Pancreas and Islet Transplantation

Citation preview

Pancreas and Islet Transplantation Edited by

Nadey Hakim St Mary’s Hospital London UK

Robert Stratta UT Medical Group Memphis USA

Derek Gray Churchill Hospital Oxford UK

1

Whilst every effort has been made to ensure that the contents of this book are as complete, accurate and up-to-date as possible at the date of writing, Oxford University Press is not able to give any guarantee or assurance that this is the case. Readers are urged to take appropriately qualified medical advice in all cases. The information in this book is intended to be useful to the general reader, but should not be used as a means of self-diagnosis or for the prescription of medication.

Pancreas and Islet Transplantation

1 Great Clarendon Street, Oxford OX2 6DP Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Bangkok Buenos Aires Cape Town Chennai Dar es Salaam Delhi Hong Kong Istanbul Karachi Kolkata Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi São Paulo Shanghai Taipei Tokyo Toronto and an associated company in Berlin Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York © Oxford University Press, 2002 The moral rights of the author have been asserted Database right Oxford University Press (maker) First published 2002 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose this same condition on any acquirer A catalogue record for this title is available from the British Library Library of Congress Cataloging in Publication Data Pancreas and islet transplantation / edited by Nadey Hakim, Robert Stratta, Derek Gray. 1. Pancreas–Transplantation. 2. Islands of Langerhans–Transplantation. I. Hakim, Nadey S., 1958- II. Stratta, R.J. (Robert J.) III. Gray, Derek. [DNLM: 1. Pancreas Transplantation. 2. Islets of Langerhans Transplantation. WI 830 P1876 2002] RD546.P364 2002 617.5′570592–dc21 2001052324 ISBN 0 19 263255 8 (hbk. : alk. paper) 10 9 8 7 6 5 4 3 2 1 Typeset by EXPO, Malaysia Printed in Great Britain on acid-free paper by Biddles Ltd, Guildford & King’s Lynn

Preface

Overwhelming clinical evidence has accumulated to support the contention that pancreas transplantation is no longer investigational and indeed, is now regarded as a treatment of choice for uraemic diabetic patients. The results of pancreas transplantation are at least equivalent, if not superior, to other solid organ transplants. A kidney/pancreas transplant may not only be life enhancing, but life saving, compared to alternative treatment options. It is therefore not surprising that pancreas transplantation will become an increasingly important treatment option in the management of diabetic patients with or without renal failure. This book provides an authoritative account on the current status of all organ pancreas transplantation and other transplantation. Pancreas transplantation has rapidly moved from an experimental procedure asociated with high rates of morbidity and mortality to a mainstream technique with excellent patient and grafit survival. Over 15,000 pancreas transplants have already been performed. However pancreas transplantation’s value must be balanced against the risk of the operative procedure and the innovative long-term immunosuppressive therapy. Therefore, the indications for this procedure and the selection of the patients are critical to ensure low mortality and any improvement in quality of life. Consistent islet allograft success has recently been achieved by using multiple donors and steroid free immunosuppressive regime. Perhaps in the future there will be a role for both islet and pancreas transplantation with islet transplantation reserved for diabetic individuals with a low insulin requirement, while the pancreas transplant may be preferable for those with a high insulin requirement or insulin resistance such as those with type II diabetes. N.H., R.S., and D.G.

This page intentionally left blank

Contents

List of contributors ix 1 History of pancreas transplantation 1

David E. R. Sutherland and Carl G. Groth 2 The nature of the problem: why do we need pancreatic transplantation? 15

Martin Press 3 Experimental models in whole organ pancreatic transplantation 27

Vassilios E. Papalois and Nadey S. Hakim 4 Pretransplant medical evaluation for pancreas transplant candidates 45

Jerry McCauley and Robert J. Corry 5 Indications for kidney and pancreas transplantation and patient selection 59

Konstantinos N. Haritopoulos and Nadey S. Hakim 6 Indications for solitary pancreas transplantation 67

Robert J. Stratta 7 Donor management and selection for pancreas transplantation 79

Eldo Ermenegildo Frezza and Robert J. Corry 8 Procurement and benchwork preparation of the pancreatic graft 89

Nadey S. Hakim and Vassilios E. Papalois 9 Pancreas preservation 95

Hans U. Spiegel and Daniel Palmes 10 Surgical techniques of pancreas transplantation 115

Venkatesh Krishnamurthi and Stephen T. Bartlett 11 Pancreas transplantation with portal-enteric drainage 125

Robert J. Stratta, M. Hosein Shokouh-Amiri, Hani P. Grewal, and A. Osama Gaber 12 Pancreas transplantation: early perioperative management 145

Alan J. Koffron and Dixon B. Kaufman 13 Surgical complications of pancreas transplantation 155

Enrico Benedetti, Pierpaolo Sileri, Angelika C. Gruessner, and Luca Cicalese 14 Medical and urological complications of pancreas and kidney/pancreas

transplantation 167 Phuong-Thu T. Pham, Phuong-Chi T. Pham, and Alan H. Wilkinson

viii

CONTENTS

15 Diagnosis and treatment of pancreatic rejection 191

Alan J. Koffron and Dixon B. Kaufman 16 Histology of the pancreas transplant 205

David K. Klassen 17 Pancreas transplantation: effects on secondary complications of diabetes

mellitus 219 R. Paul Robertson 18 Pancreas transplantation: long-term aspects and effect on quality of life 229

Robert J. Stratta 19 Immunosuppression and diabetogenicity 247

Rahul M. Jindal, V.K. Revanur, and Alan G. Jardine 20 The economics of pancreas transplantation 277

Robert J. Stratta 21 Gene therapy for diabetes 291

Muralidhar Karanam, Z. Song, and Rahul M. Jindal 22 An historical view of the development of cellular and islet transplantation 305

Derek W. R. Gray 23 Fetal and neonatal pancreatic tissue transplantation 325

Bernard E. Tuch 24 Experimental approaches to the prevention of islet rejection 339

Norma Sue Kenyon 25 Clinical effectiveness of islet transplantation 355

Alberto M. Davalli, Federico Bertuzzi, and Antonio SecchIi Index 369

List of contributors

Stephen T. Bartlett, MD University of Maryland, Department of Surgery, Organ Transplant Baltimore, USA Enrico Benedetti, MD University of Illinois at Chicago, Department of Surgery–Division of Transplantation, 840 South Wood Street, CSB M/C 958 room 402, Chicago, 60612 IL, USA Federico Bertuzzi Department of Medicine, Istituto Scientifico San Raffaele, Milan, Italy Luca Cicalese, MD Department of Surgery, University of Illinois at Chicago, Chicago, USA Robert J. Corry, MD Professor of Surgery, Director Pancreas Transplantation Thomas E. Starzl Transplantation Institute, University of Pittsburgh School of Medicine, USA Alberto M. Davalli Department of Medicine, Istituto Scientifico San Raffaele, Milan, Italy Eldo Ermenegildo Frezza Thomas E. Starzl Transplantation Institute, University of Pittsburgh School of Medicine, USA

A. Osama Gaber Department of Surgery (Division of Transplantation), University of Tennessee Memphis, Memphis, Tennessee, USA Derek W. R. Gray D. Phil, FRCP, FRCS The Nuffield Department of Surgery, University of Oxford, Oxford, UK Hani P. Grewal Department of Surgery (Division of Transplantation), University of Tennessee Memphis, Memphis, Tennessee, USA Carl G. Groth, MD, PhD Professor of Transplantation Surgery Karolinska Institute, Huddinge Hospital, Huddinge, S-141 86, Sweden Angelika C. Gruessner, MS, PhD Department of Surgery, University of Minnesota, USA Nadey S. Hakim, MD, PhD, FRCS, FICS, FACS Surgical Director of Transplant Unit, St Mary’s Hospital, Praed Street, Paddington, London W2 1NY, UK

x

LIST OF CONTRIBUTORS

Konstantinos N. Haritopoulos, MD PhD FICS Transplant Fellow, Transplant Unit, St Mary’s Hospital, London, UK Alan G. Jardine, BSc MD FRCP Senior Lecturer in Nephrology, Department of Surgery, University of Glasgow, UK Rahul M. Jindal, MD, FRCS, MSc Director of Transplantation University Department of Surgery, Western Infirmary, Church Street, Glasgow G11 6NT, UK Muralidhar Karanam Post-doctoral Scientist, University of Glasgow, Glasgow, UK Dixon B. Kaufman, MD, PhD Northwestern University Medical School, Department of Surgery, Division of Transplantation, Chicago, USA Norma Sue Kenyon, Ph.D Diabetes Research Institute, University of Miami School of Medicine, 1450 NW 10th Avenue, Miami, Florida 33136, USA David K. Klassen, MD Department of Medicine, Medical Director of Transplantation, University of Maryland Medical Systems, 22 S. Greene St.–N3W143, Baltimore, MD 21201, USA, Alan J. Koffron, MD Northwestern University Medical School, Department of Surgery, Division of Transplantation, Chicago, USA

Venkatesh Krishnamurthi, MD Staff Surgeon, Director Pancreas Transplant, Cleveland Clinic Foundations, Northwestern, Ohio, USA Jerry McCauley MD, MPH, FACP Associate Professor of Medicine and Surgery, Director of Transplantation Nephrology, Renal-Electrolyte Division and Thomas E. Starzl, Transplantation Institute, University of Pittsburgh School of Medicine, USA Daniel Palmes, MD Münster University Hospital, Department of General Surgery–Surgical Research, Waldeyer Strasse 1, D-48149 Münster, Germany Vassilios E. Papalois, MD, PhD, FICS Consultant Transplant Surgeon Transplant Unit, St Mary’s Hospital, Praed Street, Paddington, London W2 1NY, UK Phuong-Chi T. Pham, MD Olive View-UCLA Medical Center, Department of Medicine, Division of Nephrology, Sylmar, CA 91342, USA Phuong-Thu T. Pham, MD University of California at Los Angeles School of Medicine, Department of Medicine, Division of Nephrology, Kidney and Pancreas Transplantation, Los Angeles, CA 90095, USA Martin Press FRCP Consultant Endocrinologist, Royal Free Hospital, London, NW3 2QG, UK

LIST OF CONTRIBUTORS

V.K. Revanur, FRCS Registrar in Surgery, Department of Surgery, University of Glasgow, UK R. Paul Robertson, MD Scientific Director/CEO, Pacific Northwest Research Institute, 720 Broadway, Seattle, WA 98122, USA Antonio Secchi Department of Medicine, Istituto Scientifico San Raffaele, Milan, Italy M. Hosein Shokouh-Amiri Department of Surgery (Division of Transplantation), University of Tennessee Memphis, Memphis, Tennessee, USA Pierpaolo Sileri, MD Department of Surgery, University of Illinois at Chicago, Chicago, USA Z. Song Post-doctoral Scientist, University of Glasgow, Glasgow, UK Hans Ulrich Spiegel, MD Münster University Hospital, Department of General Surgery–Surgical Research, Waldeyer Strasse 1, D-48149 Münster, Germany

Robert J. Stratta, MD Department of Surgery (Division of Transplantation), University of Tennessee Memphis, 956 Court Avenue, Suite A202, Memphis, TN 38163-2119, USA David E. R. Sutherland, MD, Ph.D. Professor of Surgery, Head, Division of Transplantation Director, Diabetes Institute for Immunology and Transplantation, Department of Surgery, University of Minnesota, Minneapolis, Minnesota, USA Bernard E. Tuch Professor of Medicine and Director of the Pancreas Transplant Unit, Prince of Wales Hospital and the University of New South Wales, Sydney, Australia Alan H. Wilkinson, MD, FRCP Professor of Medicine, Department of Medicine, Division of Nephrology, Medical Director, Kidney and Pancreas Transplantation, University of California at Los Angeles School of Medicine, 200 UCLA Medical Plaza, Suite 365, Los Angeles, CA 90095, USA

xi

This page intentionally left blank

Chapter 1

History of pancreas transplantation David E.R. Sutherland and Carl G. Groth

Introduction The history of pancreas transplantation is intertwined with that of other organ transplants and with the evolution of treatment of diabetes mellitus. Once diabetes was recognized as a consequence of a defect in the internal secretion of the pancreas, the idea of pancreas transplantation took root. From the late 1920s to the mid-1960s, various investigators worked out some of the surgical details of pancreas allotransplantation in dogs [1–8], setting the stage for clinical application [9]. The first pancreas allograft in a human was performed by Drs William D. Kelly and Richard C. Lillihei on 16 December 1966 at the University of Minnesota, a duct ligated segmental graft done simultaneous with a kidney in an uraemic diabetic recipient that resulted in immediate insulin independence (Fig. 1.1). The graft was still functioning 2 months later when the recipient died of sepsis related to surgical complications [10]. Surgical techniques are critical in pancreas transplantation and dominated the seminars organized to forward the field in the decade or two that followed the first case [11,12]. However, from a start where only an occasional pancreas transplant resulted in long-term insulin independence of the recipient [13] to the current highly successful application [14], the development of new immunosuppressants and the evolution of modern antirejection protocols that enveloped the organ transplant field in general were equally important [15-17]. By the mid-1980s, the symposia organized to facilitate the

Fig. 1.1 Technique for segmental transplantation used by Kelly and Lillehei in the first human pancreas transplant [10].

2

HISTORY OF PANCREAS TRANPLANTATION

application of pancreas transplantation reflected the broad interests — surgical, immunological, metabolic — of a maturing discipline [18–20]. The steady improvement in pancreas transplant outcome was documented by the International Pancreas and Islet Transplant Registry (IPTR), founded at the Lyon meeting [11] in 1980 [13,21]. Annual reports have been forthcoming from the IPTR since that time [14]. Patient selection is an important element in the history of pancreas transplantation, with most centres largely limiting its application to diabetic renal allograft recipients who were obligated to immunosuppression in any case. This bias was enforced by the observation that pancreas transplants alone (PTA) in non-uraemic diabetic recipients were associated with a substantially higher rejection rate than simultaneous pancreas kidney (SPK) transplants in uraemic recipients [22], a gap that was narrowed only in the last decade [14]. Nevertheless, a few programmes persisted in the application of pancreas transplantation in nonuraemic diabetic patients [23], accepting the trade-off of immunosuppression to achieve insulin independence per se. The history of the pancreas transplantation can be described in the context of individual contributions to programme development; the evolution, refinements, and physiology or new surgical techniques; antirejection strategies and methods to diagnose rejection episodes; observations on the consequences of euglycaemia on metabolism and diabetic secondary complications; and expansion in application.

Surgical techniques Following the first pancreas transplant with Kelly [10], Lillihei and associates [9,24] went on to perform an additional 13 whole pancreas transplants between 1966 and 1973 (the first five with a cutaneous graft duodenostomy; the next seven with internal exocrine drainage via a Roux-en-Y duodenal jejunostomy (Fig. 1.2); the last with only the papilla of Vater retained for anastomosis to the

Fig. 1.2 Technique of whole pancreaticoduodenal transplant with enteric drainage via a Roux-en-Y limb of recipient bowel, as devised by Lillehei in the initial Minnesota series [9].

D.E.R. SUTHERLAND AND CARL G. GROTH

recipient bowel). Of the first 14 cases at the University of Minnesota, 11 were in consecutive uraemic patients (10 with a simultaneous kidney transplant) and a subsequent three in non-uraemic patients. The complication rate was high, but one patient was insulin independent for over a year and died with a functioning graft after returning to dialysis because of renal allograft thrombosis secondary to arterial stenosis. This was the longest functioning pancreas graft in the world until a series of simultaneous pancreas kidney (SPK) transplants with segmental grafts by Gliedman et al., at Montefiore Hospital in New York beginning in the early 1970s, produced a recipient whose new pancreas functioned (insulin independent) for 5 years [25]. Several surgeons [26–33] scattered around the world performed one or two pancreas transplants in the late 1960s and early 1970s (reviewed in [34,35]), but by the mid-1970s only two centres were systematically offering pancreas transplantation to diabetic patients [36,37]. With the thought that the early failures were related to complications of the graft duodenum, both groups did segmental grafts (body and tail). In Stockholm, Groth et al. [36,38,39] evolved enteric drainage for duct management, while in Lyon, Dubernard et al. [37,40] obliterated the exocrine pancreas by duct injection of a sclerosing synthetic polymer. The latter technique was very safe, although long term-function was still rare. Meanwhile, the University of Minnesota initiated a clinical islet transplant programme with the first case in 1974 [41]. Although clinical islet autografts were successful [42],none of the initial allograft islet recepients became insulin independent [43], and in 1978 pancreas transplants were resumed at the University of Minnesota with a new surgical team [44]. The first few transplants in the new Minnesota series were also segmental grafts with the duct left open into the peritoneal cavity [45]. The very first recipient of an open duct graft (pancreas after kidney, PAK) achieved long-term insulin independence (for 17 years until dying with a functioning graft after being thrown off a horse [46]). However, the open duct technique was technically successfully only half of the time [47], and the University of Minnesota programme went on to use all of the duct management techniques that were extant or subsequently evolved (duct injection, enteric drainage, and urinary drainage) [48]. By the early 1980s, several programmes emerged and forwarded the field. One of the most important developments was modification of urinary drainage by Sollinger et al. [49] at the University of Wisconsin with direct anastomosis of the pancreatic duct to the bladder, remarkedly reducing the acute complication rate that Gliedman et al. [29] had using the ureter. Starzl et al. [50] at the University of Pittsburg resurrected the whole pancreatic duodenal technique of Lillihei et al. [9] for enteric drainage, and shortly thereafter Nghiem and Corry at the University of Iowa reported retention of the duodenum segment for anastomosis to the bladder [51]. Bladder drainage (BD) via the graft duodenum was quickly adapted by most American centres and was the dominant surgical technique for management of the pancreas graft exocrine secretions well into the 1990s [14]. Bladder drainage became popular because even if it did not have a lower complication rate than enteric drainage, the acute morbidity or severity of the complication was less. In addition, for solitary pancreas transplants, a decrease in urine amylase was shown by Prieto et al. to be an early marker for rejection that preceded hyperglycaemia by several days [52]. For SPK transplants, this was not so important, since the elevation of serum creatinine is a marker for rejection in the kidney that usually preceded exocrine or endocrine dysfunction as a sign of concomitant rejection in the pancreas from the same donor, allowing for early treatment and reversal [53]. Chronic complications of BD of pancreaticoduodenal transplants (infections, haematuria, acidosis, and dehydration) lead to the need for enteric conversion in about 10 per cent of cases [14]. The first conversion was reported by Tom et al. [54] from the University of Cincinnati in 1987. Segmental pancreas transplantation did not completely disappear as whole pancreaticoduodenal transplants evolved into a safe technique. Indeed, the fact that a segmental graft could be done at all

3

4

HISTORY OF PANCREAS TRANPLANTATION

led to the first use of living donors for pancreas transplantation at the University of Minnesota in 1979 [55], and more than 100 have been done since this time (over 10 per cent of the transplants at this institution) [46]. Besides the choice of where to drain (or not drain) the pancreas graft exocrine secretions, there is also a choice as to the site of transplantation and of venous drainage (systemic or portal). Most pancreas transplants are placed in the pelvis with vascular anastomosis to the recipient iliac artery and vein. With the expansion of all organ transplants, techniques were developed to preserve the arterial vessel to both the pancreas and liver when removed and transplanted separately. In the 1980s, several groups reported separately [56–58] the technique of retaining the entire hepatic artery to the liver by using a Y graft of the donor common external internal iliac artery to join the splenic and superior mesenteric arteries feeding a whole pancreas graft. Drainage of segmental pancreas transplant venous effluent into the recipient's portal system was done in a few patients in the 1980s. Calne [59], at Cambridge, was the first, accessing the recipient's splenic vein transabdominally and performing a ductogastrostomy for exocrine drainage. The Barcelona group also used the recipient's splenic vein, but via a retroperitoneal approach with a ducto-uterostomy for exocrine drainage [60]. The portal system was accessed, via the superior mesenteric vein in Stockholm [61,62], and via the inferior mesenteric vein at the University of Minnesota [63], combined with enteric drainage by both groups. In the early 1990s, Rosenlof et al. [64] at the University of Virginia, reported on a series of enteric-drained whole pancreaticoduodenal transplants with anastomosis of the graft portal vein to the recipients superior mesenteric vein (Fig. 1.3), and by the mid-1990s a few groups adapted this technique as their routine [65–67].

Fig. 1.3 Technique of whole pancreaticoduodenal enteric-drained pancreas transplant and portal drainage as devised by Rosenlof et al. [64] at the University of Virginia in 1990.

D.E.R. SUTHERLAND AND CARL G. GROTH

Systemic insulin levels in pancreas transplant recipients are lower with portal venous drainage, and it is thus more physiological, but no studies have shown that portal drainage gives any further advantage over that achieved with systemic drainage in ameliorating secondary complications of diabetes. To summarize the surgical evolution of pancreas transplantation, it went from predominantly whole organ transplantation in the late 1960s and early 1970s to mostly segmental transplantation in the late 1970s and early 1980s and then back to whole organ transplantation. Enteric drainage, duct injection, and BD were successively dominant from the early 1970s to the mid-1980s. The latter still predominates for solitary pancreas transplant, but enteric drainage has now re-emerged as dominant for SPK transplants. [Indeed, the whole pancreaticoduodenal transplant technique with systemic venous and enteric exocrine drainage devised early on by Lillihei (Fig. 1.2) is nearly identical to that currently used by most surgeons, although many forgo a Roux-en-Y limb of recipient bowel.] This shift is explained by the fact that although the immediate technical complications rate is nearly the same for enteric drainage and bladder drainage, the chronic complications rate is higher for bladder drainage. Early diagnosis and treatment of pancreas graft rejection (before hyperglycaemia), however, remains a problem for solitary pancreas transplants (PAK and PTA), while for SPK transplants from the same donor an elevation in serum creatinine is a surrogate marker for rejection likely involving both organs. For solitary pancreas transplants with BD, rejection will always eventually lead to a decline in urinary amylase (UA). Although not entirely specific, UA is 100 per cent sensitive, while an elevation in serum pancreatic enzymes (also not specific) does not always occur (lower sensitivity) [52]. For these reasons, urinary drainage for pancreas transplantation has persisted more than 30 years after its introduction by Gliedman et al. [25] using the ureter and nearly 20 years after direct drainage to the bladder by Sollinger et al. [49].

Antirejection treatment The immunosuppressive regimens used in pancreas allograft recipients are the same as those for other solid organs. From the 1960s to the early 1980s the only universally available maintenance immunosuppressants were azathioprine and steroids. In 1980, 1-year pancreas graft survival rate overall was 20 per cent [13]. Following the introduction of cyclosporin for general use in the mid1980s, 1-year graft survival rates reached approximately 75 per cent for SPK and approximately 50 per cent for PAK and PTA cases [68]. Tacrolimus and mycophenolate mofetil (MMF) came into general use in the mid-1990s [16,17], and current 1-year pancreas graft survival rates are over 80 per cent in all categories [14]. It should be noted that the transplant surgeons instrumental in introducing ciclosporin A, tacrolimus, and MMF, Calne et al.[15], Starzl et al.[16], and Rayhill et al. [17], respectively, were also pioneers and innovators in the surgical techniques for pancreas transplantation [49,50,59]. Also of interest is the fact that the diabetogenic effect of cyclosporin was first reported from its use in pancreas transplant recipients [69]. The other calcineurin inhibitor, tacrolimus, also had this property [16]. Nevertheless, the incidence of drug-induced hyperglycaemia has not been any higher in pancreas transplant recipients than other organ allograft recipients, and long-term insulin independence rates, are highest in pancreas recipients given one or the other of these drugs [14]. The diagnosis and treatment of rejection episodes are based on the same principle as for other organ allografts. Graft dysfunction leads to a tentative diagnosis, a confirmative biopsy is done if possible, and if positive, immunosuppression is temporarily increased. Pancreatic graft biopsies were first done by laparotomy [70], then transcystoscopically for bladder-drained grafts [71], and currently percutaneously, regardless of the surgical technique originally used [72,73].

5

6

HISTORY OF PANCREAS TRANPLANTATION

Impact of pancreas transplantation on the understanding of pathogenesis of diabetes mellitus and its associated complications Following the introduction of exogenous insulin therapy for the treatment of diabetes in the 1920s, diabetic patients survived to develop complications affecting the eyes, nerves, kidneys, and other tissues. For decades, debates ranged over whether or not the complications were secondary to disordered metabolism. In the early 1980s, the Diabetes Control and Complications Trial (DCCT) was organized to settle the debate. The results, first published in 1993 [74], showed that the lower the glycosylated haemoglobin levels (a reflection of the average blood sugar levels) were in the study patients, the lower the probability that retinopathy, neuropathy, or nephropathy would develop. However, in the 1980s, well before the DCCT study was published, observations in pancreas transplant recipients had already answered the question [75–79]. Since the 1970s it has been known that normal kidneys transplanted into diabetic recipients develop diabetic nephropathy [80]. In 1985, Bohman et al. [75] in Stockholm, showed that in recipients of simultaneous kidney and pancreas transplants in which neither organ was rejected, lesions of diabetic nephropathy did not occur in a transplanted kidney. Shortly thereafter, Bilous et al. [79] at the University of Minnesota, showed that a successful pancreas transplant (PAK) could prevent, reverse, or stabilize lesions in a previously transplanted kidney. The data was confirmed in long-term followup studies in the 1990s [81]. Indeed, Fioretto et al. [82], at the University of Minnesota, showed that lesions of diabetic nephropathy could be reversed in the native kidneys of non-uraemic recipients of pancreas transplants alone. In regard to neuropathy, Solders et al. [76,83] at Stockholm, in studies of recipients of combined pancreas and kidney transplants (SPK) and Kennedy and associates [77,84] at the University of Minnesota, in studies of recipients of successful solitary pancreas transplants, showed neuropathy improved as opposed to its inexorable progression in those with no transplant or with a failed graft. The Minnesota group also showed that neuropathic patients had improved survival probability after successful pancreas transplantation [85]. Finally, Ramsay et al. [78], also at the University of Minnesota, showed that in patients with advanced retinopathy, retinal events occurred in up to 30 per cent during the first 3 years after successful pancreas transplant. Thereafter, they remained stable, while those with failed grafts continued to have retinal events. No patients without disease at baseline developed retinopathy as long as the pancreatic graft functioned. Thus, the scientific principle that the diabetic lesions of the eye, nerve, and kidneys are linked to disordered metabolism was fully established from the observations in pancreas transplant recipients well before the DCCT was completed. Another area of controversy in diabetes mellitus that was enlightened by pancreas transplantation involved the pathogenesis of the disease. By the early 1980s, consistent evidence implicated autoimmunity as responsible for the destruction of ␤-cells in patients with type I diabetes [86]. That this must be the case was finally shown by observations in diabetic recipients of segmental pancreas transplants from their non-diabetic identical twin counterparts carried out at the University of Minnesota in the 1980s [87]. The first three recipients in the identical twin series were done without prophylactic immunosuppression (since isografts cannot be rejected) and all three twin recipients developed biopsy confirmed autoimmune isletitis (no rejection) and recurrence of diabetes mellitus within a few months of the transplant [88]. In contrast, immunosuppressed recipients of identical twin segmental grafts did not develop recurrence of disease [89], indeed, two identical twin recipients have been insulin independent for 9 and 13 years, respectively [46].

D.E.R. SUTHERLAND AND CARL G. GROTH

Although unusual, islet-directed autoimmunity can break through immunosuppression capable of preventing allograft rejection. In 1996, Tyden et al. [90] in Stockholm documented selective ␤-cell destruction in two cadaver pancreas transplants. However, immunosuppression given to prevent allograft rejection is nearly always sufficient to suppress autoimmunity [89], given the rarity of recurrent disease [91]. In summary, observations in pancreas transplant recipients definitely proved both that type I diabetes is an autoimmune disease, and that the diabetic complications which may occur in other organs are secondary to disorder in metabolism.

Evolution of recipient selection and programme development Besides the impact on secondary complications, studies in the 1980s documented an improvement in day-to-day quality of life of a diabetic patient by the insulin independence induced by a functioning pancreas graft [92,93]. Follow-up studies confirmed this benefit [94,95], providing an impetus to consider pancreas (or islet) transplantation at any stage of diabetes. Transplantation of functioning ␤-cells is the most physiological treatment of diabetes mellitus. Ideally it should be simple, with just an injection of isolated islets. The optimism that islet transplantation could be rapidly developed to achieve this simplicity probably delayed the development of clinical pancreas transplantation. However, develop it did and in turn the surprising success of pancreas transplantation (after a rocky start) in part has probably inhibited the development of clinical islet transplantation. Immunological tolerance for clinical application has not yet been developed, and with either pancreas or islet transplantation there is a need for immunosuppression. For this reason, most programmes limited pancreas or islet transplants to uraemic diabetic recipients or renal allografts who are thus obligated to immunosuppression. As the twentieth century ended, approximately 1500 pancreas transplants were being performed annually worldwide (the majority in the United States), of which more than 80 per cent were SPK [14]. During the 1980s, solitary pancreas transplants were routine only at the University of Minnesota [96]. In the 1990s, the number of programmes promoting solitary pancreas transplants has expanded [97,98] and the results in the modern immunosuppressive era have been good [14]. Pancreas (or islet) transplantation in non-uraemic patients should predominate and the foundation for its application has been laid. In regard to islet transplantation, since 1989, several groups have succeeded in establishing insulin independence in occasional diabetic recipients by intraportal islet transplantation [99–102]. Consistent islet allograft success has recently been achieved by using multiple donors and a steroidfree immunosuppression regimen at the University of Alberta in Edmonton [103]. Thus, engrafting an adequate ␤-cell mass appears to be the critical factor for clinical islet transplant to succeed. The challenge now is to achieve consistent success with a single donor. That this should be possible is apparent from the fact that auto islet transplantation after total pancreatectomy can sustain insulin independence [104], as shown by the very first case in the late 1970s [42]. Perhaps there will be a role for both islet and pancreas transplantation, with islet transplantation reserved for diabetic individuals with a low insulin requirement, while pancreas transplant may be preferable for those with a high insulin requirement or insulin resistance such as those with type II diabetes. Pancreas transplantation will also be preferred in individuals diabetic as a result of total pancreatectomy where enteric drainage could be used to correct exocrine deficiency, as was first done in the 1980s [105]. Indeed, if an unlimited supply of ␤-cells for transplantation could be obtained (xenografts or human cell lines), the future of pancreas transplantation could primarily be to correct exocrine deficiency, for which the ground work has already been laid [105,106].

7

8

HISTORY OF PANCREAS TRANPLANTATION

References 1 Gayet R, Guillamie M. La regulation de la secretion interne pancreatique par un processus humoral, demonstree par des transplantations de pancreas. Experiences sue des animaux normaux. Coll Roy Soc Biol 1927;97:1613–14. 2 Brooks JR, Gifford GH. Pancreatic homotransplantation. Transplant Bull 1959;6:100–3. 3 DeJode LR, Howard JM. Studie in pancreaticoduodenal homo-transplantation. Surg Gynecol Obstet 1962;114:553–8. 4 Bergan J, Hoehn JG, Porter N. Total pancreatic allografts in pancreatectomized dogs. Arch Surg 1965;90:521–6. 5 Largiader F, Lyons GW, Hidalgo F, Dietzman RH, Lillehei RC. Orthotopic allotransplantation of the pancreas. Am J Surg 1967;113:70–8. 6 Merkel FK, Kelly WD, Goetz FC, Maney J. Irradiated heterotopic segmental canine pancreatic allografts. Surgery 1968;63:291–7. 7 Idezuki Y, Feemster JA, Dietzman RH, Lillehei RC. Experimental pancreaticoduodenal preservation and transplantation. Surg Gynecol Obstet 1968;126:1002–14. 8 Sutherland DER. Pancreas and islet transplantation. I. Experimental studies. Diabetologia 1981;20:161–85. 9 Lillehei RC, Simmons RL, Najarian JS, Weil R, Uchida H, Ruiz JO, et al. Pancreatico-duodenal allotransplantation: Experimental and clinical experience. Ann Surg 1970;172:405–36. 10 Kelly WD, Lillehei RC, Merkel FK. Allotransplantation of the pancreas and duodenum along with the kidney in diabetic nephropathy. Surgery 1967;61:827–35. 11 Dubernard JM, Traeger J. Pancreas and islet transplantation workshop. Transplant Proc 1980;12:1–2. 12 Land W, Landgraf R. Segmental pancreatic transplantation international workshop. Horm Metab Res 1983;13:1–104. 13 Sutherland DER. International human pancreas and islet transplant registry. Transplant Proc 1980;12 (No. 4 Suppl. 2):229–36. 14 Gruessner A, Sutherland DER. Pancreas transplant outcomes for United States cases reported to the United Network for Organ Sharing (UNOS) and non-US cases reported to the International Pancreas Transplant Registry. In: Cecka JM, Terasaki PI, eds. Clinical transplants — 1999. Los Angeles: UCLA Immunogenetics Center; 2000:51–69. 15 Calne RY, Rolles K, White DJ, Thiru S, Evans DB, McMaster P, et al. Cyclosporin A initially as the only immunosuppressant in 34 recipients of cadaveric organs: 32 kidneys, 2 pancreases, and 2 livers. Lancet 1979;2:1033–6. 16 Starzl TE, Todo S, Fung J, Demetris AJ, Venkataramman R, Jain A. FK 506 for liver, kidney, and pancreas transplantation. Lancet 1989;2:1000–4. 17 Rayhill SC, Kirk AD, Odorico JS, Heisey DM, Cangro CB, Pirsch JD, et al. Simultaneous pancreas–kidney transplantation at the University of Wisconsin. In: Cecka JM, Terasaki PJ, eds Clinical transplants 1994. Los Angeles: UCLA Tissue Typing Laboratory 1995:261–9. 18 van Schilfgaarde R, Persijn GG, Sutherland DER. International symposium on organ transplantation in diabetics. Transplant Proc 1984;16:565–6. 19 Land W, Landgraf R. The world experience in clinical pancreas transplantation. Transplant Proc 1987;19 (Suppl. 4):1–2. 20 Groth CG, Tyden G, Bolinder J. First international congress on pancreatic and islet transplantation. Diabetes (Suppl. 1)1989;38:1–335. 21 Sutherland DER, Moudry KC. Pancreas Transplant Registry: history and analysis of cases 1966 to October 1986. Pancreas 1987;2:473–88.

D.E.R. SUTHERLAND AND CARL G. GROTH

22 Groth CG, Lundgren G, Ostman J, Gunnarsson R. Experience with nine segmental pancreatic transplantations in preuremic diabetic patients in Stockholm. Transplant Proc 1980;12:68–71. 23 Sutherland DER. Pancreas transplantation in non-uremic diabetic patients. Transplant Proc 1986;18:1747–9. 24 Lillehei RC, Ruiz JO, Aquino C, Goetz FC. Transplantation of the pancreas. Acta Endocrinol 1976;83 (Suppl. 205):303–20. 25 Gliedman ML, Tellis VA, Soberman R. Long-term effects of pancreatic transplant function in patients with advanced juvenile onset diabetes. Diabetes Care 1978;1:1–9. 26 Bortagaray MC, Zelasco JF, Bava A. Homotransplante parcial del pancreas en el diabetico. Prensa Med Argent 1970;57:220–4. 27 Teixeira E, Monteiro G, De Cenzo M, Teixeira A, Bergan JJ. Transplantation of the isolated pancreas: report on the first human case. Bull Soc Int Chir 1970;29:337–44. 28 Merkel FK, Poticha SM, Nudelman EJ, Colwell JA, Bergan JJ. Pancreas transplantation in man. I. The donor and recipient operation. Arch Surg 1971;103:205–10. 29 Gliedman ML, Gold M, Whittaker J, Rifkin H, Soberman R, Freed S, et al. Clinical segmental pancreatic transplantation with ureter-pancreatic duct anastomosis for exocrine drainage. Surgery 1973;74:171–80. 30 Connolly JE, Martin DC, Steinberg T, Gwinup G, Gazzaniga AB, Bartlett RH. Clinical experience with pancreaticoduodenal transplantation. Arch Surg 1973;106:489–94. 31 Merkel FK, Ryan WG, Armbruster K, Seim S, Ing TS. Pancreatic transplantation for diabetes mellitus. Ill Med J 1973;144:477–9. 32 Largiader F, Uhlschmid G, Binswanger U, Zaruba K. Pancreas rejection in combined pancreaticoduodenal and renal allotransplantation in man. Transplantation 1975;19:185–7. 33 Brynger H, Blohme G, Claes G, Gustafsson A, Gelin LE. Transplantation of a duct-ligated pancreatic allograft to a diabetic patient. Scand J Urol Nephrol 1975 (Suppl. 29);59–62. 34 Sutherland DER. Pancreas and islet transplantation. II. Clinical trials. Diabetologia 1981;20:435–50. 35 Sutherland DER, Rynasiewicz JJ, Najarian JS. Current status of pancreas and islet transplantation. In: Brownlee M, ed. Handbook of diabetes mellitus. New York: Garland Publishing, 1981:273–416. 36 Groth CG, Lundgren G, Arner P, Collste H, Hardstedt C, Lewander R, et al. Rejection of isolated pancreatic allografts in patients with diabetes. Surg Gynecol Obstet 1976;143:933–40. 37 Dubernard JM, Traeger J, Neyra P, Touraine JL, Traudiant D, Blanc-Brunat N. A new method of preparation of segmental pancreatic grafts for transplantation: trials in dogs and in man. Surgery 1978;84:633–40. 38 Groth CG, Lundgren G, Gunnarsson R, Hardstedt C, Ostman J. Segmental pancreatic transplantation with special reference to the use of ileal exocrine diversion and to the hemodynamics of the graft. Transplant Proc 1980;12 (Suppl. 2):62–7. 39 Groth CG, Collste H, Lundgren G. Successful outcome of segmental human pancreatic transplantation with enteric exocrine diversion after modifications in technique. Lancet 1982;2:522–4. 40 Traeger J, Dubernard JM, Touraine JL, Neyra P, Malik MC, Pelissard C, et al. Pancreatic transplantation in man: a new method of pancreas preparation and results on diabetes correction. Transplant Proc 1979;11:331–5. 41 Najarian JS, Sutherland DER, Matas AJ, Steffes MW, Simmons RL, Goetz FC. Human islet transplantation: A preliminary report. Transplant Proc 1977;9:233–6. 42 Sutherland DER, Matas AJ, Najarian JS. Pancreatic islet cell transplantation. Surg Clin N Am 1978;58:365–82. 43 Sutherland DER, Matas AJ, Goetz FC, Najarian JS. Transplantation of dispersed pancreatic islet tissue in humans: autografts and allografts. Diabetes 1980;29 (Suppl.):31–44.

9

10

HISTORY OF PANCREAS TRANPLANTATION

44 Sutherland DER, Goetz FC, Najarian JS. Intraperitoneal transplantation of immediately vascularized segmental pancreatic grafts without duct ligation: A clinical trial. Transplantation 1979;28:485–91. 45 Sutherland DER, Baumgartner D, Najarian JS. Free intraperitoneal drainage of segmental pancreas grafts: Clinical and experimental observations on technical aspects. Transplant Proc 1980;12:26–32. 46 Sutherland DE, Gruessner RWG, Dunn DC, Matas AJ, Humar A, Kandaswamy R, et al. Over 1000 pancreas transplants at a single institution. Ann Surg 2000;233:463–501. 47 Sutherland DER, Goetz FC, Rynasiewicz JJ, Baumgartner D, White DC, Elick BA, et al. Segmental pancreas transplantation from living related and cadaver donors: A clinical experience. Surgery 1981;90:159–69. 48 Sutherland DE, Gores PF, Farney AC, Wahoff DC, Matas AJ, Dunn DL, et al. Evolution of kidney, pancreas, and islet transplantation for patients with diabetes at the University of Minnesota. Am J Surg 1993;166:456–91. 49 Sollinger HW, Cook K, Kamps D. Clinical and experimental experience with pancreaticocystostomy for exocrine pancreatic drainage in pancreas transplantation. Transplant Proc 1984;16:749–51. 50 Starzl TE, Iwatsuki S, Shaw BW Jr, Greene DA, Van Thiel DH, Nalesnik MA, et al. Pancreaticoduodenal transplantation in humans. Surg Gynecol Obstet 1984;159:265–72. 51 Nghiem DD, Corry RJ. Technique of simultaneous pancreaticoduodenal transplantation with urinary drainage of pancreatic secretion. Am J Surg 1987;153:405–6. 52 Prieto M, Sutherland DER, Fernandez-Cruz L, Heil JE, Najarian JS. Experimental and clinical experience with urine amylase monitoring for early diagnosis of rejection in pancreas transplantation. Transplantation 1987;43:71–9. 53 Gruessner RWG, Dunn DL, Tzardis PJ, Tomadze G, Moudry-Munns KC, Matas AJ, et al. Simultaneous pancreas and kidney transplants versus single kidney transplants and previous kidney transplants in uremic patients and single pancreas transplants in nonuremic diabetic patients: comparison of rejection, morbidity, and long-term outcome. Transplant Proc 1990;22:622–3. 54 Tom WW, Munda R, First MR, Alexander JW. Autodigestion of the glans penis and urethra by activated transplant pancreatic exocrine enzymes. Surgery 1987;102:99–101. 55 Sutherland DER, Goetz FC, Najarian JS. Living-related donor segmental pancreatectomy for transplantation. Transplant Proc 1980;12:19–25. 56 Sutherland DER, Moudry KC, Najarian JS. Pancreas transplantation. In: Cerilli J, ed. Organ transplantation and replacement. Philadelphia: Lippincott, 1987:535–74. 57 Marsh CL, Perkins JD, Sutherland DER, Corry RJ, Sterioff S. Combined hepatic and pancreaticoduodenal procurement for transplantation. Surg Gynecol Obstet 1989;168:254–8. 58 Delmonico FL, Jenkins RL, Auchincloss H Jr, Etienne TJ, Russell PS, Monaco AB, et al. Procurement of a whole pancreas and liver from the same cadaveric donor (see comments). Surgery 1989;105:718–23. 59 Calne RY. Paratopic segmental pancreas grafting: a technique with portal venous drainage. Lancet 1984;1:595–7. 60 Gil-Vernet JM, Fernandez-Cruz L, Caralps A, Andreu J, Figuerola D. Whole organ and pancreaticoureterostomy in clinical pancreas transplantation. Transplant Proc 1985;17:2019–22. 61 Tyden G, Lundgren G, Ostman J, Gunnarsson R, Groth CG. Grafted pancreas with portal venous drainage. Lancet 1984;1:964. 62 Tydén G, Wilczek H, Lundgren G, Ostman J, Gunnarsson R, Jaremko G, et al. Experience with 21 intraperitoneal segmental pancreatic transplantations with enteric or gastric exocrine diversion in humans. Transplant Proc 1985;17:331–5. 63 Sutherland DER, Goetz FC, Moudry KC, Abouna GM, Najarian JS. Use of recipient mesenteric vessels for revascularization of segmental pancreas grafts: Technical and metabolic considerations. Transplant Proc 1987;19:2300–4.

D.E.R. SUTHERLAND AND CARL G. GROTH

64 Rosenlof LK, Earnhardt RC, Pruett TL, Stevenson WC, Douglas MT, Cornett GC, et al. Pancreas transplantation: an initial experience with systemic and portal drainage of pancreatic allografts. Ann Surg 1992;215:586–95. 65 Reddy KS, Stratta RJ, Shokouh-Amiri MH, Alloway R, Egidi MF, Gaber AO. Surgical complications after pancreas transplantation with portal-enteric drainage. J Am Coll Surg 1999;189:305–13. 66 Bruce DS, Newell KA, Woodle ES, Cronin DC, Grewal HP, Millis JM, Ruebe M, et al. Synchronous pancreas–kidney transplantation with portal venous and enteric exocrine drainage: outcome in 70 consecutive cases. Transplant Proc 1998;30:270–1. 67 Philosophe B, Farney AC, Schweitzer E, Colonna J, Jarrell BE, Krishnamurthi V, et al. The superiority of portal venous drainage over systemic venous drainage in pancreas transplantation. Ann Surg 2001;234:689–696. 68 Sutherland DER, Moudry-Munns KC. International Pancreas Transplant Registry Report. In: Terasaki PI, ed. Clinical transplants 1988. UCLA Tissue Typing Laboratory, California; 1989:53–64. 69 Gunnarsson R, Klintmalm G, Lundgren G, Wilczek H, Ostman J, Groth CG. Deterioration in glucose metabolism in pancreatic transplant recipients given cyclosporin (letter). Lancet 1983;2:571–2. 70 Sutherland DER, Casanova D, Sibley RK. Role of pancreas graft biopsies in the diagnosis and treatment of rejection after pancreas transplantation. Transplant Proc 1987;19:2329–31. 71 Perkins JD, Munn SR, Marsh CL, Barr D, Engen DE, Carpenter HA. Safety and efficacy of cystoscopically directed biopsy in pancreas transplantation. Transplant Proc 1990;22:665–6. 72 Allen RD, Wilson TG, Grierson JM, Greenberg ML, Earl MJ, Nankivel BJ, et al. Percutaneous biopsy of bladder-drained pancreas transplants. Transplantation 1991;51:1213–16. 73 Gaber AO, Gaber L, Shokouh-Amiri MH, Hathaway D. Percutaneous biopsy of pancreas transplants. Transplantation 1992;54:548–50. 74 DCCT Research Group. Diabetes control and complications trial (DCCT): The effect of intensive diabetes treatment in long term complications in IDDM. NEJM 1993;329:977–86. 75 Bohman SO, Tyden G, Wilczek H, Lundgren G, Jaremko G, Gunnarsson R, et al. Prevention of kidney graft diabetic nephropathy by pancreas transplantation in man. Diabetes 1985;34:306–8. 76 Solders G, Wilczek H, Gunnarsson R, Tyden G, Persson A, Groth CG. Effects of combined pancreatic and renal transplantation on diabetic neuropathy: a two-year follow-up study. Lancet 1987;2:1232–5. 77 Van der Vliet JA, Navarro X, Kennedy WR, Goetz FC, Najarian JS, Sutherland DER. The effect of pancreas transplantation on diabetic polyneuropathy. Transplantation 1988;45:368–70. 78 Ramsay RC, Goetz FC, Sutherland DER, Mauer SM, Robinson LL, Cantrill HL, et al. Progression of diabetic retinopathy after pancreas transplantation for insulin-dependent diabetes mellitus. NEJM 1988;318:208–14. 79 Bilous RW, Mauer SM, Sutherland DER, Najarian JS, Goetz FC, Steffes MW. The effects of pancreas transplantation on the glomerular structure of renal allografts in patients with insulin-dependent diabetes. NEJM 1989;321:80–5. 80 Mauer SM, Barbosa J, Vernier RL, Kjellstrand CM, Buselmeier TJ, Simmons RL, et al. Development of diabetic vascular lesions in normal kidneys. Development of diabetic vascular lesions in normal kidneys transplanted into patients with diabetes mellitus. N Engl J Med 1976;295:916–20. 81 Wilczek H, Jaremko G, Tydén G, Groth CG. A pancreatic graft protects a simultaneously transplanted kidney from developing diabetic nephropathy: a 1–6 year follow-up study. Transplant Proc 1993;25:1314–15. 82 Fioretto P, Steffes MW, Sutherland DER, Goetz FC, Mauer SM. Reversal of lesions of diabetic nephropathy after pancreas transplantation. New Engl J Med 1998;339:69–75. 83 Solders G, Tyden G, Persson A, Groth CG. Improvement of nerve conduction in diabetic neuropathy: a follow-up study 4 years after combined pancreatic and renal transplantation. Diabetes 1992;41:946. 84 Kennedy WR, Navarro X, Goetz FC, Sutherland DER, Najarian JS. Effects of pancreatic transplantation on diabetic neuropathy. NEJM 1990;322:1031–7.

11

12

HISTORY OF PANCREAS TRANPLANTATION

85 Navarro X, Kennedy WR, Loewensen RB, Sutherland DER. Influence of pancreas transplantation on cardiorespiratory reflexes, nerve conduction, and mortality in diabetes mellitus. Diabetes 1990;39:802–6. 86 Eisenbarth GS. Type I diabetes mellitus: A chronic autoimmune disease. NEJM 1986;314:1360–8. 87 Sutherland DER, Sibley RK, Xu XZ, Michael AF, Srikanta S, Taub F, et al. Twin-to-twin pancreas transplantation: Reversal and reenactment of the pathogenesis of type I diabetes. Trans Assoc Am Physicians 1984;97:80–7. 88 Sibley RK, Sutherland DER, Goetz FC, Michael AF. Recurrent diabetes mellitus in the pancreas isoand allograft: A light and electron microscopic and immunohistochemical analysis of four cases. Lab Invest 1985;53:132–44. 89 Sutherland DER, Goetz FC, Sibley RK. Recurrence of disease in pancreas transplants. Diabetes 1989;38:85–7. 90 Tyden G, Reinholt FP, Sundkvist G, Bolinder J. Recurrence of autoimmune diabetes mellitus in recipients of cadaveric pancreatic grafts (see comments) (published erratum appears in N Engl J Med 1996;(23):1778). N Engl J Med 1996;335:860–3. 91 Papadimitriou JC, Drachenberg CB, Wiland A, Klassen DK, Fink J, Weir MR, et al. Histologic grading of acute allograft rejection in pancreas needle biopsy: correlation to serum enzymes, glycemia, and response to immunosuppressive treatment. Transplantation 1998;66:1741–5. 92 Nakache R, Tyden G, Groth CG. Quality of life in diabetic patients after combined pancreas-kidney or kidney transplantation. Diabetes 1989;38 (Suppl. 1):40–2. 93 Zehrer CL, Gross CR. Quality of life of pancreas transplant recipients. Diabetologia 1991;34 (Suppl. 1):S145–9. 94 Nakache R, Tyden G, Groth CG. Long-term quality of life in diabetic patients after combined pancreas–kidney transplantation or kidney transplantation. Transplant Proc 1994;26:510–11. 95 Gross CR, Limwattananon C, Matthees BJ. Quality of life after pancreas transplantation: a review. Clin Transplant 1998;12:351–61. 96 Sutherland DER, Chinn PL, Goetz FC, Elick BA, Najarian JS. Minnesota experience with 85 pancreas transplants between 1978 and 1983. World J Surg 1984;8:244–52. 97 Bartlett ST, Schweitzer EJ, Johnson LB, Kuo PC, Papadimitriou JC, Drachenberg CB, et al. Equivalent success of simultaneous pancreas kidney and solitary pancreas transplantation. A prospective trial of tacrolimus immunosuppression with percutaneous biopsy. Ann Surg 1996;224:440–9. 98 Stratta RJ, Weide LG, Sindhi R, Sudan D, Jerius JT, Larsen JL, et al. Solitary pancreas transplantation. Experience with 62 consecutive cases. Diabetes Care 1997;20:362–8. 99 Scharp DW, Lacy PE, Santiago JV, McCullough CS, Weide LG, Falqui L, et al. Insulin independence after islet transplantation into type I diabetic patient. Diabetes 1990;39:515–18. 100 Warnock GL, Kneteman NM, Ryan E. Long-term follow-up after transplantation of insulinproducing pancreatic islets into patients with type I (insulin independent) diabetes mellitus. Diabetologia 1992;35:89–95. 101 Gores PF, Najarian JS, Stephanian E, Lloveras JJ, Kelley SL, Sutherland DER. Insulin independence in type I diabetes after transplantation of unpurified islets from a single donor using 15deoxyspergualin. Lancet 1993;341:19–21. 102 Meyer C, Hering BJ, Grossmann R, Brandhorst H, Brandhorst D, Gerich J, et al. Improved glucose counterregulation and autonomic symptoms after intraportal islet transplants alone in patients with long-standing type I diabetes mellitus. Transplantation 1998;66:233–40. 103 Shapiro AMJ, Lakey JRT, Ryan E, Korbutt GS, Toth E, Warnock GL, et al. Insulin independence after solitary islet transplantation in type I diabetic patients using steroid free immunosuppression. N Engl J Med 2000;343:230–8.

D.E.R. SUTHERLAND AND CARL G. GROTH

104 Wahoff DC, Papalois BE, Najarian JS, Kendall DM, Farney AC, Leone JP, et al. Autologous islet transplantation to prevent diabetes after pancreatic resection. Ann Surg 1995;22(4):562–79. 105 Gruessner RWG, Manivel DC, Dunn DL, Sutherland DER. Pancreaticoduodenal transplantation with enteric drainage following native total pancreatectomy for chronic pancreatitis: A case report. Pancreas 1991;6:479–88. 106 Stern RC, Mayes JT, Weber FL Jr, Blades EW, Schulak JA. Restoration of exocrine pancreatic function following pancreas-liver-kidney transplantation in a cystic fibrosis patient. Clin Transplant 1994;8:1–4.

13

This page intentionally left blank

Chapter 2

The nature of the problem: why do we need pancreatic transplantation? Martin Press

Diabetes is one of the cruellest of diseases. Not only does it shorten life expectancy, but it can seriously compromise quality of life in the process. It remains the most common cause of blindness in people of working age, causes renal failure, impotence, neuropathy, foot ulcers, amputations, and heart disease. This makes it a very expensive disease. In the United Kingdom it accounts for some 9 per cent of the NHS budget [1]. In the United States, 14.6 per cent of the health care budget, amounting to some $100 billion per year, was spent on the 4.5 per cent of the population with diabetes [2]. Of this, half was accounted for by medical costs directly related to diabetes (mostly the treatment of advanced diabetic complications, including haemodialysis, foot ulcers, and amputations) and half by societal costs (including long-term disability and loss of earnings due to blindness, amputations, and premature death). Even without complications, a child who develops diabetes at the age of 7 years will have faced costs of $52 000 by the time he is 40 years old [3]. Patients live their lives in the shadow of these complications. And while they worry about the longterm consequences of hyperglycaemia, the short-term effect is that it makes them feel tired, constantly 'below par'. It affects their driving licence, their life insurance premium, their job, their social life. If they try to improve their control they run into an increased risk of acute hypoglycaemia, which in turn affects their life and can be frankly dangerous, not only to them but to other people (if they become hypoglycaemic at the wheel of a car, for example). It is a miserable disease.

Types of diabetes and their causes Diabetes mellitus is the condition which results from failure of the ␤-cells of the pancreas, the only cells in the body able to produce insulin. It is conventionally divided into type 1, type 2, and secondary diabetes. Secondary diabetes can in turn be divided in two. It can result from either a pancreatic lesion (such as pancreatitis, either acute or chronic, haemochromatosis, or a pancreatic cancer) or from an excess of hormones which result in insulin resistance (e.g. acromegaly due to growth hormone excess or Cushing's syndrome due to corticosteroid excess). By far the most common cause of secondary diabetes is iatrogenic steroid-induced diabetes in patients receiving corticosteroids for arthritis, inflammatory bowel disease, or asthma. Type 1 diabetes, which used to be termed insulin-dependent diabetes mellitus, and before that juvenile-onset diabetes, has its peak onset at puberty but can occur at any age. It is an autoimmune disease in which the ␤-cells are selectively destroyed by a T-cell mediated process. Antibodies, either to insulin itself or to ␤-cell antigens such as GAD (glutamic acid decarboxylase) are markers for the condition. They are frequently positive even before there is any identifiable abnormality of glucose tolerance, and

16

WHY DO WE NEED PANCREATIC TRANSPLANTATION?

often do not stay positive for long after the onset of the disease itself [4], possibly because the destruction of the ␤-cells results in the loss of the antigen. The antibodies probably do not play a direct pathogenic role, but rather reflect the ongoing clinically silent insulitis [4]. It is still unclear which antigens the T cells are responding to or what instigates the autoimmune process, although in non-obese diabetic (NOD) mice, which represent an animal model of type 1 diabetes, knockout of the GAD gene prevents T-cell activation and protects the animals from the development of diabetes [5]. In man, it is believed that the disease results from a combination of genetic and environmental factors. There is disequilibrium with certain HLA haplotypes, but concordance within identical twins is less than 40 per cent [6], implying that while genetic factors may confer a predisposition, environmental factors play at least as great a role. What the (?infectious) factors may be which instigate the autoimmune process in genetically predisposed individuals remains unclear [7]. Whatever the explanation(s) there is currently an epidemic of type 1 diabetes in northern Europe such that the incidence is doubling with each successive generation [8]. Type 2 diabetes was formally known as non-insulin-dependent diabetes mellitus and before that as maturity-onset diabetes. If there is at present an epidemic of type 1 diabetes there is in contrast a global pandemic of type 2 diabetes. While it is appreciated that there is a very strong genetic component of type 2 diabetes, such that there is a near 100 per cent concordance rate in identical twins [9], it is increasingly being appreciated that type 2 diabetes is a heterogeneous, polygenic condition, and different families probably have different genotypes underlying their condition. The prevalence of type 2 diabetes is highest (up to 50 per cent) in American Indians and in South Pacific Islanders. These are populations which have evolved to survive caloric deprivation but who are now affluent and obese. The suggestion is that the 'thrifty' genes which have in the past conferred a survival advantage when food was in short supply, are now responsible for a tendency to gain weight easily and to become insulin resistant as a result [10]. The other major recent observation has been that low birth weight is associated with the development of insulin resistance, hypertension, and type 2 diabetes in later life [11]. While this has been ascribed to the consequences of intrauterine malnutrition [12], it has also been pointed out that, since insulin is the major intrauterine growth hormone, low birth weight and subsequent insulin resistance could both represent different phenotypic manifestations of the same 'insulin-resistant' genotype [13]. Patients with type 2 diabetes are characteristically insulin resistant, and insulin resistance is a characteristic feature of obesity. However, while most patients with type 2 diabetes are overweight, even those patients who are not overweight are insulin resistant. However, this in itself is not sufficient to cause diabetes because so long as the ␤-cells are healthy they can compensate for this. Indeed, all obese and sedentary people are insulin resistant to a greater or lesser extent, but only a small percentage of such people have diabetes. A second lesion appears to be necessary in which the ␤-cells themselves are affected. It is not so much that they 'run out of insulin' as that they appear oblivious to the fact that glucose levels are rising and more insulin is needed [14]. Thus, ␤-cells in patients with type 2 diabetes may retain their ability to respond to other secretagogues such as amino acids or sulphonylureas (and of course the latter are used in the treatment of the condition as a result). Insulin levels are typically only slightly low compared to healthy controls, but they are markedly lower than they would be in a control subject given glucose to cause an equivalent degree of hyperglycaemia. The exact nature of the ␤-cell defect remains unclear, but pulses of insulin secretion lose their normal periodicity [15] and the first phase of insulin secretion in response to an acute challenge is lost [16]. Both the insulin resistance and the ␤-cell secretory defect are exacerbated by hyperglycaemia (Fig. 2.1), thus creating two vicious circles.

MARTIN PRESS

Fig. 2.1 Vicious circles in type 2 diabetes. In type 2 diabetes, hyperglycaemia results from a combination of insulin resistance and ␤-cell dysfunction. Both of these abnormalities are in turn exacerbated by hyperglycaemia. Thus, any therapeutic strategy which improves metabolic control tends to improve both insulin sensitivity and ␤-cell function.

In normal subjects the insulin response to oral glucose is very much greater than to intravenous glucose due to the effect of 'incretins'; hormones secreted by the gut which sensitize the ␤-cell to glucose. The major candidates for the incretins are GLP-1 (glucagon-like peptide 1) and GIP (glucose-dependent insulinotropic peptide), both members of the glucagon superfamily. In type 2 diabetes, secretion of GLP-1 is reduced while its actions are normal, whereas the secretion of GIP is normal but its effects are markedly impaired [17,18]. Infusion of GLP-1 normalizes ␤-cell responsiveness and analogues may in the future be developed for therapeutic use. For the time being, treatment remains with drugs which stimulate insulin secretion, such as the sulphonylureas, or drugs which increase insulin sensitivity, such as biguanides or thiazolidinediones.

Clinical course Type 1 diabetes characteristically has an abrupt onset, but in fact autoimmune destruction of the ␤cells has at that point been going on for months or even years. New-onset type 1 diabetes can thus with equal justification be viewed as 'endstage prediabetes'. On the basis of partial pancreatectomy experiments it is believed that one can manage with only 20 per cent or so of one's ␤-cells, but as soon as the number falls below that level, diabetes presents abruptly, characteristically with polyuria, polydipsia, and weight loss. Type 2 diabetes in contrast is an insidious process whereby insulin resistance appears to be present from the beginning and ␤-cell failure supervenes gradually and progressively over a period of many years [19]. Patients may present with tiredness or infections, but the acute presenting symptoms of type 1 diabetes are absent and patients are frequently totally asymptomatic. On the basis of the incidence of complications at presentation, it is believed that diabetes may have been present for 10 years or more prior to presentation [20]. Furthermore, the United Kingdom Prospective Diabetes Study (UKPDS) has illustrated vividly the fact that even after presentation, type 2 diabetes continues to progress inexorably as the ␤-cells gradually fail [21]. However, in contrast to type 1 diabetes, ␤-cell mass remains normal [22].

Actions of insulin and consequences of insulin deficiency Insulin is the body's major anabolic hormone, and affects not only carbohydrate metabolism but lipid and protein metabolism as well. Thus, patients presenting acutely with type 1 diabetes lose weight

17

18

WHY DO WE NEED PANCREATIC TRANSPLANTATION?

because they lose muscle and fat, and it is the breakdown of fat, and its metabolism to ketoacids, which results in diabetic ketoacidosis if the condition is not recognized and treated soon enough. Insulin effectively behaves as two distinct hormones with respect to its anabolic and anticatabolic effects respectively, and this is not merely an academic distinction (Fig. 2.2). We have evolved to take in calories as meals rather than grazing continuously. However, when we are not eating, there is a continuing requirement for fuel and the brain in particular cannot utilize anything but glucose to a significant extent. There is thus a need for distinct effects of insulin following a meal, when high levels result in uptake of glucose by muscle and fat in particular, and in a parallel stimulation of fat uptake by adipose tissue under the influence of lipoprotein lipase, an enzyme which is induced by insulin. In contrast, in between meals and particularly overnight, when insulin levels are much lower, the regulation of glucose production by the liver and of free fatty acid release from adipose tissue under the influence of hormone-sensitive lipase, another enzyme regulated by insulin, results in an appropriate rate of release of these fuels to match requirements. Thus, in the postabsorptive state, insulin levels are regulated exquisitely by the pancreatic ␤-cells to maintain exactly the right hepatic glucose production rate to keep fasting glucose levels constant. If the ␤-cells start to fail, the resultant low levels of insulin would normally be the signal to the liver that the subject is hypoglycaemic, and glucose production increases steeply. Similarly, adipose tissue releases greatly increased amounts of free fatty acids which are in turn converted by the liver to ketoacids. Conversely, if the ▪-cells are working normally, the rise

Fig. 2.2 Regulation of blood glucose levels by insulin. A near-constant blood glucose level is maintained despite a constant glucose uptake by the brain, and an intermittent input of large amounts of glucose following a meal. This is achieved by changes in plasma insulin which turns two 'taps'. In the fasting state, when insulin is present at low concentrations, insulin acts predominantly on the liver to regulate glucose production. Following a meal raised insulin concentrations 'turn off' this tap and 'turn on' a tap to increase peripheral glucose uptake, predominantly in muscle.

MARTIN PRESS

Fig. 2.3 Dose-response curves of glucose kinetics as a function of insulin concentrations. Hepatic glucose production is regulated by insulin at low concentrations of insulin (5 to 10 mU/l). This is thus the major determinant of the fasting glucose concentration. Following a meal, insulin concentrations rise by an order of magnitude (to 50 to 100 mU/l). At these levels, hepatic glucose production is completely suppressed and peripheral uptake is stimulated, thus minimizing the hyperglycaemic excursion following food. Because of the steepness of these dose-response curves, even slightly too much or too little insulin will have major effects on the blood glucose concentration in an insulin-treated patient.

in insulin levels following a meal rapidly shuts off hepatic glucose production and free fatty acid release from adipose tissue. It is the steepness of the dose–response curves which should be appreciated (Fig. 2.3). Slight underor over-replacement of insulin results in disproportionate changes in glucose and fatty acid release from the liver and adipose tissue, respectively. Given that the half-life of insulin in the circulation is only 2 to 3 min, and that insulin levels (in a non-diabetic subject) vary continuously to match requirements, the problems involved in the physiological replacement of insulin become obvious. It is surprising that diabetes can be managed as well as it can with 2 to 4 daily subcutaneous injections of insulin.

Metabolic profiles of diabetes and diabetic monitoring Diabetic control is assessed by monitoring glycosylated haemoglobin (HbA1c) every few months and by home glucose monitoring by the patient himself. HbA1c is formed from HbA in the red cells as they circulate. The reaction is non-enzymic and the amount of HbA1c formed is a function of the mean glucose concentration to which the red cells are exposed. Since the red cell survival time is about 120 days, the average age of the red cells in the circulation at any one time is 60 days, and the HbA1c thus provides a guide to what average glucose levels have been over the past 6 to 8 weeks. In some assays, misleading values can be encountered in patients with haemoglobinopathies, and HbA1c values will also be misleading if red cell survival is shortened, for example in a patient with menorrhagia. Some confusion tends to arise because a normal mean blood glucose concentration is about 5 mmol/l and a

19

20

WHY DO WE NEED PANCREATIC TRANSPLANTATION?

normal level of HbA1c is about 5 per cent. The fact that these figures are numerically identical does not mean that one can translate a given HbA1c into a numerically identical mean glucose level. An HbA1c of 10 per cent translates into a mean glucose level of about 15 mmol/l, although there is considerable variability from one patient to another. This same interpatient variability means that HbA1c cannot be used to diagnose diabetes either. The correct way to use HbA1c measurements is to compare one clinic visit to another in the same patient, in order to ascertain whether diabetic control is getting better or worse. While the HbA1c gives rough guidance on whether diabetic control is adequate, it does not give any information on what to do if it is not. For this one requires the patient's own finger pricks to give information on what glucose levels are at different times of the day. Armed with this information, the aim is to tailor the treatment regimen to the individual patient. It is often possible, just by looking at glucose levels at different times of the day, to tell whether a patient has type 1 or type 2 diabetes. In type 2 diabetes, fasting glucose levels are increased, and, because hepatic glucose production does not fall to zero after a meal, there is an exaggerated rise in glucose level following food. However, the effect is relatively smooth and predictable and varies little from one day to the next. In contrast, type 1 diabetes, in which insulin is totally deficient, is characterized by gross instability and marked day-to-day variability. The incidence of hypoglycaemia, which is rare in type 2 diabetes and virtually always results from too much insulin or too great a dose of a sulphonylurea, is some 20-fold higher in type 1 diabetes since it is not possible with conventional insulin treatment to gauge the dose of insulin required exactly. We teach patients that glucose levels are elevated following food and reduced as a result of exercise or insulin treatment. However, there are clearly many more influences at work than this, and looking at a continuous sensor readout it is often impossible to tell what is causing particular perturbations in diabetic control (Fig. 2.4). One well-recognized influence on diabetic control is the so-called dawn phenomenon, whereby insulin requirements rise between about 4.00 and 8.00 a.m. This appears to result from the effects of growth hormone which is typically secreted as soon as a patient achieves deep sleep at between 11.00 p.m. and 1.00 a.m. These growth hormone peaks induce insulin resistance some 4 h later [23]. If the subject does not have diabetes, the pancreas is able to compensate for this by secreting more insulin but if the ␤-cells are not able to do this, glucose levels will rise. Matters are made worse in the diabetic patient by the fact that this is characteristically the time of day when insulin concentrations are declining as the effects of the previous night's intermediate-acting insulin injection wears off. The effect is most pronounced in teenagers, presumably because they secrete more growth hormone, and is made worse by the fact that growth hormone secretion varies from one night to another. Diabetes is notoriously unstable in teenagers, who tend to get the blame for a phenomenon which is, at least in part, physiological. Insulin resistance is also induced by counter-regulatory hormones secreted during exercise and as a result of hypoglycaemia. The idea that counter-regulatory hormone secretion during hypoglycaemia results in rebound hyperglycaemia (the Somogyi phenomenon) [24] is, however, less of a problem than it might otherwise be because patients with diabetes tend to have impaired secretion of counterregulatory hormones in response to hypoglycaemia [25], but this is a function of how well or badly their diabetes has been controlled.

Insulin regimens While many different types of insulin are available, they can broadly be divided into quick acting and intermediate or long acting. A solution of insulin itself (soluble or regular insulin) will typically start

MARTIN PRESS

Fig. 2.4 Twenty-four hour subcutaneous glucose profile in a patient with type 1 diabetes. The recording is obtained with the MiniMed continuous glucose monitoring system, with a glucose electrode implanted percutaneously into adipose tissue. The arrows represent insulin injections, the first three being preprandial soluble insulin and the fourth a bedtime injection of an isophane insulin. The five circles indicate the patient's own finger-prick blood glucose measurements, which are used in the calibration of the sensor. The sensor tracing indicates that the patient's diabetic control is much worse than he would know from his finger-pricks, with unrecognized episodes of both hypo- and hyperglycaemia. It also illustrates the instability of glucose levels in type 1 diabetes. Clearly there must be other determinants of glucose levels besides insulin, food, and exercise.

to take effect following subcutaneous injection within half an hour. Its peak effect will be seen at 2 to 3 h but a continued effect is seen up to 6 h after an injection. Ultra quick-acting insulin analogues are now also available, which start to act immediately, exert their peak action at 1 to 2 h, and wear off over about 4 h [26]. Intermediate-acting insulins are divided into the isophane and the lente types of insulins. Both have a gradual onset of action with peak effects at 6 to 12 h and a duration of action of 16 to 20 h or more. Insulin analogues which have a less pronounced peak are starting to become available. Intermediateacting insulins come as non-uniform suspensions and considerable variation is seen not only in the amount of insulin actually injected between one injection and the next (depending on how well the insulin is mixed) [27] but also as a result of major differences in absorption rates from one injection to the next [28]. Finally, insulin can be given by insulin pumps. These consist of small battery-operated syringe pumps which are typically worn on the belt or carried in the pocket. The syringe, which contains only quick- or ultra quick-acting insulin, is connected via a tube to a small indwelling subcutaneous catheter. The pump delivers insulin continuously (the basal rate) and thus has advantages over traditional intermediate-acting insulins both in terms of the reliability of absorption [29] and the lack of peaking and waning of the insulin effect. Furthermore, the pump can be programmed to give an increased basal rate during the second half of the night to compensate for

21

22

WHY DO WE NEED PANCREATIC TRANSPLANTATION?

the dawn phenomenon and mimic a physiological overnight insulin profile. Boluses of quick- or ultra quick-acting insulin are given before a meal via the pump. Current pumps are in no way automatic, however, and patients have to do fingerpricks to 'tell' their pump how much insulin to give. Studies in patients treated with an insulin pump show either an improvement in HbA1c or a reduction in the risk of hypoglycaemia, or both [30–32]. A normal HbA1c is rarely attainable, however, and the risk of longterm complications, while reduced, is far from abolished. Furthermore, any attempt to improve diabetic control further is always limited by an increased risk of hypoglycaemia. Without a pump, the best one can do to approximate a physiological insulin profile is to give an injection of quick- or ultra quick-acting insulin before each meal and a long-acting insulin at bedtime. Patients who wish to be less intensive with their management very often settle for two injections a day, each comprising a mixture of quick- and intermediate-acting insulins. Thus, following the morning injection, the quick-acting fraction covers the hyperglycaemic effect of breakfast and the intermediate-acting insulin covers the afternoon, while following the evening injection the quick-acting insulin covers the evening meal and the intermediate-acting insulin then lasts through the night. The problem with all of these regimens is that they represent only a very poor approximation to the patient's actual insulin needs which vary on a minute-to-minute basis. This is in marked contrast to other endocrine failure syndromes. The hypothyroid patient is perfectly well replaced with a single tablet of thyroxine a day, and even the hypoadrenal patient typically requires two or at most three doses of hydrocortisone per day. These hormones are relatively long acting and requirements vary little.

Does tight diabetic control matter? When insulin was discovered in 1921 and started to be used clinically, it was assumed that this would provide the cure for type 1 diabetes. Certainly, in the sense that people rarely succumbed to what had until then been a fatal disease, this was indeed the case. However, during the 1930s and 1940s patients who did not die from diabetic ketoacidosis nevertheless developed what became known as the complications of diabetes. These included renal failure, blindness, amputations, impotence, and heart disease. Thus, it became apparent not only that diabetes still reduced lifespan but more importantly, that it destroyed quality of life. The question that arose as to whether these complications were the direct result of the hyperglycaemia, or whether they, like the ␤-cell failure, arose from some unidentified underlying metabolic abnormality, has only recently been answered. Seminal studies by Pirart in the 1970s established that patients whose diabetic control was worst, as judged by their blood and urine glucose levels, were indeed at an increased risk from what Pirart described as the triopathy of retinopathy, nephropathy, and neuropathy [33]. He failed to find such a relationship, however, with peripheral vascular disease or coronary heart disease (and the exact relationship between hyperglycaemia and coronary heart disease remains controversial [34]). More importantly, he failed to demonstrate that the relationship between the hyperglycaemia and the complications was one of cause and effect. This would require a randomized prospective trial which finally came in 1993 with the publication of the Diabetes Control and Complications Trial (DCCT) [35]. Some 1441 patients with type 1 diabetes were randomized to be left on their existing insulin regime with one or two injections of insulin per day and an HbA1c of 9.0 per cent, or to be tightly controlled either with multiple injections or with insulin pumps. The aim was to normalize glucose metabolism as far as possible but in the event the study showed vividly the impossibility of normalizing control completely. Although a relationship was shown between glucose levels and the same three complications that Pirart had identified 20 years previously (retinopathy, neuropathy, and nephropathy), the price of bringing the HbA1c down from

MARTIN PRESS

9 to 7 per cent was a three-fold increase in the incidence of severe hypoglycaemia from 19 to 62 episodes per 100 patient years. No data was available on whether macrovascular disease was prevented with intensified control since the patient group was too young. The UKPDS compared different ways of managing type 2 diabetes and concluded that the incidence of coronary heart disease was also a function of the prevalent glucose levels, although no significant reduction could be demonstrated with 'tighter' control, possibly because there was insufficient difference between the 'tight' and 'less tight' study groups [21]. Furthermore, strategies for achieving control which resulted in weight gain (sulphonylureas or insulin) may have substituted one cardiac risk factor for another, since the lowest incidence of coronary heart disease who observed in a small subgroup treated with metformin, the only drug which improves diabetic control without causing weight gain [36].

The practicalities of super tight control It is apparent from these various studies that the degree of diabetic control which is achievable with conventional insulin regimens is quite simply insufficient to prevent diabetic complications. Thus, for example, although patients in the intensively treated group in the DCCT had a significantly reduced risk of microangiopathic complications, 10 to 15 per cent had nevertheless developed microalbuminuria, retinopathy, or neuropathy by the end of the 9-year study period [35]. And this was despite frequent clinic attendances to see doctors, nurses, dietitians, and psychologists, and even more frequent phone calls. The problem in type 1 diabetes is that because the blood glucose is unstable and variance is markedly increased, attempts to reduce the average level to normal inevitably result in hypoglycaemia at times when the glucose level is lowest. This risk is substantially less in patients with type 2 diabetes, because of its lower variance, but in both types of diabetes improved control inevitably results in weight gain.

Closed loop systems In order to achieve tight diabetic control without a concomitant increase in the risk of hypoglycaemia, the only solution is a closed loop system, whereby insulin levels are adjusted on a moment-tomoment basis according to requirements (Fig. 2.5). This is of course only what a healthy pancreatic ␤-cell does. There are only two ways of achieving this in a patient whose ␤-cells are no longer

Fig. 2.5 Schematic illustration of closed loop feedback system for regulation of blood glucose. In the future, such a system may be practical by linking a continuous glucose sensor to an insulin pump. For the present, the only way to achieve this is with a transplant of ␤-cells, either as a whole pancreas or as isolated islets.

23

24

WHY DO WE NEED PANCREATIC TRANSPLANTATION?

functioning: either by using a continuous glucose sensor attached to an insulin pump, or by transplanting functioning islets, either in a whole vascularized pancreas or as isolated islets. The first approach, with a closed loop insulin pump, was first tried more than 25 years ago [37]. At that time the machines used were enormous and only potentially useful during surgery or labour. Blood glucose levels were sampled ex vivo in blood withdrawn by a slow continuous pump from an arm vein, and the appropriate amount of glucose or insulin, decided by a computer, was infused into the contralateral arm. At the time it was believed that while it would probably be possible to miniaturize the pump and the glucose sensor in due course, there was no prospect of being able to miniaturize the computer! A portable closed loop insulin pump was reported by a Japanese group 15 years ago [38], but there were problems with indwelling sensors in terms of their biocompatibility, stability, and accuracy. These problems seem finally to have been largely overcome [39], but major concerns about the safety and reliability of an implanted closed loop insulin pump mean that this approach is still some years away from routine clinical use. Thus, at present the only available strategy able to provide a closed loop system is ␤-cell transplantation, whether as a whole vascularized pancreas or as isolated islets of Langerhans. The remainder of this book addresses the practicalities of these two approaches.

References 1 Currie CJ, Kraus D, Morgan CL, Gill L, Stott NC, Peters JR. NHS acute sector expenditure for diabetes: the present, the future and excess in-patient cost of care. Diabet Med 1997;14:686–92. 2 Rubin RJ, Altman WM, Mendelson DN. Health care expenditure for people with diabetes mellitus, 1992. J Clin Endocrinol Metab 1994;78:809A–F. 3 Songer TJ. The economic costs of NIDDM. Diabet Metab Rev 1992;8:389–404. 4 Lernmark A, Falorni A. Immune phenomena and events in the islets in insulin-dependent diabetes mellitus. In: Textbook of diabetes, Pickup J, Williams G, ed. Blackwell Science, 2nd edn 1997;15.1–15.23. 5 Yoon J-W, Yoon C-S, Lim H-W, Huang QQ, Kang Y, Pyun KH, et al. Control of autoimmune diabetes in NOD mice by GAD expression or suppression in ␤-cells. Science 1999;284:1183–7. 6 Barnett A, Eff C, Leslie RDG, Pyke D. Diabetes in identical twins. A study of 200 pairs. Diabetologia 1981;20:87–93. 7 Yoon J-W. Environmental factors in the pathogenesis of insulin-dependent diabetes mellitus. In: Textbook of diabetes, Pickup J, Williams G, ed. Blackwell Science, 2nd edn 1997;14:1–14. 8 Bingley PJ, Gale EAM. Rising incidence of insulin-dependent diabetes mellitus in Europe. Diabetes Care 1989;12:289. 9 Newman B, Selby JV, Slemenda C, Fabsitz R, Friedman GD. Concordance for type 2 (non-insulindependent) diabetes mellitus in male twins. Diabetologia 1987;30:763–8. 10 Neel JV, Weder AS, Julius S. Type II diabetes, essential hypertension and obesity as 'syndromes of impaired genetic homeostasis': the 'thrifty genotype' hypothesis enters the 21st century. Perspect Biol Med 1998;42:44–74. 11 Barker DJ, Hales CN, Fall CH, Osmond C, Phipps K, Clark PM. Type 2 (non-insulin-dependent) diabetes mellitus, hypertension and hyperlipidaemia (syndrome X): relation to reduced fetal growth. Diabetologia 1993;36:62–7. 12 Barker DJ. Intrauterine programming of adult disease. Mol Med Today 1995;1:418–23. 13 Hattersley AT, Tooke JE. The foetal insulin hypothesis: an alternative explanation of the association of low birth weight with diabetes and vascular disease. Lancet 1999;353:1789–92. 14 Porte D. ␤-cells in type II diabetes mellitus. Diabetes 1991;40:166–80.

MARTIN PRESS

15 Polonsky KR, Given BD, Hirsch LJ, Tillil H, Shapiro ET. Abnormal patterns of insulin secretion in non-insulin dependent diabetes mellitus. N Engl J Med 1988;318:1231-9. 16 Luzi L. Effect of the loss of first phase insulin secretion on glucose production and disposal in man. Am J Physiol 1989;257:E241–6. 17 Vaag A, Holst JJ, Vølund A, Beck-Nielsen H. Gut incretin hormones in identical twins discordant for non-insulin dependent diabetes mellitus (NIDDM): evidence for decreased glucagon-like peptide 1 secretion during oral glucose ingestion in NIDDM twins. Eur J Endocrinol 1996;135:425–32. 18 Nauck MA, Heimesaat MM, Ørskov C, Holst JJ, Ebert R, Creutzfeld W. Preserved incretin activity of glucagon-like peptide-1 (7–36 amide), but not of synthetic gastric inhibitory polypeptide in patients with type 2 diabetes mellitus. J Clin Invest 1993;91:301–7. 19 Swinburn BA, Gianchandari R, Saad MF, Lillioja S. In vivo ␤-cell function at the transition to early non-insulin dependent diabetes mellitus. Metabolism 1995;44:757–64. 20 Roxburgh MA, Vaughan NJA, New J. Using large cohorts from population based diabetes registers to develop risk equations for type 2 diabetes. Diabetologia 2000;43(Suppl.1):A1. 21 UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes. Lancet 1998;352:837–53. 22 Rahier J, Sempoux C, Moulin P, Guiot Y. No decrease of the ␤-cell mass in type 2 diabetic patients. Diabetologia 2000;43(Suppl.1):A65. 23 Campbell PJ, Bolli GB, Cryer PE, Gerich JE. Pathogenesis of the dawn phenomenon in patients with insulin-dependent diabetes mellitus: accelerated glucose production and impaired glucose utilization due to nocturnal surges in growth hormone. N Engl J Med 1985;312:1473–9. 24 Somogyi M. Exacerbation of diabetes by excess insulin action. Am J Med 1959;26:169–91. 25 Frier BM, Fisher BM, Gray CE, Beastall GH. Counter-regulatory hormonal responses to hypoglycaemia in type 1 diabetes: evidence for diminished hypothalamic-pituitary hormonal secretion. Diabetologia 1988;31:421–9. 26 Holleman F, Hoekstra JBL. Insulin Lispro. N Engl J Med 1997;337:176–83. 27 Jehle PM, Micheler C, Jehle DR, Breitig D, Boehm BO. Inadequate suspension of neutral protamine Hagedorn (NPH) insulin in pens. Lancet 1999;354:1604–7. 28 Lauritzen T, Faber OK, Binder C. Variation in 125I-insulin absorption and blood glucose concentration. Diabetologia 1979;17:291–5. 29 Lauritzen T, Pramming S, Deckert T, Binder C. Pharmacokinetics of continuous subcutaneous insulin infusion. Diabetologia 1983;24:326–9. 30 Bode BW, Steed RD, Davidson PC. Reduction in severe hypoglycemia with long-term continuous subcutaneous insulin infusion in type 1 diabetes. Diabetes Care 1996;19:324–7. 31 Melki V, Renard E, Lassmann-Vague V, et al. Improvement of HbA1c and blood glucose stability in IDDM patients treated with lispro insulin analog analog in external pumps. Diabetes Care 1998;21:977–82. 32 Boland EA, Grey M, Oesterle A, Fredrickson L, Tamborlane WV. Continuous subcutaneous insulin infusion. A new way to lower risk of severe hypoglycemia, improve metabolic control, and enhance coping in adolescents with type 1 diabetes. Diabetes Care 1999;22:1779–84. 33 Pirart J. Diabète et complications dégénératives. Présentation d'une étude prospective portant sur 4400 cas observés entre 1947 et 1973. Diabète Metab 1977;3:97–107 and 173–82. 34 Barrett-Connor E. Does hyperglycaemia really cause coronary heart disease? Diabetes Care 1997;20:1620–3. 35 The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–86.

25

26

WHY DO WE NEED PANCREATIC TRANSPLANTATION?

36 UK Prospective Diabetes Study (UKPDS) Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes. Lancet 1998;352:854–65. 37 Albisser AM, Leibel BS, Ewart TG, Davidovac Z, Botz CK, Zingg W. An artificial endocrine pancreas. Diabetes 1974;23:389–96. 38 Shichiri M, Kawaniori R, Hakui N, Asakawi N, Yamasaki Y, Abe H. The development of wearable-type artificial endocrine pancreas and its usefulness in glycaemic control of human diabetes mellitus. Biomed Biochem Acta 1984;43:561–8. 39 Renard E, Costalat G, Moran B, Kolopp M, Apostol D, Lauton D, et al. First human experience with combined implantation of a long-term i.v. glucose sensor and an i.p. insulin pump. Acta Diabetol 2000;37:166.

Chapter 3

Experimental models in whole organ pancreatic transplantation Vassilios E. Papalois and Nadey S. Hakim

Introduction The first successful pancreas transplant in humans was performed by Kelly and Lillehei on 17 December 1966 in the Department of Surgery at the University of Minnesota [1]. It was a combined transplant of the tail of the pancreas and the kidney to a female diabetic recipient with endstage renal failure. Since then, the results of pancreatic transplantation have dramatically improved. There is a constant effort for improvement of results in all centers that perform pancreatic transplantation worldwide. The improvements include refinement of surgical technique, to minimize the changes of postoperative surgical complications such as graft pancreatitis and thrombosis, methods of drainage of the exocrine portion of the pancreas, the early detection of rejection, and the effect of pancreas transplantation in establishing an insulin-independent status, as well as in preventing or reversing the secondary complications of diabetes. To address these issues, many experimental models in small and large animals have been developed over the years. The importance of the pancreas for glucose control was initially demonstrated in 1890 when Minkowski performed a total pancreatectomy in a dog, which eventually developed diabetes mellitus and died [2]. Later, the discovery of insulin by Banting and Best in 1922, after extensive experimentation in the dog model, was one of the main breakthroughs of medical science in the twentieth century [3]. The first successful experimental allotransplantation of the pancreas in the dog model was performed in 1927 by Gayet and Guillaumie [4]. In the 1950s, the pioneers in the field of experimental pancreatic allotransplantation in the large animal model were Brooks and Gifford [5] and later, in the 1960s, initially De Jode and Howard [6] and later Lillehei [7] pioneered the field. Finally, in the early 1970s, Orloff et al. developed a reliable model for whole organ pancreas transplantation in the rat [8]. Three animals are now used worldwide for experimental pancreatic transplantation — the rat, the dog, and the pig. In this chapter, we present the principles of anatomy and surgical technique for pancreatic allotransplantation in these three animal models, in addition to the most interesting recent experimental developments and the issues raised in relation to pancreatic transplantation.

The rat model Anatomy The pancreas of the rat is more diffuse, softer, and less clearly delineated than the human pancreas. Its location is approximately the same as that of the pancreas in humans, although it is more spread out [9] (Fig. 3.1). The right lobe and body of the pancreas are embedded in the mesoduodenum and the

28

EXPERIMENTAL MODELS

Fig. 3.1 Gross anatomy of the rat pancreas showing the right lobe and the body in the mesoduodenum. The left lobe is not shown. Pancreas (P), liver (L), duodenum (D), pancreatic duct (PD).

beginning of the mesojejunum. The left lobe is a branched, flattened part that runs along the dorsal aspect of the stomach, embedded in the dorsal leaf of the greater omentum and along the splenic artery towards the hilum of the spleen. The pancreatic acini are drained by 15 to 40 excretory ducts which fuse to form at least two, and sometimes five to eight, main ducts that open into the bile duct. Often, small ducts open directly into the duodenum. The largest collecting duct is from the left lobe and it is the first duct to enter the bile duct. The arterial blood supply to the pancreas is provided by two main arteries, the superior and inferior pancreaticoduodenal arteries (Fig. 3.2). The superior pancreaticoduodenal artery runs caudally in the mesoduodenum and is a branch of the gastroduodenal artery which, in turn, is a branch of the hepatic artery or, sometimes, a direct branch of the coeliac artery. The coeliac artery in the rat is similar to that of the human, arising from the abdominal aorta as the first unpaired branch. The inferior pancreaticoduodenal artery is a branch of the superior mesenteric artery which, in turn, arises from the aorta as the second unpaired branch 3 to 5 mm caudal to the coeliac artery. The superior and inferior pancreaticoduodenal arteries connect with each other. The splenic artery gives a number of small branches to the left lobe of the pancreas. The venous drainage of the pancreas is into the hepatic portal vein (Fig. 3.3). The superior pancreaticoduodenal vein drains the right lobe and body of the pancreas and is the last tributary to enter the portal vein directly. The inferior pancreatic vein drains the left lobe of the pancreas into one of the splenic veins, which in turn drain into the portal vein. The portal vein in the rat is similar to the human portal vein, draining the superior mesenteric vein, gastrosplenic vein, inferior mesenteric vein, right gastroepiploic vein, and veins from the pancreas. The rat duodenum is relatively longer than the human duodenum and is more mobile, since it is suspended by a mesentery, the mesoduodenum. It arises in the midline from the distal end of the pylorus,

V.E. PAPALOIS AND NADEY S. HAKIM

Fig. 3.2 Arterial vascular supply to the rat pancreas. The right lobe and the body of the pancreas are supplied by a branch of the gastroduodenal (superior pancreaticoduodenal) and a branch of the superior mesenteric (inferior pancreaticoduodenal) artery. The left lobe is supplied by branches of the splenic artery.

Fig. 3.3 Venous drainage of the rat pancreas. The right lobe and body drain directly into the portal vein via the superior pancreaticoduodenal vein. The left lobe is drained via the inferior pancreatic vein which drains into one of the splenic veins, which drain into the portal vein.

29

30

EXPERIMENTAL MODELS

runs transversally towards the right abdominal wall, rises dorsally after an initial curve along the right margin of the liver towards the right kidney, curves as a transfer duodenum towards the mid-sagittal plane, and after another right-angled turn runs cranially as the ascending duodenum.

Technique A variety of injectable anaesthetic agents and inhalant agents have been used for anaesthetizing rats for pancreaticoduodenal transplantation. One of the most classic is an injectable anaesthetic cocktail consisting of ketamine (50 mg/kg), acepromazine (1 mg/kg), and xylazine (5 mg/kg), mixed together in a single solution and given intramuscularly [10]. The depth of anaesthesia is controlled by rigidly monitoring the lead reflex and muscle movement. Operations are performed at a constant room temperature to minimize hypothermia and dehydration. During the procurement of the rat pancreas for pancreaticoduodenal transplantation, the vasculature of the graft consists of a segment of the aorta containing the arteries that ultimately supply the pancreas and duodenum, and a segment of the hepatic portal vein containing the venous tributaries that drain the graft [11]. In the recipient, the graft aortic segment and graft portal vein are anastomosed with the host aorta and inferior vena cava, respectively, below the level of the renal vessels (Fig. 3.4) When transplantation of the pancreas without the duodenum is performed, the pancreatic ducts that enter the bile duct and duodenum are meticulously isolated, ligated, and divided. The small arter-

Fig. 3.4 (a) During the procurement of the rat pancreas for pancreaticoduodenal transplantation, the vasculature of the graft consists of a segment of the aorta containing the arteries that ultimately supply the pancreas and duodenum, and a segment of the hepatic portal vein containing the venous tributaries that drain the graft. In the recipient, the graft aortic segment and graft portal vein are anastomosed with the host aorta (arrow) and inferior vena cava (double arrow), respectively, below the level of the renal vessels. The duodenum of the graft is anastomosed end-to-side to the recipient's duodenum. (indicated by the triple arrow) (b) When transplantation of the pancreas without the duodenum is performed, the pancreatic ducts that enter the bile duct and duodenum are meticulously isolated, ligated and divided. The small arteries and veins between the duodenum and pancreas are ligated and divided and the pancreas is carefully separated from the duodenum. The pancreatic blood vessels in the mesoduodenum are carefully preserved.

V.E. PAPALOIS AND NADEY S. HAKIM

ies and veins between the duodenum and pancreas are ligated and divided, and the pancreas is carefully separated from the duodenum. The pancreatic blood vessels in the mesoduodenum are carefully preserved.

Studies Prevention of post-transplant graft pancreatitis Post-transplant pancreatitis is still a frequently occurring complication of whole organ pancreas transplantation and it is usually attributed to microcirculatory disorders caused by cold ischaemia time and reperfusion injury. Vollmar et al. [12] developed an interesting experimental model for the prevention of reperfusion injury following pancreatic transplantation. They studied the effect of providing nitric oxide (known for its vasodilating nature) on the postischaemic microvascular reperfusion injury that occurs after pancreas transplantation. Heterotopic isogeneic pancreaticoduodenal transplantation was performed in Sprague Dawly rats after 16 h of cold storage of the graft at 4° C in histidine-tryptophane-ketoglutarate (HTK) solution. A second group of rats received intravenous Larginine (a nitric oxide provider) immediately before (50 mg/kg) and during the first 30 min of reperfusion (100 mg/kg). Intravital fluorescence microscopy was used for analysis of functional capillary density, capillary diameters, and capillary red blood cell velocity in exocrine pancreatic tissue during the first 120 min after reperfusion. Histology served for a quantitative assessment of inflammatory response (leucocytic tissue infiltration) and endothelial disintegration (oedema formation). In L-arginine treated animals, functional capillary density of exocrine tissue of pancreatic grafts was found slightly higher after 30 and 60 mins, and significantly higher after 120 min of postischaemic reperfusion, compared with the untreated pancreatic grafts. This was accompanied by a significant increase of capillary diameters. In addition, there was a significant attenuation of both leucocytic tissue infiltration and oedema formation in the L-arginine treated animals, when compared with the non-treated controls. The authors concluded that besides reduction of leucocyte-dependent microvascular injury, L-arginine improves postischaemic microvascular reperfusion, by capillary dilatation. These results suggest that a supplement of nitric oxide during reperfusion is effective in attenuating exocrine microvascular reperfusion injury.

Reversibility of rejection It is well known in pancreatic transplantation that abnormal blood glucose levels usually indicate irreversible rejection. It is very important to know at which stage the rejection progress remains reversible. A very interesting model to investigate this was developed by Konigsrainer et al. [13]. A total of 54 Lewis rats became diabetic after intravenous injection of streptozotosin and they were allotransplanted with whole pancreases from Brown Norway rats. All recipients received orally 50 mg/kg of cyclosporin for 5 days. The animals were divided into six groups. On the sixth day after the transplant, cyclosporin was discontinued in the first group for 2, the second for 4, the third for 6, the fourth for 8, the fifth for 9, and the sixth group for 10 days. Two animals of each group were euthanized at the end of the immunosuppression-free interval for histological assessment of rejection. The remaining animals were treated with intravenous methylprednisolone 7 mg/kg for 3 consecutive days and continuation of cyclosporin at a dose of 15 mg/kg orally for another 3 weeks and they were then euthanized for histological assessment of the pancreatic graft. At the end of the immunosuppression-free interval, the number of animals that became diabetic were as follows: five out of nine in group 4, seven out of 11 in group 5, and eight out of 11 in group 6 and there was evidence of rejection in the pancreatic grafts of all animals (two from each group) that were sacrificed. The severity of rejection correlated to the length of immunosuppression-free interval. The histology returned to normal after

31

32

EXPERIMENTAL MODELS

antirejection therapy in four animals (57 per cent) of group 1, in two animals (28 per cent) of group 2, and in one animal (11 per cent) of groups 3 and 4, respectively. Although there were no animals in groups 5 and 6 with normal graft histology after treatment, there were still four (36 per cent) and three (27 per cent) animals, respectively, that were normoglycaemic and that had pancreatic grafts with well-preserved islets. The authors concluded that hyperglycaemia and severe rejection with endothelitis and isletitis could be reversed. Therefore, they suggested that elevated blood sugars should not be considered as an endpoint for pancreas transplantation and a trial of enhanced immunosuppression is justified in patients with advanced pancreas allograft rejection.

Graft immunoprotection It has been shown that transplantation of the pancreas in combination with the liver can protect the pancreas allograft from rejection [14]. An interesting question that arises from this observation is if ongoing pancreatic rejection could be reversed by subsequent transplantation of the liver. An excellent model to investigate this was developed by Wang et al. [15]. Transplantation was performed between PVG (donors) and DA (recipients) rats. The first group of animals was PVG to PVG pancreas isogeneic grafts. The second group was PVG to DA pancreas allografts, groups 3 to 5 were PVG to DA pancreas allografts followed by liver transplantation on days 2, 4, and 6 respectively, and in group 6 there was PVG to DA pancreas allografts 4 weeks after liver transplantation. The results show that pancreas allografts in group 2 were rejected between the seventh and 13th postoperative day. Liver transplantation prevented subsequent pancreas allograft rejection in group 6. Ongoing rejection was reversed by liver transplantation with subsequent graft acceptance in groups 3 to 5. Significant graft infiltrating lymphocyte apoptosis was demonstrated at 2 weeks in pancreas transplants associated with liver grafting. Graft-versus-host disease was not detected in the pancreas recipients. The authors concluded that although pancreas allografts in the PVG to DA combination rejected rapidly with a median survival time of 9 days, liver transplantation could protect subsequent pancreas grafts from rejection and reverse ongoing pancreas graft rejection with subsequent pancreatic acceptance. Graft infiltrating lymphocyte apoptosis could be associated with the process of graft acceptance.

