overwhelming majority are consultants from during organ procurement. a clinical background in intensive care. They are paid In some countries transplant coordinators are (by NHSBT) for approximately half a day to 1 day per responsible for most or all of the communication with week.Theirroleistoactasalocal“Lead”fordonation the donor family, whereas in others they only talk to through raising local awareness, establishing training, the donor’s family when requested by the local hos- and ensuring that protocols for donation are in place pital team. Recent data concerning the achieved con- based on national guidelines. The task force described sent rates suggest that specifically trained coordinators minimal notification criteria that would ensure that are more successful in this regard than ICU person- all potential donors would be referred to the DTC nel. This is another argument in favor of the establish- network, at the appropriate time, and this will not ment of local in-house donor coordinators. The direc- only help to maximize actual donation, but will also tive on organ transplantation and the accompanying provide an accurate and consistent measurement of action plan therefore explicitly support the introduc- the performance of every hospital toward maximizing tion of in-house coordination in Europe. donation.

339 Section 8: The transplant service

Table 40.1 Definitions of brain death and cardiac death donors Donation after brain death (DBD) Donation after cardiac death (DCD)

Potential donor A person whose clinical condition is suspected to fulfill A person in whom the cessation of circulatory and brain death criteria. respiratory functions is anticipated to occur within a time frame that will enable organ recovery. Eligible donor A medically suitable person who has been declared dead A medically suitable person who has been declared dead based on neurological criteria as stipulated by the law of based on the irreversible absence of circulatory and the relevant jurisdiction. respiratory functions as stipulated by the law of the relevant jurisdiction, within a time frame that enables organ recovery. Actual donor A consented eligible donor: In whom an operative incision was made with the intent of organ recovery for the purpose of transplantation and/or From whom at least one organ was recovered for the purpose of transplantation. Utilized donor An actual donor from whom at least one organ was transplanted.

Europe Donation Committee, and the chairs of these commit- The crucial role of intensive care nurses and doc- tees. The chair will typically come from a non-clinical tors in the identification and maintenance of potential background and have a high local profile within the donors has already been noted and is acknowledged hospital or the local community. The members of the in all European countries. In some countries, intensive committee will include clinicians, nurses, faith leaders, care personnel are legally obliged to report potential ethics, management, and others. The committee will donors to the national donor coordination organiza- receive (every 6 months) the donation activity for their tions. In day-to-day practice, the effect of such legis- hospital based on the PDA, assess progress, identify lation has been limited: raising the awareness around local obstacles and solutions, and report to the man- organ donation and continuous training of ICU per- agement board of the hospital. sonnel on the other hand is of pivotal importance. Europe Data on donor potential is not collected in Europe; part The potential for donation of the problem is that generally agreed definitions are United Kingdom not established. Recently the World Health Organiza- The UK Potential Donor Audit (PDA) was established tion (WHO) suggested a set of definitions for DBD in 2003 and collects data on every patient who dies in andDCD.Althoughdataonactualandonutilized every intensive care unit in the country. The hierarchi- donors are typically available, information on eligible cal structure of the PDA allows the reasons for which and especially potential donors is lacking in almost all a potential donor failed to become an actual donor to countries so far. be identified, and the data can be used both nation- In Spain, a systematic retrospective review of all ally and locally. The most recent data (UK Transplant ICU deaths is performed; this allows clear identifica- Registry, unpublished) shows that for potential donors tion of all potential donors and helps to increase aware- after brain death, 25% of patients were not subjected ness of potential donors among the ICU personnel in to brainstem tests, 11% of patients certified dead were the future. not referred to the DTC network, and the refusal rate The Donor Action program works in a similar for donation was 39%. Thus the overall conversion rate manner. One of the steps of this initiative to increase in 2008/2009 was only 51%. There are many reasons organ donation is a systematic medical record review for these figures, not all of which represent a failure (MRR). The MRR uses specially designed forms to col- of the system, but the availability of such data allows lect retrospective and prospective data on critical care focused efforts, both nationally and locally, on areas unit mortalities. These data are used to establish the forimprovement.Atalocallevel,afurtherrecommen- unit’s donor potential and to pinpoint where and when dation of the task force was that every hospital should potential donors failed to convert into actual donors. A have a “non-clinical champion,” and in practice this is recent comparison of the conversion rates in countries being achieved through the establishment of a hospital with an opting-out versus an opting-in system showed

340 Chapter 40: UK and European service – legal and operational framework

Table 40.2 Donor and transplant rates per million population (2008) Deceased donors Kidneys Livers Hearts Lungs

UK 14.7 23.0 11.9 2.1 2.3 Spain 34.2 44.9 24.0 6.3 4.2 Germany 14.6 26.6 13.7 4.7 3.3 France 25.3 41.9 15.9 6.0 3.4 Netherlands 12.8 21.5 8.0 1.6 3.4 Italy 21.1 26.9 17.8 5.7 1.6 Norway 20.5 37.6 16.5 8.1 6.3 Sweden 16.5 30.7 15.8 4.9 5.6

a significantly higher conversion rate in the countries patients with severe brain injury is futile and/or inap- with presumed consent, related to a lower number of propriate and that as a result treatment is withdrawn refusals in these countries. before the patient reaches the stage of brainstem death. There is also the likelihood that changes in the clini- Donor and transplant rates in the United cal management of such patients, such as decompres- Kingdom and Europe sive craniectomy and stenting/coiling of intracerebral The United Kingdom has, for many years, had one arteriovenous malformations, are changing the out- of the lower deceased donation rates in Europe while come for such patients. Finally, there is also anecdo- moving in recent years to achieve one of the higher tal evidence that critical management decisions about living donor rates. There are many reasons for this, patients are being made at an even earlier stage, i.e., and it is important to recognize that a wide range in the Department of Emergency Medicine, with the of factors influence both the number of possible or result that patients who are thought to have untreatable potential deceased donors and the proportion of those or unsurvivable injuries are not being admitted to crit- that progress to become actual donors (the conver- ical care facilities. Any developments that reduce mor- sion rate). Although every country must endeavor tality from major brain injury are clearly welcome, but to maximize the conversion rate, it is not realistic it is important to understand these changing patterns to expect every country to achieve the same actual of clinical care in order to focus efforts on optimizing deceased donor rate. Factors likely to have an impact donation in the most appropriate ways. include the mortality rates from head injuries (par- The demographics of deceased organ donors are ticularly road traffic accidents) and various forms of changing in the United Kingdom as elsewhere, with intracerebral hemorrhage, the number of intensive more donors being older and obese and fewer dying care beds, and the donation infrastructure. Of note in from trauma. As a result, an increasing proportion the United Kingdom are two important observations. of donated organs are “marginal,” but the propor- First, the PDA shows that the number of patients in tion of “offered” organs that are in fact retrieved and ICU in whom brainstem death is a possible diagno- transplanted has stayed remarkably constant, imply- sis has fallen by almost a third in the last 5 years, as ing that a greater proportion of transplanted organs are has the number of patients in whom death is diag- also increasingly marginal. There is an interesting (but nosed by these tests. Thus there has been a marked unanswered) question regarding the appropriate “dis- fall in the number of patients from whom organ dona- card rate” of donated organs; too low suggests that the tion (DBD) is possible. Second, the number of DCD criteria for donation/transplantation may be too rigid, donors has risen from 128 to 335 in the same time whereas too high suggests inappropriate retrieval of period. There is considerable speculation (but little organs from some possible donors. There are consider- evidence) that increasingly a decision is being made at able differences between the main European countries: a relatively early stage that further active treatment of although the donor rate in Spain is 2.36 times that of

341 Section 8: The transplant service

Table 40.3 Risk levels for donor organs Category Definition

Standard risk Evaluation process did not identify any transmittable disease Calculated risk Evaluation identified a transmissible disease, but recipient has same disease or protective serology Increased but acceptable risk Evaluation identified a transmissible disease, but organ utilization is justified by the recipient-specific health situation Unacceptable risk (absolute In these cases the donor is not suitable for transplantation. contraindication) Typically this category includes: r HIV 1 or 2 seropositivity r HBsAg and HDV contemporaneous seropositivity r Current neoplastic conditions (some exceptions apply) r Systemic infections caused by agents for which treatments are not feasible r Documented prion diseases Not assessable risk Evaluation process does not allow an appropriate risk assessment for transmittable diseases HBsAg: Hepatitis B virus surface antigen; HDV: hepatitis D virus; HIV: human immunodeficiency virus.

the United Kingdom, the kidney transplant rate is only publishing guidance. Updated guidance was published 1.95 times greater, still a very significant advantage, but in 2011 and emphasizes the importance of counseling suggesting a higher discard rate in Spain. and consenting potential recipients about specific risk factors for imparied outcome of organ transplantation. Safety and quality of donors United Kingdom All human organs donated after death for transplan- Europe tation potentially carry a risk of disease transmission Currently there are neither common binding Euro- from the donor to the recipient. The main risks are pean rules on donor evaluation nor on possible con- of transmission of an infectious agent or of malignant traindications to transplantation. As indicated earlier cells from a primary tumor elsewhere in the donor. in this chapter, the process of evaluation of donor Although all possible steps are taken to minimize the suitability is influenced by the limited availability of risk of any such transmission, transplantation differs organs. In every case the individual balance between fundamentally from many forms of tissue transplan- risks and expected benefits for the recipient have to tation and blood transfusion. The interval between be taken into account prior to acceptance of an organ donation and transplantation is measured in hours for transplantation. In addition, time constraints sur- rather than days or weeks (thus limiting the testing that rounding the donation process and due to the lim- can be performed), and many forms of organ trans- ited ischemic tolerance of the donor organs may pre- plantation are life-saving, with the alternative to use of clude performing certain screening procedures. In an available organ likely to be the death of the recipient the past, the Council of Europe has published sev- within days or weeks in some circumstances. A sensi- eraldocumentsonthissubject.Inthecourseofthe ble approach therefore should allow some discretion Alliance O project, in which representatives from sev- for informed decision making about a possible trans- eral European organ procurement and organ exchange plant by the patient, his or her family, and the clinicians organizations participated, the following recommen- responsible for the patient’s care. dations were developed: the evaluation of the suitabil- Guidance on the testing of organ donors and deci- ity of the donor has to be based on medical history, sion making about the use of organs with known risk physical examination, instrumental as well as labora- factors comes from the Advisory Committee on the tory tests, and histological examination and/or post- Safety of Blood, Tissues and Organs, the committee mortem examination with the aim to clarify issues that established by the Department of Health in 2008 to have emerged during the previous evaluation steps or succeed the Advisory Committee on the Microbiolog- stilltobeinvestigated.Basedonthefindingsgathered ical Safety of Blood, Tissues and Organs for Trans- during the donor evaluation, the donor (organs) can plantation, which had previously been responsible for be attributed to a risk level (Table 40.3).