Prevention of diabetic nephropathy The effect of pancreas transplantation in preventing the progression of diabetes in the native kidneys or the recurrence of diabetes in the transplanted kidney the case of combined kidney/pancreas transplantation is of great importance. Spadella et al. developed a very interesting experimental model to study this [16]. Ninety Lewis rats were randomly assigned to three experimental groups. The first group included 30 non-diabetic control rats, the second group included 30 Alloxan-induced diabetic control rats, and the third group included 30 Alloxan-induced diabetic rats that received isogeneic pancreatic grafts from normal donor Lewis rats. Each group was further divided into three subgroups of 10 rats, which were sacrificed at 1, 3, and 6 months post-transplant, respectively. The kidneys of five rats in each subgroup were studied and 50 glomeruli and tubules from each kidney were analysed by light microscopy by two different investigators in a double-blind study. The presence of glomerular basement membrane thickening, mesangial enlargement, and Bowman's capsule thickening in the kidneys of those rats were measured. There was no significant difference for these three parameters between normal controls, diabetic controls, and rats that received pancreatic transplantation at 1 and 3 months post-transplant. There was significant increase in all these three parameters for the diabetic controls, when compared to normal non-diabetic rats and rats that received pancreatic transplantation at 6 months. However, there was no difference between the normal controls and the rats that received pancreas transplantation. The authors concluded that pancreas transplantation in Alloxan-

V.E. PAPALOIS AND NADEY S. HAKIM

induced diabetic rats prevents the development of kidney lesions, beginning approximately at 6 months after transplantation.

Reversal of vascular complications of diabetes There is insufficient evidence as to whether the transplantation of the whole pancreas can reverse vascular complications associated with diabetes. Pieper et al. [17] investigated whether pancreatic transplantation in experimental diabetes reverses established defects in endothelium-dependent relaxation. Streptozotocin-induced diabetic rats underwent whole organ pancreas transplantation after 12 weeks of disease. Endothelial function was evaluated 4 weeks after transplantation and compared with that of the control and age-matched diabetic animals. Blood was taken for analysis of glucose, insulin, total glycosylated haemoglobin, and plasma amino acid levels. Descending thoracic aortas were isolated, sectioned into rings, and mounted in isolated tissue baths. In precontracted rings, endotheliumdependent relaxation to acetylcholine was performed and compared with endothelium-independent relaxation to nitroglycerine as a control. Pancreatic transplantation normalized the increases in glucose and total glycosylated haemoglobin level and decreased serum insulin levels. Diabetes resulted in impaired relaxation to acetylcholine without altering relaxation to nitroglycerine. Pancreatic transplantation completely restored the defective relaxation to acetylcholine. This study suggests for the first time that one aspect of vascular complications (endothelial dysfunction) is amenable to the surgical intervention of pancreas transplantation.

The dog model Many experimental studies for pancreatic auto- and allotransplantation are using the dog as an experimental model. The first reason is that the dog is an animal which is extremely tolerant to general anaesthesia as well as to intraoperative and postoperative stress. As a result, there are much fewer intraoperative problems as well as postoperative morbidity and mortality resulting in better

Fig. 3.5 Canine pancreatic allograft following vigorous cellular and vascular rejection.

33

34

EXPERIMENTAL MODELS

quantitative and qualitative transplant studies. There is also evidence that, following pancreatic allotransplantation, rejection is much stronger in the dog model [18] (Fig. 3.5). This offers an excellent test for the effectiveness of various new immunosuppressive regimens, as well as of methods for tolerance induction following pancreatic transplantation.

Anatomy The dog pancreas starts from the splenic hilum and extends along the major curvature of the stomach and along a large part of the duodenum. The major difference from the human pancreas is that there is no discreet distinction between head, body, and tail of the pancreas. There is a much more simple distinction between the left part (tail) of the pancreas, which extends from the splenic hilum all the way to the take-off of the splenic artery, and the right part of the pancreas which extends after the take-off of the splenic artery to the pylorus, and from the pylorus all the way along the duodenum (Fig. 3.6). The experimental models for pancreatic transplantation in the canine model are using only the tail which has sufficient islet mass to keep a dog normoglycaemic [18]. The blood supply to the tail of the pancreas comes from the splenic artery which takes off from the coeliac artery. The veins of the tail drain into the splenic vein, which drains eventually into the portal vein. Although this is the vascular anatomy of the tail of the canine pancreas in the majority (77.9 per cent) of cases, there are

Fig. 3.6 The dog pancreas has a left part (tail), which extends from the splenic hilum all the way to the take-off of the splenic artery, and a right part which extends after the take-off of the splenic artery to the pylorus, and from the pylorus all the way along the duodenum. The experimental models for pancreatic transplantation in the canine model are using only the tail which has sufficient islet mass to keep a dog normoglycaemic. The blood supply to the tail of the pancreas comes from the splenic artery which takes off from the coeliac artery.

V.E. PAPALOIS AND NADEY S. HAKIM

anatomical variations [19,20] and knowledge of their existence is vital for proper procurement and transplantation of the segmental (tail) pancreatic graft (Fig. 3.7). In 12.1 per cent of cases, there is an extra vein for the tail of the pancreas that drains directly in the portal vein. In 8.8 per cent of cases, the artery supplying the tail takes off from the superior mesenteric artery. In 0.8 per cent of cases, the artery supplying the tail takes off from the superior mesenteric artery and the vein of the tail drains in the superior mesenteric vein. Finally, in 0.4 per cent of cases, the artery supplying the tail takes off from the superior mesenteric artery and the vein drains directly in the portal vein.

Technique A simple and effective way to anaesthetize a dog is as follows [18]: approximately 45 min prior to the induction of anaesthesia, 5 mg of diazepam are given intramuscularly to the animal. After inserting an angiocatheter in one of the veins of the dorsal aspect of one front limb, 15 mg/kg of sodium phenobarbital is given to the animal followed by intubation and connection to a Harvard anaesthesia machine which provides a respiratory rate of 12 to 14/min giving the animal 300 to 400 cm3 of air each time. An initial intravenous dose of phentanyl (1 cm3) is given prior to the initiation of the operation. During the procedure, 50 mg of sodium phenobarbital (5 per cent concentration) and phentanyl (1 cm3) are given intravenously to the animal every 45 to 60 min. Following the procurement, the vessels of the tail of the pancreas (usually the splenic artery and vein) are anastomosed to the external iliac artery and vein of the recipient by using classic vascular anastomosis techniques (Fig. 3.8). As in clinical transplantation, the exocrine portion of the pancreas can be drained either in the bladder or in the bowel and can also be managed by an intraductal Neoprin injection. However, for experimental purposes only, there are two more methods that can be used for controlling the exocrine portion of the pancreas after the transplantation. The first is a simple ligation of the pancreatic duct. This method is not being used clinically because it eventually leads to the development of fibrotic tissue within the pancreatic graft. However, for experimental purposes, it is very effective because it is very simple and at the same time it takes at least 2 to 4 months for the development of fibrotic tissue, which is a period of time which is usually good enough for completing the observations of most experimental studies. The second method for controlling the exocrine portion of the pancreas is the free drainage of the pancreatic duct into the peritoneal cavity. Clinically this can very shortly lead to pancreatic ascites. Experimentally though, and especially in the dog model, the animal can absorb a large amount of the pancreatic fluid without the development of ascites.

Fig. 3.7 The experimental models for pancreatic transplantation in the canine model are using only the tail which has sufficient islet mass to keep a dog normoglycaemic. The blood supply to the tail of the pancreas (A, PT) comes from the splenic artery (A, SA) which takes off from the coeliac artery (A, CA). The veins of the tail drain into the splenic vein (A, SV), which drains eventually into the portal vein (A, PV). Although this is the vascular anatomy of the tail of the canine pancreas in the majority (A, 77.9 per cent) of the cases, there are anatomical variations and knowledge of their existence is vital for proper procurement and transplantation of the segmental (tail) pancreatic graft. In 12.1 per cent (B) of cases, there is an extra vein for the tail of the pancreas that drains directly in the portal vein (arrow). In 8.8 per cent (C) of cases, the artery supplying the tail takes off from the superior mesenteric artery (arrow). In 0.8 per cent (D) of cases, the artery supplying the tail takes off from the superior mesenteric artery (arrow) and the vein of the tail drains in the superior mesenteric vein (double arrow). Finally, in 0.4 per cent (E) of cases, the artery supplying the tail takes off from the superior mesenteric artery (arrow) and the vein drains directly in the portal vein (double arrow). see overleaf

35

36

EXPERIMENTAL MODELS

This page intentionally left blank

V.E. PAPALOIS AND NADEY S. HAKIM

Fig. 3.8 The vessels of the tail of the canine pancreas (usually the splenic artery and vein) are anastomosed to the external iliac artery (arrow) and vein (double arrow) of the recipient. The pancreatic duct (PD) is on free drainage into the peritoneal cavity. Clinically this can very shortly lead to pancreatic ascites. Experimentally though, and especially in the dog model, the animal can absorb a large amount of the pancreatic fluid without the development of ascites.

Studies Prevention of cold ischaemia injury Ischaemic injury of the microvascular endothelium during cold preservation causes a disturbance of vascular microcirculation after reperfusion and results in graft failure. In the canine model, the twolayer preservation method (which continuously supplies sufficient oxygen to the pancreas) leads to continued adenosine triphosphate (ATP) production which is essential to maintain cellular integrity and extends the period of preserved pancreatic viability [21,22]. The aim of a study by Kuroda et al. [23] was to clarify the role of the oxygenation of the pancreatic graft by the two-layer method in the viability of the microvascular endothelium during preservation. Following partial pancreatectomy, segmental (tail) canine pancreases were either autotransplanted immediately (with simultaneous completion pancreatectomy) without preservation (control group), preserved by simple cold storage in Euro Collins solution or by the two-layer method using Euro Collins for 48 h, and then autotransplanted. The viability of vascular endothelial cells was evaluated by using the nuclear trypan blue uptake technique. Pancreatic tissue perfusion was measured with hydrogen gas clearance. Although all grafts that were preserved in just Euro Collins solution failed, graft survival for the control and the double-layer method groups was 100 per cent. The percentage of trypan blue positive vascular endothelium in the Euro Collins group was significantly higher compared to the control group. The two-layer method significantly decreased trypan blue uptake compared to the Euro Collins group, although the uptake was still significantly higher compared to the control group. Pancreatic tissue perfusions after 2 h of reperfusion were in inverse proportion to trypan blue uptake. Specifically, pancreatic tissue perfusions in the Euro Collins group were significantly lower than the control group but the two-layer method improved pancreatic tissue perfusions to levels similar to those of the control group. The authors concluded that oxygenation of the pancreas during preservation by the two-layer method protects the microvascular endothelium from cold ischaemic injury. Consequently, pancre-

37

38

EXPERIMENTAL MODELS

atic microcirculation and tissue perfusion can be maintained after reperfusion, resulting in prolonged graft survival.

Differential diagnoses of early rejection versus graft pancreatitis It is particularly difficult to distinguish between early rejection and graft pancreatitis when early rejection produces an elevated serum amylase level. In a study by Suzuki et al. [24], it was investigated if peripancreatic fluid cytology (an alternative diagnostic tool to graft biopsy that has the potential of causing haemorrhage and pancreatic fistula) can differentiate early acute rejection from graft pancreatitis. The dogs used in this study received either a segmental pancreas allograft or autograft. This study included five groups: allografts without immunosuppression (group A), allografts with immunosuppression (group B), autografts without immunosuppression (group C), autografts with immunosuppression (group D), and autografts treated by 45 min of pretransplant warm ischaemia to induce acute graft pancreatitis (group E). A closed section drainage catheter was placed next to the graft to collect peripancreatic fluid daily after the transplant. Peripancreatic fluid cytology was compared to the corresponding histology through the observation period. Peripancreatic fluid cytology performed on day 1 showed similar neutrophil accumulations in all groups. In sharp contrast, on days 3 and 6, group A had dramatically increased mononuclear cell concentrations in peripancreatic fluid cytology, whereas groups B, C, and D showed significantly lower concentrations. Conversely, in group E, numerous degenerating neutrophils with a marked to moderate increase in necrotic tissue fragments were observed by peripancreatic fluid cytology on days 3 and 6. In terms of graft histology on days 3 and 6, group A showed interstitial mononuclear cell infiltration indicting an active rejection process, whereas groups B, C, and D had minimal inflammatory cell infiltration. In group E, graft pancreatitis was histologically confirmed on days 3 and 6. These results suggest that peripancreatic fluid cytology after pancreas transplantation could be a safe, simple, and a useful diagnostic tool for discriminating early graft rejection from post-transplant graft pancreatitis.

Effect of pancreatic transplantation in preventing the development of diabetic complications In experimental models of canine diabetes, retinopathy, neuropathy, and nephropathy have been shown to develop within 5 years. The aim of the study by Hawthorne et al. [25] was to determine in the canine model whether glucose control provided by segmental duct-occluded pancreas autografts could prevent the long-term complications of diabetes. Thirty-five out-bred mongrel dogs underwent segmental pancreas autotransplantation with residual pancreatectomy. Follow-up over 5 years included endocrine, retinal fundus photography, fluorescein angiography, and nerve conduction studies. Long-term survival was achieved in 14 dogs for 4 to 5 years, and in three dogs for 3 to 5 years. Glycosylated haemoglobin levels remained within normal limits. Fundus photography and fluorescein angiography demonstrated the absence of retinal vascular aneurysms, capillary leakage, and obliteration. Retinal digests show no vascular changes and normal endothelial/pericyte ratios. Nerve conduction was normal and histology of nerves revealed normal density of myelinated fibres and absence of intrafascicular vessels and glucogen deposits, with no changes in the spectrum of fibre diameters and ovoids. Renal histology revealed no evidence of nephropathy with normal glomerular basement membranes. The authors concluded that the duct-occluded segmental pancreatic autografts are capable of providing satisfactory metabolic control for up to 5 years, thereby preventing development of long-term microvascular complications of diabetes. This is a very important study which demonstrates the effect of pancreas transplantation in preventing the complications of diabetes without the intervention of the phenomenon of rejection. In humans, as well as in many experiments

V.E. PAPALOIS AND NADEY S. HAKIM

with allotransplantation, the intervention of rejection and early graft loss does not allow the proper study of the effect of pancreas transplantation itself in preventing the long-term complications of diabetes.

The pig model The pig is a large animal suitable for experimental pancreas transplantation due to its anatomy and transplant immunology, both of which are very similar to humans [26]. However, it is an animal very sensitive to anaesthesia and pancreas transplantation (and especially combined kidney and pancreas) can result in significant intra- and postoperative morbidity and mortality.

Anatomy The pig pancreas anatomically looks much like the dog pancreas. The basic anatomical difference is that there is a portion of the pancreatic tissue that takes off from the lobe of the pancreas that runs along the duodenum, and forms a ring that goes around the portal vein.

Technique The technique of giving anaesthesia to the pig is very similar to that described for the dog. The pig pancreas can be procured with a cuff of donor aorta containing the coeliac access and the superior mesenteric artery, and the portal vein would be used for drainage of the graft. However, there is a technique (Fig. 3.9) for combined en bloc procurement of the pancreas and the left kidney for transplantation that decreases preservation, operation, and clamp time [27]. The donor aorta (with coeliac access, superior mesenteric artery, and left renal artery) is procured and is anastomosed en bloc to the recipient's aorta in a side-to-oblique fashion. The portal vein is anastomosed end-to-side to the distal vena cava, and the left renal vein end-to-side to the left common iliac vein. The donor duodenum is usually anastomosed to the bladder to allow monitoring of rejection by measuring of urine amylase. The donor duodenum can also be anastomosed to the bowel. Due to the fact that there is great resemblance between the immunological events following human and pig pancreas allotransplantation, the pig model has been mainly used to investigate the phenomenon of rejection following pancreas transplantation.

Studies Rejection in single versus combined pancreas and kidney transplantation Clinically, the incidence of reversible renal allograft rejection episodes appears to be higher in recipients of simultaneous kidney pancreas (SPK) transplants than in kidney transplants alone (KTA) [28] although other data show that kidney graft survival is similar in SPK and KTA recipients [29]. Conversely, the rate of irreversible pancreas allograft rejection appears to be higher in pancreas transplant alone (PTA) than in SPK recipients [30]. More information to clarify these clinical phenomena could be obtained by monitoring the progress of kidney and/or pancreas allograft rejection in the immediate postransplant period. Therefore Gruessner et al. [31] studied the incidence and histological severity of the rejection process in the pig model following SPK, PTA, and KTA allotransplantation. SPK and PTA recipients were made diabetic pretransplant by intravenous streptosotosin (150 mg/kg) and SPK and KTA recipients underwent native nephrectomy. Cyclosporin, azathioprine, and prednisone were given in tapering doses from the time of transplantation. To ensure that the course of rejection could be observed, the initial doses of the immunosuppressive drugs were reduced

39

40

EXPERIMENTAL MODELS

Fig. 3.9 En bloc procurement of the porcine pancreatic graft (PG) and the left kidney graft (GK) for transplantation. The graft aorta (GA) [with coeliac access (GCA), superior mesenteric artery (GSMA) and left renal artery (GRA)] is procured and is anastomosed en bloc to the recipient's aorta (RA) in a side-tooblique fashion (arrow). The graft portal vein (GPV) is anastomosed end-to-side to the recipient's distal inferior vena cava (RIVC) (double arrow), and the left renal vein (GRV) end-to-side to the recipient's left common iliac vein (RCIV) (triple arrow). The pancreatic graft duodenum (PGD) is anastomosed to the recipient's urine bladder (RUB) to allow monitoring of rejection by measuring of urine amylase. The graft ureter (GU) is anastomosed to the RUB. The PG is placed medially to the caecum.

by 50 per cent at the end of the first week and by another 50 per cent by the end of the second week post-transplant. Grafts were biopsied weekly to grade histological severity of intestinal and vascular rejection. Pancreas allograft biopsies showed a significantly lower incidence of moderate/severe tubulointerstitial rejection in SPK than PTA recipients at 1 and 2 weeks post-transplant. Likewise, 1 and 2 weeks post-transplant, vascular rejection was moderate/severe in significantly fewer SPK than PTA grafts. In contrast to the pancreatic grafts, kidney allograft biopsies did not show any difference in the incidence of moderate/severe tubulointerstitial rejection for SPK and KTA recipients at 1 and 2 weeks post-transplant. However, very importantly, the incidence of moderate/severe vascular rejection was significantly lower in the SPK than KTA renal allografts at 1 and 2 weeks post-transplant. The authors concluded that histological progression of pancreas allograft rejection is less severe in SPK than PTA recipients, suggesting that the kidney downregulates or dilutes the immune attack to the pancreas. In addition, renal allograft rejection is not increased by the addition of the pancreas; if anything the incidence of kidney vascular rejection is decreased by the presence of the pancreas.

V.E. PAPALOIS AND NADEY S. HAKIM

Correlation of rejection of the duodenum with rejection of the pancreas following pancreaticoduodenal transplantation A very interesting study was designed to assess the correlation of rejection of the duodenum and the pancreas [32] since transcystoscopic biopsy of the duodenum of a bladder-drained pancreaticoduodenal graft could provide vital information for the presence and severity of pancreatic rejection. Pancreatic duodenal tissue from pancreaticoduodenal transplants in 32 out-bred Yorkshire landraised pigs was examined. After Streptosotosin-induced hyperglycaemia, the animals were transplanted and treated with prednisone, azathioprine, and cyclosporin. To assure that the phenomenon of rejection will progress, immunosuppression was reduced by 50 per cent weekly and discontinued over 3 weeks. The tissue samples were obtained at necropsy at various time points. Each organ was graded for interstitial rejection and vascular rejection separately as: no, mild, moderate, and severe. All but one animal rejected the organs. The results showed that, concordance of type (interstitial, vascular) and grade of duodenal and pancreas rejection occurred in 47 per cent of cases. Discordant cases usually show higher grades of rejection in the pancreas, but the opposite can also occur. It seems that if duodenal biopsies are positive, they are likely to be representative of pancreatic pathology, but when negative they do not rule out rejection of the pancreas. Finally, interstitial rejection appears to precede vascular rejection, suggesting that factors released during interstitial rejection play a role in endothelial cell activation and vascular rejection.

Differences in rejection grading after simultaneous pancreas and kidney transplantation Clinical observations suggest that recipients of multiorgan transplants from the same donor can express the clinical manifestations of rejection of one organ while the other remains unaffected [33]. If this is the case histologically as well, is a question that needs to be answered. In a well-designed study [34] the authors investigated this phenomenon in 36 diabetic (streptozotocin induced) bilateral nephrectomized immunosuppressed (cyclosporin, azathioprine, prednisone) pig recipients of simultaneous (same donor) pancreas (bladder-drained) and kidney allografts by grading the histological intensity of rejection in biopsies of each organ at definite intervals post-transplant. As in similar previous studies, immunosuppression was gradually reduced to assure that rejection will progress. Interstitial rejection was graded as absent, mild, moderate, and severe in, respectively, 8, 25, 42, and 25 per cent of pancreas versus 4, 12, 27, and 57 per cent of kidney biopsies at 1 week; and 0, 43, 29, and 29 per cent of pancreases versus 10, 0, 30, and 60 per cent of kidney biopsies at 2 weeks. Although the distribution of grades was similar in the two organs the grade of rejection for each pair at 1 week was discordant in 75 per cent (42 per cent differed by 1 and 33 per cent by ≥ 2 grades) and at 2 weeks in 57 per cent (29 per cent by 1 and 29 per cent by ≥ 2 grades) of the recipients. The inability to use the severity of interstitial rejection in one organ to predict the findings in the other is exemplified by the fact that for the two pancreases without interstitial rejection at 1 week, the corresponding kidney showed moderate or severe rejection, and for the one kidney without rejection, the corresponding pancreas showed moderate rejection. Vascular rejection grades (absent, mild, moderate, severe) also showed a similar distribution for the pancreas (57, 30, 9, and 4 per cent) versus kidney biopsies (50, 38, 0, and 12 per cent) at 1 week and 2 weeks (57, 29, 0, and 14 per cent for the pancreas versus 78, 11, 0, and 11 per cent for the kidney biopsies). However, the grading of vascular rejection in organ pairs was disynchronous in 54 per cent of the recipients at 1 week and 29 per cent of the recipients at 2 weeks. No vascular rejection in the pancreas with rejection in the kidney was seen in 23 per cent of the recipients in 1 week and in 0 per cent of the recipients at 2 weeks, while no rejection in the kidney with

41

42

EXPERIMENTAL MODELS

rejection in the pancreas was seen in 23 per cent of the recipients at 1 week and in 29 per cent of the recipients at 2 weeks. These data demonstrated that histologically severe rejection can be present in one organ while the other is not affected. The severity of histological rejection usually varies between the organs of a pair, but there is no consistent pattern whether the kidney or the pancreas is more or less afflicted. This study confirmed in histology the phenomenon of clinical manifestations of rejection of one organ without rejection of the other but further studies are necessary to investigate why the recipient expresses different immunological responses to two organs that come for the same donor and have the same expression of human leucocyte antigen (HLA) antigens. A possible explanation is the expression of different subclasses of antigens in each organ that play an important role in triggering the phenomenon of rejection.

Summary Experimental studies to investigate detection and prevention of graft pancreatitis and rejection following pancreas transplantation as well the effect of pancreas transplantation on the secondary complications of diabetes can be successfully conducted in the rat, dog, and pig model. The rat model has the advantage of using small, less developed animals with relatively small experimental cost; conversely, the results are not always applicable to humans. The dog model offers experimentation with minimum morbidity and mortality and a strong immunological response to allotransplantation which is a really good test for investigating the efficacy of any new immunoregulation method. However, it is not physiologically and immunologically so similar to humans and results from studies in the dog model cannot always be translated in clinical practice. Finally, the pig, although it is a model related to significant morbidity and mortality during experimentation, it is very similar to humans from the physiological and immunological point of view. Therefore, most of the studies in the pig model (especially those related to rejection) can, usually, have clinical application.

References 1 Kelly W, Lillehei R, Merkel F, et al. Allotransplantation of the pancreas and duodenum along with the kidney in diabetic nephropathy. Surgery 1967;61:827. 2 Minkowski O, Mering VJ. Diabetes mellitus after pancreas extirpation. Arch Exp Path Pharmakol 1889;26:371. 3 Banting FG, Best G. The internal secretion of the pancreas. J Lab Clin Med 1922;7:251. 4 Gayet R, Guillaumie M. La regulation de la secretion interne pancreatique par un processus normal demontré par la transplantations du pancreas. Cr Soc Biol 1927;97:1613. 5 Brooks JR, Gifford GH. Pancreatic homotransplantation. Transplant Bull 1959;6:100. 6 De Jode LR, Howard JM. Studies in pancreatic duodenal homotransplantation. Surg Gynecol Obstet 1962;114:553. 7 Lillehei RC, Simmons RL, Najarian JS, et al. Pancreaticoduodenal allotransplantation. Experimental and clinical experience. Ann Surg 1972;172:405. 8 Orloff MJ, Lee S, Charters AC, et al. Long-term studies of pancreas transplantation in experimental diabetes mellitus. Ann Surg 1975;182:198. 9 Orloff MJ, Orloff MS. Whole pancreas transplantation in the rat. In: Cramer DV, Podesta LG, Makowka L, ed. Handbook of animal models in transplantation research, Section III. Boca Raton: CRC Press, 1994.

V.E. PAPALOIS AND NADEY S. HAKIM

10 Orloff MJ, Greenleaf G, Girard B. Reversal of diabetic somatic neuropathy by whole pancreas transplantation. Surgery 1990;108:179. 11 Lee S, Orloff MJ. Techniques of vascular anastomoses and organ transplantation in the rat. Proc Int Macrosurg Soc 1975;1:131. 12 Vollmar B, Janata J, Yamauchi JI, Menger MD. Attenuation of microvascular reperfusion injury in rat pancreas transplantation by l -arginine. Transplantation 1995;67:950. 13 Konigsrainer A, Mark W, Hechenleitner P, et al. At what stage does pancreas allograft rejection become irreversible? An experimental study. Transplantation 1997;63:631. 14 Wang C, Sun J, Wang L, et al. Combined liver and pancreas transplantation induces pancreas allograft tolerance. Transplant Proc 1997;29:1145. 15 Wang C, Sun J, Li L, et al. Conversion of pancreas allograft rejection to acceptance by liver transplantation. Transplantation 1998;62:188. 16 Spandella CT, Mercadante MC, Schellini SA, et al. Effect of pancreas transplantation on the prevention of nephropathy in alloxan induced diabetic rats. Brazil J Med Biol Res 1996;29:1019. 17 Pieper GM, Adams MB, Roza AM. Pancreatic transplantation reverses endothelial dysfunction in experimental diabetes mellitus. Surgery 1998;123:89. 18 Papalois VE. Contribution to the study of experimental pancreatic allotransplantation. PhD Dissertation Thesis, Paschalides Medical Publications, Athens, Greece, 1992. 19 Florack G, Sutherland DER, Cavallini M, et al. Technical aspects of segmental pancreatic allotransplantation in dogs. Am J Surg 1983;146:565. 20 Ortega-Serrano J, Mendoza-Aroca A. Experimental left pancreas autotransplantation in dogs. Anatomical aspects. Res Surg 1990;2:72. 21 Matsamuto S, Kuroda Y, Hamano M, et al. Direct evidence of pancreatic tissue oxygenation during preservation by the two layer method. Transplantation 1996;62:1667. 22 Kawamura T, Kuroda Y, Suzuki Y, et al. Seventy two hour preservation of the canine pancreas by the two layer (Euro Collins solution/perfluorochemical) cold storage method. Transplantation 1989;47:776. 23 Kuroda Y, Fujita H, Matsumoto S, et al. Protection of canine pancreatic microvascular ebdothelium against cold ishemic injury during preservation by the two layer method. Transplantation 1997;64:948. 24 Suzuki Y, Kuroda Y, Tanioka Y, et al. Peripancreatic fluid cytology. Detection of early rejection versus graft pancreatitis after canine pancreatic transplantation. World J Surg 1997;21:880. 25 Hawthorne WJ, Wilson TG, Williamson P, et al. Long-term duct-occluded segmental pancreatic autografts. Absence of microvascular diabetic complications. Transplantation 1997;64:953. 26 Cooper DKC, Ye Y, Rolph L, et al. The pig a potential organ donor for man. In: Cooper DKC, Kemp E, Reemtsma K, White DJG, eds. Xenotransplantation. Heidelberg: Springer-Verlag, 1991. 27 Gruessner RWG, Tzardis P, Schechner R, et al. En bloc simultaneous pancreas and kidney allotransplantation in the pig. J Surg Res 1990;49:366. 28 Nakache R, Mainetti L, Tyden G, et al. Renal transplantation in diabetes mellitus: influence of simultaneous pancreas transplantation on outcome. Transplant Proc 1990;22:624. 29 Sutherland DER, Gruessner A, Moudry-Munns KC, et al. Tabulation of cases from the International Pancreas Transplant Registry (IPTR) and analysis of United Network for Organ Sharing (UNOS) United States of America (USA) Pancreas Transplant Registry Data according to multiple variables. Transplant Proc 1993;25:1707. 30 Sutherland DER, Gruessner RWG, Gillinham K, et al. A single institution's experience with solitary pancreas transplantation: a multivariate analysis leading to improved outcome. In: Terasaki PI, ed. Clinical transplants 1991. Los Angeles: UCLA Tissue Typing Laboratory, 141, 1992. 31 Gruessner RWG, Nakhleh R, Tzardis P, et al. Rejection in single versus combined pancreas and kidney transplantation in pigs. Transplantation 1993;56:1053.

43

44

EXPERIMENTAL MODELS

32 Gruessner RWG, Sutherland DER, Tzardis P, et al. Correlation of rejection of the duodenum with rejection of the pancreas in a pig model of pancreaticoduodenal transplantation. Transplantation 1993;56:1353. 33 Sollinger HW, Stratta RJ, D'Alessandro AM, et al. Experience for simultaneous pancreas-kidney transplantation. Ann Surg 1988;208:475. 34 Gruessner RWG, Nakhleh R, Tzardis P, et al. Differences in rejection grading after simultaneous pancreas and kidney transplantation in pigs. Transplantation 1994;57:102.

Chapter 4

Pretransplant medical evaluation for pancreas transplant candidates Jerry McCauley and Robert J. Corry

Introduction Pancreas transplantation has rapidly moved from an experimental procedure associated with high rates of morbidity and mortality to a mainstream technique with excellent patient and graft survival. The rate of pancreas transplantation increased from less than 200 per year in 1987 to approximately 1000 per year in 1998. [1,2] Recent Medicare funding for simultaneous kidney and pancreas transplantation (SPK) and pancreas after kidney (PAK) transplantation in the United States will likely fuel more rapid growth. The recent proliferation of new immunosuppressant agents will also likely contribute further to improve graft survival in patients with PAK and pancreas transplantation alone (PTA) and increase the rate of transplantation in these patients as well. The medical evaluation prior to transplantation is crucial since many arrive with pre-existing cardiac disease and other complications of diabetes, which may substantially increase the risk of graft loss and death. The indications for pancreas transplantation is in evolution and may be highly affected by the preferences of the transplant centre, improvements in allograft survival related to improved immunosuppression, and surgical technique. The indications for pancreas transplantation are also evolving and are not uniformly accepted by the rapidly growing number of pancreas transplant centres. Likewise, the contraindications for pancreas transplantation are changing as patient and graft survival continues to improve. As with most forms of transplantation, more challenging patients will likely be referred for pancreas transplantation. The medical evaluation of pancreas transplant recipients will certainly become more demanding as the art and science of pancreas transplantation progresses.

Genetics and pathogenesis of type 1 diabetes Evaluation of the potential pancreas transplant recipient must begin with determination of the type of diabetes. The American Diabetes Association (ADA) has recently reclassified diabetes on aetiological grounds, (Table 4.1) [3]. In the revised classification primary and secondary categories of type 1 diabetes have been eliminated. Patients with low or absent insulin production (type 1 diabetics) benefit from replacement of pancreatic ␤-cells capable of producing insulin. Accordingly, type 1 diabetes of the immune or idiopathic type are the most common candidates for pancreatic transplantation. Although most patients with type 1 diabetes are young (< 25 years) at the time of initial diagnosis, type 1 diabetes may develop at any age; even in the geriatric population. The older description of maturity-onset diabetes of the young (MODY) was abandoned for a new classification that identifies the specific genetic defects leading to diabetes. Mutations of the gene hepatic nuclear factor 1␣ and 4␣ are responsible for MODY3 and 1, respectively. A mutation of the glucokinase gene is responsible for

46

PRETRANSPLANT MEDICAL EVALUATION

Table 4.1 Aetiological classification of diabetes mellitus (adapted from Diabetes Care 1998; 21:S5–S19) (1) Type 1 diabetes (a) immune mediated (b) idiopathic (2) Type 2 diabetes (may range from predominantly insulin resistance with relative insulin deficiency to a predominantly secretory defect with insulin resistance) (3) Other specific types (a) Genetic defects of b-cell function (i) chromosome 12, HNF-1a (MODY3) (ii) chromosome 7, glucokinase (MODY2) (iii) chromosome 20, HNF-4a (MODY1) (iv) mitochondrial DNA (b) Others (c) Gestational diabetes mellitus HNF, hepatic nuclear factor; MODY, maturity-onset diabetes of the young.

MODY2. Mutations of mitochondrial DNA results in ineffective conversion of proinsulin to insulin and an effective insulinopenic state. A long list of other specific causes of diabetes form much of the remainder of the new classification of diabetes but most of these patients such as type 2 diabetics are not candidates for pancreatic transplantation. Finally, gestational diabetes retained in the new classification. Autoimmune type 1 diabetes develops after immune destruction of the ␤-cells in the islets of Langerhans of the pancreas [4]. The clinical manifestations of type 1 diabetes was once felt to result from a sudden illness such as a viral infections which initiated rapid destruction of ␤-cells and an immediate need for insulin therapy. It has now become clear that a chronic process of ␤-cell destruction precedes the clinical illness by many years. Hyperglycaemia is an insensitive measure of pancreatic function and mass. At least 70 per cent of the ␤-cells in a normal pancreas must be lost before hyperglycaemia develops [5]. The genetic markers of autoimmune type 1 diabetes are present from birth [6]. These patients are felt to have a genetic predisposition, which is activated by some environmental stimulus. The major gene associated with type 1 diabetes is located on chromosome 6 in association with genes related to immune recognition [4]. Both susceptibility and resistance to type 1 diabetes has been localized to HLA DR and DQ genotypes [7]. Either HLA DR3, DQB1*0201 or HLA DR4, DQB1*0302 is present in greater than 90 per cent of type 1 diabetes. In fact, if patients have both HLA DR3 and DR4, the lifelong risk of type 1 diabetes is even greater. Forty-five per cent of the general white population in the United States have either DR3 or DR4 genotypes [7]. Although this is a common antigen in the American white population, the presence of protective genotypes (DR4, DQB1*0302/DR3, DQB1*0201) and others may account for the low prevalence of type 1 diabetes. Other genes are associated with susceptibility to type 1 diabetes. A recent genome-wide search based upon the human genome project has located at least 20 chromosomal regions associated with susceptibility to type 1 diabetes [8]. It is now clear that the genetic predisposition and clinical expression of type 1 diabetes varies by race and geographical origin. Type 1 diabetes is most prevalent in descendants of northern Europe and less common in other ethnic groups such as those of African, Asian or native North American descent [4,9]. In Europe, the prevalence was demonstrated to have a strong North–South gradient with the highest rate in Finland and the lowest in southern Europe [4,10]. Destruction of pancreatic islet cells is mediated by autoimmune mechanisms. Genetic predisposition does not appear to be sufficient to initiate this process. Patients who are destined to develop type

J.McCAULEY AND R.J. CORRY

1 diabetes have a genetic susceptibility but appear to require an environmental event or multiple events to initiate the autoimmune process leading to isletitis and ultimately leading to destruction of ␤-cells [4]. Potential environmental factors associated with the development of diabetes include viral infection (Coxsackie, enteroviruses, rubella), dietary factors (cow's milk in infant formula, nitrates in drinking water), and others [4].

The evaluation process Simultaneous kidney and pancreas transplantation (see also chapter 5) Approximately 88 per cent of all pancreas transplants are performed as SPK procedures. [1]. Evaluation for the potential SPK transplant recipient is similar to the evaluation for the kidney transplant alone (KTA) recipients with relatively small modification [11]. The protocol for recipient evaluation is detailed in Table 4.2. At our centre, the transplant nurse co-ordinator obtains the initial history. The focus of the history centres on the presentation of diabetes (age at diagnosis for diabetes and immediate need for insulin therapy), the complications of diabetes, prior cardiovascular disease, and factors that might increase the urgency for pancreas transplantation (absence of hypoglycaemic symptoms with documented hypoglycaemia). The history of renal failure is sought in addition to medical or surgical past history. The nurse co-ordinator's preliminary assessment forms the basis for Table 4.2 Pancreas and kidney transplant evaluation protocol at the University of Pittsburgh Professional evaluation

Nephrologist, surgeon, nurse co-ordinator, social worker, dentist

Laboratory studies

General: creatinine, electrolytes, calcium, phosphorus, SGOT, SGPT, GGTP, alkaline phosphatase, total cholesterol, triglyceride, LDL, HDL, amylase, lipase, total protein, albumin, CBC, platelet count, PT, PTT, RPR Diabetes related: C-peptide, hemoglobin A1c

Viral serology

CMV, hepatitis B, hepatitis C, Epstein-–Bart, herpes zoster, herpes simplex, varicella

Other studies

ECG, chest X-ray, PPD and controls, urine C& S

Immunological evaluation

ABO type, HLA typing, DR typing, circulating antibody, quick PRA, crossmatch

Urological evaluation

Ultrasound of kidneys and right upper quadrant

Cancer screening

Women 35 or older: mammogram All women: gynecology examination and Papanicolaou smear Men ≥ 40 years old: PSA All patients ≥ 50 years old: sigmoidoscopy

C & S, Clinitest and Acetest; CBC, complete blood cell count; CMV, cytomegatovirus; GGTP, g-glutanyltranspeptidase; HDL/LDL, low and high-density lipoproteins; PPD, punified protein derivative of tuberculin; PRA, panel reactive antibody; PSA, prostatespecific antigen; PT, prothrombin time; PTT, partial thromboplastin time; RPR, rapid plasmin regain test; SGOT, serum glutamate oxaloacetate transaminase; SGPT, serum glutamate pyruvate transaminase.

47

48

PRETRANSPLANT MEDICAL EVALUATION

the physician and surgeons evaluation and helps to streamline the process. The nephrologist performs a complete history, physical examination, and reviews medical records with the intention of identifying all previous medical conditions but pays special attention to pre-existing cardiovascular disease and complications of diabetes. The surgeon's evaluation also aim to identify prior medical problems but gives particular emphasis to problems that might cause technical complications with the transplant procedure itself. The social worker's assessment of the patient is crucial given the burden of longstanding diabetes, which may have resulted in depression or other serious psychosocial problems. The laboratory evaluation for pancreas transplant recipients is similar to that for renal transplant candidates. The general laboratory studies, viral serology, and tissue typing is identical. Since pancreas transplantation is only appropriate for patients who are insulinopenic, demonstration of the level of insulin production by the patient's native pancreas is required. An absent or very low C-peptide is considered by most centres to be sufficient to confirm the diagnosis of type 1 diabetes. In addition, a measure of long-term glycaemic control (haemoglobin A1c) is obtained. More specific studies such as nerve conduction velocity, among others, are performed by some centres in an attempt to document the changes in diabetic complications after transplantation. Screening for pre-existing cancers is an important aspect of the medical evaluation of the pancreas recipient. Our policy is to follow the recommendations of major organizations for cancer preventive health services in all of our patients and at times to require screening at earlier ages given the potentially prohibitive risk of mortality for cancers developing after transplantation. The abdominal ultrasound screens for evidence of renal carcinoma and cholelithiasis. An asymptomatic renal carcinoma usually does not require delay in listing for a pancreas transplant outside that needed for convalescence after nephrectomy. At our centre, gallstones are removed prior to transplantation. This approach, however, is not universally accepted. Some centres do not require pretransplant cholecystectomy but remove the gallbladder only if the usual indications are present. Women 35 years of age or older obtain screening mammograms and all sexually active women undergo gynaecological evaluations including Papanicolaou smears. Men 40 years of age or older obtain prostate-specific antigen determinations and all patients 50 years or older perform a screening sigmoidoscopy. The goal of the medical evaluation is to identify risk factors that may adversely affect postoperative morbidity or mortality and to correct or minimize them prior to transplantation. The potential pancreas transplant recipient is known to be at risk for accelerated atherosclerosis due to diabetes but may also have many of the risk factors present in the general dialysis population. The cardiac evaluation of all potential pancreas transplant recipients will be described in detail later in this chapter. During the initial evaluation, however, each of the known risk factors for atherosclerosis should be addressed and the patient should be counseled regarding all modifiable risk factors. The American Heart Association and the International Task Force for Prevention of Coronary Heart Disease have advanced specific measures for primary prevention [12,13]. These measures were not explicitly developed for patients with endstage renal disease (ESRD) but appear prudent until more specific recommendations tailored to these patients are developed. The major elements of these two reports include: (a) smoking cessation; (b) blood pressure control; (c) correction of hyperlipidaemia; (d) increase in physical activity; (e) weight reduction; and (f) oestrogen replacement which is determined on an individual basis. Kasiske et al. reported that approximately 25 per cent of the dialysis patients evaluated for renal transplantation were current smokers, which is similar to the general population [14]. Hypertension is also highly prevalent in the ESRD population including diabetes and has been estimated to be 70 to 85 per cent. Diabetics with severe autonomic neuropathy tend to develop orthostatic hypotension. Once ESRD has occurred and the patients are persistently hypervolaemic, the orthostatic component is attenuated unless they become volume depleted after dialysis. Hyperlipidaemia is also common in

J.McCAULEY AND R.J. CORRY

ESRD patients and may be a particular problem for diabetics who have recently initiated dialysis since the nephrotic syndrome due to diabetic nephropathy may persist. Physical inactivity is very common in ESRD and may be particularly difficult in diabetics with labile blood glucose control. In addition, the other complications of diabetes such as neuropathy, and diffuse atherosclerotic disease may make developing a regular exercise programme challenging. Obesity is a growing problem in the dialysis population and maintaining the appropriate body weight is even more difficult for the diabetic. Tight blood glucose control typically increases body weight, and exercise for the purpose of weight reduction is difficult. Despite the difficulties in controlling these risk factors, any evaluation of the potential candidate for pancreas transplantation should include counselling in these areas. The indications for SPK in patients with ESRD or near ESRD are relatively straightforward (Table 4.3). Once type 1 diabetes has been established, the severity of diabetic complications should be assessed. The American Diabetes Association has recently released a position paper on pancreas transplantation [15]. This group suggested that the potential SPK candidate should: (a) already plan to have a kidney transplant; (b) meet the medical indications and criteria for kidney transplantation; (c) have significant clinical problems with exogenous insulin therapy; (d) not have excessive surgical

Table 4.3 Indications for pancreas transplantation (adapted from ADA [15]) Simultaneous kidney and pancreas transplant

Pancreas after kidney transplant

Pancreas transplant alone

(1) Endstage renal disease with type 1 diabetes with other diabetic compli cations (2) Near endstage renal disease with other diabetic complications (3) Planned bilateral nephrectomy in dia betic with significant other diabetic complications (4) Prior renal transplant which is failing in a type 1 diabetic Prior kidney transplant in type 1 diabetic with other diabetic complications (1) Patients with a history of frequent acute severe metabolic complications requiring medical attention (2) Clinical and emotional problems with insulin therapy that are so severe as to be incapacitating (3) Consistent failure of other therapeutic approaches Other potential indications: (4) Presence of diabetic complications which are progressive and unresponsive to intensive insulin therapy (5) Early diabetic nephropathy associated with other diabetic complications (6) Subcutaneous insulin resistance (7) Following total pancreatectomy (8) Insulin allergy (case report)

49

50

PRETRANSPLANT MEDICAL EVALUATION

risk for the dual procedure. Although most of these recommendations are self-explanatory, having significant clinical problems with exogenous insulin is less concrete. Patients who have frequent admissions for diabetic ketoacidosis, or extreme lability of glucose control fit into this group. Likewise, those with hypoglycaemia without prodromal symptoms could develop severe brain damage or death if such an event occurs while the patient is unattended. Many consider hypoglycaemic unawareness to be an absolute indication for SPK. Recent animal and human studies suggest that hypoglycaemic unawareness is largely if not completely due to recurrent or chronic hypoglycaemia [16]. When hypoglycaemia is prevented, the awareness of hypoglycaemia frequently returns. The problem of hypoglycaemic unawareness can be present in any patient with insulin-dependent diabetes mellitus (IDDM) but those involved in intensive insulin regimes aimed at normoglycemia are two to four times more likely than patients treated with conventional insulin protocols [17]. In addition to recurrent hypoglycaemia, other risk factors for hypoglycaemia include duration of diabetes, presence of autonomic neuropathy, and strict glycaemic control [18]. Once patients have developed diabetic nephropathy and renal insufficiency, other diabetic complications are almost always present. The diagnosis of diabetic nephropathy, however, is seldom confirmed by renal biopsy in patients with longstanding diabetes. It is usually assumed the renal failure is due to diabetes. Patients without other endorgan complications of diabetes and renal failure may have another underlying renal disease. Those with short duration diabetes (even if the C-peptide confirms type 1 diabetes), absence of retinopathy or other complications are at high risk for having non-diabetic renal disease. Candidates without other diabetic complications but who have renal failure may not be ideal for SPK since the other diabetic complications may not develop in the future. In such atypical cases, a renal biopsy may be useful if the patient does not have longstanding ESRD. Patients with multiple severe diabetic complications including severe atherosclerotic disease also may not be candidates for SPK. Such patients have a very high risk of perioperative complications and may not experience a significant improvement in the quality of life outside of the simple cessation of insulin treatment. Whether pancreatic transplantation reverses the complications of diabetes such as gastropathy or neuropathy, and stabilizes the retinopathy, is controversial [19]. This is largely based upon the lack of randomized studies and limited follow-up in case series. It is, however, generally agreed that the more severe the complication, the less likely it is to fully reverse. If there is little chance of diabetic complications reversing and cardiovascular risk is great, such patients should opt for KA or continue dialysis therapy depending upon the patient's particular clinical situation.

Pancreas after kidney The evaluation of potential recipients for pancreas after kidney (PAK) transplantation is similar to that for the SPK. Approximately 10 per cent of all pancreas transplants are PAK [1]. The PAK patients fall into two groups, those with a prior kidney transplant alone and those with prior SPK in those with a failed pancreas allograft. Since these patients have ESRD, the benefit of a functioning pancreas transplant is similar to the SPK group. In all potential PAK recipients, a detailed evaluation of the current renal allograft must be made. Patients with poor graft function may be candidates for SPK instead. The presence of evidence for significant chronic allograft nephropathy should raise the questions of SPK with removal of the failing renal allograft at the time of the SPK. Patients with marginal renal function and without a recent allograft biopsy should be assumed to have chronic allograft nephropathy. Confirmation of this diagnosis by biopsy is preferred but a clinical picture of a chronically elevated creatinine with or without proteinuria may be sufficient. A 24-h urine collection for creatinine

J.McCAULEY AND R.J. CORRY

clearance and protein should be obtained to better assess the level of renal impairment. No exact cutoff has been determined for creatinine clearance before replacing the prior renal allograft but patients with creatinine clearances less than approximately 30–40 cc/min and significant proteinuria will likely have progressive loss of renal function and require dialysis or transplantation in the future. The added nephrotoxicity of cyclosporin or tacrolimus after the pancreas transplant may accelerate the deterioration in renal function. The decline in glomerular filtration rate (GFR) from nephrotoxicity varies by the drug blood levels achieved and the prevailing renal function at the time of transplantation. Brennan et al. have demonstrated that the serum creatinine and creatinine clearance after a dose of cyclosporin may predict renal function in candidates for PTA and by inference PAK patients with marginal renal function [20]. If it is decided that SPK may be too early (patients with creatinine clearance > 30–40 cc/min), these patients must be followed with periodic assessments of their renal function if they are placed on the waiting list for PAK. If the patient has good allograft function, the evaluation is identical to that for the SPK. For patients who were transplanted greater than 1 year at the time of the evaluation for PAK, all studies should be repeated. This is particularly true for the cardiovascular and immunological evaluations. There has been a growing interest in performing living related renal transplantation followed by PAK (KTA – LD + PAK). Obtaining a planned renal transplant quickly rather then waiting for a pancreas transplant has been particularly attractive to some patients and some transplant centres have offered this option as the preferred approach if a donor is available. For many patients, however, this has been a difficult decision given the superior graft survival of the SPK. A recent cost-utility analysis has demonstrated that SPK is the optimal procedure even if a living kidney donor is available [21]. In this analysis SPK dominated (more effective and less expensive) living related followed by PAK strategy. The major factor causing the KTA – LD + PAK option to be less cost-effective was the inferior graft survival compared to SPK. From this analysis, in order for KTA – LD + PAK to be as costeffective as SPK the former 5-year pancreas survival would need to exceed 86 per cent which has not been possible even in the large volume centres with the best graft survivals. The pre-transplant evaluation of KTA – LD + PAK option is similar to the cadaveric kidney PAK scenario. It is particularly important to inform patients of the predictably inferior pancreas survival if this option is chosen.

Solitary pancreas (see also Chapter 6) Pancreas transplantation alone was performed in only 3 per cent of the patients in the International Pancreas Transplantation Registry [1]. As with all other pancreas transplants, PTA allograft has continued to improve from less than 50 per cent 1-year graft survival in the mid-1980s to 76.6 per cent in 1998 [22]. The indications for PTA are listed in Table 4.3. As with other forms of pancreas transplantation, the ADA has proposed indications for PTA [15]. Unlike SPK or PAK recipients, candidates for PTA do not otherwise require immunosuppression to prevent the renal allograft rejection with their predictable side-effects. There is also the additional surgical risk, which should carry a relatively small morbidity and mortality. Candidates for PTA should not have severe renal insufficiency requiring imminent renal replacement therapy. Such patients would be more appropriate candidates for preemptive SPK. It is expected that candidates for PTA should have near normal renal function and do not have severe disabling diabetic complications. The ADA has suggested that: (a) patients with a history of frequent acute severe metabolic complications requiring medical attention; (b) those with clinical and emotional problems with insulin therapy that are so severe as to be incapacitating; and (c) those with consistent failure of other therapeutic approaches should be considered for PTA. In addition the above recommendations, potential candidates for PTA should have diabetic complications

51

52

PRETRANSPLANT MEDICAL EVALUATION

which are progressive. Transplantation so early in the course of diabetes that few if any diabetic complications exist would pose significant risk with little certainty of demonstrable benefit. Insulin allergy is an unusual potential indication for PTA. The evidence for this indication is in one case report only [23]. A young woman with longstanding type 1 diabetes developed severe urticaria, which quickly developed into angio-oedema and respiratory distress after treatment with human insulin which she had taken for many years. Beef and pork insulin did not improve the symptoms. Attempts at desensitization were unsuccessful and the fear of a life-threatening anaphylactoid reaction prompted the referral for PTA. Her only diabetic complications included mild retinopathy, one episode of diabetic ketoacidosis, and rare hypoglycaemic episodes. There was no evidence for renal dysfunction. She developed normoglycaemia after the successful PTA without further episodes of insulin allergy. The incidence of insulin allergy has decreased from approximately 50 per cent to 2 and 10 per cent with the advent of human insulin [23]. Allergy to human insulin also occurs and may be due to additives such as zinc, protamine, non-insulin proteins, aggregates of insulin molecules, and animal proteins [23,24]. Pancreas transplantation for this indication should be rare and all other measures should be exhausted if this is the only indication for PTA.

Candidate risk factors for pancreas transplantation The major objective of the pretransplant recipient evaluation is to identify factors which increase the risk of death, graft loss or major morbidity after pancreas transplantation (Table 4.4). Many of the known risk factors are not peculiar to pancreas transplantation but must be considered during the evaluation period. As with renal transplant alone recipients, the following factors must be considered to increase the risk to potential pancreas transplant recipients: (a) increasing age; (b) obesity; (c) adverse psychosocial factors; (d) pre-existing cardiovascular disease; (e) chronic viral infection (hepatitis B or

Table 4.4 Contraindications to pancreas transplantation Absolute contraindications Severe cardiac disease

Severe peripheral vascular disease Malignancy(1)

(1) Coronary artery disease not amenable to anatomical correction (2) Cardiomyopathy with low ejection fraction not corrected by surgical or medical management Aortoiliac disease without possibility of surgical correction Recently diagnosed (requires appropriate waiting period after treatment) (2) Metastatic and untreatable

Relative contraindications (centre specific) HIV infections

Acceptable at some centres if no AIDSdefining illness

Severe diabetic complications unlikely to improve after pancreas transplantation

Only advantage to patient is cessation of insulin and relaxatin of dietary restrictions. Such patients are usually high-risk candidates for cardiac and other complications

J.McCAULEY AND R.J. CORRY

C, parvovirus, HIV); (f) gastrointestinal disorders (peptic ulcers, pancreatitis, diverticulosis); (g) chronic pulmonary disease (chronic restrictive or obstructive disease) chronic fungal disease (histoplasmosis, etc.); and (h) previously treated malignancy. The approach to evaluation of these problems in the renal transplant setting has been examined in great detail and will not be discussed further here. The risk factors specifically affecting pancreas transplant recipients has been examined previously by Gruessner et al [25]. This report retrospectively examined the factors predicting patient and graft survival in addition to the risks for technical complications in KTA, SPK, PAK, and PTA recipients between 1986 and 1993 with 5-year follow-up. Factors considered for the model included: (a) recipient age greater or less than 45 years; (b) obesity; (c) hypertension; (d) blindness in at least one eye; (e) known cardiac disease previous myocardial infarction, coronary bypass, or percutaneous angioplasty; (f) peripheral vascular disease (previous cerebrovascular accident or transient ischaemic attack, bypass, angioplasty, or amputation); and (g) retransplantation (previous pancreas transplant). Patient survival was adversely affected by recipient age greater than or equal to 45 years in the SPK (relative risk (RR) 3.0] and PAK groups (RR 5.86). Recipients of PTA were adversely affected by recipient age. Cardiac disease increased the risk of mortality only in SPK recipients (RR 3.78). For PAK patients only a previous pancreas transplant and peripheral vascular disease increased the risk of death. Manske et al. reported a more pessimistic view of candidate factors after pancreas transplantation [26]. This report summarizes the patient outcomes from 1987 to 1993 in 173 consecutive IDDM patients. This centre offered pancreas transplants to high-risk recipients with advanced diabetic complications and those with underlying cardiac disease although every attempt was made to correct any anatomical lesions prior to transplantation. They also offered living related renal transplantation first followed by pancreas transplantation as the preferred option if possible. In this high-risk group of patients 3-year patient survival was 68 per cent for SPK, 86 per cent for living related, and 90 per cent for KTA. A Cox proportional hazard model (which included type of organ transplant in the model) identified age in 5-year increments (RR = 1.5, P = 0.001), history of congestive heart failure (RR = 2.7, P = 0.03), and SPK (RR = 3.1, P = 0.02) as predicators of increased risk of mortality. During this period the International Pancreas Transplant Registry was reporting 84 per cent, 3-year patient survival for SPK and a response to this study by Secchi et al. emphasized that excellent results were possible with SPK if patient selection differed from the Manske report [27]. These workers emphasize the importance of excluding patients with severe macroangiopathy (previous strokes, severe dilated cardiomyopathy, and amputations). They likewise favoured SPK over living related renal transplant followed by pancreas transplantation given the superior graft survival with the former. Despite this careful evaluation, cardiovascular events were the major cause of death in all groups (13 per cent KTA, 8 per cent in kidney with segmental pancreas, and 6 per cent in SPK). These two reports and the aggregate experience of the International Pancreas Transplant Registry suggest that excellent patient and allograft survival should be the expectation after pancreas transplantation of all types. Performing these procedures in patients with severe diabetic complications and marginally corrected cardiac disease may lead to unacceptable morbidity and mortality.

The cardiac evaluation in the pancreas transplant candidate Cardiovascular disease (CVD) is the leading cause of death in diabetic ESRD patients. Candidates for pancreas transplantation are known to have accelerated atherosclerosis due to diabetes itself, other non-diabetes related factors seen in the general population (smoking, lack of exercise, etc.) and may have an added risk from renal failure. Patients with chronic renal disease appear to have greater risk of developing CVD than the general population. Mortality from CVD has been estimated to be 10 to 20

53

54

PRETRANSPLANT MEDICAL EVALUATION

times that of the general population [28]. There is no consensus on the nature of the cardiovascular evaluation in the diabetic patient with ESRD. Some centres perform coronary angiography in all patients and others attempt risk stratification, performing angiography in selected high-risk patients. The cardiac evaluation in diabetics is complicated by the high prevalence of silent ischaemia and cardiomyopathy. Studies have found that exercise stress tests with or without thallium or dobutamine echocardiography are of limited predictive value in diabetics [29]. Only 45 per cent of diabetics are capable of reaching 70 to 80 per cent of their maximum predicted heart rate. One recent study, however, suggested that dipyridamole thallium stress tests might be predictive of perioperative cardiac events [30]. In this study only one of 111 (0.9 per cent) patients with a normal study developed a perioperative cardiac event and none of those with fixed defects or reversible defects with less than 50 per cent narrowing of coronary arteries on angiography developed perioperative events. This optimistic view of non-invasive cardiac evaluations has not been the experience of most caring for these patients. Manske et al. have developed an algorithm which attempts to identify low-risk diabetic patients who can safely avoid angiography in the pretransplant evaluation [31]. This group found an 88 per cent prevalence of coronary artery disease in diabetics over 45 years of age and therefore recommended angiography in all diabetics in this group. In young patients without smoking history or ST–T wave changes on electrocardiogram (ECG), or in those with diabetes for less than 25 years, the risk of coronary disease was found to be lower. These patients were recommended to undergo a dipyridamole thallium stress test as the initial study. Any abnormality on this study would require angiography, even in the low-risk group. Although this was a relatively small study, this approach probably represents the most useful way to screen diabetics. Once significant coronary artery disease has been found, anatomical correction by coronary revascularization or angioplasty should be performed before transplantation. Coronary revascularization prior to transplantation in the diabetic has already been shown to decrease the frequency of cardiac events and mortality [32]. Many centres are advocating the use of dobutamine echocardiography as the preferred non-invasive screening study for coronary artery disease in diabetic and non-diabetic renal transplant candidates. It has the advantage of providing information on the presence of ischaemia, wall motion and ejection fraction. As with dipyridamole thallium stress testing this study may not be as accurate in patients with ESRD. Herzog et al. have reported the only study which directly compares dobutamine stress echocardiography (DSE) with coronary angiography [33]. Fifty candidates for renal transplantation underwent DSE followed by coronary angiography. Of these patients 39 (78 per cent) were diabetic with 10 (26 per cent) being type 1 and the remaining type 2. Twenty of the 50 dobutamine studies were positive for inducible ischaemia. Three false negative dobutamine studies were found in patients with more than 70 per cent stenosis and for false negative studies in patients with more than 75 per cent stenosis (including patients with > 70 per cent). The sensitivity and specificity of DSE were respectively 52 and 74 per cent for 50 per cent or greater stenosis, 75 and 71 per cent for stenosis greater than 70 per cent, and 75 and 76 per cent for stenosis greater than 75. The positive and negative predictive values were respectively 70 and 57 per cent for greater than or equals 50 per cent, 45 and 90 per cent for greater than 70 per cent and 60 and 87 per cent greater than 75 per cent stenosis. The authors appropriately conclude that DSE is a useful but imperfect screening test for angiographicallly defined coronary artery disease. Several algorithms have been advocated to screen for coronary artery disease in pancreas transplant candidates. They all attempt to define the low- and high-risk patient and perform angiography as the first diagnostic study only in the high-risk group. Williams has suggested an algorithm which is similar to the Manske approach but attempts to avoid cardiac screening in low-risk patients [34].

J.McCAULEY AND R.J. CORRY

Fig. 4.1

Low-risk patients could potentially avoid a cardiac evaluation altogether if the risk of coronary artery disease appeared to be negligible. Such patients might be placed directly on the waiting list. Our approach to cardiac evaluation (an adaptation of the Manske and Williams approaches) is displayed in Fig. 4.1. In practice, most centres consider a non-invasive stress test to be the minimum evaluation in all pancreas transplant candidates. Patients having a positive stress test require coronary angiography. Those with lesions greater than 70 per cent should undergo angioplasty or coronary artery bypass grafting before being placed on the waiting list. Candidates with inoperable diffuse disease should not be transplanted. Candidates with lesions less than 70 per cent could be approved for transplantation but must be re-evaluated every 6 months to 1 year with a DSE. These patients should also have annual DSE after transplantation. High-risk candidates are those with a prior history of coronary artery disease and/or multiple risk factors. Since the presence of peripheral arterial disease and/or carotid artery disease increases the chances of having coronary artery disease, such patients would be considered to be at high risk and probably should proceed directly to angiography.

Organ allocation system The United Network of Organ Sharing (UNOS) administers the allocation of pancreatic organs in the United States and has devised a system based upon geographical origin of the organ, number of antigen mismatches, and waiting time [35]. Whole pancreata are first allocated locally, regionally, then nationally. The transplant centre in the local allocation scheme can chose a patient waiting for an isolated pancreas, kidney-pancreas (SPK), or solid organ–islet combination from the same donor. If there is a zero antigen mismatched with any patient in the nation, the pancreas is mandated to be allocated to that patient. Within each waiting list, the organs are allocated based upon blood type

55

56

PRETRANSPLANT MEDICAL EVALUATION

compatibility and waiting time on the transplant list. Blood type O organs are mandated to be transplanted into type O recipients. If the organ is not used at the local level it is allocated regionally then nationally. The organs are allocated by blood type and waiting time in the following sequence: (a) isolated pancreas candidates with one A, B, or DR antigen mismatch, then (b) isolated pancreas candidates with two A, B, or DR antigen mismatches, then (c) isolated pancreas candidates with three A, B, or DR antigen mismatches, then (d) combined kidney pancreas candidates if a kidney is available, then (e) isolated pancreas candidates with four more A, B, or DR antigen mismatches [35]. As mentioned earlier, zero antigen mismatches are mandated to be shared at the national level. If the whole pancreas has not been used locally, regionally, or nationally, it can then be used for islet transplantation. The host organ procurement organization offers the pancreas for islet transplantation locally, regionally, then nationally. At the regional and national levels allocation of the organ is determined by HLA matching, medical urgency, and waiting time. Candidates with zero HLA antigen mismatches receive 3 points, those with one mismatch 2 points, and patients with two mismatches receive 1 point. Potential recipients with three or more antigen mismatches do not receive points. Medical urgency for islet cell transplantation is divided into two groups. Status 1 patients are those who have already received an islet cell transplant within the previous 3 weeks and are considered to be the most urgent. Patients awaiting the first islet cell transplant are assigned to status 2. Patients with the longest waiting time are given 1 point. A fraction of a point is assigned to subsequent patients based upon their waiting time compared to the others on the list.

References 1 1999 Annual Report of the US Scientific Registry for Transplant Recipients and the Organ Procurement and Transplantation Network: Transplant Data: 1989–1998. US Department of Health and Human Services, Health Resources and Services Administration, Office of Special Programs, Division of Transplantation, Rockville, MD: UNOS, Richmond VA. 2 Sutherland D. Pancreas and pancreas–kidney transplantation. Curr Opin Nephrol Hypertens 1998;7:317–25. 3 Annoymous. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 1998;21:S5–S19. 4 Atkinson MA, Maclaren NK. Mechanisms of disease: The pathogenesis of insulin-dependent diabetes mellitus. N Engl J Med 1994;331:1428. 5 Bonner-Weir S, Trent DF, and Weir GC. Partial pancreatectomy in the rat and subsequent defect in glucose-induced insulin release. J Clin Invest 1983;71:1544–53. 6 McCulloch DK, Palmer JP. The appropriate use of B-cell function testing in the preclinical period of type 1 diabetes. Diabet Med 1991;8:800. 7 Todd JA, Bennett JC. A practical approach to identification of susceptibility genes for IDDM. Diabetes 1992;41:1029–34. 8 Davies, JL, Kawaguchi, Y, Bennett ST, et al. A genome-wide search for human type 1 diabetes susceptibility genes. Nature 1994;371:130. 9 Diabetes Epidemiology Research International Mortality Study Group. Major cross-country differences in risk of dying for people with IDDM. Diabetes Care 1991;14:49–54. 10 Green A, Gale EAM, Patterson CC. Incidence of childhood-onset insulin-dependent diabetes mellitus: the EURODIABAC study. Lancet 1992;339:905–9. 11 McCauley J. Evaluation of the potential renal allograft recipient. Stamford: Appleton & Lange, 1997:43–72.

J.McCAULEY AND R.J. CORRY

12 Grundy SM, Balady GJ, Criqui MH, Fletcher G, et al. Guide to primary prevention of cardiovascular disease: A statement for healthcare professionals from the Task Force on Risk Reduction. Circulation 1997;95:2329–31. 13 Assmann G, Carmena R, Cullen P, Fruchart J, Jossa F, et al. Coronary heart disease: reducing the risk: A world wide view. International Task Force for the Prevention of Coronary Heart Disease. Circulation 1999;100(18):1930–8. 14 Kasiske BL, et al. The adverse effects of cigarette smoking in renal transplant recipients. J Am Soc Nephrol 2000;11(4):753–9. 15 Anonymous. Pancreas transplantation for patients with diabetes mellitus. Diabetes Care 1998;21:S79. 16 Bolli GB. Counterregulatory mechanisms to insulin-induced hypoglycemia in humans: relevance to the problem of intensive treatment for IDDM. J Ped Endo Met 1998;11(Suppl.1):103–15. 17 The DCCT Research Group: Epidemiology of severe hypoglycemia in the diabetes control and complications trial. Am J Med 1991;90:450–9. 18 Kendall D, Rooney DP, Smets YFC, Bolding LS, Robertson RP. Pancreas transplantation restores epinephrine response and symptom recognition during hypoglycemia in patients with long-standing type 1 diabetes and autonomic neuropathy. Diabetes 1997;46(2):249–57. 19 Hricik DE. Combined kidney–pancreas transplantation. Kidney Int 1998;53:1091–102. 20 Brennan DC, Stratta RJ, Lowell JA, Miller SA, Taylor RJ. Cyclosporine challenge in the decision of combined kidney–pancreas versus solitary pancreas transplantation. Transplantation 1994;57(11):1606–11. 21 Douzdjian V, Escobar F, Kupin W, Venkat KK, Abouljoud M. Cost–utility of living-donor kidney transplantation followed by pancreas transplantation versus simultaneous pancreas–kidney transplantation. Clin Transplant 1999;13(1):51–8. 22 1999 Annual Report of the US Scientific Registry for Transplant Recipients and the Organ Procurement and Transplantation Network: Transplant Data: 1989–1998. US Department of Health and Human Services, Health Resources and Services Administration, Office of Special Programs, Division of Transplantation, Rockville, MD; UNOS, Richmond, Virginia, pp. 156. 23 Oh HK, Provenzano R, Hendrix J, El-Nachef MW. Insulin allergy resolution following pancreas transplantation alone. Clin Transplant 1998;12(6):593–5. 24 Simmond JP, Russell GI, Cowley AJ, et al. Generalized allergy to porcine and bovine monocomponent insulins. Br Med J 1980;281(6236):355–6. 25 Gruessner RWG, Dunn DL, Gruessner AC, Matas AJ, Najarian JS, Sutherland DER. Recipient risk factors have an impact on the technical failure and graft survival rates in bladder-drained pancreas transplants. Transplantation 1994;57:1598–606. 26 Manske C, Wang Y, Thomas W. Mortality of cadaveric kidney transplantation versus combined kidney–pancreas transplantation in diabetic patients. Lancet 1995;346:1658–62. 27 Secchi A, Caldara R, Di Carlo V, Guido P. Mortality of cadaveric kidney transplantation versus combined kidney–pancreas transplantation in diabetic patients. Lancet 1996;347:827. 28 Foley RN, Parfrey PS, Sarnak MJ. Clinical epidemiology of cardiovascular disease in chronic renal disease. Am J Kidney Dis 1998;32(5):S112–19. 29 Morrow CE, Schwartz, Sutherland DER, et al. Predictive value of thallium stress testing for coronary and cardiovascular events in uremic diabetic patients before renal transplantation. Am J Surg 1983;146:331–5. 30 Mistry BM, Bastani B, Solomon H, Hoff J, Aridge D, Lindsey L, et al. Prognostic value of dipyridamole thallium-201 screening to minimize perioperative cardiac complications in diabetics undergoing kidney or kidney–pancreas transplantation. Clin Transplant 1998;12(2):130–5. 31 Manske CL, Thomas W, Wang Y, Wilson R. Screening diabetic transplant candidates for coronary artery disease: Identification of a low risk subgroup. Kidney Int 1993;44:617–21.

57

58

PRETRANSPLANT MEDICAL EVALUATION

32 Manske CL, Wang Y, Rector T, Wilson R, et al. Coronary revascularization in insulin dependent diabetic patients with chronic renal failure. Lancet 1992;340:998–1002. 33 Herzog CA, Marwick TH, Pheley AM, White CW, Rao VK, Dick CD. Dobutamine stress echocardiography for the detection of significant coronary artery disease in renal transplant candidates. Am J Kidney Dis 1999;33(6):1080–90. 34 Williams ME. Management of the diabetic transplant recipient. Kidney Int 1995;48:1660. 35 Annual Report of the US Scientific Registry for Transplant Recipients and the Organ Procurement and Transplantation Network: Transplant Data: 1989–1998. US Department of Health and Human Services, Health Resources and Services Administration, Office of Special Programs, Division of Transplantation, Rockville, MD; UNOS, Richmond, Virginia, pp. 418–19.

Chapter 5

Indications for kidney and pancreas transplantation and patient selection Konstantinos N. Haritopoulos and Nadey S. Hakim

Diabetes mellitus (DM) afflicts 6 per cent of the general population and is currently the third most common disease and the fourth leading cause of death by disease in the United States [1,2]. Of the estimated 16 million diabetic patients in the United States, 4 million take insulin and 1 million have IDDM (insulin-dependent diabetes mellitus) (type I), juvenile onset. Nearly 30 000 new cases of IDDM are diagnosed each year, and the incidence is increasing [2]. The syndrome of IDDM includes not only abnormal glucose metabolism but also specific long-term complications such as retinopathy, nephropathy, and neuropathy [3]. Although exogenous insulin therapy is effective at preventing acute metabolic decompensation and is life-saving, the majority of patients with diabetes will develop one or more endorgan complications during their lifetime [3,4]. Tight glucose control is even more important than previously recognized, as demonstrated by a multicentre trial — the Diabetes Control and Complications Trial (DCCT) [5]. The results of this study demonstrated that intensive control of glucose can significantly reduce (but not reliably protect against) the long-term complications of diabetes. Therefore, in patients who fail other ways of intensive insulin therapy, pancreas transplantation (PTx) may be the best available treatment option because it appears to be the single most effective method in achieving tight glucose control in an ambulatory setting [6]. Nearly 13 000 cases of PTx have been reported to the International Pancreas Transplant Registry (IPTR) and over 170 centres worldwide have performed a PTx [7,8]. This number is probably higher since it is not obligatory for transplant centers not in the United States to report their number of PTx to the IPTR. Insulin independence approaches 82 per cent at 1 year with a 94 per cent patient survival [9]. There are various categories for performing a PTx; either simultaneous pancreas and kidney (SPK), pancreas after kidney (PAK), or pancreas transplant alone (PTA). The majority are SPK transplants (90 per cent), 4 per cent are PAK and less than 6 per cent are PTA [10]. However, unlike liver, lung, and heart transplantation, PTx is not a life-saving procedure and its value must be balanced against the risks of the operative procedure and the inevitable long-term immunosuppressive therapy. Therefore, the indications for this procedure and the selection of patients are critical to ensure low mortality and an improvement in quality of life.

Criteria for pancreas transplantation All recipients should be type I DM, since no benefits have been shown conclusively in type II DM patients because of the peripheral insulin resistance and pre-existing C-peptide secretion [11] (Table

60

INDICATIONS FOR KIDNEY AND PANCREAS TRANSPLANTATION

Table 5.1 Criteria for pancreas transplantation Inclusion criteria ● Presence of IDDM ● Presence of secondary diabetic complications ● Ability to withstand surgery and immunosuppression ● Socio-psychological suitability ● Ability to understand the therapeutic nature of pancreas transplantation ● Ability to understand and comply with the long-term immunosuppression and need for follow-up Exclusion criteria ● Insufficient cardiovascular reserve (a) angiography indicating non-correctable coronary artery disease (b) ejection fraction below 50% (c) recent myocardial infarction ● Malignancy ● Ongoing drug or alcohol abuse ● History of non-compliance with treatment ● Major psychiatric illness ● Active infection ● Age over 65 years ● Lack of well-defined secondary diabetic complications ● Extreme obesity (> 50% of ideal body weight) ● Psychological instability

5.1). C-peptide deficient patients make better candidates for PTx and the majority of pancreas transplant recipients in the United States are C-peptide deficient. However, some IDDM patients that have normal C-peptide levels can achieve good glucose control following PTx [12]. Nevertheless, PTx is an operation that is reserved for patients who have developed secondary complications due to DM. Only 50 per cent of patients with diabetes of 20 years or more will develop severe neuropathy, retinopathy, or nephropathy with 30 per cent progressing to endstage renal failure [7]. In view of the difficulty in predicting those patients who will develop chronic diabetic complications, PTx is currently performed in those patients with established complications. Pancreas transplantation is a difficult operation and to achieve good results it is essential that patients who are possible transplant recipients are in good general health. For the best results, patients should be carefully selected and any pre-existing disease should be scrupulously assessed and evaluated. This is of particular importance in coronary artery disease and, if suspected, it is better to diagnose and treat early by coronary artery bypass grafting before proceeding to PTx [13]. Coronary angiography should be performed in cases in which the history, physical examination, or non-invasive cardiac studies (stress test) reveal an abnormality [14,15]. In addition, angiography of the cerebral arteries is indicated when the carotid Doppler is not normal [16]. Characteristics such as age over 45, diabetes for more than 25 years, a smoking history, longstanding hypertension, previous major amputations due to peripheral vascular disease, or history of cerebrovascular disease are among the usual indications for performing cardiac catheterization [14–15]. A history of previous myocardial infarction, angioplasty, or coronary artery bypass grafting are not necessarily contraindications to PTx. In these cases, stress thallium and echocardiographic imaging in combination with coronary angiography are helpful in defining the operative risk. Echocardiography provides a functional assessment whereas coronary angiography illustrates the anatomical details. Major amputations secondary to severe peripheral vascular disease or severe visual impairment are not considered absolute contraindi-

K.N. HARITOPOULOS AND NADEY S. HAKIM

cations [17]. This is especially true for uraemic patients who can benefit from the reversal of uraemia by an SPK transplant. However, it should be borne in mind that these patients are both unlikely to derive improved function at the endorgan level from the restoration of glucose homeostasis, and that the iliac atherosclerosis these patients are bound to have will potentially complicate the technical procedure. The optimum age of potential recipients should also be considered. There is evidence to suggest that patients who are over 50 years of age with a long duration of diabetes and more advanced diabetic complications have little chance of PTx being able to reverse or stabilize retinopathy, nephropathy, or neuropathy. The situation is quite different in younger patients (less than 50 years old) whose disease may be less advanced and the improvement of diabetic complications more likely [18]. Furthermore, patients who approach 50 years of age represent a high-risk group in terms of patient and graft survival [18–20]. Potential recipients should also be evaluated in regards to their emotional and sociopsychological suitability to undergo a pancreas transplantation. They ought to understand the therapeutic nature of pancreas transplantation as well as the need for long-term immunosuppression and follow-up. Patients who have had a history of non-compliance with treatment are not good candidates for PTx since recipients are required to take their immunosuppression drugs continuously and meticulously. Moreover, contraindications for PTx include untreated malignancy, active infection and HIV seropositivity, severe obesity (over 50 per cent of ideal body weight), an insufficient cardiovascular reserve, as well as age over 65 years.

Recipient groups The degree of nephropathy plays a role in determining the type of pancreas transplantation (Table 5.2). If the creatinine clearance is below 40 ml/min or 0.67 ml/s then an SPK transplant is indicated. If the creatinine clearance is above 40 ml/min or 0.67 ml/s on cyclosporin a PAK transplant is indicated, whereas in cases where the creatinine clearance is above 70 ml/min or 1.17 ml/s then the most suitable Table 5.2 Criteria for selection of candidates for SPK, PTA and PAK transplants Entry criteria for SPK transplant ● Diabetic nephropathy (creatinine clearance < 40 ml/min) ● Patient on dialysis or very close to starting dialysis ● Failure of previous renal allograft Entry criteria for PTA transplant ● Presence of two or more diabetic complications (a) proliferative retinopathy (b) early nephropathy (creatinine clearance > 70 ml/min and proteinuria > 150 mg/24 h but < 3 g/24 h) (c) presence of overt peripheral or autonomic neuropathy (d) vasculopathy with accelerated atherosclerosis ● Hyperlabile diabetes as defined by (a) frequency and severity of episodes of ketoacidosis (b) frequency and severity of episodes of hypoglycaemia (c) hypoglycaemia unawareness (d) impairment of quality of life Entry criteria for PAK transplant ● Patients with stable function of previous renal allograft that meet the criteria for PTA

61

62

INDICATIONS FOR KIDNEY AND PANCREAS TRANSPLANTATION

operation is a PTA transplant. In those patients who have an intermediate renal function (i.e. clearance between 45 and 70 ml/min or 1.17 ml/s or proteinuria above 2 g/24 h) a ciclosporin challenge test can be performed to assess the functional renal reserve to help determine the type of transplant [21].

Simultaneous pancreas and kidney transplantation This is the preferred procedure in type I diabetics with endstage or near endstage renal disease (ESRD) [22]. Patients with IDDM with impending or ESRD who have minimal or limited secondary complications of diabetes and are between the ages of 20 and 40 years are considered optimal candidates for SPK. One exception to this is the young diabetic patient for whom a suitable living related renal transplant is available. A living related renal transplant offers excellent long-term results with less immunosuppression than is required for SPK transplantation. However, not all IDDM patients with renal failure are acceptable candidates. It has been reported that only 64 per cent of diabetic patients screened are actually accepted for SPK [23], severe cardiovascular illness has been identified as the main criteria for limiting patient selection [24]. Patients who have undergone angioplasty or coronary artery bypass can be accepted if they have adequate left ventricular function without demonstrable ischaemia. Blindness, history of major amputation, or history of cardiac disease are considered to be relative contraindications to SPK [14]. Although these diabetes-related problems are not reversible, a number of patients are well adjusted to these complications and can lead productive lives after dual organ transplantation with facilitated rehabilitation. Another factor that has to be considered is the timing of the transplantation. Pre-emptive transplantation offers the additional advantage of halting the diabetic complications before uraemia develops. Pre-emptive transplantation refers to the use of transplantation for primary renal replacement therapy before dialysis commences. The aim is to take advantage of the possible benefits of transplantation over dialysis. These include improved survival, reduced costs, and reduced morbidity [25]. Furthermore, if the increasing waiting times, the variable progressive nature of diabetic complications along with the diminished survival that IDDM patients have on dialysis are taken into consideration, it can be argued that SPK transplantation should be carefully thought for as a potential treatment for diabetic patients before dialysis.

Pancreas transplantation alone (see also Chapter 6) Ideally, solitary PTx should be performed before the development of diabetic complications such as the need for a kidney transplant. However, at present no reliable markers exist to predict, before the appearance of early lesions, which diabetic patients will develop complications. In a few American centers, including Minneapolis, Omaha, and Wisconsin, PTA has been reserved for those patients with very unstable diabetes or a hypoglycaemic unawareness that is life-threatening [17,26–27]. Indications for PTA include two or more diabetic complications including evidence of early diabetic nephropathy such as microalbuminuria, proteinuria or early histological changes but with relatively preserved renal function (creatinine clearance > 70 ml/min). Other indications may include glucose hyperlability defined as frequent episodes of hypoglycaemia without frank symptoms which lead to a significantly poor quality of life and increased risk of trauma or sudden death. Nevertheless, when non-uraemic diabetics are concerned, the morbidity and mortality associated with the longterm immunosuppression and the surgical procedure itself, must be weighed against the benefits of reversing or halting the progression of secondary endorgan diseases, reducing the risks of hypogly-

K.N. HARITOPOULOS AND NADEY S. HAKIM

caemic events, and improving quality of life. In summary, PTA is only appropriate in non-uraemic patients where the problems of diabetes itself are perceived to be more serious than the potential problems of immunosuppression. In the diabetic patients whose metabolic control is so fragile that their life is chaotic, PTA may be their only hope for a better lifestyle [28].

Pancreas after kidney transplantation In IDDM patients with a well-functioning kidney transplant, sequential PAK transplant has been advocated because these patients are already receiving chronic immunosuppression. The benefits of a subsequent PAK transplantation are improved quality of life and the fact that a functioning pancreatic allograft will likely prevent or reverse early diabetic changes in the existing kidney transplant. However, only patients with stable and adequate renal transplant function (creatinine clearance ≥ 50 ml/min) should be considered for PAK transplants. Patients with marginal function of their transplanted kidneys should instead be considered for an SPK (or no pancreas) transplant because intensified calcineurin inhibitor therapy used postoperatively can induce renal failure [17]. In Minneapolis, the preferred technique is PAK, advocating a living related renal donation first in order to correct uraemia. Then, 12 months later, a cadaver PTx follows. This sequence maximizes the availability of cadaver kidneys to other patients and improves the recipient’s well being. However, its limitation is the lower rate of pancreas graft survival. However, graft survival rates have improved and it is hoped that with future advances in immunosuppression the survival rates will be ever higher [29].

Living related pancreas transplantation Worldwide, of all pancreas transplants reported to the IPTR between 1966 and 1997 only 105 have been from live donors [30] (Table 5.3). After the kidney, the pancreas was the first solid organ to be transplanted successfully using a live donor [31]. There are certain categories of patients with IDDM who can benefit from PTx from a living donor. These include highly sensitized patients with a low probability of a negative crossmatch against a cadaveric donor, but who have a negative crossmatch

Table 5.3 Criteria for live related pancreas donation Recipients inclusion criteria ● ● ●

Highly sensitized patients with low probability of a negative crossmatch against a cadaveric donor Patients for whom high-dose immunosuppression is undesirable Patients who have been a long time on the waiting list and have had already a kidney transplant that is affected by recurrence of diabetes

Donors exclusion criteria ● History of IDDM in any first-degree relative ● History of gestational diabetes ● Donor age less than the age of diagnosis of IDDM in the recipient plus 10 years ● Body mass index > 27 kg/m2 ● Age > 45 years ● Impaired glucose tolerance or diabetes by the American National Diabetes Data Group criteria ● HbA1c level > 6% ● Glucose disposal rate < 1% during intravenous glucose tolerance tests ● Presence of elevated titre of islet cell autoantibodies

63

64

INDICATIONS FOR KIDNEY AND PANCREAS TRANSPLANTATION

against a relative. Moreover, diabetic patients who will probably not tolerate high-dose immunosuppression or patients who want to have a minimum dose of immunosuppression can benefit from a very well-matched pancreatic graft from a living related donor. In addition, patients with IDDM and a potential living donor for a kidney transplant who can gain from a successful PTx and wish to avoid a second operation for receiving a pancreatic graft from a cadaveric donor, can receive a pancreas and a kidney from the same living related donor. Furthermore, patients with brittle IDDM and diabetic recipients of a kidney transplant that has been affected by the recurrence of DM and have been on the waiting list for a long time are candidates for a living related pancreas transplant. However, living related pancreas transplantation involves certain risks for the donor as well as the anticipated risks for the recipient. Partial pancreatectomy is by no means an easy operation and a careful selection of potential donors has to be made. Individuals must be excluded from consideration if they meet any of the following criteria: (a) a history of type I diabetes in any first-degree relative; (b) a history of gestational diabetes; (c) age younger than 10 years greater than age of diagnosis of type I in recipient; (d) a body mass index above 27 kg/m2; (e) age older than 45 years; (f) impaired tolerance or diabetes by National Diabetes Group criteria [32]; (g) haemoglobin A1c level above 6 per cent; (h) a glucose disposal rate below 1 per cent during intravenous glucose tolerance tests; or (i) the presence of elevated titre of islet cell autoantibodies. In addition, potential donors who have any of the following criteria should be evaluated by endocrinologists before a decision is made: (a) a glucose value greater than 150 mg/dl; (b) basal, fasting insulin values greater than 20 ␮U/ml; (c) an acute insulin response to glucose or arginine above 300 per cent basal insulin; (d) clinical evidence of insulin resistance; and (e) evidence of more than one autoimmune endocrine disorder [30]. All potential donors have to be advised that none of the current available tests can absolutely exclude the possibility of a later development of diabetes following hemipancreatectomy. Hence, long-term follow-up of the metabolic function of donors is essential. Until 1994 in Minneapolis, living related pancreas transplantation was reserved only for patients without uraemia (PTA) or for diabetic patients who had previously received a kidney transplant (PAK) [33]. The first successful living kidney and pancreas donor transplantation was performed in Minneapolis in 1994 and ever since more than 20 such transplants have been carried out [30,34]. The obvious advantage of a live related SPK transplant is that the recipient does not have to go through two separate surgical procedures. In conclusion, living donor SPK transplants can be done safely and successfully in selected uraemic patients with type I DM.

Conclusion Ideally, pancreatic transplantation should be performed earlier in the course of the disease, before the appearance of secondary complications. As the results of patients and graft survival have considerably improved during the last 5 years, the indications for PTA have become the main subject of debate in pancreatic transplantation. Both the advantages and disadvantages of PTx and insulin therapy have to be carefully considered for specific populations of diabetic patients as well as for each patient individually. In non-uraemic diabetic patients the choice of PTx has to be balanced against the risks of surgery and long-term immunosuppression on one hand and the complications of DM and insulin therapy on the other. At present, candidates for PTx can be selected among the various categories of diabetics. In patients with labile diabetes who have developed secondary complications, or those with frequent hypoglycaemic episodes, PTx with the accompanying immunosuppression is an acceptable alternative. In the future, PTx could be used prophylactically to prevent secondary complications. This could be based

K.N. HARITOPOULOS AND NADEY S. HAKIM

on the study of genetic susceptibility to complications. Currently, there are no such genetic markers although microalbuminuria has been demonstrated as an early marker of nephropathy in some cases, and elevated Na+/Li+ countertransport has been implicated as a potential marker in predicting the development of nephropathy [35]. Furthermore, candidates for PTA could be selected in the future from the prepubertal diabetic population. It is known that the earlier the onset of diabetes, the greater the risk of complications. PTx can be considered when the patients are mature enough to participate in the choice of treatment.

References 1 Harris M, Hadden WC, Knowles WC, Bennett PH. Prevalence of diabetes and impaired glucose tolerance and plasma glucose levels in the US population aged 20–74 years. Diabetes 1987;36:523–34. 2 Libman I, Songer T, Lapore R. How many people in the US have IDDM? Diabetes Care 1993;16:841–2. 3 Nathan DM. Long-term complications of diabetes mellitus. N Engl J Med 1993;328:1676–84. 4 Clark CM, Lee DA. Prevention and treatment of the complications of diabetes mellitus. N Engl J Med 1995;332:1210–17. 5 The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;29:977–86. 6 Robertson RP. Pancreatic and islet transplantation for diabetes: cures or curiosities? N Engl J Med 1992;327:861–8. 7 Sutherland DER. Pancreas transplants. Br J Surg 1994;81:2–4. 8 Gruessner AC, Sutherland DER. Analyses of pancreas transplant outcomes for United States cases reported to the United Network for Organ Sharing (UNOS) and non-US cases reported to the International Pancreas Transplant Registry (IPTR). In: Cecka JM, Terasaki PI, ed. Clinical transplants 1999. Los Angeles: UCLA Tissue Typing Laboratory, 2000:51–69. 9 International Pancreas Transplant Registry. Worldwide pancreas transplants. IPTR Newsletter 1997;9:1–12. 10 Sutherland DER, Gruessner A, Moudry-Munns K. International pancreas transplant registry report. Transplant Proc 1994;26:407–11. 11 White SA, Nicholson ML, London JM. Vascularized pancreas allotransplantation — clinical indications and outcome. Diabet Med 1999;16:533–43. 12 Sasaki TM, Gray RS, Ratner RE, Currier C, Aquino A, Barhyte DY, et al. Successful long-term kidneypancreas transplants in diabetic patients with high C-peptide levels. Transplantation 1998;65:1510–12. 13 Brayman KL, Weber M. Sutherland DER. Pancreatic and islet transplantation. In: Trede M, Carter DC, eds. Surgery of the pancreas. London: Churchill Livingstone, 1997:637–65. 14 Stratta RJ, Taylor RJ, Lowell JA, Sindhi R, Sudan D, Castaldo P, et al. Clinical transplants 1994. Los Angeles: UCLA Tissue Typing Laboratory, 1994:265–81. 15 Stratta RJ, Weide LG, Sindhi R, Sudan D, Jerius JT, Larsen JR, et al. Solitary pancreas transplantation. Diabetes Care 1997;20:362–8. 16 Dubernard JM, Tajra LCF, Lefrancois N, Dawahra M, Martin C, Thivolet C, et al. Pancreas transplantation: results and indications. Diabetes Metab 1998;24:195–9. 17 Odorico JS, Leverson GE, Becker YT, Pirsch JD, Knechtle SJ, D’Alessandro AM, et al. Pancreas transplantation at the University of Wisconsin. In: Cecka JM, Terasaki PI, ed. Clinical transplants 1999. Los Angeles: UCLA Tissue Typing Laboratory, 2000:199–210. 18 Allen R. Pancreas transplantation. In: Forsythe JLR, ed. A companion to specialist surgical practice — Transplantation surgery. London: Saunders, 1998:167–202.

65

66

INDICATIONS FOR KIDNEY AND PANCREAS TRANSPLANTATION

19 Stratta RJ. Mortality after vascularised pancreas transplantation. Surgery 1998;124:823–30. 20 Freise CE, Stock PG, Meizer JS. Increased morbidity and mortality of simultaneous pancreas and renal transplantation in patients over 49 years of age. Transplant Proc 1998;30:292. 21 Brennan DC, Stratta RJ, Lowell JA, Miller SA, Taylor RJ. Cyclosporine challenge in the decision of combined kidney–pancreas versus solitary pancreas transplantation. Transplantation 1994;57:1606–11. 22 Shapiro R, Jordan ML, Scantlebury, Vivas CA, Jain A, Chakrabarti P, et al. Simultaneous pancreas–kidney transplantation at the University of Pittsburgh. In: Cecka JM, Terasaki PI, ed. Clinical transplants 1999. Los Angeles; UCLA Tissue Typing Laboratory, 2000:217–21. 23 Stratta RJ, Taylor RJ, Wahl TO, Duckworth WC, Gallagher TF, Knight TF, et al. Recipient selection and evaluation for vascularized pancreas transplantation. Transplantation 1993;55:1090–6. 24 Elkhammas EA, Demirag A, Henry ML. Simultaneous pancreas–kidney transplantation at the Ohio State University Medical Center. In: Cecka JM, Terasaki PI, ed. Clinical transplants 1999. Los Angeles: UCLA Tissue Typing Laboratory, 2000:211–15. 25 Stratta RJ, Taylor RJ, Ozaki CF, Stevenson Bynon J, Miller S, Knight TF, et al. A comparative analysis of results and morbidity in type I diabetes undergoing preemptive versus postdialysis combined pancreas–kidney transplantation. Transplantation 1993;55:1097–103. 26 Gruessner RW, Sutherland DER, Najarian JS, Dunn DL, Gruessner AC. Solitary pancreas transplantation for non-uraemic patients with labile insulin-dependent diabetes mellitus. Transplantation 1997;64:1572–77. 27 Stratta RJ, Taylor RJ, Bynon JS, Lowell JA, Sindh R, Wahi TO, et al. Surgical treatment of diabetes mellitus with pancreas transplantation. Ann Surg 1994;220:809–17. 28 Remuzzi G, Ruggenenti P, Mauer SM. Pancreas and kidney/pancreas transplants: experimental medicine or real improvement. Lancet 1994;343:27–31. 29 Sutherland DER, Gruessner RWG, Najarian JS, Gruessner AC. Solitary pancreas transplants: a new era. Transplant Proc 1998;20:280–1. 30 Gruessner RWG, Kendall DM, Drangstveit MB, Gruessner AC, Sutherland DER. Simultaneous pancreas–kidney transplantation from live donors. Ann Surg 1997;226:471–82. 31 Sutherland DER, Goetz FC, Najarian JS. Living-related donor segmental pancreatectomy for transplantation. Transplant Proc 1980;12:19–25. 32 National Diabetes Data Group. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 1979;28:1039–57. 33 Sutherland DER, Gruessner RWG, Dunn D, Moudry-Munns KC, Gruessner A, Najarian JS. Pancreas transplants from living-related donors. Transplant Proc 1994;26:443–5. 34 Gruessner RWG, Sutherland DER. Simultaneous kidney and segmental pancreas transplant from living related donors — the first two successful cases. Transplantation 1996;61:1265–8. 35 Dubernard JM, Martin C, Lefrancois N, Thivolet CH, Dawahra M, Petruzzo P, et al. Advances in pancreas transplantation — 1999 Indications and Contraindications. J Pediatr Endocrinol Metab 1999;12:765–70.

Chapter 6

Indications for solitary pancreas transplantation Robert J. Stratta

Introduction After a decade of controversy surrounding the therapeutic validity of pancreas transplantation (PTX), the procedure has become accepted as the preferred treatment for patients with insulindependent diabetes mellitus (IDDM) and advanced diabetic nephropathy. Vascularized PTX is currently the only available form of autoregulating total endocrine replacement therapy that reliably achieves an insulin-independent euglycaemic state and normal glucose homeostasis resulting in the successful management of diabetes mellitus. Whether PTX represents a treatment or cure for IDDM is debatable. The trade-offs for normal glucose homeostasis are the operative risks of the transplant procedure and the need for chronic immunosuppression. Free islet grafts have the same potential but do not approach PTX in terms of consistency of results. With improvements in organ retrieval and preservation technology, refinements in diagnostic methodology and surgical techniques, advances in clinical immunosuppression and anti-infective prophylaxis, and experience in donor and recipient selection, success rates for PTX have continued to improve. From 1966 through to July 2000, over 14 000 PTXs were performed worldwide and reported to the International Pancreas Transplant Registry (IPTR) [1]. In the last decade, the majority (83 per cent) of PTXs have been performed in combination with a kidney transplant [simultaneous kidney and pancreas transplant (SPK)] in patients with endstage diabetic nephropathy. The current 1-year actuarial patient, kidney, and pancreas (with complete insulin independence) graft survival rates after SPK are 95, 92, and 84 per cent respectively [1,2]. Solitary PTXs comprise the remaining activity, including either sequential pancreas after kidney transplants (PAK, 12 per cent) or PTX alone (PTA, 5 per cent).The current 1-year patient survival rate after solitary PTX is 95 per cent, and the 1-year actuarial pancreas graft survival rates are 73 per cent for PAK and 70 per cent for PTA [1,2]. The differences in graft survival rates between SPK, PAK, and PTA transplants have been attributed to an increased rate of graft loss due to rejection and thrombosis after solitary PTX. PTX should be considered an acceptable therapeutic alternative to continued insulin therapy in diabetic patients with imminent or established endstage renal disease (ESRD) who have had or plan to have a kidney transplant, because the successful addition of a pancreas does not jeopardize patient survival, may improve kidney graft survival, and will restore normoglycaemia [3]. In the diabetic patient with a well-functioning kidney transplant, sequential PAK is advocated because these patients are already obligated to chronic immunosuppression and the additional risk is primarily that of the surgical procedure [4]. The proportion and total number of PAKs performed annually has steadily increased in recent years due to a number of factors: (a) the

68

INDICATIONS FOR SOLITARY PANCREAS TRANSPLANTATION

results have continued to improve with advances in immunosuppression (the current 1-year pancreas graft survival rate for PAK patients receiving antibody induction in combination with tacrolimus and mycophenolate mofetil therapy is 83 per cent); (b) living donor kidney transplantation followed by PAK expands the kidney donor pool and permits pre-emptive transplantation as a scheduled event; (c) waiting times for PAK are much shorter than for SPK; (d) increasing the number of solitary PTXs may improve donor utilization and enhance sharing; (e) PAK may be better tolerated with less morbidity compared to SPK (for instance, the patient may be in better medical condition at the time of PAK-not on dialysis); and (f) the results of the kidney may be better when it is not transplanted simultaneous with a pancreas due to less surgical and immunological morbidity [4,5]. Although many transplant centers are experienced in SPK, only a few centers have accumulated experience in PAK. For a patient with diabetes but no uraemia, the choices are between exogenous insulin with the burden of diabetic management and the risk of acute and chronic diabetic complications versus the surgical risk of PTX and medical risks of chronic immunosuppression [6,7]. The Diabetes Control and Complications Trial (DCCT) has clearly shown that improved glycaemic control lowers the risk of secondary diabetic complications [8]. However, intensive insulin therapy did not result in normalization of glycosylated haemoglobin levels, was associated with a three-fold increased risk of severe hypoglycaemia, and was more resource-intensive. The results of the DCCT provide a strong rationale for PTA transplantation provided that the burden of immunosuppression is more benign than the burden of diabetes over the long term. Because of the uncertainty about the relative risk of each treatment option, PTA transplantation is restricted by necessity to those patients who already have demonstrated a propensity to early diabetic complications that are (or predictably will be) worse than the potential undesirable side-effects of chronic immunosuppression [3]. For this reason, most PTA transplants are done in patients with long-term diabetes who have specific problems, such as metabolic lability, neuropathy, or hypoglycaemic unawareness. In selected cases, PTA transplantation may be considered as a treatment option to prevent, stabilize, or reverse secondary complications of diabetes [9]. However, in the absence of glucose hyperlability, establishment of insulin independence alone would not justify the need for chronic immunosuppression if an effect on secondary diabetic complications could not be achieved.

Donor selection Donor selection and organ procurement are of paramount importance to the success of solitary PTX. Most heart-beating donors who have been declared brain dead and are appropriate for kidney, liver, and heart donation are also suitable for pancreas donation (Table 6.1) [10,11]. Although there is some evidence to suggest that donor hyperglycaemia may have an adverse effect on initial and long-term allograft function, the presence of hyperglycaemia or hyperamylasaemia, as such, are not contraindications for pancreas donation. In general, ideal pancreas donors range in age from 10 to 40 years and range in weight from 30 to 80 kg. Management of the multiple organ donor includes aggressive resuscitation to maintain haemodynamic stability, organ perfusion, and oxygenation. Resuscitative efforts usually result in significant hyperglycaemia and intensive control with insulin may have a favourable effect on initial allograft function and survival. Intravenous colloid fluids and mannitol are given to minimize pancreatic edema. Judicious administration of vasopressors such as dopamine is indicated to maintain a systolic blood pressure above 90 mmHg and promote diuresis.

R.J. STRATTA

Table 6.1 Cadaveric pancreas organ donation Indications Declaration of brain death Informed consent Age 6–55 years (ideal 10–40) Weight 30–100 kg (ideal 30–80) Haemodynamic stability with adequate perfusion and oxygenation Normal glycosylated haemoglobin level (only needed in case of severe hyperglycaemia, extreme obesity, or positive family history of diabetes) Absence of infectious or transmissible diseases (i.e. tuberculosis, syphilis, hepatitis, AIDS) Negative serology (HIV; hepatitis B and C) Absence of malignancy (unless skin or low-grade brain cancer) Absence of parenchymal/intrinsic pancreatic disease Contraindications History of diabetes mellitus (type 1 or 2, or gestational) Previous pancreatic surgery Moderate to severe pancreatic trauma Pancreatitis (active acute or chronic) Significant intra-abdominal contamination Major (active) infection Chronic alcohol abuse Recent history of intravenous drug abuse Recent history of homosexuality or high-risk sexual behaviour Prolonged hypotension or hypoxaemia with evidence for significant endorgan (kidney, liver) damage Severe atherosclerosis Inexperienced retrieval team Severe fatty infiltration of pancreatic parenchyma Severe pancreatic oedema Severe obesity (> 150% ideal body weight or BMI > 30 kg/m2) Risk factors Massive transfusions Prior splenectomy Mild to moderate obesity (< 150% ideal body weight, BMI > 27.5 kg/m2) Aberrant hepatic artery anatomy Positive VDRL/RPR serology Prolonged length of hospital stay Donor age above 45 years Cardiovascular or cerebrovascular cause of brain death Mild to moderate fatty infiltration Mild to moderate pancreatic oedema Donor instability Mild pancreatic trauma Mild to moderate atherosclerosis 1 AIDS, acquired immunodeficiency syndrome; 2 BMI, body mass index; 3 HIV, human immunodeficiency virus; 4 RPR, rapid plasma reagin; 5 VDRL, venereal disease research laboratory.

69

70

INDICATIONS FOR SOLITARY PANCREAS TRANSPLANTATION

As the results of PTX have improved and experience has increased, previous contraindications for pancreas donation have become mere risk factors for a successful outcome (Table 6.1). According to IPTR data, the following variables are associated with an increased risk of pancreas allograft thrombosis: (a) donor age above 40 years; (b) cardiovascular or cerebrovascular cause of brain death; and (c) pancreas preservation time above 24 h [1,2]. The results of anecdotal experience suggest that: (a) over 150 per cent ideal body weight or body mass index (BMI) over 30 kg/m2 may be associated with an increased risk of early pancreas graft loss (due to thrombosis, pancreatitis, infection, or primary non-function); (b) a donor liver biopsy with more than 25 to 30 per cent macrovesicular steatosis may be associated with a fatty pancreas leading to an increased risk of early graft loss; and (c) fatty infiltration of the pancreas (as opposed to peripancreatic fat) may be associated with an increased risk of early graft loss [12,13]. Notably, pancreas grafts from female donors over 45 years of age appear to fare better than pancreas grafts from male donors over 45 years independent of age, cause of brain death, or body habitus. The presence of donor obesity or a fatty pancreas may be an underappreciated cause of early graft loss following PTX. However, some overweight donors may have little or no fatty infiltration of the pancreas, while some thin donors may have significant fatty infiltration of the pancreas, so the correlation is not absolute. It is also important to distinguish fatty infiltration of the pancreatic parenchyma from peripancreatic fat deposition. The latter is not uncommon, but is not necessarily associated with an adverse outcome. The most reliable way to make this distinction is to mobilize completely the spleen, body, and tail of the pancreas up into the operative field so that the anterior and posterior aspects of the gland can be carefully visualized and palpated to determine the quality of the organ. There are no data currently available regarding the utility of donor pancreas biopsies, particularly with regard to steatosis. The importance of an experienced retrieval team must be emphasized for the in situ assessment of pancreatic anatomy. Pancreas donors may be categorized as ideal, good, or marginal. By using donor age, weight, and cause of brain death as the three most important factors, one can usually make a rapid and accurate assessment of the quality of the donor pancreas prior to actual intraoperative assessment, which is the second most important factor. Solitary PTX donors need to be either ideal or good. If either the donor or recipient are marginal, there is a greater likelihood of a poor outcome. Unlike SPK, the inherent risk of thrombosis is much higher (two to three times) in PAK and PTA [14]. For this reason, cold ischaemia should be kept to a minimum and serious consideration should be given to routine perioperative anticoagulation in the recipient.

Recipient selection Specific selection criteria for solitary PTX are based on the presence of early diabetic complications or hyperlabile diabetes with adequate cardiac and renal functional reserve [10,11,15]. Indications for PTX include IDDM (type 1 or 2) and the predicted abilities to tolerate the operative procedure, the requisite immunosuppression after transplantation, and possible associated complications (Table 6.2). Patient selection is aided by a comprehensive medical evaluation before transplantation (Table 6.3) performed by a multidisciplinary team that confirms the diagnosis of IDDM, determines the patient’s ability to withstand the operative procedure, establishes the absence of any exclusion criteria (Table 6.4), and documents endorgan complications for future tracking after transplantation [10,11,15]. The primary determinants for recipient selection are the presence of diabetic complications, degree of nephropathy, and cardiovascular risk (Table 6.2). With increasing experience, previous absolute have become relative contraindica-

R.J. STRATTA

Table 6.2 Indications for pancreas transplantation: eligibility guidelines Medical necessity Presence of insulin-treated diabetes mellitus (1) documentation of insulin dose (2) type 1 or 2 diabetes Ability to withstand surgery and immunosuppression (as assessed by pretransplant medical evaluation) (1) adequate cardiopulmonary function (a)cardiac stress testing ± coronary angiography to rule out significant coronary artery disease or other cardiac contraindications (b)patients with significant coronary artery disease should have it corrected pretrans plant (2) absence of other organ system failure (other than kidney) Emotional and sociopsychological suitability Presence of well-defined diabetic complications (any two) (1) proliferative retinopathy (2) nephropathy (hypertension, proteinuria, or decline in GFR) (3) symptomatic peripheral or autonomic neuropathy (4) microangiopathy (5) accelerated atherosclerosis (macroangiopathy) (6) glucose hyperlability, insulin resistance, or hypoglycaemia unawareness with a significant impairment in quality of life Absence of any contraindications Financial resources Type of pancreas transplant Specific entry criteria based on degree of nephropathy (1) simultaneous kidney–pancreas transplant: creatinine clearance below 30 ml/min (2) sequential pancreas after kidney transplant: creatinine clearance ≥ 40 ml/min (on calcineurin inhibitor); > 55 ml/min if not on calcineurin inhibitor (3) pancreas transplant alone: creatinine clearance above 60–70 ml/min and 24-h protein excretion < 2 g Primary determinants for recipient selection are the presence of diabetic complications, degree of nephropathy, and cardiovascular risk GFR, glomerular filtration rate.

tions, and relative contraindications have become risk factors for PTX (Table 6.4). Binocular blindness or a history of a major amputation are not necessarily contraindications for solitary PTX, provided that the patient is well adjusted to these otherwise irreversible diabetic complications. Inclusion and exclusion criteria for PTX are listed in Tables 6.2 and 6.4. Selection criteria for solitary PTX are based on the presence of progressive diabetic complications associated with either hypoglycaemia unawareness or exogenous insulin failure with hyperlabile diabetes resulting in a significant impairment in quality of life despite optimization of insulin therapy [16]. In the absence of indications for kidney transplantation, PTA transplantation should only be considered in patients who exhibit one or more of the following criteria: (a) a history of frequent, acute, and severe metabolic complications (hypoglycaemia, hyperglycaemia, diabetic ketoacidosis) requiring medical attention; (b) clinical and emotional problems with exogenous insulin therapy that are incapacitating; or (c) consistent failure of insulin-based management to prevent acute complications. Diabetic patients with a creatinine clearance above 60 to 70 ml/min and evidence of overt diabetic complications or unawareness of hypoglycaemic

71

72

INDICATIONS FOR SOLITARY PANCREAS TRANSPLANTATION

Table 6.3 Evaluation of the pancreas transplant candidate Interviews and consultations History and physical examination by nephrologist, endocrinologist, and transplant surgeon Ophthalmology evaluation including visual acuity, fluorescein angiography, retinal fundus photography with retinopathy score, and slit-lamp examination Transplant co-ordinator and medical social worker interview including completion or quality of life questionnaire Gynaecology consultation for all females (pelvic examination with Papanicolaou smear) Dental evaluation When indicated, additional evaluations may be required by orthopaedic surgery, podiatry, psychology, psychiatry, neurology, or gastroenterology Cardiovascular, respiratory, and peripheral vascular evaluations Standard testing includes orthostatic vital signs, 12-lead electrocardiogram, chest radiograph, echocardiography, and exercise treadmill, stress thallium, or dobutamine stress echocardiography Additional studies may include arterial blood gases, 24-h Holter monitoring, autonomic and peripheral vasomotor reflexes, Doppler arterial studies, ankle/brachial index, transcutaneous oxygen monitoring, plethysmography, carotid Doppler examination, aortography with run-off, or pulmonary function tests as indicated Cardiology consultation with or without coronary angiography as indicated Metabolic and endocrine evaluation Standard testing includes fasting blood glucose, glycohaemoglobin, and fasting lipid panel (cholesterol, triglycerides and HDL cholesterol) Fasting and stimulated C-peptide levels are used to assess type of diabetes if needed Additional studies may include oral or intravenous glucose challenge, anti-insulin and islet cell antibodies, proinsulin level, and lipoprotein profile Genitourinary/renal evaluation Standard testing includes electrolytes, blood urea nitrogen, serum creatinine, urinalysis with culture, and 24-h urine for protein and creatinine clearance Voiding cystourethrogram and urodynamics when indicated Radiometric glomerular filtration rate if needed In addition, kidney biopsy may be indicated Calcineurin inhibitor challenge test when indicated Hormonal profiles as indicated Evaluation of erectile dysfunction when indicated Serology and immunology evaluation ABO blood type and HLA tissue type Cytotoxic antibodies Viral titres (Epstein–Barr virus, herpes simplex virus, varicella-zoster virus, human immunodeficiency virus, hepatitis B virus, hepatitis C virus, and cytomegalovirus); PCR quantitation when indicated VDRL/FTA test for syphilis Other laboratory tests Complete blood count with differential and platelets, prothrombin time, partial thromboplastin time, chemistry profile, amylase, lipase Abdominal ultrasound of kidneys and gallbladder Mammography in females over 35 years Haemoccult × 3; contrast studies or endoscopy when indicated When indicated, nerve conduction studies, gastric emptying scan, electromyography Hypercoagulable work-up (when indicated) FTA, fluorescent treponemal antibody; HDL, high-density lipoprotein; PCR, polymerase chain reaction; VDRL, Venereal Disease Research Laboratory

R.J. STRATTA

Table 6.4 Absolute and relative contraindications and risk factors for pancreas transplantation Absolute contraindications Insufficient cardiovascular reserve; one or more of the following (1) coronary angiographic evidence of significant non-correctable or untreatable coronary artery disease (2) recent myocardial infarction (3) ejection fraction below 30% Active infection History of malignancy diagnosed within past 3 years (excluding non-melanoma skin cancer) Positive HIV serology Positive hepatitis B surface antigen serology Active, untreated peptic ulcer disease Ongoing substance abuse (drug or alcohol) Major ongoing psychiatric illness Recent history of non-compliance Inability to provide informed consent Any systemic illness that would severely limit life expectancy or compromise recovery Significant, irreversible hepatic or pulmonary dysfunction Positive crossmatch Relative contraindications Age less than 18 or greater than 65 years Recent retinal haemorrhage Symptomatic cerebrovascular or peripheral vascular disease Absence of appropriate social support network Extreme obesity (> 150% ideal body weight or BMI > 30 kg/m2) Active smoking Severe aortoiliac vascular disease Risk factors History of myocardial infarction, congestive heart failure, previous open heart surgery, or cardiac intervention History of major amputation or peripheral bypass graft History of cerebrovascular event or carotid endarterectomy History of hypercoagulable syndrome BMI, body mass index; HIV, human immunodeficiency virus.

symptoms are potential candidates for PTA transplantation (Table 6.2). A creatinine clearance over 60 to 70 ml/min is usually required for PTA transplantation because immunosuppression can cause accelerated deterioration of renal function in those with a lower creatinine clearance before transplantation. In 233 PTA transplants performed at the University of Minnesota, 36 patients (15.4 per cent) subsequently required a kidney transplant [17]. The actuarial probability of this occurrence was 4 per cent at 1 year and 10 per cent at 5 years. This risk was present regardless of whether or not the patient exhibited long-term pancreas function. Diabetic patients that have previously received a renal allograft, whether from a living or a cadaveric donor, are considered potential candidates for PAK if the creatinine clearance is above 40 ml/min on either cyclosporin or tacrolimus immunosuppression [15]. If the patient is not on a calcineurin inhibitor for their kidney transplant, a minimum creatinine clearance of 55 ml/min is recommended (Table 6.2). Adding a calcineurin inhibitor (or increasing the dose for PAK) generally results in a 25 per cent decline in baseline creatinine clearance. It is unclear how to factor the level of proteinuria into the decision analysis, particularly since the use of a

73

74

INDICATIONS FOR SOLITARY PANCREAS TRANSPLANTATION

calcineurin inhibitor will decrease the actual level of proteinuria. However, in the presence of massive proteinuria (> 2 g/day) or nephrotic syndrome, the risk of accelerated nephropathy due to drug toxicity must be considered. In this setting, a trial of high-dose cyclosporin or tacrolimus prior to PTX (cyclosporin or tacrolimus challenge) may be indicated to ascertain the drug effect on the patient’s creatinine clearance, serum creatinine, and protein excretion [18,19]. In general, if the serum creatinine is above 2.5 mg/dl in a male or above 2.0 mg/dl in a female, then it may be ‘too late’ to proceed with a PAK and the patient may have to wait to ‘qualify’ for a pre-emptive SPK [15,20]. A baseline kidney biopsy is important to document, quantify, and monitor the progression of nephropathy after PAK. However, there is little data or experience in using the biopsy results to guide recipient selection. The presence of chronic allograft nephropathy might be a contraindication to proceeding with PAK in a patient with otherwise stable renal allograft function. The cardiac status of each candidate must be assessed carefully because significant (and silent) coronary artery disease is not uncommon in this population. The cardiac evaluation consists of a non-invasive functional assessment such as an exercise or a pharmacological stress test in addition to echocardiography. Coronary angiography is reserved for specific indications such as age over 45 years, diabetes for more than 25 years, a positive smoking history, longstanding hypertension, previous major amputation due to peripheral vascular disease, history of cerebrovascular disease, or cases in which the history, physical examination, or non-invasive cardiac studies reveal an abnormality [10,11,15]. A history of previous myocardial infarction, angioplasty, stenting, or coronary artery bypass grafting are not contraindications for PTX, as excellent outcomes have been reported in patients with previous cardiac interventions [17]. However, sudden cardiac death, in the absence of significant structural heart disease, continues to be a major cause of cardiac mortality after PTX [21]. For this reason, a number of centres are beginning to test cardiac autonomic function in these patients using laboratory evoked cardiovascular tests and 24-h heart rate variability measurements [22]. The new methodology may be able to detect alterations in autonomic function prior to the onset of disabling symptoms. In general, age over 65 years, heavy smoking, a left ventricular ejection fraction below 30 per cent, recent myocardial infarction, and severe obesity (over 150 per cent ideal body weight or BMI over 30 kg/m2) are usually viewed as contraindications for PTX (Table 6.4) [10,11,15]. Most patients below 45 years of age are acceptable candidates for PTX, provided that no significant coronary artery disease is present. Diabetic patients older than 45 years of age are not candidates until proven otherwise, and need to be undergo an extensive cardiovascular and peripheral vascular evaluation. Transplant recipients tend to age at a faster rate than non-transplant patients. For example, a diabetic patient who is 5 to 10 years postkidney transplantation is physiologically much different than a diabetic patient who is the same age and being considered for PTA transplant. A good rule of thumb is to add the number of years after kidney transplantation to the patient’s actual age to calculate physiological age (this also applies to the number of years on dialysis pretransplant). Recipient weight criteria are similar to those used for donor selection. Potential male recipients above 100 kg and female recipients above 80 kg, depending on their height and body habitus, have a higher rate of surgical complications after PTX. For this reason, a BMI over 30 kg/m2 is considered an absolute contraindication and BMI over 27.5 kg/m2 as a relative contraindication for solitary PTX. IDDM is associated with a number of coagulation abnormalities including hyperhomocysteinaemia, activated protein C resistance, increased levels of fibrinogen and von Willebrand’s factor, decreased levels of antithrombin III, and protein S deficiency [23]. In contrast to other

R.J. STRATTA

transplanted organs, the pancreas is susceptible to thrombosis because of its low microcirculatory flow based on collateral circulation. In the absence of the antiplatelet and anticoagulative effects of uraemia, solitary PTX recipients may be uniquely prone to vascular thrombosis [14]. For these reasons, a hypercoagulable work-up is indicated (Table 6.5). Ideally, solitary PTX should be performed before diabetic complications are present and before the need for a kidney transplant [24–30]. At present, there are no reliable early markers to predict, before the earliest complications appear, which diabetic individuals are at risk for progressive complications. Diabetic patients with repeated episodes of ketoacidosis, hypoglycaemia unawareness, or glucose hyperlability may derive immediate benefit from PTA with an enhanced quality of life simply by achieving an insulin-free state with improved counter-regulation. Most PTX recipients find the transition to transplantation easier than continued insulin therapy. In addition to improving quality of life and rehabilitation, there is now compelling evidence that PTX is not only acutely life-enhancing, but chronically life-saving [31]. It is hoped that the beneficial changes in carbohydrate and lipid metabolism that occur early after PTX will translate into long-term improvements in diabetic endorgan complications and decrease the risk of atherosclerotic vascular disease. In addition to correcting dysmetabolism and freeing the patient from exogenous insulin therapy, data on the course of secondary complications after PTX are emerging [9,31]. With regard to nephropathy [32], preliminary evidence suggests that successful PTA transplantation can induce regression of early but not advanced microscopic lesions of diabetic nephropathy and stabilize renal function, while successful PAK can prevent the recurrence of diabetic nephropathy in the kidney transplant. The progression of diabetic retinopathy appears to be less favourably influenced by a functioning PTX. However, with longer follow-up (more than 4 years), data are accumulating to suggest that retinopathy may be stabilized. Peripheral and autonomic neuropathies improve or stabilize in most PTX recipients, which may actually translate into a survival advantage. Improvements in nerve conduction velocity, gastric function, cardiac function, and a beneficial effect on microcirculatory blood flow have been demonstrated [9]. These effects Table 6.5 Hypercoagulation work-up If a patient has a history of any of the following conditions, then the hypercoagulation work-up should be ordered Autoimmune disease (Wegener’s, lupus, polyarthritis) Post-thrombosis (deep venous thrombosis; pulmonary embolism, stroke, myocardial infarction, recurrent dialysis access thromboses) Vasculitis Family history of thrombosis History of thrombosis with loss of allograft The hypercoagulation work-up includes the following tests Prothrombin time, partial thromboplastin time, fibrinogen Antithrombin III activity Functional protein C & S levels Lupus anticoagulant studies Factor V Leiden, factor VII, factor VIII, factor IX activity Anticardiolipin antibody assay Antiphospholipid antibody Thrombin time Homocysteine level

75

76

INDICATIONS FOR SOLITARY PANCREAS TRANSPLANTATION

may place patients at a lower overall risk for the development of peripheral ulcers or amputations. There is also evidence that a functioning PTX may ablate the hyperlipidaemic effects of immunosuppression and actually improve lipid metabolism over time. However, long-term studies are needed to document and characterize fully the effects of successful PTX on the diabetic condition. A history of compliance with medication regimens and scheduled follow-up is an important factor in patient selection. Other exclusion criteria that are applicable to all solid organ transplant recipients include the presence of active infection or recent malignancy, active substance abuse or dependence, recent history of non-compliance or psychiatric illness, and positive human immunodeficiency virus or hepatitis B virus serology (Table 6.4) [10,11,15]. Solitary PTX has assumed an increasingly important role in the treatment of IDDM and currently accounts for more than 20 per cent of PTX activity in the United States. In the future, advances in immunosuppressive strategies and diagnostic technology will only enhance the already good results achieved with solitary PTX. Further documentation of the long-term benefits and effects of PTX may lead to wider availability and acceptance, particularly from a reimbursement standpoint. Effective control of rejection with earlier diagnosis or better prevention may soon permit solitary PTX to become an accepted treatment option in diabetic patients without advanced complications. Such a policy, if applied correctly, might actually reduce the number of diabetic patients requiring kidney transplantation in the future. Other strategies for the treatment of IDDM are being actively investigated, including islet cell and fetal pancreas transplants, gene therapy, implantable insulin pumps, and biohybrid artificial pancreas units. Although any or all of these complementary methods may have a role in the treatment of IDDM in the future, it will be difficult for these alternative strategies to improve on the metabolic efficiency of the vascularized PTX that is achieved at present.

Acknowledgements We gratefully acknowledge the expertise of Joyce Lariviere in the preparation of this manuscript.

References 1 Sutherland DER, Gruessner, AC. International Pancreas Transplant Registry Update. IPTR Newsletter 2000;12:1–23. 2 Gruessner AC, Sutherland DER. Analysis of pancreas transplant outcomes for United States cases reported to the United Network for Organ Sharing (UNOS) and non-US cases reported to the International Pancreas Transplant Registry (IPTR). In: Cecka JM, Terasaki PI, ed. Clinical transplants 1999. Los Angeles: UCLA Immunogenetics Center, 2000, 51–69. 3 Robertson RP, Davis C, Larsen J, Stratta R, Sutherland DER. Pancreas and islet transplantation for patients with diabetes. Diabetes Care 2000;23:112–16. 4 Humar A, Kandaswamy R, Ramcharan T, et al. Pancreas after kidney transplants: a viable alternative to simultaneous pancreas–kidney transplants. Ann Surg 2001; (in press). 5 Humar A, Sutherland DER, Ramcharan T, et al. Optimal timing for a pancreas transplant after a successful kidney transplant. Transplantation 2000;70:1247–50. 6 Robertson RP. Pancreatic and islet transplantation for diabetes — cures or curiosities? N Engl J Med 1992;327:1861–8. 7 Stratta RJ. Vascularized pancreas transplantation: the ultimate treatment for insulin dependent diabetes. Br Med J 1996;313:703–4.

R.J. STRATTA

8 The Diabetes Control and Complications Trial (DCCT) Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulindependent diabetes mellitus. N Eng J Med 1993;329:977–86. 9 Stratta RJ. Impact of pancreas transplantation on complications of diabetes. Curr Opin Organ Transpl 1998;3:258–73. 10 Stratta RJ. Pancreas transplantation. In: Soper NA, ed. Problems in general surgery. New York: Lippincott-Williams and Wilkins, 1998;15(3):43–65. 11 Stratta RJ, Taylor RJ, Gill I. Pancreas transplantation: a managed cure approach to diabetes. In: Wells SA ed. Current problems in surgery, St Louis: Mosby, 1996;33(9):711–808. 12 Stratta RJ. Donor age, organ import, and cold ischemia: effects on early outcome after simultaneous kidney–pancreas transplantation. Transplant Proc 1997;29:3291–92. 13 Stratta RJ. Graft failure after solitary pancreas transplantation. Transpl Proc 1998;30(2):289. 14 Troppmann C, Gruessner AC, Benedetti E, et al. Vascular graft thrombosis after pancreatic transplantation: univariate and multivariate operative and nonoperative risk factor analysis. J Am Coll Surg 1996;182:285–316. 15 Stratta RJ, Taylor RJ, Wahl TO, et al. Recipient selection and evaluation for vascularized pancreas transplantation. Transplantation 1993;55:1090–6. 16 Stratta RJ. Solitary pancreas transplantation: the extreme or the ultimate. UNOS Update 1996;12:22–3. 17 Sutherland DER, Gruessner RWG, Dunn DL, et al. Lessons learned from more than 1000 pancreas transplants at a single institution. Ann Surg 2001;233:463–501. 18 Brennan DC, Stratta RJ, Lowell JA, Miller SA, Taylor RJ. Cyclosporine challenge in the decision of combined kidney–pancreas versus solitary pancreas transplantation. Transplantation 1994;57:1606–11. 19 Lane JT, Ratanasuwan T, Mack-Shipman LR, et al. Cyclosporine challenge test revisited: does it predict outcome after solitary pancreas transplantation? Clin Transplant 2001;15:28–31. 20 Stratta RJ, Taylor RJ, Ozaki CF, et al. A comparative analysis of results and morbidity in type 1 diabetics undergoing preemptive versus post-dialysis combined pancreas–kidney transplantation. Transplantation 1993;55:1097–103. 21 Hathaway DK, El-Gebely S, Cardoso S, Elmer DS, Gaber AO. Autonomic cardiac dysfunction in diabetic transplant recipients succumbing to sudden cardiac death. Transplantation 1995;59:634–7. 22 Cashion AK, Hathaway DK, Milstead EJ, Reed L, Gaber AO. Changes in patterns of 24-hour heart rate variability after kidney and kidney–pancreas transplant. Transplantation 1999;68:1846–50. 23 Kamal K, Powell RJ, Sumpio BE. The pathobiology of diabetes mellitus: implications for surgeons. J Am Coll Surg 1996;183:271–89. 24 Sutherland DER, Stratta RJ, Gruessner AC. Pancreas transplant outcome by recipient category: single pancreas versus combined kidney–pancreas. Curr Opin Organ Transpl 1998;3:231–41. 25 Stratta RJ, Weide LG, Sindhi R, et al. Solitary pancreas transplantation: experience with 62 consecutive cases. Diabetes Care 1997;20:362–8. 26 Gruessner RWG, Sutherland DER, Najarian JS, Dunn DL, Gruessner AC. Solitary pancreas transplantation for non-uremic patients with labile insulin-dependent diabetes mellitus. Transplantation 1997;64:1572–7. 27 Bartlett ST, Schweitzer EJ, Johnson LB, et al. Equivalent success of simultaneous pancreas kidney and solitary pancreas transplantation. Ann Surg 1996;224:440–52. 28 Odorico JS. Current status of isolated pancreas transplantation. Graft 1999;2:82–5. 29 Gruessner AC, Sutherland DER, Gruessner RWG. Solitary pancreas transplants: improving results and factors that influence outcome. Transplant Proc 1997;29:664–5. 30 Sutherland DER, Gruessner RWG, Najarian JS, Gruessner AC. Solitary pancreas transplants: a new era. Transplant Proc 1998;30:280–1.

77

78

INDICATIONS FOR SOLITARY PANCREAS TRANSPLANTATION

31 Stratta RJ, Pancreas transplantation: long-term aspects and effect on quality of life. In: Hakim NS, Stratta RJ, Gray DWR, eds. Pancreas and islet transplantation. Oxford: Oxford University Press, 2002; 89–94. 32 Fioretto P, Steffes WM, Sutherland DER, Goetz FC, Mauer M. Reversal of lesions of diabetic nephropathy after pancreas transplantation. N Eng J Med 1998;339:69–75.

Chapter 7

Donor management and selection for pancreas transplantation Eldo Ermenegildo Frezza and Robert J. Corry

Introduction Pancreas transplantation is a well-established treatment for patients with insulin-dependent diabetes. Over the last decade, the procedure has been refined, and immunosuppressive regimens have improved such that, at major centres, 1-year graft survival routinely exceeds 80 per cent [1]. Recent evidence supports the beneficial effect of pancreas transplantation on the maintenance of normoglycaemia, and the arrest and possible reversal of diabetic complications such as vasculopathy, neuropathy, and nephropathy [2–4]. With improved outcome, the number of patients receiving pancreas transplants has risen steadily. Nearly 1000 transplants were performed in the United States in 1996, as compared to approximately 300 in 1988 [1]. Despite this, the waiting time for a pancreas continues to increase as more candidates are accepted for transplantation, particularly after Medicare’s recent recognition of the procedure. Utilization of pancreas grafts varies from 0 to more than 70 per cent among different procurement regions [5]. The shortage of appropriate organs for transplantation is further compounded by the discarding of relatively normal pancreases, or their lack of retrieval. In an effort to expand our pancreas donor pool, we have liberalized the criteria defining a suitable pancreas donor. There has been a reluctance to use organs from haemodynamically unstable or nonheart-beating donors. With demand for organs far exceeding the supply of optimal donors, we have instituted an aggressive policy that does not exclude organs based on donor age or haemodynamic status. Indeed, had we excluded pancreas from unstable donors or donors over 45 years of age, the number of patients receiving a transplant at our institution would have been halved. With routine surveillance using percutaneous needle biopsy of the transplanted pancreas, some centres are now reporting equivalent survival results among recipients of simultaneous pancreas and kidney (SPK) transplantation and pancreas transplant after kidney (PAK) [6]. We have shown that the use of marginal donor [7] has not adversely affected overall pancreas graft survival at our institution. In fact, it has allowed us to expand our donor pool and virtually double the number of pancreas transplants performed. The most important determinant of pancreas suitability is inspection of the organ by an experienced pancreas transplant surgeon. We avoid transplanting organs with calcifications, extensive fibrosis, or fatty infiltration. Also of importance is determination of the adequacy of the back-table flush, as judged by the quality of the effluent draining from the portal vein. Brisk drainage of solution installed under the force of gravity, free of blood (this sometimes requires placement of a clamp across the splenic hilum), with no evidence of blood within the organ are signs of a good flush. Although these criteria are somewhat subjective, they are superior to rejecting organs, sight unseen, based only on demographics and laboratory values. Therefore, we do

80

DONOR MANAGEMENT AND SELECTION

not restrict use of pancreatic grafts simply based on donor age or haemodynamic status. Adoption of these strategies may allow for a reduction in the number of organs discarded and expand the availability of pancreas grafts for transplantation.

Donor selection All donors without a history of diabetes should be considered eligible. Donor hyperglycaemia or hyperamylasaemia are not considered contraindications. Selection of pancreas donor grafts for transplantation is based principally on: (a) careful inspection of pancreatic parenchyma for evidence of fibrosis and extent of fatty infiltration; (b) examination of the donor vessels for presence and severity of atherosclerotic plaque; and (c) the quality of the back-table venous efflux of chilled flush solution. An immediate return of chilled solution flushed into the organ by force of gravity is considered adequate. Organs with extensive fatty infiltration, calcifications, or fibrosis are not harvested. Pancreas with major traumatic injuries are rejected, although minor injuries, such as capsular tears, are not considered contraindications. Table 7.1 outlines the donor risk factors analysed including donor age, sex, cause of death (gunshot wound, closed head injury, cerebrovascular accident, anoxia, or other miscellaneous causes), vasopressor requirement at the time of harvest, amylase, lipase, and glucose. Table 7.1 and 7.2 summarize the donor and recipient factors that, in our experience, must be considered in assessing the outcome of pancreas transplantation.

Table 7.1 Donor factors analysed to assess pancreas transplant outcome in Pittsburgh experience Risk factor* Age (mean: 29, range: 6–62 years) ≤ 45 years > 45 years Sex male female not recorded Cause of death gunshot wound closed head injury cerebrovascular accident anoxia other not reported Vasopressors None or low dose High dose Laboratory values amylase (mean 136, range 6–1411) lipase (mean 116, range 9–1373) glucose (mean 285, range 100–648) * Including

three non-heart-beating donors.

Frequency (%) 83.9 16.1 46.0 48.2 5.8 24.8 17.5 40.9 8.0 2.9 5.8 62.8 37.2

E. ERMENEGILDO FREZZA AND R.J. CORRY

Table 7.2 Recipient factors analysed to assess pancreas transplant outcome Risk factor

Frequency (%)

Age (mean: 40, range: 27–58 years) Sex Male Female

46.7 53.3

Operation

SPK PAK PTA

85.4 8.8 5.8

Drainage

Bladder Enteric

21.2 78.8

Complications

Thrombosis Donor-related Donor-unrelated (n = 14) or none (n = 86)

9.5 17.5 73.0

Cold ischaemia time, hr (mean: 17.4, range: 7–31) Laboratory values (48 h peak) Amylase (IU/dl) (mean: 480, range: ≤ 2000 > 400 Lipase (IU/dl) (mean: 2804, range: 79–24,445) ≤ 2000 > 2000 Glucose (mg/dl) (mean: 220, range: 58–826) < 300 > 300

46–2039) 56.9 42.3 58.4 40.9 78.1 21.9

Marginal donors Pancreas donors are considered marginal if they were retrieved from donors over 45 years of age or from donors haemodynamically unstable at the time of harvest. Donors were considered to be haemodynamically unstable if they required high-dose dopamine (> 10 µ/kg/min) or at least two vasopressors at the time of harvest. By these criteria more than 69 per cent of patients with grafts from marginal donors were transplanted in our experience [7]. The oldest graft transplanted was from a 62-year-old donor. Haemodynamically unstable donors represented 32 per cent of the grafts. Only in 4 per cent of case was a pancreas utilized from a donor who was both over age 45 and on highdose vasopressors.

Assessment of the pancreas allograft The ultimate decision depends on the judgement of the transplant surgeon, who is in the best position to assess the suitability of the pancreas based on the recipient’s clinical condition and the complete donor profile provided by the field co-ordinator. Indeed, in many instances the teams can only assess the suitability of the organ by direct visualization or intraoperative biopsy at the time of procurement. The pancreas is inspected for colour and consistency and presence or absence of masses, nodularity, or trauma. On palpation, the normal pancreas is soft and smooth, with no hard nodules and little or no fat. Pale appearance suggests ischaemia, whereas a yellowish colour may reflect other underlying microvesicular or macrovesicular steatosis (fatty infiltration). Fatty infiltration is often a consequence

81

82

DONOR MANAGEMENT AND SELECTION

of donor obesity, or alcoholism. A large percentage of macrosteatosis is associated with poor graft function. Fatty pancreas should therefore be carefully examined by an experienced surgeon at the time of harvesting to determine the extent and distribution of fat.

The cadaveric donor Cadaveric organ donation begins with the determination of brain death in a given patient and the identification of such an individual as a potential donor. The success of organ recovery depends on a team effort among health-care professionals and personnel from the local organ procurement organization (OPO). The demand for organ transplantation in the United States increased by 16 per cent per year [8]. Unfortunately, the number of organs donated annually has not changed much. Following the enactment of the Organ Transplant Act in 1984, the creation of a single national organization with regional offices, the United Network for Organ Sharing (UNOS), assumed the responsibility for obtaining and distributing donor organs to transplant centres. Several retrospective studies analysed the organ donation patterns in different states, being the reason that each state has a different pattern of urban and rural areas. For this reason, it is not always feasible to compare the result of a Kentucky study [9] with a study performed in Pennsylvania [10] or New Jersey [11]. In fact, the first has more non-urban population than the other two states. Moreover, there are factors that are characteristic for a rural or an urban population in a given region or state. There is a continued need for public education and an examination of the factors that contribute to an insufficient number of donors. These factors range from general distrust of the health-care system to religious myths and misperceptions regarding consent to donation. Such fears are especially common among some minority groups. One study identified three possible causes of the loss of a potential donor: (a) failure to identify a potential donor; (b) failure to ask the family to donate; and (c) family refusal to donate [12]. Investigators in New Jersey attempted to ascertain the reason for a recent increase in organ donations. Several factors seemed to correlate with an increase in the rate of conversion from a potential donor to an active donor. These were: (a) a better application of the paediatric donation New Jersey Bill [13]; (b) the acceptance of brain dead abused children; (c) a door-to-door campaign performed by UNOS in the years 1992 and 1993; and (d) an increase in donation in the Afro-American, Hispanic, and Asian, groups that historically did not donate as much [11]. Interestingly enough, the non-urban hospital area had a better donation than the large urban areas, particularly if there was a level 2 trauma centres. Surprisingly, the presence of a transplant centre in the same areas, did not significantly increase the rate of donation [11]. In general, the diagnosis of brain death precedes any consideration for organ donation. This determination is made independently of the transplantation team. The criteria for brain death are listed in Table 7.3.

Table 7.3 Criteria for establishing brain death [16] The presence of deep coma without induced hypothermia or the use of depressant drugs No spontaneous respiratory motion in the absence of muscle relaxants No spontaneous movements in the absence of decerebration and decorticate posturing No response to deep painful stimuli Absent cranial reflexes, i.e. fixed and dilated pupils, no corneal reflexes, no oculocephalic reflex (doll’s eyes)

E. ERMENEGILDO FREZZA AND R.J. CORRY

Table 7.4 Contraindications to cadaveric donation [16] Absolute Chronic renal disease Malignant tumour (except primary brain tumours) Positive HbsAg test Positive HIV test Untreated systemic bacterial, fungal, or viral infections Prolonged cold ischaemia

Relative Hypertension Diabetes Bacterial sepsis (treated) Anti-HCV positive Age > 70 years Age < 2 months Prolonged warm ischaemia

ATN, acute tubular necrosis; HbsAg, hepatitis B surface antigen; HCV, hepatitis C virus; HIV, human immunodeficiency virus.

Once a person with irreversible brain injury has been referred, determining the potential donor’s suitability is best co-ordinated by the OPO in consultation with the transplant surgeons. When relative contraindications exist, the donor is sometimes described as marginal (Table 7.4). For example, some centres prefer not to use organs from very young donors because of a higher risk of technical complications and reports of decreased graft survival. However, such organs can be successfully transplanted into selected recipients [for example, first transplant, low body weight, low panel reactive antibody (PRA) level], and should not be routinely excluded from the donor pool. Several centres have reported good results with paediatric kidneys in adult recipients [14,15]. Similarly, organs from donor older than 60 or even 70 years can often be used; however, older donors may have other medical problems (for example, hypertension, cardiac disease) that may make their organs less suitable [16]. The use of organs from hepatitis C virus (HCV) antibody-positive donors has been controversial. A consensus seems to have been reached to transplant these organs into HCV-positive recipients (with appropriate informed consent) in view of the lack of associated morbidity [17–19]. In situations in which a patient has profound, irreversible, neurological injury and strict criteria for brain death cannot be strictly met, some donors can serve as non-heart-beating donors. In these situations, consent is obtained, and the patient is extubated in the operating room; once the heart has ceased beating, the organs are rapidly removed. There are other circumstances in which organs can be recovered from non-heart-beating donors when sudden circulatory arrest takes place in an uncontrolled situation [20,21].

Management of the donor Once brain death has been declared in a potential organ donor candidate, maintenance of organ perfusion becomes the paramount issue in management. Haemodynamic stability and cardiac function must be maintained, as well as physiological electrolyte balance and pH, adequate urine output, glucose level, adequate arterial oxygenation, temperature regulation, and control of infection. Once consent for donation has been obtained from the patient’s family, the organ procurement field co-ordinator undertakes a careful work-up of the candidate. A careful history is obtained with special focus on previous medical history, social behaviour, and other risk factors for transmissible disease. Knowledge of the mechanism of brain death and duration of cardiopulmonary arrest prior to resuscitation is crucial. The pancreas is not among the organs most commonly injured in trauma. Therefore,

83

84

DONOR MANAGEMENT AND SELECTION

sometimes traumatic injuries are not discovered in time. We suggest, based on this, that the knowledge of peritoneal lavage and or abdominal CT results is important in the work-up of the traumatically injured donor. It is important to know the presence of any pancreatitis in the past or recent history of alcohol use. Minimum laboratory evaluation must include ABO blood type, glucose level, amylase, lipase, liver function tests, electrolytes, complete blood cell count, prothrombin time, partial thromboplastin time, viral serologies for hepatitis, cytomegalovirus, and HIV. Additional laboratory screening is directed by the donor’s medical and social history. Complete cultures should be performed on tissues from donors who have been admitted to hospital for prolonged periods or who have known foci of infection. The above information, together with a summary of haemodynamics and donor vital statistics (age, height, and weight), is used to generate a donor profile to assess suitability of the patient for organ donation. As experience with ‘marginal’ donors has increased [7], it has become apparent that donors should not be summarily rejected on the basis of age, laboratory, or haemodynamic thresholds. The use of the super-rapid in situ cold flush technique has allowed successful use of both livers and kidneys in situations in which cardiac arrest has occurred and there is the capability for immediate retrieval of the organs, but not yet the pancreas. Aggressive intravenous crystalloid infusion is necessary to keep urine output greater than 1 ml/kg/h. Excessive urine losses due to diabetes insipidus may warrant the use of vasopressin, although it may sometimes be desirable instead to replace the high urinary output intravenously. Severe hypernatraemia and hypokalaemia, which can occur in donors with diabetes insipidus, may be avoided by frequent electrolyte monitoring. Persistent untreated hypokalaemia may result in ventricular tachycardia and cardiac arrest. In addition, it is necessary to avoid infections in potential donors [22]. The rapid infusion of large volumes of cold fluids, such as crystalloid solutions, blood, or blood products, can aggravate pre-existing hypothermia and create pancreatic oedema, which is considered a relative contraindication for harvesting. All efforts must be directed at maintaining haemodynamic stability by continuous replacement of blood loss and urine output with crystalloid. Use of adrenergic agonists should be avoided if possible, to minimize peripheral vasoconstriction and reduced renal and hepatic perfusion. Muscle relaxants are usually required to optimize surgical exposure, because spinal reflexes sometimes prevent abdominal relaxation. Full intravenous heparinization (300 U/kg) is performed before cross clamping of the aorta. The need for an anaesthesiologist continues until the time of aortic cross-clamping and in situ hypothermic perfusion [16]. With the development of multiple organ recovery techniques the time to procure each solid organ should be minimized to decrease any unnecessary ischaemic injury. We suggest retrieval of liver and pancreas en bloc to expedite the harvest and avoid the possibility of instability in the donor. The in situ division of the pancreas could create long operating time and is not superior to a well-flushed pancreas divided on the back-table. Brain-dead cadavers are often hyperglycaemic owing to the administration of steroids and intravenous infusion of large amounts of dextrose-containing solutions. Such donors also often exhibit a resistance to insulin and require high doses to restore normoglycaemia. Nevertheless, grafts taken from hyperglycaemic cadaver donors have functioned perfectly in recipients. Hyperamylasaemia is also often present in brain-dead cadavers and is usually not associated with any apparent pancreatic injury. Hyperamylasaemia is a known consequence of isolated brain trauma. Normal function in the recipient without the occurrence of pancreatitis in grafts taken from hyperamylasaemic donors has been documented. Thus the only contraindication to the use of a brain-dead

E. ERMENEGILDO FREZZA AND R.J. CORRY

cadaver as a pancreas donor is a history of diabetes, intra-abdominal trauma with bacterial contamination, direct injury to the pancreas, or abnormalities of the pancreas on gross inspection. Whole or segmental pancreas grafts can be obtained from virtually every cadaver donor, regardless of what other organs are also procured. One should never forgo obtaining a liver from a donor simply to use the pancreas; removal of both organs from the same donor is technically feasible [16].

Complications related to mismanagement of the donor It is very important to be careful during the preparation of the donor and the harvesting of the pancreas. Two common complications that may be related to mismanagement of the donor are graft thrombosis and pancreatitis. Thrombosis and subsequent ischaemic necrosis of the graft are complications that have contributed to graft failure in a number of patients. The high incidence of this problem has been explained by the fact that the pancreas is a so-called ‘low-flow’ organ. Haemodynamic instability prior to the harvesting can be detrimental to the pancreas. Removal of the spleen in combination with postclamping oedema, which, to a varying degree, always follows ex vivo organ preservation, may be sufficient to facilitate the occurrence of thrombosis. Therefore the most important preventive factors are: (a) continued monitoring of the potential donor, with control of any hypotensive episode; and (b) meticulous surgical technique. Graft thrombosis usually presents in the initial 12 to 24 h after transplantation and is signalled by a sudden elevation in the serum glucose level or by failure of the glucose level to fall towards normal. The diagnosis can be confirmed by a radionuclide perfusion scan to delineate flow to the pancreas. If the diagnosis is confirmed, the graft must be removed expeditiously to prevent septic or vascular complications. Pancreatitis may occur early as well as late after pancreatic transplantation. Early postoperative pancreatitis is usually a consequence of: (a) haemodynamic instability prior to the harvesting; and (b) injury to the pancreas during procurement and preservation. After transplantation, there is inevitably a rise in the serum amylase level, which usually peaks after 24 to 48 h. Hyperamylasaemia usually is not associated with abdominal pain and clinical signs of pancreatitis. Treatment of pancreatitis in this setting is the same as for any other form of acute pancreatitis. Pancreatitis in the early post-transplantation period is usually self-limited. Should pancreatitis progress, however, thrombosis may ensue, necessitating removal of the graft, since the inflamed pancreas may become the focus of a septic process, resulting in pancreatic abscess, anastomotic failure, peripancreatic abscess, peritonitis, and pancreatic fistula.

Related pancreas donor The same rationale for use of related donor for kidney transplants also applies to use of related donors for segmental pancreas transplants (Table 7.5). A portion of the body and tail of the pancreas can be removed from a living donor, based on a vascular pedicle of the splenic vessels. The spleen of the donor can survive on collateral circulation, and the remainder of the body, the head, and the uncinate process of the pancreas is sufficient to maintain normoglycaemia in the donor. Criteria that prospective living-related donors must meet before being evaluated are listed in Table 7.5. Prospective donors undergo oral and intravenous glucose tolerance tests. All glucose values during standard oral glucose tolerance tests (OGTT) must be within an arbitrarily defined normal range (fasting < 105 mg/dl, 60-min value 185 mg/dl, 90-min value: 160 mg/dl, and 120-min value: 140 mg/dl) [22,23].

85

86

DONOR MANAGEMENT AND SELECTION

Table 7.5 Criteria for selection of living related pancreas donors [22] Pre-evaluation Recipient and donor discordant for diabetes for at least 10 years Donor at least 10 years older than age of onset of diabetes in recipient In cases of sibling donation, no family members other than the proband are diabetic Post evaluation Normal oral glucose tolerance test (OGTT)result by criteria of Fajans and Conn and of the Natural Diabetes Data Study Group Delta insulin > 90 IU/ml for sum of 0-, 60-, 120- and 180-min values during cortisonestimulated OGTT minus sum during standard OGTT according to technique of Fajans and Conn No islet cell antibodies Other metabolic parameters normal

Conclusions Simultaneous pancreas–kidney transplantation is now considered an excellent therapeutic option for patients with renal failure secondary to type I diabetes mellitus. Since the first pancreas transplant was performed in 1966 by Kelly and Lillehei, the surgical technique, immunosuppressive therapy, and antibiotic have continually evolved. Patient and graft survival rates after pancreas transplantation (PTX) have approached those of other solid organ transplants. According to the last Pancreas Registry, patient and graft survival rates [24] at 1 year for patient, kidney, and pancreas were 94, 90, and 83 per cent respectively for SPK; 95 and 71 per cent for pancreas transplant alone, and 95 and 71 per cent for PAK transplant. Enteric drainage is becoming more common, increasing from 2 per cent of cases in 1988 to 12 per cent in 1995, to 33 per cent of cases in 1998 [24,25]. The high incidence of morbidity after PTX is possibly related to multiple factors: underlying diabetes mellitus and uraemia (in SKP recipients), heavy immunosuppression (because the pancreas appears to be a highly immunogenic organ), low blood flow based on no collateral circulation, unique complications related to exocrine secretions, and the transplantation procedure involving two hollow viscera (donor duodenum and recipient urinary bladder or small bowel) [26]. Judicious management of the donor could therefore play an important role in the success of the transplant and prevention of early postoperative complications.

References 1 Gruessner A, Sutherland DER. Pancreas transplantation in the United States (US) and in non US as reported to the United Network for Organ Sharing (UNOS) and the international pancreas transplant registry. In: Terasaki PI, Cecka JM, ed. Clinical transplant 1994. Los Angeles: UCLA Tissue Typing Laboratory, 1996;47–9. 2 Gfesser M, Nusser J, Muller-Felber W, Abendroth D, Land W, Landgraf R. Cross-sectional study of peripheral microcirculation in diabetic patients with microangiopathy: Influence of pancreatic and kidney transplantation. Acta Diabetol 1993;30:79. 3 Muller-Felber W, Landgraf R, Scheuer R. et al. Diabetic neuropathy 3 years after successful pancreas and kidney transplantation. Diabetes 1993;42:1482. 4 Wilczek HE, Jaremko G, Tyden G, Groth CG. Evolution of diabetic nephropathy in kidney grafts. Transplantation 1995;59:51.

E. ERMENEGILDO FREZZA AND R.J. CORRY

5 Stratta RJ, Bennett L. Pancreas underutilization in the United States: analysis of United Network for Organ Sharing data. Transplant Proc 1997;29:3309. 6 Bartlett ST, Schweitzer EJ, Johnson LB, et al. Equivalent success of simultaneous pancreas kidney and solitary pancreas transplantation. Ann Surg 1996;224:440. 7 Kapur S, Bonham CA, Dodson FS, Dvorchik I, Corry RJ. Strategies to expand the donor pool for pancreas transplantation. Transplantation 1999;67:284–90. 8 Davidson MN, Devney P. Attitudinal barriers to organ donation among black Americans. Transplant Proc 1991;23:2531–2. 9 Garrison RN, Bentley FR, Rauqe GH, et al. There is an answer to the shortage of organ donors. Surg Gynecol Obstet 1991;173:391–6. 10 Nathan HM, Jarrell BE, Broznik B, et al. Estimation and characterization of the potential renal organ donor pool in Pennsylvania. Transplantation 1991;51:142–9. 11 Frezza EE, Krefski LRN, Valenziano CP. Factors influencing the potential organ donation: a 6-yr experience of the New Jersey Organ and Tissue Sharing Network. Clin Transplant 1999;13:231–40. 12 Gortmaker SL., Beasley CL, Brigham LE, et al. Organ donor potential and performance Size and nature of the organ donor shortfall. Crit Care Med 1996;24:432–9. 13 New Jersey Bill 1-1277; supplementing P.L. 1967, c. 234. 14 Ratner LE, Flye MW. Successful transplantation of cadaveric en-bloc paired kidneys into adult recipients. Transplantation 1991;51:273. 15 Darris F, Jordan ML, et al. Transplantation of pediatric en-bloc kidneys under FK506 immunosuppression. Transplant Proc 1991;23:3089. 16 Scantlebury VP. Cadaveric and living donation. In: Renal transplantation, Shapiro R, Simmons RL, Starzl TE, ed. Appleton & Lange, 1997. 17 Brunson ME, Lau JY, et al. Non-A, non-B hepatitis and elevated serum aminotransferases in renal transplant patients. Transplantation 1993;56:1364. 18 Stempel GA, Lake J, et al. Hepatitis C: its prevalence in end-stage renal failure patients and clinical course after transplantation. Transplantation 1993;55:273–6. 19 Morales JM, Campistol JM, et al. Transplantation of kidneys from donors with hepatitis C antibody into recipients with pre-transplantation anti-HCV. Kidney Int 1995;47:236–40. 20 Anaise D, Smith R, Ishiman M, et al. An approach to organ salvage from non heart beating cadaver donors under existing legal and ethical requirements for transplantation. Transplantation 1990;49:290–4. 21 Matsuno M, Kozaki M, Sakurai E, et al. Effect of combination in situ cooling and mechanic perfusion preservation on non-heart-beating donor kidney procurement. Transplant Proc 1993;25:1516–17. 22 Sutherland DE, Moudry KC, Najarian JS. Pancreas transplantation. In: Organ transplantation and replacement, Cerilli GJ, ed. Lippincott, 1988. 23 Humar A, Gruessner RW, Sutherland DE. Living related donor pancreas and pancreas-kidney transplantation. Br Med Bull 1997;53:879–91. 24 Gruessner AC, Sutherland DE. Analysis of pancreas transplants for United States (US) and non-US pancreas transplants as reported to the International Pancreas-Transplant Registry (IPTR) and to the United Network for Organ Sharing (UNOS). In: Cecka JM, Terasaki PI, ed. Clinical transplants 1998. Los Angeles: UCLA Tissue Typing Laboratory, 1999:53–71. 25 Sutherland DE, Gruessner A. Pancreas transplant results in the UNOS United States of America Registry compared with non-USA data in the international registry. In Terasaki P, Cecka M, ed. Clinical transplants 1995 Los Angeles: UCLA Tissue Typing Laboratory;1994:47. 26 Reddy KS, Stratta RJ, Shokouh Amiri MH, Alloway R, Egidi MF, et al. Surgical complications after pancreas transplantation with portal-enteric drainage. J Am Coll Surg 1999;189:305–13.

87

This page intentionally left blank

Chapter 8

Procurement and benchwork preparation of the pancreatic graft Nadey S. Hakim and Vassilios E. Papalois

Donor selection The selection criteria for pancreas donors are similar to those for other solid organs [1,2]. Ideally, donor age ranges between 18 and 45 years and donor weight between 40 to 100 kg. A history of diabetes mellitus, acute necrotizing pancreatitis, and chronic pancreatitis are obvious contraindications. Despite a reported history of pancreatitis, it is always worthwhile examining the pancreas at the time of procurement to assess the suitability of the organ. It should be remembered that after brain death the serum amylase levels are often high in the absence of pancreatitis, and hyperglycaemia may occur in the absence of diabetes. The direct inspection of the pancreas is therefore justified before deciding whether or not to procure.

Multiorgan procurement of the pancreas, liver and kidneys This chapter describes the technique we are using in our institution, which, in contrast to the classical technique [3], requires that most of the dissection for donor pancreatectomy is done after crossclamping and intravascular flushing of the abdominal organs [4,5]. We believe that this technique is faster and minimizes handling of the pancreas and blood loss. This is of great importance since it has been demonstrated that pancreatic manipulation and donor instability are factors associated with post-transplant primary non-function [6] and graft pancreatitis [7]. The cadaver-donor pancreatectomy is usually part of a standard multi organ retrieval. The procurement starts with the dissection of the vascular anatomy shared by the pancreas and the liver. Arterial anomalies in the blood supply of the liver are not uncommon and may influence the decision whether or not to procure the pancreas. Generally, in such situations, priority is given to the liver because it is a live-saving procedure. The hepatic artery is isolated and dissected. It is traced down to its origin at the coeliac axis. During the dissection, the gastroduodenal artery, right gastric artery, and coronary vein are identified, ligated, and divided. After the coeliac axis is identified, the splenic artery is isolated and looped. The left gastric artery is ligated and divided. The coeliac axis is followed down to its origin. The dissection of the hepatoduodenal ligament is completed by isolating the portal vein. The origin of the superior mesenteric artery is isolated, dissected, and looped just above the left renal vein. At this point, a nasogastric tube is advanced into the duodenum. Then 400 ml of antibiotic solution (containing 750 mg cefuroxime, 50 mg amphotericin B, and 500 mg amikacin) are injected and the tube is withdrawn back into the stomach. Contrary to other pancreatic procurement techniques, this is all that is required before intravascular flushing, as the complete dissection of the pancreas takes place after cross-clamping. Heparin

90

PROCUREMENT AND BENCHWORK PREPARATION OF THE GRAFT

(70 U/kg) is given intravenously. To prepare for intravascular flushing of the abdominal organs with University of Wisconsin (UW) preservation solution, the aorta is ligated just above its bifurcation and a perfusion cannula inserted into the aorta. The cannula tip is placed between the take-offs of the ligated inferior mesenteric artery and renal arteries. Another infusion cannula is inserted into the inferior mesenteric vein for portal perfusion of the liver. A Kocher manoeuvre is performed to expose the posterior surface of the head of the pancreas as well as the underlying aorta and inferior vena cava. In co-ordination with the thoracic procurement team, the supracoeliac aorta is cross-clamped and both arterial and portal flush begun, the suprahepatic vena cava is divided supradiaphragmatically, and venous return is vented into the chest. Two litres of UW solution are perfused through the aorta and 1 litre through the portal vein. After 1 litre is delivered through the aorta, it is advisable to tighten the vessel loop around the splenic artery to prevent overflushing of the pancreas, which can lead to post-transplant pancreatitis. After flushing is complete, the pancreas and the liver are usually separated in situ. The coeliac axis remains with the liver. The splenic artery is divided just distal to its take-off from the coeliac axis. The liver is removed first, the aorta at the level of the coeliac axis, and the superior mesenteric artery is incised laterally on the left side, so that injury to the renal arteries is avoided. The Carrel aortic patch is divided between the coeliac axis and the superior mesenteric artery. The inferior vena cava is divided proximal to the entrance of the renal veins. The portal vein is divided halfway between the pancreas and the hepatic hilum, with sufficient lengths left for both liver and pancreas. The dissection of the pancreas is done trying to avoid excessive manipulation of the pancreatic parenchyma. The gastrocolic ligament divided from the pylorus to the gastrosplenic ligament, to expose the anterior surface of the pancreas through the lesser sac. The transverse colon is mobilized completely from the hepatic and splenic flexures, which allows good exposure and dissection of the pancreas from the adjacent retroperitoneal structures. All short gastric vessels are divided, separating the stomach from the spleen. The retroperitoneal attachments of the spleen as well as the splenocolic ligament are freed, resulting in complete splenic mobilization. The spleen is used as a handle to avoid injuring the pancreas. The posterior attachments of the pancreas are taken down, with the pancreas separated from the left kidney and the left adrenal gland. Retroperitoneal pancreatic dissection is performed to the level of the aorta. Here coeliac ganglions and lymphatics are divided, exposing the superior mesenteric artery and more distally the left renal vein. The GIA 60 (Gastro-Intestinal Anastomosis 60) stapler is used to divide the duodenum just distal to the pylorus. A second pass of the GIA stapler divides the fourth portion of the duodenum proximal to the ligament of the Treitz. The take-offs of the renal arteries, as well as their relationship to the superior mesenteric artery, are identified. There is no need for an aortic cuff to remain with the superior mesenteric artery. Finally at the level of the ligament of Treitz, the mesenteric root distal to the uncinate process of the pancreas is divided with the TA 90 stapler. The pancreas is removed and inspected on the back-table. The splenic artery can be marked with a 6–0 Prolene suture, making it easy to identify should it retract on the pancreatic tissue. The distal duodenal stump is opened by cutting the staple line. The duodenum is emptied of the residual bile or duodenal contents and restapled with a GIA 60. The entire common iliac artery and its bifurcation is procured en bloc and packaged with the pancreas. The pancreas is preserved in UW solution, kept cold in a box-container, and transported to the recipient hospital.

Vascular variations in combined pancreas/liver procurement If an aberrant right hepatic artery, or even the common hepatic artery, arises from the superior mesenteric, the pancreas can still be procured with the liver. The aberrant artery usually is the first

N.S. HAKIM AND V.E. PAPALOIS

branch off the superior mesenteric artery, whereas the origin of the inferior pancreaticoduodenal artery is more distal. In this case, the superior mesenteric artery is divided just distal to the aberrant right hepatic artery, the aberrant branch and its aortic cuff including the take-off of the coeliac axis and the superior mesenteric artery, remain with the liver. If the distal superior mesenteric artery, including the inferior pancreaticoduodenal artery, is preserved, an iliac Y draft can be used for arterial reconstruction of both the distal superior mesenteric artery and the splenic artery. It is necessary to preserve the inferior pancreaticoduodenal artery to provide sufficient blood supply to the head of the pancreas and the duodenum, because the gastroduodenal artery with its main branch, the superior pancreaticoduodenal artery, is usually ligated in combined pancreas/liver procurement. If the inferior pancreaticoduodenal artery is less than 3 mm in diameter and its origin is proximal to the aberrant right hepatic artery, and the liver team needs the proximal superior mesenteric artery, the pancreas procurement should be abandoned. However, if the inferior pancreaticoduodenal artery diameter is more than 3 mm, successful arterial reconstruction using a Y graft may be feasible.

Pancreas without liver procurement If the liver is not procured, dissection of the hepatoduodenal ligament is performed close to the hepatic hilum. The hepatic artery is ligated distal to the origin of the gastroduodenal artery. This provides additional blood supply to the head of the pancreas and the duodenum via the superior pancreaticoduodenal artery. The portal vein is divided proximal to its bifurcation, giving additional length. The aortic cuff encompassing the take-offs of the coeliac axis and the superior mesenteric artery is left intact and remains with the pancreas.

Benchwork preparation of the pancreas for transplantation We describe our own technique of pancreas preparation. The pancreas benchwork preparation begins by removing the spleen. The splenic hilar vessels are stapled with Ethicon vascular staplers. After the spleen is removed, attention is turned to the donor duodenum. The proximal duodenal staple line is over-run with 4–0 Prolene running suture and inverted using a 4–0 Prolene in a Lambert fashion to bury the suture line. Excess distal duodenum is excised by carefully ligating small vessels up to the level of the uncinate process. The mesenteric root which was divided with a TA 90 stapler, is reinforced with a vascular stapler. Any peripancreatic lymphatic tissue or small vessels that have not been ligated previously are identified, ligated or stapled. Unless the coeliac axis and the superior mesenteric artery are procured on a common aortic cuff, the vascular reconstruction has to be done. The splenic artery and the superior mesenteric artery are identified and prepared for reconstruction by removing surrounding lymphatic ganglionic tissue. The most common technique uses an arterial extension iliac Y graft of donor, common, external, and internal iliac artery obtained at the time of procurement (Fig. 8.1). The internal iliac artery of the extension graft is anastomosed end-to-end to the splenic artery of the pancreatic graft with running 7–0 Prolene sutures. The external iliac artery of the extension graft is anastomosed end-to-end to the superior mesenteric artery of the pancreatic graft with running 7–0 Prolene sutures. If a donor graft cannot be used and the superior mesenteric artery has an aortic patch an is of good size, an interposition graft can be used for reconstruction. Depending on the size of the splenic artery, either the donor external or internal iliac artery is anastomosed end-to-end to the splenic artery with running sutures. An arteriotomy is made on the anterior surface of the superior mesenteric artery and an end-to-side anastomosis created between the proximal end of the iliac artery and the superior mesenteric artery with 7–0 Prolene sutures. If the splenic artery is of sufficient length and can be mobilized all the way to

91

92

PROCUREMENT AND BENCHWORK PREPARATION OF THE GRAFT

(a)

(b)

(c)

N.S. HAKIM AND V.E. PAPALOIS

(d)

(e)

Fig. 8.1 Bench work.preparation for pancreas transplantation. (A) Placement of four 7–0 prolene stay suture to the superior mesenteric artery of the pancreatic graft prior to its anastomosis with the external iliac artery of the Y graft. (B) anastomosis of the superior mesenteric artery to the external iliac artery with running 7–0 Prolene suture. (C) The Y graft after the completion of the anastomosis of its two limbs to the superior mesenteric and splenic arteries of the pancreatic graft. (D) The portal vein of the pancreatic graft. (E) The pancreatic graft after the completion of the bench work preparation, ready for revascularization.

93

94

PROCUREMENT AND BENCHWORK PREPARATION OF THE GRAFT

the superior mesenteric artery without tension, a direct end-to-side anastomosis can be made between the splenic artery and the superior mesenteric artery. The bench work procedure takes in our hand on average 45 minutes with excellent results and minimal bleeding on revascularization. We believe that the meticulous procurement and benchwork preparation techniques that we described are vital as the transplant procedure can be carried out with minimal difficulties and no complications.

References 1 Papalois VE, Hakim NS. Pancreas transplantation. In: Hakim NS, ed. Introduction to organ transplantation. London: Imperial College Press, 1997:189–200. 2 Papalois VE, Hakim NS. Pancreatic transplantation. Surgery (British) 1998;16:44–8. 3 Sutherland DER, Ascher NL. Whole pancreas donation from a cadaver. In: Simmons RL, Finch ME, Ascher NL, Najarian JS, ed. Manual of vascular access, organ donation and transplantation. New York; Springer-Verlag, 1984:144–52. 4 Papalois VE, Hakim NS. Successful procurement of 50 pancreatic grafts using a simple and fast technique. Int Surg 1998;83:327–9. 5 Papalois VE, Hakim NS. Pancreas and islet transplantation. In: Hakim NS, Danovitch GM Transplantation surgery. London: Springer-Verlag, 2002:211–33. 6 Tropmann C, Gruessner A, Benedetti E, Papalois BE, Matas A, Sutherland DER et al. Delayed endocrine pancreas graft function after simultaneous pancreas–kidney transplantation: incidence, risk factors and impact on long-term outcome. Transplantation 1996;61:1323–30. 7 Benedetti E, Coady N, Asolati M, Dunn T, Stormoen B, Bartholomew A, et al. A prospective randomised clinical trial of perioperative treatment with octreotide in pancreas transplantation. Am J Surg 1998;175:14–17.

Chapter 9

Pancreas preservation H.U. Spiegel and D. Palmes

General section Introduction The evolution of organ transplantation into a clinical reality is due mainly to the development of surgical technique, the introduction of immunosuppressant drugs with better selectivity and fewer sideeffects, and the devising of more effective means of organ preservation and protection. The immense success of organ transplantation has broadened its indications to include conditions which do not threaten life, but where the gain in quality of life outweights the risks of lifelong immunosuppression, for example, pancreas transplants in patients with insulin-dependent type I diabetes mellitus (IDDM) who are not adequately controlled even by carefully adjusted insulin therapy [1–5]. The steadily rising demand for donor organs can no longer be met by the numbers of potential organ donors. New strategies will therefore have to be devised for procuring better supplies of donor organs, for example, enlargement of the donor pool by recruitment of living donors and of ‘marginal’ donors with minor impairment of organ function; adequate utilization of available donor organs must be also ensured. Organ preservation is of crucial importance for two reasons: first, because it provides the time necessary for the complex interdisciplinary co-operation; and secondly, because it enables surgeons to utilize ‘marginal’ donor organs, thereby enlarging the potential donor pool. This section presents an overview of the goals and strategies of organ preservation. The physiological and pathophysiological events occurring during preservation and the demands made on the techniques and solutions will first be outlined in general terms and their clinical relevance to pancreas preservation will then be discussed.

Goals of organ preservation The preservation of organs has two purposes. First, it must ensure the viability of the organs and rapid recovery of their functional capacity. The disorders of function arising after organ transplantation can be divided into three categories. Irreversible loss of function after transplantation is described as ‘primary non-function’ (PNF), whereas irreversible but limited impairment of function is termed ‘primary dysfunction’ (PDF) (Fig. 9.1). If the organs springs into life only after a period of latency, the term ‘delayed graft function’ (DGE) is applied. The second goal is to provide an adequate time for testing donor-recipient compatibility and for the necessary logistical preparations. A distinction is made between short-term organ preservation for up to 24 h, which is adequate for transplantation under emergency conditions. Intermediate-term organ preservation allows this period to be extended for up to 96 h, enabling the transplantation to be performed under elective conditions. Although there have been experimental studies of the successful preservation of kidney, pancreas, heart, and liver over intermediate times, in everyday clinical practice

96

PANCREAS PRESERVATION

Fig. 9.1 Primary non function: causes and aetiology.

intermediate-time preservation has been confined to renal transplants, because of the fear of the dysfunctions mentioned above. Long-term preservation by the freezing of organs aims at an entirely different goal, the creation of an organ and tissue bank which would supply the ideal transplant for the recipient at any chosen time. By extreme lowering of temperature it should be possible to arrest cell metabolism completely. However, only by storage in liquid nitrogen at –196° C cessation of all metabolic activity can be guaranteed. Whereas successful long-term preservation of individual cells and tissues such as cornea, heart valves, and pancreatic islet cells is feasible, difficulties arise in the case of larger solid organs because of delayed penetration of freezing throughout the entire organ and, similarly, problems with uneven thawing. Rapid warming leads to premature activation of metabolism in the outer layers, whereas slow warming may be followed by renewed recrystallization of areas which have already thawed. For this reason long-term preservation by freezing is not an adequate method for solid organs. Consideration will therefore be restricted to the principles and techniques of short- and intermediate-term preservation.

Strategies of organ preservation With the cutting off of the circulation and the onset of ischaemia, viability of the organs is restricted to a period of 30–60 min by lack of oxygen, substrates, and energy and by the accumulation of metabolic endproducts. To maintain viability for the necessary time and to ensure rapid resumption of function after transplantation, two strategies are available. The first strategy is to simulate the physiological milieu of the organ during the extracorporeal phase. Optimal maintenance of an explanted organ during this phase can be guaranteed by continuous perfusion with whole blood, because this provides ideal supplies of oxygen, substrates, and cofactors, coupled with satisfactory elimination of metabolic products. This obvious idea was seized up early in the development of transplant medicine and was put into practice by using heart–lung machines for continuous perfusion of organs with their own blood or by connecting the organs to an intermediate host [6–8]. However, the physiological advantages of perfusing organs with blood came up against various logistical, medical, and ethical problems which precluded its routine use in everyday practice and caused its abandonment. In the technique of bloodless perfusion the organs are perfused with solutions based on cryoprecipitated plasma, human albumin or other colloid materials, so as to circumvent the problems arising from the use of blood. Efforts were also made to maintain aerobic energy metabolism by oxygenation and enrichment with substrates such as glucose, fatty acids, amino acids, phosphates, and adensine.

H.U. SPIEGEL AND D. PALMES

The mode of perfusion that comes closest to the physiological situation is continuous or intermittent machine perfusion. This comprises a closed system within which 800–1200 ml of solution is circulated through a heat exchanger and a membrane oxygenator and pumped through the organ at a perfusion pressure of around 30 mmHg. Metabolic endproducts can be removed by regularly changing the perfusion fluid. In everyday practice the use of continuous or intermittent organ perfusion is restricted by the relatively high technical and logistical effort required. Furthermore, it has not proved successful in organs with a pronounced tendency to oedema, such as the pancreas. Lastly, when using perfusion or nutrient solutions based on plasma or human albumin, the possibility of transmitting vital diseases must not be forgotten. The second strategy for organ preservation is to slow down the metabolic processes in the ischaemic or anoxic organ by hypothermia or by drugs, so as to prevent cell death from energy deficit and to maintain the viability of the organ until it has been implanted and reperfusion has been started. The most important item is hypothermia. The use of hypothermia for organ preservation is based on the fact that energy-consuming and autolytic metabolic processes, like all enzymatic reactions, attain their maximum reaction velocity only within a narrow temperature range. If the temperature drops below that range, the reactions will slow down. In human beings and most other warm-blooded animals, cooling by 10° C will slow metabolic activity by a factor of 1.5 to 2.0. This correlation is expressed by the van’t Hoff coefficient Q10: Q10 = (k2/k1)10/(t2–t1, where k1 and k2 denote the reaction velocities at temperatures t1 and t2. This indicates a slowing of reaction velocity by a factor of 12 when the temperature of the organ is reduced to 0° C. From this correlation it is also clear that hypothermia does not completely suppress metabolism, but can merely prolong the interval before the onset of cell death. Yet this is not the only constraint which hypothermia involves. Organs which have been stored in simple Ringer lactate solution, even for a short time show marked cell swelling and massive loss of intracellular potassium and magnesium. To understand the background behind these reactions we must first visualize the physiological situation in a normal cell and the consequences of cold ischaemia. Under physiological conditions the cell is surrounded by a milieu containing high concentrations of sodium and chloride and a low concentration of potassium ions. Furthermore, the extracellular concentration of calcium ions is roughly 1000 times greater than it is in the intracellular space. There is an electrical potential across the cell membrane, with its negative pole in the cell and positive charges on the exterior. Numerous energy-consuming processes serve to maintain this status. The necessary substrate is usually adenosine triphosphate (ATP), which is synthesized in the mitochondria by a process which consumes energy. It in turn supplies energy for the membrane-based Na/K-ATPase, which working against the concentration gradients, pumps three sodium ions out of the cell and two potassium ions into the cell. Among other important proteins is Na/Ca-ATPase, which is responsible for transport of calcium out of the cell, and the voltage-dependent calcium channels. Under conditions of ischaemia and anoxia anaerobic glycolysis comes into action to provide energy by the breakdown of glucose to lactate. Even though this metabolic pathway is considerably less efficient than aerobic glycolysis via the citrate cycle, it is nevertheless adequate to maintain the integrity of the cell in the system described above. However, though organs can tolerate ischaemia for a limited time, after this time has ended further synthesis of ATP is impossible and metabolism comes to a standstill. This leads to uncompensated influx of sodium ions and loss of intracellular potassium.

97

98

PANCREAS PRESERVATION

Collapse of the membrane potential is followed by massive influx of chloride and calcium. The membrane proteins undergo denaturation and water streams into the dying cell. Hypothermia cannot prevent these processes but can slow them down to a great extent. By the reduction in metabolic rate, the cell is enabled, by using anaerobic glycolysis, to eke out its limited energy reserves for a considerably longer time. The activity of Na/K-ATPase is certainly reduced, but because it is less temperature-sensitive than other enzymes, its residual activity allows it at least partially to maintain the membrane potential. Nevertheless, there is a relative preponderance of ion flows along the concentration gradients, so that in addition to the losses of intracellular potassium and magnesium there is an influx of sodium, carrying with it an influx of water which leads to intracellular oedema. The conditions of hypothermic ischaemia have many effects on intracellular calcium balance. Owing to accumulation of lactate, anaerobic glycolysis leads to the development of intracellular acidosis and consequently to a shift from protein-bound calcium to free Ca2+ ions. The cellular energy deficit leads to the escape of calcium from compartments with high concentrations, such as the mitochondria, into the intracellular space. Loss of the membrane potential results in an increased influx of calcium via the voltage-dependent calcium channels. At the same time the efflux of calcium via the energy-dependent Na/Ca-channels is considerably reduced. The marked increases in intracellular calcium concentration activate phospholipases A1, A2, and C, which catalyse hydrolysis of the phospholipids of the cell membranes. This reaction is detrimental to the transplant in two respects: first, it destabilizes the cell membrane with further accentuation of the electrolyte imbalance; and secondly, these processes cause the liberation of arachidonic acids. The latter are then broken down to eicosanoids, and because of their pronounced chemotactic activity these induce an increased reaction on the part of the non-specific immune system [9]. Ischaemia leads to loss of energy-rich compounds, in particular ATP, and to accumulation of breakdown products in the form of hypoxanthine. This substance is normally broken down to urate by xanthine dehydrogenase. However, under conditions of ischaemia, owing to the deficit of reduced NAD as acceptor, this pathway is diverted via xanthine oxidase. During reoxygenization, the breakdown of hypoxanthine leads to sudden liberation of oxygen radicals. Under ischaemic conditions, therefore, hypothermia can prevent rapid cell death, but there will nevertheless be an intracellular energy deficit, accompanied by electrolyte shifts with intracellular oedema and abnormalities of membrane potential. The damage that occurs during reperfusion is triggered by the influx of calcium, the activation of arachidonic acid metabolism and other biochemical changes.

Principles of organ preservation In the early days of organ transplantation various preservative solutions were developed with the object of mitigating the consequences of ischaemia and hypothermia, and of minimizing reperfusion damage. The ingredients of the solutions used to maintain the physiological milieu and to act upon the pathophysiology of the ischaemia–reperfusion lesion are described in detail below.

Electrolytes The electrolytes added to preservative solutions serve two purposes. First, they are intended to stabilize the cell membrane; and secondly, to prevent the emergence of an osmotic gradient between the extracellular and intracellular compartments during hypothermic storage. Solutions with an electrolyte composition resembling that of intracellular fluid, for example University of Wisconsin (UW) or Euro Collins (EC), have high concentrations of potassium and phosphate and low concentrations

H.U. SPIEGEL AND D. PALMES

of sodium and chloride. Because of the absence of concentration gradients between the intracellular and extracellular spaces, they are claimed to avoid the efflux of potassium out of the cell, and hence both to stabilize the cell membrane and to prevent the emergence of an osmotic gradient. Experimental and clinical studies have shown, however, that the electrolyte concentration of preservation solutions is of relatively little importance for their oncotic and cell-protective action. Comparison of the potassium-rich UW and EC solutions with Bretschneider’s histidine–tryptophan– ketoglutarate(HTK) solution, which has an ionic composition almost exactly the same as that of extracellular fluid, showed no differences with regard to the prevention of electrical charge exchanges or of cellular oedema. Furthermore, a high potassium concentration in the extracellular space, for example in UW solution, has proved to be deleterious, because it constricts the blood vessels and damages vascular endothelium. Conversely, addition of magnesium, for example in Marshall’s solution, has proved valuable. Because of the possibility of precipitation with phosphates the addition of calcium to preservation solutions should be avoided.

Impermeants Because the sodium–potassium pump has stopped working, the cell membrane potential drops, and sodium and chloride can therefore stream into the cell without hindrance. The resulting osmotic gradient leads to influx of water, culminating in cellular oedema. The development of cellular oedema can be avoided by adding certain substances to the preservative solutions. These substances can diffuse into the interstitial spaces, but they cannot permeate through the cell wall or at most can penetrate only very slowly. These substances are called impermeants. Large molecules which are not required for energy metabolism or other physiological processes within the cell have proved the most effective substances for this purpose. In the version of UW solution intended for use in heart–lung machines, gluconate (MW 196) has been used with great success, and the same is true of lactobionate (MW 358) added to the UW storage solution. Other polysaccharides and disaccharides such as raffinose (MW 594) in UW solution or mannitol (MW 182) in Marshall’s solution have been used with equal success, whereas the employment of metabolizable monosaccharides such as glucose (MW 180) in the EC solution or the use of colloids alone have proved ineffective. Other substances, which diffuse into the cell much more slowly than simple electrolytes, for example histidine (MW 155) or tryptophan (MW 204) contained in Bretschneider’s HTK solution, have given satisfactory results. Similar effects are produced by the use of citrate in Marshall’s solution or of phosphate in the EC solution [10].

Colloids During the ‘flush out’, that is, the rinsing out of the blood in the transplant at the beginning of perfusion, interstitial oedema frequently arises during continuous machine perfusion or during reperfusion. This oedema compresses the capillaries and hence hinders complete rinsing out of the blood during the ‘flush out’, and causes patchy distribution of the preservative solution and incomplete capillary reperfusion. Attempts have been made to compensate for the transcapillary fluid loss by using colloids, that is, substances of high molecular weight [for example, hydroxyethyl starch (HES) in UW solution or polyethylene glycol in EC solution], which do not escape from the blood vessels and thereby create a colloid-osmotic pressure gradient against the interstitial tissues [11–13]. However, in clinical and experimental practice it soon became evident that the colloid additive, though one of the basic prerequisites for long-term machine perfusion, is not necessary for simple hypothermic storage. Because one of the effects of colloid additives is to increase the viscosity of the

99

100

PANCREAS PRESERVATION

perfusion solution, one disadvantage, especially during liver perfusion, was inadequate washing out of the larger and smaller bile ducts whose blood supply comes from capillaries alone [14].

Buffers During tissue ischaemia, anaerobic glycolysis and glycogenolysis (Pasteur effect) lead to the accumulation of acid valencies such as lactate. This causes cell damage, affecting the lysosomes and mitochondria in particular. The intracellular acidosis, by its effect on various enzymes such as phosphorlipase A2, also modifies the reactions occurring during reperfusion. There are three measures which can compensate for these effects. By avoiding the use of glucose as an initial substrate for anaerobic glycolysis it was hoped to prevent the development of acidosis from the outset. Furthermore, by alkalinization, as in UW solution, and by the addition of buffer substances such as histidine in HTK solution, any lowering of pH due to tissue ischaemia can be neutralized.

Energy substrates Despite the reduction in cell metabolic activity during hypothermic storage, there is still some consumption of energy, and because of the anaerobic conditions and the lack of substrates, it cannot be completely made good. This leads to ATP breakdown with the emergence of endproducts such as adenosine, inosine, and hypoxanthine, which pass freely through the cell membrane and are hence lost from the cell. When reperfusion begins these substances necessary for ATP synthesis are lacking just when they are needed for regeneration of the sodium–potassium pump and for other energy-consuming functions. Attempts have been made to counteract this pathophysiological mechanism by adding various substances to preservative solutions. Energy-rich precursors such as adenosine in UW solution, adenosine, and ribosine in heart–lung machine UW solution, and phosphate in UW and EC solutions, are added so that they can serve as substrates for ATP synthesis. Another approach is to add allopurinol, as in UW solution, so as to inhibit xanthine oxidase and hence block the breakdown of adenosine nucleotides. However, the value of ATP precursors is still uncertain. Some experiments have shown more rapid ATP regeneration owing to the increased availability of these substances. Conversely, it is known that the absence of these energy precursors enhances ischaemia tolerance [15,16]. Oxygen persufflation is yet another measure which can be used to regenerate cellular energy metabolism, because aerobic glycolysis within the cell provides more energy than is given by anaerobic glycolysis. For this purpose oxygen is introduced into the organ by continuous bubbling via its blood vessels (‘persufflation’), and the oxygen is then allowed to escape by puncture channels made on the surface of the organ with a fine needle. Care must be taken that the organ remains immersed in the preservative solution and does not float up to the surface because of the gas which it contains. Although oxygen persufflation has been shown to slow down ATP breakdown in renal transplants in experimental animals, this procedure has not found favour in clinical practice, first, because of serious technical difficulties which dealing with larger organs; and secondly, because damage to vascular endothelium by reactive oxygen metabolites cannot be excluded [17,18]. During continuous machine perfusion synthetic perfluorochemicals have been used as erythrocyte substitutes with the aim of improving oxygen transport. However, for simple hypothermic storage, these additives have proved useful only in the case of pancreas preservation [19–21].

Oxygen radical scavengers During ischaemia ATP is metabolized to hypoxanthine and xanthine, while xanthine dehydrogenase is converted to xanthine oxidase. The start of reperfusion is followed by oxydation of hypoxanthine and

H.U. SPIEGEL AND D. PALMES

xanthine by xanthine oxidase, and the release of oxygen radicals. These damage the cell membrane and lead to chemotaxis and activation of neutrophil granulocytes. Preservative solutions contain ingredients which provide adequate protection from oxygen radicals by three mechanisms. UW solution contains allopurinol which inhibits xanthine oxidase, thus preventing the formation of oxygen radicals. By adding oxygen scavengers, such as glutathione in UW solution or histidine in HTK solution, free oxygen radicals can be captured or reduced in numbers. Lastly, by adding vasoactive substances [prostaglandins, eicosanoids, nitric oxide] efforts have been made to ensure optimum vascular diameters during reperfusion. This will allow an adequate influx of oxygen and nutrients, and will regulate the influx of oxygen so as to keep the formation of oxygen radicals within tolerable bounds [9].

Summary The cell damage caused by ischaemia can be minimized by hypothermia, which will in principle damp down all the energy-dependent reactions of the cell. In addition, the composition of the extracellular fluid to which the ischaemic organ is to be exposed can be adapted to the altered conditions by adding appropriate substances to the preservation solution (Table 9.1) The most important measure is to correct the serious osmotic imbalance associated with hypothermia. Simple lowering of extracellular sodium concentration to the levels existing within the cell, together with reduction of extracellular calcium concentration on the levels in the cytosol, has proved to be inadequate. To prevent cellular oedema it is necessary to add osmotically active substances, which should be so far as possible metabolically inactive. These substances should not diffuse into the cell, or only to a limited extent, and should also take over protective functions (for example, oxygen radical scavengers, buffer substances). Moreover, by buffering the extracellular space, ischaemic damage can be still further reduced, because low pH values during hypothermia lead to a decrease in the activity of certain enzymes, for example phosphofructokinase (Fig. 9.2).

Special section Introduction Diabetes mellitus is the principal cause of chronic renal impairment and blindness in adults and leads to amputations and impotence in more cases than any other disease. Type I IDDM is the most frequent chronic disease of children [22–25]. Diabetes mellitus is not merely a disorder of intermediary metabolism, but also causes specific lesions in blood vessels and the nervous system such as retinopathy, nephropathy, or neuropathy. In the last two decades it has become clear that these ‘microangiopathies’ are mainly due to hyperglycaemia. Acute metabolic decompensation can be averted by exogenous insulin therapy and properly timed injections can often ward off many of the complications of diabetes mellitus. However, even in well-stabilized patients, insulin injections do not always achieve control of intermediary metabolism as good as that provided by physiological endogenous insulin secretion, which reacts to the smallest shifts in glucose concentration. Transplantation of the pancreas is the only mode of treatment for IDDM which offers reliable independence from insulin injections and which can normalize glycated haemoglobin levels. Pancreas transplants, besides improving the patient’s quality of life, also aim to correct the secondary complications of IDDM by ensuring long-term euglycaemia [22–25]. The crucial factor is the transplantation of the cells of the islets of Langerhans; this can be done in three ways [22,23]. The simplest way would

101

102

Table 9.1 Strategies of organ preservation Hypothermia (delays catabolism, which leads to tissue destruction) Organ preservation Physical cell millieu ● osmotically active substances (suppress cell swelling) ● electrolytes (prevent the formation of an osmotic gradient between extracellular and intracellular spaces and stabilize the cell membrane) ● buffer substances (extracellular and intracellular pH neutralization) ● colloids (facilitate the initial flushout; during continuous machine perfusion) Biochemical cell millieu ● metabolites for regenerating energy production ● inhibition of the breakdown of important structural proteins (protease inhibitors) Minimization of reperfusion damage ● reduction of oxygen radical formation (inhibition of xanthine oxidase) ● scavenging of oxygen radicals already formed ● use of vasoactive substances to ensure optimal vascular diameter during reperfusion

H.U. SPIEGEL AND D. PALMES

Fig.9.2 Organ preservation: pathophysiology and therapeutic principles. AMP, adenosine monophosphate; ATP = adenosine triphophose, adenosine triphosphate.

be to implant isolated islet cells in the hope that they would colonize the recipient. This mode of therapy is still in a very early clinical stage and demands methods of preparing and processing the pancreas that are completely different from classical organ preservation. [26]. The remainder of this section will be restricted to the preservation of the entire pancreas. A second possibility is to transplant the tail of the pancreas, which contains the majority of the roughly 1 million islet cells. The third possibility is to transplant the entire pancreas. This is normally the method of choice and is usually combined with renal transplantation (simultaneous pancreas and kidney transplantation). There are certain exceptions, such as pancreatic transplantation after the performance of a kidney transplant, usually following a living donation, and the rare cases of isolated pancreatic transplantation in nonuraemic but metabolically unstable patients. These latter forms of pancreatic transplantation differ mainly in that they are performed as ‘bench surgery’ and are summarized elsewhere (Chapter 8).

Pancreas procurement The aim of any organ transplantation is optimal transplant function, and this can be achieved by adequate donor management, the best possible organ protection, and by minimizing organ damage during the implantation. Damage to a transplant may lead to disorders of functions, sometimes amounting to primary failure of the organ (PNF). When there is a shortage of donor organs this problem may be aggravated by the use of transplants with minor pre-existing damage (‘marginal donors’). Because of its filamentous structure and the absence of any capsule, the pancreas has proved highly susceptible to stresses during harvesting and preservation (Table 9.2) [27]. In principle, damage to the donor organ can be divided into five categories as follows.

Pre-preservation injury Even before explantation, potential donor organs can suffer damage from the donor’s previous illness or fatal illness, during hypotensive or ischaemic phases and even in the course of intensive care. Absolute contraindications prohibiting the use of a donor for a pancreas transplant include diabetes

103

104

PANCREAS PRESERVATION

Table 9.2 Checklist for pancreas preservation Perfusion via

Aorta Superior mesenteric vein, splenic vein Omit intravascular flush

Perfusion pressure and volume

100 cm H2O transarterial, maximum 1000 ml perfusion solution (beware of overperfusion)

Packaging

First bag containing pancreas and preservatives solution without ice (do not forget vessel interponate), organ dish filled with fluid and ice, second sterile bag around the dish

Preservative solution and ischaemia tolerance

EC solution obsolete UW solution is the gold standard, safe limit 12 h cold ischaemia time, cold ischaemia time can be prolonged to up to 96 h by adding prostanoid inhibitors, NO synthetase inhibitors, PAF antagonists and by the two-layer storage technique HTK solution: equally good results at cold ischaemia times of up to 24 h [46]

Special features

Marked tendency to oedema because of absence of capsule (beware of overperfusion, continuous perfusion obsolete, good results from omission of intravascular flush) Preservation solution must also be suitable for multiorgan harvesting and must not compromise any subsequent islet cell isolation Prolongation of cold ischaemia time in connection with pancreas preservation is worthwhile only in connection with renal ischaemia tolerance

NO = nitric oxide; PAF, platelet-activating factor.

mellitus, or a history of acute or chronic pancreatitis. Because of the connection between nicotine abuse and ␤-cell function, even though it is not yet fully clear, a history of smoking is a relative contraindication to the use of such a donor [28].

Harvesting injury Damage may occur during explantation from inexpert dissection of the organ, or from inadequate perfusion and organ cooling, possibly leading to warm ischaemia as well. Because it has no capsule, the pancreas is particularly susceptible to damage from manipulation, ischaemia, or changes in its milieu. If portal perfusion is in progress at the same time, hyperperfusion (> 1000 ml perfusion solution) or the generation of retrograde portal overpressure in the pancreas must be avoided [23,24]. To divert exocrine pancreatic secretion, a segment of duodenum has to be resected and sterilized, but fortunately it will tolerate ischaemia longer than the pancreas [27].

Cold preservation injury The duration of cold ischaemia is critical for transplant function. For example, the incidence of primary organ failure and the retransplantation rate rise significantly with increasing cold ischaemia time.

H.U. SPIEGEL AND D. PALMES

Rewarming injury Prolonged warm ischaemia time or inadequate cooling of the transplant during implantation will lead to additional damage.

Reperfusion injury In addition to cold ischaemia time, reperfusion damage plays a crucial part in the inception of primary organ failure. Cardiac problems, which may arise, for example, during liver transplantation because of the escape of preservation solutions with high potassium concentrations into the systemic circulation, can be avoided by the use of washout solutions before reperfusion. [13]

Pancreas preservation Euro Collins solution In the late 1960s Collins inaugurated the era of hypothermic organ preservation by introducing a hyperosmolar solution having an ionic composition almost identical with intracellular fluid, but without colloid additives. With this ‘Collins solution’ he succeeded under experimental conditions in prolonging hypothermic kidney preservation to over 48 h. The first Collins solutions (C2, C3, C4) had high concentrations of potassium, magnesium sulfate, and glucose. Procaine was added to the C3 and C4 solutions to inhibit sodium influx into the cell and phenoxybenzamine was used to induce vasodilatation. As experimental and clinical experience grew, various modifications were made to C3 and C4, resulting in solutions, which was fundamentally similar to C2 (Table 9.3). Phenoxybenzamine and procaine were no longer used as additives. In the event of circulatory stasis, tissue cholinesterase could metabolize procaine to para-aminobenzoic acid, which is nephrotoxic. Phenoxybenzamine rapidly precipitates out of solution. In addition the magnesium additive was withdrawn, because precipitates of magnesium phosphate complexes had often been observed in the early Collins solution. To increase the osmolarity of EC solution, the glucose concentration was raised from 120 to 180 mmol/l. However, the use of glucose as an osmotically active substance proved less than optimal for two reasons. First, long-term preservation of organs with increased glucose permeability, such as liver and pancreas, presents problems, because with the gradual diffusion of glucose into the intracellular space the osmolarity of the solution diminishes and it cannot be relied upon to prevent cell oedema [29–31]. This means that when EC solution is used to preserve the pancreas it gives a maximum ischaemia time of only 4 h before serious cell oedema develops, one of the causes being the intracellular diffusion of glucose [32,33]. This results in the ‘no flow phenomenon’, a condition in which capillary blood flow through the transplant is obstructed by cell oedema. Secondly, glucose is the initial substrate for the anaerobic glycolysis which takes place during hypothermia, and if glucose is available in large amounts it may induce tissue acidosis. Numerous experimental studies have shown that the preservation of organs, in particular of the pancreas, can be optimized by replacement of glucose by other, metabolically inert sugars, such as sucrose or mannitol, and by the addition of HES to raise osmolarity [34]. In 1976 EC solution was recommended as the preservative solution of choice by the Preservation Working Committee, which is made up of members from the transplant centres associated with Eurotransplant. Nowadays, for the reasons stated above it is considered inferior to other preservative solutions such as UW or HTK solution, and is no longer employed for clinical pancreas conservation [35].

University of Wisconsin solution UW solution was developed in 1987 by Belzer and Southard for preservation of the pancreas [33,36,37]. The aims of UW solutions are to minimize cell swelling during hypothermic ischaemia,

105

106

PANCREAS PRESERVATION

Table 9.3 Constituents of preservation solutions Constituents

EC

Colloids ●

UW

HTK

CR

50

HES

Impermeants ● ● ● ● ● ● ●

glucose mannitol lactobionate raffinose histidine tryptophan ␣-ketoglutarate

Osmolarity

182

10 30 100 30 180 2 1

355

320

57.5 10 7.4

25

310

300

Buffers ● ● ●

PO4– HCO3– pH

7.2

1 7.2

6.8

O2 radical scavengers ● ●

glutathione allopurinol

3 1

3 1

Electrolyte concentrations ● ● ● ● ●

Na+ K Cl– MgSO4 lactate

10 115 15

30 120 5

15 9 50 4

130 5 112 + 28

Additives ● ● ●

adenosine insulin dexamethasone

100 8 0.5

1 +

prevent acidosis, avoid interstitial oedema, capture oxygen radicals, especially during reperfusion, and supply precursor materials for the energy metabolism of the organ. The effects of the various constituents of UW solution are achieved by their presence in the extracellular space and not by equilibration of the transplant to the preservative solution. The solution is viscous (4.8 cP at 1° C) and has an electrolyte composition similar to that of intracellular fluid with high potassium concentrations (135 mmol/l) and low sodium concentrations (35 mmol/l) with a pH of 7.4. The pH is maintained by using phosphate as a buffer. To stabilize the membrane potential magnesium, sulfate and lactobionate are added to calcium chelators. Acting as impermeant substances, HES, lactobionate, and reffinose prevent the emergence of cell oedema during hypothermia. Furthermore, addition of HES — colloid substance — enabled UW solution to be used for machine perfusion of the kidneys. Adenosine is added to aid ATP regeneration (Table 9.3). The importance of the adenosine in UW solution has been shown in many experimental studies in which UW solution, with adenosine in various concentrations, or none at all, was employed [15,16,38]. Although there was no substantial difference between UW solutions containing 1 and 100 mmol/l, complete removal of adenosine from the solution was followed by poorer results. The use

H.U. SPIEGEL AND D. PALMES

of allopurinol inhibits both the breakdown of adenosine nucleotides and the generation of oxygen radicals [39]. Furthermore, glutathione additive will capture free oxygen radicals. On prolonged storage, however, the reduced form of glutathione is converted into the oxidative form, and the reducing action of glutathione as an oxygen radical scavenger is lost. When using UW solution which has been in store for some time, it has proved useful to add glutathione immediately before use. In the light of early experience with UW solution under experimental and clinical conditions for liver, kidney, heart and intestine preservation, certain modifications of its composition were made. First of all, HES was considered unnecessary for simple hypothermic preservation. For example, in the successful long-term preservation of rabbit kidneys, a better initial transplant function was demonstrated in kidneys which had been preserved in UW solution free from HES. To maintain osmolarity, in the modified UW solution the concentration of raffinose has been doubled to 60 mmol/l. Secondly, the ratio of sodium and potassium had been reversed, because the high potassium concentrations in the original UW solution (designed to simulate intracellular fluid) proved damaging to cells and had vasoconstrictor effects [40,41]. Besides this, overloading of the systemic circulation with potassium diffusing out of the transplant caused cardiac arrhythmias, sometimes progressing to ventricular fibrillation. Lastly, it proved unnecessary to add insulin or dexamethasone [42]. Since the early 1990s heart, lung, liver, pancreas, kidney, and intestinal transplants have been performed with great success with the modified UW solution. For pancreas preservation under hospital conditions, cold ischaemia times of up to 12 h have been achieved with UW solution. Prostanoid inhibitors [43], NO synthetase inhibitors, and platelet-activating factor (PAF) antagonists, used as additives, prolonged pancreas preservation under experimental conditions for up to 96 h [44]. Using UW solution, ischaemia tolerance times (safe limits) of 18 h for liver and up to 36 h for kidney have been achieved. At present, therefore, it is the preferred perfusion solution and the ‘gold standard’ for multiorgan harvesting [45–47].