342 Chapter 40: UK and European service – legal and operational framework

The final decision of organ suitability has to be r Outcome: Maximize the health benefit from every taken by the clinician caring for the potential recip- donated organ ient. In cases in which a patient becomes a potential r Program driven: Good survival results, either for recipient of an organ with an increased risk, a trans- the transplant center or the overall national plant team member must explain the specific risks and transplant program informed consent should be obtained from the recipi- r Public policy: Cost effectiveness ent or his or her family. Later in 2010, an EU Directive r Lobby groups: Appeals to the media by individual on Organs was published that could harmonize donor patients or groups representing patients with a evaluation in the EU member countries. particular disease r Willing to pay: With money or with organ Regulation and Oversight in the United donation r Social worth: Status, job, children or other Kingdom and Europe dependants, lifestyle (“good” or “bad”) The EU Directive on the Standards of Quality and r Queuing: First come, first served Safety of Human Organs intended for Transplan- r Random selection tation has now been adopted, and it will require implementation by 2012. While it is not possible to In practice, the over-riding principle is that of national describe the details of the directive, it is clear that equity of access to transplantation, such that similar every member state is to be required to establish a patients with similar severity of disease should have competent authority that will have considerable reg- an equal chance of receiving a transplant regardless ulatory and oversight responsibilities. In a number of where they live, their race or gender, and, as far of countries, many of these functions are already in as possible, the biological determinants such as blood place through the various established authorities as groupandhumanleukocyteantigen(HLA)pheno- described earlier in this chapter, and at present it would type. However, a number of other factors are also taken appear that the directive is unlikely to add a signif- into account, such as waiting time, the ideal of offer- icant bureaucratic burden. Harmonization of mini- ing well-matched organs to younger patients, and the mum safety and quality standards will allow easier specific needs of individual patients such as those with and safer cross-border organ sharing where this is multiple anti-HLA antibodies or with fulminant organ desirable. failure for whom death may be expected within days or weeks without transplantation. Added to this must be the different safe storage (cold ischemic) times for Organ allocation and transplantation different organs and the need to minimize transport times within safe limits. Allocation – principles and practice It is not possible to describe in detail in this chap- United Kingdom ter the different allocation schemes for the differ- ent organs, and indeed they are all subject to regu- Open, transparent, fair,andequitable are the terms fre- lar review and revision. However, several key steps quently used to describe the requirements of any organ toward the development of a satisfactory scheme can allocation system. These are clearly spelled out in the be described. First, they should be based on as much WHO Guiding Principles that were adopted in May informationasisavailable,ondonorandrecipientfac- 2010 (Guiding Principle 9). The key question for any tors that influence outcome and on the demographics organization responsible for devising and implement- of both donor and recipient pools. Second, they should ing allocation rules is “What do we want to achieve?” bebasedonpatientneedsandoutcomesratherthanon and in the United Kingdom, this responsibility rests the perceived interests of individual centers. Third, the primarily with the organ-specific Advisory Groups of advisory groups have available to them sophisticated NHSBT. computer simulation tools that, although not neces- There are numerous, often conflicting, possible sarilypredictingtheexactoutcomeofanyproposed principles for organ allocation, of which the most fre- scheme, are invaluable in comparing the benefits and quently proposed include: disadvantages of a number of proposed alternatives r Need: Give priority to the sickest patient to inform a balanced judgment regarding the optimal

343 Section 8: The transplant service

scheme. These tools have, in fact, proved to be remark- ilar to the situation described for organ procurement, ably accurate in predicting the observed outcomes of organ allocation was in the early days of transplanta- changes that have been introduced. tion taken care of by the transplant centers, and the To summarize the various schemes very briefly: organs were allocated locally. Quickly, (mandatory) exchange with other centers was introduced to address National Kidney Allocation Scheme the needs of specific patients, with this allocation being 1. Introduced in 2006, this allocates all kidneys from patient-oriented. On the other hand, regional alloca- DBDs nationally, whereas kidneys from DCDs are tion was typically center-oriented, giving the trans- allocated by the local retrieval center. plant team the choice to select the most suitable patient 2. There are five hierarchical tiers, the highest being from the waiting list at their center. Organs for which pediatric recipients who are highly sensitized and no suitable recipient could be found in the donor cen- for whom the kidney is a zero-mismatch, the ter or region were offered to other transplant centers lowest being adult patients who are not highly or regions within the donor country. Some countries sensitized and for whom the kidney is more than a established multi-national cooperation: the largest is zero-mismatch. Eurotransplant (Austria, Belgium, Croatia, Germany, 3. Within each tier, the selected recipient is chosen Luxemburg, the Netherlands, Slovenia), but there is on a points score that includes waiting time, HLA also Scandiatransplant (Denmark [including Green- match, and age points combined: age difference land], Finland, Iceland, Norway, and Sweden) and between the donor and recipient, additional the United Kingdom, with mandatory exchange rules points for HLA homozygous patients, and blood between the participating countries. group points that allocate a certain proportion of Some of the historical elements persist in most blood group O kidneys to blood group B European countries to a certain extent. For example recipients. mandatory exchange for specific patient groups such as high urgent, highly immunized, or pediatric recip- National Liver, Heart and Lung Allocation Schemes ients is in place in most countries. In some cases, These schemes are broadly similar in that each trans- this mandatory exchange is combined with a regional plant center is allocated a number of local donor hos- balancing or pay-back system. Besides the rules for pitalsandisabletochoosearecipientfromtheirown mandatory exchange criteria, organ-specific allocation waiting list for an organ from a local donor. This is rules have been developed in most countries based on over-ridden by the Super-Urgent Liver scheme and the above-mentioned ethical and medical criteria. In the Urgent Heart scheme, whereby patients meeting general, medical expert groups are responsible for the defined urgency criteria are listed nationally and take establishment of these rules; in some countries, other priority over local allocation if the organ is suitable. stakeholders such as patient groups and legal experts are also involved in the process. In many countries National Pancreas Allocation Scheme (e.g., France [kidneys], Hungary, Portugal, and Spain), A national patient-based allocation scheme for theresultingmatchlistisincaseofregionalallocation kidney–pancreas and pancreas-alone transplants, not binding, but is rather considered as a recommen- incorporating the need for islet transplantation, has dation. been developed and recently introduced. In some countries (Switzerland, Eurotransplant countries [with the exception of non-renal organs in Europe Austria], France [non-renal organs], and the UK [see Organ allocation rules vary substantially between above]), a strict patient-oriented allocation system has the different European countries. Some basic ethical been developed. Due to the expected shorter ischemic principles apply for all countries: equity and justice time and the therefore expected better result of local are considered important, priority is given to urgent and regional allocation, some preference to regional patients, and outcome after transplantation is taken allocation is also given in these patient-oriented sys- into account. The largest differences exist concerning tems. Independent of the allocation rules, the final the relative weight assigned to local allocation in the decision concerning acceptance of an organ offer and different countries. This can be explained based onthe the transplantation of a specific patient always remains historical development of organ transplantation. Sim- with the transplant center.

344 Chapter 40: UK and European service – legal and operational framework

Organ tracking Transplantation, a common European donor identifi- cation number is planned that will allow easier com- United Kingdom munication. It is clearly necessary to be able to track all organs (and tissues) from an individual donor to the several Transplant registry recipients of different organs and tissues. One of the main reasons for this is the possibility of potential or United Kingdom actual disease transmission from the donor to one or more of the recipients. This possibility may be appar- In organ donation and transplantation, comprehen- ent very early after transplantation if relevant results sive and accurate data are essential. The UK Transplant become available from microbiological screening tests Registry (part of NHSBT, and formerly known as UK or postmortem examination of the donor after the Transplant) and the transplant database serve a num- organs have been transplanted. It may also occur many ber of functions: years later, for example, if a recipient sero-converts for 1. Maintain data on all patients waiting for an organ any of the relevant viruses (in particular, hepatitis and transplant human immunodeficiency virus) or develops a tumor, 2. Record data on all organ donors (living and where suspicion arises that the donor may have been deceased) thesourceofthedisease.Amechanismmustbein 3. Document organ allocation place that allows clinicians caring for the other recip- 4. Record data on all transplant operations ients to receive notification of such a possibility and 5. Collect follow-up data on all recipients until to respond appropriately. The UK Transplant Registry transplant failure or death performs this function in the United Kingdom, and 6. Collect follow-up data on all living donors the procedures that are in place are currently being 7. Collect data submitted to the Potential Donor strengthened. Audit Europe There are several specific databases in the United King- dom, such as that on antibody-incompatible trans- In order to both maintain public trust and to increase plants and the paired-exchange living donor program. recipient safety, it is mandatory that the final use of It is not possible to provide detailed examples of all any donor organ offered for transplantation can be these activities, but the Annual Activity Report and the identified. The complete system of tracing and link- references quoted below give an indication of the ways age has to be achieved with strict adherence to the in which the UK Transplant Registry is used. relevant data safety regulations established both on the European and the sometimes even more strict national level. In general, this task is the responsibil- Europe ity of the organ procurement and/or allocation orga- Currently there exists no comprehensive overview nization. Typically, donor and recipient information of how the collection of transplant-related data are is stored after pseudonymization using unique donor organized and structured in the different European and recipient identification numbers. countries. However, in April 2008, the European Par- Inthecaseofcross-borderallocationinvolving liament endorsed a resolution on organ donation multiple (European) organ exchange organizations, and transplantation asking for monitoring of post- organ tracing is a little more complex. Currently every transplant and post-donation results. For this pur- organ exchange organization has its own donor cod- pose, a common method of data analysis is required ing system as described above. Therefore, it is neces- in order to allow optimal comparability of results sary that in case of organ exchange between organiza- across member states. The aim is to ensure the qual- tions, both the donor number of the procuring organ ity and safety of organ donation and transplantation exchange organization and the donor number of the to reduce transplant risks. To achieve this, the Com- transplanting organization is stored. In the course of mission and the member states have been asked to the implementation of the new EU Directive of the launch a pan-European database and communica- European Parliament and of the Council on Standards tion network to interconnect the national databases of Quality and Safety of Human Organs Intended for and provide them with a platform for exchange of

345 Section 8: The transplant service

comprehensive data on organ donation and transplan- UK Department of Health. Legal issues relevant to tations: The European Framework for the Evaluation non-heart beating organ donation. Available at: http:// of Organ Transplants. www.dh.gov.uk/en/Publicationsandstatistics/ Publications/PublicationsPolicyAndGuidance/DH 108825 Further reading UK Department of Health. Organs for transplants: a report Collett D, Sibanda N, Pioli S, Bradley JA, Rudge C. The UK from the Organ Donation Taskforce. Available at: http:// scheme for mandatory continuous monitoring of early www.dh.gov.uk/en/Publicationsandstatistics/ transplant outcome in all kidney transplant centers. Publications/PublicationsPolicyAndGuidance/DH Transplantation 2009; 88: 970–5. 082122 Johnson RJ, Fuggle SV, Mumford L, Bradley JA, Forsythe UK Department of Health. The potential impact of an opt JL, Rudge CJ, Kidney Advisory Group of NHS Blood out system for organ donation in the UK: an and Transplant. A New UK 2006 National Kidney independent report from the Organ Donation Taskforce. Allocation Scheme for deceased heart-beating Available at: http://www.dh.gov.uk/en/ donor kidneys. Transplantation 2010; 89: Publicationsandstatistics/Publications/ 387–94. PublicationsPolicyAndGuidance/DH 090312

346 Section 8 The transplant service Chapter US transplant service – legal and operational framework 41 Walter K. Graham, Richard S. Luskin, and Francis L. Delmonico

Key points clearthatitwouldbenecessarytouseorgansfrom r deceased donors if there was to be significant appli- The Organ Procurement and Transplantation cation of transplant therapy. In these early years, the Network (OPTN) includes all hospitals that surgeons responsible for performing the transplants provide solid organ transplantation as well as also took responsibility for recovering the organs from all organ procurement organizations (OPOs) deceased donors. However, there were severe limi- and tissue-typing laboratories. r tationsontheprocess.Therewasnoformalsystem The OPTN provides support and oversight for identifying potential donors. Organs had to be for every organ transplant in the country for removed very rapidly once the patient became asys- both living donor and deceased donor tolic and could not be preserved ex vivo for any transplants. r extended period of time. This severely limited the United Network for Organ Sharing (UNOS) availability of organs and meant that most organs were is a private membership corporation that is recovered at the transplant center itself. There was very also a non-profit, charitable organization that limited sharing of organs among transplant centers, under federal contract established and and all recovery-related activities were carried out by continues to operate the OPTN. physicians and surgeons. r UNOS collects and maintains all data for the In 1968, the first independent (i.e., non–transplant Scientific Registry of Transplant Recipients center based) organ recovery organization, the Inter (SRTR), which maintains data on every Hospital Organ Bank, was formed by transplant hos- organ transplant recipient in the United pitals in the Boston area. Similar organizations fol- States, including annual follow-up data for lowed in other metropolitan areas with multiple trans- the life of the recipient and/or graft. plant programs as a way to streamline and improve r UNOS maintains a centralized computer the quality of the recovery process. At about the same network linking all organ procurement time, development of the concept of brain death, and organizations and transplant centers. When a thus the ability to maintain the heart beat and respi- donated organ from a deceased donor is rations of potential donors on a ventilator, expanded available, that program will rank order the the pool of potential donors significantly. With the candidates on the waiting list and then offer 1972 passage of the End-Stage Renal Disease Pro- the donated organ according to the rank gram and initiation of Medicare payment for kidney order of the match. transplants of end-stage renal disease patients of all ages, demand for kidneys increased significantly. To The first successful human organ transplant took place meet this demand, deceased organ recovery needed to at the Peter Bent Brigham Hospital in Boston, Mas- expand to include donors at non-transplant centers. sachusetts, in 1954. The living kidney donor was the The logistics of the recovery process made it imperative identical twin brother of the recipient. Although sub- that a new kind of transplant professional, the dona- sequent transplants were limited initially to other tion and transplant coordinator, needed to be created. closely related donors and recipients, it quickly became Thesenewcoordinators,whetherbasedatatransplant