Histidine–Tryptophan–Ketoglutarate solution Bretschneider’s HTK solution has been in use since 1961 for experimental work and since 1971 as a cardioplegic solution in open-heart surgery. Its organ protective action depends on the following principles. The impermeant aminoacids histidine, tryptophan and ␣-ketoglutarate retard the unavoidable acidosis due to ischaemia and buffer critically low pH values. In addition, the impermeant aminoacids and the added mannitol maintain cell volume regulation under conditions of ischaemia. Because of the low extracellular sodium concentration (reduced to approximately intracellular levels) together with elevation of potassium concentration to roughly twice the extracellular norm, the load on the cellular ion pumps is relieved. By the omission of glucose, economies are made in inessential energy-consuming processes, such as the utilization of the lactate transport system when the anaerobic glycolysis rate reaches high levels. Finally, histidine, tryptophan, and ␣-ketoglutarate are capable of capturing oxygen radicals, thus minimizing the severity of reperfusion damage (Table 9.1). In contrast to other preservative solutions, the mode of action of HTK solution depends on a socalled ‘equilibration’ of the extracellular compartment, that is, by perfusion with a low-viscosity fluid (viscosity of HTK solution: 1.8 cP at 1° C) the temperature and composition of the extracellular fluid are rendered practically identical with those of the preservation solution [48–50]. Complete equilibration takes time, because the separate constituents of the extracellular fluid shift at different speeds as they come into line with the composition of the preservative solution. Calcium equilibration needs more time than sodium equilibration. This difference in kinetics can be evened out, for example, by adding magnesium, calcium antagonists, and calcium complexing agents. Because the perfusion time for organ equilibration is so much longer than perfusion with ‘single flush-out’ with other perfusion

107

108

PANCREAS PRESERVATION

solutions, the so-called ‘high volume concept’ of HTK solution has arrised. The amount of perfusion fluid needed for the purpose is calculated from the dry weight of the transplant, among other factors. For example, for complete equilibration of the kidneys (dry weight 250 to 400 g) roughly 3 litres of HTK solution will be required and for complete equilibration of the liver (dry weight 500 to 600 g). 10–15 litres of HTK solution will be needed. In general, when using HTK solution for organ preservation, the perfusion volume required (150 ml/kg body weight) will be roughly three times that needed for preservation with UW solution (standard perfusion: 50 ml/kg body weight) The considerably more viscious UW-solution (4.8 cP at 1° C) is purely a rinsing solution (‘flush solution’), and its effect are brought about the length of its stay (dwell time) in the extracellular compartment. The perfusion volume required is therefore only about one-tenth of the amount of HTK solution that is necessary. The concept advanced by Southard et al. does not see equilibration as necessary [42]. HTK solution has been used for heart transplants since 1986, and cold ischaemia times of 6–8 hr have been achieved [51–53]. HTK solution has also been in routine use for kidney and liver preservation since 1987 [54,55]. Since 1991 Eurotransplant has recommended HTK solution as an alternative of equal value of UW solution for kidney and liver transplantation. In clinical practice HTK solution is only recently used for pancreas preservation. Experimental studies of pancreas preservation in pigs after 24 h cold ischaemia time showed results comparable with those achieved by UW solution. In these circumstances HTK solution has three theoretical advantages over UW solution. First, the lower potassium concentration avoids arterial spasm and endothelial cell damage during perfusion. Secondly, due to the lower viscosity of HTK solution, arteriovenous flow rate is higher, the flushing out the pancreas is faster, and organ temperature drops more rapidly. Thirdly, because of the good buffering capacity of HTK solution anaerobic glycolysis is slower and lactate levels fall. However, a perfusion time of 4 min did not prevent occurrence of initial pancreatic oedema. This oedema, which did not affect pancreatic function, can be prevented by prolonging equilibrium times to about 10 min, but not by the addition of HES [56,57].

Two layer storage technique The two-layer storage technique provides a method for normobaric oxygenation of the pancreas during cold storage in EC or UW solution. It was originally developed as an alternative to continuous perfusion, and proved advantageous for kidney preservation because of the increased oxygen input, but when tried for pancreas preservation it led to oedema formation. The pancreas is stored in a suspension made from EC or UW solution and perfluorochemical (PFC). PFC is biologically inert, dissolves at estimated 10 times more oxygen than water at 4° C, and can therefore be used as a normobaric oxygen carrier. As the specific gravity of PFC is 1.95 and hence greater than that of EC or UW solution, PFC separates clearly from the preservative solution. The pancreas floats on the PFC surface and is surrounded by preservative solution. During the cold storage period the PFC is continuously oxygenated at a flow rate of 50 to 100 ml/min, so that the oxygen dissolved in the PFC can diffuse directly into the lower surface of the pancreas [58] (Fig. 9.3). Besides the advantage of additional oxygenation, there is also evidence that by the combined use of PFC and UW solution ATP can be synthesized by direct phosphorylation of adenosine, which is present in UW solution [59]. Membrane stability and mitochondrial function can be sustained by the high ATP concentration, and the release of lysosomal enzymes can be avoided. In addition, high ATP levels ensure rapid regeneration of the Na/K pump during reperfusion and prevent the emergence of acidosis from anaerobic metabolism [60]. The quality of ‘marginal’ donor organs, for example, pancreases from non-beating heart donors damaged by ischaemia, can be improved [61]. The two-layer storage technique, used in combination with EC solution, can achieve pancreas preservation for over 72 h [62]. However, the combination of PFC with UW solution proved superior,

H.U. SPIEGEL AND D. PALMES

Fig.9.3 Two-layer storage technique. PFC, perfluorochemical; UW, University of Wisconsin (solution)

owing to the higher concentrations of colloids and adenosine. Kin et al. [60] reported a survival rate of 57.1% of pancreas transplants after being stored in UW/PFC solution for 96 h. Kuroda et al. studied the preservation of pancreases which had already suffered ischaemic damage with warm ischaemia times of 60 min in UW/PFC and was able to make significant increases in cold ischaemia times as compared with simple storage in UW solution [61].

Flushout solution (Carolina rinse solution) The flushing out of the transplant before reperfusion has in general two purposes. It removes potentially harmful components of the preservation solution and accumulated metabolic endproducts, and equalizes the temperature between the hypothermic transplant and the normothermic recipient. It also serves to check that the anastomoses are leak-proof before starting reperfusion. For many years Ringer lactate was used as rinsing solution. The Carolina rinse (CR) solution, developed by Lemaster and colleagues at the University of North Carolina, is a further development of Ringer lactate solution. It has the same ionic composition as Ringer lactate but also contains osmotically active substances (albumin or HES), antioxidants (allopurinol, glutathione, desferrioxamine), energy substrates (adenosine, glucose, insulin, fructose), vasodilators (calcium channel blockers), and glycine. It is slightly acid, with a pH of 6.8 [63,64] (Table 9.3). Besides flushing out the transplant, CR solution was also successful in reducing reperfusion damage still further. It was therefore preferred over Ringer lactate as a flushing solution. Under experimental and clinical conditions a temperature of 20° C has proved ideal. Colloid additives, magnesium, glucose, fructose, and insulin have been found to be unnecessary. Urushihara et al., using a modified CR solution containing lactobionate as buffer and osmotic substance together with Na-famostat, a synthetic protease inhibitor, was able to reduce ischaemia/reperfusion damage after pancreas transplantation, and achieved significantly longer cold ischaemia times than were possible without rinsing solution [65–67].

Summary Owing to its filamentous structure and the absence of a capsule, the pancreas has a greater tendency to oedema formation and ischaemia-associated transplant pancreatitis than other solid organs. The main aims in pancreas preservation are therefore the avoidance of cell oedema and transplant pancreatitis. Whereas continuous organ perfusion, for example, of the kidney, achieves longer preservation times than simple hypothermic storage, this technique has proved unsuccessful for the pancreas because of its marked tendency to oedema. Conversely, by omitting the intravascular flushout of the

109

110

PANCREAS PRESERVATION

pancreas before cold storage, the incidence of acute transplant thrombosis and pancreatitis can be still further reduced [60,68,69]. When choosing the appropriate preservation solution, consideration must be given to the points mentioned above (the tendency to oedema formation and acute transplant pancreatitis) and also to its suitability for multiorgan harvesting, because the pancreas is usually resected en bloc with the liver during multiorgan harvesting. Furthermore, subsequent islet cell isolation must not be compromised. For example, Contractor et al. showed that UW solution inhibits the collagenase digestion phase employed for the isolation of human islet cells [70]. The gold standard for pancreas preservation is still UW solution, and in clinical practice it can achieve ischaemia tolerance of up to 12 h. However, current studies indicate that owing to its lower viscosity, lower potassium concentration, and higher buffer capacity, HTK solution presents an alternative of at least equal value to UW solution. The cold storage time of the pancreas can be prolonged to intermediate cold ischaemia times by new methods such as the two-layer storage technique. However, further prolongation of the cold ischaemia times of the pancreas would be useful only if it will combine with prolongation of renal ischaemia tolerance, because the vast majority of pancreatic transplants are performed in conjunction with renal transplantation.

References 1 Baker A, Dhawan A, Heaton N. Who needs a liver transplant? (new disease specific indications). Arch Dis Child 1998;79:460–4. 2 Brunkhorst R, Schlitt HJ. Kidney transplantation. Indications, results, pre- and postoperative care. (German). Internist (Berl) 1996;37:264–71. 3 Burdelski M, Rogiers X. Liver transplantation in metabolic disorders. Acta Gastroenterol Belg 1999;62:300–5. 4 Keck BM, Bennett LE, Fiol BS, Daily OP, Novick RJ, Hosenpud JD. Worldwide thoracic organ transplantation: a report from the UNOS/ISHLT International Registry for Thoracic Organ Transplantation. Clin Transplant 1999;12:35–49. 5 Kendall DM, Robertson RP. Pancreas and islet transplantation. Challenges for the twenty-first century. Endocrinol Metab Clin N Am 1997;26:611–30. 6 Gilsdorf RB, Clark SD, Leonard AS: Extracorporal recipient shunt homograft kidney perfusion: a model for organ resuscitation and function evaluation. Trans Am Soc Artif Intern Organs 1965; 11:219. 7 Ackermann JR, Fisher AJ, Barnard CN. Live storage of kidneys: a preliminary communication. Surgery 1966;60:720–4. 8 Lavender AR, Forland M, Rams JJ, Thompson JS, Russe HP, Spargo BH. Extracorporeal renal transplantation in man. JAMA 1968;203:265–71. 9 Clavien PA, Harvey PR, Strasberg SM. Preservation and reperfusion injuries in liver allografts. An overview and synthesis of current studies. Transplantation 1992;53:957–78. 10 Howden B, Rae D, Jablonski P, Marshall VC, Tange J. Studies of renal preservation using a rat kidney transplant model. Evaluation of citrate flushing. Transplantation 1983;35:311-14. 11 Ploeg RJ, Boudjema K, Marsh D, Brujin JA, Gooszen HG, Southard JH, et al. The importance of a colloid in canine pancreas preservation. Transplantation 1992;53:735–41. 12 Ploeg RJ, van Bockel JH, Langendijk PT, Groenewegen M, van der Woude FJ, Persijn GG, et al. Effects of preservation solution on results of cadaveric kidney transplantation. The European Multicentre Study Group. Lancet 1992;18:129–37. 13 Todo S, Tzakis A, Starzl TE. Preservation of livers with UW or Euro Collins solution. Transplantation 1988;46:925–6.

H.U. SPIEGEL AND D. PALMES

14 Kadmon M, Bleyl J, Kuppers B, Otto G, Herfarth C. Biliary complications after prolonged University of Wisconsin preservation of liver allografts. Transplant Proc 1993;25:1651–2. 15 Sumimoto R, Kamada N, Jamieson NV, Fukuda Y, Dohi K. A comparison of a new solution combining histidine and lactobionate with UW solution and Euro Collins for rat liver preservation. Transplantation 1991;51:589–93. 16 Sumimoto R, Southard JH, Belzer FO. Livers from fasted rats acquired resistance to warm and cold ischemia injury. Transplantation 1993;55:728–3. 17 Rolles K, Foreman J, Pegg DE. A pilot clinical study of retrograde oxygen persufflation in renal preservation. Transplantation 1989;48:339–42. 18 Rolles K, Foreman J, Pegg DE. Preservation of ischemically injured canine kidneys by retrograde oxygen persufflatio. Transplantation 1984;38:102–6. 19 Kuroda Y, Fujino Y, Kawamura T, Suzuki Y, Fujiwara H, Saitoh Y. Excellence of perfluorochemical with simple oxygen bubbling as a preservation medium for simple cold storage of canine pancreas. Transplantation 1990;49:648–50. 20 Kuroda Y, Fujino Y, Kawamura T, Suzuki Y, Fujiwara H, Saitoh Y. Mechanisim of oxygenation of pancreas during preservation by a two-layer (Euro Collins solution/perfluorochemical) cold-storage method. Transplantation 1990;49:694–6. 21 Urushihara T, Sumimoto R, Sumimoto K, Jamieson NV, Ito H, Ikeda M, et al. A comparison of some simplified lactobionate preservation solutions with standard UW solution and Euro Collins solution for pancreas preservation. Transplantation 1992;53:750–4. 22 Hakim NS. Pancreas transplantation. Am Roy Coll Surg Engl 1998;80:313–51. 23 Cicalese L, Giacomoni A, Rastellini C, Benedetti E. Pancreatic transplantation: a review. Int Surg 1999;84:305–12. 24 Mayers JT, Dennis VW, Hoogwerf BJ. Pancreas transplantation in type 1 diabetes: hope vs reality. Cleve Clin J Med 2000;67:281–6. 25 Freise CE, Narumi S, Stock PG, Melzer JS. Simultaneous pancreas-kidney transplantation: an overview of indications, complications, and outcomes. West J Med 1999;170:11–18. 26 Berney T, Ricordi C. Islet transplantation. Cell Transplant 1999;8:461–4. 27 D’Alessandro AM, Southard JH, Love RB, Belzer FO. Organ preservation. Surg Clin N Am 1994;74:1083–95. 28 Todd K, Kleinman R, Brunicardi FC. Influence of preoperative donor factors on the performance of the isolated perfused human pancreas. Transplant Proc 1994;26:552. 29 Andrews PM, Bates SB. Improving Euro Collins flushing solution’s ability to protect kidneys from normothermic ischemia. Miner Electrolyte Metab 1985;11:309–13. 30 Bretan PN Jr, Baldwin N, Martinez A, Stowe N, Scarpa A, Easley K, et al. Improved renal transplant preservation using a modified intracellular flush solution (PB-2). Characterization of mechanisms by renal clearance, high performance liquid chromatography, phosphorus-31 magnetic resonance spectroscopy, and electron microscopy studies. Urol Res 1991;19(2):73–80. 31 Grino JM, Alsinal J, Castelao AM, Sabate I, Mestre M, Gil-Vernet S, et al. Low-dose cyclosporine, antilymphocyte globulin, and steroids in first cadaveric renal transplantation. Transplant Proc 1988;20:18–20. 32 Dafoe DC, Campbell DA Jr, Marks WH, Borgstrom A, Merion RM, Berlin RE, et al. Detrimental effects of four hours of cold storage on porcine pancreaticoduodenal transplantation. Surgery 1986;99:170–7. 33 Wahlberg JA, Southard JH, Belzer FO. Development of a cold storage solution for pancreas preservation. Cryobiology 1986;23:477–82. 34 Arita S, Asano T, Suzuki S, Amemiya H, Isono K. The efficacy of CMH (Collins modified with HES) solution in canine pancreatic graft preservation. Transplant Proc 1995;27:3035–6.

111

112

PANCREAS PRESERVATION

35 Sutherland DE, Gruessner AC, Gruessner RW. Pancreas transplantation: a review. Transplant Proc 1998;30:1940–3. 36 Belzer FO, D’Alessandro AM, Hoffman RM, Knechtle SJ, Reed, A, Pirsch JD, et al. The use of UW solution in clinical transplantation. A 4-year experience. Ann Surg 1992;215(6):579–83; discussion 584–5. 37 Belzer FO, Kalayoglu M, D’Alessandro AM, Pirsch JD, Sollinger HW, Hoffmann R, et al. Organ preservation: experience with University of Wisconsin solution and plans for the future, Clin Transplant 1990;4:73–7. 38 Ametani MS, D’Alessandro AM, Southard JH. The effect of calcium in the UW solution on preservation of the rat liver. Am Transplant 1997;2:34–8. 39 Astier A, Paul M. Instability of reduced glutathione in commercial Belzer cold storage solution. Lancet 1989;2:556–7. 40 Moen J, Claesson K, Pienaar H, Lindell S, Ploeg RJ, McAnulty JE et al. Preservation of dog liver, kidney, and pancreas using the Belzer-UW solution with a high-sodium and low-potassium content. Transplantation 1989;47:940–5. 41 Hesse UJ, Troisi R, Jacobs B, Berrevoet F, DeLaere S, Maene L, et al. 24 hours cold preservation of the porcine pancreas with HTK solution. In: Hesse UJ ed. Organ preservation with HTK and UW solution — update on the clinical use and experimental studies. Lengerich: Pabst Science Publishers, 1999. 42 Southard JH, van Gulik TM, Ametani MS, Vreugdenhil PK, Lindell SL, Pienaar BL, et al. Important components of the UW solution. Transplantation 1990;49:251–7. 43 Marshall VC. Organ and tissue preservation. In: Chapman JR et al. Ed., Organ and tissue donation for transplantation. London: Arnold, 1997. 44 Hotter G, Closa D, Pi F, Prats N, Fernandez-Cruz L, Bulbena O, et al. Nitric oxide and arachidonate metabolism in ischemia-reperfusion associated with pancreas transplantation. Transplantation 1995;15:417–21. 45 Erhard J, Lange R, Scherer R, Kox WJ, Bretschneider HJ, Gebhard MM, et al. Comparison of histidine–tryptophan–ketoglutarate (HTK) solution versus University of Wisconsin (UW) solution for organ preservation in human liver transplantation. A prospective, randomized study. Transplant Int 1994;7:177–81. 46 Groenewoud AF, Thorogood J. Current status of the Eurotransplant randomized multicenter study comparing kidney graft preservation with histidine–tryptophan–ketogluterate, University of Wisconsin, and Euro Collins solutions. The HTK Study Group. Transplant Proc 1993;25:1582–5. 47 van Gulik TM, Nio CR, Cortissos E, Klopper PJ, van der Heyde MN. Comparison of HTK solution and UW solution in 24- and 48-hour preservation of canine hepatic allografts. Transplant Proc 1993;25:2554. 48 Bretschneider HJ, Helmchen U, Kehrer G. Nierenprotektion. Klin Wochenschr 1988;66:817–27. 49 Bretschneider HJ. Organübergreifende Prinzipien zur Verlängerung der Ischämietoleranz. Leopoldina 1992;37:161–74. 50 Dreikorn K. Organ preservation (German). Zentralbl Chir 1992;117(12):642–7. 51 Holscher M, Groenewoud AF. Current status of the HTK solution of Bretschneider in organ preservation. Transplant Proc 1991;23:2334–7. 52 Krohn E, Stinner B, Fleckenstein M, Gebhard MM, Bretschneider HJ. The cardioplegic solution HTK: effects on membrane potential, intracellular K+ and Na+ activities in sheep cardiac Purkinje fibres. Pflugers Arch 1989;415:269–75. 53 Reichenspurner H, Russ C, Uberfuhr R, Nollert G, Schluter A, Reichart B, et al. Myocardial preservation using HTK solution for heart transplantation. A multicenter study. Eur J Cardiothorac Surg 1993;7:414–19.

H.U. SPIEGEL AND D. PALMES

54 Hatano E, Tanaka A, Shinohara H, Kitai T, Satoh S, Inomoto T, et al. Superiority of HTK solution to UW solution for tissue oxygenation in living related liver transplantation. Transplant Proc 1996;28:1880–1. 55 Kallerhoff M, Blech M, Gotz L, Kehrer G, Bretschneider HJ, Helmchen U, et al. A new method for conservative renal surgery — experimental and first clinical results. Langenbecks Arch Chir 1990;375:340–6. 56 Troisi R, Hesse UJ. Electrolyte equilibration during perfusion with histidine–tryptophan–ketoglutarate (HTK) solution of procine pancreas: in vivo results. In: Hesse UJ, ed. Organ preservation with HTK and UW solutions – update on the clinical use and experimental studies. Lengerich: Pabst Science Publishers, 1999. 57 Leonhardt U, Tytko A, Exner B, Barthel M, Stockmann F, Kohler H et al. The effect of different solutions for organ preservation on immediate postischemic pancreatic function in vitro. Transplantation 1993;55:11-14. 58 Kuroda Y, Kawamura T, Suzuki Y, Fujiwara H, Yamamoto K, Saitoh Y. A new, simple method for cold storage of the pancreas using perfluorochemical. Transplantation 1988;46:457–60. 59 Kuroda Y, Matsumoto S, Fujita H, Tanioka Y, Sakai T, Hamano M, et al. Resuscitation of ischemically damaged pancreas during short-term preservation at 20 degrees C by the two-layer (University of Wisconsin solution/perfluorochemical) method. Transplantation 1996;61:28–30. 60 Kin S, Stephanian E, Gores P, Mass A, Flores H, Nakai I, et al. 96-hour cold-storage preservation of the canine pancreas with oxygenation using perfluorochemical. Transplantation 1993;55:229–30. 61 Kuroda Y, Morita A, Fujino Y, Tanioka Y, Ku Y, Saitoh Y. Successful extended preservation of ischemically damaged pancreas by the two-layer (University of Wisconsin solution/perfluorochemical) cold storage method. Transplantation 1993;56:1087–90. 62 Kawamura T, Kuroda Y, Suzuki Y, Fujiwara H, Fujino Y, Yamamoto K, et al. Seventy-two-hour preservation of the canine pancreas by the two-layer (Euro Collins solution/perfluorochemical) cold storage method. Transplantation 1989;47:776–8. 63 Gao W, Takei Y, Marzi I, Currin RT, Lemasters JJ, Thurman RG. Carolina rinse solution increases survival time dramatically after orthotopic liver transplantation in the rat. Transplant Proc 1991;23:648–50. 64 Gao WS, Takei Y, Marzi I, Lindert KA, Caldwell-Kenkel JC, Currin RT, et al. Carolina rinse solution — a new strategy to increase survival time after orthotopic liver transplantation in the rat. Transplantation 1991;52:417–24. 65 Urushihara T, Sumimoto K, Sumimoto R, Ikeda M, Fukuda Y, Dohi K. Rinse solution containing a protease inhibitor and Na-lactobionate increase graft survival after rat pancreas preservation. Transplant Proc 1994;26:559–60. 66 Urushihara T, Sumimoto K, Sumimoto R, Ikeda M, Fukuda Y, Dohi K. Nafamostat mesilate rinse solution improves graft survival after rat pancreas and heart preservation. Transplant Proc 1995;27:786–7. 67 Urushihara T, Sumimoto K, Sumimoto R, Ikeda M, Yamanaka K, Okugawa K, et al. Prevention of reperfusion injury after rat pancreas preservation using rinse solution containing nafamostat mesilate. Transplant Proc 1996;28:1874–5. 68 Morel P, Moudry-Munns K, Balakumar , Najarian JS, Dunn DL, Sutherland DE. Influence of preservation time on the early function of pancreas transplants. Transplant Proc 1990;22:527–8. 69 Wright FH, Wright C, Ames SA, Smith JL, Corry RJ. Pancreatic allograft thrombosis: donor and retrieval factors and early postperfusion graft function. Transplant Proc 1990;22(2):439–41. 70 Contractor HH, Johnson PR, Chadwick DR, Robertson GS, London NJ. The effect of UW solution and its components on the collagenase digestion of human and porcine pancreas. Cell Transplant 1995;4:615–19.

113

This page intentionally left blank

Chapter 10

Surgical techniques of pancreas transplantation Venkatesh Krishnamurthi and Stephen T. Bartlett

Insulin-dependent diabetes mellitus (IDDM) affects 1 million people in the United States and 30 000 new patients annually [1]. In addition to being the leading cause of endstage renal disease (ESRD), IDDM is also the leading cause of blindness, amputations, and impotence in adults. Data from the Diabetes Control and Complication Trial (DCCT) conclusively demonstrate that tight glucose control reduces the incidence of secondary complications [2]. The benefits of intensive control were not without risk however, as severe hypoglycaemia episodes and weight gain occurred more frequently in patients treated with intensive control. Moreover, as was concluded by the DCCT Research Group, there was no minimal glycaemic threshold below which complications would not occur [3]. These findings reinforce the belief that exogenous insulin administration is, at best, an imperfect solution for IDDM. Pancreas transplantation is the only treatment for IDDM that consistently establishes an insulinindependent euglycaemic state with complete normalization of glycosylated haemoglobin levels. Since its early stages over 30 years ago, this procedure has undergone considerable evolution and expansion. According to data from the International Pancreas Transplant Registry (IPTR), over 1000 pancreas transplant have been performed yearly at American centres since 1994. Throughout this period, overall graft and patient survival rates at 3 years remain above 70 and 90 per cent, respectively [4]. Several factors including donor management and selection, organ procurement and preservation, and advances in immunosuppressive medications have resulted in these improved outcomes. In addition to these factors, refinements in surgical techniques have been also been instrumental in driving pancreas transplantation to its current level of success.

Historical background A review of the history behind pancreas transplantation is helpful to understanding the evolution of the current operative technique. Kelly and Lillehei at the University of Minnesota performed the first successful pancreas transplant in 1966 [5]. In their report of two cases, the first was a segmental ductligated pancreas graft was transplanted into an extraperitoneal position in the left iliac fossa. Vascular reconstruction was performed to the recipient external iliac vessels. Graft loss occurred within 2 months as a result of severe rejection and graft pancreatitis. In the second case, a whole-organ pancreaticoduodenal graft was transplanted, also in an extraperitoneal position with similar vascular reconstruction. In contrast to the first case however, the distal end of the bowel segment was diverted as a cutaneous jejunostomy.

116

SURGICAL TECHNIQUES

In subsequent cases of pancreas transplantation vascular reconstruction to the iliac vessels remained the standard practice but the management of exocrine secretions continued to be the major obstacle. Consequently investigators devised various methods of controlling the exocrine drainage. In 1973, Gleidman et al. described segmental pancreatic transplantation with anastomosis of the pancreatic duct to the ureter [6]. This procedure was technically challenging and necessitated an ipsilateral nephrectomy. Moreover, anastomotic leaks were common and their avoidance required long-term placement of a silicone stent. Groth et al. developed a more physiological solution to manage the exocrine secretions by performing a Roux-en-Y pancreaticojejunostomy [7]. As in previous cases, pancreatic and enterocutaneous fistulae occurred in all of these patients. In an attempt to obviate the exocrine pancreatic function, Dubernard et al. reported on a new technique of segmental pancreas transplantation using duct occlusion with neoprene, a synthetic rubber [8]. Although the exocrine function was eliminated, the resulting pancreatitis and foreign body inflammatory reaction led to graft damage and consequently stimulated interest in developing new techniques for duct management. Other groups performed segmental pancreas transplants with free intraperitoneal drainage of the exocrine secretions [9,10]. Recurrent ascites from drainage of pancreatic secretions remained a problem that often required graft irradiation and occasionally led to graft loss. According to data from International Human Pancreas and Islet Transplant Registry, 105 pancreas transplants had been performed at 23 institutions up to 1980. Long-term graft survival was poor as only five pancreas allografts functioned beyond 1 year [11]. In approximately three-quarters of these cases, a segmental pancreas graft was transplanted and in the remaining cases, a pancreaticoduodenal graft was used. In a more detailed analysis by the same registry, five techniques for duct management were identified and all were associated with a less than 50 per cent 1-year pancreas graft survival. Specifically, enteric drainage (ED) with a Roux-en-Y loop had a 1 year survival of 41 per cent; duct injected grafts, 36 per cent; urinary drainage, 26 per cent; and open duct, 14 per cent [12]. These data led to a renewed pursuit for alternative techniques of controlling the exocrine secretions. Starzl et al. at Pittsburgh described three cases of pancreas transplant using variable lengths of duodenum for anastomosis to the recipient gastrointestinal tract [13]. Intestinal complications occurred frequently as the first two patients developed protein-losing enteropathy and severe rejection of the bowel segment. In both patients multiple enteric reconstructions were required and finally resulted in formation of a side-to-side duodenojejunostomy, which was used as a primary means of exocrine drainage in the last patient. Interestingly, their described technique of ED is almost identical to the current practice. Simultaneously, other groups in Wisconsin and Iowa showed renewed interest in urinary drainage techniques. Solllinger et al., at the University of Wisconsin, developed a new method of pancreaticocystostomy. Ten cases of segmental pancreas transplant were performed and in each case, a direct anastomosis between the pancreatic duct and the bladder mucosa was performed [12]. In a subsequent report, the same authors described a technique of pancreaticocystostomy with a whole organ graft. In this technique, a button of duodenum surrounding the ampulla of Vater was implanted into the bladder [14]. Finally, Nghiem and Corry at the University of Iowa refined the method of duodenocystostomy which remains the most common technique of duct management utilized today [15,16].

Current techniques The various surgical techniques of pancreas transplantation can broadly be classified according to the type of exocrine drainage performed. Bladder drainage (BD) continues to be the most common method of duct management primarily because of the ability to monitor urinary amylase as a marker

V. KRISHNAMURTHI AND S.T. BARTLETT

for rejection. With improvements in immunosuppression and a reduction in the frequency of surgical complications combined with the unique metabolic and urological complications posed by BD, increasing numbers of transplant programmes are adopting ED as a primary method of duct management. In 1998, ED was used in over 50 per cent of simultaneous pancreas and kidney (SPK) transplants as compared to only 6 per cent in 1994 [4]. Pancreas graft survival rates at 1 year in this patient group were similar between these two duct management techniques (83 per cent for BD and 82 per cent for ED). A more significant difference in pancreas graft survival according to the type of duct management is notable in recipients of solitary pancreas transplants although these differences appear to be a result of a higher technical failure rate among solitary pancreas transplants. In a recent survey of 121 active pancreas transplant centres recognized by the International Pancreas and Islet Transplant Association, 78 (64 per cent) reported their preferred surgical technique of pancreas transplantation [17]. Thirty different surgical techniques of pancreas transplantation are described; however, these differences are largely variations on the same theme. In addition to the major differences of BD versus ED, subtle modifications, depending on the location of the anastomoses, the use of arterial and venous extension grafts, and ‘hand-sewn’ versus stapled anastomosis of the duodenal segment account for the large number of different techniques.

Whole organ pancreaticoduodenal transplantation Pancreas transplantation can be performed through either a lower quadrant extraperitoneal incisions or through a mid-line intraperitoneal approach. With a retroperitoneal approach, it is recommended that the peritoneum be opened to facilitate absorption of peripancreatic secretions. In general, a midline intraperitoneal incision is preferred as this approach allows for maximum flexibility and for the performance of concomitant procedures such as simultaneous kidney transplantation, peritoneal dialysis catheter removal, nephrectomy, and appendectomy. Additionally, the risk of wound infection appears to be lower through a single incision [18,19]. Following routine exposure of the abdomen, lymphatics overlying the iliac artery and vein are divided. The right iliac artery and vein are preferred sites for implantation as there is more favourable anatomy in the right iliac venous system.

Venous Reconstruction The preferred method of venous drainage remains controversial. Systemic venous (SV) drainage, the method used in over 90 per cent of reporting centres, is an established surgical technique that is associated with excellent long-term results [17,20]. A theoretical disadvantage of SV drainage, however, relates to the high levels of insulin in the peripheral circulation. Hyperinsulinaemia has been shown in experimental systems to be associated with insulin resistance and altered lipid metabolism [21–23]. Portal venous (PV) drainage, a more physiological method that eliminates hyperinsulinaemia, is gaining interest among pancreas transplant centres. Following the initial description by Calne in 1984, PV drainage of pancreas allografts underwent several technical modifications [24–26]. The current technique of PV drainage used in most centres is based on the technique described by Shokouh-Amiri et al. [27]. Although follow-up is limited, several centres, including ours, have shown excellent graft survival rates and a reduced number of surgical complications with PV drainage [28–31]. The techniques of SV drainage involves anastomosis of the donor portal vein to the recipient iliac (common or external) vein or vena cava (Fig. 10.1). During the back-table preparation the donor portal vein should be mobilized to the confluence of the splenic and superior mesenteric branches. This often involves division of several pancreaticoduodenal branches. In addition, the recipient common and external iliac veins should be fully mobilized by dividing all internal iliac and lumbar

117

118

SURGICAL TECHNIQUES

Fig. 10.1 Simultaneous pancreas and kidney transplant with systemic venous and bladder drainage.

branches, effectively placing this vein anterolateral to the iliac artery on the right side and medial to the external iliac artery when performed on the left side. This manoeuvre decreases tension on the venous anastomosis, allows it to be performed under improved exposure, and reduces the risk of venous thrombosis [32–34]. In cases of SV drainage, we perform an end-to-side anastomosis directly between the portal vein and the iliac vein using fine (number 6-0 or 7-0) polypropylene suture. Several authors have described the use of venous extension grafts to facilitate the venous anastomosis. We, along with others, feel the use of extension grafts to be avoided due to the increased incidence of venous thrombosis [33,34]. Portal venous drainage remains our preferred technique of venous drainage (Fig. 10.2) see also Chapter 11. In this technique, the transverse colon is reflected cephalad, exposing the small bowel mesentery. In many instances the superior mesenteric vein (SMV) or large-calibre tributary may be immediately visualized. When this is not the case, the peritoneum in the root of the mesentery is incised, the mesenteric lymphatics are divided between ligatures, and the SMV is exposed. A sufficient length (3 to 4 cm) of the SMV is then circumferentially mobilized. This may require division of small draining branches. In preuraemic recipients we administer intravenous heparin (50 U/kg) prior to clamping the vein. An end-to-side anastomosis between the portal vein and the SMV is then performed using number 7-0 polypropylene suture. Once the venous anastomosis is complete, we occlude the donor portal vein with a bulldog clamp and restore the venous drainage of the bowel. At

V. KRISHNAMURTHI AND S.T. BARTLETT

Fig. 10.2 Simultaneous pancreas and kidney transplant with portal venous and enteric drainage.

this point, a tunnel is made in the small bowel mesentery adjacent to the venous anastomosis and through which the arterial graft is passed to the retroperitoneum. In addition to its potential physiological advantage, PV drainage may be technically easier to perform in comparison to an SV drainage procedure. Complete mobilization of the iliac vein can be time-consuming and technically challenging, particularly in a deep pelvis. With PV drainage, a large tributary of the SMV can be isolated quickly and the anastomosis is generally performed in a wellexposed area of the operative field. Additionally, anastomosis to a mesenteric vein does not appear to increase the risk of thrombosis. In our experience, thrombosis of the portal venous-drained pancreas allograft is limited to the donor portal vein and does not extend into the mesenteric venous circulation. These observations are similar to those seen by other groups with a large experience with PV drainage [35].

Arterial Reconstruction Arterial reconstruction of the pancreas allograft begins with the back-table preparation of the organ. Although this is discussed in detail elsewhere in this text (Chapter 8), a brief description here is necessary for completeness. In the majority of cases, a donor Y-iliac artery extension graft is used to join the superior mesenteric and splenic arteries on the pancreas (Fig. 10.3). In the setting of significant atherosclerosis involving the donor iliac arteries, alternate reconstruction must be considered. In selected

119

120

SURGICAL TECHNIQUES

Fig. 10.3 Use of the donor iliac Y graft in back-table reconstruction of the pancreas CBD, common bile duct; IPA, inferior pancreaticoduodenal artery; SMA, superior mesenteric artery; SMV, superior mesenteric vein; SPA, superior pancreaticoduodenal artery.

cases, particularly when the liver is not procured, a common patch of aorta containing both the coeliac and superior mesenteric orifices may be available, thus obviating the need for arterial reconstruction when performing a systemically-drained pancreas transplant. With PV drainage however, the arterial graft must be sufficiently long to traverse the mesenteric tunnel, a feature not possible with a common aortic patch. Other options are direct end-to-side anastomosis between the splenic artery and superior mesenteric artery, anastomosis between these vessels with an interposition graft, and use of the donor brachiocephalic arterial graft [34,36,37]. Arterial reconstruction with the donor Y graft should be preferred technique in whole organ pancreas transplantation, as all other arterial reconstructive techniques are associated with an increased incidence of thrombosis [34]. Arterial revascularization of the pancreas allograft is preferentially performed to the right common iliac or external iliac artery. With SV drainage, use of the right iliac vessels is technically easier because of the more superficial course of the external iliac vein and its relative ease of mobilization as compared to the left iliac vein. In a detailed analysis on the causes of pancreas allograft thrombosis, Troppman et al. clearly demonstrated a higher incidence of thrombosis with locations other than the right iliac vessels [34]. As indicated above, when PV drainage is used, it is important to have a long Y graft to reach the site of arterial anastomosis comfortably. When dictated by patient anatomy, we frequently place any excess distal external iliac artery onto the proximal common iliac artery as an extension graft during the back-table preparation of the allograft. In cases of PV drainage the right common iliac artery is also preferred due to its relatively proximity to the mesenteric tunnel. When this vessel is not available as an inflow source, the left common iliac artery or infrarenal aorta are acceptable alternatives which are easily reached with a long Y graft. The use of vascular closure staples (VCS) in pancreas tranplantion has been introduced by Hakim [38]. This has correlated with less anastomotic bleeding, decreased anastomotic and operative times and reduced thrombotic complications.

V. KRISHNAMURTHI AND S.T. BARTLETT

Duct Management As indicated above, BD remains the most common method of managing the exocrine secretions among pancreas transplant centres worldwide. This method is almost exclusively limited to systemically drained pancreas transplant although two centres performed PV-drained transplants with BD [17]. When performing BD, the donor duodenum should be kept to the minimum possible length as this avoids metabolic complications stemming from fluid and bicarbonate loss. The pancreas is oriented with the head positioned caudally and the vascular anastomoses are completed. The bladder is mobilized by dividing the lateral attachments. A location on the dome of the bladder that provides for a tension-free anastomosis is selected. The duodenum is opened along its antimesenteric border and the duodenocystostomy is performed with a two-layer ‘hand-sewn’ approach or with a stapling instrument [39]. The hand-sewn technique is similar to a standard enteroenterostomy. The mucosal layer is performed with a running absorbable suture and the seromuscular layer with interrupted silk sutures. With both the hand-sewn and stapled techniques it is imperative not to use permanent sutures or staples along an anastomosis that may be in contact with urine as this will undoubtedly lead to stone formation. Due to recent improvements in immunosuppression and antimicrobial prophylaxis, ED is increasingly being utilized as a method of managing the exocrine secretions. As has been shown in recent reports comparing ED and BD, ED is associated with a significant reduction in urological and metabolic complications with no increase in septic complications [31]. For SPK cases, ED is now the technique of choice. For solitary transplant, BD allows monitoring of urinary amylase. In contrast to BD, with ED the length of the donor duodenum is not as critical as pancreatic secretions are reabsorbed in the distal bowel segment. When performing ED, reconstructive options include direct side-to-side anastomosis to the recipient small bowel or anastomosis to a diverting Roux-en-Y limb. The anastomosis to the Roux limb may be performed in either a side-to-side or endto-end fashion. Additionally, any of the enteric anastomosis can be performed by a ‘hand-sewn’ or stapled technique. The majority of reporting centres perform the enteric anastomosis directly to the recipient bowel (58 per cent) in a hand-sewn fashion (87 per cent) [17]. Although a diverting Roux limb has the theoretical advantage of isolating anastomotic complications from the remainder of the bowel, our experience suggests that major complications related to the enteric anastomosis are uncommon and that the creation of a Roux loop may be unnecessary [40]. Additionally, complications related to construction of a Roux limb have been observed as late as 16 months following transplantation [41].

Segmental pancreas transplantation Segmental pancreas transplantation is a therapeutic option utilized at select centres in the United States following live donor pancreas transplantation. Two European centres continue to perform segmental pancreas transplant from cadaveric donors [17]. Two pancreatic segments are obtained from dividing the graft along its neck. The vasculature of the head segments is based upon the superior mesenteric artery and portal vein. The vasculature of the remaining segment, part of the body, and tail, is based upon the splenic artery and vein. Transplantation of either of these segments has been described, although, presently the tail is the preferred segment for transplantation. Vascular reconstruction of a segmental pancreas graft is performed to the recipient iliac vessels. As in the case of systemically drained whole organ transplants, the right iliac vessels are preferred. The splenic vein is anastomosed end-to-side to the external iliac vein. The splenic artery is then anastomosed to either the external, internal, or common iliac arteries. Importantly, the internal iliac artery should be divided so that the external iliac artery can lie freely lateral to the external iliac vein. This

121

122

SURGICAL TECHNIQUES

avoids impingement of the external iliac artery on the splenic vein [42]. Systemic anticoagulation should be utilized given the increased risk of thrombosis in segmental pancreas grafts. Current methods of duct management with segmental pancreas grafts include BD and duct injection. For BD, an anastomosis is created between the pancreatic duct and bladder mucosa using absorbable suture. A second layer between the pancreatic parenchyma and detrusor muscle is completed using non-absorbable suture. When the diameter of the segmental pancreas graft is relatively small, a ductal anastomosis may be avoided by invaginating the graft into the bladder. In this technique the internal layer is secured with a continuous number 4-0 absorbable suture and the external layer is completed with interrupted number 4-0 polypropylene suture [43]. Duct injection is another alternative for controlling exocrine secretions in segmental pancreas grafts. Approximately 2.5 to 5 ml of silicone or neoprene is injected into the main pancreatic duct, which results in obliteration of the exocrine secretions. Predictably the disadvantages of duct injection are the inability to monitor exocrine function as a marker of rejection and the fibrotic changes in the segmental graft.

Summary Since its inception over 30 years ago, vascularized pancreas transplantation has undergone considerable progress. Primarily as a result of the unique complications associated with transplantation of this organ, modifications in surgical technique have been necessary to improve outcomes. Presently, graft survival rates approach 90 per cent at 1 year. Despite this level of success, the technique of pancreas transplantation remains controversial. Future efforts to reduce morbidity and minimize immunosuppression will enable pancreas transplantation to remain an important therapeutic option for selected patients with IDDM.

References 1 Bennett PH, Haffner S, Kasiske BL, Keane WF, Mogensen CE, Parving HH, et al. Screening and management of microalbuminuria in patients with diabetes mellitus: recommendations to the Scientific Advisory Board of the National Kidney Foundation from an ad hoc committee of the Council on Diabetes Mellitus of the National Kidney Foundation Am J Kid Dis 1995;25:107–12. 2 DCCT Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–86. 3 DCCT Research Group. The absence of a glycemic threshold for the development of long-term complications: the perspective of the Diabetes Control and Complications Trial. Diabetes 1996;45:1289–98. 4 Gruessner AC, Sutherland DE. Analysis of United States (US) and non-US pancreas transplants are reported to the International Pancreas Transplant Registry (IPTR) and to the United Network for Organ Sharing (UNOS). Clin Transpl 1998;53–73. 5 Kelly WD, Lillehei RC, Merkel FK, Idezuki Y, Goetz FC. Allotransplantation of the pancreas and duodenum along with the kidney in diabetic nephropathy. Surgery 1967;61:827–37. 6 Gliedman ML, Gold M, Whittaker J, Rifkin H, Soberman R, Freed S, et al. Pancreatic duct to ureter anastomosis for exocrine drainage in pancreatic transplantation. Am J Surg 1973;125:245–52. 7 Groth CG, Lundgren G, Arner P, Collste H, Hardstedt C, Lewander R, et al. Rejection of isolated pancreatic allografts in patients with with diabetes. Surg Gynecol Obstet 1976;143:933–40.

V. KRISHNAMURTHI AND S.T. BARTLETT

8 Dubernard JM, Traeger J, Neyra P, Touraine JL, Tranchant D, Blanc-Brunat N. A new method of preparation of segmental pancreatic grafts for transplantation: trials in dogs and in man. Surgery 1978;84:633–9. 9 Dickerman RM, Raskin P, Fry WJ, Elick BA. Comparison of techniques of human pancreatic transplantation. Transplant Proc 1980;12:83–5. 10 Sutherland DE, Baumgartner D, Najarian JS. Free intraperitoneal drainage of segmental pancreas grafts: clinical and experimental observations on technical aspects. Transplant Proc 1980;12:26–32. 11 Sutherland DE. International human pancreas and islet transplant registry. Transplant Proc 1980;12:229–36. 12 Sollinger HW, Cook K, Kamps D, Glass NR, and Belzer FO. Clinical and experimental experience with pancreaticocystostomy for exocrine pancreatic drainage in pancreas transplantation. Transplant Proc 1984;16:749–51. 13 Starzl TE, Iwatsuki S, Shaw BWJ, Greene DA, Van TDH, Nalesnik MA, et al. Pancreaticoduodenal transplantation in humans. Surg Gynecol Obstet 1984;159:265–72. 14 Sollinger HW, Kalyoglu M, Hoffmann RM, Deierhoi MH, Belzer FO. Experience with pancreaticocystostomy in 24 consecutive pancreas transplants. Transplant Proc 1985;17:141–3. 15 Nghiem DD, Beutel WD. Duodenocystostomy for exocrine drainage in total pancreatic transplantation: a preliminary report. Transplant Proc 1986;18:1874–6. 16 Corry RJ, Nghiem DD, Schulak JA, Beutel WD, Gonwa TA. Surgical treatment of diabetic nephropathy with simultaneous pancreatic duodenal and renal transplantation. Surg Gynecol Obstet 1986;162:547–55. 17 Di Carlo V, Castoldi R, Cristallo M, Ferrari G, Socci C, Baldi A, et al. Techniques of pancreas transplantation through the world: an IPITA Center survey. Transplant Proc 1998;30:231–41. 18 Schweitzer EJ, Bartlett ST. Wound complications after pancreatic transplantation through a kidney transplant incision. Transplant Proc 1994;26:461. 19 Douzdjian V, Gugliuzza KK. Wound complications after simultaneous pancreas–kidney transplants: midline versus transverse incision. Transplant Proc 1995;27:3130–2. 20 Sollinger HW, Odorico JS, Knechtle SJ, D’Alessandro AM, Kalayoglu M, Pirsch JD. Experience with 500 simultaneous pancreas–kidney transplants. Ann Surg 1998;228:284–96. 21 Luck R, Klempnauer J, Ehlerding G, Kuhn K. Significance of portal venous drainage after whole-organ pancreas transplantation for endocrine graft function and prevention of diabetic nephropathy. Transplantation 1990;50:394–8. 22 Hughes TA, Gaber AO, Amiri HS, Wang X, Elmer DS, Winsett RP, et al. Kidney–pancreas transplantation. The effect of portal versus systemic venous drainage of the pancreas on the lipoprotein composition. Transplantation 1995;60:1406–12. 23 Bagdade JD, Ritter MC, Kitabchi AE, Huss E, Thistlethwaite R, Gabfr O, et al. Differing effects of pancreas–kidney transplantation with systemic versus portal venous drainage on cholesteryl ester transfer in IDDM subjects. Diabetes Care 1996;19:1108–12. 24 Calne RY. Paratopic segmental pancreas grafting: a technique with portal venous drainage. Lancet 1984;1:595–7. 25 Muhlbacher F, Gnant MF, Auinger M, Steininger R, Klauser R Prager R, et al. Pancreatic venous drainage to the portal vein: a new method in human pancreas transplantation. Transplant Proc 1990;22:636–7. 26 Rosenlof LK, Earnhardt RC, Pruett TL, Stevenson WC, Douglas MT, Cornett GC, et al. Pancreas transplantation. An initial experience with systemic and portal drainage of pancreatic allografts. Ann Surg 1992;215:586–95. 27 Shokouh-Amiri MH, Gaber AO, Gaber LW, Jensen SL, Hughes TA, Elmer D, et al. Pancreas transplantation with portal venous drainage and enteric exocrine diversion: a new technique. Transplant Proc 1992;24:776–7.

123

124

SURGICAL TECHNIQUES

28 Gaber AO, Shokouh-Amiri H, Hathaway DK, Dammontree L, Kitabchi AE, Gaber LW, et al. Results of pancreas transplantation with portal venous and enteric drainage. Ann Surg 1995;221:613–24. 29 Gaber AO, Shokouh-Amiri H, Hathaway DK, Gaber LW, Elmer D, Kitabchi A et al. Pancreas transplantation with portal venous and enteric drainage eliminates hyperinsulinemia and reduces postoperative complications. Transplant Proc 1993 25; 1176–8. 30 Bartlett ST, Kuo PC, Johnson LB, Lim JW, Schweitzer EJ. Pancreas transplantation at the University of Maryland. Clin Transpl 1996;271–80. 31 Stratta RJ, Gaber AO, Shokouh-Amiri MH, Reddy KS, Egidi MF, Grewal HP, et al. A prospective comparison of systemic-bladder versus portal-enteric drainage in vascularized pancreas transplantation. Surgery 2000;127:217–26. 32 Gill IS, Sindhi R, Jerius JT, Sudan D, Stratta RJ. Bench reconstruction of pancreas for transplantation: experience with 192 cases. Clin Transpl 1997;11:104–9. 33 Sollinger HW. Pancreatic transplantation and vascular graft thrombosis (editorial; comment). J Am Coll Surg 1996;182:362–3. 34 Troppmann C, Gruessner AC, Benedetti E, Papalois BE, Dunn DL, Najarian JS, et al. Vascular graft thrombosis after pancreatic transplantation: univariate and multivariate operative and nonoperative risk factor analysis. J Am Coll Surg 1996;182:285–316. 35 Reddy KS, Stratta RJ, Shokouh-Amiri MH, Alloway R, Egidi MF, Gaber AO. Surgical complications after pancreas transplantation with portal-enteric drainage. J Am Coll Surg 1999;189:305–13. 36 Mizrahi S, Boudreaux JP, Hayes DH, Hussey JL. Modified vascular reconstruction for pancreaticoduodenal allograft. Surg Gynecol Obstet 1993;177:89–90. 37 Ciancio G, Olson L, Burke GW. The use of the brachiocephalic trunk for arterial reconstruction of the whole pancreas allograft for transplantation. J Am Coll Surg 1995;181:79–80. 38 Papalois VE, Romagnoli J, Hakim NS. Use of vascular closure staples in vascular access for dialysis, kidney and pancreas transplantations. Int Surg 1998;83:177–180. 39 Pescovitz MD, Dunn DL, Sutherland DE. Use of the circular stapler in construction of the duodenoneocystostomy for drainage into the bladder in transplants involving the whole pancreas. Surg Gynecol Obstet 1989;169:169–71. 40 Kuo PC, Johnson LB, Schweitzer EJ, Bartlett ST. Simultaneous pancreas/kidney transplantation — a comparison of enteric and bladder drainage of exocrine pancreatic secretions. Transplantation 1997;63:238–43. 41 Nymann T, Shokouh-Amiri MH, Elmer DS, Stratta RJ, Gaber AO. Diagnosis, management, and outcome of late duodenal complications in portal-enteric pancreas transplantation: case reports. J Am Coll Surg 1997;185:560–6. 42 Gruessner RW, Kendall DM, Dragstveit MB, Gruessner AC, Sutherland DE. Simultaneous pancreas–kidney transplantation from liver donors. Ann Surg 1997;226:471–80.

Chapter 11

Pancreas transplantation with portal-enteric drainage Robert J. Stratta, M. Hosein Shokouh-Amiri, Hani P. Grewal, and A. Osama Gaber

Introduction Vascularized pancreas transplantation (PTX) was first developed as a means to re-establish endogenous insulin secretion responsive to normal feedback controls. From 1966 through to July 2000, over 14 000 PTXs were performed worldwide and reported to the International Pancreas Transplant Registry (IPTR) [1]. According to IPTR data, most PTXs are performed with systemic venous delivery of insulin and either bladder or enteric [systemic-bladder (S-B) or systemic-enteric (S-E)] drainage of the exocrine secretions [2]. From 1988 through 1995, more than 90 per cent of PTX procedures were performed by the standard technique of S-B drainage using a duodenal segment conduit. Although well tolerated in most PTX recipients, S-B drainage was associated with a finite and troublesome rate of unique metabolic and urological complications resulting from altered physiology. When these complications became persistent or refractory, conversion from bladder to enteric drainage (enteric conversion) was often necessary and successful [3]. Because of a favourable experience with enteric conversion, coupled with advances in preservation, donor selection, and immunosuppression that placed the duodenal segment at a lower risk for ischaemic or immunological injury, a resurgence of interest occurred in primary enteric drainage in an effort to avoid the complications of bladder drainage. Since 1995, the number of PTX procedures performed with primary enteric drainage has steadily increased, accounting for 60 per cent of cases in 1999 [1]. In the last few years, the results of simultaneous kidney PTX (SPK) with enteric drainage have improved and are now comparable to SPK with bladder drainage [2]. Despite an evolution in surgical techniques, the majority of PTXs with enteric drainage are performed with systemic venous delivery of insulin, resulting in peripheral hyperinsulinaemia. In the non-transplant setting, chronic hyperinsulinaemia has been associated with insulin resistance, dyslipidaemia, accelerated atherosclerosis, and macroangiopathy. To improve the physiology of PTX, a new surgical technique was developed, combining portal venous delivery of insulin with enteric drainage of the exocrine secretions [portal-enteric (P-E)] [4–6]. In a recent survey of surgical techniques among PTX centres, seven reported experience with the P-E technique, of which five used a diverting Roux limb [7]. Table 11.1 provides a list of centres that have reported experience in PTX with P-E drainage. Many of these centres have adopted the P-E technique as their preferred method of PTX. However, the proportion of cases with P-E drainage has remained low and represents only 15 to 20 per cent of enteric drained PTXs [1,2]. In the most recent IPTR analysis including PTXs performed

126

PANCREAS TRANSPLANTATION WITH PORTAL-ENTERIC DRAINAGE

Table 11.1 Recent experience in pancreas transplantation with P–E drainage Center University of Tennessee, Memphis University of Chicago University of Maryland Lyon, France Toronto Louisiana State University University Hospital, Cleveland University of Edmonton, Alberta University of Virginia Duke University University of Massachusetts Bochum, Germany Northwestern University

Number of cases > > > > > > > > > > > > >

100 100 100 20 20 15 15 15 10 10 10 10 5

References 18, 21, 23, 35, 36, 45 19, 20, 24, 27 30, 31 32 33 34 * * 16, * * * 25, 26 *

* Personal communication.

between 1996 and 1999, the 1-year pancreas graft survival rates were similar for SPK with either P-E or S-E drainage, 83 and 84 per cent, respectively [2].

Historical background The history of clinical PTX largely revolves around the development and application of various surgical techniques. Experience in PTX with portal venous delivery of insulin dates back to the mid1980s. Initial attempts employed segmental PTX with either gastric (Calne 1984 [8]), pyelic (Gil-Vernet et al. 1985 [9]), or jejunal (Tyden et al. 1985 [10] and Sutherland et al. 1987 [11]) drainage. Whole organ PTX using the P-E technique was first described clinically by our group in 1992 [4] and was based on experimental work by Shokouh-Amiri et al. in a porcine model [12–14]. This new technique employed a tributary of the superior mesenteric vein to re-establish portal venous drainage and differed substantially from other initial reports of whole organ PTX with portal venous drainage. In 1990, Muhlbacher et al. described a whole organ technique involving an end-toside anastomosis between the distal splenic vein of the donor and the recipient’s portal vein in combination with bladder drainage [15]. In 1992, Rosenlof et al. applied Calne’s original technique to whole organ PTX using an end-to-side anastomosis between the donor portal vein and the recipient’s splenic vein coupled with enteric drainage [16]. In each of these other series, however, the procedure was not widely applied because of technical problems associated with the vascular reconstruction [17]. In 1993, our group reported that P-E PTX with Roux limb diversion not only achieves acceptable metabolic control and eliminates hyperinsulinaemia but was also associated with reduced postoperative complications [6]. In 1995, we compared 19 patients undergoing SPK with the P-E technique versus a concurrent and historical control group of 28 patients receiving SPK with the conventional S-B technique [18]. Actuarial patient and graft survival rates at 1 and 3 years were no different in the two groups. Metabolic and urological complications and urinary tract infections were more common in the S-B group. Metabolic control was comparable between groups, and peripheral hyperinsulinaemia did not occur in patients with P-E drainage.

R.J. STRATTA ET AL.

In 1995, Newell et al. from the University of Chicago reported their initial experience with a similar P-E technique in 12 SPK recipients compared to a retrospective control group of 12 SPK patients with S-B drainage [19]. Six-month patient and graft survival rates were comparable, and the P-E group had less acidosis, dehydration, haematuria, rejection, and need for enteric conversion. There were no differences in technical complications, and renal and pancreas allograft functions were similar. In 1996, Newell et al. presented 12-month follow-up on the same two groups with similar findings [20]. In addition, the initial length of stay and total hospital days in the first year after SPK were slightly lower in the P-E group. There were no significant differences in costs, no delay in the diagnosis of rejection, and the authors concluded that their initial results confirmed the safety and efficacy of this new technique. In 1977, Nymann et al. from our group reported improving outcomes with increased experience with the P-E technique [21]. Two groups were compared: 23 SPKs with P-E drainage performed from 1991 to 1994 versus 23 P-E PTXs [17 SPK, three PAK, three (PTA)] performed in 1995 and 1996. The latter group received tacrolimus (TAC)-based immunosuppression, while the former group was managed with cyclosporin (CyA). Cold ischaemia time and perioperative blood transfusion were significantly lower in the latter group. In addition, the incidence of technical graft loss was reduced from 26 to 9 per cent. Consequently, 1-year patient and pancreas graft survival rates were improved in the later era. In 1998, Nymann et al. analyzed 47 SPKs with graft function at 1 month, including 30 with S-B and 17 with P-E drainage [22]. All patients had received CyA-based therapy. Although the authors noted comparable patient and graft survival and surgical complication rates, the incidences of rejection, graft loss due to rejection, and the density of rejection were all lower in patients with P-E drainage. Also in 1998, Eubanks et al. from our group compared 12 solitary PTXs with S-B drainage performed from 1991 to 1995 with 16 solitary PTXs with P-E drainage performed between July 1995 and March 1997 [23]. The former group was managed with CyA and the latter group with TAC-based immunosuppression. One patient in each group experienced graft loss as a result of thrombosis. In the remaining patients, the incidence and density of rejection were lower in the more recent era, leading to an improvement in the 1-year pancreas graft survival rate to 80 per cent. In each of these studies, the authors concluded that the results of PTX with the P-E technique are now comparable to the other reported techniques. In 1998, Bruce et al. from the University of Chicago reported their updated experience with 70 consecutive SPKs with P-E drainage performed between January 1992 and August 1997 [24]. They compared this group to a ‘historical’ control group of 70 SPKs with S-B drainage performed between January 1987 and December 1994. One-year patient, kidney, and pancreas graft survival rates were comparable between groups. There were no significant differences in technical or immunological graft failure rates (no enteric or anastomotic leaks were reported). Renal and pancreas allograft functions at 1 year were similar. However, the total number of hospital days and operative complications in the first year were significantly higher in the S-B group, with the difference in these results almost entirely accounted for by a 21 per cent rate of enteric conversion in patients with S-B drainage. In addition, the authors noted a possible ‘learning curve’ effect, with improved results in the latter 35 versus the former 35 SPKs with P-E drainage. In 1998, Busing et al. reported on 70 consecutive SPKs without anastomotic complications, including two with P-E drainage [25]. Busing et al. later updated his experience to 10 SPKs with P-E drainage, including none using a Roux limb [26]. Kidney and pancreas survival rates were both 90 per cent, with one graft lost due to thrombosis. Buell et al. likewise updated the University of Chicago experience, including 16 SPKs with P-E drainage without a Roux limb [27]. This group also reported good initial results with the P-E technique in the absence of a diverting Roux limb.

127

128

PANCREAS TRANSPLANTATION WITH PORTAL-ENTERIC DRAINAGE

In 1999, Reddy et al. reported a reduction in the surgical complication rate after PTX with P-E drainage that was attributed to increased experience with the technique [28]. Also in 1999, Stratta et al. reported that the incidence of allograft pancreatectomy was not influenced by the surgical technique of implantation [29]. In 1999, Philosophe et al. from the University of Maryland reported their initial experience with 66 PTXs with P-E drainage compared to 183 PTXs with S-E drainage [30]. Graft survival rates for SPK, sequential pancreas after kidney transplant (PAK), and PTA recipients were similar according to technique. However, when stratified for human leucocyte antigen (HLA) matching, the incidence of rejection was lower in patients with P-E drainage. In a follow-up report in 2000, Philosophe et al. compared 117 solitary PTXs with P-E drainage versus 70 with S-E drainage [31]. The authors noted not only an improvement in the pancreas graft survival rate, but also a decrease in the incidence and severity of rejection in patients with P-E drainage. The authors concluded that P-E drainage may be associated with an immunological advantage. In 2000, the Lyon group reported a prospective study of 34 SPK recipients randomized to either S-E or P-E drainage with a Roux limb [32]. Patient and graft survival rates and morbidity were similar between groups. Also in 2000, Cattral et al. prospectively studied 20 SPKs with S-B drainage followed by a sequential cohort of 20 consecutive SPKs with P-E drainage [33]. One-year patient and graft survival rates were similar between groups. However, medical morbidity, cytomegalovirus (CMV) infections, and acute rejection were more common in the S-B group. Zibari et al. reported their initial experience with 17 SPKs with P-E drainage and a Roux-en-Y venting jejunostomy to monitor for rejection and prevent anastomotic leak [34]. Patient, kidney, and pancreas graft survival rates were 100, 100, and 94 per cent, respectively, after a mean follow-up of 16 months. In each of these studies, the authors concluded that SPK with P-E drainage can be performed with excellent short-term outcomes and minimal morbidity. Herein we reported the chronology of our 9-year single-centre experience with 126 PTXs with P-E drainage spanning different immunosuppressive eras.

Programme overview The University of Tennessee (UT) Memphis PTX programme began in 1989 (Fig. 11.1) [35]. Between April 1989 and September 1990, 24 consecutive SPKs were performed with S-B drainage (Fig. 11.2).

Fig. 11.1 University of Tennessee, Memphis, pancreas transplant experience by year according to type of transplant

R.J. STRATTA ET AL.

Fig. 11.2 University of Tennessee, Memphis, pancreas transplant experience by year according to technique of transplant.

The first SPK with P-E drainage was performed in October 1990, and this patient continues to enjoy excellent dual allograft function over 10 years later. Also in 1990, the first solitary PTXs were performed at our programme including both sequential PAK and PTA (Fig. 11.1). From October 1990 to December 1994, we performed 42 SPKs including 26 with P-E and 16 with S-B drainage (Fig. 11.2). During the same interval, a total of 18 solitary PTXs were performed with S-B drainage, including 13 PTA and five PAK. In 1995 and 1996, 42 consecutive PTXs (29 SPK, nine PAK, four PTA) were performed exclusively with P-E drainage. From February 1997 to March 1998, we compared 32 consecutive PTXs performed with either S-B or P-E drainage [36]. From April 1998 to May 2000, 54 consecutive SPK recipients were entered into a prospective study of S-E versus P-E drainage at our centre. From 1989 through 2000, we performed a total of 276 PTXs, including 153 with P-E, 76 with S-B, and 47 with S-E drainage (Fig. 11.3). This overall experience accumulated over a decade includes 196 SPKs, 43 PTA, and 37 PAKs (Fig. 11.4). The UT Memphis PTX programme is currently one of the seven largest centres in the United States and recently became the thirteenth

Fig. 11.3 Total number of pancreas transplants at University of Tennessee, Memphis, according to technique (1989 to 2000, n = 276).

129

130

PANCREAS TRANSPLANTATION WITH PORTAL-ENTERIC DRAINAGE

Fig. 11.4 Total number of pancreas transplants at University of Tennessee, Memphis, according to recipient category (1989 to 2000, n = 276).

Fig. 11.5 Overall experience in pancreas transplantation with portal-enteric drainage spanning different immunosuppressive eras.

centre worldwide to perform 250 PTXs. Through 1999, we had one of the largest single-centre experiences with the P-E technique, including 126 PTXs (90 SPK, 18 PAK, 18 PTA) with P-E drainage (Fig. 11.5). This report represents a case series and our collective experience with the P-E technique.

Organ preparation and transplantation Prior to transplantation, the pancreas was reconstructed on the back-table with a donor iliac artery bifurcation Y graft to the splenic and superior mesenteric arteries [37]. The P-E procedure requires that the arterial bifurcation graft be constructed intentionally long for subsequent arterialization. The donor portal vein was mobilized and dissected back to the splenic and superior mesenteric venous confluence without the need for a venous extension graft. The proximal duodenal staple line (just distal to the pylorus) was inverted with suture, and the distal duodenal closure incorporated the third and a variable length of the fourth portion of the duodenum, as previously described [35]. The

R.J. STRATTA ET AL.

closure of the mesenteric root was reinforced with a running suture. The spleen was left attached to the tail of the pancreas to be used as a handle, but in some cases, the splenic hilar structures were ligated in continuity before revascularization. The kidney was likewise prepared using standard techniques. The pancreaticoduodenal graft was repackaged separately and in sterile fashion in cold University of Wisconsin (UW) solution prior to implantation. After preparation of the organs, the recipient operation was performed through a mid-line intraperitoneal approach. The surgical technique of P-E drainage has been previously described in detail by our group (Fig. 11.6) [4–6,18,35]. The portal vein of the pancreas graft was anastomosed end-to-side to a major tributary of the superior mesenteric vein. The donor iliac bifurcation graft was brought through a window made in the distal ileal mesentery and anastomosed end-to-side to the right common iliac artery. The transplant duodenum was anastomosed to a diverting Roux-en-Y limb of recipient jejunum. Splenectomy was performed after revascularization, and an attempt was made to anchor the tail of the pancreas to the anterior abdominal wall with interrupted sutures. These anchoring sutures permitted subsequent percutaneous, ultrasound-guided pancreas allograft biopsies to be performed as needed [38].

Immunosuppression Most PTX centres use quadruple drug immunosuppression with antilymphocyte induction (ALI) because of a high incidence of rejection and the general impression that the pancreas is a highly immunogenic organ. The evolution of surgical techniques has been largely facilitated by the rapid

Fig. 11.6 Technique of pancreas transplantation with portal-enteric drainage (see text for details).

131

132

PANCREAS TRANSPLANTATION WITH PORTAL-ENTERIC DRAINAGE

changes in immunosuppressive therapy. With the recent commercial availability of potent immunosuppressive agents such as TAC and mycophenolate mofetil (MMF), the need for routine ALI therapy after PTX has been questioned [39,40]. From October 1990 to June 1995 (Era 1), 30 SPKs and P-E drainage were performed at our centre with quadruple therapy consisting of OKT3 induction in combination with CyA-Sandimmune, prednisone, and azathioprine (Fig. 11.7) [18]. CyA dosing was titrated to achieve a target 12-h trough level of greater than 300 ng/ml for the first 3 months after transplant and greater than 200 ng/ml thereafter. Azathioprine dosing was 1 to 2 mg/kg/day. Prednisone was tapered to achieve a dose of 10 mg/day by 1 year and 5 mg/day by 2 years after transplant. From July 1995 to May 1998 (Era 2), 42 SPKs and 23 solitary PTXs (11 PAK, 12 PTA) with P-E drainage received TAC, prednisone, and MMF triple therapy without antibody induction [23,39,40]. The TAC dosing was titrated to a 12-h trough level of 15 to 25 ng/ml for the first 3 months after transplant. After 3 months, TAC blood levels were maintained at 10 to 15 ng/ml in the absence of rejection. Oral MMF was begun immediately after transplant at 2 to 3 g/day in two to four divided doses. The MMF dose was reduced in patients with gastrointestinal intolerance (nausea, vomiting, diarrhoea) or when the total white blood cell count was less than 3000/mm3. The MMF was discontinued temporarily in patients with active CMV infection or septicaemia, or when the total white blood cell count was less than 2000/mm3; it was restarted later at a reduced dose. Prednisone was gradually tapered to achieve a dose of 5 mg/day at 1 year. From June 1998 to December 1999 (Era 3), 18 SPKs and 13 solitary PTXs (seven PAK, six PTA) with P-E drainage received TAC, MMF, and prednisone immunosuppression with or without either Simulect (basiliximab) or Zenapax (daclizumab) antibody induction [41] (Fig. 11.8). Half of the SPK and all of the solitary PTX recipients received either basiliximab (20 mg intravenous on day 0 and 4) or daclizumab (1 mg/kg on day 0 and then at 2-week intervals for a total of five doses) as induction therapy.

Statistical analysis Data are reported as mean and range. Renal allograft loss was defined as death with function, transplant nephrectomy, return to dialysis or to the pretransplant serum creatinine level. Pancreas graft loss was defined as death with function, transplant pancreatectomy, or the need for daily scheduled insulin therapy.

Era 1 (10/90–6/95) 30 SPKs with P-E drainage received OKT3 induction, CSA-Sandimmune, prednisone, and azathioprine

Era 2 (7/95–5/98) 42 SPKs and 23 solitary PTXs (11 PAK, 12 PTA) received tacrolimus, prednisone and mycophenolate mofetil without antibody induction

Era 3 (6/98–12/99) 18 SPKs and 13 solitary PTXs (seven PAK, six PTA) received tacrolimus mycophenolate mofetil, and prednisone ± basiliximab or daclizumab induction

Fig. 11.7 Chronology of experience in pancreas transplantation with portal-enteric drainage according to immunosuppressive era.

R.J. STRATTA ET AL.

Fig. 11.8 Regimen of immunosuppression in Era 3 with selective use of monoclonal antibodies directed against the interleukin-2 receptor (I L 2 R) as induction therapy. Maintenance immunosuppression was triple therapy with tacrolimus (F K), mycophenolate mofetil (M M F), and steroids.

Results S-B versus P-E drainage With preliminary data demonstrating the equivalence of both procedures, we designed a prospective evaluation of PTX with S-B versus P-E drainage [36]. During an 11-month period extending from February 1997 to January 1998, 32 consecutive PTXs were performed at our centre and patients were alternately assigned to either S-B or P-E drainage. The total of 16 patients were allocated to each technique. The S-B group included 11 SPK, one PAK, and four PTA recipients while the P-E group included 12 SPK, two PAK, and two PTA recipients. The two groups were well matched for donor and recipient demographic, immunological, and transplant characteristics (Table 11.2). All SPK and PAK recipients received primary immunosuppression with TAC, MMF, and steroids without ALI. The PTA recipients in both groups received OKT3 induction in addition to the above triple maintenance therapy. Patient, kidney, and pancreas graft survival rates were 88 per cent S-B versus 94 per cent P-E, 92 per cent S-B versus 93 per cent P-E, and 81 per cent S-B versus 88 per cent P-E, respectively, with a mean follow-up of 8 months (minimum of 3 months) (Table 11.3). All kidney grafts had immediate function and the incidence of early technical problems related to the pancreas allograft (pancreatitis, thrombosis) was similar in the two groups. There were no graft losses either to immunological or infectious complications in either group, but the incidence of acute rejection was slightly higher in the S-B group (44 per cent S-B versus 31 per cent P-E, P = NS). Length of stay and hospital charges for the initial hospital admission were similar between groups (Table 11.3). For all patients, the mean length of initial hospital stay was 13 days and initial hospital charges approximated US$100 000. The incidence of urological complications was doubled in the S-B group (25 per cent S-B versus 12 per cent P-E, P = NS). The S-B group was also characterized by a higher incidence of urinary tract infections (50 per cent S-B versus 19 per cent P-E, P = 0.12). Metabolic acidosis with oral bicarbonate supplementation was universal in the S-B groups, but rarely occurred with P-E drainage. Dehydration

133

134

PANCREAS TRANSPLANTATION WITH PORTAL-ENTERIC DRAINAGE

Table 11.2 Group characteristics (mean ± SD) Systemic-bladder (n = 16) Age (years) Gender female male Years of diabetes Race: Caucasian Pretransplant dialysis Prior transplant Current PRA ≥ 10% Pancreas ischaemia (h) HLA match: ABDR HLA mismatch Waiting time (months) Type of transplant SPK PAK PTA

Portal-enteric (n = 16)

38 ± 6

37 ± 7

7 (44%) 9 (56%) 25.5 ± 6 16 (100%) 11/11 (100%) 1 (6%) 1 (6%) 15.1 ± 3.4 1.8 ± 1.3 4.2 ± 1.3 4.6 ± 4.3

7 (44%) 9 (56%) 26 ± 4 16 (100%) 6/12 (50%)* 3 (19%) 1 (6%) 13.4 ± 3.0 1.8 ± 1.3 4.1 ± 1.3 4.4 ± 3.1

11 (69%) 1 (6%) 4 (25%)

12 (75%) 2 (12.5%) 2 (12.5%)

* P = 0.01. PRA, panel reactive antibody.

with the need for intravenous fluid supplementation and placement of long-term indwelling central venous catheters occurred in all patients with S-B drainage but in only 44 per cent with P-E drainage (P < 0.01). The incidence of operative complications was similar, but the relaparotomy rate was higher in the P-E group (two patients in this group required a second reoperation, while no patients in the S-B group received multiple laparotomies). In the P-E group, one patient (6 per cent) had an enteric leak with intra-abdominal infection. Two patients underwent enteric conversion in the S-B group. The incidences of major infections and CMV infection were similar between groups. We believe that this study represented the first prospective analysis comparing PTX performed by a standardized technique of P-E drainage versus the conventional technique of S-B drainage with similar immunosuppression [36]. These preliminary results suggested that whole organ PTX with P-E drainage could be performed with results comparable to the conventional technique of S-B drainage.

S-E versus P-E drainage As the number of PTXs with enteric drainage has steadily increased, we decided to compare SPK with S-E versus P-E drainage in a prospective fashion with standardized immunosuppression. During a 26-month period from April 1998 to May 2000, 54 consecutive SPK recipients were entered into a prospective study of S-E (n = 27) versus P-E (n = 27) drainage. The technique to be performed was chosen before the transplant with selection determined by an alternating methodology. The two groups were well matched for most donor and recipient demographic, immunological, and transplant characteristics (Table 11.4). The racial distribution differed slightly, with African-American patients representing 15 per cent of the S-E and 33 per cent of the P-E group.

R.J. STRATTA ET AL.

Table 11.3 Results (mean ± SD) Systemic-bladder (n = 16) Portal-enteric (n = 16) Patient survival Graft survival kidney pancreas Follow-up (months) ATN (dialysis) Early technical problems/pancreatitis Initial LOS (days) Readmissions Acute rejection Major infection Reoperations Initial hospital charges (US$) Urological complication UTI Dehydration/acidosis CMV infection Multiple reoperations *

14 (88%)

15 (94%)

11/12 (92%) 13 (81%) 7.5 ± 3.4 0 3 (19%) 13.7 ± 9 2.6 ± 1.8 7 (44%) 8 (50%) 4 (25%) 100 215 ± 54 012 4 (25%) 8 (50%) 16 (100%) 2 (12.5%) 0

13/14 (93%) 14 (88%) 8.9 ± 3.8 0 3 (19%) 12.8 ± 7 1.75 ± 1.2 5 (31%) 10 (62.5%) 4 (25%) 94 083 ± 25 873 2 (12.5%) 3 (19%) 7 (44%)* 3 (19%) 2 (12.5%)

P < 0.01.

ATN: acute tubular necrosis; CMV, cytomegalovirus; LOS, length of stay; UTI, urinary tract infection.

With regard to immunosuppression, 63 per cent of S-E and 44 per cent of P-E patients were managed with no antibody induction. The remaining S-E patients received either daclizumab or basiliximab induction, while the P-E patients received daclizumab, basiliximab, or thymoglobulin in two patients with acute tubular necrosis (ATN). Maintenance immunosuppression in both groups consisted of TAC, MMF, and steroids. Results are depicted in Table 11.5. Patient survival rates were 93 per cent S-E versus 95 per cent P-E, while kidney graft survival rates were 93 per cent in both groups. PTX survival (complete insulin independence) rates were 74 per cent after S-E versus 85 per cent after P-E drainage, with a mean follow-up of 17 months. All but three of the 54 transplanted renal allografts had immediate function. ATN, defined as the need for dialysis in the first week after transplant, occurred in one patient after S-E and two patients after P-E drainage. All three of these kidneys eventually functioned. All 54 transplanted pancreas allografts had initial function, although three were subsequently lost to thrombosis in the first week after transplant. The incidence of allograft pancreatitis, early leaks, or other technical problems related to the pancreas allograft were similar between groups. The mean length of initial hospital stay was 12.4 days in the S-E and 12.8 days in the P-E groups, respectively. Mean initial hospital charges were comparable between groups. The S-E group was characterized by a slight increase in the number of readmissions (mean 2.8 S-E versus 2.2 P-E, P = NS) and total hospital days (mean 33 days S-E versus 24 days P-E, P = NS). The incidence of acute rejection was similar (33 per cent) in both groups, with immunological pancreas graft loss occurring in three S-E patients versus one P-E patient. The incidence of major infection was 52 per cent in both groups, with one CMV infection (4 per cent) in each group. The incidence of intra-abdominal infection was slightly higher in the S-E group (26 per cent versus 11 per cent P-E, P = NS). However, the early relaparotomy rate was similar between groups (30 per cent S-E versus 26 per cent P-E). The composite

135

136

PANCREAS TRANSPLANTATION WITH PORTAL-ENTERIC DRAINAGE

Table 11.4 Demographic and transplant characteristics (range in parentheses) Systemic-enteric (n = 27) Age (years) Gender male female Weight (kg) Race Caucasian African-American Asian Pre-transplant dialysis duration (months) peritoneal dialysis haemodialysis none Hepatitis C positive Years of diabetes Type 2 diabetes Daily insulin dose (U/day) Prior kidney transplant PRA > 10% HLA match: ABDR HLA mismatch CMV D+/R– Cold ischaemia (h) kidney pancreas Waiting time (months) Immunosuppression; FK, MMF, steroids + no antibody induction daclizumab basiliximab thymoglobulin

Portal-enteric (n = 27)

40.6 (27–57)

39.2 (26–58)

19 (70%) 8 )30%) 73.9 (45–103)

16 (59%) 11 (41%) 73.3 (49–95)

22 (81%) 4 (15%) 1 (4%) 21 (78%) 16 (5–36) 7 (26%) 14 (52%) 6 (22%) 2 (7%) 24 (4–50) 4 (15%) 38 (15–80) 2 (7%) 2 (7%) 1.4 (0–4) 4.6 (2–6) 7 (26%)

18 (67%) 9 (33%) 0 18 (67%) 13 (1–46) 10 (37%) 8 (30%) 9 (33%) 1 (4%) 23.2 (9–46) 2 (7%) 44 (15–80) 3 (11%) 2 (7%) 1.4 (0–4) 4.6 (2–6) 5 (19%)

14.3 (8–23) 14.2 (7.5–22.5) 2.8 (0.1–7)

15.1 (9.5–2.6) 15.3 (10.5–23) 3 (0.25–8.5)

17 (63%) 6 (22%) 4 (15%) 0

12 8 5 2

(44%) (30%) (19%) (7%)

P = NS. CMV, cytomegaovirus; D, donor; FK, tacrolimus; HLA, human leucocyte antigen; MMF, mycophenolate mofetil; PRA, panel reactive antibody; R, recipient

endpoint of no rejection, graft loss, or death was attained by 56 per cent of S-E and 59 per cent of P-E patients (Table 11.5). These results suggested that SPK with S-E or P-E drainage could be performed with comparable short-term outcomes.

Overall results From October 1990 to December 1999, we performed 126 PTXs with P-E drainage (Fig. 11.5), including 90 SPKs and 36 solitary PTXs (18 PAK, 18 PTA). The P-E group included 69 male and 57 female patients with a mean age of 39 years (Table 11.6). The mean duration of pretransplant diabetes was 24 years (range 8 to 50). The majority of recipients were Caucasian, although 15 (12 per cent) were African-American recipients. A total of 13 patients (10 per cent) underwent pancreas retransplanta-

R.J. STRATTA ET AL.

Table 11.5 Results (range in parentheses)

Patient survival Graft survival kidney pancreas Follow-up (months) ATN (postoperative dialysis) Early technical problems Initial hospital stay (days) Initial hospital charges (US$) Readmissions No readmissions Acute rejection Anti T-cell therapy Immunological pancreas graft loss Early relaparotomy (< 3 months) Pancreas thrombosis Major infection CMV infection Intra-abdominal infection Total hospital days Event-free survival (no rejection, graft loss, or death)

Systemic-enteric (n = 27)

Portal-enteric (n = 27)

25 (93%)

26 (96%)

25 (93%) 20 (74%) 17.4 (5–30) 1 (4%) 3 (11%) 12.4 (7–30) 102 255 2.8 (0–10) 8 (30%) 9 (33%) 4 (15%) 3 (11%) 8 (30%) 2 (7%) 14 (52%) 1 (4%) 7 (26%) 33 (9–160) 15 (56%)

25 (93%) 23 (85%) 17 (5–29) 2 (7%) 2 (7%) 12.8 (7–38) 105 789 2.2 (0–10) 5 (19%) 9 (33%) 6 (22%) 1 (4%) 7 (26%) 1 (4%) 14 (52%) 1 (4%) 3 (11%) 24 (8–92) 16 (59%)

P = NS. ATN, acute tubular necrosis; CMV, cytomegalovirus.

Table 11.6 Demographic and transplant characteristics n Age (years) Gender female male Race Caucasian African-American Years of diabetes Transplant type SPK PAK PTA Prior PTX HLA match: ABDR Pancreas cold ischaemia (h)

126 39 (range 19–56) 57 (45%) 69 (55%) 111 (88%) 15 (12%) 24 (range 8–50) 90 (72%) 18 (14%) 18 (14%) 13 (10%) 1.4 (range 0–5) 13 (range 6–23)

HLA; human leucocyte antigen.

tion with the P-E technique. The majority of patients had poor HLA matching (mean 1.4, range 0 to 5), and the mean pancreas cold ischemia was 13 h (range 6 to 23). Minimum follow-up was 11 months (mean 4.6 years).

137

138

PANCREAS TRANSPLANTATION WITH PORTAL-ENTERIC DRAINAGE

Thirty patients underwent SPK with P-E drainage in Era 1 and were compared to 42 SPKs performed in Era 2 and 18 in Era 3 (Fig. 11.7). The patients in Era 1 were managed with CyA while those in Eras 2 and 3 received TAC/MMF. We also compared 23 solitary PTXs (11 PAK, 12 PTA) performed in Era 2 with 13 (seven PAK, six PTA) performed in Era 3. One-year patient survival rates after SPK (Fig. 11.9) were 77 per cent in Era 1, 93 per cent in Era 2, and 100 per cent in Era 3 (P = 0.03). The 1year kidney graft survival rates were 77 per cent in Era 1, 93 per cent in Era 2, and 94 per cent in Era 3 (P = 0.08). The 1-year pancreas graft survival rates after SPK (Fig. 11.9) were 60 per cent in Era 1, and 83 per cent in Eras 2 and 3 (P = 0.06). The most common causes of kidney graft loss were death with function and chronic rejection (Table 11.7). The overall incidence of kidney graft loss decreased from 56 per cent in Era 1 to 23 per cent in Era 2 to 11 per cent in Era 3 (P < 0.001). The most common causes of pancreas graft loss were thrombosis, death with function, chronic rejection, and infection (Table 11.7). The overall incidence of pancreas graft loss decreased from 60 per cent in Era 1 to 31 per cent in Era 2 to 22 per cent in Era 3 (P < 0.001). The incidences of rejection (63 per cent versus 33 per cent versus 39 per cent, P < 0.001) and major infection (60 per cent versus 43 per cent versus 44 per cent, P = NS) after SPK were decreased in each successive era (Fig. 11.10). The rates of thrombosis (20 per cent versus 7 per cent versus 6 per cent, P < 0.001) and early relaparotomy (47 per cent versus 31 percent versus 33 per cent, P = NS) after SPK were also decreased in each consecutive era (Fig. 11.11). The 1-year patient survival rates after solitary PTX were both 100 per cent in Eras 2 and 3, while the corresponding pancreas graft survival rates were 61 and 69 per cent, respectively (Table 11.8). The most common causes of graft loss after solitary PTX were thrombosis and chronic rejection. The overall incidence of pancreas graft loss after solitary PTX decreased from 70 per cent in Era 2 to 31 per cent in Era 3 (P = 0.02). The rates of acute rejection (57 versus 38 per cent), major infection (35 versus 31 per cent), thrombosis (22 versus 15 per cent), and relaparotomy (43 versus 38 per cent) after solitary PTX were all slightly improved in Era 3 compared to Era 2 (P = NS). This overall experience demonstrates that SPK and solitary PTX with P-E drainage can be performed with improving outcomes. Increasing experience with the P-E technique coupled with advances in immunosuppression are associated with: (a) increasing patient, kidney, and pancreas graft survival rates; (b) less medical morbidity with a decreasing incidence of acute rejection and major infection; and (c) reduced surgical complications including decreasing rates of thrombosis and relaparotomy. The P-E technique does not appear

Fig. 11.9 One-year patient and graft survival rates after SPK according to immunosuppressive era. Survival rates were similar in Eras 2 and 3 and significantly improved compared to Era 1.

R.J. STRATTA ET AL.

Table 11.7 Results (SPK)

One-year survival patient kidney pancreas Acute rejection Major infection Thrombosis Relaparotomy Overall graft loss kidney pancreas Causes of kidney graft loss DWFG chronic rejection infection acute rejection PTLD thrombosis Causes of pancreas graft loss thrombosis DWFG chronic rejection infection PTLD acute rejection

Era 1

Era 2

Era 3

(n = 30)

(n = 42)

(n = 18)

P value

(77%) (77%) (60%) (63%) (60%) (20%) (47%)

39 (93%) 39 (93%) 35 (83%) 14 (33%) 18 (43%) 3 (7%) 13 (31%)

18 (100%) 17 (94%) 15 (83%) 7 (39%) 8 (44%) 1 (6%) 6 (33%)

0.03 0.08 0.06 < 0.001 NS < 0.001 NS

17 (56%) 18 (60%)

10 (23%) 13 (31%)

2 (11%) 4 (22%)

< 0.001 < 0.001

7 (23%) 4 (13%) 2 (7%) 1 (3%) 2 (7%) 1 (3%)

5 (12%) 3 (7%) 1 (2%) 0 1 (2%) 0

1 (5.5%) 1 (5.5%) 0 0 0 0

< 0.001 NS NS NS NS NS

6 (20%) 5 (17%) 1 (3%) 3 (10%) 2 (7%) 1 (3%)

3 (7%) 2 (5%) 5 (12%) 1 (2%) 1 (2%) 1 (2%)

1 1 1 1

< 0.001 < 0.001 NS NS NS NS

23 23 18 19 18 6 14

(5.5%) (5.5%) (5.5%) (5.5%) 0 0

DWFG, death with functioning graft; PTLD, post-transplant lymphoproliferative disease.

Fig. 11.10 The incidence of acute rejection was similar in Eras 2 and 3 and significantly decreased compared to Era 1.

139

140

PANCREAS TRANSPLANTATION WITH PORTAL-ENTERIC DRAINAGE

Fig. 11.11 The incidence of pancreas allograft thrombosis was similar in Eras 2 and 3 and significantly decreased compared to Era 1.

Table 11.8 Results (solitary PTX)

PAK PTA One-year survival patient pancreas Acute rejection Major infection Thrombosis Relaparotomy Overall pancreas graft loss Causes of Graft Loss thrombosis chronic rejection infection/PTLD acute rejection primary non-function DWFG

Era 2: (n = 23)

Era 3: (n = 13)

11 (48%) 12 (52%)

7 (54%) 6 (46%)

23 (100%) 14 (61%) 13 (57%) 8 (35%) 5 (22%) 10 (43%) 16 (70%)

13 9 5 4 2 5 4

5 (22%) 4 (17%) 3 (13%) 2 (9%) 1 (4%) 1 (4%)

(100%) (69%) (38%) (31%) (15%) (38%) (31%)

2 (15%) 2 (15%) 0 0 0 0

P value NS NS NS NS NS NS NS NS 0.02 NS NS NS NS NS NS

DWFG, death with functioning graft; PTLD, post-transplant lymphoproliferative disease.

to incur any additional or unique risks, and can be performed with results comparable to the other standard techniques of PTX. We believe that this technique should be included in the repertoire of PTX, because it offers potential physiological, metabolic, and immunological advantages over the other techniques currently available.

R.J. STRATTA ET AL.

Summary The University of Tennessee, Memphis, group has made a number of important contributions to the field of PTX, including the development of a novel whole organ technique of PTX with portal venous drainage of insulin and primary enteric drainage of the exocrine secretions. The P-E technique has the potential to become the standard of care in the near future because it is more physiological, normalizes carbohydrate and lipid metabolism, and minimizes complications attributed to the transplant procedure. In addition, we have been actively involved in studying new immunosuppressive regimes to improve and simplify the care of the PTX recipient. We believe that PTX will remain an important treatment option for insulin-treated diabetic patients with complications until other strategies are developed that can provide equal glycaemic control with less or no immunosuppression and less overall morbidity.

References 1 Sutherland DER, Gruessner AC. International Pancreas Transplant Registry Update. IPTR Newsletter 2000;12:1–23. 2 Gruessner AC, Sutherland DER. Analyses of pancreas transplant outcomes for United States cases reported to the United Network for Organ Sharing (UNOS) and non-US cases reported to the International Pancreas Transplant Registry (IPTR). In: Cecka JM, Terasaki PI, ed. Clinical transplants 1999. Los Angeles: UCLA Immunogenetics Center, 2000:51–69. 3 Sindhi R, Stratta RJ, Lowell JA, et al. Experience with enteric conversion after pancreas transplantation with bladder drainage. J Am Coll Surg. 1997;184:281–9. 4 Shokouh-Amiri MH, Gaber AO, Gaber LW, et al. Pancreas transplantation with portal venous drainage and enteric exocrine diversion: A new technique. Transplant Proc 1992;24:776–7. 5 Gaber AO, Shokouh-Amiri H, Grewal HP, Britt LG. A technique for portal pancreatic transplantation with enteric drainage. Surg Gynecol Obstet 1993;177:417–19. 6 Gaber AO, Shokouh-Amiri H, Hathaway DK, et al. Pancreas transplantation with portal venous and enteric drainage eliminates hyperinsulinemia and reduces post-operative complications. Transplant Proc 1993;25:1176–8. 7 Di Carlo V, Castoldi R, Cristallo M, et al. Techniques of pancreas transplantation through the world: An IPITA center survey. Transplant Proc 1998;30:231–41. 8 Calne RY. Paratopic segmental pancreas grafting: A technique with portal venous drainage. Lancet 1984;1:595–7. 9 Gil-Vernet J, Fernandez-Cruz L, Andreu J, Figuerola D, Caraleps A. Clinical experience with pancreaticopyelostomy for exocrine pancreatic drainage and portal venous drainage in pancreas transplantation. Transplant Proc 1985;17:342–5. 10 Tyden G. Wilczek H, Lundgren G, et al. Experience with 21 intraperitoneal segmental pancreatic transplants with enteric or gastric exocrine diversion in humans. Transplant Proc 1985;17:331–5. 11 Sutherland DER, Goetz FC, Moudry KC, Abouna GM, Najarian JS. Use of recipient mesenteric vessels for revascularization of segmental pancreas grafts: technical and metabolic considerations. Transplant Proc 1987;19:2300–4. 12 Shokouh-Amiri MH, Rahimi-Saber S, Andersen AJ. Segmental pancreatic autotransplantation in the pig. Transplantation 1989;47:42–4. 13 Shokouh-Amiri MH, Falholt K, Holst JJ, et al. Pancreas endocrine function in pigs after segmental pancreas autotransplantation with either systemic or portal venous drainage. Transplant Proc 1992;24:799–800.

141

142

PANCREAS TRANSPLANTATION WITH PORTAL-ENTERIC DRAINAGE

14 Shokouh-Amiri MH, Rahimi-Saber S, Anderson HO, Jensen SL. Pancreas autotransplantation in pig with systemic or portal venous drainage: Effect on the endocrine pancreatic function after transplantation. Transplantation 1996;61:1004–9. 15 Muhlbacher F, Gnant MFX, Auinger M, et al. Pancreatic venous drainage to the portal vein: a new method in human pancreas transplantation. Transplant Proc 1990;22:636–7. 16 Rosenlof LK, Earnhardt RC, Pruett TL, et al. Pancreas transplantation: An initial experience with systemic and portal drainage of pancreatic allografts. Ann Surg 1992;215:586–97. 17 Rees M, Brons IGM. Pancreas transplantation: A historical perspective from a single institution. In: Hakim NS, Stratta RJ, Dubernard JM, ed. Second British Symposium on Pancreatic Transplantation. London: Royal Society of Medicine Press, 1998:85–9. 18 Gaber AO, Shokouh-Amiri MH, Hathaway DK, et al. Results of pancreas transplantation with portal venous and enteric drainage. Ann Surg 1995;221:613–24. 19 Newell KA, Woodle ES, Millis JM, et al. Pancreas transplantation with portal venous drainage and enteric exocrine drainage offers early advantages without compromising safety or allograft function. Transplant Proc 1995;27:3002–3. 20 Newell KA, Bruce DS, Cronin DC, et al. Comparison of pancreas transplantation with portal venous and enteric exocrine drainage to the standard technique utilizing bladder drainage of exocrine secretions. Transplantation 1996;62:1353–6. 21 Nymann T, Elmer DS, Shokouh-Amiri MH, Gaber AO. Improved outcome of patients with portalenteric pancreas transplantation. Transplant Proc 1997;29:637–8. 22 Nymann T, Hathaway DK, Shokouh-Amiri MH, et al. Patterns of acute rejection in portal-enteric versus systemic-bladder pancreas–kidney transplantation. Clin Transplant 1998;12:175–83. 23 Eubanks JW, Shokouh-Amiri MH, Elmer D, Hathaway D, Gaber AO. Solitary pancreas transplantation using the portal-enteric technique. Transplant Proc 1998;30:446–7. 24 Bruce DS, Newell KA, Woodle ES, et al. Synchronous pancreas-kidney transplantation with portal venous and enteric exocrine drainage: Outcome in 70 consecutive cases. Transplant Proc 1998;30:270–1. 25 Busing M, Martin D, Schultz T, et al. Pancreas–kidney transplantation with urinary bladder and enteric exocrine diversion: 70 cases without anastomotic complications. Transplant Proc 1998;30:434–7. 26 Busing M, Schultz T, Konzack J, Gumprich M, Bloch T. Simultaneous pancreas/kidney transplantation with portal venous and enteric exocrine drainage: First experience in Europe. Proceedings of the 7th World Congress of the International Pancreas and Islet Transplant Association. 1999;70:48 (abstract). IPITA, Sydney, Australia. 27 Buell JF, Woodle ES, Siegel C, et al. Portal-enteric drained simultaneous pancreas–kidney transplantation: To Roux or not to Roux? Proceedings of the 7th World Congress of the International Pancreas and Islet Transplant Association. 1999;57:18 (abstract). IPITA, Sydney, Australia. 28 Reddy KS, Stratta RJ, Shokouh-Amiri MH, Alloway R, Egidi MF, Gaber AO. Surgical complications after pancreas transplantation with portal-enteric drainage. J Am Coll Surg 1999;189:305–13. 29 Stratta RJ, Gaber AO, Shokouh-Amiri MH, et al. Allograft pancreatectomy after pancreas transplantation with systemic-bladder versus portal-enteric drainage. Clin Transplant 1999;13:465–72. 30 Philosophe B, Taylor JP, Schweitzer EJ, et al. Portal venous drainage in pancreas transplantation: Is there an immunologic advantage? Proceedings of the 7th World Congress of the International Pancreas and Islet Transplant Association. 1999;56:15 (abstract). IPITA, Sydney, Australia. 31 Philosophe B, Farney AC, Schweitzer EJ, et al. The superiority of portal venous drainage over systemic venous drainage in solitary pancreas transplantation. Proceedings of the 17th International Congress of the Transplantation Society. 2000;115:(AO330) (abstract). Rome, Italy.