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

347 Section 8: The transplant service

US Federal Government Role in Transplantation

Department of Health and Human Services (DHHS)

Advisory Committe on Transplantation (ACOT)

Heath Resources and Centers for Medicare Other Agencies Services Administration & Medicaid Services (HRSA) (CMS) Figure 41.1 Organization of US National Transplantation Support. Healthcare Systems Bureau (HSB)

Division of Transplartation (DOT)

SR OPTN Contractor CWBYCTP Contractor (UNOS)

center or at an independent organ procurement orga- for bone marrow and cellular transplants, the CW nization, assumed responsibility for public and profes- “Bill” Young Cellular Transplant Program. There is not sional education about donation; assistance with the a single national network for either tissue or cornea pre-recovery and intraoperative clinical management transplantation. of the donor; assisting the surgeon with the recov- ery procedure; and organ packaging, preservation, and transportation. The OPTN history and its current The National Organ Procurement and working authority The National Organ Transplant Act (NOTA) of 1984 Transplantation Network specified that the Secretary of the US Department In 1984, spurred by the shortage of organs and the of Health and Human Services was to provide for lack of oversight and coordination in the organ pro- the establishment and operation of an Organ Pro- curement system, the United States Congress passed curement and Transplantation Network in the private legislation that prescribed the establishment and oper- sector. United Network for Organ Sharing (UNOS) ation of a national network to coordinate organ trans- is a private membership corporation that is also a plantation for the country. That network is known as non-profit, charitable organization (Figure 41.1). It is the Organ Procurement and Transplantation Network located in Richmond, Virginia, and was established in (OPTN). The network includes all hospitals that pro- 1984 by the American Society of Transplant Surgeons vide solid organ transplantation as well as all organ and the South Eastern Organ Procurement Founda- procurement organizations (OPOs) and tissue-typing tion. UNOS was awarded the initial federal contract laboratories. It also includes other organizations and on September 30, 1986, to establish the OPTN. In individuals interested in organ transplantation. It pro- 1987, the Health Resources and Services Administra- vides support and oversight for every organ transplant tion (HRSA) and agency of the US Department of in the country for both living donor and deceased Health and Human Services (HHS) awarded a contract donor transplants. Although many OPOs procure tis- to UNOS to begin operating the network that UNOS sueandcorneas,thenetworkitselfdoesnotprovide had established. UNOS has been awarded successive oversight or support of tissue, cornea, or bone mar- contracts to continue operations since that time, is the row/cellular transplants. There is a separate network only organization to ever manage the OPTN, and has

348 Chapter 41: US transplant service – legal and operational framework continued to administer the contract for more than 22 the OPTN should develop policies governing the net- years and five successive contract renewals. work’s operation through the OPTN committees with As the contractor for operating the OPTN, UNOS approval by the OPTN Board of Directors. Should the collects and maintains all data for the Scientific Reg- OPTN Board desire that any such policy be made bind- istry of Transplant Recipients (SRTR). This registry, ing upon OPTN members as one of the OPTN rules established in 1987, maintains data on every organ and requirements for the purposes of Medicare eligi- transplant recipient in the United States, including bility as defined in the Social Security Act, the Board annualfollow-updataforthelifeoftherecipient would then submit that policy to the Secretary of and/or graft. HRSA awards a separate contract for the HHS for final approval as a federal regulation. Because analysis of SRTR data and producing reports to sup- OPTN policies need to be updated frequently to stay porttheworkofOPTNcommitteesandforthepublic abreast of changes in the field, and federal regulations (notcurrentlyheldbyUNOS). sometimes require long periods of time to amend, no NOTA requires that the OPTN be a private, non- policy has ever been referred to the Secretary for adop- profit entity (i.e., not a government agency) with a tion. Rather, as a way of ensuring policy compliance, Board of Directors to govern the network. The law the OPTN has relied on the contractual agreement of stated that the OPTN should establish a list of all its members. However, because submission of data to persons who need organ transplants as well as a the OPTN is required by the OPTN Final Rule, which national system to match organs with patients wait- is a federal regulation, failure to submit the data to ing for them, maintain a 24-hour telephone service the OPTN, which are required on data forms approved to facilitate matching, assist OPOs in the distribution by the federal Office of Management and Budget, of organs, and coordinate transportation of organs. It would be considered by HHS to be a violation of the also required the OPTN to collect, analyze, and pub- requirements of Section 1138 of the Social Security lish data about organ donation and work to increase Act. the supply of donated organs. The OPTN evaluates member compliance through In 1987, Congress enacted additional legislation routine monitoring of data submission and transplant stating that in order for a hospital that transplanted activity,aswellasconfidentialmedicalpeerreviewof organs or an OPO to be eligible to receive Medi- member complaints. If a member is found to be con- care and Medicaid funds, such hospital or OPO must sistently substandard in performance or is in violation belong to and abide by the rules and requirements of of OPTN requirements, the OPTN provides notice to the OPTN. Since that time, all transplant hospitals and the offending member that they are planning to take organ procurement organizations in the U.S. have been an adverse action against the member as provided in members of the OPTN. the OPTN bylaws, which gives the member the oppor- HHS has determined that OPTN actions may be tunity to appeal. If the offense is serious enough, the regarded as such “rules and requirements” once they OPTN Board of Directors makes recommendations to have been adopted as federal regulations, and only the Secretary of HHS regarding adverse actions that HHS is empowered to deny membership in the OPTN may be taken against the member. As a result of these or participation in the Medicare program. HHS can- enforcement measures, compliance with OPTN poli- not delegate this power to the OPTN as a private body. cies continues to be very high. Because adoption of OPTN policies as a federal reg- ulation is a rare occurrence, each member hospital and OPO has agreed by private contract to abide by The OPTN Final Rule the policies and bylaws of the OPTN. This contractual ThefederalregulationsthatgoverntheOPTNarecom- agreement gives the OPTN the ability to take certain prehensive. They specify the organization and gover- consequential adverse actions designed to bring mem- nance of the network, including detailed requirements bers into compliance when they fail to follow the poli- for the composition of the OPTN Board of Direc- cies and bylaws. tors. In 2010, the OPTN included all 249 US hospi- Effective March 16, 2000, HHS issued federal reg- tals that perform organ transplantation, all 58 OPOs, ulations (the OPTN Final Rule) establishing a reg- all 59 independent histocompatibility laboratories, 17 ulatory framework for the structure and operation medical/scientific organizations or societies, 2 busi- of the OPTN. Under the terms of those regulations, ness members, and 10 individuals.

349 Section 8: The transplant service

OPTN Organization Figure 41.2 Organization of the US Organ Procurement and Transplantation Network. Board of Directors (44)

Executive Committee (12)

Executive Regions Committees Director/Staff (320) (11) (21)

Review External OPTN OPS IT Admin Boards

Policy development 6 TheOPTNFinalRulegivestheOPTNBoardof 6 1 Directors responsibility for developing policies for the 7 OPTN in specific areas: 9 2 r Equitable allocation of cadaveric organs 5 8 10 r Testing of organ donors and follow-up of 11 transplant recipients to prevent the spread of 3 infectious diseases 4 r Reducing inequities resulting from socioeconomic status r The training and experience of transplant Figure 41.3 UNOS and OPTN Regions. surgeons and transplant physicians in designated transplant programs r Nomination of officers and members of the Board During the public comment period, the proposal is of Directors reviewed by all of the other OPTN committees, which r Policies on such other matters as the Secretary may submit comments to the originating committee. may direct UNOS and the OPTN are divided into 11 geographic regions of the United States, and each region will con- OPTN policies and bylaws govern the procedural duct a meeting of its members twice a year for the aspects of policy development, allocation of donated purpose of reviewing policy proposals and providing organs, and collection of transplant data nationwide. feedback to the originating committee and the Board Members and the public are included in the pro- of Directors (see below and Figure 41.3). The timing cess through solicitation of feedback via the pub- of the publication of policy proposals is coordinated lic comment process. UNOS maintains an extensive with these regional meetings so that all regions have number of committees made up of transplant profes- an opportunity to comment on all proposals. Each sionals, patients, donors, donor family members, and region has members representing each committee and others interested in the OPTN’s issues (Figure 41.2). a regional councilor who conducts the meeting and The committees are responsible for developing pol- represents the region on the Board of Directors. This icy proposals. Once a proposal is developed, UNOS’ provides an opportunity for members in a region to staff prepares a written rationale for the proposed pol- give direct input to their representatives. icy, including the scientific data analysis and any other At the end of the public comment period, the orig- background considered by the committee in develop- inating OPTN committee considers and responds to ing the proposal. That proposal and rationale are then each comment received and finalizes the policy pro- publishedfora90-dayperiodwhenanyoneinthe posal for consideration by the Board of Directors. The membership or general public may submit comments Board then decides whether to approve the policy, to UNOS regarding the proposal. amend it, return it to the committee, or reject it.

350 Chapter 41: US transplant service – legal and operational framework

The development of UNOS regions table allocation of organs. Those policies meet certain criteria: As the OPTN contractor, UNOS uses its 11 geographic r regions as the OPTN’s regions. This regional structure Basedonsoundmedicaljudgment r Seek to achieve the best use of donated organs was developed to facilitate organ allocation and pro- r vide individuals with the opportunity to identify con- Preserve the ability of a transplant program to cerns regarding organ procurement, allocation, and declineanofferofanorganornottousetheorgan transplantation that are unique to their particular geo- for the potential recipient r graphic area. Be specific for each organ type or combination of The basis for the regions is historical organ shar- organ types r ing relationships. Prior to the creation of the OPTN, Be designed to avoid wasting organs, to avoid there were seven regional organ banks that were futile transplants, to promote patient access to funded by grants from HHS. The original UNOS transplantation, and to promote the efficient regions were based on the territories covered by those management of organ placement r seven regional organizations. Over time, several of the Notbebasedonthecandidate’splaceofresidence regions were split into more workable sizes. For exam- or place of listing, except to the extent required by ple, the territory covered by the South Eastern Organ consideration of the foregoing principles Procurement Foundation, one of the original seven An important factor for the OPTN in its policy devel- grantees, was divided into regions 3 and 11. The state opment deliberations is the fact that NOTA limits the ofNewYorkwassplitoffoftheNewEnglandregionso policy considerations for organ allocation policy to that it became region 9, and so on. medical criteria. The final rule amplifies this restriction by requiring that the patient rankings resulting from Listing requirements the allocation policy be based on objective and mea- A potential transplant candidate is referred by his or surable medical criteria. Therefore, allocation policies herdoctortoatransplantcenterforevaluation.The do not take into account social criteria such as social transplant center runs a number of tests and considers worth or economic criteria. Thus organ allocation pol- the patient’s mental and physical health, as well as his icy does not consider, for example, whether a patient or her social support system. If the center decides to is an unemployed vagrant as opposed to the CEO of accept this person as a transplant candidate, they will a major corporation. Likewise, the underlying cause add his or her medical profile to the OPTN’s national of a disease such as drug or alcohol addiction is not patient waiting list for organ transplant. The candidate considered by the allocation policies. An individual isnotplacedonarankedlistatthattime.Rather,hisor transplant program may consider whether a patient her information is kept in a constantly updated, com- has maintained a suitable period of abstinence as part puterized database. Whichever OPTN member places of its evaluation of whether the patient is an accept- the patient on the waiting list must pay a “patient reg- able candidate, but once a patient is on the OPTN wait- istration fee” to the OPTN. ing list, whether the patient’s disease was caused by OPTN policies permit a patient to be listed at mul- substance abuse may not be considered because it is tiple transplant hospitals. Organs are to be offered and understood to be “social criteria.” transplanted only for patients who are on the OPTN TheOPTNFinalRuleincludesperformancegoals waiting list. Should a patient be unable to receive a for allocation policies. In addition to the use of objec- transplant, the program may indicate the patient’s sta- tive and measurable medical criteria, the policies are tus on the waiting list as being inactive. Once the to provide for candidate rankings that give priority to patient has undergone transplantation, the program is patients based on medical urgency with patients hav- required to remove the patient from the list within 24 ing the most urgent need receiving highest priority. hours. Also, organs are to be allocated over as broad a geo- graphic area as feasible, considering the principles for allocation included in the Final Rule, and the amount Organ allocation of inter-program variance among performance mea- Specific direction is given by the OPTN Final Rule for suresistobereducedtoaslowalevelascanreasonably the principles to be included in policies for the equi- be achieved.