R.J. STRATTA ET AL.

32 Petruzzo P, Da Silva M, Feitosa LC, et al. Simultaneous pancreas — kidney transplantation: Portal versus systemic venous drainage of the pancreas allografts. Clin Transplant 2000;14:287–91. 33 Cattral MS, Bigam DL, Hemming AW, et al. Portal venous and enteric exocrine drainage versus systemic venous and bladder exocrine drainage of pancreas grafts: Clinical outcome of 40 consecutive transplant recipients. Ann Surg 2000;232:688–95. 34 Zibari GB, Aultman DF, Abreo KD, et al. Roux-en-Y venting jejunostomy in pancreatic transplantation: A novel approach to monitor rejection and prevent anastomotic leak. Clin Transplant 2000;14:380–5. 35 Stratta RJ, Gaber AO, Shokouh-Amiri MH, et al. Experience with portal-enteric pancreas transplant at the University of Tennessee-Memphis. In: Cecka JM, Terasaki PI, ed. Clinical transplant 1998. Los Angeles: UCLA Tissue Typing Laboratory, 1000:239–53. 36 Stratta RJ, Gaber AO, Shokouh-Amiri MH, et al. A prospective comparison of systemic-bladder versus portal-enteric drainage in vascularized pancreas transplantation. Surgery 2000;127:217–26. 37 Stratta RJ. Pancreas transplantation. Prob Gen Surg 1998;15:43–65. 38 Gaber AO, Gaber LW, Shokouh-Amiri MH, Hathaway D. Percutaneous biopsy of pancreas transplants. Transplantation 1992;54:548–50. 39 Reddy KS, Stratta RJ, Shokouh-Amiri H, et al. Simultaneous kidney-pancreas transplantation without anti-lymphocyte induction. Transplantation 2000;69:49–54. 40 Stratta RJ, Gaber AO, Shokouh-Amiri MH, et al. Evolution in pancreas transplantation techniques: Simultaneous kidney–pancreas transplantation using portal-enteric drainage without anti-lymphocyte induction. Ann Surg 1999;229:701–12. 41 Lo A, Stratta RJ, Alloway RR, et al. Limited benefits of induction with monoclonal antibody to interleukin-2 receptor in combination with tacrolimus, mycophenolate mofetil, and steroids in simultaneous kidney–pancreas transplantation. Transplant Proc 2001;33:1701–03.

143

This page intentionally left blank

Chapter 12

Pancreas transplantation: early perioperative management Alan J. Koffron and Dixon B. Kaufman

Introduction The collective advances in cadaveric organ donor management, candidate pretransplant evaluation, graft preservation, immunosuppression, and surgical technique have all combined to improve longterm outcomes of pancreas transplantation. The careful management of recipients during the perioperative course is the initial determinant for optimizing outcomes, since significant morbidity early postoperatively can result in prolonged hospital admission and threaten a patient’s well being. This chapter provides an overview of the essential aspects of perioperative care with particular emphasis on the various potential early complications of pancreas transplantation.

Preoperative evaluation and management The final evaluation of the transplant candidate admitted immediately preoperatively is paramount to ensuring optimal patient and graft outcomes. While time is usually limited in an effort to minimize graft cold ischaemia time, a thorough recipient evaluation must not be taken with any less enthusiasm. The re-evaluation admission also allows time to review the sequence of transplant events with the patient and family members. Specific medical and surgical questions can be addressed. In addition, this is the appropriate time to obtain informed consent if the patient is to be included in any study protocols. The pancreas transplant recipient is admitted to the hospital, re-evaluated, and then a final decision made as to whether or not to proceed with transplantation. Because patients may be on the waiting list for years, significant progression of previously insignificant medical problems may have occurred. Careful assessment and comparison to previous diagnostic evaluation is pivotal in this regard. The reevaluation process is similar to that for other transplant recipients, emphasizing work-up for infectious disease or other acute medical issues that would contraindicate surgery. There are several special considerations for the diabetic patient. Careful management of diabetes pretransplant is important for patients not allowed to eat prior to surgery. Meticulous perioperative glucose management is obviously a primary concern. The basics of the preoperative evaluation are the history and physical examination. This defines and directs further evaluation and diagnostics. Each organ system susceptible to chronic complications of diabetes should be considered. This includes the nervous system (peripheral and autonomic), vascular tree (cardiac and peripheral vessels), and renal systems. Routine preoperative evaluation is detailed in Table 12.1.

146

EARLY PERIOPERATIVE MANAGEMENT

Table 12.1 Routine preoperative laboratory evaluation of the potential recipient Complete blood cell count with differential, platelets Electrolytes, glucose, amylase, liver function tests, lipids, blood urea nitrogen, creatinine Prothrombin time, partial thromboplastin time Blood type and crossmatch Gram stain and culture of indwelling venous or peritoneal catheters Electrocardiogram Chest X-ray Additional cardiopulmonary testing

The most important area of preoperative evaluation is that of cardiac disease and cardiac reserve. Suspicion of previously undiagnosed cardiac disease is obtained through a detailed history, specifically, enquiry about hospital admissions for congestive heart failure, infarction or other cardiac diseases should be made. If the history, physical examination, and electrocardiogram (ECG) raise the index of suspicion that significant progression of cardiovascular disease has occurred, then it may be necessary to proceed with additional testing to rule out conclusively significant coronary artery disease. If stress testing shows evidence of significant cardiac disease (either newly diagnosed or progressive from initial evaluation), cardiac catheterization should be considered and abnormalities treated prior to transplant. In most cases it may be prudent to proceed directly to coronary arteriography, then the transplant can proceed with postangiography dialysis to eliminate the contrast material in patients with endstage renal disease. If progression of cardiac disease has occurred that requires revascularization (cardiac bypass or angioplasty), then transplant surgery would be deferred. Careful pretransplant evaluation can help avoid difficult intraoperative decision-making or unanticipated postoperative limb ischaemia in patients with unsuspected severe peripheral vascular disease. Progression of peripheral vascular disease should be assessed by history and physical examination. Patients expressing complaints of significant symptoms attributable to peripheral vascular disease, or presenting with physical signs of vascular insufficiency may need urgent preoperative testing (non-invasive arterial duplex examination or angiography). Significant aortoiliac arterial disease should be repaired prior to or at the time of transplantation. Many potential pancreas recipients will undergo simultaneous renal transplantation for diabetic nephropathy. They should be treated in a similar fashion to kidney transplant candidates. It is necessary to first determine if dialysis is required prior to transplantation. This includes assessment of electrolyte, acid–base, and volume status. Adequate preoperative dialysis markedly simplifies perioperative management and minimizes the risk of hyperkalaemic crisis during surgery. A relatively dehydrated preoperative state allows for the liberal use of crystalloid and blood products during surgery. In addition, a well-dialysed patient will not require immediate dialysis, with its associated risks of heparinization, hypotension, and risk of renal allograft acute tubular necrosis (ATN), if the kidney allograft [in simultaneous pancreas kidney (SPK) recipients] fails to function immediately following transplantation. If the patient is on peritoneal dialysis, occult peritonitis should be quickly ruled out by Gram stain while the culture results of the peritoneal fluid are pending. In addition to routine preoperative assessment and preparation, consideration to the specific technique of pancreas transplant should be considered. A bowel preparation is routinely performed in patients that will undergo enteric drainage of the pancreas transplant.

A.J. KOFFRON AND D.B. KAUFMAN

The transplant procedure and anaesthetic considerations Preparation Appropriate preoperative intravascular lines are placed, as dictated by the overall health status of the recipient. All patients should have a central venous line placed to titrate intra- and postoperative fluids to a central venous pressure (CVP) of 12 to 14 cm H2O in order to ensure adequate perfusion of the transplanted organs. Arterial lines and Swan–Ganz catheters should be used if the cardiovascular status of the recipient is compromised or a difficult operative course is suspected. As a result of the widespread vascular disease of these patients, haemodynamic instability may result in myocardial or central nervous system injury. Therefore, all attempts to maintain recipient homeostasis should be employed, to avoid catastrophic consequences.

Anaesthesia Anaesthesia personnel play a critical role during pancreas transplantation. The surgeon and anaesthesiologist must work together and communicate pertinent information to one another. The surgeon must take an active role in helping manage fluid administration, correcting electrolyte and glycaemic abnormalities. Particular attention needs to be paid to the metabolic parameters of these patients, particularly acid–base balance, electrolytes and glucose levels. As large volumes of fluids are required, glucose-containing fluids should be used very sparingly or hyperglycaemia and an attendant hyperosmolarity may occur. It is reasonable to monitor blood glucose levels hourly. Maintenance of blood glucose levels as 120 to 160 mg per cent with intermittent administration of regular insulin is desired. Treatment of hyperkalaemia (serum potassium > 5.0 mg per cent) can be achieved by hyperventilating the patient, administration of bicarbonate, or starting a glucose/insulin/calcium cocktail to drive potassium ions into the intracellular compartment and protect the myocardium. At the time of reperfusion of the pancreas, bleeding may occur (particularly from the donor superior mesenteric vessels), requiring aggressive fluid administration and blood transfusion. Liberal use of colloid solutions (albumin, hetastarch) may help reduce the total crystalloid requirements, possibly avoiding severe facial/laryngeal oedema postoperatively. In addition, vascular isolation of the extremities (particularly in SPK patients where both lower extremities are affected) may cause severe metabolic acidosis. Preparation and treatment for this reperfusion phenomenon is essential to avoid haemodynamic instability. Following reperfusion of the pancreas, it is not uncommon for fluid requirements to exceed expectations for optimal haemodynamics. Though not proven, reperfusion of the pancreas allograft causes a septoid-like response akin to acute pancreatitis. Multiple medications are administered intraoperatively. Broad-spectrum prophylactic intravenous antibiotics should be started immediately preoperatively, certainly prior to the skin incision, redosed intraoperatively (given agent clearance and operative duration) and continued 3 to 5 days postoperatively, to diminish the incidence of postoperative wound infections [1]. Intraoperative diuretic therapy is often given to help diminish the likelihood or severity of ATN [2] and to promote early graft function. Induction immunosuppression with ornithine ketoacid aminotransferase 3, thymoglobulin, antithymocyte ␤-globulin and so on is often administered intraoperatively and can result in changes in respiratory parameters due to systemic cytokine responses associated with initial dosing. The operating room routine, with airway control, is an optimal setting in which to manage these temporary reactions.

147

148

EARLY PERIOPERATIVE MANAGEMENT

Postoperative management Very early postoperative management Immediate postoperative care begins in the postanaesthesia recovery room with airway management as highest priority, ensuring successful extubation and airway protection. Of note, many pancreas transplant recipients require large volumes of intraoperative fluid, which may expand the pulmonary interstitium and the upper airway and face. Although rare, substantial facial and laryngeal oedema may preclude early, safe extubation in some recipients. Vital signs are monitored frequently and a continuous oxygen saturation monitoring employed. A complete chemistry profile, complete blood count, coagulation survey, chest radiograph (to include analysis of intraoperatively placed lines), and ECG are typically obtained. Close observation and documentation of hourly urine output is critical to determine the early degree of initial function of the kidney transplant (in SKP transplant recipients) as well as anticipating the intravenous fluid replacement necessary. Urinary output may range widely and can be as low as drops or greater than 1 litre per hour. Postoperative fluid replacement must be thoughtfully approached. Assessment of volume status is important to avoid volume overload or depletion. Central venous pressure monitoring is a useful guide to intravascular volume status. If a brisk diuresis is occurring, it is not uncommon for electrolyte abnormalities to develop including hypokalaemia, hypocalcaemia, and hypomagnesaemia. Replacement of potassium in intravenous fluid must be approached with extreme caution. In those patients with voluminous urine output, the urinary concentration of potassium can be unexpectedly low. It is prudent to measure urinary potassium concentration prior to considering adding potassium to the intravenous fluids. The potassium concentration in the intravenous fluids should not exceed that in the urine. An abrupt cessation of brisk urinary output must be quickly assessed. Suspicion that the Foley catheter is occluded by a blood clot should prompt immediate irrigation. Importantly, an acute renal arterial thrombus will manifest as abrupt cessation of urine output. Very early vascular problems may be reversed and the kidney salvaged if acute renal artery thrombosis is suspected (usually in the recovery room) and the patient re-explored. Early significant postoperative bleeding would manifest as hypotension, tachycardia, decreased urine output, and lower than expected haemoglobin level. When the patient is stable and the early postoperative laboratory evaluation complete, the patient is typically transferred to the transplant general care unit. Routine intensive care unit observation is usually required, especially in those patients with medical comorbidities.

Postoperative management Initial postoperative management (Table 12.2) should be accomplished in a setting where there is a limited nurse to patient ratio, as the acuity of care is relatively high. Postoperative monitoring, including central venous pressure, ECG, blood pressure, and continuos oxygen saturation monitoring, at a minimum, is routinely followed. Initial intravenous fluids usually do not contain glucose but as the serum glucose drops below 120 mg/dl, this may be added to the replacement fluid. A useful fluid solution is 1 per cent dextrose/0.45 normal saline with 20 mmol sodium bicarbonate per litre. Diastolic blood pressures greater than 100 mmHg should be controlled with appropriate antihypertensive agents. It is important to note that patients with type I diabetes frequently have autonomic neuropathy resulting in recumbant hypertension and orthostatic hypotension. It is a good guideline to consider only sitting or standing blood pressures when making decisions regarding adjusting

A.J. KOFFRON AND D.B. KAUFMAN

Table 12.2 Postoperative management 24-h Electrocardiogram monitoring 24-h SpO2 monitoring 24-h arterial blood pressure, central venous pressure monitoring Swan–Ganz haemo- and oxyparameters as required Fluids to maintain central venous pressure > 10; replace urine output cm3 for cm3 Blood pressure control Laboratory studies glucose every 2 h arterial blood gases every 4 h and as required creatinine, K, Ca, Mg, PO4 every 6 h cardiac enzymes every 8 h (as indicated) daily serum amylase Nasogastric tube decompression

antihypertensive medications. We have observed that by postoperative day 5 only a minority of patients require antihypertensive therapy. A nasogastric (NG) tube is left in place in an attempt to avoid problems associated with gastric distention, gastroparesis, colonic dysmotility syndrome, and postoperative adynamic ileus. After the first 24 h postoperatively, the patient may be moved to a regular surgical floor setting, if medically appropriate. The NG tube is discontinued with return of bowel function, and oral intake is resumed.

Post transplant complications and surveillance Pancreas transplant recipients may fall victim to several common post-transplant complications in the early recovery period (48 h following transplant). Awareness, meticulous care, and surveillance can often prevent or promptly diagnose allograft and patient complications. Table 12.3 outlines the potential complications in the early postoperative period.

Thrombosis Early pancreatic allograft thrombosis (first week post-transplant) has historically been a frequent problem and occurred in up to 5 to 10 per cent of recipients. This is heralded by a sudden rise in serum glucose, and occasionally associated with pain directly over the pancreatic graft, or ipsilateral lower extremity swelling (venous thrombosis) or ischaemic pain (arterial thrombosis). Venous thrombosis is more frequent than arterial thrombosis. If this occurs, operative intervention (either thrombectomy and vascular revision, or graft excision) is indicated. Fortunately, as a result of improved preservation solution (University of Wisconsin solution), technical modifications, and widespread use of postoperative anticoagulation, this is currently an unusual complication. Anticoagulation therapy is used by many centres to reduce the incidence of pancreatic graft thrombosis. While there is no established dogma regarding the optimal anticogulation regimen, most centres employ a combination therapy approach. Intravenous heparin may be used alone, while others prefer Lovenox combined with an antiplatelet agent such as aspirin, Persantine, or Ticlid. The concern with anticoagulation therapy centres around the need to inhibit thrombosis yet simultaneously avoid postoperative haemorrhage by overaggressive therapy. Although this matter complicates postoperative patient care, the management of postoperative haemorrhage is more acceptible than

149

150

EARLY PERIOPERATIVE MANAGEMENT

Table 12.3 Potential early complications Thrombosis arterial, venous Haemorrhage pancreatic graft anastomotic Infection bacterial or fungal peripancreatic fluid superficial wound urinary tract Metabolic Acidosis hyperkalaemia, hypokalaemia, hypocalcaemia, hypomagnesaemia dehydration Gastrointestinal anastomotic leak (enteric-drained graft) mechanical obstruction Urological haematuria bladder anastomotic leak (bladder-drained graft) urethral injury/stenosis

allograft thrombosis (and attrition). Figure 12.1 provides a useful clinical algorithm to assist in postoperative patient management in this context.

Haemorrhage This complication is usual to any major vascular procedure. Ligatures or cut surfaces of the pancreastic graft may contribute multiple sites of bleeding. The use of postoperative anticoagulation, administered to reduce the frequency of graft thrombosis, also increases the risk of this complication. Clinical suspicion, serial blood counts, and attention to abdominal drain effluents often reveal postoperative haemorrhage. Frequently, discontinuation of anticoagulants/antiplatelet agents, correction of coagulation abnormalities by administration of platelets, DDAVP, vitamin K, fresh-frozen plasma, cryoprecipitate, and so on, and medical support is all the therapy that is required. Figure 12.1 details the clinical pathway dealing with postoperative haemorrhage. Early operative intervention (coeliotomy with evacuation and control of haemorrhage) should be considered to avoid haemodynamic instability and the resultant graft and recipient sequelae.

Infection Bacterial infection, either superficial or deep wound, can occur in up to 15 per cent of patients. Wound complications initially appear as superficial drainage. These are either simple incisional wound infections or more complicated pelvic or peripancreatic abscesses. Superficial wound infections must be opened and a sample of fluid submitted to microbiology for identification of the infectious organism and its sensitivity to antibiotics. Superficial infections can be treated successfully with local wound care. Wounds are generally allowed to close by secondary intention; in contrast, wound dehiscence requires urgent surgical repair. Complicated pelvic or peripancreatic abscesses can be

A.J. KOFFRON AND D.B. KAUFMAN

Fig. 12.1

approached in several ways. Appropriate antibiotics targeted at Gram-positive and Gram-negative organisms are used in all instances. Combined with these, well-defined fluid collections may be drained with percutaneous methods. Larger, loculated collections usually require surgical drainage and sometimes reoperation. Fluids should be carefully cultured as resistant bacteria and fungi may be identified requiring a change of the current antibiotic regimen. The increased risk of intra-abdominal infections may be increased because of peripancreatic fluid formation as a sequela of reperfusion injury of the pancreas allograft. Some groups routinely employ intra-abdominal closed-suction drainage to minimize the accumulation of peripancreatic fluid. Urinary tract infections are quire common in the pancreas recipient. There are specific reasons for this problem. Many of the diabetic recipients will have altered bladder function as a result of neuropathic changes and have significant residual volumes following voiding which promotes bacterial infection. Functional alterations as a result of surgical trauma to the bladder may also occur (pancreatic bladder drainage, ureteral implantation). Duodenal contamination may provide a bacterial nidus for infectious problems. In addition, alkaline and protein pancreatic secretions may support bacterial

151

152

EARLY PERIOPERATIVE MANAGEMENT

growth following bladder drainage of pancreas grafts. If significant urinary bladder dysfunction is suspected, urological evaluation and possibly self-catheterization may be indicated.

Metabolic This complication is induced by the procedure of draining the exocrine pancreas, especially to the bladder. The pancreas excretes large fluid volumes rich in bicarbonate. Enteric-drained pancreas recipients initially may have difficulty compensating for this added loss (in the context of postoperative bowel dysfunction), and bladder-drained recipients have continuing losses via the urine. In addition, immunosuppression often creates post-transplant bowel dysfunction in these patients, amplifying this situation. Careful monitoring of serum electrolytes and acid–base balance is necessary. Electrolyte and fluid repletion is frequently necessary to avoid dehydration (manifested by increased blood urea nitrogen and creatinine, and metabolic acidosis). Patients should be started on fluid and bicarbonate supplementation early and educated regarding this entity prior to discharge to prevent severe dehydration and possible graft loss.

Gastrointestinal Anastomotic leak (enteric-drained graft) is usually a devastating clinical scenario, and is the reason bladder drainage was popular in the early period of pancreatic transplantation. This occurs as a result of an ischaemic duodenal stump, technical errors, or duodenal stump blowout. Patients typically present with abdominal pain, fever, leucocytosis, and suspicious abdominal drain effluent character (high amylase content). Computed tomographic (CT) study of the abdomen is helpful but many recipients will have some degree of peripancreatic fluid collection, and early surgical intervention (and usually pancreatectomy) is pivotal in preventing widespread abdominal contamination, enzymatic damage, and sepsis. Mechanical obstruction is occasionally encountered in the context of SPK enteric-drained pancreas recipients. This is thought to occur due to internal herniation around the duodenojejunostomy, a fixed point in the abdominal viscera. Those patients with prolonged return of bowel function, persistently high NG tube volumes, and abdominal distention require evaluation (abdominal radiographs) to exclude this entity. Treatment of this complication is the same as non-transplant patients presenting with suspected mechanical bowel obstruction.

Urinary tract Haematuria is common in the immediate postoperative period and is usually self-limited. A friable, erythematous duodenal mucosa and a bleeding suture line at the duodenocystostomy or ureterocystostomy (in SPK recipients) may be responsible. Small amounts of haematuria require only close observation, but larger clots may need continuous bladder irrigation or direct cytoscopic evaluation and cautery. Bladder leaks can occur in the post-transplant period, although infrequently. These are usually a result of an ischaemic area along the either an anastomosis or the end of the transplanted duodenum (or ureter in SPK patients). The diagnosis of bladder leak begins with a high index of clinical suspicion backed by imaging studies. These may include an ultrasound or a CT can showing a fluid collection. The fluid is then percutaneously accessed and the amylase concentration (and/or creatinine concentration in SPK recipients) is compared to that in the serum. To localize the leak, a nuclear medicine study,

A.J. KOFFRON AND D.B. KAUFMAN

or a retrograde cystogram, may be useful. Asymptomatic bladder leaks can be treated with simple Foley catheter drainage, however, if accompanied by peritonitis, exploration and repair is necessary. There is an increased incidence of urethritis and urethral stenosis following pancreas transplantation. This is thought to be caused by an initial urethral injury (catheter placement) followed by secondary damage by activated pancreatic exocrine enzymes in bladder-drained graft recipients. Careful instrumentation and meticulous care may prevent this entity.

Rejection Although hyperacute rejection is essentially non-existent in pancreas transplantation, acute rejection continues to be a significant clinical problem. Acute pancreatic graft rejection, in the immediate posttransplant period, is difficult to diagnose because of ischaemia/surgically altered serum and urinary amylase levels. A high index of suspicion is needed in pancreas alone patients, and serum creatinine levels provide evidence of rejection in SPK recipients.

Summary The pancreas transplant recipient has the potential for enormous benefit through a successful transplant endeavour. For this to become reality, careful pretransplant re-evaluation and preparation is essential. Once transplanted, these recipients require meticulous medical care, with specific attention to those clinical pitfalls that these patients frequently encounter. Attention to detail during this early period will help assure that as many recipients as possible will find long-term relief from diabetes and its sequelae.

References 1 Tilney NL, Strom TB, Vineyard GC, Merrill JP. Factors contributing to the declining mortality rate in renal transplantation. N Engl J Med 1978;299:1321. 2 van Valenberg PL, Hoitsma AJ, Tiggeler RG, Berden JH, van Lier HJ, Koene RA. Mannitol as an indespensible constituent of an intraoperative hydration protocol for the prevention of acute renal failure after renal transplantation. Transplantation 1992;44:784.

153

This page intentionally left blank

Chapter 13

Surgical complications of pancreas transplantation Enrico Benedetti, Pierpaolo Sileri, Angelika C. Gruessner, and Luca Cicalese

Pancreas transplantation is currently considered an extremely effective therapeutic option for selected patients affected by type I diabetes. According to the International Pancreas Transplantation Registry 2000 Midyear Update [1], over 1400 pancreas transplants have been performed annually worldwide since 1995. In 1999, the majority of cases were simultaneous kidney and pancreas transplants (SPK, 86 per cent), followed by pancreas after kidney (PAK, 10 per cent), and pancreas transplant alone (PTA, 4 per cent). Current 1-year and 5-year patient survival rates exceed 90 and 80 per cent, respectively, without significant differences between the different procedures for cases performed between 1996 and July 2000. Pancreatic graft survival rates at 1 year were 84 per cent for SKP, 73 per cent for sequential PAK, and 70 percent for PTA. These excellent outcomes, much improved in comparison to older data, have been achieved due to a marked decrease in immunological graft loss and technical failures over the last 10 years. This chapter will focus on surgical complications after pancreas transplantation. The surgical challenge of pancreas transplantation mimics the well-known problems related to pancreatic surgery in general. Pancreatic grafts are susceptible to a unique set of surgical complications mostly related to the exocrine secretions and the low microcirculatory blood flow of the gland. While the incidence of surgical complications has markedly decreased over the years (Figs 13.1, 13.2), technical failures are still one of the leading cause of graft loss after pancreas transplant. A recent multivariate analysis of registry data has identified several risk factors for technical failures (Table 13.1).

Fig. 13.1 Overall technical failure rate after pancreas transplantation. Data from the International Pancreas Transplant Registry (IPTR): 2000 Midyear Update Report [1].

156

SURGICAL COMPLICATIONS

Fig. 13.2 Overall technical failure rate divided by specific recipient category. Data from the International Pancreas Transplant Registry (IPTR): 2000 Midyear Update Report [1]. Table 13.1 Risk factors for technical failures (1 January 1996 to 1 August 2000). Data from the International Pancreas Transplant Registry (IPTR): 2000 Midyear Update Report [1] variables

SPK

PAK

PTA

BD vs. ED DnrAge [45 yrs] Dnr Cause Death HLA MM A HLA MM B HLA MM DR Preservation Time [12 hrs] Recip.Age [45 yrs] Body Mass Index [25] Re-transpl. vs. Primary Transpl. Anti T cell Therapy FK506 MMF

*

*

** ** **

*** **

* ** **

** *

* *

*

** **

For a systematic review of technical complications after pancreas transplant, it is critical to summarize the current technical approach to this challenging procedure. Cadaver pancreas transplantation is routinely performed using the whole pancreas with a duodenal segment; the graft is always placed intraperitoneally. Revascularization is achieved using for arterial inflow the splenic and superior mesenteric arteries joined in various fashion (most commonly with a Y graft of donor iliac artery) while the portal vein is always used for venous drainage (either systemic or portal). A small number of cases is still performed using distal pancreatic segmental grafts vascularized by splenic vessels from living related donors. In the last 10 years, the exocrine secretions have been preferentially drained in the bladder via a duodenocystostomy, either hand-sewn or stapled, with good result [2]. However, the increased awareness of the specific complications associated with bladder drainage and the recent demonstration of the safety and effectiveness of enteric drainage [3], have reversed this trend. In 1998, 60 per cent of the pancreas transplants have been enterically drained either using a Roux-en-Y or a direct side-to-side anastomosis. By using enteric drainage, it is possible to avoid completely the significant urological and metabolic complications associated with bladder-drained grafts. Registry data, however, still suggest a small overall graft survival advantage and decreased incidence of techni-

E. BENEDETTI ET AL.

Fig. 13.3 Overall technical failure rates after pancreas transplant with bladder vs. enteric drainage. These data, obtained from the 2000 Midyear Update Report of the International Pancreas Transplant Registry (IPTR) suggest a small overall graft survival advantage and decreased incidence of technical complications in bladder drained versus enteric-drained grafts [1].

Fig. 13.4 Technical failure rates after bladder vs. enteric-drained pancreas transplantation divided by specific recipient category. Data from the International Pancreas Transplant Registry (IPTR): 2000 Midyear Update Report [1].

cal and immunological complications in bladder-drained versus enteric-drained grafts, especially for PTA and PAK (Figs 13.3, 13.4). It is possible to observe all the typical complications of major abdominal surgery in after pancreas transplant, such as prolonged ileus, bowel obstruction, cholecystitis, persistent gastroparesis, superficial and deep wound infection, and fascial dehiscence. However, this chapter focuses on surgical complications specific to the procedure, with special emphasis on those more likely to result in major morbidity and mortality as well as graft failure.

Vascular complications Vascular complications after pancreas transplant are more common than in any other solid organ transplant. Several factors contribute to this serious problem, including the complexity of the arterial reconstruction, the low microvascular pancreatic flow, and the frequent need to rely on collateral arterial flow to vascularize the head of the pancreas and the donor duodenum. The latter problem is due to the fact that the gastro duodenal artery is routinely ligated in liver–pancreas cadaver donors and the

157

158

SURGICAL COMPLICATIONS

superior mesenteric artery only, via the inferior pancreaticoduodenal artery, supplies the vascularization of the pancreatic head. Vascular complication may involve the intrinsic pancreatic circulation, the reconstructed arterial supply, or the recipient vessels. Furthermore, parenchymal pathology due to pancreatitis or rejection may increase the intrapancreatic vascular resistance to the point of causing secondary thrombosis of the main vessels. Thrombosis, haemorrhage, mycotic pseudoaneusysm, arterio-venous fistula, and recipient lower extremity ischaemia after pancreas transplantation have all been described.

Thrombosis Thrombosis is the most frequently observed surgical complication associated with pancreatic transplantation. Most recent registry data suggests a current rate of graft thrombosis of 6 per cent in SPK, 7 per cent in PAK, and 9 per cent in PTA: nearly half of the non-immunological graft failures. Usually, graft thrombosis is an early event that manifests itself within the first 2 weeks after transplant in nearly 90 per cent of cases [4]. The typical clinical presentation of pancreatic graft thrombosis in a patient with previously functioning graft is the sudden onset of hyperglycaemia. If the aetiology is portal vein thrombosis, haematuria is very common and can be the only sign for grafts with poor endocrine function. If the pancreatic graft is bladder-drained, disappearance or sudden decrease of urinary amylase may be diagnostic. Occasionally, the patient may complain of sudden onset of abdominal pain, especially after portal vein thrombosis that causes significant swelling of the thrombosed graft. The presence of graft thrombosis is easily confirmed by duplex ultrasonography while angiography may be occasionally used to confirm the diagnosis. Graft pancreatectomy remains the most common treatment after acute graft thrombosis and is usually unavoidable. Troppmann et al. have published a multivariate analysis of risk for vascular thrombosis in a cohort of 438 pancreas transplant recipients transplanted at the University of Minnesota between 1986 and 1994 [4]. In this series the overall rate of graft thrombosis was 12 per cent (5 per cent arterial, 7 per cent venous). An accurate regression analysis of numerous donor and recipient parameters identified several significant risk factors. Donor-related risk factors were age above 45 years and cause of death due to cerebrovascular causes. The best bench reconstruction resulted to be the donor iliac Y graft as opposed to a Carrel patch containing coeliac and superior mesenteric take-off or an interposition graft between splenic and mesenteric artery, both significant risk factors for arterial thrombosis. The presence of a portal vein extension graft and the left-sided implantation in the recipient both increased significantly graft thrombosis in PAK recipients, found to be at increased risk as a category. Finally, the presence of early graft pancreatitis was found to be a significant factor in solitary pancreas transplants. The latest registry data show a highly significant increase in thrombosis rate for PTA and PAK when enteric drainage is used; the difference is present but not statistically significant in SPK (Table 13.2).

Haemorrhage While minor bleeding can be treated conservatively with blood transfusion and correction of coagulopathy, intra-abdominal haemorrhage is a not infrequent cause of relaparotomy early after transplant. In a cohort of 142 pancreas transplant recipients, intra-abdominal bleeding resulted in 23 relaparotomies, for an incidence of 16 per cent [5]. An important risk factor is the use of anticoagulation in pancreas transplant recipient at high risk for vascular thrombosis. Registry data suggest that massive bleeding has been responsible for graft loss in 0.2 per cent of SPK, 0.3 per cent of PAK, and 0 of PTA in the last 5 years (Table 13.2).

E. BENEDETTI ET AL.

Table 13.2 Reasons for graft loss due to technical failures according to 2000 Midyear Update Report of the International Pancreas Transplant Registry (IPTR) [1] SPK

Gft Thr Infections Pancreatitis Anastomotic leak Bleed

PAK

PTA

BD

ED

p

BD

ED

p

BD

ED

p

5.4%

6.3%

0.317

4.9%

9.7%

0.029

4.8%

14.6%

0.008

0.5%

1.1%

0.077

1.2%

2.5%

0.254

0.0%

1.0%

0.232

0.4% 0.3%

1.0% 0.1%

0.050 0.323

0.0% 0.6%.

0.4% 0.0%

0.214 0.220

0.7% 0.0%

1.0% 0.0%

0.800 –

Mycotic pseudoaneurysms These are typically a consequence of intra-abdominal infection and may become symptomatic very late after transplant. The patient may present with generalized sepsis or more specific signs such as unilateral iliac vein thrombosis, loss of distal pulses in the affected extremity, tender or pulsatile mass, or, more dramatically, with massive intra-abdominal bleeding. Sometimes one or more ‘sentinel bleeding’ precede the major bleeding episode. Graft pancreatectomy, if not previously performed, is almost always necessary. Furthermore, the affected extremity required emergent revascularization, usually achieved with extra-anatomical bypass, such as axillofemoral of cross femorofemoral bypass. The index of suspicion for this rare but potentially fatal complication should be increased by history of intra-abdominal infection after pancreas transplant.

Other vascular complications These include arteriovenous fistulas, very often localized at the site of the ligation of the mesenteric vessels or temporary consequence of a traumatic pancreatic graft biopsy. This condition rarely requires operative management and may be successfully treated either with observation or with interventional radiological techniques. Finally, distal ischaemia of the limb homolateral to the pancreas transplant site has been described. While standard vascular surgery techniques can be successfully used for the revascularization, it is worth mentioning that donor iliac artery, if available, can be successfully used as a biological conduit if the iliac artery injury is detected intraoperatively [6].

Intra-abdominal infections The occurrence of intra-abdominal infection (IAI) after pancreas transplant is an extremely serious complication that reduces significantly both patient and graft survival. One of the major progresses in pancreas transplantation over the last 10 years has been a sharp reduction of the IAI rate, an achievement that has substantially contributed to the current success of the procedure. in the early days of pancreas transplantation a rate of 22 per cent of IAI with a mortality rate of 27 per cent was reported at the University of Minnesota [7]. Even in a more recent study covering 445 cases (45 per cent SPK, 24 per cent PAK, and 31 per cent PTA) performed during the ‘cyclosporin era’ (1986 to 1994), the same institution reported a 20 per cent rate of IAI, still with devastating effects on pancreas transplant outcomes [8]. In fact, the 1-year patient survival in recipients with IAI was only 76 per cent versus 92 per cent in non-infected patients while the 1-year graft survival was 17 per cent versus 63 per cent. A similar negative effect of IAI on pancreas transplant outcomes has been reported by other groups as

159

160

SURGICAL COMPLICATIONS

well [9,10]. More recent data, in pancreas transplant recipients treated mostly with FK506-based immunosuppression protocols, have demonstrated a significant improvement in IAI rate. In a multi centre trial involving 154 pancreas transplant recipients the incidence of IAI was only 3.9 per cent [11]. Recent registry data reports a rate of graft loss due to IAI of less than 1 per cent (Table 13.2). This extraordinary progress has been achieved with a combination of improved surgical technique, better bacterial and fungal prophylaxis, reduced rejection rate with newer immunosuppressive protocols, more careful donor selection and standardized surgical technique. A number of factors can contribute to the development of IAI after pancreas transplantation. Pancreatitis caused by reperfusion injury, contamination and bacterial translocation from the donor duodenum, anastomotic leaks, and graft necrosis have all been associated with the development of IAI. Furthermore, the combination of underlying long-term juvenile diabetes and the post-transplant immunosuppression increases the risk of IAI. In an extensive multivariate analysis of possible risks factor for IAI, Gruessner et al. demonstrated a significantly increased risk associated with older donor age, longer preservation time, older recipient age, enteric drainage versus bladder drainage, vascular graft thrombosis, and retransplant versus primary transplant [8]. IAI occur usually during the initial 90 days after pancreas transplant and manifest themselves almost always with abdominal pain, distension, and ileus [8]. If the patient does not need emergency exploration, abdominal computed tomography (CT) scan is the radiological technique of choice. Since IAI are associated with anastomotic leak in about 30 per cent of cases, the use of bladder contrast during CT scan is advisable in bladder-drained grafts. CT-guided percutaneous drainage of the fluid collection for bacterial and fungal culture as well as evaluation of the creatinine and amylase concentration in the fluids is a very valuable diagnostic and therapeutic tool. It is almost always worth leaving a drainage catheter within the fluid collection: in selected cases this may represent the definitive treatment and always stabilize the patient's condition. The perfusion of the graft should always be assessed by duplex ultrasonography: the incidence of IAI is particularly high after graft thrombosis, especially if the graft is not removed in a timely fashion. In about half of cases the infection is generalized to the entire abdominal cavity while in the remaining is limited to the peripancreatic area. If the patient fails to respond to conservative management, surgical exploration should be carried out without delay. In most cases multiple laparotomies may be necessary to control the infection. In over 50 per cent of cases graft pancreatectomy is indicated. Since IAI are life-threatening situation, the principle 'life before transplanted organ' should guide the decision-making process. The most common bacteria responsible for IAI are Staphylococcus species while Candida albicans is the most common fungal agent. Intra-abdominal fungal infections are especially dangerous since they are associated with higher morbidity and mortality than bacterial infection [12]. Proper fungal and bacterial prophylaxis should always be instituted after pancreas transplantation. Many transplant centres rely on fluconazole for fungal prophylaxis while the combination of imipenem or piperacillin with vancomycin is quite popular for antibacterial prophylaxis. IAI leading to graft loss are considered a relative contraindication to retransplantation because of the high risk of recurrent infection. Interestingly, the same pathogen agents that doomed the first transplant, even after numerous years almost always sustain the recurrent IAI [13]. The risk for IAI in SPK recipients is increased by pretransplant dialysis; therefore the ideal management for these patients should be a pre-emptive SPK [14]. If dialysis cannot be avoided, the modality selected does not influence the rate of post-transplant. The marked improvement of IAI rates after pancreas transplantation should be regarded as a brilliant success for the entire field of pancreas transplantation. A methodical analysis of the nature of the

E. BENEDETTI ET AL.

problem and the risk factors associated with it, together with significant progress in infection prophylaxis and immunosuppression, have drastically reduced the incidence of this serious complication.

Graft pancreatitis Early after pancreas transplantation graft pancreatitis is manifested by prolonged hyperamylasaemia and has been observed in as many as 35 per cent of pancreas transplant recipients [15]. This event can be related to donor factors such as haemodynamic instability and vasopressor administration, or to procurement and perfusion injury (excessive flush volume or perfusion pressure), as well as preservation and reperfusion injury. Pancreatitis is responsible for graft loss usually when complicated by IAI: the rate of this event is currently less than 0.5 per cent according to registry data (Table 13.2). The diagnosis of graft pancreatitis is based on the presence of hyperamylasaemia in combination with abdominal pain and distension, radiological evidence of pancreatic oedema, and inflammation by ultrasonography or CT scan. Sometimes the diagnosis is made during explorative laparotomy based on findings of pancreatic oedema and saponification. Complications of graft pancreatitis include pancreatic abscess, necrosis, perigraft infection, sterile pancreatic and peripancreatic fluid collection, and pseudocyst. Hyperamylasaemia simulating graft pancreatitis can be caused by anastomotic leaks and pancreatic ascites (due to leak or pancreatic necrosis with injury of the pancreatic duct). Late after pancreas transplantation, in bladder-drained graft, the most common cause of graft pancreatitis is urinary retention causing reflux pancreatitis. Placement of Foley catheter and judicious use of octreotide can usually achieve the cure of this relatively common condition. However, on occasion, enteric conversion of the graft is needed for recurrent reflux pancreatitis. Graft pancreatitis can be treated with the somatostatin analogue octreotide that is also effective as a prophylactic agent to decrease the incidence of technical complications by inhibiting the pancreatic graft exocrine secretion [15, 16].

Anastomotic leak The occurrence of an anastomotic leak is a highly morbid complication of pancreas transplantation. Almost invariably anastomotic leaks are complicated by intra-abdominal infection with the previously described serious implications. In a study from the University of Minnesota, Hakim et al. reported an incidence of duodenal leak of 10 per cent in bladder-drained pancreas transplant recipients [17]. In a recent report Pirsch et al. found the incidence of leaks in bladder-drained pancreas transplants to be 12 per cent while in enteric-drained graft was 5 per cent [3]. However, current registry data for SPK recipients suggest an incidence of 1 per cent of pancreatic leak causing graft loss in enteric-drained as opposed to 0.4 per cent in bladder-drained grafts (P = 0.05) (Table 13.2). It seems intuitive that leakage in the abdominal cavity of an enteric anastomosis may have more detrimental effect than a bladder anastomosis, whereby immediate decompression can be obtained simply by placing a Foley catheter. The debate about the best way to handle exocrine pancreatic secretion after pancreas transplant is still quite open. The clinical presentation is invariably characterized by abdominal pain and distension as well as fever in about half of the cases [17]. A rise in serum amylase can be documented in about half of the patients. Routine cystogram can be diagnostic in 82 per cent of cases but abdominal CT scan with bladder contrast injection is the gold standard since it can confirm the diagnosis in 96 per cent of cases. CT-guided drain placement within the fluid collection is essential to attempt conservative management and to confirm the diagnosis. Presence of high amylase or, for bladder-drained graft, elevated

161

162

SURGICAL COMPLICATIONS

creatinine in the drained fluid is suggestive of the diagnosis. In bladder-drained cases a positive culture for the same micro-organism in the urine and in the peritoneal fluid is almost pathognomonic. The location of the leak was in 36 per cent of cases the duodenocystostomy, in 28 per cent either the proximal or distal duodenal stump, and unidentified in the remaining. In the Minnesota series 12 patients with anastomotic leak out of 42 (29 per cent) were treated conservatively with indwelling Foley catheter and peritoneal drainage; of these seven underwent elective enteric conversion later on while the remaining five healed without further treatment. Therefore the success rate of conservative management alone for anastomotic leak in bladder-drained pancreas transplantation is quite low: only about 12 per cent. The operative management consists either in primary repair of the leak (possible in two-third of cases) or enteric conversion; if the graft does not appear viable, pancreatectomy may be indicated. Anastomotic leak from enteric-drained pancreas grafts may be quite challenging and can be rarely managed with conservative measures. Even the diagnosis is not easy, especially if a Roux-en-Y loop has been used for the anastomosis. Abdominal CT scan with oral contrast can demonstrate leaks from side-to-side duodenoenterostomy. High amylase content in the drained fluid adds further evidence to the diagnosis of the leak. Exploratory laparotomy with repair of the leak or repair of the duodenoenterostomy is often necessary.

Complications related to bladder drainage (see also Chapter 14) The management of the exocrine secretion after pancreas transplant is still an area of intense debate. Bladder drainage is a safe and well-established technique with the great advantage of allowing the monitoring of amylasuria as a marker for rejection and at present is the most common procedure for solitary pancreas transplant. However, for SPK the presence of a kidney graft from the same donor allows a simple and effective monitoring tool for rejection (serum creatinine). Therefore, especially for SPK, many transplant centres are presently switching to enteric drainage (over 60 per cent of cases in 1998). The rationale behind this trend is the occurrence of numerous urological and metabolic complications in bladder-drained pancreas transplant recipients. The continuous loss of fluids and bicarbonate leads to dehydration and metabolic acidosis. Although oral bicarbonate and increased oral fluid intake are often sufficient to control the majority of patients, enteric conversion is occasionally required to permanently resolve the problem [18]. Graft pancreatitis can occur as a consequence of back pressure from the bladder, which is unable to empty adequately because of functional or mechanical obstruction. The majority of reflux graft pancreatitis can be managed by short-term Foley catheter drainage, but recurrent pancreatitis may require enteric conversion [18]. The most serious complications of the procedure are urological, including anastomotic leaks, haematuria, recurrent urinary tract infections (UTI), and urethritis with possible urethral disruption. Ploeg et al. reported 192 episodes of urological complication in 121 patients in a series of 232 bladder-drained SPK transplants (52 percent), with an average of 1.6 episodes per patient [19]. Of these urological complications, 172 (90 per cent) were directly related to the pancreas transplant; recurrent UTI accounted for 35 per cent, chronic haematuria for 22 per cent, urinary leak for 22 per cent, and urethral lesions for 7 per cent of these complications. Over 50 per cent of recurrent UTIs can be managed with appropriate antibiotic therapy; however, some patients may require cystoscopic removal of foreign bodies such as suture material from the duodenocystostomy or, in rare cases, enteric conversion [20].

E. BENEDETTI ET AL.

Chronic haematuria can be controlled with cystoscopic fulguration in the majority of cases, but a few patients do require enteric conversion for permanent cure [20]. Urinary leaks may come from the duodenocystostomy or from either the proximal or the distal duodenal stump. Minor leaks can be managed by Foley catheter drainage, but failure of conservative management mandates operative intervention consisting of either primary repair or enteric conversion [18, 20]. Finally, urethral complications, although infrequent, can be very serious in this setting. The first reported case of enteric conversion of previously bladder-drained pancreas transplant was performed because of severe urethritis [21]. In 1986, Tom et al. reported a patient with a bladderdrained pancreas transplant who developed severe urethritis and inflammation of the glans penis that were unresponsive to standard therapy and corrected only by performing an enteric conversion [21]. High levels of trypsin (77 U/ml versus normal, 0.2 U/ml) and chymotrypsin (3290 U/ml versus normal, 15 U/ml) were demonstrated in the urine of this patient. Activation of proteolytic enzymes after pancreas transplant can be explained by the documented presence of enteropeptidase production in the donor duodenum or alternatively, as a consequence of ongoing UTI [22]. Elkhammas et al. reported six patients with 'dysuria syndrome' after bladder-drained pancreas transplantation [23]. Although all patients had documented extravasation from the bulbous urethra on voiding cystourethrogram, they were successfully treated by either suprapubic cystostomy or Foley catheterization for a period ranging from 3 to 6 weeks. In contrast, Stephanian et al. reported on seven patients with severe urethritis after bladder-drained pancreas transplant who failed conservative management and were successfully treated with enteric conversion [18]. These investigators, however, did not report on patients with dysuria who were successfully treated with conservative management. At the University of Wisconsin, nine of 12 patients who developed dysuria after bladder-drained pancreas transplant were eventually required to undergo conversion to control the symptoms [19]. Interestingly, urethritis after pancreas transplant has been reported almost exclusively in men with a history of severe dysuria [18, 23]. The only exception was a woman who required enteric conversion for severe urethritis in a University of Minnesota study [18]. Urethral instrumentation, either cystoscopy or Foley catheter placement, has been reported to result frequently in urethral perforation. Quite often a palpable, tender, perineal mass was present on physical examination of these patients [18]. All the above-mentioned complications can be successfully treated by enteric conversion. The current technique consists in a side-to-side ileoduodenostomy and several groups have documented the efficacy and safety of this approach [24, 25].

Conclusions Pancreas transplantation remains a formidable surgical challenge with numerous and serious technical pitfalls. However, the constant progress in surgical technique, infection prophylaxis, immunosuppressive strategies, as well as a better understanding of the problem, has allowed a remarkable improvement in the rate of technical complication. The procedure currently appears to be a safe and effective therapy for patients affected by terminal complications of juvenile diabetes.

References 1 Sutherland DE, Gruessner AC. International Pancreas Transplant Registry (IPTR): 2000 Midyear Update Report. University of Minnesota, Minneapolis, MN, USA.

163

164

SURGICAL COMPLICATIONS

2 Sollinger HW, Ploeg RJ, Eckhoff DE, Stegall MD, Isaacs R, Pirsch JD, et al. Two hundred consecutive simultaneous pancreas–kidney transplants with bladder drainage. Surgery 1993;114(4):736–43; discussion 743–4. 3 Pirsch JD, Odorico JS, D'Alessandro AM, Knechtle SJ, Becker BN, Sollinger HW. Post-transplant infection in enteric versus bladder-drained simultaneous pancreas–kidney transplant recipients. Transplantation 1998;66(12):1746–50. 4 Troppmann C, Gruessner AC, Benedetti E, Papalois BE, Dunn DL, Najarian JS, et al. Vascular graft thrombosis after pancreatic transplantation: univariate and multivariate operative and nonoperative risk factor analysis. J Am Coll Surg 1996;182(4):285–316. 5 Troppman C, Gruessner AC, Dunn DL, Sutherland DER, Gruessner RWG. Surgical complications requiring early relaparotomy after pancreas transplantation. Ann Surg 1998;227(2):255–68. 6 Benedetti E, Baraniewski HMN, Asolati M, Pollak R, Schuler JJ. Iliac reconstruction with arterial allograft during pancreas–kidney transplantation. Clin Transplant 1997;11:459–62. 7 Husse UJ, Sutherland DER, Najarian JS, Simmons RL. Intraabdominal infections in pancreas transplant recipients. Ann Surg 1986;203:153–62. 8 Gruessner RWG, Sutherland DE, Troppmann C, Benedetti E, Hakim N, Dunn DL, et al. The surgical risk of pancreas transplantation in the cyclosporine era: an overview. J Am Coll Surg 1997;185(2):128–44. 9 Ozaki CF, Stratta RJ, Taylor RJ, Langnas AN, Bynon JS, Shaw BW Jr. Surgical complications in solitary pancreas and combined pancreas–kidney transplantation. Am J Surg 1992;164(5):546–51. 10 Douzdjian V, Abecassis MM, Cooper JL, Smith JL, Corry RJ. Incidence, management and significance of surgical complications after pancreatic transplantation. Surg Gynecol Obstet 1993;177(5):451–6. 11 Gruessner RWG, Burke GW, Stratta R, Sollinger H, Benedetti E, Marsh C, et al. A multicenter analysis of the first experience with FK506 for induction and rescue therapy after pancreas transplantation. Transplantation 1996;61(2):261–73. 12 Benedetti E, Gruessner AC, Troppmann C, Papalois BE, Sutherland DE, Dunn DL, et al. Intraabdominal fungal infections after pancreatic transplantation: incidence, treatment, and outcome. J Am Coll Surg 1996;183(4):307–16. 13 Benedetti E, Troppmann C, Gruessner AC, Sutherland DE, Dunn DL, Gruessner WG. Pancreas graft loss caused by intra-abdominal infection. A risk factor for a subsequent pancreas retransplantation. Arch Surg 1996;131(10):1054–60. 14 Papalois BE, Troppmann C, Gruessner AC, Benedetti E, Sutherland DE, Gruessner RW. Long-term peritoneal dialysis before transplantation and intra-abdominal infection after simultaneous pancreas–kidney transplantation. Arch Surg 1996;131(7):761–6. 15 Stratta RJ, Taylor RJ, Lowell JA, Bynon JS, Cattral M, Langnas AN, et al. Selective use of Sandostatin in vascularized pancreas transplantation. Am J Surg 1993;166(6):598–604; discussion 604–5. 16 Benedetti E, Coady NT, Asolati M, Dunn T, Stormoen BM, Batholomew AM et al. A prospective randomized clinical trial of perioperative treatment with octreotide in pancreas transplantation. Am J Surg 1998;175(1):14–17. 17 Hakim NS, Gruessner AC, Papalois BE, Troppmann C, Dunn DL, Sutherland DER, et al. Duodenal complication in bladder-drained pancreas transplantation. Surgery 1997;121(6):618–24. 18 Stephanian E, Gruessner RG, Brayman K, Gores P, Dunn DL, Najaran JJ, et al. Conversion of exocrine secretions from bladder to enteric drainage in recipients of whole pancreatico-duodenal transplants. Ann Surg 1992;216:663–72. 19 Ploeg RJ, Eckoff DE, D'Alessandro AM, Stegall MD, Knechtle SJ, Pirsch JD et al. Urological complications and enteric conversion after pancreas transplantation with bladder drainage. Transplant Proc 1994;26:458–9.

E. BENEDETTI ET AL.

20 Sollinger HW, Sasaki TM, D'Alessandro AM, Knechtle SJ, Belzer FO. Indications for enteric conversion after pancreas transplantation with bladder drainage. Surgery 1992;112:842–6. 21 Tom WW, Munda R, First MR, Alexander JW. Autodigestion of the glans penis and urethra by activated transplant pancreatic exocrine enzymes. Surgery 1987;102:99–101. 22 Nghiem DD, Gonwa TA, Corry RJ. Metabolic effects of urinary diversion of exocrine secretions in pancreatic transplantation. Transplantation 1987;43:70–5. 23 Elkhammas EA, Henry ML, York JP, Tesi RT, Ferguson RM. Pancreas transplantation and dysuria. J Urol 1994;152:881–3. 24 Van der Werf WJ, Odorico JS, D'Alessandro AM, Knechtle SJ, Pirsch JD, Kalayoglu M et al. Enteric conversion of bladder-drained pancreas allografts: experience in 95 patients. Transplant Proc 1998;30(2):441–2. 25 West M, Gruessner AC, Metrakos P, Sutherland DE, Gruessner RW. Conversion from bladder to enteric drainage after pancreaticoduodenal transplantation. Surgery 1998;124(5):883–93.

165

This page intentionally left blank

Chapter 14

Medical and urological complications of pancreas and kidney/pancreas transplantation Phuong-Thu T. Pham, Phuong-Chi T. Pham, and Alan H. Wilkinson

Simultaneous pancreas and kidney transplantation (SPKTx) is a well-established therapeutic option for selected patients with endstage renal disease (ESRD) secondary to type I diabetes mellitus. Refinements in surgical techniques and important advances in immunosuppressive therapy have allowed patient and graft survival to reach levels comparable to those seen in kidney alone (KA) transplantation. Nevertheless, morbidity in pancreas transplants remains high, in part due to special problems that are inherent to pancreas transplantation. Variations in surgical techniques may result in different types and incidence of both surgical and medical complications. Urinary diversion of exocrine secretions of pancreas allografts may result in unique acid–base and electrolyte disorders, whereas systemic venous drainage of the endocrine system may lead to peripheral hyperinsulinaemia and peripheral insulin resistance. Similar to recipients of other solid organ transplants, pancreas transplant patients are at risk for developing complications related to long-term immunosuppression. These may include infectious complications, cardiovascular disease, dyslipidaemia, hypertension, bone disease, post-transplant lymphoproliferative disease (PTLD), and skin disease or skin malignancies. This chapter describes the medical and urological complications of pancreas transplantation both in the early postoperative period and in long-term post-transplant follow-up. Suggested preventive measures and management of complications are also presented.

Fluids, electrolytes, and acid–base disturbances Metabolic acidosis (Table 14.1) The predominant cause of metabolic acidosis among pancreatic transplants with bladder drainage (BD) of exocrine secretions has been attributed to the high volume of duodenopancreas allograft bicarbonate secretion. While metabolic acidosis has been reported to occur between 1 week and 40 months posttransplant, its severity is typically worst during the early post-transplant period and improves with time [1–3]. Generally, in patients with normal renal function, the acid load associated with the excessive urinary bicarbonate loss can be adequately compensated via enhanced renal distal tubular ammonium excretion and avid proximal tubular bicarbonate retention. The loss of renal function can therefore unmask the high urinary bicarbonate loss as a severe metabolic acidosis. Common causes of renal failure among recipients of SPK transplants with BD that may uncover the metabolic acidosis associated with urinary bicarbonate loss include renal allograft rejection, immunosuppressive therapy induced nephrotoxicity (tacrolimus, cyclosporin), and urological complications of renal allografts.

168

MEDICAL AND UROLOGICAL COMPLICATIONS

Table 14.1 Aetiologies of decreased serum bicarbonate in SPK recipients HCO3– loss

H+ retention Enhanced acid production

Bladder drainage of exocrine secretions Renal failure allograft failure drug nephrotoxicity urological complications (obstructive) Hyperkalaemia Volume depletion Sepsis Diarrhoea (antibiotic induced) Respiratory alkalosis acute allograft rejections pain pulmonary emboli pneumonia sepsis

+ + +

+ + +

+ + + + +

+

In an analysis of functional pancreas allografts beyond 30 days at the University of Cincinnati in 1987, Munda et al. reported the presence of chronic metabolic acidosis in eight out 10 patients, which was accentuated during periods of renal allograft dysfunction including acute tubular necrosis in one patient, acute renal rejection in two patients, and ciclosporin nephrotoxicity in three patients [3]. This observation was supported by a negative correlation between serum bicarbonate and serum creatinine (correlation coefficient = 0.726, P = 0.0075) [3]. Similarly, Nghiem et al. observed large urinary loss of bicarbonate with concomitant metabolic acidosis exacerbated during periods of renal dysfunction among patients with urinary diversion of exocrine secretions, a complication unique to their pancreaticoduodenal transplants with duodenocystomy as compared to duodenojejunostomy [4]. In a different report, Burke et al. described severe metabolic acidosis in five out of 106 patients who underwent whole pancreaticoduodenal transplants with BD, four of whom had renal allograft rejection [5]. Although the fifth patient did not have renal allograft rejection, he/she suffered from severe dysuria accompanied by pelvic pain only relieved by Foley catheter drainage. Subsequent conversion of the whole pancreaticoduodenal transplants from BD to enteric drainage (ED) successfully eliminated the severe metabolic acidosis in all five patients [5]. Despite advances in surgical technique and medical care, more recent studies continue to report high rates of postoperative complications with metabolic acidosis and dehydration among patients with exocrine BD especially when compared to ED. During the period of January 1987 to April 1995 at the University of Chicago, Bruce et al. reported 37 episodes of readmission for rehydration or metabolic acidosis among 50 patients with BD [1]. In contrast, Kuo et al. reported no readmission at the University of Maryland Medical System for metabolic acidosis among their 23 consecutive SPK with primary ED during the period from July to November 1995 [6]. During the period February 1997 to January 1998 at the University of Tennessee-Memphis, Stratta et al. reported 16 out of 16 patients with systemic BD versus seven out of 16 patients with portal ED required Hickman catheter insertion for dehydration/acidosis [7]. In a report from the University of Chicago, Newell et al. also revealed greater postoperative complications with metabolic acidosis [9/12 (75 per cent) versus 1/12 (8 per

P.-T.T. PHAM ET AL.

cent)] and dehydration [7/12 (58 per cent) versus 0/12 (0 per cent)] among patients undergoing pancreas transplantation with BD compared to portal venous and enteric exocrine drainage [8]. Beside its association with duodenopancreatic exocrine secretions of bicarbonate, metabolic acidosis has also been reported to occur despite marginal pancreas allograft function due to rejection. This observation raises the possibility that other mechanisms may play a role in the exacerbation of metabolic acidosis among pancreas transplant recipients independent of high exocrine secretions. We speculate that potential exacerbating factors in the maintenance of metabolic acidosis relevant to patients undergoing pancreaticonduodenal transplants include persistent hyperkalaemia, volume depletion, and urinary obstruction. Hyperkalaemia is known to reduce the production of ammonia, a molecule required for efficient urinary acid secretion. Hyperkalaemia may occur in association with progressive renal failure, ciclosporin toxicity, and urinary tract obstruction and has to be promptly treated as dictated by its underlying aetiology. In addition, volume depletion, a common problem accompanying multiple other metabolic complications among pancreas transplant patients, may also exacerbate metabolic acidosis. Under normal conditions, sodium delivery to the distal collecting tubule results in sodium reabsorption through sodium channels thereby creating a negative luminal charge favourable to acid and potassium secretion. Potassium secretion is also coupled with distal sodium reabsorption via a different mechanism. Following distal sodium reabsorption, intracellular potassium level is increased as potassium is exchanged for sodium via the basolateral Na+,K+-ATPase pump. The rise in intracellular potassium in turn creates a favourable chemical gradient for potassium secretion. Distal sodium delivery and good urinary flow are therefore important components in maintaining the electrochemical gradient necessary for appropriate acid and potassium secretions. When volume depletion is present, avid proximal sodium reabsorption limits distal sodium delivery, hence, the kidney's ability to generate a favourable electrochemical gradient necessary for efficient acid and potassium secretion. Conversely, when there is urinary tract obstruction and urinary stasis from any cause, dissipation of the electrochemical gradient can also limit both acid and potassium secretion and uncover the high urinary bicarbonate loss associated with BD of exocrine secretions. Interestingly, in a review of 75 consecutive recipients of combined kidney pancreas transplants with duodenocystostomy, Ketel et al. reported requirement for treatment of metabolic acidosis in 51 patients (67 per cent), of whom 40 per cent were ‘complicated by hyperkalaemia (K+ > 6 meq/l)’ [9]. In another review of 150 combined kidney pancreas transplantation with systemic venous and exocrine bladder drainage at the Ohio State University hospitals, Elkhammas et al. reported that 36 per cent of readmissions were due to one or all of a triad of dehydration, acidosis, or hyperkalaemia [2]. The cause and effect of hyperkalaemia and volume depletion on the reported metabolic acidosis were not analysed in either review. Although all current literature on pancreas transplants focus on metabolic acidosis as a complication in association with a fall in bicarbonate, it should be noted that a drop in bicarbonate may also occur as a compensatory change secondary to respiratory alkalosis. Among any solid organ transplant recipients, respiratory alkalosis may occur in association with various conditions including acute graft rejections, sepsis, pain, pulmonary emboli, and any acute pulmonary process and therefore, may herald a more ominous clinical course. In summary, metabolic acidosis is a well-documented complication associated with pancreaticoduodenal transplants with BD. despite this association, a blood gas should be performed to rule out any respiratory alkalosis when there is an unexplained fall in serum bicarbonate. The severity of the metabolic acidosis is generally worse during the early post-transplant period and improves with time [1–3]. Exacerbating factors include deteriorating renal function, hyperkalaemia, volume depletion, and urological complications. Treatment options include evaluation for and correction of any underlying exacerbating cause, bicarbonate supplementation (an average of 1 up to 4 mmol/kg/day) with or

169

170

MEDICAL AND UROLOGICAL COMPLICATIONS

without acetazolamide 250 mg orally twice a day [10], or surgical conversion of exocrine secretions from BD to ED [3,5,10]. The use of the carbonic anhydrase acetazolamide as a supplement to bicarbonate replacement therapy has been reported by several centres because of its ability to decrease bicarbonate secretion by pancreatic ductal cells [2,9,11] hence minimize subsequent urinary loss.

Volume depletion Patients who suffer the most from volume depletion appear to be those who receive pancreaticoduodenal transplants with BD due to the large urinary loss of sodium bicarbonate. As expected, volume depletion occurs more frequently during the first 3 postoperative months when metabolic acidosis is most problematic [9]. Most studies to date indicate a lower rate of readmissions for volume depletion among patients with ED compared to BD. The consequences of volume depletion may be deleterious as it may give rise to cardiovascular collapse, acute tubular necrosis, worsening of renal function, hence, more severe metabolic acidosis. In addition, profound orthostatic hypotension may occur due to residual diabetic autonomic dysfunction. Management includes intravenous fluids, bicarbonate supplementation, high salt diet, and fludrocortisone (Florinef) therapy.

Metabolic complications Hyperinsulinaemia Despite achievement of euglycaemia following pancreas transplantation, concerns arise with respect to hyperinsulinaemia due to its potential role as an independent risk factor, and its association with other well-accepted risk factors, for macrovascular disease. Hyperinsulinaemia was described in the early history of pancreas transplantation, but its aetiology was only speculative until the introduction of the more physiological pancreatic transplantation using portal drainage of endocrine secretions and ED of exocrine secretions [12]. In 1990, Diem et al. were among the first to analyse the presence and cause of hyperinsulinaemia in type I diabetic recipients of pancreas allografts with systemic versus portal drainage of endocrine secretions [13]. In this study, basal levels of insulin and C-peptide and the changes of these levels in response to intravenous glucose and arginine in 30 pancreas allograft recipients with systemic venous and portal drainage were determined and compared to eight similarly immunosuppressed nondiabetic kidney recipients and 28 non-diabetic control subjects. Basal insulin levels were lowest in control subjects (66 ± 5 pmol), followed by non-diabetic kidney recipients [77 ± 17 pmol, not significant (NS) versus control] and pancreas recipients (204 ± 18 pmol, P < 0.0001 versus control). Stimulated insulin levels were lowest in control subjects (416 ± 44 pmol), followed by pancreas recipients (763 ± 91 pmol, P < 0.01 versus control), and non-diabetic kidney recipients (589 ± 113 pmol, NS versus control). Basal and stimulated insulin levels in two pancreas recipients with portal venous drainage, however, were within normal range. Integrated acute C-peptide responses were not statistically different. Non-glucose stimulated insulin secretion with arginine revealed similar insulin and C-peptide results. The authors concluded that recipients of pancreas allografts with systemic venous drainage have greater basal and stimulated insulin levels compared to any of the group studied. In addition, both basal and glucose-stimulated insulin to C-peptide ratios in pancreas recipients were found to be significantly greater than those in control subjects. In general, insulin and C-peptide are synthesized in equimolar amounts in pancreatic ␤-cells. Whereas insulin undergoes a 50 per cent first-pass hepatic clearance, C-peptide is primarily metabolized by the kidneys and excreted in the urine. Given the greater insulin to C-peptide ratios, the authors speculated that there is an escape

P.-T.T. PHAM ET AL.

from first-pass hepatic insulin clearance rather than increased insulin secretion with iliac venous drainage compared to portal drainage [13]. Another mechanism suggested for hyperinsulinaemia in pancreas allograft trans-plantation involves absence of feedback inhibition for insulin secretion, a neurally-mediated process which is lost in the denervated transplanted pancreas [14]. Finally, immunosuppressive therapy induced insulin resistance with subsequent hyperinsulinaemia has also been suggested [13].

Insulin resistance Chronic hyperinsulinaemia with systemic drainage of endocrine secretions and long-term therapy with immunosuppressive agents such as prednisone have been suggested to contribute to insulin resistance among pancreaticoduodenal allograft recipients. Although the mechanisms of insulin resistance are still being determined, it has been suggested that insulin resistance is due to the reduction of peripheral glucose utilization rather than resistance to suppression of hepatic glucose production [15]. The site of insulin resistance has been proposed to be reduced insulin-stimulated non-oxidative glucose metabolism of peripheral tissues with resulting decreased capacity to store glucose as glycogen [15,16].

Urological complications Urological complications are generally more problematic among pancreaticoduodenal allograft recipients with BD. In a comparison study between 19 recipients of combined kidney pancreas transplantation with portal venous ED and 28 recipients of standard systemic bladder transplantation, Gaber et al. reported requirement for cystoscopic evaluation for urological complications in 53.6 per cent versus 0 per cent, respectively, at 6 to 42 month follow-up [12]. Common urological complications related to BD of exocrine secretions include haematuria, recurrent urinary tract infections (UTI), duodenal segment leak, dysuria, urethritis, and urethral stricture and disruption, and reflux pancreatitis.

Haematuria Haematuria is one of the most common urological complications occurring in 12 to 25 per cent [2,8, 17–21] of pancreaticoduodenal allograft recipients with BD. Haematuria has been classified as early and chronic [20]. In the first 2 to 3 day postoperative period, mild haematuria is almost a universal finding due to recent surgical manipulation of the bladder and mucosa of the duodenal allograft. Observation alone is generally sufficient. On occasion, severe postoperative haematuria may occur, and although rare, blood transfusions may be required. A potential cause of severe postoperative haematuria is a bleeding vessel at the anastomotic site. More frequently, however, no causal factor can be directly identified. Initial conservative management with frequent irrigation of the Foley catheter and correction of any existing coagulopathy is recommended. In most cases of persistent haematuria requiring cystoscopy, haematuria can resolve with simple clot evacuation with placement of a triple lumen Foley catheter and subsequent bladder irrigation. In cases where a bleeding vessel can be identified, fulguration has been reported to be effective [18,20]. Refractory haematuria may require enteric conversion [18]. Chronic haematuria has been described to occur from 4 weeks to over 4 years following transplantation [20]. Cystoscopy is generally required for diagnosis. The most commonly reported cause of chronic haematuria is ulcerations in the duodenal segment or granulation tissue in the anastomotic site and may be associated with cytomegalovirus (CMV) infection [17,20,21]. When no obvious source is identified with cystoscopy, a thorough evaluation of the entire urinary tract system with a

171

172

MEDICAL AND UROLOGICAL COMPLICATIONS

computed tomographic (CT) pyelogram is recommended to rule out potential sources from the combined transplanted kidney or native kidneys.

Urinary tract infections Urinary tract infections are discussed under infectious complications.

Bladder/duodenal leak Bladder/duodenal leaks have been reported to occur in 9.8 to 15 per cent of pancreas transplant recipients with standard systemic BD [2,17,18]. Bladder/duodenal leaks can also be classified into early (within the first 4 to 8 weeks) and late (after 8 weeks) postoperative leaks due to the different mechanisms and sites involved [20]. Whereas early leaks have been primarily attributed to primary anastomotic leak (poor closure of the suture line between the bladder and duodenal segment of the allograft), late leaks have been described to occur either at the lateral duodenal staple line or ulcerated/perforated areas in the duodenal segment [20]. Patients with bladder/duodenal leaks typically present with an acute and often severe lower abdominal pain, fevers, leucocytosis, and elevated serum amylase and creatinine [17,20]. Both voiding cystogram and cystograffin or technetium-99m voiding cystourethrogram (VCUG) have been advocated as diagnostic tools for bladder/duodenal leaks [20]. For small leaks, however, technetium-99m VCUG has been reported to be the most sensitive diagnostic test [20]. Early small leaks may be managed conservatively with 2 to 3 weeks Foley catheter placement and observation [17, 20]. With larger leaks, surgical exploration and repair or even enteric conversion may be required. For all patients with late leaks secondary to duodenal ulcerations, enteric conversion has been recommended [20].

Urethritis and urethral stictures Urethritis, a predominantly male complication [2,18,20], has been reported to occur in 3 to 8 per cent of pancreas transplant recipients with standard systemic BD [2,17,22]. Urethritis is presumed to occur secondary to the exposure of epithelial cells to digestive pancreatic enzymes [20] and/or increased urinary pH in association with large exocrine secretions of bicarbonate. Although urethritis is benign when treated early, it can evolve into urethral strictures and disruptions with delayed treatments. At initial presentation, urethritis may be alleviated with short-term (2 to 3 weeks) Foley catheter placement [2, 17, 20]. With persistent symptoms of urethritis, evaluation with a urethrogram and/or cystoscopy is required. Persistent symptoms generally dictate early enteric conversion to avoid urethral stricture and disruption [20].

Reflux pancreatitis Pancreatitis has been reported to occur in 11.4 to 52 per cent [1,8,12,18,20] of Systemic-bladder recipients compared to 5.3 to 25 per cent [8,12] of Portal-enteric recipients. The aetiology of pancreatitis in S-B patients is generally presumed to be due to urinary reflux and may be treated conservatively with Foley catheter drainage for several days. Reflux pancreatitis may be diagnosed in patients with acute pain over the pancreas allograft in association with elevated serum amylase, absence of urinary leak, oedema of the pancreas on CT, and resolution of symptoms within 24 h following Foley catheter placement [20].

P.-T.T. PHAM ET AL.

Infectious complications Incidence and types Infection is common and remains an important cause of morbidity following pancreas transplantation. An overall infection rate of 50 to 84 per cent over variable follow-up periods has been reported [7,23,24]. Within the first 3 months, infections commonly involve the urinary tract and surgical wounds including intra-abdominal abscesses, peritonitis, and infected fluid collections. In large series involving more than 300 BD pancreas transplant recipients, UTI occurred in 36 to 62 per cent of patients [17,21]. Since the resurgence of interest in ED pancreas transplants over the last half decade, there have been several reports comparing post-transplant infections in ED versus BD pancreas. In most series, the incidence of UTI was reduced by 50 to 60 per cent in ED compared to BD recipients [12,24]. Intra-abdominal infections occurred with similar frequency in both groups, with a reported incidence of 15 to 19 per cent [6,17,25,26]. The higher incidence of UTI in the BD pancreas have been attributed to alkalinization of urine, alteration of the mucosal integrity of the bladder caused by pancreatic enzymes, retained bladder suture, stone formation around the sutures, and prolonged Foley catheter drainage. In most series reported, the Foley catheter is removed on postoperative day 3 in ED pancreas and on day 12 to 14 in BD pancreas. Of interest, in a series of 500 SPKTx patients involving 388 BD and 112 ED recipients, Sollinger et al. found a significantly higher rate of intra-abdominal wound infections in patients who had previously undergone peritoneal dialysis [21]. It is speculated that this dialytic modality may be a risk factor for subsequent septic intra-abdominal complications. As with any other patient, prolonged use of vascular access devices, drainage catheters, indwelling stents or other surgical instrumentations, and extended intubation all predispose patients to the common nosocomial bacterial and candidal infections of the vascular access sites, surgical wounds, lungs, and urinary tract [27]. Other potential sources of infections peculiar to the transplant population include infection of the gastrointestinal tract and hepatobiliary system by immunomodulating viruses such as CMV, hepatitis B, and hepatitis C viruses (HBV, HCV) [27].

Aetiology The time to occurrence of different infections in any type of solid organ transplantation follows a ‘time table’ pattern [27]. Infections in the first month after transplantation are most frequently caused by bacterial micro-organisms. Although bacterial pathogens may vary from centre to centre, commonly isolated micro-organisms reported in pancreas or SPK transplant patients include aerobic Gram-negative bacterial (Escherichiae coli, Enterobacteriaceae, Pseudomonas) and aerobic Gram-positive bacteria (Enterococcus, Staphylococcus aureus, coagulase-negative Staphylococcus) [23,28,29]. After the first post-transplant month, infections with immunomodulating viruses including CMV, other human herpes virus, Epstein–Barrvirus (EBV), HBV, and HCV may occur either due to the overall state of immunosuppression, exogenous infection, or reactivation of latent disease. Repeated courses of antibiotics increase the risk of fungal infections whereas infections with immunomodulating viruses may render the patients more susceptible to opportunistic infections due to Pneumocystic carinii, Asperiggilus, and Listeria monocytogenes [27]. Beyond 6 months following transplantation, the risk of infection in patients with good allograft function is similar to that of the general population and include common respiratory viral infections and UTI. These patients are usually maintained on a relatively low level of immunosuppression. In contrast, patients who experience multiple episodes of rejection are the most likely candidates for chronic viral infections and superinfection to opportunistic organisms, usually as a result of repeated

173

174

MEDICAL AND UROLOGICAL COMPLICATIONS

exposure to heavy immunosuppression [30]. Causative opportunistic agents include Pneumocystic carinii, L. monocytogenes, N. cardia asteroides, Cryptococcus neoformans, Aspergillus, and geographicalrestricted mycoses (coccidioidomycosis, histoplasmosis, blastomycosis, and paracoccidioidomycosis [27,30–32]. Fishman et al. have advocated the use of life-long prophylactic therapy with trimethoprim-sulfamethoxazole in high-risk candidates. In addition, anti fungal prophylaxis should be considered and environmental exposure minimized [33]. Suggested prophylactic therapy in recipients of SPK is shown in Table 14.2.

Cytomegalovirus infection Cytomegalovirus infection occurs primarily after the first month post-transplantation and continues to be a significant cause of morbidity the first 6 month after organ transplantation. Cytomegalovirus infection may occur in the setting of primary infection in a seronegative recipient, reactivation of endogenous latent virus, or superinfection with a new virus in a seropositive recipient. Primary CMV infection often results in more severe disease than reactivation or superinfection. The clinical manifestations of CMV infection spans the spectrum of asymptomatic seroconversion, mononucleosis-like syndrome or flu-like illnesses with fever and leucopenia and/or thrombocytopenia, to widespread tissue invasive disease. The latter may result in clinical hepatitis, oesophagitis, gastroenteritis, colitis, retinitis, and pneumonitis among others. The transplanted organ appears to be more susceptible to the direct effects of CMV infection than are the native organ [27,33]. Hence, clinically significant hepatitis is more commonly seen in liver allograft recipients, CMV pancreatitis in pancreas allograft recipients, and severe CMV pneumonitis in lung or heart-lung transplant recipients. Cytomegalovirus-related bleeding ulcer from the duodenal segment in ED pancreas transplant has been described [34]. Donor and recipient seropositive status, and the use of blood products from CMV seropositive donor, are well-established risk factors for CMV infection. Other factors associated with an increased risk of CMV infection include the use of antilymphocyte antibodies, prolonged or repeated courses of antilymphocyte preparations [35], and episodes of allograft rejection. Although the cause/effect of allograft rejection and CMV infection remain largely conjectural, several studies have suggested a bidirectional relationship mediated via inflammatory cytokines [27,31,33,36,37]. Management of CMV infection consists of preventive (prophylactic and/or pre-emptive therapy) and therapeutic measures. Over the last several years, various prophylactic and pre-emptive protocols Table 14.2 Suggested prophylactic therapy for recipients of SPK transplants Comments Trimethoprim-sulfamethoxazole (TMP/SMX) (80/400 mg) one tablet q.d. × 3 months

Its routine use reduces or eliminates the incidenceof PCP, L.monocytogenes, N. asteroides, and T. gondii In renal transplant recipients, TMP/SMX reduces the incidences of UTI from 30–80% to less than 5–10%

Bimonthly intravenous or aerosolized pentamidine, dapsone, or atovaquone*

Replaces TMP/SMX for patients with sulfa allergies

Fluconazole 200 mg one tablet q.d. × 2 months

Close monitoring of CSA/Tac levels when starting and stopping antifungal agents

Aciclovir/Ganciclovir

For CMV prophylaxis, see Table 14.3.

*

In order of efficacy.

PCP, Pneumocystis carinii.

P.-T.T. PHAM ET AL.

have been developed to reduce the rate of CMV infections. Observations drawn from various studies reveal that oral aciclovir provides effective CMV prophylaxis solely in recipients of seronegative donor organ(s). Oral/intravenous ganciclovir provides superior prophylactic/pre-emptive therapy against CMV reactivation [38–40]. The routine use of CMV hyperimmune globulin and intravenous immune globulin result in no added benefit compared to a regimen consisting of antiviral agents alone. Its use should probably be limited to high-risk candidates such as during antilymphocyte therapy in recipients with primary CMV exposure [41,42]. Suggested CMV prophylaxis protocol is shown in Table 14.3. Although there is no well-defined protocol for the prevention of CMV infection, therapy should be directed based on the intensity of immunosuppression, the risk of reactivation such as during antilymphocyte antibody therapy, and the seropositive status of the donor and/or the recipient. Clinical CMV disease should be treated with intravenous ganciclovir and continued until clearance of viraemia as assessed by antigenaemia or polymerase chain reaction (PCR). The usual course consists of 2 to 4 weeks of intravenous therapy. In patients with tissue invasive disease, Fishman and Rubin advocate the use of intravenous ganciclovir followed by 2-month course of oral ganciclovir in seropositive individuals and for 3 to 4 months in those with primary infection [27]. Relapse may develop as a result of premature discontinuation of intravenous therapy [27,33]. The use of oral ganciclovir in the presence of high viral load may confer ganciclovir-resistance due to low bioavailability achieved by oral ganciclovir therapy.

Post-transplant bone disease Osteoporosis Post-transplant decline in bone marrow density (BMD) is most pronounced in the first 6 months and correlates with higher glucocorticoid exposure in the early post-transplant period. The rate of bone loss varies from 3 to 10 per cent and is most apparent at sites of cancellous bone, particularly the lumbar spine or axial skeleton [19,32,43–46]. This early rapid decrease in BMD is usually followed by

Table 14.3 CMV prophylaxis protocol CMV (–) recipients of a CMV (–) organ Aciclovir 400 mg q.h.s. for 3 months CMV (–) recipients of a CMV (+) organ During antibody treatment DHPG 2.5 mg/kg i.v. b.i.d., then following antibody treatment Cytovene 1000 mg p.o. t.i.d. for 3 months adjusted to renal function If no antibody treatment: Cytovene 1000 mg p.o. t.i.d. for 3 months CMV (+) recipients of a CMV (–) organ During antibody treatment DHPG 2.5 mg/kg i.v. b.i.d. then following antibody treatment Cytovene 1000 mg p.o. t.i.d. for 3 months adjusted to renal function If no antibody treatment: aciclovir 400 m.g. q.h.s., order CMV DNA q. 2 weeks × 3 CMV (+) recipients of a CMV (+) organ During antibody treatment: DHPG 2.5 mg/kg i.v. b.i.d. then following antibody treatment Cytovene 1000 mg p.o. t.i.d. for 3 months adjusted to renal function If no antibody treatment: aciclovir 400 mg q.h.s., order CMV DNA q. 2 weeks × 3 DHPG = 9-(1,3-dihydroxy-2-propoxymethyl) guanine.

175

176

MEDICAL AND UROLOGICAL COMPLICATIONS

a slower rate of bone loss and is related to the cumulative steroid dose. A decrease in BMD averaging 1.7 per cent per year in later post-transplant years has been reported [46]. Reduced bone mass is found in nearly all transplant patients within 5 years of transplantation; however, recovery or even an increase in BMD has also been described. The most important effect of glucocorticoids on the skeletal system involves direct inhibition of osteoblastogenesis and induction of apoptosis of osteoblasts and osteocytes [47]. Other adverse effects of glucocorticoids on bone mechanical integrity include inhibition of intestinal absorption of calcium, enhancement of renal excretion of calcium, secondary hyperparathyroidism, and direct suppression of gonadal hormone secretion. Experimental animal models suggest that ciclosporin and tacrolimus may also contribute to bone loss following transplantation by stimulating bone resorption [48]. However, the effect of ciclosporin and tacrolimus in human subjects remains speculative, as these agents are usually used in conjunction with glucocorticoids. Furthermore, bone histology in renal transplant recipients fails to demonstrate increased osteoclast-mediated bone resorption. Interestingly, recent studies have suggested a beneficial effect of ciclosporin on bone remodelling which may counterbalance corticosteroid-induced osteopenia and osteoporosis [49].

Osteonecrosis Osteonecrosis occurs in 6 to 8 per cent of patients in the first few years after transplantation. The most commonly affected bone is the femoral head and neck, first described in 1964 in renal transplant recipients [50]. Avascular necrosis is now a well-recognized complication following organ transplantation. It may affect other weight-bearing bones and the humeral head. Early avascular necrosis of the femoral head commonly presents with hip pain and/or referred knee pain. Symptoms may be aggravated by weight bearing but may also be paradoxically worse at night. magnetic resonance imaging is the most sensitive technique for early detection. Plain radiographs are of limited diagnostic value in the early stage. Predisposing factors for the development of avascular necrosis include greater exposure to intravenous methylprednisolone, low bone mass, hyperparathyroidism, and a history of local trauma.

Bone fracture Despite corrections of many of the metabolic, bone, and mineral abnormalities associated with diabetes mellitus and renal failure following successful kidney/pancreas transplantation, the incidence of fracture is still increased compared to the general population. It is speculated that insulin-dependent diabetes mellitus and kidney/pancreas transplantation per se are important risk factors for fractures. The incidence of fractures has been reported to occur in 2 per cent per year in non-diabetic renal transplant recipients, in 5 per cent per year in patients with preexisting diabetes mellitus, and in 12 per cent per year in recipients of kidney/pancreas transplantation [51–53]. The increased risks of fractures in diabetic patients, particularly of the appendicular skeleton, may result from a complex interplay between greater steroid exposure, pre-existing cortical osteopenia or low bone turnover disease associated with insulin-dependent diabetes mellitus, diabetes-related bone deformities, peripheral neuropathy associated stress fractures, and post-transplant persistent hyperparathyroidism. Predisposing risk factors for fractures in transplant patients are similar to the general population and include pre-existing severe osteoporosis, hypogonadism, and postmenopausal state. The risk of fractures in transplant recipients, however, has been shown to be independent of age. Interestingly, a high prevalence of limb fractures, particularly foot fractures, has been reported following SPKTx. In a cohort study consisting of 600 recipients of solid organ transplant, fracture involving the limb was found to be universal among SPKTx recipients. While an unexpectedly higher incidence of foot fracture was recorded among KA transplant recipients, both axial

P.-T.T. PHAM ET AL.

and limb fractures occurred at similar frequency among liver and heart transplant recipients [52]. In another study designed to assess the long-term outcome of kidney/pancreas transplantation, Bruce et al. reported that over half of the limb fractures observed occurred in the foot [1].

Management of post-transplant bone disease Management of post-transplant bone disease has largely been based on studies involving postmenopausal osteoporosis and glucocorticoid-induced osteopenia in non-transplant settings. Comparable data in organ transplant recipients are currently limited. Nevertheless, management should be directed at reducing risk factors. Early ambulation and physical exercise should be encouraged. Caution should be exercised to prevent stress fractures in diabetic patients with peripheral neuropathy. Adjunctive preventive/therapeutic measures include vitamin and mineral supplements, anti bone resorptive therapy, and hormonal replacement therapy.

Vitamins and minerals Adequate calcium and vitamin D analogue supplementation is generally recommended following transplantation to prevent rapid bone loss in the first post-transplant year. Recognition and correction of the hypocalcaemic and hypercalciuric effect of corticosteroid should be an integral part of post-transplant management. In long-term stable transplant recipients, corticosteroids should be kept at a safe minimum. In patients with pre-existing secondary hyperparathyroidism due to endstage renal failure, hypercalcaemia, and hypophosphataemia may develop following a successful renal transplant due to the combined effect of persistent hyperparathyroidism and high calcitriol level. A conservative therapeutic approach with phosphate supplement is generally recommended. Adjunctive therapy with 1,25 vitamin D (Rocaltrol) has also been suggested to be beneficial. Whether 1- ␣-hydorxyvitamin D2 (Hectorol), a vitamin D analogue, may be advantageous in post-transplant residual secondary hyperparathyroidism and hypercalcaemia is currently not known and awaits studies. This newer agent has been shown to be effective in treating secondary hyperparathyroidism in dialysis patients and more favourable compared to other vitamin D analogues due to their lower hyperphosphataemic and hypercalcaemic effects [54]. Parathyroidectomy is indicated in patients with tertiary hyperparathyroidism or persistent severe hypercalcaemia (> 12.5 mg/dl for more than a year), symptomatic hypercalcaemia, nephrolithiasis, persistent metabolic bone disease, calcium-related renal allograft dysfunction, or vascular calcification and calciphylaxis [55].

Antibone resorptive agents Biphosphonates Biphosphonates inhibit bone resorption and have been shown in increase bone mineral density in post menopausal women and in glucocortocoid-induced osteoporosis, particularly at the lumbar spine and trochanter [56]. Its use has been extended to post-transplant bone disease. Fan et al. first conducted a prospective, randomized, and placebo-controlled trial to assess the effectiveness of pamidronate in preventing bone loss in the immediate post-transplant period [57]. In 14 recipients of renal transplant who were randomized to receive biphosphonate treatment, intravenous pamidronate was given at the time of transplantation and at 1 month postoperatively. During the 12month follow-up, a 6.4 per cent reduction in BMD at the lumbar spine (P < 0.05) and a 9 per cent reduction in BMD at the femoral neck (P < 0.005) were observed in the placebo-treated group (n = 12) whereas no significant changes were observed in the pamidronate-treated group [57]. Despite these favourable results, the use of biphosphonate in preventing bone loss and its impact in reducing

177

178

MEDICAL AND UROLOGICAL COMPLICATIONS

the incidence of fracture following transplantation remains to be established. Nevertheless, biphosphonate therapy may be justifiable in potential high-risk candidates including patients with pre-existing fractures or severe osteoporosis, patients with diabetes mellitus, postmenopausal women, and recipients of kidney/pancreas transplants [53].

Calcitonins Calcitonin inhibits osteoclastic action by a direct receptor-mediated signal. Similar to biphosphonates, calcitonin has been shown to be effective in postmenopausal as well as in glucocorticoid-induced osteoporosis. There are limited data on the use of calcitonin in recipients of organ transplants. In a small series consisting of 16 kidney transplant recipients with established osteopenia and osteoporosis [BMD ≤ 1.5 standard deviation (SD) below normal], calcitonin has been shown to increase lumbar spine BMD by 3.2 per cent (n = 16, P = 0.034); of note, however, BMD was also increased by 1.8 per cent in the control group (n = 15, P = 0.265) and the difference in BMD changes among the groups were not statistically significant [58]. Whether anti bone resorptive agents may have an adverse effect on bone remodelling and subsequent fracture risk following transplantation is unknown due to the complex nature of posttransplant bone disease. Large-scale clinical trials with long-term follow-up are still needed.

Hormone replacement Unless contraindicated, oestrogen replacement therapy should be prescribed in postmenopausal women. In patients with an intact uterus, progesterone must also be given to prevent endometrial cancer. Testosterone deficiency has also been implicated in the development of bone loss and fractures following transplantation. However, the role of hormonal replacement therapy in men with hypogonadism is less well defined [59] and treatment should be tailored to the individual patient. Attention should be given to potential side-effects of androgen therapy including hyperlipidaemia, hepatic enzyme abnormalities, and prostatic hypertrophy.

Post-transplant macrovascular disease Despite risk factor modification including restoration of renal function and achievement of euglycaemia, macrovascular disease remains an important cause of morbidity and mortality after SPK transplantation during long-term follow-up. In one of the world's largest series of SPK transplants involving 500 SPK recipients followed over a 12-year period, cardiovascular events including myocardial infarction (MI), cardiac arrest, and arrythmias were found to be the leading cause of death (38 per cent), followed by infection (17 per cent), malignancy (9 per cent), and cerebrovascular accident (CVA) (5.5 per cent) [21]. Studies on the impact of pancreas transplantation on the progression of cardiac, cerebrovascular, and peripheral vascular disease (PVD) have yielded variable results. The literature on the incidence of macrovascular complications in recipients of SPK compared to that of KA in uraemic diabetic patients will be reviewed followed by a discussion on the potential risk factors for post-transplant macrovascular disease. In a retrospective study consisting of 39 SPK transplants and 65 consecutive diabetic patients who received KA during the same period, Morissey et al. reported a higher incidence of peripheral vascular complications (PVC) in SPK recipients despite fewer risk factors in the pretransplant period [60, 61]. Peripheral vascular complications, defined as any mid foot or limb amputation, ischaemic ulceration requiring treatment, and lower extremity bypass surgery or angioplasty, were comparable between groups prior to transplantation. Pancreas transplant was performed with BD of exocrine secretions and systemic delivery of insulin. During a mean follow-up of more than 4 years, 35 new episodes of PVC were observed in 18 of 39 SPK recipients (P = 0.005). During the same period, 32 new episodes

P.-T.T. PHAM ET AL.

of PVC developed in 20 of 65 KA transplant recipients (P = 0.005). overall, 46 per cent of SPK recipients experienced clinically significant PVC compared with 31 per cent of an older cohort of diabetic patients undergoing KA transplant. The authors concluded that a functioning pancreas does not alter the progression of PVD and may even accelerate peripheral vascular complications [60,61]. Knight et al. reported similar incidence of PVC in type I diabetic patients undergoing SPK transplant compared to a cohort of KA transplant recipients [62]. Whole organ pancreas transplant with portal-systemic venous drainage was performed in all cases. Mean follow-up for the SPK and KA groups were 22 ± 13 months and 22 ± 12 months, respectively. During follow-up, two of 20 SPK recipients underwent major amputation 1 year post-transplantation. Vascular disease was present in the pretransplant period in both patients. A third patient underwent a single toe amputation 1 year post-transplant for the treatment of infection. Among the KA cohort, two of 17 patients suffered major vascular complications within 3 months of transplantation requiring below knee amputation (BKA) in one and revascularization procedure followed by a toe amputation for advanced ischaemic vasculopathy in the other. In both patients, vascular disease was present prior to transplantation. A third patient suffered a toe infection that was successfully treated with antibiotics. The authors concluded that the addition of a pancreas transplant does not alter the progressive nature of PVD in diabetic patients with a history of pretransplant vascular complication [62]. In a study conducted by Biesenbach et al., the incidence of progressive macroangiopathy in SPK recipients was comparable to that of uraemic diabetic patients undergoing KA transplant [63]. Segmental pancreas transplantation with systemic BD was performed in all cases. The mean observation period was 69 ± 37 for SPK recipients and 70 ± 33 months for diabetic patients with KA transplants. During follow-up, progression of cerebrovascular disease (CVD) and coronary heart disease (CHD) was observed in four of 11 SPK recipients (30 per cent). Five of 11 SPK recipients (45 per cent) showed progression of PVD. In the cohort with KA transplant, progression of CVD and CHD occurred in four of 10 patients (40 per cent) and progression of PVD was observed in five of 10 patients (50 per cent). The incidence of macroangiopathic complication was not significantly different between SPK and KA transplants despite amelioration of risk factors in patients with a functioning pancreas allograft. The mean values of haemoglobin A1c (HbA1c) and serum triglyceride (TG) were significantly lower in patients with SPK transplants than in patients with KA transplants (P < 0.001) [63]. In a series consisting of 335 SPK transplant performed at the University of Wisconsin between 1966 and December 1995, Becker et al. showed no significant differences in cardiac death rates between SPK recipients and age-matched diabetic cadaveric kidney (DM-cadaveric, n = 147) or diabetic living donor kidney recipients (DM-live donor n = 160). However, age stratifying the patient cohorts revealed significantly fewer cardiac deaths in SPK recipients (n = 202) compared to DM-live donor (n = 59, P = 0.005 versus SPK) and DM-cadaveric (n = 57, P = 0.004 versus SPK) recipients between ages 30 and 39. Long-term survival as measured by the ratio of observed/expected life expectancy was significantly increased in SPK transplant compared with age-matched DM-cadaveric and DM-live donor kidney transplant recipients. This study represents the largest series reported with a mean follow-up of 13 ± 5.9 years [64]. Although studies on the impact of pancreas transplantation or the progression of macrovascular disease have yielded variable and even conflicting results, well-established risk factors for atherosclrotic vascular disease include: hyperinsulinaemia, systolic and diastolic hypertension, dyslipidaemia, diabetes mellitus, and obesity. These factors may either be present in the pretransplant period and persist post-transplantation, or may develop due to transplantation-related complications. As reversal of microvascular complications with pancreas transplantation has been shown to require prolonged

179

180

MEDICAL AND UROLOGICAL COMPLICATIONS

periods of euglycaemia [65,66], long-term beneficial effects of SPKTx on macrovascular disease may not be apparent unless studies with longer follow-up are done.

Hyperinsulinaemia Systemic venous drainage of pancreas allograft results in peripheral hyperinsulinaemia and insulin resistance which are independent risk factors for CVD and coronary artery disease. Portal-drained pancreas transplant eliminates systemic hyperinsulinaemia by restoring the physiological first-pass hepatic clearance by insulin. Whether this effect translates into fewer macrovascular complications in pancreas transplants remains to be determined. There is currently no data comparing cardiovascular complications in recipients of portal versus systemic drained pancreas transplantation. In addition to complications inherent in the surgical techniques, corticosteroid therapy has also been suggested to be an important contributing factor in post-transplant insulin resistance.

Hypertension Successful SPK transplantation has been reported to lower the incidence of hypertension by restoring renal function and euglycaemia. Nevertheless, hypertension continues to be a common complication following SPK transplantation with a reported incidence of 49 to 89 per cent [67–69]. Pre-existing hypertension, ciclosporin and tacrolimus immunosuppression, and corticosteroid therapy have all been implicated in the development of post-transplant hypertension. The use of large quantities of sodium bicarbonate to correct metabolic acidosis in bladder drained pancreas allograft has also been suggested to contribute to post-transplant hypertension. In a series of 25 SPK recipients, Raja et al. found a higher sodium bicarbonate dosage among hypertensive SPK recipients compared to their normotensive counterparts (122 ± 11.3 mmol/day versus 96 ± 8.6 mmol/day, respectively, P < 0.05) [69]. A high correlation between sodium bicarbonate administration and an increase in nocturnal blood pressure has also been reported [70]. Acute renal allograft rejection, renal artery stenosis and retained native kidneys are yet other aetiological factors of post-transplant hypertension.

Dyslipidaemia Although SPKTx has resulted in a favourable lipid profile including lowering of postprandial TG and total serum cholesterol levels and improvement in high-density lipoprotein (HDL) cholesterol, dyslipidaemia remains a significant risk factor for cardiovascular disease in the post-transplant period. Suggested aetiological factors for post-transplant dyslipidaemia include corticosteroid therapy, ciclosporin and tacrolimus immunosuppression, age, diet, rapid weight gain, hyperinsulinaemia, preexisting hypercholesterolaemia, and the use of ␤ -blockers and diuretics. Rapamune, a newer immunosuppressant that is structurally similar to tacrolimus, has also been shown to have significant hyperlipaemic effect, particularly hypertriglyceridaemia [71]. Corticosteroids and ciclosporin have been shown to be independently associated with elevated total serum cholesterol level. However, while corticosteroid therapy is associated with an increase in both low-density lipoprotein (LDL) and HDL cholesterol, ciclosporin therapy leads to an increase in total and LDL cholesterol with little or no effect on HDL cholesterol. Studies on the effect of steroid withdrawal on lipoprotein profiles of ciclosporin-treated kidney or SPK transplant recipients revealed a 17 per cent reduction in total serum cholesterol and a parallel 18 per cent reduction in HDL cholesterol levels. In diabetic recipients of kidney or SPK transplant, the ratio of total to HDL cholesterol remained unchanged following steroid withdrawal. In contrast, in non-diabetic kidney transplant recipients, the ratio actually increased following withdrawal of prednisone [67]. Whether prednisone withdrawal has

P.-T.T. PHAM ET AL.

Table 14.4 Goals: LDL < 100 mg/dl, TG < 200 mg/dl, HDL > 45 mg/dl LDL < 100 mg/dl

100–130 mg/dl

> 130 mg/dl

No drug therapy

Step I and II NCEP diet Suggested drug therapy

TG < 200 mg/dl HMGCoA RIa Re sinb Niacin

TG 200–400 mg/dl HMGCoA RIa Niacin

TG > 400 mg/dl Combination Rx HMGCoA RIa and (fibrate, niacin)

HDL < 35 mg/dl: weight management in conjunction with regular exercise programme. a HMGCoA RI (HMGCoA reductase inhibitor) are the most effective drugs and should be the agents of first choice. Start at low doses in patients on CSA and FK506. Monitor for myositis and hepatic enzyme elevations, particularly in patients receiving combination therapy. b Bile acid sequestrants should probably not be taken at the same time as CSA.

a salutary effect on the overall cardiovascular risk remains to be elucidated. Suggested guidelines for pharmacological treatments of dyslipidaemia is shown in Table 14.4.

Obesity Obesity is a well-established risk factor for accelerated atherosclerotic heart disease and is frequently associated with comorbid conditions including hyperinsulinaemia and insulin resistance, diabetes mellitus, and hypertension. Pretransplant obesity has variably been shown to be a predictor for increased weight gain post-transplant [72–74].

Diabetes mellitus The diabetogenic effect of immunosuppressive therapy including corticosteroids, cyclosporin, and tacrolimus has been well recognized and is discussed elsewhere.

Transplant-associated malignancies Recipients of organ transplants are at an increased risk of developing certain neoplastic complications compared to the general population. Post-transplant lymphoproliferative disorder and carcinoma of the lips and skin are seen with increased frequency in recipients of organ transplants compared to the general population. Data from the Cincinnati Transplant Tumor Registry (CTTR) revealed a 24 per cent incidence of lymphoma in transplant recipients compared to 5 per cent in the general population. While lip cancer is infrequently seen in the general population (reported incidence of 0.2 per cent), it occurs in 6.7 per cent of transplant recipients. Of those with lip cancer, nearly half also develop skin cancer [75]. Similar to post-transplant infectious complications, the time to occurrence of different types of malignancies following transplantation appears to follow a 'time-table' pattern. Post-transplant lymphoproliferative disease occurs early after transplantation, while the incidence of skin cancers increases with the length of follow-up after transplantation. Both the intensity and duration of immunosuppression have been suggested to be important risk factors. Interestingly, transplant recipients show no increase and even a decrease in the incidence of neoplasms commonly seen in the general population such as cancer of the lung, breast, prostate, colon, and invasive uterine cancer [76].

181

182

MEDICAL AND UROLOGICAL COMPLICATIONS

Table 14.5 Time of appearance of neoplasms following transplantation and initiation of immunosuppression (adapted from [75]) Type of cancer

Median in months

Lymphomas Kaposi's sarcoma Carcinomas of kidney Sarcomas (excluding Kaposi's) Carcinomas of cervix Hepatobiliary carcinomas Skin cancers Carcinomas of vulva/perineum

12 12.5 41 43.5 46 67.5 69 113.5

All cancers

46

The CTTR data on the time of appearance of different neoplasms following solid organ transplantation is shown in Table 14.5.

Post-transplant lymphoproliferative disorder Post-transplant lymphoproliferative disorder is a well-documented neoplastic complication of organ transplantation with a reported incidence of 1 to 2 per cent in renal transplant recipients, 2 to 5 per cent in liver and heart transplant recipients, 2 to 3 per cent in pancreass/SPK transplant recipients, and 6 to 8 per cent in recipients of lung transplants [77,78]. The majority of PTLD are non-Hodgkin's lymphoma of B-cell origin. It often involves the transplanted organ causing allograft dysfunction. Extranodal presentation is also common. Invasion of the central nervous system, liver, lungs, kidneys, and gastrointestinal tract has been well described. Important risk factors for the development of PTLD include primary EBV infection [79]; the intensity of immunosuppression particularly with protocols consisting of antilymphocyte antibody (ALA) including ATG, MALG, and/or OKT3 [80,81]; and antecedent history of CMV disease. It has recently been suggested that ciclosporin and tacrolimus may enhance the development of EBV-associated PTLD by directly promoting the survival of EBV-infected B-cells. In an in vitro model using spontaneous lymphoblastoid cell lines (SLCL) derived from transplant recipients with PTLD, ciclosporin and tacrolimus have been shown to augment SLCL growth through enhanced cell viability in a concentration-dependent manner. Specifically, ciclosporin and tacrolimus trated EBV-transformed cells are protected from spontaneous cell death as well as activated-induced apoptosis [82]. Interestingly, sirolimus, a potent immunosuppressant that blocks both calcium-dependent and calcium independent Bcell activation, has no effect on cell growth or viability of SLCL. There has been no consensus on the optimal management of PTLD. Nevertheless, reduction or discontinuation of immunosuppressive therapy particularly ALA, ciclosporin, tacrolimus, or mycophenolate mofetil should be the first line of treatment. Concomitant aciclovir or ganciclovir therapy has been reported to be beneficial and curative in benign polyclonal B-cell proliferation. The role of antiviral therapy in B-cell monoclonal malignant transformation is less well-defined. Fifty to ninety per cent mortality has been reported despite antiviral therapy [83]. Surgical resection or radiation therapy has been suggested for localized disease. In lesions not amenable to surgery, sequential approach with interferon-␣ and chemotherapy with anthracycline-based regimen have been used with favourable results [84]; however, its routine recommendation awaits further clinical studies. Current investigational therapeutic measures include anti-interleukin (IL) 6 antibody, and anti-B-cell

P.-T.T. PHAM ET AL.

monoclonal antibodies such as anti-CD20 (rituximab), anti-CD21, anti-CD24, or anti-CD40 antibodies [84–87]. Although the optimal treatment for established PTLD remains to be defined, antiviral preventive strategies targeting both CMV and EBV have been shown to decrease the incidence of PTLD and should be considered in all organ transplant recipients. The dose and duration of prophylactic and/or pre-emptive therapy should be tailored to each individual. In high-risk candidates (donor positive and recipient negative for EBV), a protocol consisting of pre-emptive intravenous ganciclovir for a minimum of 100 days combined with serial monitoring of peripheral blood for EBV by PCR has been shown to decrease the overall incidence of PTLD from 10 to 5 per cent in paediatric orthotopic liver transplantation (OLT) recipients on tacrolimus-based immunosuppression [88]. In a series of 109 lung transplant recipients reported by Levine et al., PTLD occurred in 1.8 per cent, an incidence much lower than previously reported for lung transplant recipients [89]. Their standard infection prophylaxis protocol include postoperative intravenous CMV immune globulin and intravenous ganciclovir for 14 days, followed by oral aciclovir 800 mg three times a day for 3 months and 400 mg orally, three times a day lifelong [88]. Of the first 190 pancreas transplants performed at the University of Maryland Medical System (including recipients of SPK/PAK/pancreas alone transplants), PTLD occurred in five patients within 6 weeks following transplantation, for an incidence of 2.6 per cent [77]. Ganciclovir prophylaxis was not given in four of five cases in whom the risk of CMV infection was felt to be low (donor and recipient negative for CMV). Pretransplant EBV antibody was positive in all five. In the subsequent series of 40 pancreas transplants, the authors advocated the use of ganciclovir prophylaxis in all cases. No PTLD was observed during the 7-month follow-up period. Although further observation is needed, the results suggest a beneficial effect of antiviral prophylaxis in reducing the risk of PTLD.

Skin cancers Skin cancers have been noted to be the most common cancers among all de novo post-transplant tumors. Skin cancers may occur 20 to 30 years earlier in immunosuppressed patients compared to the general population and the incidence of skin cancers may be up to 20 and seven times higher in sunexposed areas and sun-unexposed areas, respectively [90]. Risk factors for skin cancers among immunosuppressed post-transplant recipients include lightskin colour, intensity of sun exposure, and duration of follow-up following transplantation. Geographical risks for the development of skin cancers include residence in Australia, apparently due to higher sun exposure, and countries at higher latitudes such as Canada, Sweden, and Scotland, presumably due to malignant changes in papillomavirus-induced cutaneous warts in association with immunosuppression and sunlight exposure, among other factors [75]. The clinical presentation of skin cancers differs in transplant recipients compared to the general population in several aspects [75,76,90]. First, the incidence of squamous cell carcinoma is approximately twice that of basal cell carcinoma (up to 16 : 1 in Australia) among transplant recipients compared to a ratio of 0.2 to 1.0 in the general population. Second, multiple cancerous lesions of different types may be observed (i.e. concomitant appearance of squamous cell carcinoma and basal cell carcinoma). Third, the incidence of malignant melanoma may occur twice more frequently (5.4 versus 2.7 per cent). Fourth, skin cancers in transplant recipients tend to behave more aggressively with higher incidence of metastatic disease and mortality (5.8 versus to 1 to 2 per cent) compared to the general population. Not surprisingly, patients with skin cancers are more prone to develop other malignancies compared to transplant recipients free of skin cancers.

183

184

MEDICAL AND UROLOGICAL COMPLICATIONS

Conclusion In conclusion, SPKTx has evolved to become a widely acceptable therapeutic option for selected patients with ESRD secondary to type I diabetes mellitus. We have presented a historical view on the medical and urological complications following pancreas transplantation. The resurgence of ED of exocrine pancreatic secretions has resulted in a decreased incidence of UTI and urological complications as well as in fewer metabolic complications associated with primary BD. Portal venous drainage of pancreas transplant eliminates systemic hyperinsulinaemia and peripheral insulin resistance associated with systemic venous drainage. Post-transplant complications including infections, atherosclerotic vascular disease, dyslipidaemia, hypertension, bone disease, and transplant-associated malignancies continue to be important causes of morbidity and mortality following organ transplantation, which emphasize the need for more aggressive preventive and prophylactic therapy. Cardiovascular risk factor modification including blood pressure control and treatment of dyslipidaemia, weight reduction in obese patients, smoking cessation, and physical exercise should be encouraged. Dermatolgoy surveillance should be an integral part of post-transplant follow-up, especially in long-term survivors. Efforts to prevent or improve transplant-associated complications will unquestionably improve the quality of life in SPK transplant recipients beyond restoring renal function and euglycaemia.

References 1 Bruce DS, Newell KA, Josephson MA, Woodle ES, Piper JB, Millis JM, et al. Long-term outcome of kidney–pancreas transplant recipients with good graft function at one year. Transplantation 1996;62,451–6. 2 Elkhammas EA, Henry ML, Tesi RJ, Ferguson RM. Combined kidney/pancreas transplantation at the Ohio State University Hospitals. Clin Transpl 1992;191–7. 3 Munda R, Tom WW, First MR, Gartside P, Alexander JW. Pancreatic allograft exocrine urinary tract diversion. Pathophysiology. Transplantation 1987;43:95–9. 4 Nghiem DD, Gonwa TA, Corry RJ. Metabolic effects of urinary diversion of exocrine secretions in pancreatic transplantation. Transplantation 1987;43:70–3. 5 Burke GW, Gruessner R, Dunn DL, Sutherland DE. Conversion of whole pancreaticoduodenal transplants from bladder to enteric drainage for metabolic acidosis or dysuria. Transplant Proc 1990;22:651–2. 6 Kuo PC, Johnson LB, Schweitzer EJ, Barlett ST. Simultaneous pancreas/kidney transplantation — a comparison of enteric and bladder drainage of exocrine pancreatic secretions. Transplantation 1997;63:238–43. 7 Stratta RJ, Gaber AO, Shokouh-Amiri MH, Egidi MF, Grewal HP, Reddy KS, et al. Experience with portal-enteric pancreas transplant at the University of Tennessee-Memphis. Clin Transplant 1998;239–53. 8 Newell KA, Bruce DS, Cronin DC, Woodle ES, Millis JM, Piper JB, et al. Comparison of pancreas transplantation with portal venous and enteric exocrine drainage to the standard technique utilizing bladder drainage of exocrine secretions. Transplantation 1996;62:1353–6. 9 Ketel B, Henry ML, Elkhammas EA, Tesi RJ, Ferguson RM. Metabolic complications in combined kidney/pancreas transplantation. Transplant Proc 1992;24:774–5. 10 Stephanian E, Gruessner RW, Brayman KL, Gores P, Dunn DL, Najarian JS, et al. Conversion of exocrine secretions from bladder to enteric drainage in recipients of whole pancreaticoduodenal transplants. Ann Surg 1992;216:663–72.

P.-T.T. PHAM ET AL.

11 Dreiling DA, Jonowitz HD, Halpern M. The effect of a carbonic anhydrase inhibitor, diamox, on human pancreatic secretion. Pancreatic Physiol 1968;54:765–7. 12 Gaber AO, Shokouh-Amiri MH, Hathaway DK, Hammontree L, Kitabchi AE, Gaber LW, et al. Results of pancreas transplantation with portal venous and enteric drainage. Ann Surg 1995;221:613–24. 13 Diem P, Abid M, Redmon JB, Sutherland DE, Robertson RP. Systemic venous drainage of pancreas allografts as independent cause of hyperinsulinemia in type I diabetic recipients. Diabetes 1990;39:534–40. 14 Tajra LC, Martin X, Benchaid M, Dawhara M, Lefrancois N, Dubernard JM. Long-term metabolic control in pancreas transplant patients according to three techniques. Transplant Proc 1998;30:268–9. 15 Christiansen E, Vestergaard H, Tibell A, Hother-Nielsen O, Holst JJ, Pedersen O, et al. Impaired insulin-stimulated nonoxidative glucose metabolism in pancreas–kidney transplant recipients. Diabetes 1996;45:1267–75. 16 Luzi L, Secchi A, Facchini F, Battezzati A,Staudacher C, Spotti D, et al. Reduction of insulin resistance by combined kidney–pancreas transplantation in type I (insulin-dependent) diabetic patients. Diabetological 1990;33:549–56. 17 Elkhammas EA, Henry ML, Yilmaz S, Pelletier RP, Bumgardner GL, Ferguson RM. Simultaneous pancreas–kidney transplantation at the Ohio State University Medical Center. Clin Transpl 1997;167–72. 18 Rayhill SC, Kirk AD, Odorico JS, Heisey DM, Cangro CB, Pirsch JD, et al. Simultaneous pancreas–kidney transplantation at the University of Wisconsin.Clin Transplant 1995;00:261–9. 19 Smets YFC, van der Pijl, Ringers J, de Fijter JW, Lemkes HH, Hamdy NA. Pattern of bone loss after simultaneous pancreas kidney transplantation: a prospective study. Transplant Proc. 1998;30:326. 20 Sollinger HW, Messing EM, Eckhoff DE, Pirsch JD, D'Alessandro AM, Kalayoglu et al. Urological complications in 210 consecutive simultaneous pancreas–kidney transplants with bladder drainage. Ann Surg 1993;218:f516–70. 21 Sollinger HW, Odorico JS, Knechtle SJ, D'Alessandro AM, Kalayoglu M, Pirsch JD. Experience with 500 simultaneous pancreas – kidney transplants. Ann Surg 1998;228:284–96. 22 Sollinger HW, Knechtle SJ, Reed A, D'Alessandro AM, Kalayoglu M, Belzer FO, et al. Experience with 100 consecutive simultaneous kidney–pancreas transplants with bladder drainage. Ann Surg 1991;214:703–11. 23 Lumbreras C, Fernandez I, Velosa J, Munn S, Sterioff S, Paya CV. Infectious complications following pancreatic transplantation: incidence, microbiological and clinical characteristics, and outcome. Clin Infect Dis 1995;20:514–20. 24 Stratta RJ,Gaber AO, Shokouh-Amiri MH, Reddy KS, Egidi MF, Grewal HP, et al. A prospective comparison of systemic-bladder versus portal-enteric drainage in vascularized pancreas transplantation. Surgery 2000;127:217–26. 25 Elkhammas EA, Henry ML, Yilmaz S, Awad A, Skaf S, Bungardner GL, et al. Impact of intraabdominal fluid collections following simultaneous pancreas–kidney transplantation on graft and patient loss. Transplant Proc 1998;30:263. 26 Odorico JS, Becker YT, Van der Werf W, Collins B, D'Alessandro AM, Knechtle SJ, et al. Advances in pancreas transplantation: the University of Wisconsin experience. Clin Transpl 1997:157–66. 27 Fishman JA, Rubin RH. Infection in organ transplant recipients. N Engl J Med 1998;338:1741–51. 28 Secchi A, Caldara R, La Rocca E, Martinenghi S, Bernardi M, Bonfatti D. et al. Simultaneous kidney and pancreas transplantation at the San Raffaele Scientific Institute: clinical experience and results. Clin Transplant 1994;255–3. 29 Smets YF, van der Pijl JW, van Dissel JT, Ringers J, de Fijter JW, Lenkes HH. Infectious disease complications of simultaneous pancreas kidney transplantation. Nephrol Dial Transplant 1997;12:764–1.

185

186

MEDICAL AND UROLOGICAL COMPLICATIONS

30 Fishman JA. Pneumocystis carinii and parasitic infections in transplantation. Infect Dis Clin N Am 1995;9:1005–44. 31 Rubin RH, Wolfson JS, Cosimi AB, Tolkoff-Rubin NE. Infection in the renal transplant patient. Am J Med 1981;70:405–1. 32 Ruegsegger P, Medici TC, Anliker M. Corticosteroid-induced bone loss: a longitudinal study of alternate day therapy in patients with bronchial asthma using quantitative computed tomography. Eur J Clin Pharmacol 1983;25:615–20. 33 Ginns LC, Cosimi AB, Morris PJ, ed. Infection in the organ transplant recipient Transplantation. Blackwell Science, Inc. Malden, Massachusetts 747–69. 34 Barone GW, Webb JW, Hudec WA. The enteric drained pancreas transplant: another potential source of gastrointestinal bleeding. Am J Gastroenterol 1998;93:1369–71. 35 Bailey TC, Powderly WG, Storch GA, Miller SB, Dunkel JD, Woodward RS, et al. Symptomatic cytomegalovirus infection in renal transplant recipients given either Minnesota antilymphoblast globulin (MALG) or OKT3 for rejection prophylaxis. Am J Kidney Dis 1993;21:196–201. 36 O'Grady JG, Alexander GJ, Sutherland S, Donaldson PT, Harvey F, Portmann B, et al. Cytomegalovirus infection and donor/recipient HLA antigens: interdependent co-factors in pathogenesis of vanishing bile-duct syndrome after liver transplantation. Lancet 1988;2 (8606): 302–5. 37 Von Willbrand E, Pettersson E, Ahonen J, Havry P. CMV infection, class II antigen expression, and human kidney allograft rejection. Transplantation 1986;42:364–7. 38 Flechner SM, Avery RK, Fisher R, Mastroianni BA, Papajcik DA, O’Malley KJ, et al. A randomized prospective controlled trial of oral acyclovir versus oral ganciclovir for cytomegalovirus prophylaxis in high-risk kidney transplant recipients. Transplantation 1998;66:1682–8. 39 Goral S, Ynares C, Dummer S, Helderman JH. Acyclovir prophylaxis for cytomegalovirus disease in high-risk renal transplant recipients: is it effective? Kidney Int 1996;57:S62–5. 40 Shen GK, Alfrey EJ, Knoppel CL, Dafoe DC, Scandling JD. Eradication of cytomegalovirus reactivation disease during high-dose acyclovir and targeted intravenous ganciclovir in kidney and kidney/pancreas transplantation. Transplantation 1997;64:931–3. 41 Patel R, Snydman DR, Rubin RH, Ho M, Pescovitz M, Martin M, Paya CV. Cytomegalovirus prophylaxis in solid organ transplant recipients. Transplantation 1996;61:1279–89. 42 Stratta RJ, Taylor RJ, Bynon JS, Lowell JA, Cattral MS, Frisbie K, et al. Viral prophylaxis in combined pancreas–kidney transplant recipients. Transplantation 1994;57:506–12. 43 Grotz WH, Mudinger FA, Rasenack J, Speidel L, Olschewski M,Exner VM, et al. Bone loss after kidney transplantation: a longitudinal study in 115 graft recipients. Nephrol Dial Transplant 1995;10:2096–100. 44 Horber FF, Casez JP, Steiger U, Czerniak A, Montandon A, Jaeger P. Changes in bone mass early after kidney transplantation. J Bone Miner Res 1994;9:1–9. 45 Julian BA, Laskow DA, Dubovsky J, Dubovsky EV, Curtis JJ, Quarles LD. Rapid loss of vertebral mineral density after renal transplantation. N Engl J Med 1991;325:544–50. 46 Pichette V, Bonnardeaux A, Prudhomme L, Gagne M, Cardinal J, Ouimet D. Long-term bone loss in kidney transplant recipients: a cross-sectional and longitudinal study. Am J Kidney Dis 1996;28:105–14. 47 Weinstein RS, Jilka RL, Parfitt AM, Manolagas SC. Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. Potential mechanisms of their deleterious effects on bone. J Clin Invest 1998;102:274–82. 48 Movsowitz C, Epstein S, Fallon M, Ismail F, Thomas S. Cyclosporin-A in vivo produces severe osteopenia in the rat: effect of dose and duration of administration. Endocrinology 1988;123:2571–7.

P.-T.T. PHAM ET AL.

49 Westeel FP, Mazouz H,Ezaitouni F, Hottelart C, Ivan C,Fardellone P, et al. Cyclosporine bone remodeling effect prevents steroid osteopenia after kidney transplantation. Kidney Int 2000;58:1788–96. 50 Starzl TE, Marchioro TL, Porter KA, Moore GA, Rifkind D, Waddell WR. Renal homotransplantations: late function and complications. Ann Int Med 1964;61:470–7. 51 Chiu MY, Sprague SM, Bruce DS, Woodle ES, Thistlethwaite JR Jr, Josephson MA. Analysis of fracture prevalence in kidney–pancreas allograft recipients. J Am Soc Nephrol 1998;9:677–83. 52 Ramsey-Goldman R, Dunn JE, Dunlop DD,Stuart FP, Abecassis MM, Kaufman DB, et al. Increased risk of fracture in patients receiving solid organ transplants. J Bone Miner Res 1999;14:456–63. 53 Weber TJ, Quarles LD. Preventing bone loss after renal transplantation with biphosphonates: we can … but should we? Kidney Int 2000;57:735–7. 54 Frazao JM, Elangovan L, Maung HM, Chesney RW, Acchiardo SR, Bower JD, et al. Intermittent doxercalciferol (1-␣-hydroxyvitamin D2) therapy for secondary hyperparathyroidism. Am J Kidney Dis 2000;36:550–61. 55 Pham PC, Pham PT. In: Nissenson AR, Fine RN, eds. Dialysis therapy. third edition. Chapter 24. Parathyroidectomy. Hanley & Belfus, Inc., Philadelphia, PA. 2002, 410–15. 56 Adachi JD, Bensen WG, Brown J, Hanley D, Hodsman A, Josse R, et al. Intermittent etidronate therapy to prevent corticosteroid-induced osteoporosis. N Engl J Med 1997;337:382–7. 57 Fan SL, Almond MK, Ball E, Evans K, Cunningham J. Pamidronate therapy as prevention of bone loss following renal transplantations. Kidney Int 2000;57:684–90. 58 Grotz W H, Rump LC, Niessen A, Schmidt-Gayk H, Reichelt A, Kirste G, et al. Treatment of osteopenia and osteoporosis after kidney transplantation. Transplantation 1998;66:1004–8. 59 Eastell R, Boyle IT, Compston J, et al. Management of male osteoporosis: report of the UK Concensus Group. Q J M ed. 1998;91:71–92. 60 Morrissey PE, Shaffer D, Madras PN, Sahyoun AI, Monaco AP. Progression of peripheral vascular disease after combined kidney–pancreas transplantation in diabetic patients with end-stage renal failure. Transplant Proc 1997;29:662–3. 61 Morrissey PE, Shaffer D, Monaco AP, Conway P, Madras PN. Peripheral vascular disease after kidneypancreas transplantation in diabetic patients with end-stage renal disease. Arch Surg. 1997;132:358–62. 62 Knight RJ, Schanzer H, Guy S, Fishbein T, Burrows L, Miller C. Impact of kidney–pancreas transplantation on the progression of peripheral vascular disease in diabetic patients with end-stage renal disease . Transplant Proc 1998;30:1947–9. 63 Biesenbach, Margreiter R, Konigsrainer, Bosmuller C, Janko O, Brucke P et al. Comparison of progression of macrovascular diseases after kidney or pancreas and kidney transplantation in diabetic patients with end-stage renal disease. Diabetologia 2000;43:231–4. 64 Becker BN, Brazy PC, Becker YT, Odorico JS, Pintar TJ, Collins BH, et al. Simultaneous pancreas–kidney transplantation reduces excess mortality in type I diabetic patients with end-stage renal disease. Kidney Int 2000;57:2129–35. 65 Cheung AT, Perez RV, Basadona GP, Cox KL, Bry WI. Microangiopathy reversal in successful simultaneous pancreas–kidney transplantation. Transplant Proc 1994;26:493–5. 66 Fioretto P, Steffes MW, Sutherland DER, Goetz FC, Mauer M. Reversal of lesions of diabetic nephropathy after pancreas transplantation. N Engl J Med 1998;339:69–75. 67 Hricik DE, Bartucci MR, Mayes JT, Schulak JA. The effects of steroid withdrawal on the lipoprotein profiles of cyclosporine-treated kidney and kidney – pancreas transplant recipients. Transplantation 1992;54:868–71. 68 La Rocca E, Gobbi C, Ciurlino D, Di Carlo V, Pozza G, Secchi A. Improvement of glucose/insulin metabolism reduces hypertension in insulin-dependent diabetes mellitus recipients of kidney–pancreas transplantation. Transplantation 1998;65:390–3.

187

188

MEDICAL AND UROLOGICAL COMPLICATIONS

69 Raja MR, Lerner L, Morris M. Hypertension with combined pancreas–kidney transplantation in patients with diabetic nephropathy. Transplantation 1993;55:1187–8. 70 Marx MA, Gardner SF, Ketel BL. Diurnal blood pressure variation in kidney–pancreas transplant recipients. Am J Hypertens 1996;9:823–7. 71 Brattstrom C, Wilczek HE, Tyden G, Bottiger Y, Sawe J, Groth CC. Hypertriglyceridemia in renal transplant recipients treated with sirolimus. Transpl Proc 1998;30:3950–1. 72 Bumgardner GL, Henry ML, Elkhammas E, Wilson GA, Tso P, Davies E, et al. Obesity as a risk factor after combined pancreas/kidney transplantation. Transplantation 1995;60:1426–30. 73 Merion RM, Twork AM, Rosenberg L, Ham JM, Burtch GD, Turcotte et al. Obesity and renal transplantation. Surg Gynecol Obstet 1991;172:367–76. 74 Palmer M, Schaffner F, Thung SN. Excessive weight gain after liver transplantation. Transplantation 1991;51:797–800. 75 Ginns LC, Cosimi AB, Morris PJ, ed. Transplantation. Neoplastic complications of organ transplantation. Blackwell Science; Inc. Malden, Massachusett, 770–86. 76 Penn I. The changing pattern of posttransplant malignancies. Transplant Proc 1991;23:1101–3. 77 Bartlett ST, Kuo PC, Johnson LB, Lim JW, Schweitzer EJ. Pancreas transplantation at the University of Maryland. Clin Transpl 1996;271–80. 78 Kew II CE, Lopez-Ben R, Smith JK, Robbin ML, Cook WJ, Gaston RS, et al. Posttransplant lymphoproliferative disorder localized near the allograft in renal transplantation. Transplantation 2000;69:809–14. 79 Ho M, Miller G, Atchison RW, Breinig MK, Dummer JS, Andiman W, et al. Epstein–Barr virus infections and DNA hybridization studies in posttransplantation lymphoma and lymphoproliferative lesions: the role of primary infection. J Infect Dis 1985;152:876–86. 80 Jamil B, Nicholls K, Becker GJ, Walker RG. Impact of acute rejection therapy on infections and malignancies in renal transplant recipients.Transplantation 1999;68:1597–19. 81 Swinnen LJ, Costanzo-Nordin MR, Fisher SG, O’Sullivan EJ, Johnson MR, Heroux AL, et al. Increased incidence of lymphoproliferative disorder after immunosuppression with the monoclonal antibody OKT3 in cardiac-transplant recipients. N Engl J Med 1990;323:1723–8. 82 Beatty PR, Krams SM, Esquivel CO, Martinez OM. Effect of cyclosporine and tacrolimus on the growth of Esptein-Barr virus-transformed B-cell lines. Transplantation 1998;65:1248–55. 83 Hanto DW, Gajl-Peczalska KJ, Frizzera G, Arthur DC, Balfour HH Jr, McClain K, et al. Epstein – Barr virus (EBV) induced polyclonal and monoclonal B-cell lymphoproliferative diseases occuring after renal transplantation: clinical, pathologic, and virologic findings and implications for therapy. Ann Surg 1983;198:356–69. 84 Davis CL, Wood BL, Sabath DE, Joseph JS, Stethman-Breen C, Broudy VC. Interferon-alpha treatment of posttransplant lymphoproliferative disorder in recipients of solid organ transplants. Transplantation 1998;66:1770–9. 85 Milpied N, Vasseur B, Parquet N, Garnier JL, Antoine C, Quartier Pet al. Humanized anti-CD20 monoclonal antibody (rituximab) in post transplant B-lymphoproliferative disorder: a retrospective analysis on 32 patients. Ann Oncol 2000;11:S113–16. 86 Oertel SHK, Anagnostopoulos I, Bechstein WO, Liehr H, Riess HB. Treatment of posttransplant lymphoproliferative disorder with the anti-CD20 monoclonal antibody rituximab alone in an adult after liver transplantation: a new drug in therapy of patients with posttransplant lymphoproliferative disorder after solid organ transplantation? Transplantation 2000;69:430–2. 87 Paya CV, Fung JJ, Nalesnik MA, Kieff E, Green M, Gores G, et al. Epstein–Barr virus-induced posttransplant lymphoproliferative disorders. Transplantation 1999;68:1517–25.

P.-T.T. PHAM ET AL.

88 McDiarmid SV, Jordan S, Kim GS, Toyoda M, Goss JA, Vargas JH, et al. Prevention and preemptive therapy of posttransplant lymphoproliferative disease in pediatric liver recipients. Transplantation 1998;66:16 04–11. 89 Levine SM, Angel L, Anzueto A, Susanto I, Peters JI, Sako EY, et al. A low incidence of posttransplant lymphoproliferative disorder in 109 lung transplant recipients. Chest 1999;116:1273–7. 90 Danovitch GM. Handbook of kidney transplantation. Long term post-transplant management and complications. Philadelphia, Pennsylvania: 154–86.

189

This page intentionally left blank

Chapter 15

Diagnosis and treatment of pancreatic rejection Alan J. Koffron and Dixon B. Kaufman

Introduction Pancreatic transplantation is intended to re-establish euglycaemia without exogenous insulin supplementation. Pancreatic transplantation is the only treatment for type I diabetes that normalizes haemoglobin A1c levels for as long as the graft functions [1]. Through improved metabolic control, the course of many secondary complications of diabetes including diabetic neuropathy [2], the incidence of autonomic neuropathy-associated sudden death [3], and onset of diabetic nephropathy in both nonuraemic patients [4] and those diabetics having undergone renal transplantation [5] may be markedly improved. The most important threat to loss of a functioning pancreas allograft is acute rejection. Any discussion of rejection must take into account the three different scenarios in which pancreas transplantation takes place. The largest recipient population (85 to 90 percent of pancreas transplants) is the diabetic-uraemic patient [6] treated by simultaneous pancreas and kidney (SPK) transplant. Pancreas after kidney (PAK) transplant is performed in approximately 10 per cent of American cases of pancreas transplantation [7,8]. Pancreas transplant alone (PTA) occurs in about 5 per cent of American cases and is indicated in the setting of the non-uraemic diabetic for the purpose of correcting diabetes for patients with extreme lability of metabolic control and poor quality of life [9]. Oneyear pancreatic survival rates exceed 80 per cent in SPK recipients, whereas pancreas graft survival for PTA recipients and PAK recipients is inferior, explained largely by the incidence of loss due to acute rejection [10]. While the most common cause for graft failure within the first year post-transplant is technical, rejection is the second most common cause for graft failure, with marked differences by recipient category. In SPK recipients, technical failures account for 63 per cent and rejection for 31 per cent of all graft losses; for PAK, technical failures for 42 per cent and rejection for 53 per cent: for PTA, technical failures for 36 per cent and rejection for 61 per cent [10]. Thus, for solitary pancreas recipients (PAK, PTA), rejection is the most common cause of graft failure, for SPK recipients, technical failures are more common. Therefore, if only technically successful pancreas transplants are considered, rejection is by far the most common cause of graft loss for all recipient categories. Acute rejection and chronic rejection are responsible for 47 and 52 per cent of rejection-related graft losses, respectively [10]. Pancreatic allograft monitoring and the diagnosis of rejection have been the most difficult problems in transplantation and the most menacing aspect of pancreas transplantation itself. Undertreatment or overtreatment of a suspected rejection episode are the most serious causes of morbidity and mortality of the pancreas transplant recipient. This chapter reviews both the various diagnostic modalities available in the diagnosis of pancreatic allograft rejection, and the treatment of this most elusive clinical entity.

192

DIAGNOSIS AND TREATMENT OF PANCREATIC REJECTION

Pathophysiology of pancreas rejection Pancreas allograft rejection is similar to rejection of other transplanted solid organs and characterized as hyperacute, acute, and chronic. Hyperacute rejection is a difficult diagnosis to make because of the relatively high incidence of early organ failure due to vascular thrombosis. The few cases published describe negative crossmatches in recipients with high panel-reactive antibody levels [11]. Acute pancreatic graft rejection is primarily a function of cell-mediated cytotoxicity. The initial cellular targets of rejection are the endothelial cells, acinar, and ductal epithelial cells. Islets, and specifically ␤-cells, are not primary targets of alloimmune rejection [11]. Interestingly, recurrent autoimmune destruction of ␤-cells has been reported in recipients receiving a living donor pancreas graft from an identical twin. Islets, however, may be involved late in rejection and may also stop functioning before becoming involved with inflammatory cells [12–14]. Rejection is usually described as two events, including rejection of the parenchymal tissues and rejection of the vascular tissue. This is a dynamic process with connection of the two processes. As with many organ types, little information is available regarding the pathogenic mechanism of chronic rejection of the pancreas. Many possible aetiologies have been suggested that may also apply to chronic rejection of the pancreas. The most notable possible contributing factors include multiple acute rejection episodes, systemic or organ-specific infection, hyperlipidaemia, and the toxic effects of calcineurin inhibitors on vascular endothelial and smooth muscle cells [15]. Chronic rejection in the pancreas is characterized by arterial narrowing and interstitial fibrosis with variable loss of acinar and islet tissue [12,14]. This will progressively cause ischaemic damage to the acinar and islet tissues, resulting in extensive pancreatic fibrosis. Acute rejection, however, remains the most clinically relevant process in pancreas transplantation.

Clinical diagnosis of pancreatic rejection The clinical presentation of pancreas allograft rejection is very different to that of kidney rejection. In most cases of pancreatic allograft rejection, clinical symptoms are subtle or non-existent. Only 5 to 20 per cent of patients with pancreatic graft rejection present with obvious clinical symptoms remarked by the recipient [16,17]. The pancreatic graft undergoing acute rejection becomes inflamed. Patients experience pain and discomfort due to surrounding peritoneal irritation, but it is difficult to distinguish clinically from benign graft pancreatitis. Fever as a clinical symptom of rejection is not common, partly due to maintenance immunosuppressive therapy with prednisone. But if the workup for infection is negative, fever is highly suspicious of rejection. Likewise, paralytic ileus and acute abdomen are rarely seen but can be caused by rejection-induced pancreatitis. Inflammation of the surrounding organs, such as the small and large intestine, may result in adynamic ileus or diarrhoea, respectively. Even in the presence of overt clinical symptoms, the diagnosis of rejection, if a biopsy is not obtained, is usually a composite decision based on clinical and laboratory criteria. Destruction of the ␤-cells occurs relatively late following initial injury of acinar tissue. Therefore, hyperglycaemia is considered a very late sign of rejection and is associated with a poor outcome.

Laboratory diagnosis of pancreatic rejection Serum markers of rejection These are the most commonly relied upon parameters to guide decision-making regarding subsequent imaging studies or biopsy. It is established that profound destruction of exocrine pancreatic

A.J. KOFFRON AND D.B. KAUFMAN

tissue occurs prior to significant deterioration in endocrine pancreatic function [18]. Therefore, hyperglycaemia is a late parameter of rejection and usually apparent only after extensive destruction of the islets has taken place. The sign of hyperglycaemia is not useful to diagnose acute rejection that is likely to be reversed. It is more likely a sign of development of type II diabetes. Measurement of Cpeptide will determine if that has occurred. The problem of using pancreas-specific serum markers to detect rejection lies in the pathophysiology of the exocrine pancreas: rejection — as well as pancreatitis, infection, or preservation injury — leads to temporary or constant damage of acinar tissue, with subsequent enzyme and cytokine release. Thus, the causes of destruction of pancreas acinar tissue are multiple and, with pancreas-specific serum parameters only, difficult to differentiate. An increase in serum amylase may occur with rejection and precede a decline in urinary amylase (in recipients with bladder-drained pancreas), but unfortunately this is non-specific [19-21]. Posttransplant hyperamylasaemia can be caused by any process causing pancreatic inflammation by virtue of bystander acinar tissue damage. In addition, serum amylase is also derived in large part from other tissues, including salivary glands and intestine. A multitude of other serum markers have been introduced to facilitate the prompt diagnosis of rejection. Most have not reached the level of clinical relevance, either because they are not universally available or are not consistently reliable.

Serum creatinine In the context of SPK transplantation, it is the kidney allograft that is the best indicator of a rejection episode. Rejection of the kidney allograft will manifest as a rise in serum creatinine. This will prompt ultrasound and biopsy of the kidney allograft, and if rejection is diagnosed, antirejection therapy is instituted. If there is a concurrent pancreas graft rejection process, the antirejection therapy will reverse the process in both organs.

Serum anodal trypsinogen Serum anodal trypsinogen (SAT) has been the focus of several experimental and clinical investigations. Initial work in a porcine model compared serum levels of immunoreactive anionic and cationic trypsin [22]; rejection was heralded by a significant increase in immunoreactive anodal trypsinogen by at least 4 or more days before hyperglycaemia or histological evidence of rejection in about 80 per cent of cases. A decrease in immunoreactive cationic trypsin was less sensitive for rejection. Several studies of SPK recipients showed elevated anodal trypsinogen levels during clinically diagnosed rejection episodes [20,23,24] but, in the absence of histological evidence of rejection, these studies lack clarity. SAT levels are frequently elevated in the early post-transplant period, which may reflect preservation or procurement injury (pancreatitis) rather than rejection; this underscores the need to correlate SAT with pancreatic core biopsy findings. SAT levels may be influenced by renal dysfunction [22–4], pancreatitis, trauma, and outlet obstruction. One study included both renal biopsies and SAT levels performed on SPK and PAK recipients, finding SAT a reliable marker of pancreas rejection in all cases [25]. Unlike urinary amylase, SAT does not depend on the type of management of exocrine pancreatic secretion and therefore can be used to monitor bladder-drained or enteric-drained pancreatic grafts. One study [24] compared urinary amylase, SAT, and serum amylase in acute rejection episode in bladder-drained SPK recipients; all rejection episodes were biopsy-proven. It was determined that serum amylase, although less specific, was as sensitive as SAT, but did not correlate with successful treatment of the rejection process. Both serum amylase and serum creatinine levels positively correlated with SAT at a time when urinary amylase changes were not yet apparent. The usefulness of SAT

193

194

DIAGNOSIS AND TREATMENT OF PANCREATIC REJECTION

in diagnosing pancreatic graft rejection must be determined in PTA recipients and the frequency with which an increase in SAT and serum amylase precedes a decrease in urinary amylase needs to be calculated and correlated with biopsy proven rejection or absence of rejection.

Other serum markers Plasma pancreatic secretory trypsin inhibitor (PSTI), a pancreatic exocrine protein, a marker of acute pancreatitis [26,27], has been studied in the context of rejection but while possessing high sensitivity, lacks specificity particularly in the presence of renal dysfunction [28,29). In contrast, pancreasspecific protein (PASP) lacks both sensitivity and specificity [30–32]. Pancreatic elastase 1 is difficult to interpret in the presence of renal dysfunction [33,34], and interleukin 2 (IL-2) and its soluble receptor (SIL-2R) alone have not been shown to be reliable in diagnosing rejection [35]. Several other markers of rejection have been investigated including neopterin [36,37], phospholipase A2 [29,38], thromboxane, and prostacyclin [39] showing poor clinical application.

Summary on utility of serum markers of rejection Despite the wide variety of serum and urinary tests, not a single marker currently can reliably predict pancreatic graft rejection. For SPK recipients, kidney biopsies in conjunction with serial serum creatinine levels have been used to diagnose pancreatic graft rejection. For recipients of solitary pancreatic transplants (PTA, PAK), serum creatinine levels are of no diagnostic value in this context and laboratory diagnosis by pancreas-specific markers appears to be even more important. In bladder-drained pancreatic transplants, urinary amylase remains the most common parameter due to its simplicity and ubiquitous availability. The need for reliable laboratory parameters is most clinically urgent in enteric-drained, injected solitary pancreatic transplant, and pancreatic islet cell transplant recipients. Some plasma pancreas-specific markers show clinical diagnostic promise, particularly SAT; yet clinical usefulness must be correlated with biopsy specimens from PTA recipients. Employing a panel of serum and urinary parameters and carefully assessing the patient’s clinical course can help detect rejection early and reverse the process when biopsies cannot be obtained. However, pancreatic biopsies remain the gold standard for diagnosing and grading rejection episodes.

Urine markers of rejection Urine amylase Bladder drainage was a widely used technique for the management of exocrine secretion in pancreatic transplantation. Originally advocated to reduce the surgical and infectious complication rate, it also allows graft exocrine function to be monitored by measuring pancreatic enzymes secreted directly into the urine [40,41]. It is mostly used in recipients of PAK and PTA. It is becoming less frequently used in SPK transplant recipients because monitoring renal allograft function serves as a better indication of rejection (and a surrogate marker of pancreas graft rejection), and there is less morbidity of enteric drainage. In clinical and experimental studies, exocrine pancreatic rejection has been shown to precede endocrine rejection [12,42,43], where islets are spared during the early, interstitial phase of rejection [44]. Serial urine amylase measurement has emerged as a very common surveillance and diagnostic laboratory test: a reduction in urinary amylase activity (relative hypoamylasuria) is the most commonly used biochemical marker of acute rejection in the pancreas alone and PAK. By monitoring urinary amylase levels, antirejection treatment can begin before hypoglycaemia occurs [43]. Urinary amylase measurements are simple, without morbidity, relatively inexpensive, and can be performed by most laboratories.

A.J. KOFFRON AND D.B. KAUFMAN

One of the limitations of urinary amylase monitoring is that a decrease in activity does not necessarily mean rejection. Reduced urinary amylase levels may be caused by other factors such as preservation injury in the early post-transplant period, pancreatitis, fibrosis, thrombosis, ductal obstruction, prolonged fasting, hydration status, and diuresis [45–8]. In addition, exocrine function may not directly reflect the integrity of ␤-cells [49]. In these situations, hypoamylasuria may lead to unnecessary treatment and sequelae. A few studies have correlated hypoamylasuria and pancreaticoduodenal biopsy results. Munn et al. [50] noted that histological rejection was found in 64 per cent of episodes of hypoamylasuria (urinary amylase decrease > 50 per cent from baseline). Nankivell et al. [46] reported a 60 per cent sensitivity when urinary amylase activity was correlated with protocol biopsies. Benedetti et al. [51] found a urinary amylase activity sensitivity of 100 per cent (stable urinary amylase levels meaning no rejection) and a specificity of 30 per cent; the predictive value of a positive test (a > 25 per cent decrease from baseline urinary amylase levels) was 53 per cent; of a negative test, 100 per cent. Therefore, stable post-transplant urinary amylase levels reliably exclude rejection, and rejection is associated with a decline in urinary amylase activity. Monitoring of urinary amylase levels is the most common method of evaluation pancreatic function in this context.

Tissue and cell diagnosis of rejection (see also Chapter 16) Needle core biopsy This is the gold standard for the diagnosis of pancreas allograft rejection in the context of PAK and PTA. For most solid organ transplants, histological evaluation of graft biopsies became the standard assessment for rejection early on. For pancreas transplantation, the development was different for two reasons. It is rare that isolated pancreatic rejection occurs in SPK recipients without simultaneous renal allograft rejection. In these patients, most rejection episodes involve either the kidney alone or the kidney and the pancreas simultaneously [52]. This observation has promoted the perception that pancreatic graft rejection can be monitored indirectly by relying on serum creatinine changes or kidney graft biopsies. For SPK transplants, the kidney serves as excellent surrogate marker for rejection. In recipients of solitary pancreas transplants (PTA, PAK), serum creatinine levels or kidney biopsies cannot be used as markers of rejection, and given the inadequacies of laboratory parameters, biopsies are therefore essential for monitoring solitary pancreas transplants. It should be emphasized that in SPK recipients, isolated pancreatic graft rejection can occur and pancreatic graft biopsies may become necessary if a change in exocrine or endocrine laboratory parameters occurs without an elevation in serum creatinine. In the early experience with pancreas transplantation, pancreatic biopsies were only reluctantly performed, due to the potential complications such as pancreatitis, pancreatic fistulas, visceral injury, and bleeding. Before the development of special percutaneous biopsy needles and new imaging techniques [ultrasound, computed tomography (CT), magnetic resonance imaging (MRI)], tissue diagnosis usually required laparotomy and graft biopsy. In this situation complications are well recognized [53,54]. Currently, the vast majority of pancreatic graft biopsies are obtained either percutaneously or cystoscopically and only rarely by laparotomy. Percutaneous and cystoscopic biopsies should be done under ultrasound or CT guidance to enhance the diagnostic value and safety of these procedures. Percutaneous biopsies of bladder-drained pancreas grafts, developed in 1990, are successful in over 90 per cent of attempts [55], when performed under radiographic guidance, and is proven to be more reliable than biochemical assays in the diagnosis of rejection [56].

195

196

DIAGNOSIS AND TREATMENT OF PANCREATIC REJECTION

Cystoscopic transduodenal pancreas biopsy (CTDB) achieves equal success [51], especially when combined with ultrasound guidance [58–60]. In addition, this approach allows isolated duodenal biopsy. Animal studies have shown that the duodenum is reflective of pancreatic pathology, in twothirds of cases [61], which can reliably diagnose, although not exclude, pancreas rejection [62,63]. More recently, it is suggested that in those cases in which rejection is present, and both duodenal and pancreas tissues were examined, over 80 per cent were concordant regarding the presence of rejection [64]. However, duodenal biopsy is not as specific as pancreas biopsy, and a technical problem with duodenal biopsy is mucosal sloughing, which obscures the findings. Cystoscopic biopsies are associated with a low complication rate including haematuria and pancreatitis [51,59,60]. Unfortunately, this approach is only applicable to bladder-drained pancreatic grafts. Currently, most centres prefer to employ ultrasound-guided, percutaneous biopsy, performed under local anaesthesia. If unable to obtain tissue for histology, or overlying bowel prohibits sampling, the cystoscopic approach is employed for bladder-drained graft. Laparotomy and biopsy is reserved for enteric/bladder-drained grafts inaccessible by the aforementioned approaches and the risks of empiric antirejection therapy outweigh that of surgery.

Fine-needle aspiration biopsy Fine-needle aspiration biopsy (FNAB), although employed originally for safety has been found to be more technically difficult and to have a high failure rate approaching 30 per cent, sampling error, and difficult histological interpretation [55].

Cytology Inflammatory cells appear in pancreas secretions, and therefore are useful in the diagnosis of early rejection [65]. Cytological examination of the pancreatic exocrine drainage has been performed by at least two methods [66–69] and used as a screening and diagnostic tool for the detection of rejection. Features characteristic of acute rejection include increased overall cellularity, with over 5 per cent lymphocytes, eosinophil granules, and necrotic epithelial cells. The presence of lymphocytes in pancreatic secretions was found to precede the change in amylase level in one study [65], and a larger study found pancreatic secretion cytology to have a sensitivity and specificity of 87 and 97 per cent, respectively [70]. Advantages of monitoring urinary cytology or pancreatic drainage cytology is that rejection can be detected before clinical symptoms and before a decline in urinary amylase occurs. However, persistent lymphocyturia (following antirejection treatment), viral infection, and pancreatitis limit the clinical application of this technique [71]. Cytological monitoring should be confirmed by biopsy [68,69] and therefore, reserved to assist in diagnosis for those patients where the only remaining option is laparotomy.

Imaging techniques for diagnosis of rejection While the wide variety of imaging techniques have been helpful in a variety of pathological conditions associated with pancreatic transplantation (pancreatitis, graft thrombosis, anastomotic leak, intraabdominal infection) most techniques are of no diagnostic value in the early diagnosis of rejection. There are virtually no studies correlating radiographic findings with allograft biopsies, raising doubt to the true clinical applicability of diagnostic radiology in the context of pancreatic rejection.

A.J. KOFFRON AND D.B. KAUFMAN

Scintography A multitude of tracers have been studied in pancreatic transplant recipients [6]. Clinically, the most frequently employed tracing agent is technetium-99 DPTA, due to effective visualization of the simultaneously transplanted kidney (in SPK recipients). Computer analysis can subsequently generate a quantitative measure of blood flow to the pancreas (technetium index or TI). Unfortunately, poor visualization of the pancreas is common and has been reported despite normal pancreatic allograft function. Therefore, in clinical practice, the results are usually interpreted in conjunction with laboratory data and clinical findings. The obvious short-coming of scintigraphic methods is that while they are capable of differentiating grafts with perfusion detects, they are not capable of distinguishing subtle changes, such as early rejection versus pancreatitis.

Ultrasound, computed tomography, and magnetic resonance imaging Ultrasonic examination of the pancreatic graft has not been helpful in diagnosing rejection, although patterns of inhomogeneous echogenicity have been linked with acute rejection [72]. Unfortunately, this finding can also be present in the presence of graft preservation injury, pancreatitis, or (partial) thrombosis [73,74]. Graft inflammation and enlargement has been noted in association with acute rejection or pancreatitis. Duplex ultrasonography has been successfully used in the immediate postoperative period for reasons other than rejection. The presence of low splenic vein velocity, associated with absence of pulsatile flow and inadequate glucose control, suggest partial splenic vein thrombosis. For other purposes, such as rejection, duplex ultrasonography has not been helpful [75]. It has been suggested that Doppler sonography of intrapancreatic arterial flow and determination of resistive index values can help diagnose acute rejection. However, the intra-abdominal location of the pancreas, which causes occasional difficulty with visualization (that is, overlying viscera and intestinal gas), and the lack of sensitivity and specificity for resistive indices (as demonstrated for renal and hepatic transplants), have tempered enthusiasm for the use of Doppler sonography in diagnosing rejection [74]. Likewise CT studies have not been able to reliably diagnose (early) rejection. Graft inhomogenicity can be seen, but is non-specific [74,76]. Irreversible rejection diminishes graft size due to shrinkage and fibrosis [72,76]. CT studies in pancreas recipients are best used to detect major parenchymal abnormalities (for example, thrombosis, oedema, haemorrhage, pseudocyst) and abdominal fluid but not to diagnose rejection [74]. No studies correlating the images with biopsies are reported. A few MRI studies have been conducted in pancreas grafts during rejection. One group noted changes in tissue water content, such as inflammatory oedema during rejection and a decrease in congestion after effective treatment [77]. However, false-positive results were noted, especially during recovery from acute rejection and in the immediate postoperative period; the false-negative rate was low, suggesting that rejection is unlikely if MRI is negative. In a retrospective study comparing nuclear medicine, ultrasound, and MRI, it appeared that MRI had the highest sensitivity and specificity in detecting rejection [78], although further studies are needed to determine whether MRI is more useful in suggesting rejection that CT or ultrasound studies. In summary, imaging techniques can help diagnose a variety of pathological conditions after pancreatic transplantation improving the clinical course in many patients. Unfortunately, current technology is not capable of diagnosing early pancreatic allograft rejection.

197

198

DIAGNOSIS AND TREATMENT OF PANCREATIC REJECTION

Immunosuppression (see also Chapter 19) Over the past decade, pancreas transplantation results have improved significantly primarily due to advances in immunosuppression. The principles of immunosuppressive therapy for pancreas recipients are similar to those applied to recipients of other solid organ allografts. However, the amount of induction and maintenance immunosuppression required appears to be more than for other solid organ transplants. This is based on a higher incidence of rejection episodes well described in the transplant literature. Combination immunotherapy has been the basis of clinical immunosuppression since the initiation of pancreas transplantation. The advent of more effective immunomodulating agents has reduced the frequency and severity of pancreatic allograft rejection episodes. However, acute rejection continues to be the most challenging event in the course of pancreatic graft recipients. Acute cellular rejection has been treated by a myriad of agents, each with differing mechanisms of action, indications, and efficacies. The oldest immune response-modifying drugs are the corticosteroids, which have been used to suppress inflammation and immune-mediated diseases for over five decades. Their immunosuppressive and anti-inflammatory effects are complex and doserelated, causing inhibition of proinflammatory genes, lymphocyte population alterations, and many other effects. Corticosteroids are typically the first-line agents for the treatment of mild cellular rejection. In this scenario, corticosteroids may be effective in approximately 30 to 50 per cent of patients [79]. The treatment of more severe rejection includes protein-based drugs designed to target specific cell membrane receptors or their ligands and are typically antibodies. These agents are highly selective and potent. This class of immunomodulating agents consists of lymphocyte immune globulins [antilymphocyte globulin (ALG), antithymocyte globulin (ATG), and thymoglobulin] and OKT3. These agents are typically used to induce immunosuppression and reverse acute cellular rejection, with varying degrees of success.

Polyclonal T-cell depleting antibodies Antilymphocyte globulin and ATG are polyclonal antibody preparations against human lymphocyte membrane structures. They are derived from sera of animals immunized against human lymphocytes or thymocytes. The immunosuppressive effect of these agents is achieved through the binding of antibodies to lymphocyte membrane proteins. This has three effects: lysis of circulating lymphocytes, altered traffic of antibody-coated cells, and altered function of affected cells.

Monoclonal T-cell depleting antibody In contrast to the above agents, OKT3 is a murine-derived monoclonal antibody against the human lymphocyte CD3 moiety. OKT3 reacts with the CD3 complex by blocking the function of T lymphocytes by inhibiting signal transduction of the T-cell receptor complex. Clinical experience has shown that prompt, aggressive therapy is vital to the treatment of pancreatic allograft rejection. In many transplant centres, pancreatic rejection episodes are treated with a 7- to 14-day course of monoclonal (OKT3) or polyclonal (ALG, ATG, thymoglobulin) antibody [80–2]. A representative study included 46 rejection episodes in 21 recipients (13 SPK, eight PAK) treated with monoclonal antibody (OKT3) [17]. Indications for OKT3 use included steroid- or antilymphoblast

A.J. KOFFRON AND D.B. KAUFMAN

globulin-resistant rejection. OKT3 was administered for 14 days, concomitant with corticosteroids. In 62 per cent of recipients, OKT3 rescue therapy was successful, but there were significant differences by recipient category. In SPK recipients, 92 per cent of rejection episodes were responsive to OKT3, for PAK recipients, only 13 per cent OKT3 therapy was most successful for recipients with early rejection, for SPK recipients with rejection, and for rejection not associated with hyperglycaemia (late rejection). Salvage treatment with OKT3 appears safe and effective in reversing pancreatic rejection [17]. The conclusions of this study are difficult to interpret as pancreatic rejection was based on clinical criteria, not on pancreatic biopsies. There have been no large, randomized trials comparing thymoglobulin to OKT3 in the treatment of rejection. However, thymoglobulin, has been compared to ATG in the context of induction of immunosuppression [83] and in the treatment of rejection [84] in kidney transplantation. In these studies, thymoglobulin was found to be significantly more effective in both the incidence of rejection and its effective reversal. With the advent of multiple effective rejection therapies, it is plausible to match the drug that is most appropriate for the severity of rejection. The increasing diagnostic usefulness of percutaneous and cystoscopic biopsy techniques may help determine whether pancreatic rejection treatment can be tailored to the individual patient by severity of the rejection episode as graded by a pathologist. It is conceivable that minimal or mild pancreatic rejection may only require steroid boluses or recycling of the steroid taper. Antibody therapy may be necessary only to reverse moderate and severe rejection episodes. These hypotheses must be tested in clinical studies that include histological grading and antirejection agents with particular attention to the type of pancreas transplant (SPK, PAK, PTA).

Summary and conclusion In summary, the results for all types of pancreas transplantation have improved over the past 5 years in large part due to advances in immunosuppression. As with other organ transplants, episodes of acute rejection appear to predispose to late pancreas transplant loss. With decreasing rates of pancreas rejection and immunological graft failure, this may translate into improved long-term survival for patients transplanted under mycophenolate and/or tacrolimus immunosuppression.

References 1 Morel P, Goetz F, Moudry-Munns KC, et al. Long term metabolic control in patients with pancreatic transplants. Ann Intern Med 1991;115:694–9. 2 Van der Vliet JA, Navarro X, Kennedy WR, et al. The effect of pancreas transplantation on diabetic polyneuropathy. Transplantation 1988;45:368–70. 3 Navarro X, Kennedy WR, Sutherland DER. Autonomic neuropathy and survival in diabetes mellitus: Effects of pancreas transplantations. Diabetologia 1991;34(Suppl.1):S108–12. 4 Bilous RW, Mauer SM, Sutherland DER, Steffes MW. Glomerular structure and function following successful pancreas transplantation for insulin-dependent diabetes mellitus. Diabetes 1987;36:43A. 5 Bilous RW, Mauer SM, Sutherland DER, et al. The effects of pancreas transplantation on the glomerular structure of renal allografts in patients with insulin-dependent diabetes. N Engl J Med 1989;321:80–5.

199

200

DIAGNOSIS AND TREATMENT OF PANCREATIC REJECTION

6 Jacobson SH, Fryd DS, Sutherland DER, Kjellstand CM. Treatment of the diabetic patients with endstage renal failure. Diabetes Metab Rev 1988;4:191–200. 7 Cecka JM, Terasaki PI. The UNOS scientific renal transplant registry. In: Terasaki PI, Cecka JM, ed. Clinical Transplants 1993, Los Angeles:UCLA Tissue Typing Laboratory, 1993:1–18. 8 Sutherland DER, Gores P, Farney A, et al. Evolution of kidney, pancreas and islet transplantation for diabetes at the University of Minnesota. Am J Surg 1993;166:456–91. 9 Sutherland DER. Present status of pancreas transplantation alone in nonuremic diabetic patients. Transplant Proc 1994;26:379–83. 10 Sutherland DER, Moudry-Munns KC, Gruessner A. Pancreas transplant results in United Network for Organ Sharing (UNOS) United States of America (USA) Registry with a comparison to non-USA data in the international registry. In: Terasaki PI, Cecka JM, ed. Clinical transplants 1993. Los Angeles: UCLA Tissue Typing Laboratory, 1993:47–69. 11 Sibley RK. Pancreas Transplantation. In: Sale GE, ed. The pathology of organ transplantation. Boston, Massachusetts, 1990:179–215. 12 Sibley RK, Sutherland DER. Pancreas transplantation: An immunohistologic and histopathologic examination of 100 grafts. Am J Pathol 1987;128:151–70. 13 Nakhleh RE, Gruessner RWG, Swanson PE, et al. Pancreas transplant pathology: A morphologic immunohistochemical, and electron microscopic comparison of allogeneic grafts with rejection, syngeneic grafts, and chronic pancreatitis. Am J Surg Pathol 1991;15:246–56. 14 Nakhleh RE, Sutherland DER. Pancreas rejection: Significance of histopatholgic findings with implication of classification for rejection. Am J Surg Pathol 1992;16:1098–107. 15 Paul LC, Fellstrom B. Overview: Chronic vascular rejection of the heart and the kidney. Have rational treatment options emerged? Transplantation 1992;53:1169–79. 16 Chen H, Wu J, Luo H, Daloze P. Synergistic effect of rapamycin and cyclosporine in pancreaticoduodenal transplantation in the rat. Transplant Proc 1992;24:892–3. 17 Stratta RJ, Sollinger HW, D'Alessandro AM, et al. OKT3 rescue therapy in pancreas-allograft rejection. Diabetes 1989;38:74–8. 18 Dragstedt LR. Some physiologic problems in surgery of the pancreas. Ann Surg 1943;118:576–93. 19 Tyden G, Gunnarsson R, Ostman J, Groth CG. Laboratory findings during rejection of segmental pancreatic allografts. Transplant Proc 1984;16:715–17. 20 Ploeg RJ, D'Alessandro AM, Groshek M, et al. Efficacy of human anodal trypsinogen for detection of rejection in clinical pancreas transplantation. Transplant Proc 1994;26:531–3. 21 Cheng SS, Mun SR. Posttransplant hyperamylasemia is associated with decreased patient and graft survival in pancreas allograft recipients. Transplant Proc 1994;26:428–9. 22 Borgstrom A, Marks WH, Dafoe DC, et al. Immunoreactive anionic and cationic trypsins in serum after experimental porcine pancreatic transplantation. Surgery 1986;100:841–8. 23 Marks WH, Borgstrom A, Sollinger H, Marks C. Serum immunoreactive anodal trypsinogen and urinary amylase as biochemical markers for rejection of clinical whole-organ pancreas allografts having exocrine drainage into the urinary bladder. Transplantation 1990;49:112–15. 24 Douzdjian V, Cooper JL, Abecassis MM, Corry RJ. Markers for pancreatic allograft rejection: Comparison of serum anodal trypsinogen, serum amylase, serum creatinine and urinary amylase. Clin Transplant 1994;8:79–82. 25 Perkal M, Marks C, Lorber MI, Marks WH. A three-year experience with serum anodal trypsinogen as a biochemical marker for rejection in pancreatic allografts. False positives, tissue biopsy, comparison with other markers, and diagnostic strategies. Transplantation 1992;53:415–19. 26 Matsuda K, Ogawa M, Shibata T, et al. Postoperative elevation of serum pancreatic secretory trypsin inhibitor. Am J Gastroenterol 1985;80:694–8.

A.J. KOFFRON AND D.B. KAUFMAN

27 Kitahara T, Takatsuka Y, Fujimoto K, et al. Radioimmunoassay for human pancreatic secretory trypsin inhibitor: Measurement of serum pancreatic secretory trypsin inhibitor in normal subjects and subjects with pancreatic diseases. Clin Chim Acta 1980;103:135–43. 28 Suzuki Y, Kuroda Y, Sollinger HW, et al. Plasma pancreatic secretory trypsin inhibitor as a marker of pancreas graft rejection after combined pancreas–kidney transplantation. Transplantation 1991;52:504–7. 29 Suzuki Y, Kuroda Y, Sollinger HW, Saitoh Y. Plasma phospholipase A2 and pancreatic secretory trypsin inhibitor as markers for pancreas graft rejection. Transplant Proc 1994;26:538–40. 30 Fernstad R, Skoldefors H, Pousette A, et al. A novel assay for pancreatic cellular damage: III. Use of a pancreas-specific protein as a marker of pancreas graft dysfunction in humans. Pancreas 1989;4:44–52. 31 Fernstad R, Tyden G, Brattstrom C, et al. Rejection of pancreas grafts–pancreas-specific protein: New serum marker for graft rejection in pancreas-transplant recipients. Diabetes 1989;38:55–6. 32 Nyberg G, Olausson M, Norden G, et al. Pancreas specific protein (PASP) monitoring in pancreas transplantation. Transplant Proc 1991;23:1604–5. 33 Linder R, Sziegoleit A, Brattstrom C, et al. Pancreatic elastase 1 after pancreatic transplantation. Pancreas 1991;6:31–6. 34 Linder R, Sziegoleit A, Peters B, et al. Serum pancreatic elastase 1 as marker of pancreatic graft damage. Transplant Proc 1990;22:1595. 35 Abendroth D, Capalbo M, Illner WD, et al. Critical analysis of rejection markers sIL-2R, urinary amylase, and lipase in whole-organ pancreas transplantation with exocrine bladder drainage. Transplant Proc 1992;24:786–7. 36 Brattstrom C, Tyden G, Reinholt FP, et al. Markers for pancreas-graft rejection in humans. Diabetes 1989;38:57–62. 37 Konigsrainer A, Tilg H, Reibnegger G, et al. Pancreatic juice neopterin excretion — A reliable marker of pancreas allograft rejection. Transplant Proc 1992;24:907–8. 38 Funakoshi A, Yamada Y, Ito T, et al. Misaki A, Kono M: Clinical usefulness of serum phospholipase A2 determination in patients with pancreatic diseases.Pancreas 1991;6:588–94. 39 Johnson BF, Thomas G, Wiley KN, et al. Thromboxane and prostacyclin synthesis in experimental pancreas transplantation: Changes in parenchymal and vascular prostanoids. Transplantation 1993;56:1447–53. 40 Gliedman ML, Gold M, Whittaker J, et al. Pancreatic duct to ureter anastamosis in pancreatic transplantation. Am J Surg 1973;125:245–52. 41 Sollinger HW, D'Alessandro AM, Stratta RJ, et al. Combined kidney–pancreas transplantation with pancreaticocystostomy. Transplant Proc 1989;21:2837–8. 42 Prieto M, Sutherland DER, Fernandez-Cruz L, et al. Urinary amylase monitoring for early diagnosis of pancreas allograft rejection in dogs. J Surg Res 1986;40:597–604. 43 Prieto M, Sutherland DER, Fernandez-Cruz L, et al. Experimental and clinical experience with urine amylase monitoring for early diagnosis of rejection in pancreas transplantation. Transplantation 1987;43:73–9. 44 Schulak JA, Drevyanko TF. Experimental pancreas allograft rejection: Correlation between histologic and functional rejection and the efficacy of antirejection therapy. Surgery 1985;98:330–6. 45 Munn SR, Engen DE, Barr D, et al. Differential diagnosis of hypoamylasuria in pancreas allograft recipients with urinary exocrine drainage.Transplantation 1990;49:359–62. 46 Nankivell BJ, Allen RDM, Bell B, et al. Factors affecting measurement of urinary amylase after bladderdrained pancreas transplantation.Clin Transplant 1991;5:392–7. 47 Moukarzel M, Benoit G, Charpentier B, et al. Is urinary amylase a reliable index for monitoring whole pancreas endocrine graft function? Transplant Proc 1992;24:925–6.

201

202

DIAGNOSIS AND TREATMENT OF PANCREATIC REJECTION

48 Lievens MM. Potential pitfalls in the determination of amylase activity in the urine of pancreastransplanted patients with bladder drainage. Transplantation 1990;50:526–7. 49 Henry ML, Osei K, O'Dorisio TM, et al. Concomitant reduction in urinary amylase and acute firstphase insulin release predict pancreatic allograft transplant rejection in type I diabetic recipients. Clin Transplant 1991;5:112–20. 50 Munn SR, Engen DE, Barr D, et al. Differential diagnosis of hypoamylasuria in pancreas allograft recipients with urinary exocrine drainage. Transplantation 1990;49:359–62. 51 Benedetti E, Najarian JS, Sutherland DER, et al. Correlation between cystoscopic biopsy results and hypoamylasuria in bladder-drained pancreas transplants. Surgery 1995;118:864–72. 52 Gruessner RWG, Dunn DL, Tzardis PJ, et al. Simultaneous pancreas and kidney transplants versus single kidney transplants and previous kidney transplants in uremic patients and single pancreas transplants in nonuremic diabetic patients. Comparison of rejection, morbidity and long-term outcome. Transplant Proc 1990;22:622–3. 53 Sutherland DER, Casanova D, Sibley RK. Monitoring and diagnosis of rejection: Role of pancreas graft biopsies in the diagnosis and treatment of rejection after pancreas transplantation.Transplant Proc 1987;19:2329–31. 54 Casanova D, Gruessner R, Brayman K, et al. Retrospective analysis of the role of pancreatic biopsy (open and transcystic technique) in the management of solitary pancreas transplants. Transplant Proc 1993;25:1192–3. 55 Allen RDM, Wilson TG, Grierson JM, et al. Percutaneous biopsy of bladder-drained pancreas transplants. Transplantation 1991;51:1213–16. 56 Martinenghi S, Dell'Antonio G, Secchi A, et al. Percutaneous microbiopsy for the diagnosis of rejection in whole bladder-diverted pancreas transplantation. Transplant Proc 1991;26:526. 57 Perkins JD, Munn SR, Marsh CL, et al. Safety and efficacy of cystoscopically directed biopsy in pancreas transplantation.Transplant Proc 1990;22:665–6. 58 Gruessner RWG, Nakhleh R, Gruessner A, et al. Streptozotocin-induced diabetes mellitus in pigs. Horm Metab Res 1993;25:199–203. 59 Jones JW, Nakhleh RE, Casanova D, et al. Cystoscopic transduodenal pancreas transplant biopsy: A new needle. Transplant Proc 1994;26:527–8. 60 Lowell JA, Bynon JS, Nelson N, et al. Improved technique for transduodenal pancreas transplant biopsy. Transplantation 1994;57:752–3. 61 Nakhleh RE, Sutherland DER, Tzardis P, et al. Correlation of rejection of the duodenum with rejection of the pancreas in a pig model of pancreaticoduodenal transplantation. Transplantation 1993;56:1353–6. 62 Nakhleh RE, Sutherland DER, Tzardis P, et al. Correlation of rejection of the duodenum with rejection of the pancreas in a pig model of pancreaticoduodenal transplantation. Transplantation 1993;56:1353–6. 63 Nakhleh RE, Sutherland DER, Benedetti E, et al. Diagnostic utility and correlation of duodenal and pancreas biopsy tissues in pancreaticoduodenal transplants with emphasis on therapeutic use. Transplant Proc 1995;27:1327–8. 64 Nakhleh RE, Sutherland DER, Benedetti E, et al. Diagnostic utility and correlation of pancreatic and duodenal cystoscopically directed biopsies of pancreaticoduodenal transplants. Mod Pathol 1994;7:134A. 65 Reinholt FP, Tyden G, Bohman SO, et al. Pancreatic juice cytology in the diagnosis of pancreatic graft rejection. Clin Transplant 1988;2:127–33. 66 Klima G, Margreiter R, Konigsrainer A, et al. Pancreas juice cytology (PJC) for early detection of pancreas allograft rejection. Transplant Proc 1989;21:2782–3.

A.J. KOFFRON AND D.B. KAUFMAN

67 Kubota K, Reinholt FP, Tyden G, et al. Findings in pancreatic juice cytology compared with histologic findings in the pancreatic graft. Transplant Proc 1990;22:670. 68 Radio SJ, Stratta RJ, Taylor RJ, Linder J. The utility of urine cytology in the diagnosis of allograft rejection after combined pancreas–kidney transplantation. Transplantation 1993;55:509–16. 69 Kendall T, Radio SJ, Stratton RJ, et al. Confirmation of acute pancreas rejection by core biopsy in patients monitored by urine cytology. Mod Pathol 1994;7:132A. 70 Klima G, Margreiter R. Pancreatic juice cytology in the monitoring of pancreas allografts. Transplantation 1989;48:980–5. 71 Radio SJ, Stratta RJ, Taylor RJ, Linder J. The utility of urine cytology in the diagnosis of allograft rejection after combined pancreas–kidney transplantation. Transplantation 1993;55:509–16. 72 Patel B, Markivee CR, Manhanta B, et al. Pancreatic transplantation: Scintigraphy, US, and CT. Radiology 1988;167:685–7. 73 Letourneau JG, Maile CW, Sutherland DER, Feinberg SB. Ultrasound and computed tomography in the evaluation of pancreatic transplantation. Radio Clin N Am 1987;25:345–55. 74 Letourneau JG. Sonography, CT, and MRI of pancreas allografts. In: Letourneau JG, Day DL, Ascher NL, ed. Radiology of organ transplantation. St Louis: Mosby-Year Book, 119:257–66. 75 Nghiem DD, Ludrosky L, Young JC. Evaluation of pancreatic circulation by duplex color Doppler flow sonography. Transplant Proc 1994;26:466. 76 Moulton JS, Munda R, Weiss MA, Lubberg DJ. Pancreatic transplants: CT with clinical and pathologic correlation. Radiology 1989;172:21–6. 77 Yuh WTC, Hunsicker LG, Sato Y, et al. Application of magnetic resonance imaging in pancreas transplant. Diabetes 1989;38:27–9. 78 Yuh WTC, Wiese JA, Abu-Yousef MM, et al. Pancreatic transplant imaging. Radiology 1988;167:679–83. 79 Gruessner RWG, Sutherland DER. Clinical diagnosis in Pancreatic rejection. In: Solez K, Racusen LC, Billingham ME, ed. Solid organ transplant rejection, mechanisms, pathology and diagnosis. New York: Marcel Dekker, 1996:455–99. 80 Sutherland DER. Immunosuppression for clinical pancreas transplantation. Clin Transplant 1991; (special issue) 5:549–53. 81 Sollinger HW, Knechtle SJ, Reed A, et al. Experience with 100 consecutive simultaneous kidney–pancreas transplants with bladder drainage. Ann Surg 1991;214:703–11. 82 Stratta RJ, Taylor RJ, Bynon SJ, et al. Patterns of rejection after combined pancreas–kidney transplantation. Transplant Proc 1994;26:524–5. 83 Brennan DC, Flavin K, Lowell JA, et al. A randomized, double-blinded comparison of Thymoglobulin versus ATGAM for induction immunosuppressive therapy in adult renal transplant recipients. Transplantation 1999;67:1011–18. 84 Gaber AO, First MR, Tesi RJ, et al. Results of the double-blind, randomized, multicenter, phase III clinical trial of thymoglobulin versus ATGAM in the treatment of acute graft rejection episodes after renal transplantation.Transplantation 1998;66:29–37.

203

This page intentionally left blank

Chapter 16

Histology of the pancreas transplant David K.Klassen

Introduction The early detection and effective treatment of acute rejection has been a major challenge in pancreas transplantation. Numerous non-invasive markers of pancreas rejection and function have been studied such as serum or urine levels of amylase, lipase, or anodal trypsinogen or functional measures such as glucose disappearance rates. [1–3]. Although widely used, none of these biochemical markers or functional measures have proven to be specific for acute rejection. Similarly imaging studies such as magnetic resonance imaging (MRI) scanning or ultrasound have not shown adequate sensitivity or specificity on which to base treatment of rejection [4–6]. The diagnosis of acute pancreas rejection has until recently relied on evidence of renal allograft rejection in patients receiving simultaneous kidney and pancreas transplantation. In these patients, renal allograft dysfunction is used as a surrogate marker for pancreas rejection as most rejection episodes occur simultaneously in both the kidney and pancreas, and the renal allograft is easily biopsied. This approach is limited to patients receiving a kidney and pancreas from the same donor. It is however not completely reliable since a significant incidence of isolated pancreas rejection in simultaneous kidney and pancreas recipients has been shown to occur [7]. The significantly lower graft survival rates of isolated pancreas transplants, that is pancreas after kidney and pancreas transplant alone, has been attributed at least in part to the difficulty in diagnosing acute rejection. As a result relatively few isolated pancreas transplants are done compared to simultaneous pancreas and kidney transplantations procedures [8]. The use of simultaneous kidney and pancreas transplantation has become more difficult as recipient waiting lists lengthen and issues of equitable organ donation to diabetic patients have arisen. Histological assessment of allograft biopsies has been the basis of rejection diagnosis in most solid organ transplants, but has been only recently been applied to pancreas transplantation. In the past, direct histological assessment of pancreas rejection was largely based on the evaluations of specimens obtained from open surgical biopsies or from transplant pancreatectomy specimens [9–11]. In bladder-drained pancreas transplants, cystoscopically obtained transduodenal pancreas allograft biopsies have been used [12]. All of these procedures are relatively invasive and these specimens were only infrequently obtained and often late in the course of acute rejection episodes. Biopsies obtained in this fashion provided little information on the histology of early pancreas rejection. Antirejection therapy is less likely to be effective when the diagnosis of rejection is delayed. As a result, our understanding of the histology of pancreas allograft rejection and the relationship of pancreas allograft histological findings to changes in clinically available biochemical markers of pancreas inflammation was limited. Because of the difficulty in obtaining biopsy material, the clinical application of this information was limited. These early studies of the histology of pancreas allograft rejection included patients with silastic and prolamine duct obstructed grafts. This resulted in non-specific histological changes which complicated the interpretation of such biopsies. More recently the development of percutaneous pancreas allograft biopsy, and the demonstrated safety of this technique, has allowed the application of histological assessments in the diagnosis of pancreas allograft dysfunction and rejection

206

HISTOLOGY OF THE PANCREAS TRANSPLANT

[13,14]. Previous impressions that percutaneous core biopsy of the pancreas was hazardous because of excessive risks of bleeding, fistula development, and pancreatitis, have not been borne out. In experienced hands, a single 18-gauge percutaneously obtained core needle biopsy of a pancreas allograft has a very high yield of tissue adequate for histological diagnosis and is well tolerated by patients [7]. The use of percutaneous pancreas allograft biopsy has allowed an early diagnosis of acute rejection and provided data on the evolution of pancreas allograft rejection over time. The technique can be repeated as often as clinically indicated and can be applied in protocol biopsies for clinical research. In some transplant programmes protocol biopsies have been adopted as a component of standard clinical practice [15]. Routine use of percutaneous pancreas allograft biopsy has allowed the graft survival of isolated pancreas transplants to approach that obtained in simultaneous kidney pancreas recipients at some centres [16]. The clinical utility of pancreas allograft biopsy is dependent on the ability to assess the histology of the biopsy specimen accurately and reproducibly, and provide an interpretation which is predictive of clinical outcome. Drachenberg et al. at the University of Maryland have analysed a large number of percutaneous pancreas allograft biopsies obtained early in the course of acute rejection as well as in later stages of rejection. They have developed a grading scheme for acute pancreas rejection which is reproducible and which is predictive of the response to therapy and graft survival [17].

Acute rejection Drachenberg et al. demonstrated that in pancreas allograft recipients the earliest and most subtle evidence of rejection consists of septal inflammatory infiltrates and associated perivenular infiltrates. Subsequently, rejection evolves to include other interlobular structures followed by the acinar parenchyma and arterial structures. These results were consistent with studies in animal models of unmodified pancreas rejection which also suggested that the first histological changes in rejection occur in the fibrous septal areas with infiltrates around veins and capillaries followed by progressive acinar involvement and arterial endotheliitis [18–20]. Prior work by Sibley, Nakhleh, and Sutherland had retrospectively examined the histological features of failed and functioning pancreatic grafts in patients. They identified histological features associated with a high probability of graft failure. Their studies of grafts with advanced degrees of rejection based the diagnosis of rejection on the identification of arterial endotheliitis and vasculitis. The University of Maryland histological criteria for classification of acute rejection were developed from the systematic assessment of 26 histological features. Biopsies from patients with a clinical diagnosis of acute rejection were compared with those from non-rejecting pancreas allograft biopsies obtained as protocol biopsies as well as with samples from native pancreata with a variety of non-immunological diseases. Venous endotheliitis, the presence of eosinophils, activated lymphocytes, acinar inflammation, and septal inflammation were all significantly associated with acute rejection. Septal inflammation was the most common finding in acute rejection being observed in 92 per cent of biopsies. Acinar inflammation and ductal inflammation were observed in 71 and 64 per cent of biopsies, respectively. Eosinophils were commonly seen and were identified in the fibrous septa and acini in 82 per cent of biopsies with features of rejection. Arteritis and arterial endotheliitis were associated almost exclusively with rejection, but occurred very infrequently. Ductal inflammation was seen in both rejection and in various native pancreas disease states, but was not seen in protocol biopsies of pancreas allografts. Ductal inflammation can thus not be considered specific for rejection. The presence of neutrophils, plasma cells, ductal cell necrosis or atypia, and interstitial oedema were not significantly different between biopsy samples from patients with rejection and those with other pathological processes. Coagulation necrosis of the acinar

D.K. KLASSEN

parenchyma and enzymatic fat necrosis was found in early graft thrombosis without significant inflammatory infiltrates. Enzymatic necrosis associated with a prominent infiltrate of neutrophils was seen in acute non-rejection mediated pancreatitis. Based on these histological findings a grading scheme was constructed with six levels of acute rejection ranging from normal (grade O) to severe rejection (grade V) (Table 16.1). Grade I findings termed ‘inflammation of undetermined significance’ consist of sparse, purely septal lymphocytic infiltrates. There is no venous endotheliitis or acinar involvement (Fig. 16.1). The initial impression that any degree of septal lymphocytic infiltrate represented rejection was discarded when protocol allograft biopsies were found to exhibit similar changes in a significant number of cases. In rare cases, sparse septal infiltrates were noted to proceed to the development of rejection or occur during the resolution of rejection following treatment. Patients with grade I biopsies had a uniformly good outcome

Table 16.1 University of Maryland acute rejection grading scheme. Grades range from normal (grade 0) to severe acute rejection (grade V) Grade 0: Normal Unremarkable pancreatic parenchyma without inflammatory infiltrates Grade I:

Inflammation of undetermined significance Sparse, purely septal mononuclear inflammatory infiltrates No venous endotheliitis or acinar involvement identified

Grade II: Minimal rejection Purely septal inflammation with venous endotheliitis (attachment of lymphocytes to the endothelium with associated endothelial damage and lifting of the endothe lium from the basement membrane) In the absence of venous endotheliitis a constellation of at least three of the following four histological features (A) septal inflammatory infiltrates composed of a mixed population of small and large (‘activated’) lymphocytes (B) eosinophils (C) acinar inflammation in rare (up to two) focia (D) ductal inflammation (permeation of inflammation cells through the ductal basement membrane) Grade III: Mild rejection Septal inflammatory infiltrates composed of a population of small and large (‘activated’) lymphocytes with associated acinar inflammation in multiple (3 or more) focia Eosinophils, venous endotheliitis, ductal inflammation, and evidence of acinar single cell injury may be seen depending on sampling. The latter is manifested as cellular drop-out (apoptosis-pyknotic cell death), or necrosis (oncotic cell death) Grade IV: Moderate rejection Arterial endotheliitis and/or necrotizing arteritis (vasculitis). Features described in grade III are usually present Grade V: Severe rejection Extensive acinar lymphoid or mixed inflammatory infiltrates with multicellular focal or confluent acinar cell necrosis Depending on sampling vascular and ductal lesions may be demonstrated aInflammatory

focus is defined as a collection of at least 10 mononuclear cells.

207

208

HISTOLOGY OF THE PANCREAS TRANSPLANT

Fig. 16.1 Pancreas allograft grade I inflammation of undetermined significance. Note sparse septal lymphocytic infiltrates without acinar involvement. Increasing degree of infiltration with venous endotheliitis is grade II.

with no graft loss over 19 months of follow-up. Biopsies with purely septal inflammation associated with venous endotheliitis and occasionally rare foci of acinar inflammation were termed grade II — ‘minimal rejection’. These biopsies showed no evidence of acinar lymphocytic infiltrates. Progression of the rejection process to include three or more foci of acinar inflammation were considered grade III — ‘mild rejection’ (Fig. 16.2). Patients with grade II and grade III biopsies were found to have an increased incidence of graft loss over the period of follow-up when compared to grades 0 and I. These patients with minimal or mild forms of rejection require antirejection treatment. Patients with grade I findings in most cases do not merit acute antirejection therapy but should be carefully observed for biochemical evidence of continued inflammation. Such patients may merit follow-up biopsies. Advanced stages of rejection, (grades IV and V) are characterized by progressive arterial endotheliitis and necrotizing vasculitis respectively (grade IV) (Fig. 16.3) which ultimately results in multicellular focal or confluent acinar cell necrosis (grade V) (Fig. 16.4). These patients have a relatively high inci-

Fig. 16.2 Grade III mild acute rejection in a pancreas allograft. Acinar lymphocytic infiltrates of T lymphocytes and eosinophils are present.

D.K. KLASSEN

Fig. 16.3 Grade IV pancreas allograft rejection. Diffuse lymphocytic infiltrates spread beyond the septal areas into acinar tissue. There is arterial vasculitis noted in septal vessels.

Fig. 16.4 Grade V pancreas allograft rejection. There is fibrinoid necrosis of arterial structures and acinar cell necrosis.

dence of graft loss secondary to rejection, with graft losses of 38 per cent for grade 4 and 100 per cent for grade 5, over the period of follow-up despite the use of aggressive antirejection therapy. The higher grades of rejection (grades III to V) generally caused no diagnostic problems. Rarely, patients with normal biopsies have had persistent evidence of allograft dysfunction and have shown some degree of rejection on subsequent biopsies. This suggested the possibility of sampling error. The incidence of sampling error in these pancreas biopsies however is probably low. In 28 simultaneous pancreas and kidney recipients with evidence of pancreas dysfunction who had simultaneous kidney and pancreas biopsies, 18 cases showed simultaneous rejection in both organs, four cases showed isolated pancreas rejection, and two cases showed isolated renal rejection. This suggests that the potential for false-negative pancreas biopsies based on sampling error is likely no greater than that of renal allograft biopsies. Of note, the presence of peripancreatic abscesses or intra-abdominal infection may cause changes in the superficial pancreatic tissue. These changes may consist of septal inflammatory infiltrates and increased fibrosis and occasional evidence of acinar inflammation. These findings are limited to

209

210

HISTOLOGY OF THE PANCREAS TRANSPLANT

superficial portions of the biopsy core but may on occasion be confused with acute cellular rejection. The histological interpretation of such biopsies must take into consideration the clinical context. Many pancreas transplant programme use levels of serum amylase and lipase to screen for evidence of the development of acute rejection. The correlation between changes in serum amylase and lipase and the grade of histological rejection on biopsy has been studied by Papadimitriou et al. [21]. The levels of serum amylase and lipase rose proportionally to the degree of exocrine parenchymal injury. Rising serum lipase was more closely correlated with the histological grade of rejection than was serum amylase. Because of the wide variation in serum enzyme levels between individual patients the rejection grade could not be distinguished based on enzyme levels alone between individuals. Neither could rejection be distinguished from non-rejection mediated pancreactic parenchymal damage based on enzyme levels (Table 16.2). The level of serum glucose showed no overall statistical relationship to the degree of pancreas rejection, although hyperglycaemia was more common in patients with grade IV and V rejection. The response to therapy with steroids or antilymphocyte antibody and the 1-year graft survival in relation to biopsy grade is shown in Table 16.3. The response to therapy was defined as return of serum pancreatic enzymes to baseline. The poor response seen in patients with grade 0 and I rejection reflects a non-immunological basis of pancreatic parenchymal injury in these cases. The best responses to therapy with steroids alone were seen in the grade II and III rejection cases. Grade IV rejection had a poor response to steroid therapy as the sole modality. The addition of antilymphocyte antibody therapy to corticosteroids resulted in improved graft survival particularly in the grade IV rejection group. One-year graft survival correlated with biopsy rejection grade and decreased with increasing grade of histological acute rejection.

Chronic rejection The clinical picture of a chronically failing pancreas allograft can be difficult to appreciate. Many patient’s initial presentation is overt hyperglycaemia which may be unmasked by infection or other physiological stress. There is no marker of progressive loss of pancreatic functional reserve similar to serial measurements of serum creatinine or glomerular filtration rate in kidney transplantation. Potentially reversible causes of hyperglycaemia such as the development of insulin resistance due to corticosteroids or impaired insulin release from excessive tacrolimus or ciclosporin must be excluded. Patients with chronic rejection and no superimposed acute rejection do not show signs of allograft inflammation and thus may have normal serum amylase and lipase levels. C-peptide levels may be low but are still measurable and may in fact be within the normal range. There are characteristic but not specific radiological findings associated with chronic pancreas graft rejection. These include a proTable 16.2 The relationship between acute rejection biopsy grade and serum amylase, lipase and glucose at the time of biopsy. Amylase and lipase values are expressed as international units per liter. Serum glucose values are expressed as mg/dl (from [21] with permission) Grade 0 I II III IV V

Amylase 285 (± 67) 165 (± 81) 172 (± 76) 192 (± 78) 214 (± 92) 186 (±116)

Lipase 961 618 756 838 932 1009

(± (± (± (± (± (±

342) 412) 386) 396) 472) 592)

Glucose 155 106 115 116 102 167

(± (± (± (± (± (±

17) 20) 19) 20) 23) 29)

Table 16.3 The relationship between acute rejection biopsy grade and the response to therapy with corticosteroids (CS) or antilymphocyte antibody (Ab) therapy. One-year graft survival is expressed as per cent (adapted from [21]) Rejection grade 0 I II III IV V

Number 23 32 30 48 11 7

% response to treatment CS CS/Ab 0 33 36 50 86 89 68 85 0 71 – 17

1-year graft surviving 100 88 100 83 80 40

211

212

HISTOLOGY OF THE PANCREAS TRANSPLANT

gressive decrease in graft size on ultrasound and MRI, and evidence of decreased parenchymal perfusion. Chronic rejection in the pancreas allograft is histologically characterized by fibrosis of lobular tissue and loss of acinar and islet tissue. Studies of patients with chronic rejection who have had serial pancreas allograft biopsies have demonstrated that this interstitial fibrosis is progressive over time [22]. Often active fibroblastic proliferation is noted together with chronic inflammation and oedema. The pancreatic lobules become progressively fragmented at the periphery showing a ragged border, and develop fibrosis and atrophy. In contrast normal pancreas parenchyma is characterized by exocrine lobules with smooth contours separated by relatively inconspicuous fibrous septa. Large vessels and ducts are accompanied a proportional amount of fibrous tissue. In chronically rejecting pancreas allografts, evidence of chronic vascular rejection is frequently seen, similar to that found in renal, cardiac, and hepatic allografts with chronic rejection [10,23,24]. It is characterized by variable narrowing of the arterial lumen with concentric fibroproliferative intimal endarteritis. Low-grade lymphocytic infiltration is commonly seen and clear-cut evidence of acute cellular rejection is also frequently seen and coexisted in over 80 per cent of biopsies with chronic rejection at the University of Maryland. This may be related to a selection bias since most biopsies were done to evaluate abnormalities in serum amylase and lipase. In contrast to acute rejection, biochemical markers of inflammatory tissue destruction such as serum amylase or lipase are normal or minimally elevated in the absence of superimposed acute rejection. Evidence of chronic rejection was observed as early as 2 month posttransplant. The findings of progressive fibrosis, and acinar loss consistent with developing chronic rejection, are predictive of ultimate graft failure. A classification scheme for chronic rejection diagnosed with needle biopsy has been proposed and is based on the amount of fibrosis observed relative to the total volume of tissue in the biopsy sample (Table 16.4) [24]. This grading scheme correlates with clinical outcome. The time to graft failure decreases with increasing biopsy grade. Grade I chronic rejection is characterized by the thickening of interlobular fibrous septa. Lobules remain smooth although there may be focal acinar atrophy (Fig. 16.5). Progression of the fibrotic process and the development of grade II histological chronic rejection is characterized by fibrous septal expansion to include 30 to 40 per cent of the biopsy core. Lobules demonstrate atrophy and fragmentation at their periphery (Fig. 16.6). Moderate chronic rejection (grade III) has fibrosis pro-

Table 16.4 Histological grading scheme for chronic pancreas allograft rejection developed at the University of Maryland. Grades are based on the degree fibrosis in percutaneous needle biopsies (from [24]) C–0

C–I

C–II

C–III

C–IV

Normal pancreas parenchyma Lobules have smooth contours separated by thin fibrous septa Minimal chronic rejection Fibrous septa are accentuated. Lobules are smooth with focal acinar atrophy Mild chronic rejection Lobules are atrophic in their periphery. Fibrous septa represent 30 to 40% of tissue core Moderate chronic rejection Fibrous tissue is more than 50% of tissue core. Significant atrophy of lobules Severe chronic rejection Fibrotic core with rare residual pancreatic structures

D.K. KLASSEN

Fig. 16.5 Grade I chronic rejection of a pancreas allograft. There is mild accentuation of interlobular fibrous septa. Lobules are smooth with some focal atrophy.

Fig. 16.6 Grade II chronic rejection of a pancreas allograft. Fibrosis is increased to 30 per cent or more of the biopsy. Lobules appear fragmented and atrophic at their periphery. Increasing fibrosis to greater than 50 per cent is designated grade III.

Fig. 16.7 Grade IV severe chronic pancreas allograft rejection. The biopsy core is nearly all fibrous tissue. Only residual pancreatic acinar tissue is apparent.

213

214

HISTOLOGY OF THE PANCREAS TRANSPLANT

gressing to include over 50 per cent of the biopsy sample. In severe chronic rejection (grade IV, Fig. 16.7) only residual acinar or islet structures remain.

Recurrent diabetes mellitus Clinical and histological evidence of recurrent autoimmune-mediated diabetes mellitus in pancreas allografts recipients has been reported by Sibley and Sutherland [25]. Their initial report described four patients, three of whom were recipients of pancreas transplants from their identical twin sibling and one of whom was from an HLA identical non-twin sibling. The identical twin recipients were not treated with any immunosuppression. They developed complete loss of pancreatic function within 6 to 12 weeks following transplant. Histologically these grafts showed mononuclear cell infiltrates centred upon the islets consisting of varying numbers of T lymphocytes and monocytes. Follow-up biopsies revealed a selective destruction of islet ␤-cells. The fourth patient was initially treated with low dose ciclosporin. A biopsy for evaluation of pancreas allograft dysfunction showed isletitis and the patient ultimately resumed insulin therapy although he did retain some degree of pancreatic function based on the amount of insulin required to maintain euglycaemia. However, pancreas allografts in patients treated with current immunosuppressive regimens appear to be relatively resistant to recurrent cell-mediated diabetes mellitus. Drachenberg et al. systematically evaluated the islets in 100 percutaneous needle biopsies [22]. In this group of biopsies, islet inflammation was never seen independently of allograft rejection, and was always associated with inflammation of the surrounding acinar structures. The degree of islet inflammation and occasional evidence of islet necrosis correlated with degree of acute rejection and was seen only in higher grades of acute rejection. Isolated isletitis suggestive of a selected cell-mediated cytotoxicity against ␤-cells was not observed.

Drug toxicity Both ciclosporin and tacrolimus are known to be associated with the development of post-transplant diabetes mellitus. Animal studies have shown morphological abnormalities in the islets of animals receiving these drugs [26].Diminished ␤-cell density, decreased insulin synthesis, and secretion have been reported. Similar findings have now been reported in recipients of pancreas allografts treated with ciclosporin and tacrolimus. Islet cell morphology was studied in patients with pancreas allografts randomized to receive either ciclosporin or tacrolimus [27]. These histological findings were compared with samples from patients receiving only corticosteroids or from native pancreatectomy specimens. Patients treated with tacrolimus were noted to have islet cells with striking cytoplasmic swelling and vacuolization. Diminished immunoperoxidase staining for insulin also was noted in the majority of biopsies from the tacrolimus-treated patients. Apoptosis was demonstrated by in situ hybridization in occasional cells. Apoptosis was absent in islets from patients not treated with tacrolimus or cicylosporin, and immunoperoxidase staining for insulin was normal. Islets from patients receiving ciclosporin showed the same changes as patients treated with tacrolimus, but to a lesser degree. Electron microscopic studies showed a marked decrease or complete absence of endocrine secretory granules most notably in islets from patients treated with tacrolimus. The periphery of the islets, an area composed of predominantly of non-␤-cells was lss affected than the central portions. Overall, the islets from patients receiving tacrolimus and ciclosporin both had an average of 40 to 50 per cent of cells showing degenerative changes. Islet cell vacuolization correlated with both the mean and peak levels of tacrolimus and ciclosporin prior to the biopsy. Follow-up biopsies in a small number of

D.K. KLASSEN

patients found reductions in drug dose or changing to alternative therapies resulted in substantial resolution of the tacrolimus or ciclosporin induced histological changes.