351 Section 8: The transplant service

Figure 41.4 Donation service areas in the WA` US (58 total areas). WT HD HH 21 LN VT ME MI 1 OR 23 KY 34 1 MA WY 39 D 4 SO Yn 5 7 2 58 NY IA 33 6 RI NE 3 CT N 12 53 48 38 36 9 PA NJ 56 13 51 31 CH 52 10 DE MO L 32 37 ND CA 35 WY YA UT CO 50 14 X3 49 KY 28 AZ MU 22 55 TX TN 29 17 NC 44 26 SC 57 54 43 OK 41 30 47 AR 18 15 27 H 45 25 42 US AL GA 16 LA FL PR & USM 21 46 19 AX 20 8 24

Prior to NOTA there were more than 120 OPOs, by the Board of Directors. The OPO then offers the some independent and some transplant-hospital organ to designated transplant programs in accord- based. However, each operated independently and ing to the rank order of the match. The transplant there was no effective system for sharing organs among program may accept or refuse the offer. The match recovery regions. As part of NOTA’s implementation, is patient-specific as required by NOTA. The organ in 1987 HRSA solicited proposals from organiza- may not simply be offered to a transplant program tions to become the designated OPO for a specific for the program to use at its discretion. If an organ is geographic region of the United States. In 1988, offered to a transplant program for a certain patient, 74 organizations were selected and given exclusive but the program decides to decline the offer, the organ responsibility for all activities related to the procure- is then offered to the next patient on the ranked OPTN ment of organs from deceased donors. These newly match list, even if the patient is at another institution. designated OPOs, all non-profit, were a mix of pre- Exceptions can be made in certain circumstances, but viously existing OPOs and new organizations formed those are not the norm. The match for each donor by the merger of OPOs that had served overlapping organ will be different and unique to the circumstances geographic regions. By 2010, the number of OPOs of the donor and the patients waiting. Factors affect- was reduced to 58 through mergers of geographically ing ranking may include tissue match, blood type, adjacent OPOs (Figure 41.4). Some consolidations length of time on the waiting list, immune status, and were voluntary, whereas others were the result of HHS the distance between the potential recipient and the enforcement of standards of performance. donor. For heart, liver, lung, and intestines, the poten- tial recipient’s degree of medical urgency is also con- Identification of the organ recipient sidered. The organ is offered to the transplant team Under its OPTN contract with HRSA, UNOS main- for the first person on the match list. Often, the top- tains a centralized computer network linking all organ ranked patient will not get the organ for one of sev- procurement organizations and transplant centers. eral reasons. When a patient is selected, he or she This computer network is accessible 24 hours a day, must be available, healthy enough to undergo major 7 days a week, with organ placement specialists in surgery, and willing to undergo transplantation imme- the UNOS Organ Center always available to answer diately. Also, a laboratory test to measure compatibil- questions. When a donated organ from a deceased ity between the donor and recipient may be necessary. donor is available, the OPO procuring the organ must For example, patients with high antibody levels often enter information about the donor into the OPTN prove incompatible to the donor organ and cannot contractor’s computer system and execute the com- receive the organ because the patient’s immune system puter match program. That program will rank order would reject it. Once a patient is selected and contacted the candidates on the OPTN waiting list according to and all testing is complete, the transplant procedure is the organ allocation policies that have been adopted scheduled.

352 Chapter 41: US transplant service – legal and operational framework

Transplant program requirements mendations to the Secretary of HHS in 2002 after a year’s deliberations. Included among those recom- and evaluation mendations were several regarding the need for pro- TheOPTNFinalRulestatesthatorgansmayonly tections for living donors. Specifically, the ACOT rec- be offered to transplant programs that are “desig- ommended that qualifications of programs perform- nated programs.” In order to become designated, a ing living donor transplants be verified, that ethical transplant program must be approved as a Medicare principles and informed consent standards be imple- transplant program or be part of a federal govern- mented for all living donors, and that certain informed ment hospital (e.g., Veterans Administration hospi- consent practices be instituted, including the require- tals) or meet certain requirements defined by the ment of an independent donor advocate at each trans- OPTN. Those criteria include minimum standards for plant center performing living donor transplantation. the education, training, and experience of a primary In June 2006, HRSA issued a directive to the OPTN to surgeon and a primary physician for the transplant develop and enforce policies regarding living donors program. OPTN Bylaws require that all programs and living donor transplant recipients. meet the criteria for a primary surgeon and physician In response, UNOS, together with the American regardless of Medicare certification, and OPTN mem- Society of Transplant Surgeons and the American bership approval may include a site review designed to Society of Transplantation, convened a consensus con- ensure that the necessary resources and personnel are ference to develop guidelines for the psychosocial eval- available to provide transplantation services. uation of prospective living kidney donors who have The OPTN oversees transplant center performance neither a biological nor longstanding emotional rela- by monitoring center-specific reports provided by the tionship with the transplant candidate (i.e., unrelated SRTR contractor on a quarterly basis. In these reports, donors). Those recommendations were adopted by center-specific data are compared with expected risk- the OPTN, which also developed membership crite- adjusted outcomes using three well-defined criteria ria for transplant programs that perform living donor adopted by the OPTN for quality assurance. If a pro- transplantation and a series of guidelines/policies for gram has results that meet all three criteria, the OPTN providing informed consent to potential donors as Membership and Professional Standards Committee well as the medical evaluation of potential living (MPSC) commissions a group of reviewers who inves- donors. tigate potential causes for inferior outcomes and then Transplant centers are required to submit reports report back to the MPSC. Based on their findings, the to the OPTN on all transplants, including living donor MPSC makes recommendations regarding the neces- transplants, even when both donor and recipient are sary steps required to improve outcomes. The OPTN from another country and in the United States only for continues to monitor performance until outcomes are the transplant. Transplant programs are also required improved. If serious deficiencies persist and are not to report all instances of living donor deaths and fail- corrected, the OPTN may take actions against the ure of the living donor’s native organ function to the transplant center, including a letter of warning, a let- OPTN within 72 hours after the program becomes ter of reprimand, declare the transplant center to be aware of such an event. In 2005, HRSA determined that on probation, or declare it to be a “member not in good failure by a transplant center to report required data standing.” If the problem is serious enough, the OPTN totheOPTNwouldbeaviolationoftheOPTNFinal may recommend that HHS remove the transplant pro- Rule, and as such would be a violation of OPTN “rules gram’s “designated” status, which will deny the trans- and requirements,” which could lead to the entire hos- plant center access to deceased donor organs for trans- pital being made ineligible to receive Medicare or Med- plant. icaid payments.

Living donor organ transplantation Success and ongoing challenges Neither NOTA nor the OPTN Final Rule issued in of the US system 2000 included living donor transplantation as part Organ transplantation is different from any other of the OPTN’s charge. The Advisory Committee on medical service because transplantation practice does Transplantation (ACOT) issued its first set of recom- not rely solely on professional expertise. The success

353 Section 8: The transplant service

of transplantation necessitates either a willing living part of medicine and surgery in the United States can donor or suitable deceased organ donor. Society has an claim such successes for its entire field. oversight responsibility to assure a just distribution of deceased organs and the proper care of the live donor. The OPTN system fulfills that societal responsibility Further reading in that every provider of organ transplantation partici- Luskin RS, Glazier AK, Delmonico FL. Organ donation and dual advocacy. NEnglJMed2008; 358: 1297–8. pates in the network under a model of self-governance with federal oversight. The practitioners who provide McDiarmid SV, Pruett TL, Graham WK. The oversight of solid organ transplantation in the United States. Am J the medical and surgical services also develop and Transplant 2008; 8: 739–44. implement the standards and policies by which the Organ Procurement and Transplantation Network. field is governed with federal oversight. One of the Available at: http://optn.transplant.hrsa.gov greatest strengths of the system is the fact that data for every organ, every transplant, every organ donor, Pretagostini R, Gabbrielli F, Fiaschetti P, et al. Risk management systems for health care and safety and every transplant recipient are collected and main- development on transplantation: a review and a tained in the OPTN databases for analyses and publi- proposal. Transplant Proc 2010; 42: 1014–6. cation. These data are widely available to researchers, Tuttle-Newhall JE, Krishnan SM, Levy MF, McBride V, enabling the field to continually improve. The data Orlowski JP, Sung RS. Organ donation and utilization in are assessed for performance of individual institutions, the United States: 1998–2007. Am J Transplant2009; 9: and through a process of confidential medical peer 879–93. review, efforts are made to improve those institutions United Network for Organ Sharing. Available at: http:// that are not performing at an acceptable level. No other www.unos.org

354 Section 8 The transplant service Chapter Conclusions

42 Clive J. Lewis

Key points plant occurred in 1931 as part of gender reassign- r Transplant medicine has more Nobel Prize ment surgery, but the patient died from acute rejec- winners than any other field. tion. Partial success was seen with a functioning uterus for approximately 3 months in 2000. Potential recipi- r Uterus and ovary transplant may allow ents include those with congenital absence of a uterus successful pregnancy. (e.g., Mayer-Rokitansky-Kuster-Hauser syndrome) or r Biological heart valves are an essential part of hysterectomy due to a number of causes, including the cardiac surgeon’s armamentarium. obstetric complication (e.g., postpartum hemorrhage), r Thymustransplantisanimportant pelvic inflammatory disease, fibroids, and cervical or therapeutic option for infants with Di George ovarian malignancies. Experimental animal studies syndrome. r have established the feasibility of uterus transplant, Cell transplant and tissue engineering is a including donor retrieval techniques, cold ischemic significant area of research and may provide tolerance of greater than 24 hours, and the surgi- great strides in future treatment for a wide cal implant procedure, although problems remain variety of diseases. with vascular anastomosis and thrombosis. Pregnan- cies have been reported in syngenic animal mod- Transplantation has evolved enormously over the last els. Other remaining issues include ethical concerns 50 years, with many significant milestones and with and immunosuppression regimes with their inher- the contribution of many individuals of all disci- ent complications. Human uterine transplant remains plines to make transplant medicine an exciting and experimental but may soon reach a stage for further research-focused area with more Nobel prizes than attempts. any other field! Every conceivable cell, tissue, and organ has been transplanted either in an experimental modelorfullclinicaltransplantprogram.Thisbook Ovary transplantation has considered the most commonly transplanted tis- The objective of ovarian transplantation is similar sues and solid organs, yet much research continues to that of uterus transplantation in that successful in the cell laboratory, in animal models, and in early engraftment would result in fertility. Research in this human clinical studies. This chapter briefly outlines area was largely taken over by the successes of in vitro anumberoftheseareasthathavenotalreadybeen fertilization and uterine implantation. However, in discussed. select situations (e.g., following chemo/radiotherapy for childhood cancers), implantation of autotrans- planted previously frozen ovary (partial or total) has Uterus transplantation been reported to give rise to successful pregnan- Humanuterustransplantmayhavearoleasafuture cies. Syngeneic ovary transplantation has also been method to treat uterine causes of infertility. How- reported. Both of these methods have the advantages ever, there remain significant barriers, including sur- of avoiding immunosuppression and providing long- gical and ethical factors. The first human uterus trans- term benefits of endogenous estrogen.

Organ Transplantation: A Clinical Guide, ed. A.A. Klein, C.J. Lewis, and J.C. Madsen. Published by Cambridge University Press. C Cambridge University Press 2011.

355 Section 8: The transplant service

Cardiac valves and vessels from donor infants undergoing cardiac surgery and, following culture, is usually placed ectopically (e.g., When hearts are not procured for human transplan- into quadriceps muscle). Approximately 50% require tation, the valves and vessels may be harvested for immunosuppression. Reported mortality rates are useinconventionalcardiacsurgery.Aorticandpul- 25% at 1 year, although thereafter, survival is excellent monary valves may be taken together with a length with minimal subsequent mortality over many years. of artery (homograft) or excised alone and placed Recipients also appear to develop stable immuno- in valve support structure or conduits for ventricu- constitution and responses. Complications include loarterial valve replacement. These have the advan- graft-versus-host disease and long-term immunosup- tage of immune privilege and therefore do not require pression. immunosuppression. Homograft tissue is likely to lead to the development of human leukocyte antigen sen- sitization, which may be important in patients who Cell-based therapies may require organ transplant in the future. The use Transplanting cells derived from fetal tissue (embry- of animal tissue (xenograft) for transplantation has onic stem cells; ESCs) have the potential to improve a been long considered one answer to the shortage of wide variety of human disease because they can differ- donor organs. Primate to human organ transplants entiate and replace damaged tissues without the need have failed due to rapid immune recognition and rejec- for immunosuppression and their inherent complica- tion. Significant advances in immunobiology will be tions. This area of research is likely to make the biggest needed before xenotransplantation becomes a real- ever impact on human disease. A number of stem cell ity. However, animal valves (particularly porcine and therapeutics exist, but most are at experimental stages. bovine) have been successfully used in human cardiac Only bone marrow transplantation is in current clin- valvereplacementformanyyearswithouttheproblem ical use. Potential areas of treatment include spinal of rejection. cord injury (neural ESC – oligodendrocyte progeni- tor cells; the subject of a clinical trial), diabetes melli- Thymus tus (insulin-producing ESC-derived beta cells), heart Thymustransplantationhasbeenusedtotreatthe failure (ESC-derived cardiac stem cells), and degen- rare condition of complete athymia in Di George erative disease, such as juvenile macular degenera- syndrome. Di George anomaly is characterized by tion (subretinal transplantation of ESC-derived reti- varying defects of the heart, thymus, and parathy- nalstemcells),Parkinson’sdisease(substantianigra– roid glands. Infants may present with very low grafted ESC-derived dopaminergic cells), and Hunt- T-cell counts or develop oligoclonal T cells that pro- ington’s disease (striatum-grafted ESC-derived striatal duce a rash and adenopathy and are at increased cells). This is likely to be a very exciting area of devel- risk of infection. Thymus for transplant is obtained opment in the next decade.

356 Index

Note: Page numbers followed by f refer to figures; page numbers followed by t refer to tables.

A1-antitrypsin deficiency, 140t, 140, treatment, 26 (See also portal venous, 195f, 194–5, 201, 221, 222t Immunosuppression; specific 196 Abdominal organ transplantation drugs and drug classes) reperfusion, bile duct, 196–8 history, 3–4 Acute viral hepatitis, 223 Angiotensin-converting enzyme ABO incompatibility Adaptive immune system effector (ACE) inhibitors, 273t desensitization protocols, response, 13–14 Angiotensin receptor blocker (ARB) 232 Adaptive immunity. See Acute inhibitors, 273t kidney transplantation recipient rejection Antibodies in transplant arteriopathy sensitization, 244–5 Adult-to-Adult Living Donor Liver 40–1 living donor liver transplantation, Transplantation Cohort Study Antibody induction 188 (A2ALL), 183 immunosuppression, 21, 22–3, pediatric heart transplantation, 116, AEB071 (sotrastaurin), 27–9 25t, 24–5, 97 117t Age Antibody-mediated rejection (AMR) ACE (angiotensin-converting enzyme) heart transplantation recipient chronic, 267–8, 271 inhibitors, 273t selection, 66 donor-specific antibody (DSA) in Activator protein 1 (AP-1), 12, as infection factor, 66 development of, 239, 242t, 13 kidney transplantation recipient 242–3 Acute liver failure (ALF), 223t, 223, selection, 237t, 236–7 kidney transplantation, 263 224 as lung transplantation comorbidity, outcome following desensitization, Acute rejection 124–5 247 adaptive immune system effector Alagille syndrome, 222t, 221–2 overview, 99f, 98–9 response, 13–14 Alcohol-related malignancy, 36. See pathology, 13–14 allorecognition pathways, 11f, 10–11 also Smoking, alcohol abuse risk assessment, crossmatch/ antibody-mediated rejection (AMR) Alefacept, 29–30 antibody screening based, 240, (See Antibody-mediated Alemtuzumab (Campath) 241t, 242t rejection (AMR)) desensitization protocol, 246 treatment, 26, 159 CD4+ T cells in, 11f, 13 historical applications of, 21 Antigen-presenting cells (APCs), 11f, CD8+ T cells (cytotoxic T intestinal transplants, 308t 10–11 lymphocytes (CTLs)) in, 11f, 13, islet transplantation trials, 300 Antihypertensive medications, 259, 14 in maintenance 273t clinical presentation, 152 immunosuppression, 25 Anti-lymphocyte globulin (ALG), costimulation pathways, 11–12 mechanism of action, 16 19–20 in heart recipient mortality, 104, overview, 22, 97 Antimetabolites mechanism of action, 105t trends in, 4 16 HLA, non-HLA antibodies clinical ALG (anti-lymphocyte globulin), Anti-proliferative agents ADRs, 264 relevance, 240t, 239–40 19–20 Anti-thymocyte globulin (ATG, rATG) immunological synapse, 12 Allograft transplantation history historical applications of, 19–20 immunosuppression (See abdominal organs, 3–4 in maintenance Immunosuppression) cardiothoracic, 4–6 immunosuppression, 25 initiation, 11f, 10–11, 13 combined heart/lung, 6–7 overview, 22, 97 innate immune system effector future directions, 8 AP-1 (activator protein 1), 12, 13 response, 14–15 lung, 7 ARB (angiotensin receptor blocker) interleukin-17 (IL-17) in, 11f, 14 overview, 1–3 inhibitors, 273t leukocyte recruitment, 12 Alloway, 26–7 Artificial Heart Program, 76–7 lipid rafts, 12 Alpha blockers, 273t Aspergillus infection, 50–1, 153–4, 158, memory cells, 12–13 Anastomoses 228 overview, 9 donor lung, 142 ATG. See Anti-thymocyte globulin prophylaxis, 25t, 24–5, 26 hepatectomy, caval, 194f, 193–4 (ATG, rATG) re-transplantation, 68 hepatic artery, 196, 197f ATGAM trial, 19–20 Th2 cells in, 13–14 historical applications, 2, 5, 7 Atherosclerosis, 326t

357 Index

Atraumatic intracranial bleeds (aICB), drug interactions, 23t definition, 38 74 mechanism of action, 16 glomerular basement membrane Autograft transplantation history, 2 overview, 23t, 23, 273t duplication in, 42f, 41–2 Autoimmune hepatitis (AIH), 222 reduction, withdrawal, 270b, 270, heart, 43, 44f Avascular necrosis, 326t 271b, 271 liver, 43f, 42–3, 215–16 Azathioprine side effects, ADRs, 263–4, 270, lung, 44f, 43–4 historical applications of, 3, 6, 19–20 276–7 overview, 38 mechanism of action, 16 Calcium channel blockers, 273t transplant arteriopathy (See overview, 97–8 Campath. See Alemtuzumab Transplant arteriopathy) reduction, withdrawal, 270b, 270, (Campath) transplant glomerulopathy, 42f, 271b, 271 Cancer. See Malignancy 41–2 side effects, ADRs, 276–7 Candida infection, 50–1, 153–4, 158, treatment, reversibility, 44–5 Azithromycin, 160–1 228 Clevian classification system, 185t Carbon monoxide poisoning, 71 Coagulopathy, 92–3 Basal cell carcinoma (BCC), 32 Cardiac allograft vasculopathy Coccidioidomycosis, 228 Basiliximab clinical presentation, diagnosis, Collaborative Transplant Study, 73 islet transplantation trials, 300 107t, 107 Combined heart-kidney, heart-liver in maintenance outcomes, 108f, 107–8 recipient selection, 68 immunosuppression, 25 overview, 43, 44f, 106t, 106 Combined heart/lung transplantation mechanism of action, 16 pathophysiology, 106–7 historically, 6–7 overview, 22–3, 97 pediatric heart transplantation, recipient selection, 127 Belatacept, 29, 300 118f, 118–19 surgical procedure, 135–6 Benzodiazepines, 90 treatment, 108, 109t Complement cascade in acute Beta blockers, 273t Cardiothoracic transplantation, 4–6, rejection, 14–15 B7 family, 11–12 31. See also under Combined Composite tissue allotransplantation Biliary atresia, 221, 222t, 226 heart/lung transplantation; (CTA), 313–14 Biologics, 29. See also Monoclonal Heart transplantation; Lung Corneal transplantation antibodies transplantation anatomy, 330f, 330–1 BK virus CD28, 11–12 donor cornea processing, 331 management, 268–9, 271 CD80/86, 11–12 history, 1–2, 330 overview, 48–9 CD154, 11–12 immune privilege in, 331–2 Bladder transplantation, 31 CD27/CD70, 11–12 outcomes, 333t, 333–4 Bortezomib, 29 CD279/CD274, 11–12 rejection clinical features, treatment, BOS. See Bronchiolitis obliterans CD8+ T cells (cytotoxic T 333f, 333 syndrome (BOS) lymphocytes (CTLs)), 11f, 13, rejection immunology, 332–3 Brain, brainstem death physiology 14 surgical techniques, 331, 331t, 331f cardiovascular changes, 54–5 Cell-based therapies, 356. See also Corticosteroids. See also specific drugs endocrine changes, 55–6 Hematopoietic stem cell desensitization protocol, 246 inflammatory, immunological transplantation graft rejection management, 100 changes, 56 Cerebral oximetry, 89 historical applications of, 3, 19–20 overview, 53–4 Chemokines in acute rejection, 12 intestinal transplants, 308t pulmonary changes, 55 Chemotherapy. See also specific drugs as lung transplantation comorbidity, temperature regulation, historical applications of, 6 124 consumptive coagulopathy, 56 PTLD treatment, 35 in maintenance Bronchiolitis obliterans syndrome trends in, 4 immunosuppression, 25 (BOS) usage historically, 3 mechanism of action, 15 in children, 168, 172 Child-Pugh scores, 175t, 175–6 pediatric liver transplantation, development, comorbidities, 157, Chlorpheniramine, 308t 225–6 160t Cholangiocarcinoma, 177t reduction, withdrawal, 270b, 270, overview, 159–61 Cholangitis, recurrent, 177t 271b, 271 severity criteria, 160t Chronic allograft nephropathy, 267t, side effects, ADRs, 264, 276–7 in stem cell transplantation, 326t 266–7, 270, 272f, 284 Crigler-Najjar syndrome, 225 Burkholderia spp., 123, 165–6 Chronic hepatic encephalopathy, 177t Cryptococcus infection, 228 Byetta (exenatide), 299 Chronic obstructive pulmonary CTA (composite tissue disease (COPD) allotransplantation), 313–14 Calcineurin inhibitors (CNIs). See also lung transplantation recipient CTLA-4, 11–12, 29 Cyclosporine; Tacrolimus selection, 125, 126t, 140t, 140 Cutaneous transplantation. See Skin (FK-506, Fujimycin) as stem cell complication, 326t transplantation clinical study data, 23, 97–8 Chronic rejection pathology Cyclophosphamide, 3, 322