Cytomegalovirus Infection Although, tissue invasive cytomegalovirus (CMV) infection is a significant problem in solid organ transplantation, it has only occasionally been reported in pancreas allografts [10,28]. A histological diagnosis of CMV pancreatitis is made by the identification of cytoplasmic or intranuclear inclusions and cytomegaly in epithelial, endothelial, or stromal cells. Histologically identifiable CMV allograft pancreatitis can coexists with acute rejection. This complicates the diagnosis. Acinar cell inflammation in acute allograft rejection and in CMV allograft pancreatitis are indistinguishable from each other if virally infected cells are not demonstrated. Systematic evaluation of multiple sections and confirmation of the diagnosis by immunoperoxidase staining for CMV antigens is appropriate. Patients with pancreas biopsies showing non-specific inflammation, but with a clinical suspicion of CMV infection will occasionally have positive immunoperoxidase staining for CMV antigens even in the absence of typical viral cytopathic changes. Cytomegalovirus pancreatitis has been demonstrated predominantly in patients serologically negative for CMV prior to transplantation who received allografts from CMV positive donors. Repeat biopsies following prolonged courses of antiviral medication have demonstrated resolution of the CMV-related histological changes. If rejection and CMV infection co-exist, concurrent therapy with antirejection medication and intravenous antiviral medication has resulted in resolution of acute rejection and the CMV pancreatis on follow-up biopsies with maintenance of allograft function.

Post-transplant lymphoproliferative disease in pancreas allografts The development of Epstein–Barr virus related post-transplant lymphoproliferative disease (PTLD) within the pancreas allograft and its histological characteristics has been reported [29]. In these cases the main diagnostic issue is differentiating PTLD from severe acute rejection. Differentiation between these two entities is important in the early stages of Epstein–Barr virus related PTLD because avoiding unnecessary treatment with immunosuppression is crucial for effective therapy of the lymphoproliferative process. In four cases evaluated at the University of Maryland, the histological features characteristic of lymphoma were nodular and expansile infiltrates composed of a high proportion of atypical plasmacytoid B-cells making up to 40 to 70 per cent of the infiltrate. Reed–Sternberg like cells were noted in two patients. The infiltrates in the parenchyma were random with no apparent affinity for the acinar tissue. Arterial structures were not involved in PTLD unless there was concurrent vascular rejection. Foci of necrosis and infiltration of venous walls were seen with PTLD as well as acute rejection. Biopsy features suggesting rejection included an infiltrate consisting of mixed small and large lymphocytes, the majority of which are T cells with a smaller component of mature plasma cells, and variable numbers of eosinophils. Cytologically atypical cells suggests PTLD. For screening purposes, immunoperoxidase stains for B and T cells are extremely helpful. The dense aggregates of cells noted in the PTLD patients were never seen in biopsy specimens from patients with pure rejection. In situ hybridization for Epstein–Barr virus encoded RNA was positive only in samples from the patients with PTLD. Demonstration of immunoglobulin light chain restriction leads to an unequivocal diagnosis of lymphoproliferative disorder, but is not seen in all cases of PTLD. Thus the assessment of the morphological differences between PTLD and acute rejection coupled with the selective use of stain-

215

216

HISTOLOGY OF THE PANCREAS TRANSPLANT

ing for B and T lymphocytes and Epstein–Barr viral antigens, allows the correct diagnosis of PTLD in pancreas allografts even in the small tissue samples from needle biopsies in most cases.

References 1 Sutherland DER, Gruessner RWG, Gores PF. Pancreas and islet transplantation: an update. Transplant Rev 1994;8:185–206. 2 Perkal M, Marks C, Lorber MI, et al. A three-year experience with serum anodal trypsinogen as a biochemical marker for rejection in pancreatic allografts: false positives, tissue biopsy, comparison with other markers and diagnostic strategies. Transplantation 1992;53:415–19. 3 Elmer DS, Hathaway DK, Bashar Abdulkarim A, et al. Use of glucose disappearance rates (kG) to monitor endocrine pancreas allografts. Clin Transplant 1998;12:56–64. 4 Yuh WT, Hunsicker LG, Nghiem DD, et al. Pancreatic transplants: evaluation with MR imaging. Radiology 1989;170:171–7. 5 Wong JJ, Krebs TL, Klassen DK, et al. Evaluation of acute pancreatic transplant rejection: morphologydoppler analysis versus guided percutaneous biopsy. Am J Roentgenol 1996;166:803–7. 6 Krebs TL, Daly B, Wong-You-Cheong JJ, et al. Acute pancreatic transplant rejection: evaluation with dynamic contrast-enhanced MR imaging compared with histopathology. Radiology 1999;210:437–42. 7 Klassen DK, Hoehn-Saric EW, Weir MR, et al. Isolated pancreas rejection in combined kidney pancreas transplantation: results of percutaneous biopsy. Transplantation 1997;61:974–9. 8 Gruessner AC, Sutherland DER. Analysis of United States (US) and non-US pancreas transplants as reported to the international pancreas transplant registry (IPTR) and to the united network for organ sharing (UNO In Cecka M, Terasaki, PI, ed. Clinical transplants 1998. California: UCLA Tissue Typing Laboratory, Los Angeles, 53–7. 9 Sibley RK, Sutherland DER. Pancreas transplantation: an immunohistologic and histopathologic examination of 100 grafts. Am J Pathol 1982;128:151–70. 10 Sibley RK. Pathology of pancreatic transplantation. In: Sale GE, ed. The pathology of organ transplantation Massachusetts: Butterworths Stoneham, 719–213. 11 Nakhleh RE, Sutherland DER. Pancreas rejection: significance of histopathologic findings with implications of classification of rejection. Am J Surg Pathol 1992;16:1098–107. 12 Perkins JD, Munn SR, Marsh CL, et al. Safety and efficiency of cystoscopically directed biopsy in pancreas transplantation. Transplant Proc 1990;22:665–6. 13 Allen RD, Wilson TG, Grierson JM, et al. Percutaneous biopsy of bladder-drained pancreas transplants. Transplantation 1991;51:1213–16. 14 Gaber AO, Gaber LW, Shokouh-Amiri MH, et al. Percutaneous biopsy of pancreas transplants. Transplantation 1992;54:548–50. 15 Stratta RJ, Taylor RJ, Grune MT, et al. Experience with protocol biopsies after solitary pancreas transplantation. Transplantation 1995;60:1431–7. 16 Bartlett ST, Schweitzer EJ, Johnson LB, et al. Equivalent success of simultaneous pancreas kidney and solitary pancreas transplantation: a prospective trial of tacrolimus immunosuppression with percutaneous biopsy. Ann Surg 1996;24:440–9. 17 Drachenberg CB, Papadimitriou JC, Klassen DK, et al. Evaluation of pancreas transplant needle biopsy: reproducibility and revision of histologic grading system. Transplantation, 1997;63:1579–86. 18 Steiniger B, Klempnauer J. Distinct histologic patterns of acute prolonged and chronic rejection in vascularized rat pancreas allografts. Am J Pathol 1986;124:253–62.

D.K. KLASSEN

19 Carpenter HA, Barr D, Marsh CL, et al. Sequential histopathologic changes in pancreaticoduodenal allograft rejection in dogs. Transplantation 1989;48:764–8. 20 Allen RD, Grierson JM, Ekberg H, et al. Longitudinal histopathologic assessment of rejection after bladder-drained canine pancreas allograft transplantation. Am J Pathol 1991;138:303–12. 21 Papadimitriou JC, Drachenberg CB, Wiland A, et al. Histologic grading of acute allograft rejection in pancreas needle biopsy: correlation to serum enzymes, glycemia and response to immunosuppressive treatment. Transplantation 1998;66:1741–5. 22 Drachenberg CB, Papadimitriou JC, Weir MR, et al. Histologic findings in islets of whole pancreas allografts: lack of evidence for recurrent cell-mediated diabetes mellitus. Transplantation 1996;62:1770–3. 23 Nakhleh RE. Pathology of pancreatic transplantation. In: Solez K, Racusen LC, Billingham ME, ed. Solid organ transplant New York: Marcel Dekker 261–76. 24 Drachenberg CB, Papadimitriou JC, Klassen DK et al. Chronic pancreas allograft and proposal of a grading scheme. Transplant Proc,1999;31:614. 25 Sibley RK, Sutherland DER, Goetz F, et al. Recurrent diabetes mellitus in the pancreas iso- and allograft: a light and electron microscopic and immunohistochemical analysis of four cases. Lab Invest 1985;53:132–44. 26 Hirano Y, Fujihira S, Ohara H, et al. Morphological and functional changes of islets of Langerhans in FK506-treated rats. Transplantation 1992;53:889–94. 27 Drachenberg CB, Klassen DK, Weir MR, et al. Islet cell damage associated with tacrolimus and cyclosporine: morphological features in pancreas allograft biopsies and clinical correlation. Transplantation 1999;68:396–402. 28 Klassen DK, Drachenberg CB, Papadimitriou JC, et al. CMV allograft pancreatitis: diagnosis, treatment and histologic features. Transplantation 2000;69:1968–71. 29 Drachenberg CB, Abruzzo LV, Klassen DK et al. Epstein–Barr virus-related posttransplantation lymphoproliferative disorder involving pancreatic allografts: histologic differentiation from acute allograft rejection. Hum Pathol 1998;29:569–77.

217

This page intentionally left blank

Chapter 17

Pancreas transplantation: effects on secondary complications of diabetes mellitus R. Paul Robertson

The secondary complications of diabetes mellitus are usually categorized as acute or chronic. Acute complications include hyperglycaemia and ketoacidosis as well as hypoglycaemia caused by exogenous insulin treatment. Chronic complications include the subcategories of microvascular and macrovascular disease. Microvascular diseases can lead to retinopathy, nephropathy, and neuropathy. Macrovascular disease can lead to accelerated atherosclerosis, which, in turn, leads to an increased incidence of myocardial infarction, stroke, and peripheral gangrene. Additional complications include various skin lesions (diabetic dermopathy, necrobiosis lipoidica diabeticorum, candidiasis), and skin ulcers. Bones and joints can be affected by diabetic arthropathy, Dupuytren’s contractures, and Charcot joint. More unusual infections include necrotizing fasciitis, necrotizing myositis, and Mucor meningitis. This chapter will focus on the impact of successful pancreas transplantation on the more common secondary complications of diabetes mellitus.

Hypoglycaemia On a day-to-day basis, it is the fear of unexpected hypoglycaemia that weighs most heavily in the minds of diabetic patients requiring insulin treatment. This represents a therapeutic irony since it is the very therapy the diabetic patient becomes dependent upon to prevent severe hyperglycaemia, secondary complications, and death that leads to chronic bouts of hypoglycaemia. Chronic hypoglycaemia desensitizes patients to the symptoms of hypoglycaemia that normally serve as a warning that glucose levels are dropping to dangerously low levels. This presents the patient with the difficult option of choosing less meticulous maintenance of glycaemia and haemoglobin A 1c levels, and thereby running a greater risk of secondary complications of hyperglycaemia. The normal defence mechanisms against insulin-induced hypoglycaemia involve secretion of two hormones, glucagon and epinephrine [1]. Glucagon is released from the ␣-cells of the pancreatic islet within minutes after glucose values reach the 50 to 60 mg/dl range. Soon after, epinephrine is released from the adrenal medullae. Both hormones travel to the liver and stimulate glycogenolysis which releases glucose into the systemic circulation via the hepatic vein. In diabetes mellitus, glucagon responses begin to deteriorate within several years of the disease and thereafter become greatly diminished in most patients [2]. Eventually, the epinephrine response is also compromised although usually not as severely as the glucagon response [3]. The pathogenesis of these secretory defects remain enigmas. Defective glucagon response to hypoglycaemia is primarily a signalling problem since ␣-cells respond to other glucagon stimuli such as aminoacids [4]. Similarly, the abnormal epinephrine

220

EFFECTS ON SECONDARY COMPLICATIONS OF DIABETES MELLITUS

response cannot be ascribed to inadequate epinephrine stores or defective release mechanisms because exercise stimulates normal epinephrine responses in diabetic patients who do not have normal epinephrine responses to insulin-induced hypoglycaemia [5]. The most difficult part of this clinical problem is that patients with recurrent hypoglycaemia gradually lose their ability to sense low circulating glucose levels. Normally, the first set of symptoms to hypoglycaemia include feelings of warmth, palpitations, hunger, and sweating. This is followed by a second set of symptoms that include visual blurring, sleepiness, obtundation, and confusion when the glucose levels are very low for a prolonged period of time. The patient quickly learns to depend upon these symptoms as signals to take corrective measures to restore the glucose level to normal. However, when recurrent hypoglycaemia has reduced the threshold for symptoms, they begin at lower glucose levels so that it is possible for the patient to traverse from normal to seriously low circulating glucose levels without experiencing any symptoms. This makes possible the dangerous situation of going from an asymptomatic state to sudden lethargy, sleepiness, and unconsciousness. Fortunately, it has become appreciated in recent years that evoidance of recurrent hypoglycaemia will correct this problem [6–8]. Careful loosening of glycaemic control and avoidance of hypoglycaemia returns early symptom awareness to diabetic patients. However, this loosening of glycaemic control usually leads to higher degrees of glycaemia which increases the risk of secondary complications of chronic hyperglycaemia. The ideal treatment regimen would avoid both hypoglycaemia and hyperglycaemia, outcomes that are povided by successful pancreas transplantation. The first extensive studies of counter-regulation of hypoglycaemia following pancreas transplantation appeared in the 1990s. In one study, levels of glucose, glucagon, and catecholamines were examined in 38 successful diabetic pancreas recipients and compared with 54 type 1 diabetic non-recipients and 26 non-diabetic control subjects [9]. Glucose recovery after insulin-induced hypoglycaemia was significantly improved in the transplant recipients. Later studies by Barrou et al. documented that hepatic glucose production during hypoglycaemia was increased in successful recipients of pancreas transplantation [10] (Figs 17.1, 17.2). In 1997, Kendall et al. examined the effect of pancreas transplantation on epinephrine secretion and symptom unawareness by performing stepped hypoglycaemic clamp studies in 13 pancreas transplant recipients and matched control subjects [11]. Epinephrine responses (Fig. 17.3) in the recipients were improved but still less than those seen in either healthy control subjects or non-diabetic immunosuppressed kidney transplant recipients. Nonetheless, hypoglycaemic symptom recognition (Fig. 17.4) was significantly greater in pancreas transplant recipients than patients with type 1 diabetes and just as intense in the normal control group. These studies established that successful pancreas transplantation impoves glucagon and epinephrine responses and normalizes hypoglycaemia symptom recognition in patients with long-standing diabetes. The value of hormonal counter-regulatory responses to hypoglycaemia in patients who are not taking exogenous insulin because of successful pancreas transplantation is underscored by evidence that pancreas transplant recipients can sometimes experience hypoglycaemic reactions [12]. This is probably related to the hyperinsulinaemic state that patients experience by virtue of systemic venous drainage of the allograft which bypasses first-pass hepatic degradation of insulin [13,14].

Diabetic retinopathy The first comprehensive study of the effect of successful pancreas transplantation on diabetic retinopathy was published by Ramsey et al. [15] in 1988. Seven-field colour stereophotography and macular fluorescein angiography examinations were conducted in 22 patients with type 1 diabetes mellitus who had been successfully transplanted and compared to 16 similar patients in whom pan-

R.P. ROBERTSON

Fig. 17.1 Glucagon responses during hypoglycaemia induced by an insulin infusion (stepped hypoglycaemic clamp). Glucagon responses in type 1 diabetic patients are absent. After successful pancreas transplantation (PTx), glucagon responses are normal (from [10]).

Fig. 17.2 Hepatic glucose production during hypoglycaemic clamps in control subjects (CONT), successful recipients of pancreas transplantation (PTx), non-transplanted type 1 diabetic patients (IDDM), and non-diabetic successful recipients of kidney transplantation (KTX) (from [10]).

221

222

EFFECTS ON SECONDARY COMPLICATIONS OF DIABETES MELLITUS

Fig. 17.3 Epinephrine responses during a stepped hypoglycaemic clamp in non-diabetic kidney transplant recipients, type 1 diabetic pancreas transplant recipients, and patients with type 1 diabetes mellitus. Secretion of epinephrine during the clamp improves in successful recipients of pancreas transplantation, but does not return to normal levels (from [11]).

creas transplantation had not been successful. The majority of patients in both groups had advanced proliferative retinopathy. At a mean follow-up of 24 months no differences were observed between the two groups. Similarly, Peterson et al. [16] reported a study in eight patients followed for 12 to 49 months after combined pancreas and kidney transplantation. No post-transplantation differences between the two groups were found in visual acuity, macular oedema, capillary closure, preretinal gliosis, neovascularization, or worsening of the severity of retionpathy. Similar findings were published by Schieder and Wang [17]. Consequently, there is no evidence that successful pancreas transplantation and normalization of glycaemia has a beneficial effects on diabetic retinopathy for up to 5 years post-transplantation. It is unfortunate that such patients have not been followed for a longer period of time, as will become evident from the sections below describing the experience with diabetic nephropathy. It is possible that it may take 5 and possibly up to 10 years before the beneficial effects on retinopathy can be observed following normalization of glucose levels by pancreas transplantation.

Diabetic nephropathy In 1985 Bohman et al. [18] compared kidney graft biopsies in six patients who had undergone kidney transplantation with two patients who underwent combined pancreas and kidney transplantation. Five of the six patients receiving kidney only transplantation had changes compatible with diabetic nephropathy in the kidney transplant whereas neither of the two patients who received both pancreas and kidney had changes in the donated kidneys. In 1989, Bilous et al. [19] compared kidney biopsies

R.P. ROBERTSON

Fig. 17.4 Degree of symptomatology during a stepped hypoglycaemic clamp in non-diabetic kidney transplant recipients, type 1 diabetic pancreas transplant recipients, control subjects, and type 1 diabetic subjects. In contrast to the diminished symptom awareness of the type 1 diabetic subjects, diabetic patients undergoing successful pancreas transplantation have normal symptom awareness (from [11]).

in patients who had received a pancreas transplant after a previous kidney transplant with biopsies from patients who received a kidney transplant only. The patients who received pancreas transplants had no progression of disease in the transplanted kidney two years after pancreas transplantation and had smaller glomerular volumes and less mesangial expansion than 13 matched diabetic patients who were recipients of kidney alone transplantation without pancreas transplantation. In 1995 Wilczek et al. [20] published a series comparing 20 diabetic patients who received combined pancreas and kidney to 30 diabetic patients who received kidney only. Using light and electron microscopy, a significantly greater percentage of abnormalities were found in those patients receiving kidney transplant alone compared to those patients receiving both pancreas and kidney. Thus, these three studies amply document that transplanting a pancreas at the time of kidney transplant very effectively serves to protect the transplanted kidney from developing the lesions typical of diabetes mellitus. In 1998 Fioretto et al. [21] published the first longitudinal study in diabetic patients who received pancreas transplants alone. These investigators studied kidney function and performed renal biopsies between pancreas transplantation and 5 and 10 years thereafter in eight patients with type 1 diabetes who did not have uraemia and therefore did not need kidney transplantation. Measures of renal structure included thickness of glomerular basement membrane, thickness of tubular basement membrane, mesangial fractional volume per glomerulus, mesangial matrix fractional volume per glomerulus, and mean glomerular volume. None of these parameters showed improvement when biopsies were taken 5 years post-pancreas transplantation. However, statistically significant improvements were observed in thickness of glomerular basement membrane, thickness of tubular basement membrane, mesangial

223

224

EFFECTS ON SECONDARY COMPLICATIONS OF DIABETES MELLITUS

fractional volume per glomerulus, and mean glomerular volume when kidney biopsies were performed 10 years post-transplantation. These observations established for the first time that abnormal renal structure secondary to chronic diabetes mellitus can be improved by pancreas transplantation and normalization of glucose levels.

Diabetic neuropathy More reports about the effects of pancreas transplantation on diabetic neuropathy have been published than any other chronic complication. Although often looked upon clinically as an annoyance that can sometimes be incapacitating, several publications provide sobering testimony to the potential lethality of diabetic neuropathy, autonomic neuropathy, which is associated with cardiorespiratory arrest and up to 50 per cent mortality within 5 years of onset [22–24]. In 1990 Navarro et al. [25] reported that patients with autonomic neuropathy who successfully received pancreas transplantation had marked decrease in mortality rate during the ensuing 5 years (50 versus 10 per cent). Improvement in autonomic function was soon verified by others in 1991. In that same year several other reports appeared describing improved nerve conduction velocity after successful pancreas transplantation [26–30]. In 1993 Muller-Felber et al. [31] suggested that successful pancreas transplantation is able not only to help the progression of diabetic polyneuropathy but also to improve it to some extent. By 1997 sufficient clinical experience had accumulated to evaluate the long-term effects of pancreas transplantation on diabetic neuropathy. In a publication by Navarro et al. [32] 115 patients with functioning transplanted pancreases were compared to 92 control patients treated with insulin. Data were provided at 1, 2, 3, 5, 5, 7 and 10 years post-transplant. In the control patients, neuropathy progressively worsened during follow-up. In contrast, clinical examination of motor and sensory nerve conduction indices, and autonomic testing in the pancreas transplant group, showed stable improvement. In the same year Allen et al. [33] reported similar findings, concluding that the early transplantation of uraemic diabetic patients before onset of severe neuropathy maximized neurological recovery after simultaneous pancreas and kidney transplantation. Since there have been no reports to the contrary, it can be safely concluded that successful pancreas transplantation, by virtue of normalizing glycaemia, is associated with arrest of the progression of diabetic neuropathy.

Macrovascular disease By contrast with the secondary complications of hypoglycaemia, retinopathy, nephropathy, and neuropathy, very few published studies assessing the beneficial effects of successful pancreas transplantation on macrovascular disease have appeared. This is all the more surprising since coronary artery disease and stroke are a major complication of the disease. In 2000 Biesenbach et al. [34] examined the effect of pancreas kidney transplantation on the progression of macrovascular disease in type 1 diabetic patients with endstage renal disease. Eleven patients receiving simultaneous pancreas kidney transplantation were compared to 10 diabetic patients receiving kidney transplant alone. The patients receiving combined organ transplantation had lower haemoglobin A1c and serum triglyceride levels, but similar cholesterol concentrations and blood pressures as the patients receiving kidney transplantation alone. Despite this difference, no differences between the two groups were found in progression to macrovascular and coronary heart disease or in progression of peripheral vascular disease. In contrast two reports by Larsen et al. [35] and Fiorina et al. [36] reported beneficial effects of pancreas transplantation. In the first report measures of carotid intima media thickness were found to decrease approximately 4 years after a pancreas transplant was performed. Since this measure predicts cardio-

R.P. ROBERTSON

vascular events, the authors predicted a decrease in the overall risk of cardiovascular events. The report by Fiorina et al. evaluated a panel of atherosclerotic risk factors, a measure of endothelial dysfunction, and measurements of intima media thickness. Patients receiving a successful pancreas transplant, but not those receiving a kidney transplant alone, showed a lower atherosclerotic risk profile, normalization of endothelial dysfunction, and reduction of intima media thickness. These initial reports indicate that successful pancreas transplantation may have beneficial effects on atherosclerotic complications of diabetes.

Summary Intensive studies of the impact of pancreas transplantation on the secondary complications of diabetes, including defective hormonal counter-regulation and symptom unawareness of hypoglycaemia, renal structure, and both peripheral and autonomic nerve function, have provided solid evidence that these secondary complications of diabetes mellitus are ameliorated by normalization of glycaemia. The short-term studies that are available for diabetic retinopathy have not indicated a significant benefit. Preliminary studies of macrovascular disease are not in agreement but data exist in the larger studies suggesting that normalization of glycaemia improves metabolic and functional parameters associated with atherosclerotic risk. Consequently, it can be safely concluded that successful pancreas transplantation not only returns the recipient to normal glycaemia but also has substantial impact on quality of life [37–39] by stabilizing some of the secondary complications of diabetes.

References 1 Rizza RA, Cryer PE, Gerich JE. Role of glucagon, catecholamines, and growth hormone in human glucose counterregulation. Effects of somatostatin and combined alpha- and beta-adrenergic blockade on plasma glucose recovery and glucose flux rates after insulin-induced hypoglycemia. J Clin Invest 1979;64:62–71. 2 Bolli G, de Feo P, Compagnucci P, Cartechini MG, Angeletti G, Santeusanio F, et al. Abnormal glucose counterregulation in insulin-dependent diabetes mellitus: interaction of anti-insulin antibodies and impaired glucagon and epinephrine secretion. Diabetes 1983;32:134–41. 3 Kleinbaum J, Shamoon H. Impaired counterregulation of hypoglycemia in insulin-dependent diabetes mellitus. Diabetes 1983;32:493–8. 4 Gerich JE, Langlois M, Noacco C, Karam JH, Forsham PH. Lack of glucagon response to hypoglycemia in diabetes: evidence for an intrinsic pancreatic alpha cell defect. Science 1973;182:171–3. 5 Hirsch BR, Shamoon H. Defective epinephrine and growth hormone responses in type I diabetes are stimulus specific. Diabetes 1987;36:20–6. 6 Dagogo-Jack S, Rattarasarn C, Cryer PE. Reversal of hypoglycemia unawareness, but not defective glucose counterregulation, in IDDM. Diabetes 1994;43:1426–34. 7 Fanelli CG, Epifano L, Rambotti AM, Pampanelli S, Di Vincenzo A, Modarelli F, et al. Meticulous prevention of hypoglycemia normalizes the glycemic thresholds and magnitude of most of neuroendocrine responses to, symptoms of, and cognitive function during hypoglycemia in intensively treated patients with short-term IDDM. Diabetes 1993;43:1683–9. 8 Davis M, Mellman M, Friedman S, Chang CJ, Shamoon H. Recovery of epinephrine response but not hypoglycemic symptom threshold after intensive therapy in type 1 diabetes. Am J Med 1994;97:535–42.

225

226

EFFECTS ON SECONDARY COMPLICATIONS OF DIABETES MELLITUS

9 Diem P, Redmon JB, Abid M, Moran A, Sutherland DE, Halter JB, Robertson RP. Glucagon, catecholamine and pancreatic polypeptide secretion in type 1 diabetic recipients of pancreas allografts. J Clin Invest1990;86:2008–13. 10 Barrou Z, Seaquist ER, Robertson RP. Pancreas transplantation in diabetic humans normalizes hepatic glucose production during hypoglycemia.Diabetes 1994;43:661–6. 11 Kendall DM, Rooney DP, Smets YF, Salazar Bolding L, Robertson RP. Pancreas transplantation restores epinephrine response and symptom recognition during hypoglycemia in patients with long-standing type 1 diabetes and autonomic neuropathy. Diabetes 1997;46:249–57. 12 Cottrell DA, Henry ML, O’Dorisio TM, Tesi RJ, Ferguson RM, Osei K. Hypoglycemia after successful pancreas transplantation in type 1 diabetic patients. Diabetes Care 1991;14:111–13. 13 Diem P, Abid M, Redmon JB, Sutherland DE, Robertson RP. Systemic venous drainage of pancreas allografts as independent cause of hyperinsulinemia in type 1 diabetic recipients. Diabetes 1990;39:534–40. 14 Redmon JB, Teuscher AU, Robertson RP. Hypoglycemia after pancreas transplantation. Diabetes Care 1998;21:1944–50. 15 Ramsay RC, Goetz FC, Sutherland DE, Mauer SM, Robinson LL, Cantrill HL, et al. Progression of diabetic retinopathy after pancreas transplantation for insulin-dependent diabetes mellitus. N Engl J Med 1988;318:208–14. 16 Petersen MR.University of Michigan Pancreas Transplant Evaluation Committee: Progression of diabetic retinopathy after pancreas transplantation. Ophthalmology 1990;97:496–502. 17 Scheider A, Meyer-Schwickerath E, Nusser J, Land W, Landgraf R. Diabetic retinopathy and pancreas transplantation: a 3-year follow-up. Diabetologia 1991;34 (Suppl 1):S95–9. 18 Bohman SO, Tyden G, Wilczek H, Lundgren G, Jaremko G,Gunnarsson R, et al. Prevention of kidney graft diabetic nephropathy by pancreas transplantation in man. Diabetes 1985;34:306–8. 19 Bilous RW, Mauer SM, Sutherland DE, Najarian JS, Goetz FC, Steffes MW. The effects of pancreas transplantation on the glomerular structure of renal allografts in patients with insulin-dependent diabetes. N Engl J Med 1989;321:80–5. 20 Wilczek HE, Jaremko G, Tyden G, Groth CG. Evolution of diabetic nephropathy in kidney grafts. Evidence that a simultaneously transplanted pancreas exerts a protective effect. Transplantation 1995;59:51–7. 21 Fioretto P, Steffes MW, Sutherland DE, Goetz FC, Mauer M. Reversal of lesions of diabetic nephropathy after pancreas transplantation. N Engl J Med 1998;339:69–75. 22 Ewing DJ, Campbell IW, Clarke BF. Mortality in diabetic autonomic neuropathy. Lancet 1976;1:601–3. 23 Page MM, Watkins PJ. Cardiorespiratory arrest and diabetic autonomic neuropathy. Lancet 1978;1:14–16. 24 Ewing DJ, Campbell IW, Clarke BF. The natural history of diabetic autonomic neuropathy. Q J Med 1980;49:95–108. 25 Navarro X, Kennedy WR, Loewenson RB, Sutherland DE. Influence of pancreas transplantation on cardiorespiratory reflexes, nerve conduction, and mortality in diabetes mellitus. Diabetes 1990;39:802–6. 26 Secchi A, Martinenghi S,Galardi G, Comi G, Canal N, Pozza G. Effects of pancreatic transplantation on diabetic polyneuropathy. Transplant Proc 1991;23:1658–9. 27 Aridge D, Reese J, Niehoff M, Carney K, Lindsey L, Chun HS, et al. Effect of successful renal and segmental pancreatic transplantation on peripheral and autonomic neuropathy. Transplant Proc 1991;23:1670–1. 28 Caldara R, Sanseverino R, Lefrancois N, Martin X, Martinenghi S, Dubernard JM. Pancreas transplantation: long-term results. Clin Transplant 1991;5:260–4.

R.P. ROBERTSON

29 Gaber AO, Cardoso S, Pearson S, Abell T, Gaber L, Hathaway D, et al. Improvement in autonomic function following combined pancreas – kidney transplantation. Transplant Proc 1991;23:1660–2. 30 Solders G, Tyden G, Persson A, Groth CG. Improvement of nerve conduction in diabetic neuropathy. A follow-up study 4 yr after combined pancreatic and renal transplantation. Diabetes 1992;41:946–51. 31 Muller-Felber W, Landgraf R, Scheuer R, Wagner S, Reimers CD, Nusser J, et al. Diabetic neuropathy 3 years after successful pancreas and kidney transplantation. Diabetes 1993;42:1482–6. 32 Navarro X, Sutherland DE, Kennedy WR. Long-term effects of pancreatic transplantation on diabetic neuropathy. Ann Neurol 1997;42:727–36. 33 Allen RD, Al-Harbi IS, Morris JG, Clouston PD, O’Connell PJ,Chapman JR et al. Diabetic neuropathy after pancreas transplantation: determinants of recovery. Transplantation 1997;63:830–8. 34 Biesenbach G, Margreiter R, Konigsrainer A, Bosmuller C, Janko O, Brucke P, et al. Comparison of progression of macrovascular diseases after kidney or pancreas and kidney transplantation in diabetic patients with end-stage renal disease. Diabetologia 2000;43:231–4. 35 Larsen JL, Lynch TG, Leone JP, Erickson JM, Mack-Shipman LR, Lane JT, et al. Carotid intima media thickness (IMT) decreases after pancreas transplantation (PTX). Diabetes 2000;49:A30. 36 Fiorina PLRE, Massimo V, Minicucci F, Fermo I, Paroni R, Sblendido M, et al. Effects of kidney – pancreas transplantation on atherosclerotic risk factors and endothelial dysfunction in IDDM uremic patients. Diabetes 2000;49:A30. 37 Zehrer CL, Gross CR. Comparison of quality of life between pancreas/kidney and kidney transplant recipients: 1-year follow-up. Transplant Proc 1994;26:508–9. 38 Piehlmeier W, Bullinger M, Nusser J, Konig A, Illner WD, Abendroth D, et al. Quality in life in type 1 (insulin-dependent) diabetic patients prior to and after pancreas and kidney transplantation in relation to organ function. Diabetologia 1991;34 (Suppl 1):S150–7. 39 Barrou BBA, Bitker MO, et al. Pregnancy after pancreas transplantation: report of four cases and review of the literature. Transplant Proc 1995;27:3042–4.

227

This page intentionally left blank

Chapter 18

Pancreas transplantation: long-term aspects and effect on quality of life Robert J. Stratta

Introduction Diabetes mellitus afflicts approximately 6 per cent of the population and is the third most common disease and the fourth leading cause of death by disease in the United States. Of the estimated 16 million diabetic patients in the United States, about 10 million are diagnosed, 4 million take insulin, and 1 million have insulin-dependent diabetes mellitus (IDDM) [1]. Nearly 30 000 new cases of IDDM are diagnosed each year, and the incidence is increasing. The syndrome of IDDM includes not only abnormal glucose metabolism but also specific long-term complications such as retinopathy, nephropathy, and neuropathy. Although some patients escape the problems associated with IDDM, many patients develop symptomatic complications within an average of 15 to 20 years following diagnosis and may manifest several complications concurrently. Diabetes mellitus is currently the leading cause of blindness in adults, the principle disease cause of amputations and impotence, and ranks among the leading chronic diseases of childhood. In addition, diabetes mellitus is associated with accelerated atherosclerosis, abnormal lipid metabolism, cardiovascular disease, and directly accounts for more than 170 000 deaths per year in the United States. Life expectancy for patients with IDDM is at least 10 years less than for those without diabetes. IDDM has been found to be an independent risk factor for both coronary artery disease and cardiac death with an estimated 80 per cent of diabetic patients dying of atherosclerotic complications. The relative mortality from cardiovascular disease is increased 40-fold in IDDM patients with diabetic nephropathy as compared to the non-diabetic population [2,3]. Approximately 35 per cent of patients with IDDM develop clinical nephropathy, making diabetes the leading cause of endstage renal disease (ESRD) in the United States [2]. In IDDM patients with persistent proteinuria, greater than 75 per cent will develop ESRD (or die) after an average of 6 years. The 2000 US Renal Data System Annual Report noted that of 323 821 patients receiving either dialytic therapy or a kidney transplant up until 1998, 107 613 had diabetes, which is a prevalence rate of 33.2 per cent [3]. Furthermore, of the 85 520 new cases of ESRD in 1998, 36 904 (43.2 per cent) were listed as diabetic. In 1998, diabetes mellitus was also the single leading cause of ESRD for both cadaver donor and living donor kidney recipients. The incidence of endstage diabetic nephropathy is increasing at nearly twice the average rate compared to all other causes of ESRD. It is estimated that as many as one-half of all newly treated ESRD patients will have diabetes in the new millennium.

230

LONG-TERM ASPECTS AND EFFECT ON QUALITY OF LIFE

The discovery of insulin in 1922 changed IDDM from an acute, rapidly fatal disease into a chronic, inexorable illness. Delivering insulin in a physiological manner has been an ongoing goal and challenge since insulin was first purified for administration. Although exogenous insulin therapy is effective at preventing acute metabolic decompensation and is life-saving, most patients with IDDM have one or more endorgan complications during their lifetime. Over the past decade, it has become increasingly evident that the microvascular complications of diabetes mellitus result from hyperglycaemia. Long-term hyperglycaemia may result in excessive glycosylation of circulating and membrane-bound proteins, leading to basement membrane thickening and microangiopathy. Tight glucose control is even more important than previously recognized, as demonstrated by the results of the Diabetes Control and Complications Trial (DCCT) [4]. The DCCT demonstrated that intensive control of glucose reduced the adjusted mean risk of retinopathy by 76 per cent, slowed the progression of retinopathy by 54 per cent, reduced the occurrence of albuminuria by 54 per cent, and decreased the appearance of clinical neuropathy at 5 years by 60 per cent. In addition, intensive insulin therapy reduced the development of hypercholesterolaemia by 34 per cent and the risk of macrovascular disease by 41 per cent. The results of the DCCT clearly indicated that intensive control of glucose can significantly reduce, but not completely protect against, the long-term microvascular complications of diabetes mellitus. Furthermore, the DCCT suggested the absence of a glycaemic threshold for the development of diabetic complications, with total lifetime exposure to glycaemia as the principle determinant of risk [5]. Therefore, the goal of therapy is to achieve normoglycaemia as early and as long as possible. However, intensive therapy in the DCCT had no effect on mortality and was accompanied by a three-fold increase in the risk of severe hypoglycaemia, was more expensive, and was more resource-intensive [4]. Vascularized pancreas transplantation (PTX) was first developed as a means to re-establish endogenous insulin secretion responsive to normal feedback controls. PTX is currently the only known therapy that reliably establishes an insulin-independent euglycaemic state with complete normalization of glycosylated haemoglobin levels [6]. Increasingly, PTX is being offered to patients who either would benefit from a kidney transplant (simultaneous kidney and pancreas transplant or SPK) or have had a previously successful kidney transplant (sequential pancreas after kidney transplant or PAK) [7]. An increasing number of centres are also performing PTX alone (PA) in diabetic patients with hyperlability and severe hypoglycaemic unawareness in the absence of advanced nephropathy [8]. With improvements in organ retrieval technology, surgical techniques, clinical immunosuppression, antimicrobial prophylaxis, donor and recipient selection, and diagnostic methodology, success rates for vascularized PTX have increased dramatically [9]. As early graft survival rates have improved, the long-term consequences of PTX have become much more important, because the benefits must be balanced against the costs and risks of the procedure and the consequences of chronic immunosuppression. The propriety of PTX has been questioned because of the morbidity associated with the procedure and the lack of controlled trials that demonstrate a significant benefit on the secondary complications of diabetes. Despite these concerns, PTX has continued to gain acceptance as an important option for diabetic patients with complications because it is the single most effective method of achieving tight glucose control in the ambulatory setting. With increasing experience in PTX, short-term patient and graft survival rates have steadily improved in recent years. According to United Network for Organ Sharing (UNOS) Registry Data, the 1-year patient, kidney, and pancreas graft survival rates after SPK from 1996 to 1999 were 95, 91, and 85 per cent, respectively [9]. In addition to correcting dysmetabolism and freeing the patient from exogenous insulin therapy, data on the long-term aspects of PTX and its effect on quality of life are

R.J. STRATTA

emerging. In this chapter, we will review data pertinent to long-term outcomes and quality of life after PTX.

Long-term outcomes: single centre reports (Table 18.1) Improving short-term outcomes following PTX have been reported, but it was not until recently that long-term data have become available. One of the earliest studies on long-term outcomes following SPK with systemic-bladder (S-B) drainage was reported Sollinger et al [10]. The 5-year actuarial patient, kidney, and pancreas graft survival rates in 200 consecutive SPK recipients were 90, 80, and 79 per cent, respectively. A total of 23 patients had follow-up of greater than 5 years. These authors noted that patient and graft survival rates were ‘very stable’ after the second post-transplant year. In 1995, Sutherland and Gruessner analysed long-term pancreas graft function (> 5 years) in 596 cases from the International PTX Registry (IPTR) database [11]. For patients who had a functioning pancreas graft at 5 years, the subsequent 10-year actuarial patient survival rate was 90 per cent, and the pancreas graft survival rate was 76 per cent. Based on this analysis, the authors concluded that insulin independence over a normal lifespan is ‘almost certainly possible’ with PTX and that patients with stable endocrine function at 5 years have a low susceptibility to chronic rejection. In 1996, Bruce et al. reported on 50 SPK recipients who had good graft function at 1 year posttransplant and a minimum follow-up of 3 years [12]. Five-year actuarial patient, kidney, and pancreas graft survival rates were 94, 85, and 86 per cent, respectively, with a mean follow-up of 4.3 years. Estimated kidney and pancreas half-lives were 15 ± 2 and 23 ± 7 years, respectively. Rejection and death with functioning grafts were the major causes of graft loss. Hospital admissions, acute rejection, graft pancreatitis, dehydration, and severe infections all decreased dramatically after the first year post-transplant. Beyond 2 years post-transplant, hospital admissions became relatively infrequent, as did transplant-related complications. Psychosocial adjustment and quality of life assessment were both remarkably positive. In a similar analysis, Sudan et al. reported on 57 SPK recipients with a minimum follow-up of 4.5 years and a maximum of 7 years [13]. Ten-year actuarial patient, kidney, and pancreas graft survival rates were 93, 82, and 79 per cent, respectively, with a mean follow-up of 5.7 years. Chronic rejection was a major cause of graft loss. The number of hospital admissions decreased significantly with increasing time after SPK from a mean of 1.2 admissions per patient during the second year after SPK to a mean of 0.2 admissions by year 6. In 1997, Martin et al. evaluated the post-transplant outcome of 89 patients with pancreas graft function for more than 3 years (range 3 to 13 years) and concluded that long-term pancreas graft function is now comparable to other transplanted organs [14]. Long-term stable endocrine function was better with total (versus segmental) pancreas grafts and improvements in diabetic neuropathy (sensory and motor nerve function, bladder function) and stabilization of retinopathy were found after 5 years of normoglycaemia. Chronic rejection was the most important cause of late graft loss, but the incidence of late graft loss was low after 3 years of function. In 1997, Bloom et al. evaluated long-term pancreas allograft outcomes in 71 SPK recipients including 37 with bladder and 34 with enteric exocrine drainage [15]. Five patients in each group experienced early pancreas graft loss and were excluded from further analysis. In the remaining 61 patients, the mean follow-up for bladder and enteric drainage was 76 and 46 months, respectively. The incidence of volume depletion, acidosis, pancreatitis, and urinary tract infection was lower in patients with enteric drainage. In addition, 19 per cent of patients with bladder drainage subsequently

231

232

Table 18.1 Long-term outcomes: single centre reports Mean followup (years)

Kidney graft survival

Pancreas graft survival

5 yr 90%

5 yr 80%

5 yr 79%

6.3

5 yr 71%

5 yr 56%

5 yr 71%

34 S-E 388 S-B, 112 S-E

3.8 – –

5 yr 87% 10 yr 76%

5 yr 87% 10 yr 67%

5 yr 87% 10 yr 67%

S-B

–

5 yr 80%

5 yr 68%

5 yr 68%

SPK recipients with good dual allograft function at 1 year Bruce et al., [12] 1987–92 50/62 S-B

4.3

5 yr 94%

5 yr 85%

5 yr 86%

Sudan et al. [13]

1989–91

57/61

S-B

5.7

10 yr 93%

10 yr 82%

10 yr 79%

Peddi et al. [19]

1989–95

59/79

51 S-B, 8 S-E 4.2

5 yr 88%

5 yr 79%

5 yr 82%

Lo et al. [20]

1990–96

45/75

19 S-B

8

5 yr 84%

5 yr 79%

5 yr 74%

26 P-E

6

5 yr 92%

5 yr 81%

5 yr 88%

Authors (reference)

Time period

n

Technique

Sollinger et al. [10]

1985–92

200

S-B

–

Bloom et al. [15]

1988-96

61/71

37 S-B

Sollinger et al. [16]

1985–97

500

Henry et al. [18]

1990–96

300

Comments Survival rates stable after year 2 Improved outcomes with enteric drainage Survival superior to all other transplants except HLA-Identical KTA SPK provides a superior alternative

Pancreas half-life 23 ± 7 yrs; few complications or readmissions after year 2 Chronic rejection as major cause of graft loss DWFG and chronic rejection as major causes of graft loss Comparable long-term outcomes with P-E drainage

Table 18.1 Long-term outcomes: single centre reports (continued)

Authors (reference)

Time period

n

Technique

PTX recipients with pancreas graft function at 3 years Martin et al. [14] 1976–96 89/175 S-B, S-E, D-O

PTX recipients with pancreas graft function at 10 years Najarian et al. [17] 1978–87 34/211 1 D-O, 1 O-D, 15 S-E, 17 S-B

Mean followup (years)

Patient survival

Kidney graft survival

Pancreas graft survival

–

–

5 yr 81%

Long-term pancreas function comparable to other transplanted organs

–

–

–

Indefinite pancreas graft function seems possible; low rate of chronic rejection

Comments

S-E, systemic-enteric; D-O, duct obstruction; O-D, open duct; DWFG, xxxx.

233

234

LONG-TERM ASPECTS AND EFFECT ON QUALITY OF LIFE

required conversion to enteric drainage for intractable complications. Moreover, the number of readmissions and in-hospital days were less in patients with enteric drainage. Actuarial patient and pancreas allograft survival rates up to 4 years after transplant were similar between groups and in excess of 80 per cent. Therefore, the findings of this study suggested that long-term outcomes may be improved with enteric drainage, particularly with regard to pancreas-related morbidity. In 1998, Sollinger et al. reported their experience with 500 consecutive SPKs, including 388 with bladder drainage and 112 with enteric drainage [16]. Ten-year actuarial patient, kidney, and pancreas graft survival rates were 76, 67, and 67 per cent, respectively. Conversion from bladder to enteric drainage was required in 24 per cent of cases, but no graft was lost as a result of enteric conversion. There was no difference in 1-year graft survival rates between enteric and bladder drainage. Leading causes of pancreas graft loss were rejection in 45 patients and death with functioning grafts in 27 patients. Since June 1995, a total of 109 SPK recipients were managed with mycophenolate mofetil (MMF) therapy. In this latter group, 1-year patient and graft survival rates were in excess of 90 per cent and a marked reduction in rejection was noted. These authors concluded that 10-year graft survival rates exceed those of all other transplants, with the exception of human leucocyte antigen (HLA)-identical living related kidney grafts. In 1998, Najarian et al. analysed long-term pancreas graft function ( > 10 years) in 34 cases performed at the University of Minnesota [17]. The authors concluded that ‘indefinite pancreas graft function seems possible’, with the longest functioning graft currently at 17 years. In patients with stable endocrine function at 1 year, a low rate of chronic rejection was noted, similar to that of other transplanted organs. Also in 1998, Henry et al. reported long-term outcomes in 300 consecutive SKP recipients [18]. Five-year patient, kidney, and pancreas graft survival rates were 80, 68, and 68 per cents, respectively. Death remote to transplantation, but with functioning grafts, was the most common cause of graft loss. The authors concluded that SPK provides a superior alternative for the IDDM patient with ESRD. In 1999, Peddi et al. analysed retrospectively long-term outcomes in 59 SPK recipients with both grafts functioning at 1 year post-transplant [19]. At a mean follow-up of 50 months (range 24 to 101), there were five deaths (8.5 per cent), 11 renal allograft losses (19 per cent), and nine pancreas graft losses (15 per cent).The 5-year Kaplan – Meier patient, kidney, and pancreas graft survival rates were 88, 79, and 82 per cent respectively. Death with functioning grafts and chronic rejection were the major causes of graft loss. Pre-existing cardiac and vascular disease contributed to ongoing morbidity and mortality in these patients. In 2001, Lo et al. retrospectively reviewed long-term outcomes in SPK recipients with either portalenteric (P-E) or S-B drainage [20]. A total of 45 patients were alive with functioning grafts 1 year after SPK and were followed for a minimum of 3 years (mean 7 years) including 26 with P-E and 19 with SB drainage. In both groups, hospital admissions decreased significantly with increasing time after SPK. Renal and pancreas allograft functions were similar between the two groups. At 1 year posttransplant, stabilization in most diabetic complications was reported. Four quality of life surveys that provided 29 scores were completed 6 to 24 months (mean 18.5 months) after SPK. Improved quality of life was reported in all but one of the scales, with many dimensions showing significant improvements. At 3 years after transplant, no activity limitation was reported in 76 per cent of patients after PE versus 53 per cent after S-B drainage. Five-year actuarial patient, kidney, and pancreas graft survival rates were 92 per cent P-E and 84 per cent S-B, 81 per cent P-E and 79 per cent S-B, and 88 per cent PE and 74 per cent S-B, respectively (P = NS).The authors concluded that SPK with P-E drainage is a safe and effective method to treat advanced diabetic nephropathy and is associated with decreasing morbidity, improving rehabilitation and quality of life, and stable metabolic function over time.

R.J. STRATTA

Long-term prognosis after the first year was excellent and was at least comparable to the results achieved with S-B drainage.

Long-term outcomes: SPK versus KTA (Table 18.2) In 1997, Tibell et al. reported an 8-year patient survival rate of 86 per cent after SPK versus 47 per cent in diabetic patients undergoing kidney transplantation alone (KTA) [21]. The control group consisted of patients originally considered eligible for SPK but either the donor pancreas was not deemed suitable during procurement or the pancreatic graft was lost early after transplant. In a follow-up study from this group, Tyden et al. presented 10-year data on 14 patients with IDDM who underwent SPK versus a control group of 15 IDDM patients receiving KTA [22]. The 10-year patient survival rate was 79 per cent after SPK versus 20 per cent after KTA. The SPK recipients were noted to have normal glucose control, improved nerve conduction and autonomic function, better quality of life, and a significantly lower mortality than the control group of IDDM patients undergoing KTA. In 1999, Smets et al. performed a registry study in The Netherlands of 415 IDDM patients with ESRD between the ages of 18 to 52 years who began renal replacement therapy between 1985 and 1996 [23]. The patients were divided into two geographical areas based on whether the primary intention to treat was with SPK versus KTA. In the Leiden region, 41 (73 per cent) of 56 transplanted patients received SPK, while only 59 (37 per cent) of 158 transplanted patients in the non-Leiden area underwent SPK. The authors compared mortality in the two regions after making adjustments for age, gender, and duration of dialysis pretransplant. In the region in which SPK was the preferred treatment option, transplant recipients had a 60 per cent lower 5-year mortality (hazard ratio for mortality was 0.4). The 10-year actuarial patient survival after transplant was nearly 80 per cent in the Leiden area versus 40 per cent in the region in which KTA was the predominant treatment. The authors concluded that the 50 per cent reduction in mortality in patients transplanted in the Leiden area was attributable to the higher rate of SPK versus KTA in this region. In the largest single centre report from the University of Wisconsin, Becker et al. demonstrated that SPK recipients (n = 335) have an increased observed/expected lifespan compared with cadaveric kidney (n = 147) and living donor kidney (n = 160) IDDM recipients [24]. The annual mortality rate was 1.5 per cent for SPK recipients compared to 6.3 per cent for cadaveric kidney recipients and 3.7 per cent for living donor kidney recipients. In a follow-up study from the University of Wisconsin, Rayhill et al. used a longitudinal database to compare survival rates among IDDM patients undergoing either HLA-identical living donor KTA (n = 43), haplo identical living donor KTA (n = 87), cadaver donor KTA (n = 296), or SPK (n = 379) [25]. Patient and graft survival rates were comparable for living donor KTA and SPK, but significantly lower in cadaveric donor KTA recipients. Cardiovascular disease was the primary cause of death in all groups. Acute rejection, chronic rejection, and death with a functioning graft were the predominant causes of graft loss. Five-year patient (94 per cent) and graft (85 per cent) survival rates were slightly higher in HLA-identical living donor KTA recipients, but no differences were noted between SPK and haploidentical KTA recipients. In another follow-up study, Sollinger et al. from the University of Wisconsin analysed their single centre experience with 335 SPKs compared to IDDM patients undergoing either cadaveric or living donor KTA [26]. According to lifetable analysis, diabetic recipients attained more of their projected life expectancy when transplanted with both organs. SPK increased the observed versus expected lifespan compared to KTA (regardless of donor source), and was associated with reduced annual mortality rates. Reddy et al. studied the UNOS database and analysed 18 549 patients with IDDM and ESRD who received a kidney transplant between 1987 and 1996 [27]. Of these, 9956 underwent cadaveric donor

235

236

Table 18.2 Long-term outcomes: multi-centre registry reports

Authors (reference)

Time period

n

PTX recipients with pancreas grant function at 5 years Sutherland 1966–90 596/2337 et al. [11]

SPK vs KTA Tibell et al. [21] Tyden et al. [22] Smets et al [23] Becker et al. [24] Rayhill et al. [25]

Technique Patient survival

328 S-B, 158 D-O, 95 S-E, 15 Other

10 yr 90%

Kidney graft survival

10 yr 76%

14 SPK 15 KTA

S-E

8 yr 86% 8 yr 47%

1982–86

14 SPK 15 KTA 41/56 SPK 59/158 SPK 335 SPK 160 L-D KTA 147 CAD KTA 379 SPK 43 HLA-identical L-D KTAs 87 HAPLO-identical L-D KTAs 296 CAD KTAs

S-E

10 yr 79% 10 yr 20% 10 yr 80% 10 yr 35% 10 yr 85% 10 yr 66% 10 yr 50% 5 yr 88%

10 yr 80% 10 yr 30% 10 yr 70% 10 yr 68% 10 yr 52% 5 yr 98%

5 yr 94%

5 yr 85%

5 yr 85% 5 yr 72%

5 yr 72% 5 yr 64%

1986–95

1986–96

S-B

Comments

Insulin independence over normal lifespan almost certainly possible

Improvements in neuropathy, quality of life, and survival after SPK vs KTA

1982–86

1985–96

Pancreas graft survival

50% Reduction in mortality associated with SPK SPK increased the observed/ expected lifespan Survival rates after SPK comparable to L-D, superior to CAD KTA; more of projected life expectancy attained with SPK

Table 18.2 Long-term outcomes: multi-centre registry reports (continued)

Authors (reference)

Time period

n

Reddy et al. [27]

1987–96

Ojo et al. [29]

1988–97

4602 SPK 3991 L-D KTA 9956 CAD KTA 4718 SPK 671 L-D KTA 4127 CAD KTA

Tyden et al. [30]

1981–88

69 SPK 44 KTA

Technique

Kidney graft survival

Pancreas graft survival

8 yr 72% 8 yr 72% 8 yr 55% 10 yr 67% 10 yr 65% 10 yr 46%

8 yr 70% 8 yr 62% 8 yr 45% – – –

– – – – – –

10 yr 60%/33%a 10 yr 37%

– –

– –

Patient survival

Comments Excess initial mortality of SPK offset by reduced late mortality Projected life expectancy of 23.4 yrs after SPK; no survival benefit in patients > 50 Functioning pancreas graft contributes to superior life expectancy

S-E, systemic-enteric; D-O, duct obstruction; CAD, cadaveric donor; L-D, living donor. aPancreas

graft failure.

237

238

LONG-TERM ASPECTS AND EFFECT ON QUALITY OF LIFE

KTA, 3991 living donor KTA, and 4602 SPK. Mean follow-up was 4.8 years in survivors. The 8-year actuarial survival rates were 72 per cent for both SPK and living donor KTA recipients as compared to 55 per cent for cadaver donor KTA recipients. Using a proportional hazards model, the survival advantage for SPK recipients over cadaveric donor KTA recipients diminished but persisted after adjusting for donor and recipient variables and kidney graft function as time-varying covariants. SPK recipients had a high mortality risk relative to living donor KTA recipients for the first 18 months post-transplant [relative risk) (RR) = 2.2], but had a lower mortality risk thereafter (RR = 0.86). In SPK recipients, maintenance of a functioning pancreas graft was associated with a significant survival benefit. In a similar analysis, Hunsicker et al. analysed outcomes from the 1997 UNOS Center-Specific report in 2304 SPK recipients with kidney graft function at 1 year [28]. Of these, the pancreas graft was still functioning at 1 year in 2010 patients and had failed in 294. Presence of a functioning pancreas graft was associated with a 53 per cent reduction in mortality after 1 year, a 47 per cent reduction in total renal graft failure, and a 45 per cent reduction in renal graft failure censored for death with a functioning graft (p < 0.001 for all comparisons). Patient survival was 15 to 20 per cent better at 10 years in the patients with functioning pancreas grafts at 1 year. Each of these studies concluded that SPK is a life-saving procedure. Ojo et al. used data from the US Scientific Renal Transplant Registry and from the US Renal Data System (USRDS) database to analyse long-term outcomes in IDDM patients with ESRD who were placed on the active transplant waiting list between 1 October 1988 and 30 June 1997. A total of 13 467 evaluable wait-listed patients were followed until 30 June 1998 [29]. Time-dependent mortality risks and life expectancy were calculated for the study group which included 4718 SPK recipients, 4127 cadaveric donor KTA recipients, 671 living donor KTA recipients, and 3951 wait-listed but never transplanted patients on maintenance dialysis. Adjusted 10-year patient survival rates were 67 per cent for SPK, 65 per cent for living donor KTA, and 46 per cent for cadaveric donor KTA recipients. The excess initial mortality normally associated with kidney transplantation and the risk of early infectious death were two-fold higher in SPK recipients. However, the adjusted 5-year mortality risks (RR) using maintenance dialysis therapy as the reference were 0.40 for SPK, 0.45 for living donor KTA, and 0.75 for cadaveric donor KTA. The projected life expectancy was 23.4 years for SPK, 20.9 years for living donor KTA, and 12.6 years for cadaveric donor KTA recipients. No survival benefit was found for SPK recipients above 50 years of age. The authors concluded that SPK before the age of 50 years is associated with a long-term improvement in survival compared to either KTA or dialysis. In 2000 Tyden et al. analysed 515 patients undergoing transplantation between 1981 and 1988, including 69 SPK recipients and 44 IDDM patients receiving KTA [30]. The actual 10-year patient survival rate in non-diabetic KTA recipients was 72 per cent compared to 60 per cent after SPK. In SPK recipients in which the pancreas graft failed within 2 years, the actual 10-year patient survival rate was 33 per cent, similar to the 37 per cent survival rate seen in IDDM patients undergoing KTA. The authors concluded that a functioning pancreas graft contributes to superior life expectancy after transplantation.

Long-term metabolic aspects The ultimate goal of PTX is to restore normal glucose homeostasis to prevent secondary complications of diabetes. Several studies have previously demonstrated that short-term metabolic control is more nearly normalized by PTX than any other form of diabetic therapy. A number of recent reports have analysed long-term metabolic control after PTX. Robertson et al. reported that successful PTX

R.J. STRATTA

recipients are capable of maintaining insulin secretion that is sufficient to allow maintenance of normoglycaemia and normal glycosylated haemoglobin levels for periods extending up to 11 years [6]. These authors concluded that PTX, when successful, is the most consistently reliable means of restoring normal insulin secretion and glucose homeostasis in patients with IDDM. In 1998, Tajra et al. studied the long-term metabolic status of 80 patients with functioning PTXs with follow-up ranging from 3 to 13 years [31]. Recipients of whole organ pancreatic grafts had consistently better long-term metabolic control, but segmental pancreatic grafts were able to maintain normal serum glucose and glycosylated haemoglobin levels up to 13 years after PTX. Also in 1998, Tyden et al. studied 33 SPK recipients with graft function for at least 5 years, including 21 segmental and 12 whole organ PTXs [32]. Excellent metabolic control was documented with both techniques, although recipients of whole organ PTX had slightly higher stimulated C-peptide levels. The authors concluded that long-term glycaemic control remains normal in both recipients of segmental and whole organ PTXs. In experimental models of canine diabetes, retinopathy, neuropathy, and nephropathy have been shown to develop within 5 years. Hawthorne et al. performed segmental pancreas autotransplantation with residual pancreatectomy in 35 outbread mongrel dogs with follow-up to 5 years [33]. The authors showed that segmental pancreatic autografts are capable of providing satisfactory metabolic control for up to 5 years. None of the dogs with functioning PTX developed any evidence for retinopathy, neuropathy, or nephropathy. The authors concluded that the metabolic control achieved by a functioning PTX can prevent the development of long-term microvascular complications of diabetes. Although successful PTX results in excellent metabolic control with complete insulin independence, there have been sporadic reports of patients returning to insulin therapy either for the development of type 2 diabetes caused by insulin resistance or autoimmune disease recurrence. In 1996, Jones et al. reported an isolated case of a patient undergoing successful SPK who subsequently developed fastening hyperglycaemia despite hyperinsulinaemia [34]. This patient experienced rapid weight gain after transplantation, which may have contributed to the development of insulin resistance. However, preliminary experience with SPK has been favourable in patients with type 2 diabetes, high C-peptide levels, and presumed peripheral insulin resistance [35]. Also in 1996, Tyden et al. reported two cases of autoimmune disease recurrence after SPK [36]. Both patients developed a gradual reduction in fasting C-peptide levels, a more marked reduction in stimulated C-peptide levels, and histological evidence for selective B-cell destruction and insulitis in the absence of rejection. In one patient, histological evidence for insulitis occurred in association with the development of anti-islet cell and anti-glutamic acid decarboxylase (GAD) antibodies. However, in another study monitoring markers for humoral autoimmunity in a longitudinal fashion, no correlation was found between the detection of anti-islet cell or anti-GAD antibodies and pancreas allograft function [37]. Since chronic immunosuppression is a requisite for PTX, it is currently believed that recurrent autoimmune diabetes is not inevitable and is probably rare. In addition, because most techniques of PTX result in systemic or peripheral hyperinsulinaemia, the development of insulin resistance over the long term is a theoretical concern that has yet to be borne out by clinical studies. Based on these analyses, it is concluded that SPK is a safe and effective method to treat advanced diabetic nephropathy and is associated with decreasing morbidity, improving rehabilitation, and stable metabolic function over time. Long-term prognosis after the first year is excellent and at least provides the potential for improved survival with stabilization of diabetic complications.

Quality of life The dramatic increase in post-transplant patient and graft survival rates over the last decade has resulted in great interest in quality of life. This is of particular importance for those patients with

239

240

LONG-TERM ASPECTS AND EFFECT ON QUALITY OF LIFE

ESRD and IDDM who not only have symptoms associated with uraemia, but complications related to long-standing diabetes. Quality of life is a multi dimensional construct reflecting an individual’s perception of health, well being, and happiness. Conceptual aspects include physical function, social function, psychological function, the burden of symptoms and treatments, and sense of well being. Various questionnaires and survey instruments are available that assess indices of well being, affect, satisfaction, activities and daily living, and health-care burden. Quality of life surveys do not specifically address any particular physiological consequence, but they do take into account various factors that are most important to the transplant recipient. These factors include opportunities for social interaction, ability to return to work or maintain employment, overall energy level, and psychological adjustment. There is no question that impaired or poor quality of life is a ‘secondary complication’ of diabetes, particularly in IDDM patients with ESRD. Most quality of life studies are observational, cross-sectional, retrospective, and non-randomized in design. Another methodological difficulty remains in selecting an appropriate ‘control’ group to compare to PTX recipients such as ‘matched’ patients with diabetes, patients with diabetes on dialysis, patients with diabetes after KTA (cadaver versus living donor), or patients with diabetes after SPK with a failed pancreas (or kidney) graft. In spite of these limitations, there are at least 30 studies in the recent literature reporting on quality of life after PTX [38–46]. Twelve of these studies are prospective and longitudinal, while the remainder are cross-sectional in design, and none are randomized. With regard to methodology, five of these studies involve SPK, eight SPK versus KTA, five SPK versus failed SPK, six SPK versus failed SPK versus KTA, three SPK versus KTA versus IDDM, two SPK versus failed SPK versus IDDM, and one SPK versus IDDM patients. All but one of these studies show some improvement in quality of life after SPK, although the differences are not always significant. Numerous studies have demonstrated that successful SPK results in improvements in physical function, activities and daily living, energy level, mobility, vocational rehabilitation, social well being, communication, role function, health perception, self-image, psychological function, future expectations, sense of well being, overall satisfaction, diet flexibility, diabetes-related concerns, time to manage health, health impact on family, and autonomy [38–46]. The major benefits of PTX are an enhanced quality of life characterized by the following: (a) rehabilitation to ‘normal’ living with physical, social, and psychological well being with near normal activities and daily living and a self-perception of normality; (b) global improvement in quality of life with the perception of being healthy and having control over one’s destiny; and (c) fewer restrictions and enhanced capacities leading to an improved sense of well being and independence [45]. Freedom from daily insulin injections and blood glucose monitoring are important advantages for patients with a successful PTX. Although the long-term commitment to immunosuppression is a major trade-off, most patients with diabetes find the transition to transplantation easier than continued insulin therapy because of an improved sense of well being with fewer dietary and activity restrictions. Immunosuppression is perceived to be easier to manage and less demanding than diabetes [41]. Despite the morbidity of SPK and its increased perceived burden of treatment, when questioned most patients would opt for pancreas retransplantaion. With the increasing short- and long-term success of PTX, the emphasis has shifted from survival outcomes to health-related quality of life. A recent quantitative analysis of the literature revealed 218 studies that examined quality of life following kidney, pancreas, kidney–pancreas, heart, lung, heart–lung, liver, and bone marrow transplantation for nearly 15 000 recipients [40]. The majority of studies were cross-sectional with limited follow-up, as 56 per cent evaluated quality of life within 1 year of transplant, 36 per cent evaluated 1 to 3 years post-transplant, and 8 per cent evaluated 3 or more years following transplantation. Despite these limitations in study designs, quality of life was found to be universally and substantially improved after transplant for all quality of life domains examined

R.J. STRATTA

(physical, mental, social, and global). Physical function was more likely to demonstrate improvements than were the other domains. In the few studies that did employ longitudinal designs, quality of life was either stable or improved during the first 7 years following transplantation. These studies indicate that transplantation does improve quality of life, at least in the short-term and perhaps in the long-term as well. Although transplantation has not returned individuals to a totally normal life, the overall majority of successful recipients perceived a marked improvement in quality of life. It is known from the exercise physiology literature that even mild regular exercise and deep breathing may improve autonomic function. Preliminary data obtained from transplant recipients also indicate that moderate levels of exercise are beneficial for transplant recipients in terms of improving autonomic function. While one can logically infer the physiological consequences of treatment, such as autonomic neuropathy, are linked with patient perceptions of quality of life, this relationship has been investigated to only limited extent. In a 1994 study, a path analysis conducted with 56 patients with diabetes found that 15 per cent of the variation and functional variability could be accounted for by measures of autonomic neuropathy [38]. If relationships do exist between the physiological consequences of treatment and quality of life, then interventions to facilitate patients’ attainment of optimal quality of life may result in further improvements in autonomic function. Painter et al. studied cardiorespiratory fitness in 25 SPK and 16 KTA recipients with a mean follow-up of approximately 2 years [47]. SPK recipients were younger and in general achieved higher levels of cardiorespiratory fitness than KTA recipients. In both groups, patients who self-reported themselves as physically active scored higher than inactive patients, although this difference did not achieve significance. Because transplant recipients are living longer, interest in long-term quality of life outcomes is emerging. Given the projection that many recipients will live well into the second decade following transplantation, it is important to note that some forms of neuropathy continue to demonstrate improvement in these later years, that neuropathy is associated with mortality as well as quality of life in the transplant population, and that some interventions may be available that improve neuropathic function long term [48–50]. Therefore, it seems apparent that well-designed, longitudinal studies are needed not only to assess quantity but also quality of life as well as physiological function. The major factors perceived to affect quality of life are immunosuppression and its side-effects after SPK, rejection and physical symptoms after KTA, and diabetes and its complications after a failed pancreas allograft [41]. Furthermore, studies have shown that failure of a pancreas allograft after SPK results in increased fatigue, less energy, an increased need for social support, a net reduction in quality of life, and higher mortality when compared with both SPK and KTA [41–48]. Thus, any studies designed to evaluate the cost-effectiveness of PTX need to balance these factors, particularly a recipient’s ability to return to employment as well as cost savings from prevention or reversal of diabetic complications in addition to the costs associated with the operative procedure, immunosuppression, and graft failure. Because of the impact that PTX has on quality of life, we are conducting comprehensive, prospective studies examining quality of life changes in this patient population. Since 1990 the University of Tennessee, Memphis, has included three measurements of quality of life in order to capture as many dimensions as possible [45,51]. Functional disability is measured by the Sickness Impact Profile (SIP), while a more positive view of health is measured by the Quality of Life Index (QLI). Psychoemotional dimensions are measured with the Adult Self-Image Scale (ASIS). Each instrument has been used in the transplant population and has documented reliability and validity [45,51]. All patients are asked to participate by completing a battery of quality of life instruments that provide a multidimensional assessment of this construct. Combined, the instruments yield over 20 scores reflecting specific dimensions of quality of life such as mobility, work, family, anxiety, and independ-

241

242

LONG-TERM ASPECTS AND EFFECT ON QUALITY OF LIFE

ence; composite scores that group specific dimensions into five categories (health-related, physical, psychological, disability, and self-esteem); and one global measure of quality of life that does not separate specific dimensions. These instruments are repeated at 6 and 12 months after PTX and yearly thereafter. This provides us with prospective, longitudinal data regarding the influence of PTX on quality of life. With the improving short-term success of PTX, long-term prognosis after the first year is excellent and at least provides the potential for stabilization of diabetic complications. We retrospectively reviewed remote SPK recipients with P-E drainage; 26 patients (65 per cent) were alive with functioning grafts one year after SPK and were followed for a minimum of 3 years (mean 5 years) [52]. Hospital admissions decreased significantly from a mean of 2.4 admissions per patients in the first year to 0.6 by year 4. At 1 year post-transplant, improvements in most diabetic complications were noted. No activity limitations were reported in 80 per cent of patients at 1 year after SPK compared to 23 per cent pretransplant. Four quality of life surveys that provided 29 scores were completed 6 to 24 months (mean 18 months) after SPK. Improved quality of life was reported in all but one of the scales. Actual patient, kidney, and pancreas graft survival rates were 92, 81, and 89 per cent, respectively. These results demonstrated that SPK with P-E drainage is associated with decreasing morbidity and improving quality of life over time with intermediateterm follow-up. Improvement in quality of life is one of the major goals of PTX. Current data document the presence of poor overall quality of life in patients with diabetes as compared to their non-diabetic counterparts. It is interesting to note that the baseline quality of life reported for our more recent diabetic patients has generally improved from previous reports. We can only surmise that advances in medical care (erythropoietin) and/or earlier transplant referral may have influenced this outcome. When comparing SPK to non-diabetic KTA recipients, patients with diabetes pretransplant had a poor quality of life in two of five measures, primarily reflecting greater physical dysfunction and a less positive view of their overall health [45,51,53]. Following transplantation, quality of life improved in four of the five categories for both groups. However, at 24 months, a lingering disparity was still noted between nondiabetic KTA and SPK recipients with respect to physical function and overall health perspective.

Summary Vascularized PTX has assumed an increasingly important role in the treatment of IDDM. SPK is gaining acceptance as a viable alternative to KTA in transplant recipients with diabetes because of its ability to provide superior glycaemia control, improve quality of life, and enhance life expectancy. The greater morbidity (and early mortality) of SPK can be justified by the mounting evidence that a functioning pancreas graft may prevent, stabilize, or induce regression of diabetic complications coincident with a beneficial effect on quality as well as quantity of life. Although PTX results in euglycaemia and complete insulin independence, these results occur at the expense of hyperinsulinaemia and chronic immunosuppression. The net result of these changes on diabetic complications in the long term is currently being studied. In the short term, improvement in quality of life and possible prevention of further morbidity and mortality associated with diabetes makes PTX an important therapeutic option for selected patients with IDDM. In the future, PTX will remain an important option in the treatment of IDDM until other strategies are developed that can provide equal glycaemic control with less or no immunosuppression or less overall morbidity.

R.J. STRATTA

Acknowledgements We gratefully acknowledge the expertise of Joyce Lariviere in the preparation of this manuscript.

References 1 American Diabetes Association. Economic consequences of diabetes mellitus in the US in 1997. Diabetes Care 1998;28(2):296–309. 2 American Diabetes Association, Diabetic nephropathy. Diabetes Care 1998;21(Suppl.1):S50–3. 3 US Renal Data System. USRDS 2000 Annual Data Report: incidence and prevalence of ESRD. Am J Kidney Dis 2000;36(Suppl. 2):S37–54. 4 The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–86. 5 The Diabetes Control and Complications Trial Research Group. The absence of a glycemic threshold for the development of long-term complications: the perspective of the Diabetes Control and Complications Trial. Diabetes 1996;45:1289–98. 6 Robertson RP, Kendall D, Teuscher A, Sutherland DER. Long-term metabolic control with pancreatic transplantation. Transplant Proc 1994;26(2):386–7. 7 Sutherland DER, Gruessner AC, Gruessner RWG. Pancreas transplantation: a review. Transplant Proc 1998;30:1940–3. 8 Sutherland DER, Gruessner RWG, Najarian JS, Gruessner AC. Solitary pancreas transplants: a new era. Transplant Proc 1998;30:280–1. 9 Gruessner AC, Sutherland DER. Analysis of pancreas transplant outcomes for United States cases reported to the United Network for Organ Sharing (UNOS) and the non-US cases reported to the International Pancreas Transplant Registry (IPTR). In: Cecka JM, Terasaki PI, ed. Clinical transplants 1999. Los Angeles: UCLA Immunogenetics Center, 2000, 51–69. 10 Sollinger HW, Ploeg RJ, Eckhoff DE, Stegall MD, Isaacs R, Pirsch JD, et al. Two hundred consecutive simultaneous pancreas–kidney transplants with bladder drainage. Surgery 1993;114:736–43. 11 Sutherland DER, Gruessner A. Long-term function (> 5 years) of pancreas grafts from the International Pancreas Transplant Registry Database. Transplant Proc 1995;27:2977–80. 12 Bruce DS, Newell KA, Josephson MA, Woodle ES, Piper JB, Millis JM. et al. Long-term outcome of kidney – pancreas recipients with good graft function at 1 year. Transplantation 1996;62:451–6. 13 Sudan D, Sudan R, Stratta RS. Long-term outcome of simultaneous kidney–pancreas transplantation: analysis of 61 patients with more than 5 years follow-up. Transplantation 2000;69:550–5. 14 Martin X, Tajra LCF, Benchaib M, Dawahra M, Lefrancois N, Dubernard JM. Long-term outcome of pancreas transplantation. Transplant Proc, 1997;29:2423–4. 15 Bloom RD, Olivares M,Rehman L, Raja RM, Yang S, Badosa F. Long-term pancreas allograft outcome in simultaneous pancreas – kidney transplantation: a comparison of enteric and bladder drainage. Transplantation 1997;64:1689–95. 16 Sollinger HW, Odorico JS, Knechtle SJ, D’Alessandro AM, Kalayoglu M, Pirsch JD. Experience with 500 simultaneous pancreas–kidney transplants. Ann Surg 1998;228:284–96. 17 Najarian JS, Gruessner AC, Drangsteveit MB, Gruessner RWG, Goetz FC, Sutherland DER. Insulin independence for more than 10 years after pancreas transplantation. Transplant Proc 1998;30:1936–7. 18 Henry ML, Elkhammas EA, Bumgardner GL, Pelletier RP, Ferguson RM. Outcome of 300 consecutive pancreas–kidney transplants. Transplant Proc 1998;30:291.

243

244

LONG-TERM ASPECTS AND EFFECT ON QUALITY OF LIFE

19 Peddi VR, Munda R, Demmy AM, First MR. Long-term outcome in simultaneous kidney and pancreas transplant recipients with functioning allografts at one-year posttransplantation. Transplant Proc 1999;31:608–9. 20 Lo A, Stratta RJ, Hathaway DK, Egidi MF, Shokouh-Amiri MH, Grewal HP, et al. Long-term outcomes in simultaneous kidney–pancreas transplant recipients with portal-enteric versus systemic-bladder drainage. Am J Kidney Dis 2001;38–132–143. 21 Tibell A, Solders G, Larsson M, Brattstrom C, Tyden G. Superior survival after simultaneous pancreas and kidney transplantation compared with transplantation of a kidney alone in diabetic recipients followed for 8 years. Transplant Proc 1997;29:668. 22 Tyden G, Bolinder J, Solders G, Brattstrom C, Tibell A, Groth CG. Improved survival in patients with insulin-dependent diabetes mellitus and end-stage diabetic nephropathy 10 years after combined pancreas and kidney transplantation. Transplantation 1999;67:645–8. 23 Smets YFC, Wetendorp RGJ, Van der Pijl JW, et al. Effect of simultaneous pancreas–kidney transplantation on mortality of patients with type-1 diabetes mellitus and end-stage renal failure. Lancet 1999;353:1915–19. 24 Becker BN, Brazy PC, Becker YT, et al. Simultaneous pancreas–kidney transplantation reduces excess mortality in type 1 diabetic patients with end-stage renal disease. Kidney Int. 2000;57:2129–35. 25 Rayhill SC, D’Alessandro AM, Odorico JS, Knechtle SJ, Pirsch JD, Heisey DM, et al. Simultaneous pancreas–kidney transplantation and living related donor renal transplantation in patients with diabetes: is there a difference in survival? Ann Surg 2000;231:417–23. 26 Sollinger HW, Pirsch JD, Odorico JS, Becker BN. Is the pancreas a life-saving organ transplant? Proc 17th Int Congr Transplant Soc 2000;116 (Abstract):A0332. 27 Reddy KS, Stablein D,Taranto S, et al. Long-term survival following simultaneous kidney–pancreas transplantation versus kidney transplantation alone in patients with type 1 diabetes mellitus and renal failure. Transplant Proc 2001 (in press). 28 Hunsicker LG, Bozorgzadeh A, Rosendale JD, et al. Pancreas graft function reduces mortality and renal graft loss in simultaneous pancreas–kidney (SPK) transplants beyond one year. Proc 17th Int Congr Transplant Soc 2000;55 (Abstract):A0219. 29 Ojo AO, Meier-Kriesche HU, Hanson JA, et al. The impact of simultaneous pancreas–kidney transplantation on long-term patient survival. Transplantation 2001;71:82–90. 30 Tyden G, Tollemar J, Bolinder J. Combined pancreas and kidney transplantation improves survival in patients with end-stage diabetic nephropathy. Clin Transplant 2000;14:505–8. 31 Tajra LCF, Martin X, Benchaid M, Dawhara M, Lefrancois N, Dubernard JM. Long-term metabolic control in pancreas transplant patients according to 3 techniques. Transplant Proc 1998:30:268–9. 32 Tyden G, Bolinder J, Brattstrom C, Tibell A, Groth CG, Long-term metabolic control in recipients of segmental or whole-organ pancreatic grafts with enteric exocrine diversion and function beyond 5 years. Transplant Proc 1998;30:634. 33 Hawthorne WJ, Wilson TG, Williamson P, Stewart GJ, Allen RDM, Little J, et al. Long-term duct occluded segmental pancreatic allografts: absence of microvascular diabetic complications. Transplantation 1997;64:953–9. 34 Jones JW, Mizrahi SS, Bentley FR. Type II diabetes after combined kidney and pancreas transplantation for type I diabetes mellitus and end-stage renal disease. Clin Transplant 1996;10:574–5. 35 Sasaki TM, Gray RS, Ratner RE, Currier C, Aqino A, Barhyte DY, et al. Successful long-term kidney–pancreas transplants in diabetic patients with high C-peptide levels. Transplantation 1998;65:1510–12. 36 Tyden G, Reinholt FP, Sundkvist G, Bolinder J. Recurrence of autoimmune diabetes mellitus in recipients of cadaveric pancreatic grafts. N Engl J Med 1996;335:860–3.

R.J. STRATTA

37 Esmatjes E, Rodriguez-Villar C, Ricart MJ, Casamitjana R, Martorell J,Sabater L, et al. Recurrence of immunological markers for type I (insulin-dependent) diabetes mellitus in immunosuppressed patients after pancreas transplantation. Transplantation 1998;66:128–31. 38 Hathaway DK, Abell T, Cardoso S, Hartwig MS, El-Gebely S, Gaber AO. Improvement in autonomic and gastric function following pancreas–kidney versus kidney–alone transplantation and the correlation with quality of life. Transplantation 1994;57:816–22. 39 Gaber AO, Hathaway DK, Abell T, Cardoso S, Hartwig MS, El-Gebely S. Improved autonomic and gastric function in pancreas–kidney versus kidney–alone transplantation contributes to quality of life. Transplant Proc 1994;26(2):515–16. 40 Dew MA, Switzer GE, Goycoolea JM, Allen AS, Dimartini A, Kormos RL, et al. Does transplantation produce quality of life benefits? A) quantitative analysis of the literature. Transplantation 1997;64:1261–73. 41 Gross CR, Limwattananon C, Matthees BJ. Quality of life after pancreas transplantation: a review. Clin Transplant 1998;12:351–61. 42 Adang EMM, Engel GL, Van Hooff JP, Kootstra G. Comparison before and after transplantation of pancreas–kidney and pancreas–kidney with loss of pancreas — A prospective controlled quality of life study. Transplantation 1996;62:754–8. 43 Matas AJ, McHugh L, Payne WD, Wrenshall LE, Dunn DL, Gruessner RWG, et al. Long-term quality of life after kidney and simultaneous pancreas–kidney transplantation. 44 Secchi A, Martinenghi S, Castoldi R, Giudici D, Di Carlo V, Pozza G, Effects of pancreas transplantation on quality of life in type I diabetic patients undergoing kidney transplantation. Transplant Proc 1998;30:339–42. 45 Hathaway DK, Hartwig MS, Crom DB, Gaber AO. Identification of quality-of-life outcomes distinguishing diabetic kidney-alone and pancreas-kidney recipients. Transplant Proc 1995;27(6):3065-6. 46 Milde FK, Hart LK, Zehr PS. Pancreatic transplantation: impact on the quality of life of diabetic renal transplant recipients. Diabetes Care 1995;18:93–5. 47 Painter P, Tomlanovich S, Hector L, Ray K, Stock P, Melzer J. Cardiorespiratory fitness in pancreas–kidney transplant recipients. Transplant Proc 1998;30:651–2. 48 Navarro X, Kennedy WR, Loewenson RB, Sutherland DER. Influence of pancreas transplantation on cardiorespiratory reflexes, nerve conduction and mortality in diabetes mellitus. Diabetes 1990;39:802–6. 49 Navarro X, Sutherland DER, Kennedy WR. Long-term effects of pancreatic transplantation on diabetic neuropathy. Ann Neurol 1997;42:727–36. 50 Navarro X, Kennedy WR, Aeppli D, Sutherland DER. Neuropathy and mortality in diabetes: influence of pancreas transplantation. Muscle and Nerve 1996;19:1009–16. 51 Hathaway DK, Hartwig MS, Milstead J, et al. A prospective study of changes in quality of life reported by diabetic recipients of kidney-only and pancreas–kidney allografts. J Transplant Coord 1994;4:12–17. 52 Stratta RJ, Lo A, Hathaway DK, et al. Long-term outcome in simultaneous kidney–pancreas transplant recipients with portal-enteric drainage (Abstract) Proceedings of the 7th World Congress of the International Pancreas and Islet Transplant Association, 1999;71:P 52. 53 Hathaway DK, Wicks MN, Cashion AK, et al. Heart rate variability and quality of life following kidney and kidney–pancreas transplantation. Transplant Proc 1999;31:643–4.

245

This page intentionally left blank

Chapter 19

Immunosuppression and diabetogenicity Rahul M. Jindal, V.K. Revanur, and Alan G. Jardine

Introduction Post-transplant diabetes mellitus (PTDM) has emerged as a major side-effect of immunosuppresants [1–3] and, as recipients of organ transplants survive longer, the secondary complications of DM have assumed greater importance. Thus, the findings of the Diabetes Control and Complications Trial (DCCT) in patients with IDDM, that tight control of glucose levels significantly reduced the risk of diabetes related complications, may have implications for transplant recipients. In the DCCT, a 2 per cent difference in average haemoglobin A (HbA1c) between standard and intensive treatment groups was associated with a 60 per cent reduction in risk for diabetic retinopathy, nephropathy, and neuropathy. Furthermore, there was a continuing reduction in the risk of complications when the HbA1c was reduced below 8 per cent [4,5]. Although there are no similar studies in recipients of organ transplants who develop PTDM and the natural history of PTDM is not well defined, the importance of recognizing the possible impact of DM induced by immunosuppressive agents is clear. Moreover, it has recently been suggested that the prediabetic state (impaired glucose tolerance) is also associated with an increased risk of cardiovascular disease. It follows from this hypothesis that even a modest increase in glucose concentrations may be important, and a modifiable cardiovascular risk factor in transplant recipients [6]. Metabolic adverse effects of immunosuppressants are of critical importance, and efforts should now be directed towards finding an immunosuppressive regimen that is relatively free of metabolic effects. The immunosuppresants in current use: steroids, cyclosporin (CsA) and tacrolimus are diabetogenic, whereas mycophenolate mofetil (MMF, CellCept), azathiprine, sirolimus, and the emerging compounds — SDZ-RAD, FTY720, leflunomide, and deoxyspergulain (DSG) — do not appear to be diabetogenic. Advances, will depend on an improved understanding of the mechanisms of PTDM and the development of a unifying hypothesis [1–3].

Diabetes and risk of cardiovascular disease It is well known that there is an increased risk of cardiovascular and cerebrovascular complications in recipients of renal failure. However, it is not fully appreciated that this risk persists even after the treatment of renal failure by a renal transplant. Immunosuppressive agents not only cause glucose intolerance, but also increase serum lipids and cause hypertension. Immunosuppressants may therefore increase the risk of death by coronary disease, or stroke by three main mechanisms: dyslipidaemia, hyperglycaemia, and hypertension. Additional indirect mechanisms include changes in homocysteine,

248

IMMUNOSUPPRESSION AND DIABETOGENICITY

and activation of neuro-humoral mechanisms such as the renin –angiotensin system, nitric oxide, and endothelin. Despite the amelioration of renal failure by a successful transplant, we believe that these additional factors put the recipient in the same risk category of developing a serious cardiovascular or neurovascular event as before the transplant [7,8].

Diabetes and hyperlipidaemia The importance of dyslipidaemia in diabetes was not considered important in the past because the prevalence of high cholesterol was similar to that in the non-diabetic population. It has now been shown that unique abnormalities in the composition and metabolism of lipoproteins may occur in diabetics [9]. Several other factors related to hyperglycaemia may also contribute to the excess risk of atherosclerosis, such as deranged glycosylation or glyco-oxidation of various lipoproteins and arterial wall proteins, nephropathy and proteinuria, microvascular disease of vasa vasorum, and abnormalities in platelet function and haemostasis [10].

Definition of post-transplant diabetes mellitus The definition of PTDM has not been universally agreed. Investigators have either followed the World Health Organization (WHO) recommendation or defined their own criteria and, depending upon the definition and duration of follow-up, the incidence of PTDM has varied widely from 4 to 20 per cent. The precise definition of PTDM is particularly important for inter centre comparisons and for investigating the relative importance of individual risk factors such as dyslipidaemia, hypertension, and hyperglycaemia in long-term survivors of solid organ transplants. In contrast to criteria for the diagnosis of DM and other categories of glucose intolerance developed by the National Diabetes Data Group (NDDG) in 1979, and endorsed by the WHO in 1980 [11], various criteria have been employed to define PTDM. For example, in kidney transplant recipients, Mejia et al. [12] diagnosed PTDM when at least two determinations of serum glucose were above 11.1 mmol/l, and excluded temporary glucose intolerance induced by pulse steroids. Esmatjes et al. [13] used the NDDG criteria, while Friedman et al. [14] used the mean fasting blood glucose (FBS) of historical controls on the third day following transplantation as this time was found to be the of onset of PTDM. Sumrani et al. [15] defined PTDM as an FBS of more than 150 mg/dl on three separate occasions or an abnormal glucose tolerance test, Boudreaux et al. [16] two FBS levels greater than 140 mg/dl and an abnormal glucose tolerance test {GTT}, whereas Roth et al [17] defined glucose intolerance as three FBS exceeding 140 mg/dl, but did not perform GTT. In recipients of liver transplants, Krentz et al. [18] used the OGTT on the basis of fasting and 120 min venous whole blood glucose concentrations according to the WHO criteria. We used a broad criteria to define hyperglycaemia: an FBS over 400 mg/dl at any point or over 200 mg/dl for 2 weeks or necessitating insulin for at least for 2 weeks [19,20]. These were confirmed by OGTT as per WHO criteria. These variable definitions explain much of the variation in the reported incidence of PTDM. Although they will identify similar groups, a clearer definition — based on the WHO criteria — is required to follow the natural history of PTDM and permit the assessment of strategies to reduce glucose intolerance following transplantation.

Symptomatology of post-transplant diabetes mellitus The symptomatology of DM is well known, but chronic hyperglycaemia may occasionally be asymptomatic and may present with the consequences of diabetes such as coronary disease, stroke, retinopathy, nephropathy, or neuropathy.

R.M. JINDAL ET AL.

Hyperglycaemia after pancreas transplantation Hyperglycaemia after pancreatic transplantation can result from a number of causes including graft ischaemia, rejection, steroids, and drug-induced or auto-immune damage to islets. However, PTDM may not occur after successful vascularized pancreatic transplantation despite heavy immunosuppression. Elmer et al. [21] reported a study in which they looked at the metabolic effects of tacrolimus compared with CsA in portally drained pancreatic allografts. They concluded that tacrolimus appeared to impair glucose metabolism to a greater degree than CsA, but that long-term pancreatic function needs to be studied further. Gaber [22] has recently summarized the proceedings of the World Congress of the International Pancreas and Islet Society, Sidney, 1999. He reported that the technique of portal drainage after whole pancreas transplantation did not affect glucose control. Furthermore, the native pancreas appears more susceptible to the diabetogenic effect of immunosuppresants, raising the possibility that the denervated pancreas is able to withstand the toxicity better than the native pancreas.

Synopsis of proposed mechanisms of post-transplant diabetes The mechanism of PTDM is also not clear and a unified hypothesis has not emerged. However, some recent investigations have focused on the effects of immunosuppresants on the synthesis and secretion of insulin and on tissue uptake — the major factors responsible for regulation of insulin levels. Calmodulin may have a role in insulin secretion; CsA binds to calmodulin which, in turn, may have an inhibitory effect on insulin secretion [23]. A calmodulin inhibitor restored the insulin secretory capacity of pancreatic islets that was suppressed by CsA. It is known that the enzyme cis-peptidylpropyl isomerase A is a major binding site for tacrolimus and CsA. Inhibition of this enzyme may cause some of the side-effects which are common to both tacrolimus and CsA [24]. In the case of tacrolimus, toxic effects on the endocrine pancreas may be due to the selective localization of FKbinding protein (FKBP)-12 and calcineurin in the islets compared with acinar tissue. Neither tacrolimus nor CsA adversely affect the acinar portion of the pancreas [25]. At the molecular level, tacrolimus did not affect the glucose uptake by insulin into rat striated muscle cell line, but suppressed insulin production in the insulinoma cells. When tacrolimus was administered for 2 weeks at 10 mg/kg/day, there was a time-dependent inhibition of insulin production at the transcriptional level whereas glucagon was not affected. Interestingly, when tacrolimus was stopped, insulin mRNA transcription and insulin production returned to normal [26]. Other investigators have shown that tacrolimus did not have any effect on the number of insulin receptors, but caused inhibition of synthesis and secretion of insulin. With regard to steroids, the predominant effect in the causation of PTDM seems to be the induction of insulin resistance. Other mechanisms for steroid-induced PTDM have been suggested: decreased insulin receptor number and affinity, impaired peripheral glucose uptake in the muscle, impaired suppression of endogenous insulin production or activation of the glucose/FFA cycle. It has also been postulated that PTDM may be the result of a genetic defect in insulin secretion. However, firm evidence for this is lacking. Zimmerman et al. [27] showed that leucine (an essential amino acid) may be a criticalo factor in steroid-induced PTDM. In healthy volunteers, high-dose prednisone therapy was associated with a significant increase in the rate of leucine oxidation (leucine flux: an indicator of whole body proteolysis). This would partly explain the protein wasting observed during steroid therapy and a high rate of steroid-induced insulin resistance.

249

250

IMMUNOSUPPRESSION AND DIABETOGENICITY

Drug interaction Interaction among immunosuppresssive agents may also be important. Ost [28] suggested that renal transplant patients developed a cushingoid appearance despite the use of low-dose steroids, in combination with CsA. In this situation it was proposed that CsA may have potentiated the effect of steroids by reducing the clearance of the latter drug due to competitive inhibition, since both CsA and prednisone utilize the cytochrome P-450 for their metabolism. Both CsA and tacrolimus are potent inhibitors of the cytochrome P-450 3A4 and may thus reduce steroid metabolism by this pathway.

Effect of HLA type We investigated the effect of HLA type on the development of PTDM, but failed to find such an association in liver transplant recipients [29]. David et al. [30], however, observed an increased frequency of human leucoyte antigen (HLA)–A28 among renal transplant patients developing steroid-induced DM. D’Apice et al. [31], observed a strong association between antigens HLA B8, B18, Bw15, and Dw16, and the development of PTDM. However, these studies have been conducted in small groups of patients; larger studies need to be carried out before one can establish a definitive association between HLA antigens and PTDM. Sumrani et al. [15] reported an incidence of 11.6 per cent PTDM in recipients of renal transplantation treated in the CsA era, the risk factors in their patients were race, age, and most interestingly those with HLA A30 and Bw42 antigens but not the total dose and type of immunosuppressive therapy. PTDM also resulted in higher incidence of sepsis and poorer patient and graft survival. Hathaway et al. [32] suggested that there were four independent risk factor: age, family history, glucose intolerance (at day 4 to 7 post-transplant), and specific HLA types, while race, gender, or donor source were not predictive of PTDM. In a prospective study of renal transplant recipients from Norway reported that HLA B27 was significantly associated with PTDM. Using multiple stepwise logistic regression analysis in 173 consecutive recipients of renal transplants, other factors associated with PTDM were prednisone dose, age, family history of diabetes, and cytomegalovirus (CMV) infection. However, gender, body mass index, donor source, and CsA level did not influence PTDM.

Diabetogenicity of immunosuppressants: experimental studies Cyclosporin Studies in vitro and in rodents Studies in rodents have shown that CsA at 50 mg/kg and 15 mg/kg over 3 weeks resulted in hyperglycaemia. Pancreatic insulin extract was reduced to 33 per cent of controls and there was a 50 per cent decrease in the ␤-cell volume. There was a similar decrease in DNA synthesis and, at therapeutic levels. CsA caused irreversible inhibition of insulin secretion, although the oxidation, respiratory rate, morphology, and phospholipid content were unaffected [34]. When islets were cultured with therapeutic CsA, insulin release and islet-insulin were adversely affected, but the somatostatin-inhibitory mechanism, glucagon stores and arginine-stimulated glucagon release remained normal [35]. However, contradictory data have also been presented to show that there is a global inhibition of protein synthesis and that potentially irreversible vacuolization, degranulation, and damage to the endoplasmic reticulum of islet cells occur with CsA treatment [36]. Robertson [37] reported that CsA at therapeutic levels could irreversibly inhibit insulin secretion in rat islets. Their further studies showed that the inhibitory activity of CsA was not removed by washing the CsA-treated cultures, suggesting that CsA may have cumulative toxic effect on islet function. These findings may have important clinical implications. Andersson et al. [38] suggested that inhibition of DNA synthesis was a likely

R.M. JINDAL ET AL.

mechanism for imparied insulin synthesis. They showed that mouse islets cultured in the presence of CsA had imparied proinsulin biosynthesis and insulin release when stimulated with glucose. Insulin content and DNA synthesis were also decreased. Helmchen et al. [39] reported that severe degranulation of ␤-cells may the underlying cause. In contrast, when Nielsen et al. [40] cultured rodent islets, in the presence of therapeutic doses of CsA (100 ng/ml), for 5 days, the release of insulin was reduced by 36 per cent while the islet-insulin content was increased by 59 per cent. Glucagon content was unaffected and the authors concluded that CsA had a direct inhibitory effect on insulin release from isolated pancreatic islets. Gillison et al. [41] further showed that, under in vitro conditions, somatostatin-inhibitory mechanisms were partially intact in CsA-treated pancreata during glucoseinduced release. They also showed that CsA inhibition was specific for insulin release as glucagon stores and arginine-stimulated glucagon release were not affected by CsA. These findings contradict reports that CsA results in global inhibition of protein synthesis and, overall, the contradictory reports in this area may partly reflect difficulties in the in vitro study of CsA. A general consensus at the present time is that CsA reduces synthesis, storage, and secretion of insulin in islets and islet cells. An understanding of these mechanisms at the cellular level may assist the development of immunosuppressive agents without metabolic side-effects [42].