358 Index

Cyclosporine Diuretic-resistant ascites, 177t Face, scalp transplantation, 313–19 in combination therapy, 24 Dobutamine, 91, 95 Familial amyloidosis, 177t historical applications of, 4, 20–1 Donation after brain death (DBD) Fentanyl, 90 mechanism of action, 16 cardiovascular management, 57f, FK-506. See Tacrolimus (FK-506, monitoring, 98, 157 58–9 Fujimycin) overview, 23t, 23, 97–8 definition of death (UK, Europe), Focal segmental glomerulosclerosis reduction, withdrawal, 270b, 270, 335–6, 340t (FSGS), 268, 284 271b, 271 euvolemic donor vasoactive drug 4-IBB, 11–12 side effects, ADRs, 276–7 support, 58–9 Fujimycin. See Tacrolimus (FK-506, trends in, 4 hormone replacement therapy, 59 Fujimycin) Cystic fibrosis initial resuscitative measures, 56 Future liver remnant, 184 evaluation of, 126, 126t kidney transplantation surgical infections and, 152–3 procedure, 256 Gastroesophageal reflux (GERD, intra-operative ventilation, 140–1 post-consent donor management, GORD) as living donor lung transplant 57f, 56–7 as absolute contraindication, 123, indication, 130 respiratory management, 59 124 nutrition and, 148 Donation after cardiac death (DCD) as BOS risk factor, 160 overview, 140t heart transplantation, 60–2 Glomerular basement membrane in pediatric lung transplants, 164–5, lung transplantation surgical duplication, 42f, 41–2 166t, 171 procedure, 136 Glomerulonephritis, recurrent, 268 Cytokine levels, post-brain, brainstem Donor organs. See under Organ Gondal failure/infertility, 326t death, 56 donation Graft rejection Cytomegalovirus Donor-specific antibody (DSA), 239, acute (See Acute rejection) in children, 169 242t, 242–3. See also antibody-mediated, 40–1 overview, 48–9, 153, 158 Antibody-mediated rejection antibody-mediated rejection (AMR) pediatric liver transplantation, 228 (AMR) (See Antibody-mediated prophylaxis, 96–7, 104–5, 147–8, DVT (deep venous thrombosis) rejection (AMR)) 260t, 260 prophylaxis, 259 chemotherapy usage historically, 3 Dyslipidemia, 162, 274–5 classification, 99–100 Daclizumab history, 1–2, 6 islet transplantation trials, 300 EBV-specific cytotoxic T-Cells, 34 management, 100 mechanism of action, 16 Eculizumab, 30 monitoring, 99, 100f overview, 97 Edmonton protocol, 300 Tcellmediated(See Tcellmediated DAG (diacylglycerol), 12 Efalizumab, 29, 300 rejection) Dead donor rule (Europe), 336 Efficacy Limiting Toxicity Elimination types, 99f, 98–9 Deep venous thrombosis (DVT) (ELITE) – Symphony study, Graft to recipient body weight ratio, prophylaxis, 259 24–5 184 Demikhov, 5, 6–7 Eisenmenger syndrome, 166t Dendritic cells (antigen-presenting Empyema, 150 Heart transplantation cells (APCs)), 11f, 10–11 Enoximone, 91 allocation, UK/Europe, 344 Desensitization, 242t, 243t, 245t, Epoprostenol, 143–4 cardiac valves, vessels, 356 244–5, 246f, 246 Epstein-Barr virus chronic rejection pathology, 43, Diabetes mellitus. See also Pancreas as immunosuppression marker, 156 44f transplantation; Pancreatic islet overview, 48–9 combined heart/lung (See transplantation pediatric liver transplantation, 228 Combined heart/lung heart transplantation postoperative in PTLD, 33–4, 36 (See also transplantation) management, 96 Post-transplant donation after cardiac death (DCD), hyperglycemia, 109 lymphoproliferative disorder 60–2 kidney transplantation recipient (PTLD)) donor organ demand, 7–8 selection, 236 Eptacog alpha activated (recombinant history, 4–6, 20 lung transplantation long-term factor VIIa), 204 indications, contraindications, 64t, management, 162 Escherichia coli, 276 63–4, 64t new onset diabetes after Etanercept, 299 Mann preparation, 5 transplantation (NODAT), 274t, Etopside, 322 overview, 63 274 Everolimus pediatric (See Pediatric heart post-liver transplantation, 214t, 214 mechanism of action, 16 transplantation) pre-transplant, 274 monitoring, 98, 157 Heart transplantation donor selection, Diacylglycerol (DAG), 12 overview, 24 70–5 Di George Syndrome, 356 Exenatide (Byetta), 299 in recipient mortality, 104

359 Index

Heart transplantation long-term reduced intensity conditioning, 323, Immediate blood mediated management 324f inflammatory reaction (IBMIR), complications sources, 322t, 321–2 298, 302 cardiac allograft vasculopathy Hemolyticuremicsyndrome(HUS), Immunoglobulin (IVIg) (See Cardiac allograft 268, 284 antibody-mediated rejection vasculopathy) Hemothorax, 150 therapy, 26 chronic renal dysfunction, 110 Hepatic artery thrombosis desensitization protocol, 245–6 late cardiac allograft dysfunction, in children, 226 polyoma virus management, 271 110–11 post-liver transplantation, 208t, 209, Immunology history, 2, 3 malignancy, 110t, 110 217 Immunosuppression. See also specific metabolic syndrome, 108–9 Hepatitis drugs and drug classes; specific overview, 106, 106t acute viral, 223 procedures mortality, 104f, 103–4, 105t, 105 autoimmune, post liver acute rejection prophylaxis, 25t, post-first year survival, 106t, 105–6 transplantation, 217 24–5, 26 (See also Acute quality of life, 111 in heart donor selection, 70–1 rejection) survival rates, 102, 103f in heart recipient selection, 67 antibody induction, 21, 22–3, 25t, Heart transplantation management kidney transplant recipient 24–5, 97 during surgery, 88–93 selection, 234 antivirals, antifungals generally, 21 Heart transplantation postoperative living donor liver transplantation, calcineurin inhibitors (CNIs) (See management 183 Calcineurin inhibitors (CNIs)) atrioventricular arrhythmias, 96 overview, 48–9 clinical principles, 97–8 diabetes mellitus, 96 post liver transplantation, 216 desensitization protocol, 246 early allograft-related hemodynamic Hepatoblastoma, 222–3 drug interactions, 23t events, 94–5 Hepatocellular carcinoma, 32t, 35–6, future directions, 27, 28f hemodynamic stability 179, 183 generic drugs, 26–7 maintenance, 95–6 Hepatopulmonary syndrome, 177t, history, 4, 6, 19–21 immediate, 87 201 maintenance, 23, 25t, 25, 156–7 immunosuppression principles, Hereditary haemorrhagic malignancy with, 275–6 97–8 (See also telangiectasia, 177t mechanism of action, 15f, 15 Immunosuppression) Herpes virus negative selection (chimerism) in, infection prophylaxis, 96–7 overview, 48–9, 153 17–18 overview, 94 pediatric liver transplantation, 228 non-nephrotoxic, 271 rejection (See Graft rejection) prophylaxis, 104–5 overview, 19, 21–2 renal failure, 96 HIV infection, 67, 71, 234 pediatric lung growth rates and, 171 therapeutic drug monitoring, 98 Human leukocyte antigen (HLA) side effects, ADRs, 276–7 Heart transplantation recipient allorecognition mediation by, 10–11 small molecules, 27 selection, 63–8 clinical relevance, 240t, 239–40 T-cell depleting induction, 300 Heart transplantation surgical HUS (hemolytic uremic syndrome), tolerance induction generally, 16–17 procedure, 83–7 268, 284 Treg in, 17 Hematopoietic stem cell Hyperlipidemia, 109, 177t, 214t, trends in, 4 transplantation 214 Inducible T cell costimulator (ICOS, complications Hypertension CD278), 11–12 early, 323–4, 325t anti-hypertensive agents, 273t Infection graft vs. host disease (GvHD), causes of, 273t as absolute contraindication, 123 326t, 326–7 management, 272f, 273t, 272–3, 273t age as factor, 66 hemorrhagic cystitis, 324 post-liver transplantation, 214t, 214 bacterial, 49–50, 152–3 infections, 324, 325f pulmonary (See Pulmonary donor-derived, 47–8 late, 326t hypertension) fungal, 50–1, 104–5, 153–4, 158, venooclusive disease, 180t, Hypothermia historical applications, 5 228 324–6 Hypothyroidism, 326t in heart recipient mortality, 104–5 conditioning principles, 323 Kaposi’s sarcoma, 32–3 definitions, 321 Idiopathic pulmonary arterial kidney transplantation recipient donor lymphocyte infusions, hypertension (IPAH), 126t, 126, selection, 234–5 327 140t overview, 46–8 donors, 321–2 Idiopathic pulmonary fibrosis (IPF), parasitic, 51–2, 104–5 indications, 328f, 327–8, 329 126t, 126–7, 140t prophylaxis, 96–7, 235, 260 organisations, 329 Idiopathic pulmonary hypertension, pulmonary, 123, 147–8 outcomes, 327–9 166t risk management, 47, 104–5 principles, rationale, 320–1 IgA nephropathy, 268 sepsis, 70

360 Index

timeline, 47f, 46–7 antibody induction new onset diabetes after viral, 48–9, 67, 70–1, 104–5, 153, 158 chemotherapy usage historically, 3 transplantation (NODAT), 274t, Infertility/gondal failure, 326t donor organ demand, 7–8 274 Inositol triphosphate (IP3), 12 history, 2, 20 non-adherence, 271–2 Interleukin-2 (IL-2), 12 live donor (See Living donor kidney overview, 265 Interleukin-17 (IL-17), 11f, 14 donation) polyomavirus associated Intestinal transplantation maintenance immunosuppression, nephropathy, 268–9, 271 combined kidney, 307 25 pregnancy, 277 complications malignancy recurrence in, 31 pre-transplant diabetes, 274 acute rejection, 309–11 pediatric (See Pediatric kidney PTLD (See Post-transplant infection prophylaxis, 308 transplantation) lymphoproliferative disorder systemic sepsis treatment, 310t, skin, lip, 32 (PTLD)) 311 twins historically, 3 recurrent glomerulonephritis, 268 thrombosis prophylaxis, 308 Kidney transplantation complications renovascular disease, 268 graft type selection, 306 acute rejection, 262–3 transplant bone disease, 276 immunosuppression, 308t, 307–8 antibody-mediated rejection transplant glomerulopathy, 267, 271 indications, 305t, 307 (AMR), 263 (See also urinary tract infection (UTI), 276 intercurrent surgery, 312 Antibody-mediated rejection Kidney transplantation perioperative nutrition, 308–9 (AMR)) care outcomes, 303–4 delayed graft function, 261 antihypertensive medications, 259, pediatric, 305t, 307 graft thrombosis, 262 273t physiotherapy, 309 immunosuppression, 263–4 antiplatelet, anticoagulation therapy, post-discharge management, 311 lymphocele formation, 262 259 postoperative mortality, 311f, 311 postoperative hemorrhage, 261–2 consent, 260 pregnancy, 312 T-cell mediated rejection, 263t, 263 dialysis, 259 preoperative management, 307 urological, 262 evaluation, 258–9 process, requirements of, 305–6 wound, 261 fluid management, 260–1 recipient selection, management, Kidney transplantation long-term immunosuppression, 259 304, 305t management oliguric/anuric patient evaluation, renal function, 308 anti-proteinuric measures, 272 261 salt, water balance, 308 atypical hemolytic uremic syndrome postoperative analgesia, 261 splenectomy, 307 (HUS), 268 postoperative early phase, 260–1 surgical procedure, 307 cardiovascular morbidity, mortality prophylaxis, 260t, 259–60 timing of, 304–5 prevention, 272 Kidney transplantation recipient tissue typing, 311 chronic allograft nephropathy, 267t, selection travel, 312 266–7, 270, 272f ABO incompatibility, trends in, 303f, 303 chronic antibody mediated desensitization protocols, 232 Intra vascular ultrasound (IVUS), 107 rejection, 267–8, 271 age, 237t, 236–7 IP3 (inositol triphosphate), 12 chronic graft dysfunction, 267t, cardiovascular disease, 233 IPAH (idiopathic pulmonary arterial 266–7, 270, 272f combined liver/kidney, 179–81 hypertension), 126t, 126, 140t chronic kidney disease, 277 contraindications, 232t, 232 IPF (idiopathic pulmonary fibrosis), CNIs (See Calcineurin inhibitors diabetes, 236 126t, 126–7, 140t (CNIs)) hypercoagulable states, 233–4 Islet transplantation. See Pancreatic death with functioning graft, 266 infection, 234–5 islet transplantation dyslipidemia, 274–5 ischemic heart disease, 233 Isoproterenol, 91, 95 focal segmental glomerulosclerosis, list management, 237 268, 284 malignancy, 235 Janus kinases (Jak), 27 graft, patient survival, 265–6 obesity, 236 JC virus graft function improvement overview, 231 management, 268–9, 271 strategies, 267t, 269t, 269, 270 paired exchange, kidney list overview, 48–9 graft loss, causes, 266t, 266, 266f donation, 232–3 hypertension, 272f, 273t, 272–3, panel reactive antibody (PRA), 232 Kaposi’s sarcoma, 32t, 32–3 273t peripheral vascular disease, 233 Keratoconjunctivitis sicca, 326t IgA nephropathy, 268 preoperative evaluation, 231 Ketamine, 90 immunosuppression with primary renal disease, 236 Kidney transplantation malignancy, 275–6 psychiatric, cognitive, psychosocial acute rejection prophylaxis, 25t, infection, 276 evaluation, 237 24–5, 26 malignancy, 275f, 275f, 275–6 (See recurrence, 236 allocation, UK/Europe, 344 also Malignancy) structural heart disease, 233

361 Index

Kidney transplantation recipient donor organ demand, 7–8 electrolyte, acid-base changes, 205, selection (cont.) future directions, 219 206 timing of referral, 231 hepatitis, 216 fluids, 204 tissue typing, histocompatibility, 232 history, 3–4, 20 glucose control, 205–6 tobacco use, 233 living donor (See Living donor liver hepatopulmonary syndrome, urologic disease, 235 transplantation) 201 Kidney transplantation recipient pediatric (See Pediatric liver ICU care, 208–9 sensitization transplantation) immediate preoperative ABO incompatibility, 244–5 quality of life, 218–19 preparation, 203 acceptable mismatch programs, 244 tobacco, alcohol-related malignancy, immunosuppression, 209–10 causes of, 239 36 monitoring, 203–4 definitions, 238 Liver transplantation complications obstructive airways disease, 201 desensitization, 242t, 243t, 245t, acute rejection, 208t, 210, 215–16 post ICU care, 209–10 244–5, 246f, 246 autoimmune hepatitis, 217 postoperative care, 207–8 HLA, non-HLA antibodies clinical bile leak, 208t, 210 post-reperfusion syndrome, 206 relevance, 240t, 239–40 diabetes, 214t, 214 preoperative assessment, 200t, HLA incompatibility, 242t, 245 graft deterioration, 208t, 210 199–200, 202 options for, 243–6 hemorrhage, 208t, 209 cardiovascular function, 199–200 outcome following desensitization, hepatic artery thrombosis, 208t, 209, hematological, 201–2 247 217 metabolic, 202 paired kidney donation, 244 hyperlipidemia, 214t, 214 nutritional impairment, 202 postoperative protocols, 245t, 246–7 hypertension, 214t, 214 pulmonary, 201 quantization, 238–9 metabolic syndrome, 214t, 213–14 renal impairment, 202 risk assessment, crossmatch/ obesity, 214t, 214 pulmonary hypertension, 201, antibody screening based, 240, overview, 208t, 214t 206–7 241t, 242t primary biliary cirrhosis, 217 renal function changes, 207 virtual cross-matching, 243t, 240–3 primary non-function (PNF), 208t, respiratory function changes, 207 Kidney transplantation surgical 209 vascular access, 203 procedure primary sclerosing cholangitis, 217 veno-venous bypass, 207 cadaveric donor nephrectomy, sepsis, 208t, 210 Liver transplantation recipient 253–4 structural, 217 selection cold ischemia, 254 Liver transplantation long-term acute liver failure, 178t, 179 complications, 256–7 management clinical endpoints, 174 DCD donors, 256 autoimmune hepatitis, 217 combined liver/kidney, 179–81 en bloc, 256 conception, pregnancy, 219 contraindications, 181 Leadbetter–Politano, 255 disease recurrence, 216–17 delisting criteria, 181 live donor (See Living donor kidney follow-up, 218f, 217–18 disease severity scores, 175t, 175–6 donation) immunosuppression, 213 disease-severity selection criteria, overview, 253 malignancy, 214t, 215 176t, 176 pediatric (See Pediatric kidney metabolic syndrome, 213–14 ethical standpoints, 175t, 174–5 transplantation) mortality, 212f, 212, 213t, 213 hepatocellular carcinoma, 179 perioperative care, 255 osteoporosis, 215 overview, 173–4 standard cadaveric, 254–5 primary biliary cirrhosis, 217 re-transplantation, 179–81, 218 transplant nephrectomy, 257 primary sclerosing cholangitis, 217 variant syndromes (exceptional Kohler and Milstein, 20 quality of life, 218–19 diagnoses, MELD exceptions), rejection, 215–16 177t, 176–7, 179 Late cardiac allograft dysfunction renal impairment, 214–15 Liver transplantation surgical arrhythmia, 110–11 re-transplantation, 179–81, 218 procedure cardiac allograft failure, 111 structural complications, 217 donor tricuspid regurgitation, 111 Liver transplantation perioperative post-brainstem death, 191f, Legal issues. See under Organ donation management 190–1, 192 Leishmania, 51–2 acute liver failure, multiorgan post-cardiac death, 192 Liberase HI, 297 failure, 202–3 overview, 190 Lip, skin malignancy, 32t, 32 anesthetic technique, 203 preparation of liver, 192–3 Liver transplantation blood replacement, 204 recipient allocation, UK/Europe, 344 cardiovascular changes, 206–7 hepatectomy, caval anastomosis, chronic rejection pathology, 43f, coagulopathy, 204–5 194f, 193–4 42–3, 215–16 complications (See Liver hepatic artery anastomosis, 196, conception, pregnancy, 219 transplantation complications) 197f

362 Index

portal venous anastomosis, 195f, donor organ demand, 7–8 complications 194–5, 196 donor organ procurement, 59 cardiovascular, hemodynamic, reperfusion, bile duct historically, 7, 20, 122, 123t 151 anastomosis, 196–8 living-donor lobar (See early airway, 150–1 Living donor kidney donation Living-donor lobar lung gastrointestinal, 151 deceased vs., 248, 249f transplantation) infections, 152–4 donor evaluation, 249–50, 255 malignancy recurrence in, 31 of operation, 149 donor quality of life, 251 overview, 127 overview, 148–9 donor surgical risks, 250 tobacco, alcohol-related malignancy, pleural, 150 future directions, 252 36 reimplantation response, PGD, graft survival post-dialysis, 248, 249f Lung transplantation long-term 149–50 long-term donor outcome, 250–1 management rejection, 152 pregnancy following, 251–2 acute rejection, surveillance renal, 152 surgical procedure, 255–6 biopsies, 159t, 158–9 fluid management, hemodynamics, Living donor liver transplantation infection, 157–8 (See also Infection) 146–7 ABO-incompatible, 188 maintenance immunosuppression, immunosuppression, 148 complications 156–7 nutrition, 148 biliary, 187 malignancy, 162 overview, 145 donor, 184–5, 185t metabolic syndrome (diabetes, pulmonary edema, 146–7 rejection, 188 dyslipidemia), 162 pulmonary hypertension, 146 small-for-size syndrome (SFSS), nephrotoxicity, 161–2 ventilation, 145 186–7 osteoporosis, 162 Lung transplantation recipient vascular, 187–8 outcomes, 162f, 163f, 162–3 selection donor evaluation, 183, 184f overview, 155 absolute contraindications, 123t, donor operation, 184, 185f patient, 156 122–3, 124 dual graft, 188 post-transplant lymphoproliferative BODE index calculation, 126t graft, patient survivals, 186 disorder (See Post-transplant chronic obstructive pulmonary graft types, 184, 185f lymphoproliferative disorder disease (COPD) (See Chronic hepatitis C, 183 (PTLD)) obstructive pulmonary disease hepatocellular carcinoma, 183 surveillance, 155–6 (COPD)) indications, 182–3 venous thromboembolism (VTE), cystic fibrosisSee ( Cystic fibrosis) left vs. right lobe, 186 162 historically, 122 overview, 182 Lung transplantation management IPAH (idiopathic pulmonary paired donor exchange, 189 during surgery arterial hypertension), 126t, recipient operation, 185–6 cardiopulmonary bypass, 141–2 126, 140t recipient outcomes, 186 monitoring, vascular access IPF (idiopathic pulmonary fibrosis), rejection, 188 anesthesia maintenance, 141t, 126t, 126–7, 140t special recipient indications, 183 141 Lung transplantation surgical Living-donor lobar lung induction, 140t, 139–40 procedure transplantation obstructive disease, 140t, 140 bilateral, 134–5, 141 cadaveric vs., 131–2 overview, 139 combined heart/lung, 135–6 donor selection, size matching, 130t, pulmonary hypertension, 141 dissection, removal of native lung, 129–30 restrictive disease, 140 142 outcomes, 130, 131f, 131 suppurative disease, 140–1 donation after cardiac death (DCD), overview, 128, 129f TEE, 139 136 pediatric, 167 patient assessment, 138 donor lung anastomosis, 142 postoperative management, 131 premedication, 138 graft reperfusion, 142–3 recipient selection, 130, 131t primary graft dysfunction (PGD), overview, 133 surgical technique, 129f, 128–9 143–4 single lung, 133–4, 141 Lung transplantation RV failure, 142 Lymphoma allocation, UK/Europe, 344 single vs. bilateral, 141 diagnosis, 34 BOS (See Bronchiolitis obliterans stages of surgery (See Lung Epstein-Barr virus in, 33–4 syndrome (BOS)) transplantation surgical malignancy recurrence in, 31 chronic rejection pathology, 44f, procedure) pediatric heart transplant, 118 43–4 thoracic epidural analgesia, prevalence data, 32t combined heart/lung (See 139 treatment, 34–5 Combined heart/lung Lung transplantation postoperative transplantation) care Major histocompatibility complex comorbidities, 124–5, 126t antimicrobial prophylaxis, 147–8 (MHC), 11f, 10–11

363 Index

Malignancy Midazolam, 90 Obesity as absolute contraindication, 123 Milan criteria, 179 heart transplantation recipient in children, 36 Milrinone, 91 selection, 67 de novo types, 32t, 32 Minor histocompatibility (miH) kidney transplantation recipient donor transmission of, 31, 36 antigens, 10–11, 12 selection, 236 extracranial, 71 Mitral, tricuspid valve regurgitation, as lung transplantation comorbidity, hepatocellular carcinoma, 32t, 95 125 35–6 Monoclonal antibodies. See also post-liver transplantation, 214t, 214 immunosuppression with, 275–6 specific drugs Obliterative bronchiolitis, 44f, 43–4 Kaposi’s sarcoma, 32t, 32–3 historical applications of, 20 Obstructive airways disease, 201 kidney transplantation recipient in maintenance OKT3, 16, 20–1 selection, 235 immunosuppression, 25 Opioids, 90 liver transplantation recipient mechanism of action, 16 Organ donation (UK, Europe) selection, 177t overview, 22–3, 28f, 29, 97 allocation, 343–4 management, 37 PTLD treatment, 34–5 Bone Marrow Donors Worldwide, miscellaneous, 36 MRSA (methicillin-resistant 321 overview, 31–2, 37 Staphylococcus aureus), 152–3 critical care, emergency medicine post-transplant lymphoproliferative mTOR (mammalian target of roles, 339–40 disorder (See Post-transplant rapamycin) inhibitors, 16, 98 dead donor rule (Europe), 336 lymphoproliferative disorder Mucormycosis, 228 donor, transplant rates, 341t, 341–2 (PTLD)) Mycobacteria spp., 123 donor coordination, 338–9 prevalence data, 31–2 Mycophenolate mofetil (MMF) donor safety, quality, 342t, 342–3 recurrence, 31 in combination therapy, 25 legal issues renalcellcarcinoma,36 desensitization protocol, 246 consent, 336–7 skin, lip, 32t, 32 historical applications of, 21 definition of death, 335–6, 340t in stem cell transplantation, islet transplantation trials, 300 living donor, 337–8 326t mechanism of action, 16 overview, 335 tobacco, alcohol-related, 36 overview, 23–4, 97–8 potential for, 340t, 340–1 treatment, 37 reduction, withdrawal, 270b, 270, regulation, oversight, 343 vulva, perineum, 32t, 36 271b, 271 tracking, 345 Mammalian target of rapamycin side effects, ADRs, 276–7 transplant registries, 345–6 (mTOR) inhibitors, 16 Mycophenolate sodium (EC-MPS), Organ donation (US) MAP kinase (MAPK), 12 23–4 allocation, 352f, 351–2 Medical record review (MRR), Bone Marrow Donors Worldwide, 340–1 National Kidney Allocation Scheme, 321 Melanoma prevalence data, 32. See also 344 donor demand, 7–8, 53 Malignancy National Liver, Heart and Lung history, 347–8 MELD score, 175t, 176 Allocation Schemes, 344 listing requirements, 351 6-mercaptopurine National Pancreas Allocation Scheme, living donor, 353 historical applications of, 3 344 management, procurement mechanism of action, 16 Natural killer (NK) cells abdominal organs, 60 Metabolic syndrome in acute rejection, 12, 14–15 brain, brainstem death physiology drug-induced, 162 in transplant arteriopathy, 41 (See Brain, brainstem death hyperglycemia (diabetes mellitus), Neonatal hemochromatosis, 223–4 physiology) 109 New onset diabetes after donation after brain death (See hyperlipidemia, 109 transplantation (NODAT), 274t, Donation after brain death overview, 108, 213–14 274 (DBD)) systemic hypertension, 108–9 NF-␬B, 12 euvolemic donor vasoactive drug Methicillin-resistant Staphylococcus Nitric oxide, 143 support, 58–9 aureus (MRSA), 152–3 NK cells. See Natural killer (NK) cells hormone replacement therapy, 59 Methotrexate, 3 Non-cirrhotic metabolic liver non-heart beating donors, 60–2 Methylprednisolone disorders, 221, 222t organization, 59–60 acute rejection regimen, 26 Non-Hodgkins lymphoma, 33–4 overview, 53 historical applications of, 3, 6 Nutrition Organ Procurement and intestinal transplants, 308t cystic fibrosis, 148 Transplantation Network organ donor hemodynamic impairment, 202 (OPTN), 350f, 348–50, 350f, instability, 59 intestinal transplantation, 308–9 350 MHC (major histocompatibility lung transplantation postoperative recipient identification, 352 complex), 11f, 10–11 care, 148 successes, challenges, 353–4

364 Index

transplant program requirements, Pediatric heart transplantation, Pruritis, persistent, 177t evaluation, 353 112–120 Pulmonary hypertension United Network for Organ Sharing Pediatric kidney transplantation, 256, idiopathic, 166t (UNOS), 348f, 348–9, 350f, 351 278–285 IPAH, 126t, 126, 140t Organ Procurement and Pediatric liver transplantation liver transplantation perioperative Transplantation Network complications, 220–228 management, 201, 206–7 (OPTN), 350f, 348–50, 350f, Pediatric lung transplantation lung transplantation, 141, 146 350 allocation, 166–7 pediatric liver transplantation, 221 Orthotopic transplantation history, BOS (See Bronchiolitis obliterans portopulmonary, 177t 5–6 syndrome (BOS)) Osteoporosis complications, 168 Radiation therapy, 3, 35 as lung transplantation comorbidity, future considerations, 172 RATG. See Anti-thymocyte globulin 125 growth, 171 (ATG, rATG) overview, 326t indications/contraindications, 165f, Recombinant factor VIIa (eptacog post-lung transplantation, 162 164–5, 166t, 166 alpha activated), 204 treatment, 215 management, 168–70 Recurrent glomerulonephritis, 268 Ovary transplantation, 355 outcomes, 169f, 170f, 170–1 Renal cell carcinoma, 36 OX40, 11–12 overview, 164, 165f Renal transplantation. See Kidney quality of life, 171–2 transplantation Paired donor exchange, 189 re-transplantation, 170 Renovascular disease, 268 Pancreas transplantation surgical procedure, 167–8 Respiratory viruses, 48–9 allocation, UK/Europe, 344 Perineum, vulvar malignancy, 32t, 36 Restrictive lung disease, 326t allograft rejection, 292–3 Phosphodiesterase III inhibitors, 91 Rituximab cadaveric donor selection, 289t, Plasmapheresis (TPEX), 26 antibody-mediated rejection 288–9 Pleural effusions, 150 therapy, 26 candidate evaluation, 288 Pneumocystis jiroveci (carinii), 50–1, desensitization protocol, 245–6 complications 147, 260 mechanism of action, 16 bladder-drained patient, 291–2 Pneumothorax, 125, 150 PTLD treatment, 34–5 enteric-drained patient, 292 Polycystic liver disease (PLD), 177t hemorrhage, 291 Polyomavirus associated nephropathy, Sepsis, 70 surgical, 290 268–9, 271 SFSS (small-for-size syndrome), 177t, thrombosis, 291 Portal hypertensive bleeding, 177t 186–7 transplant pancreatitis, 291 Portal vein thrombosis (PVT), 226 Sirolimus (rapamycin) diabetes type 1, 287 Portopulmonary hypertension, 177t in combination therapy, 24, 25–6 history, 3–4 Post-reperfusion syndrome, 206 historical applications of, 21 immunosuppression, 292 Post-transplant lymphoproliferative mechanism of action, 16 indications, contraindications, 287, disorder (PTLD) monitoring, 98, 157 289 in children, 36, 169–70 overview, 24, 97–8 malignancy recurrence in, 31 diagnosis, 34 reduction, withdrawal, 270b, 270, organ allocation, 288 non-nephrotoxic 271b, 271 outcomes, 293 immunosuppression, 271 SIRS (systemic inflammatory response rationale in diabetes patients, 286–7 overview, 32t, 33–4 syndrome), 297 surgical techniques, 289f, 290f, pediatric liver transplantation, 228 Skin, lip malignancy, 32t, 32 289–90, 290f, 290 post lung transplant, 161t, 161 Skin transplantation Pancreatic islet transplantation treatment, 35f, 34–5, 275 history, 1, 2–3 auto-transplant, 301–2 WHO classification, 161t malignancy recurrence in, 31 engraftment, function, 299 PRA (panel reactive antibody), 232, Small-for-size syndrome (SFSS), 177t, future directions, 302 238–9 186–7 IBMIR, 298, 302 Prednisolone, 3, 124 Smoking, alcohol abuse. See also immunosuppression, 299–300 Prednisone, 100 Substance abuse infusion, 298f, 297–8, 299 Pregnancy post-transplantation as absolute contraindication, 123–4 living donor allo-transplant, 302 intestinal, 312 heart donor selection, 74 overview, 295 kidney, 277 heart transplantation recipient post-transplant monitoring, 300–1 live donor kidney donation, 251–2 selection, 67 processing, 296–7 liver, 219 kidney transplantation recipient recipient selection, 296 Primary graft dysfunction (PGD), selection, 233 Panel reactive antibody (PRA), 232, 143–4 Smoking, alcohol-related malignancy, 238–9 Primary hyperoxaluria type 1, 284–5 36 Parvovirus B19, 48–9 Prostaglandins, 143–4 Sotrastaurin (AEB071), 27–9

365 Index

Squamous cell carcinoma (SCC), 32 Tcells Treg, 17 Staphylococcus aureus (MRSA), 152–3 in acute rejection (See Acute Tricuspid regurgitation, 111 Starzl, T. E., 3–4, 190 rejection) Tricyclic antidepressants, 71 Statins, 98 CD8+ (cytotoxic T lymphocytes Tuberculosis, 234 Strongyloides stercoralis, 51–2 (CTLs)), 11f, 13, 14 Tumor necrosis factor (TNF)/TNF Substance abuse. See also Smoking, CD4+ in acute rejection, 11f, 13 receptor (TNFR) family, 11–12 alcohol abuse depleting induction heart donor selection, 74 immunosuppression, 300 UK Code of Practice, 336 heart transplantation donor memory, 12–13, 239 UKELD score, 175t, 176 selection, 74 sensitization of, 239 United Network for Organ Sharing heart transplantation recipient in transplant arteriopathy, 40 (See (UNOS), 348f, 348–9, 350f, 351 selection, 67 also Transplant arteriopathy) Urinary tract infection (UTI), 276 Surfactant deficiencies, 166t Teplizumab, 300 Uterus transplantation, 355 SV40 virus management, 268–9, 271 Thiazide diuretics, 273t Systemic hypertension, 108–9 Thymoglobulin, 21 Vanishing bile duct syndrome, 43f, Systemic hypotension, 95 Thymus transplantation, 356 42–3 Systemic inflammatory response T3 levels post-brain, brainstem death, Varicella zoster virus (VZV), 48–9 syndrome (SIRS), 297 55–6 Vascular anastomoses techniques TNF (tumor necrosis factor)/TNF donor lung, 142 Tacrolimus (FK-506, Fujimycin) receptor (TNFR) family, 11–12 hepatectomy, caval, 194f, 193–4 in combination therapy, 24, 25, 27–9 Toxoplasma gondii, T. cruzi, 51–2 hepatic artery, 196, 197f desensitization protocol, 246 TPEX (plasmapheresis), 26 history, 2, 5, 7 historical applications of, 21 Transesophageal echocardiography, 92 portal venous anastomosis, 195f, mechanism of action, 16 Transplant arteriopathy 194–5, 196 monitoring, 98, 157 chronic humoral rejection in kidney, reperfusion, bile duct, 196–8 overview, 23t, 23, 97–8 41f, 42f, 41–2 Vascular surgery history, 2 pediatric liver transplantation, defined, 38–9 Veno-venous bypass, 207 225–6 pathogenesis, experimental models, Ventricular assist devices (VADs), reduction, withdrawal, 270b, 270, 39–41 76–82, 89 271b, 271 pathologic diagnosis, 39f, 39, 40f Virtual cross-matching, 243t, 240–3 Tasocitinib (CP-690, 550), 27 Transplant bone disease, 276 Vulva, perineum malignancy, 32t, 36 Tcellmediatedrejection,239, 247, Transplant glomerulopathy 263t chronic rejection pathology, 42f, Wilson’s disease, 224 T cell receptor (TCR), 11f, 10–11, 41–2 12 long-term management, 267, 271 Xenograft transplantation history, 2

366