Studies in larger animal models Alejandro et al. [43] reported, in healthy beagles, that insulin secretion — in response to intravenous glucose and glucagon — was significantly inhibited by CsA (20 mg/kg/day) and persisted for 4 months after CsA was terminated. In ewes, 4 weeks of CsA therapy resulted in a greater degree of impairment of insulin synthesis than its release. Stegall et al. [44] showed that therapeutic doses of CsA impaired glucose tolerance in monkeys, which again persisted for 3 weeks on withdrawing CsA. However, when CsA was given to dogs who had intra splenic islet autografts there was not impairment of glucose metabolism. Pretreatment with CsA (1, 10, and 100 nmol), for 24 h, resulted in suppression of glucose-stimulated insulin secretion in a dose-dependent fashion. Schilfgaarde et al. [45] showed in the segmental pancreatic autotransplantation model in dogs that CsA had a detrimental effect on the function of ␤-cells in vivo, but was reversible upon lowering the CsA levels. In dogs, Wahlstrom et al. [46] reported that withdrawal of short-term CsA resulted in rapid recovery, whereas long-term CsA therapy resulted in slower recovery of glucose tolerance. Dresner et al. [47] studied the effect of oral CsA (15 mg/kg/day) on glucose homeostasis in a large animal model (ewes). Four weeks of CsA therapy caused significant changes in glucose metabolism that may not be detected by simple tests such as IVGTT. Their experiments demonstrated that insulin synthesis may be impaired to a greater degree than insulin release and confirmed the findings of others that CsA had deleterious effect on insulin secretion, proinsulin, and mRNA synthesis. Kneteman et al. [48] studied the effect of CsA on the glucose metabolism in dogs who had intra splenic islet autografts. Interestingly, there was no impairment of plasma glucose clearance or stimulated insulin secretion in the transplanted dogs. Overall these studies confirm consistent effects of CsA to impair glucose tolerance, by a variety of mechanisms including reduced transcription, synthesis, and secretion. These effects are reversible at least after short-term exposure and, interestingly, the denervated pancreas, or islets, are relatively resistant to these actions of CsA.

Tacrolimus Studies in vitro and in rodents The toxicity of tacrolimus in small animals models was first examined by Hirano et al. [49] who showed that tacrolimus caused glucose intolerance in rats given daily oral doses of 1, 5, or

251

252

IMMUNOSUPPRESSION AND DIABETOGENICITY

10 mg/kg/day for 14 days. Insulin secretion was impaired, insulin content of the pancreas was lowered and there was morphological damage to islet cells, however, these changes disappeared within 14 days after withdrawal of the drug. Tacrolimus when given in vivo at 1 mg/kg/day for 2 weeks also decreased insulin release. Treatment of human pancreatic islets, transplanted under the kidney capsule of diabetic mice, with tacrolimus (0.3, 1, and 3 mg/kg/day) did not produce significant alteration of glucose homeostasis in animals given 0.3 mg/kg for 7 days. However, those animals receiving 1 and 3 mg/kg/day had a significant delay in plasma glucose disappearance rate and impaired insulin secretion from the engrafted islets. There was no histological damage to islets by high doses of tacrolimus [50]. Carroll et al. [51] showed that tacrolimus inhibited insulin secretion at 10 and 100 ng/ml on rat pancreactic islets, whilst the combination of tacrolimus and CyA was not additive had no additive effect on glucose-induced insulin secretion. It is interesting to note that at low doses of tacrolimus there was a minimal, but reversible, effect on insulin secretion by rat islets. Ricordi et al. [52] gave parental tacrolimus to nude mice transplanted with human pancreatic islets. They found that intraperitoneal administration of tacrolimus for 1 week at 0.3 mg/kg/day did not produce significant alteration of glucose metabolism. However, higher doses (1 and 3 mg/kg/day) significantly reduced glucose disappearance rate and inhibited glucose-mediated C-peptide response.

Studies in large animal models The toxicity of tacrolimus on the pancreas in large animal models was confirmed by Strasser et al. [53]. Tacrolimus administration to normal beagles resulted in reduced glucose disappearance following intravenous glucose injection. Tacrolimus at 1 mg/kg/day reduced insulin secretion and glucose utilization, effects that persisted while tacrolimus was given. Ericzon et al. [54] found that when oral tacrolimus (1 or 10 mg/kg/day) was given to non-transplanted monkeys, hyperglycaemia was mild, but animals receiving intramuscular tracrolimus developed hyperglycaemia and died of emaciation. In the animals who underwent pacreatico duodenal transplantation, hyperglycaemia was reversible on dose reduction without breakthrough rejection. When tacrolimus was used intramuscularly there was a significant diabetogenic effect in baboons, who underwent renal transplantation, but the diabetogenicity was not observed when tacrolimus was switched to oral application at day 4 [55]. Overall, tacrolimus has a similar pattern of effects to CsA, but is more potent, and glucose intolerance can be limited by dose reduction.

Newer and emerging agents Deoxyspergualin An investigational immunosuppressant has exhibited the potential to be used for pancreatic transplantation [56]. Investigators have reported significant activity in extending allograft survival in models employing skin, kidney, heart, thyroid, and pancreatic grafts, and immunosuppressive activity has also been demonstrated in xenograft models [57].The exact immunological target and mechanism of action of DSG has not been defined but appears unique from that of CsA or tacrolimus and it was therefore inferred that DSG may be less diabetogenic. There are only a few reports of the effect of DSG on the endocrine function of the pancreas. Strandell et al. [58] reported that following islet culture at 4 ␥/ml of DSG, comparable to the pharmacological level achieved after administration of 2.5 mg/kg/day, ␤-cell function was unimpaired. Higher doses caused morphological changes, islets cultured in DSG (0.1 to 10 ␥/ml for 1 week) became fluffy with a ragged surface. Interestingly, these authors found that DSG administered to mice abrogated the diabetogenic effect of multiple doses of streptozotocin and that the withdrawal of DSG led to the development of delayed hyperglycaemia.

R.M. JINDAL ET AL.

We showed that intramuscular DSG at 5 mg/kg/day was toxic in rats as shown by weight loss, sepsis, respiratory symptoms, and liver dysfunction. However, even at this toxic dose, the pancreas was barely affected as demonstrated by IVGTT, blood glucose, and pancreatic insulin content [59]. Xenos et al [60] showed that incubation of rate and human islets with DSG at a variety of concentrations did not affect the secretory capacity of islets and the glucose metabolism was unaffected by administration of DSG at 1, 4, or 10 mg/kg for 1 week. An optimal immunosuppressant for pancreatic transplantation, either vascularized or islets, would be one that would not have any adverse effect on pancreatic function, such as inducing diabetes. DSG may be such a drug. Overall, review of the literature shows that DSG does have diabetogenic effect in high doses, but that lower doses may be useful for induction or treatment of rejection after pancreatic transplantation. The systemic toxicity seen in rats with DSG treatment is less of a concern in humans. Phase 1 clinical studies have shown minimal toxicity in cancer patients treated with a 3 h infusion of 2000 ␥/m2. The upper limit of DSG is probably in the range of 2 to 4 mg/kg/day. The use of DSG is likely to be limited to induction and treatment of acute rejection episodes as it can be given only parenterally: an oral formulation is not yet available. Uchida et al. [61] showed that the toxicity of DSG varied with the dosing time without compromising its efficacy in the rat cardiac allograft model. They suggested that DSG was better tolerated if administered in the early phase of the inactive period, an observation that may reflect circadian variation in immune function.

Sirolimus (rapamycin, Rapamune) This is a macrolide antibiotic produced by Streptomyces hygroscopicus an actinomycete. Sirolimus has been under investigation for its immunosuppressive properties in small and large animal models [62]. It has been shown to prolong graft survival in many animal models of transplantation for both heterotopic and orthotopic organ grafting, bone marrow transplantation, and also for islet cell transplantation. Whole blood levels between 5 and 60 ng/ml was seen to be adequate for allotransplants, the range of side-effects has been similar to calcineurin inhibitors with some notable exceptions such as increases in serum low-density lipoprotein (LDH), cholesterol, and triglycerides and mild reductions in platelet and white blood cell (WBC) counts. These abnormalities were reversed by dose reduction, countermeasure therapy, or discontinuation of sirolimus. When sirolimus was administered with CsA, increased toxicities were observed such as renal basophilia, pancreatic islet vacuolation and associated hyperglycaemia, thymic and testicular tubular atrophy and myocardial degeneration, and also increased fetal mortality in rodents. The increased toxicity was attributed to the significant increases in exposure levels of each compound and associated biological activity when given in combination [63]. Fabien et al. [64] reported on the impact of sirolimus monotherapy (intraperitoneally for 7 days) on glucose metabolism in diabetic mice transplanted with islet allografts. Animals that received 0.1 and 0.3 mg/kg/day showed prolongation of islet allograft survival that was not seen with 0.5, 1.0, and 5.0 mg/kg dosages. In addition, the latter group of animals showed erratic blood glucose control. There was also a decrease in insulin secretion if islet culture was prolonged for 72 h at high sirolimus levels. Andoh et al. [65] examined the effect of CsA and sirolimus combination therapy in a rat model of chronic CsA nephropathy (induced by giving CsA at the dose of 8 mg/kg/day). Rats were given sirolimus in a variety of doses, but at subtherapeutic dose (0.1 mg/kg/day) there was worsening of glucose metabolism and a potentiation of nephrotoxicity. They postulated that hyperglycaemia may be the underlying mechanism for the synergistic nephrotoxicity due to its known properties of accelerating tubulointestinal fibrosis. However, Kneteman et al. have shown the efficacy of low-dose CsA and sirolimus in preventing rejection of canine islet allografts [66]. More recently, they have examined the effect of sirolimus on the metabolic efficiency of intras splenic islet autografts in the dog model.

253

254

IMMUNOSUPPRESSION AND DIABETOGENICITY

They did not find an adverse effect of sirolimus (1 mg/kg/day orally for 1 month) given alone or in combination with CsA (levels maintained at 300 ␥/l intramuscularly in a single dose) on glucose homeostasis. An intriguing suggestion has been made that subtle hyperglycaemia may have masked nephropathy in sirolimus-treated animals [67].

SDZ-RAD This is a derivative of sirolimus bearing a hydroxyethyl chain at position 40; increasing the polarity and making it more soluble than the parent compound. Animal studies have shown it to have similar immunosuppressive profiles and a better toxicological profile than sirolimus in vivo and in combination with CsA, SDZ-RAD decreased the requirements for CsA (Neoral) by three to four times, associated with a reduction in side-effects. Combination of SDZ-RAD with tacrolimus has not been reported, but is likely to have similar benefits. In addition to in vitro immunosuppressive properties, SDZ-RAD is effective in rodent models of autoimmune diseases and allotransplantation at oral doses from 1 to 5 mg/kg/day. In unreported experiments (Novartis, data-on-file, personal communications), SDZ-RAD in single doses of 1, 10, and 30 mg/kg orally in the rat did not have any significant effect on growth hormone (GH), Ruteinizing hormone, (LH), prolactin, testosterone, cortisone, glucose, and calcium. However, after repeated oral administration of SDZ-RAD to Cynomolgus monkeys, an increased incidence of pancreatic islet cell degeneration was seen at 5 mg/kg/day. The combination of SDZ-RAD with CsA (Neoral), for 4 weeks in monkeys, showed unexpected findings of haemorrhage and arteritis in several organs including the pancreas. Thus, like sirolimus, SDZ-RAD in combination with CsA may have deleterious effects on glucose tolerance at high doses; further evaluation of the diabetogenicity profile was awaited.

Mycophenolate mofetil Platz et al. [68] reported on the immunosuppressive activity of MMF (20 or 40 mg/kg/day) in canine renal allografts when given in combination with steroids and subtherapeutic dose of CsA (5 mg/kg/day). Serious gastrointestinal side-effects were noted, but there were no adverse effects on glucose metabolism in doses sufficient to prevent rejection of kidney allografts. MMF was shown to facilitate islet allograft survival in mice. Moreover, there was development of tolerance after 30 days of treatment in some strains of mice; withdrawal of the drug resulted in the indefinite survival in about 70 per cent of the islet allografts. However, this effect has not been investigated in large animal models, and is unlikely to be relevant to human transplantation.

Leflunomide This compound has been shown to have immunoregulatory and anti-inflammatory properties [69,70]. Manna and Aggarwal [71] showed that treatment of a human T-cell line with leflunomide blocked tumour necrosil factor (TNF)-mediated nuclear factor (NF) ␬-B activation in a dose- and time-dependent manner which may be the molecular basis of its anti-inflammatory and immunosuppressive activity. Elder et al. [72] investigated the mechanism of immunosuppressive action of leflunomide in vitro and found that this compound inhibited pyrimidine and protein tyrosine kinase biosynthesis and was effective as for antirejection prophylaxis of adult porcine islets in NOD and BALB/c mice in combination with CsA [73]. In another report, leflunomide in combination with CsA and MMF was effective in preventing the destruction of fetal procine islet-like clusters placed under the kidney capsule of rats [74].Combination of leflunomide and a suboptimal dose of CsA was effective in preventing the rejection of rat allogeneic islets transplanted under the kidney capsule of STZinduced diabetic rats [75]. Rats treated with leflunomide at 10 or 20 mg/kg/day exhibited no effect on

R.M. JINDAL ET AL.

blood sugar even after 7 months of treatment, in contrast to the development of hyperglycaemia in rats receiving CsA at 10 mg/kg/day [76]. Furthermore, when syngeneic islets were placed under the kidney capsule or intraportality in STZ-diabetic rats, normoglycaemia was obtained with leflunomide 20 mg/kg/day. Leflunomide was also effectively combined with tacrolimus for islet allografts. In these reports, leflunomide did not have any adverse effect on the endocrine pancreas [77].

FTY720 This is a synthetic structural analogue of spingosine, related to the drug Myrinocin, which was isolated from culture filtrates of the fungus. Isaria sinclairii [78]. In several small animal models, FTY 720 had been shown to inhibit allogeneic mixed lymphocyte reactions and interleukin (IL)2-dependent proliferation .This agent has also been shown to prolong the survival of skin, heart, and liver allografts in the rat allogeneic model and, more recently, in the dog kidney allograft model. There was a synergistic effect of FTY720 on the immunosuppressive activity of CsA and/or sirolimus [79]. In these studies, there was no toxicity on the endocrine pancreas.

Brequinar Brequinar is a quinoline carboxylic acid derivative that inhibits pyrimidine synthesis at the level of dihydro-orotate dehydrogenase and has been examined in a number of allogeneic and xenogeneic animal models for its immunosuppressive activity. Antoniou et al. [80] used brequinar in combination with either leflunomide or tacrolimus in a heterotopic rat cardiac allotransplantation model. Brequinar in combination with either drug resulted in prolonged graft survival, but brequinar aloneat therapeutic dose (12 mg/kg orally) resulted in aortic – graft ruptures or clinical toxicity due to over immunosuppression. In toxicity studies of brequinar, the maximum tolerated dose was estimated at 5 to 10 mg/kg/day, and the toxicity increased when combined with CsA. The main side-effects were bone marrow suppression, weight loss, and thymic and villous atrophy in jejunum. Even at high doses, brequinar did not show adverse effects on the endocrine pancreas [81]. Overall, the newer and emerging immunosuppressant agents appear to have less adverse effect on the development of hyperglycaemia. However, this may reflect, in part, limited experience of their use in humans and will clearly be an important aspect of the future evaluation of these agents.

Post-transplant diabetes mellitus in recipients of solid organ transplants: clinical studies Pre-cyclosporin era Steroids continue to be the cornerstone of immunosuppressive regimens. Steroid-induced diabetes was first described by Ingle [82] and subsequently by Starzl [83] in renal transplant patients. Several mechanisms for steroid-induced PTDM have been suggested: decreased insulin receptor number and affinity [84], impaired peripheral glucose uptake in the muscle [85], impaired suppression of endogenous insulin production [86], or activation of the glucose/FFA cycle [87]. However, experimental studies have suggested that induction of insulin resistance may be the predominant defect. Both hyperinsulinaemia and hypoinsulinaemia have been reported in PTDM [88]. Ekstrand et al. [89] reported in their study of immunosuppressed kidney transplant recipients that DM was due to impaired storage of glucose as glycogen, but steroids did not have an inhibitory effect on insulin secretion. These authors also showed that kidney transplant patients were resistant to the glucoregulatory effect of exogenous insulin, but the oxidation of glucose and lipids was not affected by steroid therapy.

255

256

IMMUNOSUPPRESSION AND DIABETOGENICITY

In a further study, the same investigators [90] measured insulin sensitivity and secretion in three groups of patients (controls, kidney transplant patients who had normal OGTT, and patients who developed DM after transplantation). They showed that glucose utilization was reduced in non diabetic kidney transplant patients compared with controls, and insulin secretion was normal in relation to the degree of insulin resistance. The development of PTDM was associated with further deterioration of glucose storage, and first phase, second phase, and glucagon-stimulated insulin secretion was impaired as compared with kidney transplant patients who did not develop PTDM. They postulated that PTDM may be the result of a genetic defect in insulin secretion, a consequence of insulin resistance, toxic effects on the ␤-cell or a combination of these effects. Arner et al. [91] reported a much higher incidence of PTDM in renal transplant recipients who received steroids and azathioprine only. Twenty-five per cent of the patients developed persistent DM of whom 50 per cent required insulin. PTDM remained a significant complication of renal transplantation, until McGeown et al. [92], in 1980, showed the advantages of low dose steroids (20 mg prednisone from the day of transplantation) in kidney transplant recipients. Only four patients of 151 developed PTDM. Careful screening in renal allograft recipients however, revealed a much higher incidence of PTDM.

Cyclosporin The diabetogenic effect of CsA was reported first by Gunnarsson et al. [93] in patients where azathioprine was replaced by CyA. Tyden et al [94] suggested that CsA played an important part in the impairment of glucose metabolism in transplant recipients. Reid et al. [95] observed that CsA was concentrated in the pancreas, thus identifying a potential explanation for ␤-cell dysfunction in the long-term. Boudreaux et al. [16] compared a variety of immunosuppressive regimens in renal transplant patients, and showed that an increased incidence of PTDM (11.6 per cent) was associated with the use of CsA, despite reduction in steroids. A similar incidence has been reported by others [96, 97]. Robertson et al. [98] performed a double-blind study, assessing IVGTT and insulin secretion, in patients with multiple sclerosis randomly assigned CsA or placebo. Surprisingly, they did not find impairment of glucose tolerance or pancreatic islet function when CsA was given in conventional doses (4 to 8 mg/kg) for up to 2 years. However, it must be noted that steroids were not used. The incidence of PTDM in the CsA era has remained between 4 and 20 per cent, of whom insulin therapy will be required in approximately 40 per cent. A study of 1000 kidney transplant patients reported an incidence of 16 per cent, of whom 56 per cent became hyperglycaemic within 3 weeks of transplantation [14]. Sang-35, a CsA formulation bioequivalent to Neoral — the microemulsion pre-concentrate of CsA — has shown similar safety, tolerability, and diabetogenicity profile [99].

Tacrolimus There is convincing evidence that tacrolimus has steroid-sparing effects following transplantation [100]. This led to the notion that tacrolimus would result in decreased incidence of PTDM. Scantlebury et al. [101] reported their retrospective analysis of occurrence of DM in 24 patients who received kidney transplants. They found a 20 per cent incidence of new-onset DM with tacrolimus when compared with 7 per cent incidence with CsA. We analysed the diabetogenicity of tacrolimus plus low-dose steroids versus CsA plus higher dose steroids in a randomized, prospective trial involving 63 liver transplant recipients between February 1991 and October 1991. Tacrolimus was initiated at 0.1 mg/kg/day by continuous intravenous infusion, followed by 0.15 mg/kg given orally twice a day when oral intake was tolerated. CsA-treated patients received a single preoperative oral dose of 10 mg/kg. Post-operatively, CsA was begun intra-

R.M. JINDAL ET AL.

venously at 1 mg/kg twice a day and converted to oral therapy once adequate CsA levels were obtained. Of the 30 patients in the tacrolimus group, 30 per cent became hyperglycaemic, 23 per cent of whom required treatment with insulin, and two (7 per cent) patients became permanently insulindependent. In the CsA group 33 patients (42 per cent) became hyperglycaemic, 21 per cent of whom required insulin, and two (6 per cent) became permanently insulin-dependent. There was no significant difference between groups in the incidence of hyperglycaemia or of insulin dependence. Hyperglycaemia was often temporally related to administration of increased steroids for treatment of rejection (CsA: 12/14 patients, tacrolimus: 4/9), two patients in each group required permanent insulin treatment [20]. Krentz et al. [18] compared the diabetogenicity of CsA [mean standard deviation (SD) dose 6.3 ± 0.5 mg/kg/day] versus tacrolimus (0.13 ± 0.01 mg/kg/day) in liver transplant recipients who had their steroids withdrawn 6 weeks before the study. They found that blood glucose concentrations after glucose challenge were significantly elevated in both groups of patients compared with controls. These metabolic abnormalities were therefore independent of steroid administration and were more pronounced in tacrolimus-treated patients. In another study of liver transplant recipients [100], we analysed the steroid requirements in two groups of patients receiving either CsA or tacrolimus. We showed that the total steroids (mg) administered in CsA group was 9190.10 ± 1992.08 versus 6587.56 ± 2958.23 in the tacrolimus group (P < 0.0001). Further breakdown of steroid use (CsA group) for induction, rejection, and daily maintenance was 1600 ± 2722 ± 1739, and 4868 ± 630, respectively; while that for the tacrolimus group was 1209 ± 248, 2376 ± 910, and 3003 ± 112, respectively. Therefore, the tacrolimus group of patients had a significantly lower steroid usage for induction (P , 0.0001), maintenance (P , 0.0001), and in the total steroid requirement over 1 year (P , 0.0001). However, there was no significant difference in the steroid usage for the treatment of rejections (P = 0.485). There were no cardiovascular events in either group and graft survival in both groups was comparable. Tacrolimus allowed for lower doses of steroids at all treatment points, as originally observed by Todo et al. [55], however, the incidence of PTDM was similar in both groups at the end of 1 year. The American multicentre trial of tacrolimus and CsA reported a much higher incidence of PTDM at 1 year (19.9 versus 4 per cent) with tacrolimus [102], and that the rate of PTDM was higher in black patients (risk ratio 4.4). However, a recent 3-year follow-up showed that after the first year there was only a marginal increase in PTDM in both arms. At 36 months, 33 per cent of the tacrolimus patients who had earlier developed PTDM, no longer required insulin compared to only 17 per cent in the CsA arm [103]. In subsequent studies [104] in which patients were given tacrolimus and MMF with lower total doses of tacrolimus and steroids, the incidence of PTDM was less than 5 per cent in over 700 patients. However, a more recent meta-analysis of three randomized controlled clinical trials in renal transplantation of tacrolimus compared with CsA did suggest that tacrolimus was associated with an increased incidence of developing PTDM [105]. Studies aiming to eliminate steroids very early after transplantation by using tacrolimus/MMF combination with induction by anti-IL-2 antibodies will be watched with interest. In paediatric renal transplantation, lower tacrolimus levels and a more rapid taper of steroids resulted in the elimination of PTDM, and those patients who required insulin (on high-dose tacrolimus) became insulin-independent after reduction of the dose of tacrolimus or switch to CsA [106].

New and emerging agents A number of new immunosuppressive agents have been introduced which may be used in combination with either CsA or tacrolimus in the hope of reducing the toxicity and improving graft survival. Of these drugs, MMF and sirolimus have completed phase III trials and are available for clinical use.

257

258

IMMUNOSUPPRESSION AND DIABETOGENICITY

Mycophenolate mofetil Sollinger et al. [107] reported on phase I clinical trial and pilot rescue study of MMF in combination with CsA and prednisone in kidney transplant recipients. No adverse effects on blood glucose were reported. Ensley et al. [108] reported their findings on the use of MMF in cardiac transplant recipients who had evidence of mild rejection, MMF was substituted for azathioprine while CsA levels and steroids were maintained. With the exception of minor gastrointestinal symptoms, MMF was well tolerated, again without adversely affecting blood glucose homeostasis. Three most recent trials, in which MMF was compared with azathioprine and used in combination with CsA and steroids, suggest similar benefits. In these trials MMF (2 or 3 g/day) significantly reduced biopsy-proven rejection in the first 6 months following kidney transplantation [109–111]. A number of studies have reported similar experiences in pancreatic transplants [112]. In 109 kidney/-sol pancreas Transplants receiving MMF in a CsA-based regimen, compared to a historical group of 250 recipients of combined kidney and pancreas receiving azathioprine and CsA, MMF use reduced rejection episodes. Moreover, the MMF group had better initial graft function, shorter hospital stays, and a similar incidence of infections [113]. As a consequence, Stratta recently reported that MMF had become a part of the standard immunosuppressive therapy at most pancreas transplant centres [114]. In a randomized prospective trial of CsA/MMF/prednisone against tacrolimus/ MMF/prednisone in combined kidney and pancreas transplantation found no differences in the patient or graft survival, the rate of rejections, infections or metabolic problems. Both arms of this study received OKT3 induction [115]. However, another study of combined kidney and pancreas transplantation, using a noninduction strategy with tacrolimus, MMF, and steroids, patient and graft survival were excellent and comparable to historical controls using anti lymphocyte induction therapy [116]. Other studies have confirmed the efficacy of combined MMF and tacrolimus to achieve similar rejection rates to those achieved with induction therapy. Moreover, the combination of tacrolimus with MMF and steroids may allow steroid withdrawal in the majority of patients receiving pancreatic transplants [117]. The most recent 3-year follow-up of the European multicentre trial showed a definitive reduction in the incidence of graft loss [118]. The major toxicities were gastrointestinal upset, leucopenia, and infections. Although approximately 25 per cent of the patients discontinued MMF due to adverse events, diabetogenicity was not a major factor and, overall, at the present time MMF is likely to be used in regimes that aim to reduce rejection rates without altering glucose tolerance.

Deoxyspergualin Clinical data on the use of DSG is limited and confined to centres in Japan. DSG was used to rescue patients who had steroid-resistant rejections, where it was found to be at least as effective as antibody treatment and no specific metabolic adverse effects were seen. Investigators at the University of Minnesota used DSG to induce patients who received a combined kidney/-islet tranplant in the hope that DSG would allow for reduced doses of CsA or tacrolimus early after transplantation and facilitate islet implantation. DSG was used by Gores et al. [119], who demonstrated sustained islet function in two patients transplanted with unpurified islets (with a simultaneous kidney graft). DSG and ALG were used for induction, thus the potentially diabetogenic CsA and tacrolimus were avoided in the immediate post-operative period allowing engraftment of islets. DSG may also have had a specific effect in reducing primary non-function which is the result of a non-specific inflammatory reaction surrounding the allogeneic islet tissue [120].

Sirolimus Sirolimus has been evaluated in clinical trials to determine its efficacy for the prophylaxis and treatment of acute allograft rejection. The phase II studies demonstrated the safety, tolerability, and

R.M. JINDAL ET AL.

efficacy of sirolimus used in combination with CsA and steroids for the prophylaxis of acute renal allograft rejection [121]. The results of the two phase III pivotal studies showed that at doses 2 and 5 mg/day, sirolimus significantly reduced the incidence of biopsy-proven acute rejection at 6 months compared to both azathioprine and placebo. Patient and graft survival were excellent in both groups [122,123]. In a European study [124], first cadaveric renal transplant recipients were randomized to either CsA or sirolimus, all patients also received steroids and azathioprine. At 12 months, graft and patient survival and incidence of rejection was similar, however, the incidence of dyslipidaemia and hyperglycaemia was higher in the sirolimus group. Only one patient in each group developed insulindependent diabetes at the end of 1 year. These metabolic events were found to be concentrationdependent and improved or reversed when the target trough concentration was lowered from 30 to 15 mg/ml. In a study by Kahan et al. [125], 149 recipients of a mismatched cadaveric or living renal allograft, three groups received sirolimus or placebo with steroids and full dose of CsA while the other three groups received steroids, reduced dose CsA, and varying dose of sirolimus. There was a reduction of acute rejection episodes from 32 per cent in controls to 8.5 per cent in patients receiving sirolimus and full dose CsA. Adverse effects due to CsA, such as hypertension and PTDM, were not exacerbated but lipid and haematological abnormalities were more frequent in patients receiving sirolimus. Overall, despite concerns over the cardiovascular risk attributable to the treatable lipid alterations that accompany sirolimus use, the minor effects of low-dose sirolimus on PTDM and hypertension may make the profile of sirolimus attractive. Whether, this is associated with long-term patient benefits is being assessed in ongoing trials.

SDZ-RAD SDZ-RAD has a shorter half-life compared to sirolimus and has been suggested that this factor may confer the benefits of rapid attainment of steady state and elimination of effects after discontinuation of the drug. Kahan et al. [126] reported the safety and pharmacokinetics of a 4-week course of once daily, sequential ascending doses of SDZ-RAD capsules in renal transplant recipients on a stable regimen of CsA and prednisone. They found that the incidence of adverse effects was similar in sirolimus, and that SDZ-RAD did not have significant drug interaction with CsA. They further speculated that as SDZ-RAD reaches the steady state more rapidly, there may not be the need to give a loading dose as is recommended for initiation of sirolimus. The terminal elimination half-life of SDZRAD is about 20 h versus 60 h for sirolimus; a fact that may be beneficial in reducing the side-effects when combined with CsA or tacrolimus.

Interleukin 2 receptor blockade The introduction of two new anti-IL2 receptor monoclonal antibodies may facilitate combination therapies for early reduction of steroids. In pivotal studies of these compounds, both baxiliximab and declizumab were both well tolerated with an adverse event profile comparable to placebo [127]. Although these agents seek to reduce the incidence and severity of acute rejection without augmentating immunosuppression to the same extent as anti-CD3 antibodies, they are also likely to reduce metabolic complications. A clinical trial involving one of these monoclonal antibodies may facilitate combination therapies for early reduction of steroids. In pivotal studies of these compounds, both baxiliximab and daclizumab were both well tolerated with an adverse event profile comparable to placebo [127]. Although these agents seek to reduce the incidence and severity of acute rejection without augmentating immunosuppression to the same extent as anti-CD3 antibodies, they are also likely to reduce metabolic complications. A clinical trial involving one of these monoclonal antibodies, tacrolimus, and MMF will be undertaken in our unit. In this study one-third of the patients will have no steroids at all; a third will have standard steroid therapy withdrawn a day 7; and the final third

259

260

IMMUNOSUPPRESSION AND DIABETOGENICITY

arm will have steroids tapered and withdrawn at 3 months following transplantation. Another study planned by Roche Pharmaceuticals involves the use of daclizumab and MMF in combination with steroids and CsA (low dose or withdrawal) to optimize the renal function in recipients of renal allografts; the secondary efficacy parameters include hyperlipidaemia, PTDM, and hypertension. These studies will define the role of ‘minimalist’ drug therapy, albeit with the more potent immunosuppressive drugs currently available, for recipients of organ transplants on allograft rejection, function, and cardiovascular risk.

Leflunomide This agent has recently been approved by the American Food and Drug Administration (FDA) for the treatment of rheumatoid arthritis based on data from several double-blind trials comparing leflunomide versus methotrexate versus placebo [128]. In a randomized, double-blind, placebo-controlled 12-month study of 482 patients, both methotrexate and leflunomide were equally effective and the response rate was significantly higher than in patients receiving placebo. In these trials, leflunomide was not shown to have diabetogenic properties [128]. Leflunomide is not yet in trials for immunoprophylaxis of organ transplants, but it has been proposed that this drug may be suitable for islet transplantation due to its lack of diabetogenicity.

Brequinar A phase I study of brequinar in stable renal, liver, and heart transplant patients receiving maintenance CsA and prednisone therapy was carried out recently. Brequinar (given as a single oral dose of 0.5 to 4 mg/kg) had a lower oral clearance (12 to 19 ml/min) than seen in previous studies in cancer patients (30 ml/min) suggesting that drug interaction between brequinar and CsA may be an important issue in clinical trials [129]. In another phase I study of brequinar in combination with cisplatin, Burris et al. [130] found that coadministration of the two drugs did not affect the pharmacokinetics of either drug. Brequinar is readily soluble in aqueous solutions, can be administered intravenously or orally, and exhibits a high level of bioavailability after oral administration, it has an extended half-life making it a desirable immunosuppressive drug. Although brequinar has not exhibited specific endocrine toxicity, toxicities such as myelosuppression, mucositis in clinical studies, and reduced body weight, thymic atrophy, cellular depletion of bone marrow and splenic white pulp, and villous atrophy in the jejunum are likely to be problematic [131,132]. Furthermore, the potential synergism between brequinar and other compounds such as CsA will mean a careful evaluation of the therapeutic window.

Diabetogenicity of tacrolimus versus cyclosporin As stated above, the steroid-sparing effects of tacrolimus led to the belief that this drug would result in a decreased incidence of PTDM. The mechanisms of action of both CsA and tacrolimus are similar, but tacrolimus is 10 to 100 times more potent than CsA. In the European randomized trial comparing tacrolimus and CsA involving 545 liver transplant recipients, the overall incidence of hyperglycaemia and DM were higher in the tacrolimus group, but the use of insulin and oral antidiabetic medications was similar [133]. In the American multicentre trial comparing the efficacy of tacrolimus versus CsA in liver transplant recipients, hyperglycaemia was more frequent in the tacrolimus group (47 versus 38 per cent0 and at the end of a year, 11 patients in the tacrolimus group and four patients in the CsA were rendered insulin-dependent diabetic [134]. However, Mieles et al. [135] recorded the changes in glucose metabolism in 72 liver transplant recipients who were switched from CsA to tacrolimus and

R.M. JINDAL ET AL.

showed that there was no statistical increase in PTDM, perhaps reflecting the lower dosages required in stable transplant recipients. Reversal of tacrolimus-associated side-effects have been described for chronic nephropathy, neurotoxicity, and gastrointestinal toxicity. However, there are few reported cases where tacrolimus was replaced by CsA for poorly controlled DM. More recently, Fernandez et al. [136] compared the metabolic side-effects of low-dose maintenance tacrolimus and CsA in 14 recipients of liver transplants, who were on no or minimum doses of steroids 1 year after transplantation. They measured in vivo insulin action by euglycaemic hyperinsulinaemic clamp and arginine stimulation tests at normo- and hyperglycaemic conditions. Both insulin sensitivity and reduced insulin release in response to arginine stimulation were observed in patients receiving tacrolimus or CsA compared with controls, indicating a reduced ␤-cell secretory reserve in treated patients. Furthermore, they showed that the acute glucagon response to arginine during hyperglycaemia declined less in the tacrolimus and CsA groups, indicating a defect in the pancreactic ␤-cell–␣-cell axis. They concluded that both drugs were equally diabetogenic at conventional dosages despite the fact that both groups of patients were on equivalent dose of steroids. Thus, although tacrolimus is associated with an early increase in PTDM, the long-term incidence of PTDM is likely to remain similar to CsA-based regimens. In the long term, tacrolimus may have lesser detrimental effect on glucose metabolism because the total steroid requirement is markedly less when compared to CsA-based regimes. Whether this notion is borne out and results in reduced cardiovascular and bone disease will need to be assessed in longitudinal studies.

Effect of steroid reduction on post-transplant diabetes mellitus Investigators have used a variety of strategies to reduce steroid usage by either low-dose, alternate-day dose, or complete steroid withdrawal regimens. Fabrega et al. [137] investigated the role of steroid withdrawal in long-term stable renal transplant recipients with PTDM. Of the 12 patients analysed, acute rejection developed in two patients and chronic rejection in another three patients. At the end of 2 years, only four patients were free of insulin or oral hypoglycaemic agents, but all patients had a significant reductions in HbA1c. Accumulated evidence suggests that steroid withdrawal may be more successfully applied to liver transplant recipients than to those receiving other organs. In the case of liver transplants, McDiarmid et al. [138] found that dual therapy with CsA an azathioprine in stable liver transplant recipients was safe with a significant improvement in serum lipid profile. However, there was no difference in the blood sugars at the end of a year between the groups and, surprisingly, steroid withdrawal did not result in improvement of pre-existing DM in the control group. Rejection rates up to 50 per cent were seen in recipients of kidney transplants in whom steroid withdrawal was attempted and the large Canadian multicentre propsective trial showed a higher graft loss at 5 years in the steroid withdrawal group as compared to the placebo control group [139]. A similar problem was encountered in paediatric renal transplant recipients where a 56 per cent incidence of rejections was noted with increased graft loss [140]. Late steroid reduction, after a year from 10 mg/day to 5 mg/day, also reduced the incidence and severity of PTDM, similar to the experience of other investigators in the case of renal transplants [141]. Other investigators have successfully withdrawn steroids from transplant recipients with a low rate of breakthrough rejection and a beneficial effect on metabolic side-effects. In renal allograft recipients CsA at a dose of 6 mg/kg/day monotherapy did not result in PTDM at the end of 6.5 years [142]. In another small trial, Hricik et al. [143] withdrew steroids from seven stable renal transplant recipients with PTDM and in a single patient who received a combined kidney pancreas transplant.

261

262

IMMUNOSUPPRESSION AND DIABETOGENICITY

Of these patients, seven were able to discontinue antidiabetic therapy within 4 months of discontinuing steroids with a corresponding decline in HbA1c Jordan et al. [144] treated 77 renal transplant recipients with ongoing acute rejection of which 74 per cent could be salvaged with a significant reduction in the daily prednisone dose from about 22 to 7.5 mg/day. Furthermore, there was a small reduction in the post conversion glucose levels. Overall, it is clear that steroids can be withdrawn safely in many patients with beneficial metabolic effects. It may be possible to identify the patients least likely to suffer acute rejection and most likely to benefit metabolically. However, the risk of acute rejection and reduced graft survival nullify the benefits to the population of renal transplant recipients. As mentioned above studies are ongoing with newer monoclonal-based immunosuppressive regimes; additional studies are also in progress. In an ongoing multicentre European study sponsored by Fujisawa, tacrolimus is combined with MMF and steroids. After 3 months, either steroids or MMF are withdrawn while the third arm continues triple therapy. The recruitment phase of this trial is complete and the results should be available later this year. These studies will shape future attitude towards steroid withdrawal.

Correction of post-transplant diabetes Tolerance remains the ultimate goal of transplantation, and indeed some progress has been made in this regard, notably the induction of microchimerism by simultaneous transfusion of bone marrow with the solid organ [145,146]. However, for the foreseeable future it is likely that immunosuppression will be required to some form. The ideal combination of medications with the least side-effects should be balanced against the possibility of graft loss and rejection. The beneficial effects of steroidfree immunosuppression are greatest when these agents are not used at all or withdrawn early after transplantation. However, post-transplant recipients who may benefit from this approach are difficult to identify. A variety of assays have been attempted to detect immunological tolerance, but were not always reliable. Despite these concerns, we believe that it is appropriate to institute trials of steroid withdrawal in recipients of pancreatic transplants due to the better understanding of the risks versus benefits and the introduction of newer potent immunosuppressive agents in recent years. The goal should be to achieve near normal glcyaemia (HbA1c level no higher than 1 per cent above the upper normal limit) by diet and exercise therapy, staged introduction of oral hypoglycaemia agents, and finally insulin regimens. Reduction of PTDM should be part of an overall strategy to reduce the cardiovascular and neurovascular mortality in recipients of organ transplants. Early attempts to discontinue CsA resulted in an unacceptable loss of kidney transplant, and CsA withdrawal is not used in current clinical practice. In the case of tacrolimus, complete withdrawal of immunosuppression may be possible in selected cases of liver transplants. The risks and benefits of steroid withdrawal are discussed above [147]. Attempts to substitute azathioprine for prednisone have been made by investigators both for PTDM and hyperlipidaemia. However, these attempts have also resulted in a significant incidence of rejection episodes and prednisone had to be restarted in many patients. A further concern is danger of myelosuppression with azathioprine when increased to compensate for reduced dosages of prednisone, leading to the suggestion that prednisone should not be withdrawn in patients tolerating less than 1.5 mg/kg/day of azathioprine. Alternate-day steroid therapy, has been an attractive alternative to complete withdrawal of steroids. Curtis et al. [148] found that patients receiving alternate-day prednisone had a significant decrease in the total cholesterol, while Cattran et al. [149] found a beneficial effect on triglyceride levels when a similar protocol was used. It would seem reasonable to institute alternate-day steroids in stable trans-

R.M. JINDAL ET AL.

plant recipients at some point after transplantation, however, larger trials are indicated to study the validity of alternate-day steroids in minimizing steroid-related complications. Drug interaction between CsA and glibenclimide should be kept in mind as the coadministration of the two drugs resulted in a 57 per cent increase in CsA levels despite normal renal and liver function in six patients who developed PTDM after renal transplantation [150].

Natural history of post-transplant diabetes mellitus Shields et al. [151] examined the effect of pretransplant DM on survival after orthotopic liver transplant. They found that patient survival in the DM group was significantly worse than matched controls; the type of diabetes (insulin-treated) and patients with alcoholic liver disease did worse. Other studies found that pretransplant DM caused a significant increase in infections of all types [152]. Vesco et al. [153] in a retrospective study of 1325 consecutive renal transplant recipients in France found that 2.5 per cent of their cohort of patients developed PTDM requiring insulin at a mean of 5.7 ± 1.5 months post-transplant. Actuarial patient and graft survival rates were not significantly different. However, when compared with paired-control transplant patients, both patient and graft survival tended to better in the controls versus patients who developed PTDM. In a study from New York, the effect of PTDM on long-term graft and patient survival was examined. Twelve-year graft survival in diabetic patients was significantly worse (48 versus 70 per cent) and regression analysis showed that PTDM was a significant predictor of graft loss. Renal function at 5 years was also significantly inferior in the PTDM group. However, patient survival was similar at 12 years despite the fact that PTDM patients experienced both metabolic and secondary consequences of diabetes [154]. We retrospectively reviewed the charts of 978 patients who received a renal transplant in the CsA era between 1984 and 1999. Seventy-two (7.36 per cent ) patients had renal failure due to type 1 diabetes and seven (0.72 per cent) patients due to type II diabetes at the Western Infirmary (Glasgow, United Kingdom). Sixty-six (6.75 per cent) recipients developed PTDM. Using Kaplan–Maier life table analysis, the mean graft survival for the patients with PTDM was 8.89 years versus 11.1 years for non-diabetic patients, while for those with pre-existing type I diabetes mellitus was 9.29 years and 2.97 years for patients with type II diabetes. The mean survival of patients with type I, type II, and PTDM were 8.1, 3.7, and 11.0 years, respectively. The mean survival of the patients without pre-existing diabetes or PTDM was 13.0 years (P < 0.05; log rank test). There was no difference in the incidence of infections or rejections in the four groups. Both patient and graft survival were adversely affected by pre-existing diabetes and, to a much lesser extent, by PTDM (unpublished data, submitted for publication).

Paradoxical effects of immunosuppressants in autoimmune diabetes The inhibitory effect of CsA and tacrolimus on ␤-cell function is particularly interesting because patients receiving pancreatic grafts commonly receive one of these drugs. Moreover, some patients with type I DM are being treated with insulin and CsA. It would seem contradictory to treat DM with drugs that could decrease insulin secretion, increase insulin resistance or cause direct ␤-cell damage. There are, however, some animal data to show that CsA and tacrolimus may paradoxically benefit the patients with type I DM through its immunosuppressive action on the autoimmune component of

263

264

IMMUNOSUPPRESSION AND DIABETOGENICITY

DM [155]. In clinical diabetes prevention studies, it is difficult to differentiate glucose intolerance due to the appearance of the diabetes diathesis itself from the toxic effects of CsA on the ␤-cell. CsAinduced glucose intolerance may mask any beneficial effects of immunosuppressive therapy. Stiller et al. [156] in a study of 41 patients treated for 2 to 12 months with CsA reported a significantly higher incidence of recovery to a non-insulin state than would normally be expected. CsA was initiated at 10 mg/kg/day and monitored by radio-immunoassay (100 to 200 ng/ml). Of interest was the significant reduction of islet antibodies, maintenance of C-peptide to normal levels (up to 1 year), and minimum toxicity, except nephrotoxicity which returned to baseline when CsA was discontinued. Following a number of uncontrolled studies [157–159] CsA was evaluated in double-blind, placebo-controlled studies. These studies showed that in patients with new-onset DM, CsA did not have a beneficial effect on preservation of ␤-cell function. The positive responses were transient and nephrotoxicity was a major concern which, though reversible in some patients, caused irreversible kidney damage in a significant number. It seems that agents such as nicotinamide, intensive insulin therapy, and less toxic therapies (monoclonal antibodies and immunotherapy) may be of more benefit in human type I DM. Interestingly, BB rats treated with tacrolimus were protected from diabetes [160]. A serious shortcoming of this approach is the direct toxic insult to the ␤-cell when given tacrolimus, some of this effect is not reversible. It therefore remains to be seen if the toxicity of tacrolimus will outweigh the possible benefits of immune manipulation. Sirolimus was effective in preventing the onset and severity of disease in several animal models of autoimmune diseases such as insulin-dependent DM (IDDM), systemic lupus crythematosus (SLE), arthritis, uveoretinitis, and encephalomyelitis. Sirolimus was also effective in delaying the onset or the worsening of disease in these animal models suggesting the possible therapeutic application of this drug in autoimmune disorders [161]. The agent FTY720, which has immunosuppresive properties, although the mechanisms of action are as yet unclear, has been shown to produce dose-dependent inhibition of induction of delayed-type hypersensitivity and also inhibited joint destruction and paw oedema in a rat model of adjuvant arthritis. Another study showed that FTY720 ameliorated T-cell mediated autoimmune responses in rat model of collagen-induced arthritis and allergic encephalomyelitis. Studies are being planned to investigate the use of FTY720 in autoimmune animal models of diabetes [162]. Stosic-Grujicic et al. [163] investigated the effect of a leflunomide metabolite (A77 1726) in experimentally induced model of autoimmune diabetes produced by multiple low doses of STZ. Mice were injected intraperitoneally with A77 1726 for 10 consecutive days, either during the first 10 days of the disease or starting from day 10 after disease induction. Hyperglycaemia, pancreatic infiltrate, expression of interferon (IFN)-␥ and iNOS were reduced after early treatment in the dose range of 5 to 35 mg/kg/day while late treatment with high dose (25 mg/kg) arrested the progression of the inflammatory process.

Paradoxical effects of tacrolimus in the immunosuppression of pancreatic transplants Studies have shown that tacrolimus was effective in preventing the development of overt diabetes in rodent models of spontaneously occurring diabetes if administered early in the process; furthermore, tacrolimus treatment prevented autoimmune responses in both syngeneic and allogeneic islet transplants in this model of diabetes. However, tacrolimus treatment had to be initiated before the onset of the autoimmune process. The higher potency of tacrolimus than CsA lead investigators to treat pancreatic grafts with tacrolimus with the expectation that the better immunosuppressive profile and antiau-

R.M. JINDAL ET AL.

toimmune properties would outweigh the diabetogenicity of tacrolimus. There are data that tacrolimus was successful in rescuing rejecting pancreatic grafts on CsA in combined pancreas and kidney transplants. It was suggested that the key issue in these cases was the maintenance of trough tacrolimus levels at a relatively low level (8 to 15 ng/ml), which may reflect the growing experience that the incidence of PTDM fell progressively when the levels of tacrolimus and steroids were reduced [164]. The introduction of tacrolimus has also revived interest in solitary pancreas transplantation as an option in patients with unstable diabetes and a functioning kidney. In contrast to the results of simultaneous kidney pancreas transplant (graft survival about 75 to 90 per cent at 1 year), the results of pancreas transplant alone have remained poor (about 50 per cent at 1 year). Bartlett et al. [165] have showed that in using tacrolimus and percutaneous biopsy of the pancreas to diagnose rejections at an early stage, they were able to increase graft survival to 90.1 per cent as compared to 53.4 per cent in the CsA group. These results reflect the findings of a multicentre trial in which tacrolimus was used for induction in 82 cases and for refractory rejection in 61 cases in simultaneous kidney and pancreas transplants. The use of tacrolimus was associated with an increased survival of pancreatic allografts; only 5 per cent of the patients were switched back to CsA for PTDM [166]. Drachenberg et al. [167] recently reported on the biopsy finding obtained from pancreatic allografts randomized to receive either CsA or tacrolimus. They also correlated islet morphology with mean and peak levels of CsA and tacrolimus in serum and with glycaemia and steroid administration. Although islet damage was more severe in patients receiving tacrolimus versus CsA, these were not significant. There was a correlation between islet cell damage and serum levels of CsA or tacrolimus during the 15 days prior to biopsy; serial biopsies from two hyperglycaemic patients receiving tacrolimus showed that islet cell damage was discontinued.

Towards a specific immunosuppression for pancreas and islet grafts A meeting was organized in Lyon, France, in 1998, under the auspices of the International Pancreatic and Islet Transplantation Association to discuss PTDM. The aims of the meetings were to review the diabetogenic effects of the main immunosuppressants, analyse the mechanisms and the means of reducing diabetogenicity, and suggest an ideal immunosuppresive therapy for pancreatic and islet transplants. The recommendations below are based on the discussions at the end of the meeting [168]. 1. There is a need for a precise definition of PTDM to facilitate intercentre comparison and to study the natural history of PTDM. For the diagnosis of PTDM, blood sugar findings should be confirmed by oral GTT as per WHO criteria and serial measurements of HbA1c to monitor compliance and to predict the secondary complications. 2. Recommended therapy: quadruple immunosuppression of ATGAM, tacrolimus, MME, and prednisone. Steroid taper should begin at 3 months and be completed at 6 months. ATGAM may be given for 5 to 7 days until adequate levels of tacrolimus are obtained. Tacrolimus levels for the first 6 weeks should be 20 to 25 ng/ml, 15 to 20 ng/ml from 6 weeks to 3 months, and then 10 to 15 ng/ml form 3 to 6 months. Use of the two newer antiILE-2 receptor antibodies (Zenapax and Simulect) have been reported anecdotally in whole pancreas and islet transplants. Further evaluation of these agents over longer time is awaited. 3. New immunosuppressants such as sirolimus, MMF, DSG, brequinar, and leflunomide will need to be evaluated for their diabetogenicity and their role in pancreatic transplants.

265

266

IMMUNOSUPPRESSION AND DIABETOGENICITY

4. Studies to evaluate diabetogenicity at the molecular level may hold the key for pharmacological manipulation of current immunosuppressive regimens which may result in decreased metabolic complications. 5. Pancreatic islet transplants have not been uniformly successful. Therefore, it is not possible to make definitive recommendations on the use of immunosuppressive therapy. However, there is evidence that steroids may be detrimental to the implantation of islets. Results of islet after kidney transplantation may be improved in recipients who have had their steroids withdrawn at earlier time points. For combined kidney islet transplants, the preclinical studies using anti-CD40L antibody are encouraging. In islet transplants alone, steroids should be avoided. 6. Reduction of PTDM should be part of an overall strategy to reduce the cardiovascular and neurovascular mortality in recipients of organ transplants. In addition to a tight control of PTDM, comprehensive care of transplant recipients must include aggressive attempts to reduce other cardiovascular risk factors such as hypertension, smoking, dyslipidaemia, and obesity by a multidisciplinary team approach. 7. We recommend that stable recipients of vascularized pancreatic transplants should be candidates for steroid withdrawal protocols with steroid withdrawal commencing at 3 months post-transplant. However, there are no definitive immunological parameters to predict the success of steroid withdrawal. Until the development of such assays or clinical predictive indicators, steroid withdrawal strategies will have to be evaluated by the institution of large trials using appropriate placebo control and randomization. These trials should be undertaken cautiously after informing the patients of all potential risks; racial and geographical factors should also be taken into account when contemplating steroid withdrawal.

Summary The well-established immunosuppressants — steroids, CsA, and tacrolimus — are diabetogenic to a significant extent, while azathioprine, MMF, sirolimus, SDZ-RAD, DSG, FTY720, brequinar, and leflunomide have been shown by experimental and limited clinical trials to be free of significant diabetogenic properties. In vitro and in vivo studies have shown that sirolimus potentiates the action of CsA, and conversely, CsA has been shown to augment the action of sirolimus. However, at low doses, sirolimus does not increase the diabetogenicity of CsA, and CsA withdrawal protocols are currently under investigation. Thus sirolimus may have a role in limiting PTDM and due to improved pharmacokinetics and a shorter half-life, the rapamycin analogue, SDZ-RAD may be a safer alternative to sirolimus. The two newer anti-IL-2 receptor blockade agents have not shown diabetogenicity and may be useful in combination therapy in which steroids are completely eliminated. Evidence has accumulated that CsA and tacrolimus cause PTDM by multiple mechanisms, which may be due to a combination of decreased insulin secretion, increased insulin resistance or a direct toxic effect on the ␤-cell. In case of steroids, the induction of insulin resistance seems to be the predominant factor. Immunosuppressive agents increase the risk of death by coronary disease or stroke by three mechanisms: dyslipidaemia, hypertension, and hyperglycaemia. Post-transplant diabetes mellitus has emerged as a major side-effect of immunosuppressants. As recipients of organ transplants survive longer, the secondary complications of DM have assumed greater importance. There is a need for a precise definition of PTDM to facilitate intercentre comparison and to study the natural history of PTDM. We recommend broad criteria to define hyperglycaemia, such as a fasting blood glucose over

R.M. JINDAL ET AL.

400 mg/dl at any point or over 200 mg/dl for 2 weeks, or requiring insulin for at least for 2 weeks. These findings should then be confirmed WHO criteria, including GTT in borderline cases. We also recommend serial measurements of HbAIc to monitor compliance and to predict the secondary complications of PTDM. Steroid-sparing regiments have been shown to reduce the metabolic complications of immunosuppressants including PTDM. However, their use should be balanced against the increased incidence of rejections. Data have emerged to show that PTDM may be organ-specific, irrespective of the immunosuppression used. Tacrolimus causes a high incidence of PTDM in recipients of kidney transplants (up to 20 per cent in some reports); the diabetogenicity of CsA-based regiments is comparable to tacrolimus-based regimens in recipients of liver transplants. Studies in small and large animals have suggested that damage to the ␤cells by CsA or tacrolimus may be reversible on discontinuation of the drugs. However, in the clinical situation, these drugs must be continued indefinitely. A few clinical studies in which attempts were made to discontinue CsA resulted in an unacceptable loss of graft. In the case of tacrolimus, complete withdrawal of immunosuppression may be possible in selected cases of liver transplants. However, post-transplant recipients who may benefit from this approach are difficult to identify. A variety of assays have been attempted to detect immunological tolerance, but were not always reliable. It should be emphasized that in some early series, patients received doses of tacrolimus that were approximately two to three times higher than those currently used, which may have resulted in a higher incidence of PTDM. More recently it has been shown that tacrolimus was successful in salvaging whole pancreatic grafts which were maintained on CsA. Tacrolimus-based immunosuppression as primary therapy was also used with remarkably success in solitary whole pancreas transplants. In addition to a tight control of PTDM, comprehensive care of transplant recipients must include aggressive attempts to reduce the other cardiovascular risk factors such as hypertension, smoking, dyslipidaemia, and obesity by a multidisciplinary team approach. To optimize immunosuppression and reduce the metabolic side-effects we recommend a wider use of sequential immunosuppression by which tacrolimus-based immunosuppression may be used for induction followed by CsA/sirolimus/SDZ-RAD for maintenance. The use of MMF, sirolimus, SDZ-RAD, and monoclonal antibodies to reduce the dosages of steroids may result in a reduction of PTDM. We also recommend that stable transplant recipients who have PTDM or other risk factors for cardiovascular mortality should be candidates for steroid withdrawal protocols.

References 1 Jindal RM. Post-transplant diabetes mellitus: A review. Transplantation 1994;58:1289–98. 2 Jindal RM, Sidner RA, Milgrom ML. Post-transplant diabetes mellitus: The role of immunosuppression. Drug Safety 1997;16:242–57. 3 Weir MR, Fink JC. Risk for post-transplant diabetes mellitus with current immunosuppressive medications. Am J Kidney Dis 1999;34:1–13. 4 American Diabetes Association. Position statement — Implications of the Diabetes Control and Complications Trial. Clin Diabetes 1993;11:91. 5 Diabetes Control and Complications Trial Research Group. The absence of a glycemic threshold for the development of long-term complications: The perspective of the diabetes control and complications trial. Diabetes 1996; 45:1289–98. 6 Gerstein H, Yusuf S. Dysglycaemia and risk of cardiovascular disease. Lancet 1996;347:949–50. 7 Rao VK. Posttransplant medical complications. Surg Clin N Am 1998,78:113–32. 8 Aker S, Ivens K, Guo Z, Grabensee B, Heering P. Cardiovascular complications after renal transplantation. Transplant Proc 1998;30:2039–2. 9 Garg A. Treatment of diabetic dyslipidaemia. Am J Cardiol 1998;81:47B–51B.

267

268

IMMUNOSUPPRESSION AND DIABETOGENICITY

10 Maser RE, Wolfson SK Jr, Ellis D, et al. Cardiovascular disease and arterial calcification in insulindependent diabetes mellitus: interrelations and risk factor profiles. Pittsburgh Epidemiology of Diabetes Complications Study — V. Arterioscler Thromb 1991;11:958–65. 11 Welborn TA. The definition of diabetes. In: Nattrass M, Santiago JV, ed. Recent advances in diabetes. Edinburgh: Churchill Livingston, 1984:15–18. 12 Mejia G, Arbelaez M, Heao JE, Arango JL, Garcia A. Cyclosporine-associated diabetes mellitus in renal transplants. Clin Transplant 1989;3:260. 13 Esmatjes E, Ricart MJ, Ferrer JP, Oppenhaimer F, Vilardell J, Casamitjana R. Cyclosporines effect on insulin secretion in patients with kidney transplants. Transplantation 1991;52:500–3. 14 Friedman EA, Shyh TP, Beyer MM, Manis T, Butt KM. Posttransplant diabetes in kidney transplant recipients. Am J Nephrol 1985;5:196–202. 15 Sumrani NB, Delaney V, Ding ZK, et al. Diabetes mellitus after renal transplantation in the cyclosporine era — An analysis of risk factors. Transplantation 1991;51:343–7. 16 Boudreaux JP, McHugh L, Canafax DM, et al. The impact of cyclosporine and combination immunosuppression on the incidence of posttransplant diabetes in renal allograft recipients. Transplantation 1987;44:376–81. 17 Roth D, Milgrom M, Esquenazi V, Fuller L, Burke G, Miller J. Posttransplant hyperglycemia: Increased incidence in cyclosporine-treated renal allograft recipients. Transplantation 1989;47:278–81. 18 Krentz AJ, Dousset B, Mayer D, et al. Metabolic effects of cyclosporine A and FK506 in liver transplant recipients. Diabetes 1993;42:1753–9. 19 Jindal RM, Emre S, Meneses P, et al. Diabetogenicity of FK506 versus CyA in liver transplant recipients. Hepatology 1993;18:745A. 20 Jindal RM, Popescu I, Schwartz ME, et al. Diabetogenicity of FK506 versus cyclosporine in liver transplant recipients. Transplantation 1994;58:370–2. 21 Elmer DS, Abulkarim AB, Fraga D, et al. Metabolic effects of FK506 (tacrolimus) versus cyclosporine in portally drained pancreas allografts. Transplant Proc 1998;30:523–4. 22 Gaber AO. Effects of various immunosuppressive protocols on glucose metabolism following pancreas transplantation. Excerpta Med, 1999. 23 Krausz Y, Wollheim CB, Siegel E, Sharp GWG. Possible role for calmodulin in insulin release. Studies with trifluperazine in the rat pancreatic islets. J Clin Invest 1980; 66:603–7. 24 Harding MW, Galat A, Uehling DE, Schreiber SL. A receptor for the immunosuppressant FK506 is a cis-trans peptidyl-proyl isomerase. Nature 1989;341:758–60. 25 Hirano Y, Hisatomi A, Ohara K, Noguchi H. The effects of FK506 and cyclosporine on the exocrine function of the rat pancreas. Transplantation 1992;54:883–7. 26 Tamura K, Fujimura T, Tsutsumi T, et al. Transcriptional inhibition of insulin by FK506 and possible involvement of FK506 binding protein-12 in pancreatic beta-cell. Transplantation 1995;59:1606–13. 27 Zimmerman T, Horber F, Rodriguez N, Schwenk WF, Haymond MW. Contribution of insulin resistance to catabolic effect of prednisone on leucine metabolism in humans. Diabetes 1989;38:1238–44. 28 Ost L. Effects of cyclosporin on prednisolone metabolism. Lancet 1984;1:451. 29 Jindal RM, Sidner RA, Hughes D, et al. Metabolic problems in recipients of liver transplants. Clin Transplant 1996;10:213–17. 30 David DS, Cheigh JS, Braun DW Jr, et al. HLA-A28 and steroid induced diabetes in renal transplant patients. JAMA 1980;243:532–3. 31 d’Apice AJ, Mathews JD, Tait BD, Kincaid-Smith P. Association of HLA antigens with glucose intolerance following renal transplantation. Tissue Antigens 1978;11:423–6.

R.M. JINDAL ET AL.

32 Hathaway DK, Tolley EA, Blakely ML, Winsett RP, Gaber AO. Development of an index to predict posttransplant diabetes mellitus. Clin Transplant 1993;7:330–8. 33 Hjelmesaeth J, Hartmann A, Kofstad J, et al. Glucose intolerance after renal transplantation depends upon prednisolone dose and recipient age. Transplantation 1997;64:979–83. 34 Hahn HJ, Laube F, Lucke S, et al. Toxic effects of cyclosporine on the endocrine pancreas of Wistar rats. Transplantation 1986;41:44–7. 35 Eun HM, Pak CY, Kim CJ, McArthur RG, Yoon JW. Role of cyclosporine A in macromolecular sysnthesis of beta-cell. Diabetes 1987;36:952–8. 36 Yale JF, Roy RD, Grose M, Seemayer TA, Murphy GF, Marliss EB. Effects of cyclosporine on glucose tolerance in the rat. Diabetes 1985;34:1309–13 37 Robertson RP. Cyclosporine-induced inhibition of insulin secretion in isolated rat islets and HIT cells. Diabetes 1986;35:1016–19. 38 Andersson A, Borg H, Hallberg A, Hellerstrom C, Sandler S, Schnell A. Long-term effects of cyclosporin A on cultured mouse pancreatic islets. Diabetologia 1984;27(Suppl.):66–9. 39 Helmchen U, Schmidt WE, Siegel EG, Creutzfeldt W. Morphological and functional changes of pancreatic B cells in cyclosporin A-treated rats. Diabetologia 1984;27:416–18. 40 Nielsen J, Mandrup-Poulsen T, Nerup J. Direct effects of cyclosporine A on human pancreatic betacells. Diabetes 1986;35:1049–52. 41 Gillison SL, Bartlett ST, Curry DL. Inhibition by cyclosporine of insulin secretion — a beta cellspecific alteration of islet tissue function. Transplantation 1991;52:890–5. 42 Faraci M, Vigeant C, Yale JF. Toxic effects of cyclosporine A and G in Wistar rats. Transplant Proc 1988;20:963–8. 43 Alejandro R, Feldman EC, Bloom AD, Kenyon NS. Effect of cyclosporin on insulin and C-peptide secretion in healthy beagles. Diabetes 1989;38:698–703. 44 Stegall MD, Chabot J, Weber C, Reemtsma K, Hardy MA. Pancreatic islet transplantation in cynomolgus monkeys. Initial studies and evidence that cyclosporine impairs glucose tolerance in normal monkeys. Transplantation 1989;48:944–50. 45 Schilfgaarde van RV, Burg van der MPM, Suylichem van PTR, Goozen HG, Frolich M. Reversible suppression of canine beta cell function by cyclosporine A is dose dependent. Transplant Proc 1986;18:1556–7. 46 Wahlstrom HE, Akimoto R, Endres D, Kolterman O, Moosa AR. Recovery and hypersecretion of insulin and reversal of insulin resistance after withdrawal of short-term cyclosporine treatment. Transplantation 1992;53:1190–5. 47 Dresner LS, Andersen DK, Kahng KU, Mushi IA, Wait RB. Effects of cyclosporine on glucose metabolism. Surgery 1989;106:163–9. 48 Kneteman NM, Marchetti P, Tordjman K, et al. Effects of cyclosporine on insulin secretion and insulin sensitivity in dogs with intrasplenic islet autotransplants. Surgery 1992;111:430–7. 49 Hirano Y, Fujihira S, Ohara K, Katsuki S, Noguchi H. Morphological and functional changes of islets of Langerhans in FK506-treated rats. Transplantation 1992;53:889–94. 50 Rilo HL, Zeng Y, Alejandro R, Carroll PB, et al. Effect of FK506 on function of human islets of Langerhans. Transplant Proc 1991;23:3164–5. 51 Carroll PB, Boschero AC, Li MY, Tzakis AG, Starzl TE, Atwater I. Effect of the immunosuppressant FK506 on glucose-induced insulin secretion from adult rat islets of Langerhans. Transplantation 1991;51:275–8. 52 Ricordi C, Zeng YJ, Alejandro R, et al. In vivo effect of FK506 on human pancreatic islets. Transplantation 1991;52:519–22.

269

270

IMMUNOSUPPRESSION AND DIABETOGENICITY

53 Strasser S, Alejandro R, Shapiro ET, Ricordi C, Todo S, Mintz DH. Effect of FK506 on insulin secretion in normal dogs. Metabolism 1992;41:64–7. 54 Ericzon BG, Wijne RM, Kubota K, Kootstra G, Groth CG. FK506-induced impairment of glucose metabolism in the primate-studies in pancreatic transplant recipients and in nontransplanted animals. Transplantation 1992;54:615–20. 55 Todo S, Demetris A, Ueda Y, et al. Renal transplantation in baboons under FK506. Surgery 1989;106:444–51. 56 Jindal RM, Tepper MA, Soltys K, Cho SI. Deoxyspergualin — A novel immunosuppressant. Mt Sinai J Med 1994;61:51–6. 57 Jindal RM, Soltys K, Yost F, Beer E, Tepper MA, Cho SI. Xenotransplantation of pig pancreatic islets to rat using deoxyspergualin monotherapy. Transplant Proc 1994;26:1108–9. 58 Strandell E, Andersson A, Groth CG, Sandler S. Effects of (-) 15-deoxyspergualin on pancreatic islet B-cell function in vitro and on the development of diabetes after multiple low dose streptozotocin administration. Pharmacol Toxicol 1989;65:114–18. 59 Jindal RM, Soltys K, Yost F, Beer E, Tepper MA, Cho SI. Effect of deoxyspergualin on the endocrine function of the rat pancreas. Transplantation 1993;56:1275–8. 60 Xenos ES, Casanova D, Sutherland DE, Farney AC, Lloveras JJ, Gores PF. The in vivo and in vitro effect of 15-deoxyspergualin on pancreatic islet function. Transplantation 1993;56:144–7. 61 Uchida H, Kobayashi E, Matsuda K, et al. Chronopharmacology for deoxyspergualin: toxicity and efficacy in the rat. Transplantation 1999;67:1269–74. 62 Sehgal SN. Rapamune (RAPA, rapamycin, sirolimus): Mechanism of action immunosuppressive effect results from blockade of signal transduction and inhibition of cell cycle progression. Clin Biochem 1998;31:335–40. 63 Kahan BD. The role of rapamycin in chronic rejection prophylaxis: a theoretical consideration. Graft 1998;1:93–6. 64 Fabian MC, Lakey JR, Rajotte RV, Kneteman NM. The efficacy and toxicity of rapamycin in murine islet transplantation. Transplantation 1993;56:1137–42. 65 Andoh TF, Lindsley J, Franceschini N, Bennett WM. Synergistic effects of cyclosporine and rapamycin in a chronic nephrotoxicity model. Transplantation 1996;62:311–16. 66 Kneteman NM, Lakey JR, Wagner T, Finegood D. The metabolic impact of rapamycin (sirolimus) in chronic canine islet graft recipients. Transplantation 1996;61:1206–10. 67 Sharma K, Ziyadeh FN. Hyperglycemia and diabetic kidney disease. The case for transforming growth factor-beta as a key mediator. Diabetes 1995;44:1139–46. 68 Platz KP, Sollinger HW, Hullett DA, Eckhoff DE, Eugui EM, Allison AC. RS-61443 — a new, potent immunosuppressive agent. Transplantation 1991;51:27–31. 69 Xu X, Shen J, Mall JW, et al. In vitro and in vivo antitumor activity of a novel immunomodulatory drug, leflunomide: mechanisms of action. Biochem Pharmacol 1999;58:1405–13. 70 Jarman ER, Kuba A, Monterman E, Bartlett RR, Reske-Kunz AB. Inhibition of murine IgE and immediate cutaneous hypersensitivity responses to ovalbumin by the immunoregulatory agent leflunomide. Clin Exp Immunol 1999;115:221–8. 71 Manna SK, Aggarwal BB. Immunosuppressive leflunomide metabolite (A77 1726) blocks TNFdependent nuclear factor-kappa B activation and gene expression. J Immunol 1999; 162:2095–102. 72 Elder RT, Xu X, Williams JW, Gong H, Finnegan A, Chong AS. The immunosuppressive metabolite of leflunomide, A77 1726, affects murine T cells through two biochemical mechanisms. J Immunol 1997;159:22–7. 73 Mital D, Guo Z, Chong AS, et al. Successful xenotransplantation of adult procine islets in NOD and BALB/c mice with leflunomide and cyclosporine. Transplant Proc 1997;29:2166–7.

R.M. JINDAL ET AL.

74 Wennberg L, Karlsson-Parra A, Sundberg B, et al. Efficacy of immunosuppressive drugs in islet xenotransplantation: leflunomide in combination with cyclosporine and mycophenolate mofetil prevents islet xenograft rejection in the pig-to-rat model. Transplantation 1997;63:1234–42. 75 Guo Z, Chong AS, Shen J, et al. Prolongation of rat islet allograft survival by the immunosuppressive agent leflunomide. Transplantation 1997;63:711–16. 76 Guo Z, Chong AS, Shen J, et al. In vivo effects of leflunomide on normal pancreatic islet and syngeneic islet graft function. Transplantation 1997;63:716–21. 77 Rastellini C, Cicalese L, Leach R, et al. Prolonged survival of islet allografts following combined therapy with tacrolimus and leflunomide. Transplant Proc 1999;31:646–7. 78 Kahan BD. FTY720: A new immunosuppressive agent with novel mechanism(s) of action. Transplant Proc 1998;30:2210–13. 79 Stepkowski SM, Wang M, Qu X, et al. Synergistic interaction of FTY720 with cyclosporine or sirolimus to prolong heart allograft survival. Transplant Proc 1998;30:2214–16. 80 Antoniou EA, Deroover A, Howie AJ, Chondros K, McMaster P, D’Silva M. Immunosuppressive effect of combination schedules of brequinar with leflunomide or tacrolimus on rat cardiac allotransplantation. Microsurgery 1999;19:98–102. 81 Pally C, Smith D, Jaffee B, et al. Side effects of brequinar and brequinar analogues, in combination with cyclosporine, in the rat. Toxicology 1998;15:207–22. 82 Ingle DJ. The production of glycosuria in the normal rat by means of 17-hydroxycorticosterone. Endocrinology 1941;29:649. 83 Starzl TE. Experience in renal transplantation. Philadelphia: Sunders, 1964. 84 Rizza RA, Mandarino LJ, Gerich JE. Cortisol-induced insulin resistance in man: impaired suppression of glucose production and stimulation of glucose utilization due to a postreceptor defect of insulin action. J Clin Endocrinol Metab 1982;54:131–8. 85 Munck A. Glucocorticoid inhibition of glucose uptake by peripheral tissues: old and new evidence, molecular mechanisms and physiological significance. Perspect Biol Med 1971;14:265–9. 86 Kahn CR, Goldfine ID, Neville DM Jr, De Meyts P. Alterations in insulin binding induced by changes in vivo+ in the levels of glucocorticoids and growth hormone. Endocrinology 1978;103:1054–66. 87 Venkatesan N, Davidson MB, Huchinson A. Possible role for the glucose-fatty acid cycle in dexamethasone-induced insulin antagonism in rats. Metabolism 1987;36:883–91. 88 DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 1979;237:E214–23. 89 Ekstrand A, Ahonen J, Gronhagen-Riska C, Groop L. Mechanisms of insulin resistance after kidney transplantation. Transplantation 1989;48:563–8. 90 Ekstrand AV, Eriksson JG, Gronhage-Riska C, Ahonen PJ, Groop LC. Insulin resistance and insulin deficiency in the pathogenesis of posttransplantation diabetes in man. Transplantation 1992;53:563–9. 91 Arner P, Gunnarsson R, Blomdal S, et al. Some characteristics of steroid diabetes: a study in renal transplant recipients receiving high dose corticosteroid therapy. Diabetes Care 1983;6:23–5. 92 McGeown MG, Douglas JF, Brown WA, et al. Advantages of low dose steroid from the day after renal transplantation. Transplantation 1980;29:287–9. 93 Gunnarsson R, Klintmalm G, Lundgren G et al. Deterioration in glucose metabolism in pancreatic transplant recipients after conversion from azathioprine to cyclosporine. Transplant Proc 1984;16:709–12. 94 Tyden G, Brattstrom C, Gunnarsson R, et al. Metabolic control at 2 months to 4.5 years after pancreatic transplantation, with special reference to the role of cyclosporine. Transplant Proc 1987;19:2294–6.

271

272

IMMUNOSUPPRESSION AND DIABETOGENICITY

95 Reid M, Gibbons S, Kwok D, Van Buren CT, Flechner S, Kahan BD. Cyclosporine levels in human tissues of patients treated for one week to one year. Transplant Proc 1983;15:2434–7. 96 Yoshimura N, Nakai I, Ohmori Y, et al. Effect of cyclosporine on the endocrine and exocrine pancreas in kidney transplant recipients. Am J Kidney Dis 1988;12:11–17. 97 Nakai I, Omori Y, Aikawa I, et al. Effect of cyclosporine on glucose metabolism in kidney transplant recipients. Transplant Proc 1988;20:969–78. 98 Robertson RP, Franklin G, Nelson L. Intravenous glucose tolerance and pancreatic islet beta-cell function in patients with multiple sclerosis during 2-yr treatment with cyclosporin. Diabetes 1989;38:58–64. 99 Schroeder TJ, Cho MJ, Pollack GM, et al. Comparison of two cyclosporine formulations in healthy volunteers: bioequivalence of new Sang-35 formulation and Neoral. J Clin Pharmacol 1998;807–14. 100 Jindal RM, Popescu I, Emre S, et al. Serum lipid changes in liver transplant recipients in a prospective trial of cyclosporine versus FK506. Transplantation 1994;57:1395–8. 101 Scantlebury V, Shapiro R, Fung J, et al. New Onset of diabetes in FK506 vs. cyclosporine-treated kidney transplant recipients. Transplant Proc 1991;23:3169–70. 102 Pirsch JD, Miller J, Deierhoi MH, Vincenti F, Filo RS. A comparison of tacrolimus (FK506) and cyclosporine for immunosuppression after cadaveric renal transplantation. Transplantation 1997;63:977–83. 103 Laskow DA, Neylan JF III, Shapiro RS, et al. The role of tacrolimus in adult kidney transplantation: a review. Clin Transplant 1998;12:489–503. 104 Miller J, for the FK506-MMF Dose Ranging Kidney Transplant Study Group. Tacrolimus and mycophenolate mofetil in renal transplant recipients: one year results of a multicenter, randomised dose ranging trial. Transplant Proc 1999;31:276–7. 105 Knoll GA, Bell RC. Tarcrolimus versus cyclosporine for immunosuppression in renal transplantation: meta-analysis of randomised trial. Br Med J 1999;318:1104–7. 106 Moxey-Mims MM. Letter to the editor. Transplantation 1999;68:320. 107 Sollinger HW, for the U.S. renal transplant mycophenolate mofetil study group. Mycophenolate mofetil for the prevention of acute rejection in primary cadaveric renal allograft recipients. Transplantation 1995;60:225–32. 108 Ensely RD, Bristow MR, Oslen SL, et al. The use of mycophenolate mofetil (RS-61443) in human heart transplant recipients. Transplantation 1993;56:75–82. 109 Antoniadis A, Papachristou F, Gakis D, Takoudas D, Sotiriou I. Comparison between mycophenolate mofetil and azathioprine based immunosuppression in pediatric renal transplantation from living related donors. Transplant Proc 1998;30:4085–6. 110 European Mycophenolate Mofetil Cooperative Study Group. Placebo-controlled study of mycophenolate mofetil combined with cyclosporin and corticosteroids for the prevention of acute rejection. Lancet 1995;345:1321–5. 111 Tricontinental Mycophenolate Mofetil Renal Transplantation Study Group. A blinded, randomised clinical trial of mycophenolate for the prevention of acute rejection in cadaveric renal transplantation. Transplantation 1996;61:1029–37. 112 Kaufman DB, Leventhal JR, Stuart J, et al. Mycophenolate mofetil and tacrolimus as primary maintenance immunosuppression in simultaneous pancreas-kidney transplantation: initial experience in 50 consecutive cases. Transplantation 1999;67:586–93. 113 Sollinger HW, Odorico JS, Knechtle SJ, D’Alessandro AM, Kalayoglu M, Pirsch JD. Experience with 500 simultaneous pancreas-kidney transplants. Ann Surg 1998;228:284–96. 114 Stratta RJ. Review of immunosuppressive usage in pancreas transplantation. Clin Transplant 1999;13:1–12.

R.M. JINDAL ET AL.

115 Stegal MD, Simon M, Wachs ME, Chan L, Nolan C, Kam I. Mycophenolate mofetil decreases rejection in simultaneous pancreas-kidney transplantation when combined with tacrolimus or cyclosporine. Transplantation 1997;64:1695–700. 116 Reddy KS, Strata RJ, Shokouh-Amiri MH, et al. Simultaneous kidney-pancreas transplantation without anti-lymphocyte induction. American Society of Transplant Physicians, 17th Annual Meeting, Chicago, 1998. 117 Jordan ML, Shapiro R, Gritsch HA, et al. Long-term results of pancreas transplantation under tacrolimus immunosuppression. Transplantation 1999;67:266–72. 118 Halloran P, Mathew T, Tomlanovich S, Groth C, Hooftman L, Barker C for the International Mycophenolate Mofetil Renal Transplant Study Groups. Mycophenolate mofetil in renal allograft recipients: a pooled efficacy analysis of three randomised, double-blind, clinical studies in prevention of rejection. Transplantation 1997;63:39–47. 119 Gores PF, Najarian JS, Stephanian E, Lloveras JJ, Kelley SL, Sutherland DE. Insulin independence in type 1 diabetes after transplantation of unpurified islets from single donor with 15-deoxyspergualin. Lancet 1993;341:19–21. 120 Kenmochi T, Miyamoto M, Mullen Y. Protection of mouse islet allografts from nonspecific inflammatory damage by recipient treatment with nicotinamide and 15-deoxyspergualin. Cell Transplant 1996;5:41–7. 121 Kahan BD, Podbielski J, Napoli KL, Katz SM, Meier-kriesche HU, Van Buren CT. Immunosuppressive effects and safety of a sirolimus/cyclosporine combination regimen for renal transplantation. Transplantation 1998;66:1040–6. 122 McDonald AS for the Rapamune Global Study Group. A randomized, placebo-controlled trial of Rapamune in primary renal allograft recipients. Abstract 426, Congress of the Transplantation Society, Montreal, 1998. 123 Kahan BD for the Rapamune US Study Group. A phase III comparative efficacy trial of Rapamune in renal allograft recipients. Abstract 198. Congress of the Transplantation Society, Montreal, 1998. 124 Growth CG, Backman L, Morales J-M, et al. Sirolimus (rapamycin)-based therapy in human renal transplantation. Transplantation 1999;67:1036–42. 125 Kahan BD, Julian BA, Pescovitz MD, et al. Sirolimus reduces the incidence of acute rejection episodes despite lower cyclosporine doses in caucasian recipients of mismatched primary allografts: A phase II trial. Transplantation 1999;68:1526–32. 126 Kahan BD, Wong RL, Carter C, et al. A phase I study of a 4-week course of SDZ-RAD (RAD) in quiescent cyclosporine -prednisone-treated renal transplant recipients. Transplantation 1999;68:1100–6. 127 Kahan BD, Rajagopalan PR, Hall M, for the United States Simulect Renal Study Group. Transplantation 1999;67:276–84. 128 Strand V, Cohen S, Schiff M, et al. Treatment of active rheumatoid arthritis with leflunomide compared with placebo and methotrexate. Leflunomide rheumatoid Arthritis Investigators Group. Arch Intern Med 1999;159:2542–50. 129 Joshi AS, King SY, Zajac BA, et al. Phase I safety and pharmacokinetic studies of brequinar sodium after single ascending oral doses in stable renal, hepatic, and cardiac allograft recipients. J Clin Pharmacol 1997;37:1121–8. 130 Burris HA III, Raymond E, Awada A, et al. Pharmacokinetics and phase I studies of brequinar (DUP 785; NSC 368390) in combination with cisplatin in patients with advanced malignancies. Invest New Drugs 1998;16:18–27. 131 Buzaid AC, Pizzorno G, Marsh JC, et al. Biochemical modulation of 5-Fluorouacil with brequinar: results of a phase 1 study. Cancer Chemother Pharmacol 1995;36:373–8.

273

274

IMMUNOSUPPRESSION AND DIABETOGENICITY

132 Cody R, Stewart D, DeForni M, et al. Multicenter phase II study of brequinar sodium in patients with advanced breast cancer. Am J Clin Oncol 1993;16:526–8. 133 European FK506 Multicenter Liver Study Group. Randomized trial comparing tacrolimus (FK506) and cyclosporine in prevention of liver allograft rejection. Lancet 1994;344:423–8. 134 The US Multicenter FK506 Liver Study Group. A comparison of tacrolimus (FK506) and cyclosporine for immunosuppression in liver transplantation. N Engl J Med 1994;331:1110–15. 135 Mieles L, Gordon RD, Mintz D, et al. Glycemia and insulin need following FK506 rescue therapy in liver transplant recipients. Transplant Proc 1991;23:949–53. 136 Fernandez LA, Lehmann R, Luzi L, et al. The effects of maintenance doses of FK506 versus cyclosporine on glucose and lipid metabolism after orthotopic liver transplantation. Transplantation 1999;68:1532–41. 137 Fabrega AJ, Meslar P, Cohan J, Lash J, Pollak R. Long-term (24 month) follow-up of steroid withdrawal in renal allograft recipients with posttransplant diabetes mellitus. Transplantation 1995;60:1612–14. 138 McDiarmid SV, Farmer DA, Goldstein LI, et al. A randomized prospective trial of steroid withdrawal after liver transplantation. Transplantation 1995;60:1443–50. 139 Sinclair NRS, for the Canadian Multicenter Transplant Study Group. Low-dose steroid therapy in cyclosporine-treated renal transplant recipients with a well-functioning graft. Can Med Assoc J 1992;147:645–57. 140 Reisman L, Lieberman KV, Burrows L, Schanzer H. Follow-up of cyclosporine-treated pediatric renal allograft recipient after cessation of prednisone. Transplantation 1990;49:76–80. 141 Ratcliffe PJ, Dudley CR, Higgins RM, Firth JD, Smith B, Morris PJ. Randomized controlled trial of steroid withdrawal in renal transplant recipients receiving triple immunosuppression. Lancet 1996;348:643–8. 142 Cantarovich D, Dantal J, Murat A, Soulillou JP. Normal glucose metabolism and insulin secretion in CyA-treated nondiabetic renal allograft patients not receiving steroids. Transplant Proc 1990;22:643–4. 143 Hricik DE, Bartucci MR, Moir EJ, Mayes JT, Schulak JA. Effects of steroid withdrawal on posttransplant diabetes mellitus in cyclosporine-treated renal transplant recipients. Transplantation 1991;51:374–7. 144 Jordan ML, Shapiro R, Vivas CA, et al. FK506 ‘rescue’ for resistant rejection of renal allografts under primary cyclosporine immunosuppression. Transplantation 1994;57:860–5. 145 Jindal RM, Sahota A. The role of cell migration and microchimerism in the induction of tolerance after solid organ transplantation. Postgrad Med J 1997;73:146–50. 146 McDaniel HB, Yang M, Sidner RA, Jindal RM, Sahota A. Prospective study of microchimerism in transplant recipients. Clin Transplant 1999;13:187–92. 147 Hricik DE, Almawi WY, Strom TB. Trends in the use of glucocorticoids in renal transplantation. Transplantation 1994;57:979–89. 148 Curtis JJ, Galla JH, Woodford SY, Lucas BA, Luke RG. Effect of alternate-day prednisone on plasma lipids in renal transplant recipients. Kidney Int 1982;22:42–7. 149 Cattran DC, Steiner G, Wilson DR, Fenton SA. Hyperlipidemia after renal transplantation: natural history and pathophysiology. Ann Intern Med 1979;91:554–9. 150 Islam SI, Masuda QN, Bolaji OO, Shaheen FM, Sheikh IA. Possible interaction between cyclosporine and libenclimide in posttransplant diabetic patients. Ther Drug Monit 1996;18:624–6. 151 Shields PL, Tang H, Neuberger JM, Gunson BK, McMaster P, Pirenne J. Poor outcome in patients with diabetes mellitus undergoing liver transplantation. Transplantation 1999;68:530–5. 152 Navasa M, Bustamante J, Marroni C, et al. Diabetes mellitus after liver transplantation: prevalence and predictive factors. J Hepatol 1996;25:64–71.

R.M. JINDAL ET AL.

153 Vesco L, Busson M, Bedrossian J, Bitker MO, Hiesse C, Lang P. Diabetes mellitus after renal transplantation: Characteristics, outcome, and risk factors. Transplantation 1996;61:1475–8. 154 Miles AM, Sumrani N, Horowitz R, et al. Diabetes mellitus after renal transplantation: as deleterious as non-transplant-associated diabetes? Transplantation 1998;65:380–4. 155 Kai N, Motojima K, Tsunoda T, Kanematsu T. Prevention of insulitis and diabetes in nonobese diabetic mice by administration of FK506. Transplantation 1993;55:936–40. 156 Stiller CR, Dupre J, Gent M, et al. Effects of cyclosporine immunosuppression in insulin-dependent diabetes mellitus of recent onset. Science 1984;223:1362–7. 157 Rodger NW, Dupre J, Stiller CR, et al. The Canadian trial with cyclosporin: effects of immunosuppression in early onset type 1 diabetes mellitus. Pediatr Adolesc Endocrinol 1986;15:340. 158 Dupre J, Stiller CR, Gent M, et al. Clinical trials of cyclosporin in IDDM. Diabetes Care 1988;11(Suppl. 1):37–44. 159 Assan R, Feutren G, Sirmai J, et al. Plasma C-peptide levels and clinical remissions in recent-onset type 1 diabetic patients treated with cyclosporine A and insulin. Diabetes 1990;39:768–74. 160 Murase N, Lieberman I, Nalesnik MA, et al. Effect of FK506 on spontaneous diabetes in BB rats. Diabetes 1990;39:1584–6. 161 Martel RR, Klicius J, Galet S. Inhibition of the immune response by rapamycin, a new antifungal antibiotic. Can J Physiol Pharmacol 1977;55:48–51. 162 Yan H, Suzuki K, Li XF, et al. Immunosuppressive effect of FTY720 on autoimmune diabetes models. Transplant Proc 1998;30:3436–7. 163 Stosic-Grujicic S, Dimitrijevic M, Bartlett R. Leflunomide protects mice from multiple low dose streptozotocin (MLD-SZ)-induced insulitis and diabetes. Clin Exp Immunol 1999;117:44–50. 164 Gruessner RW. Tacrolimus in pancreas transplantation: a multicenter analysis. Tacrolimus Pancreas Transplant Study Group. Clin Transplant 1997;11:299–312. 165 Bartlett ST, Schweitzer E, Johnson LB, et al. Equivalent success of simultaneous pancreas kidney and solitary pancreas transplantation: A prospective trial of tacrolimus immunosuppression with percutaneous biopsy. Ann Surg 1996;224:440–52. 166 Gruessner RW, Burke GW, Stratta RJ, et al. A multicenter analysis of the first experience with FK506 for induction and rescue therapy after pancreas transplantation. Transplantation 1996;61:261–73. 167 Drachenberg CB, Klassen DK, Weir MR, et al. Islet cell damage associated with tacrolimus and cyclosporine: Morphological features in pancreas allograft biopsies and clinical correlation. Transplantation 1999;68:396–402. 168 Jindal RM, Dubernard J-M. Towards a specific immunosuppression for pancreas and islet grafts. Clin Transplant (in press).

275

This page intentionally left blank

Chapter 20

The economics of pancreas transplantation Robert J. Stratta

Introduction Insulin-dependent diabetes mellitus (IDDM) is a very costly disease despite current medical knowledge and treatment practices. In addition to the incalculable costs on quality of life, IDDM causes a tremendous economic burden on the individual and the health-care system. In 1997, the direct and indirect costs of diabetes approached $98 billion, which was nearly 10 per cent of total health-care expenditures in the United States for that year [1] (prices in this chapter are given in American dollars). Since diabetic nephropathy is the leading cause of endstage renal disease (ESRD), renal replacement therapy for diabetic ESRD represents a major concern in planning allocation of resources in this era of managed care [2,3]. In addition to the differential survival (outcome) between the various forms of dialysis and transplantation for the diabetic patient, the economic impact (resource utilization) of these modalities must be considered. Studies have shown that renal replacement therapy by kidney transplantation alone (KTA) for diabetic ESRD not only improves survival and quality of life but is also cost-effective in the long term when compared to dialysis [3–14].

Costs of diabetes management A number of recent studies have attempted to characterize the total annual medical expenditures for diabetes care and management. In a report from the American Diabetes Association (ADA), total medical expenditures incurred by the 7.5 million people diagnosed with diabetes in 1997 in the United States totalled $77.7 billion or $10 071 per diabetic patient, as compared with $2669 per person in non-diabetic patients [1]. Javor et al. studied a cohort of 200 adult type 1 diabetic patients from January 1993 through to June 1995 to determine the medical charges for treating diabetic ketoacidosis (DKA) episodes relative to direct medical care charges for diabetes [15]. The estimated annual medical care charge for each patient was $7855, including $13 096 per patient experiencing an episode of DKA versus $4907 per patient not experiencing an episode. The Diabetes Control and Complications Trial (DCCT) compared outcomes and resource utilization for intensive versus conventional insulin therapy in a cohort of patients from 1983 to 1992 [1618]. In the DCCT, the annual diabetes-related medical costs of intensive therapy were estimated at $4545. Medical charges not directly related to diabetes care were not included in the DCCT model. In a cost-effectiveness decision analysis model of treatment strategies for IDDM patients with ESRD, Douzdjian et al. estimated the annual cost of diabetes management (in 1996 dollars to be $13 509 [19].

278

ECONOMICS OF PANCREAS TRANPLANTATION

In 2000, Klonoff and Schwartz performed an economic analysis of interventions for diabetes, including conventional and intensive insulin therapy as defined by the DCCT [20]. In 1998 dollars, the lifetime costs of conventional versus intensive insulin therapy were estimated to be $75 815 and $114 535, respectively. Costs of improved glycaemic control in type 1 diabetes were allocated for outpatient care, inpatient care, self-care supplies, case management services, and adverse effects of therapy. Costs and cost-benefit ratios were both discounted at a rate of 3 per cent. The costs per year of life gained with intensive insulin therapy were estimated to be $32 886. Moreover, the costs per quality life year gained with intensive therapy were estimated to be 22 933. The authors concluded that improved glycaemic control of diabetes is a cost-effective strategy. Given the fact that patients being considered for pancreas transplantation have either labile or complication-prone diabetes, the estimated annual medical expenditures for diabetes care and management in these selected patients is about $14 000 to $15 000 per year.

Costs of dialysis Data is also available on the annual costs of dialysis therapy. In a recent review by Pastan and Bailey, the average costs of providing care for a patient receiving dialysis was estimated at $45 000 per year [21]. The costs of caring for patients being treated with haemodialysis ($46 000 per year) and peritoneal dialysis $41 000) were similar. In 1995, inpatient and outpatient expenditures for ESRD, including haemodialysis, peritoneal dialysis, and transplantation, totalled $13.1 billion, with 75 per cent of this cost borne by the federal government. In the 2000 Annual Data Report of the United States Renal Data System (USRDS) analyzing data through 1998, total estimated direct medical payments for ESRD by public and private sources were $16.7 billion [3,12]. Total Medicare spending per capita for all ESRD treatment modalities combined was $43 000. While Medicare spending for all dialysis patients averaged $51 000 per year, Medicare payments for transplant patients (not including payments for organ procurement) were only $18 000 per year. Haemodialysis payments averaged $53 000 per year whereas peritoneal dialysis payments averaged $47 000 per year. Costs per patient-year at risk was higher for diabetics ($50 000) than for non-diabetics ($38 000) [12]. Care for diabetic ESRD patients was more costly in each of the five age groups analysed. Adjusting for age, race, gender, and treatment modality made no difference as diabetic patients had uniformly higher costs than their non-diabetic counterparts.

Costs of kidney transplantation Transplantation represents a $4.8 billion industry in the United States. A number of studies have analysed the cost-effectiveness of kidney transplantation compared to other forms of renal replacement therapy [3–14]. Dialysis costs continue to accumulate for each subsequent year on dialysis, whereas the costs of kidney transplantation after the first year are almost exclusively related to immunosuppressant medications and follow-up. The long-term cost savings of kidney transplantation over dialysis are well known. It is useful to compare the costs of kidney transplantation to dialysis in terms of a break-even point, which is the time it takes for the initial high costs of transplantation to be recovered by saving the ongoing costs of dialysis [4,13,14,22]. Although a kidney transplant produces a net fiscal loss to the health-care system during the time before the break-even point, savings accrue beyond the break-even point for as long as the graft survives. In the recent past, this break-even point was estimated by Eggers to be at 4.6 years, depending on whether the kidney transplant was performed from a living (3.9 years) or a cadaveric (4.9 years) donor [4,13,22].

R.J. STRATTA

At present, total case rates for a kidney transplant and initial hospital admission are approximately $40 000. Schweitzer et al. recently analysed 184 consecutive adult recipients of laparoscopically procured living donor kidney transplants with an estimated average first year cost of $89 939 [14]. Annual transplant costs each subsequent year were estimated to be $16 043, and did not take into account expenditures related to the management of diabetes. The average cost to Medicare for organ acquisition was estimated from actual hospital charges to be $19 802. Treatment with tacrolimus and mycophenolate mofetil for 1 year at the average doses and reimbursement rates in this study was estimated to cost Medicare $10 743. Using these cost figures, Schweitzer et al. reported a break-even point for Medicare reimbursement of 2.7 years after living donor kidney transplantation [14]. In an updated analysis from the Health Care Finance Administration (HCFA) by Eggers, the break-even point was calculated to be 3.1 years [23]. In 2000, Smith et al. performed a Medicare payment analysis by analysing more than 5 million Medicare payment records for claims made between 1991 and 1996 [24]. By using USRDS and United Network for Organ Sharing (UNOS) Registry Data, economic information was obtained on 42 868 cadaveric and 13 754 living donor kidney transplant recipients. Average total Medicare payments (exclusive of organ acquisition costs) for all kidney transplants in the United States were $39 534 and $24 652 for cadaveric and living donor recipients, respectively (P < 0.0001) during the first post-transplant year. The average number of readmissions was 2.5 for cadaveric and 2.28 for living donor kidney transplant recipients. For patients who had Medicare as the primary payor, the transplant-related service charges accrued during the initial hospital admission were significantly higher for cadaveric donor ($79 730) versus living donor ($69 547, P < 0.0001) kidney transplant recipients. However, Medicare payments were similar regardless of donor source and averaged $28 447 for living donation versus $28 483 for cadaveric donation. Five-year cadaveric donor transplant average charges of $280 793 were significantly higher compared to the 5-year living donor transplant average charges of $223 529. Five-year Medicare payments were also significantly higher for cadaveric donor transplants ($118 099) compared to living donor transplants ($96 060, P < 0.0001). The authors concluded that Medicare payments are remarkably lower for living compared to cadaveric donor kidney transplants in every category [24].

Costs of simultaneous kidney pancreas transplantation Despite the fact that the outcomes of simultaneous kidney pancreas transplantation (SPK) continue to improve, escalating costs of medical care threaten to severely limit its application. Similar to studies analysing the cost-effectiveness of kidney transplantation versus dialysis, economic data on pancreas transplantation are accumulating Table 21.1). In 1993, Evans et al. reported on data obtained in conjunction with the National Cooperative Transplant Study [25]. Valid financial data were collected on 133 pancreas transplant performed in 1988. Due to outliers, transplantation procedure charges (from date of transplant to discharge) were reported as statistical medians. Professional fees were considered separately and excluded. The median charge for SPK was $55 888, with a median length of hospital stay of 21 days. Total SPK charges ranged between $33 733 and $93 848 for 50 per cent of the cases studied. Half of the patients had a hospital length of stay between 16 and 33 days. In comparison, the median charge for KTA was $33 504, with a median length of hospital stay of 14 days. For nearly 50 per cent of the transplants studied, the level of reimbursement that the hospital received was 80 per cent or less. This study reported also that the proportion of health maintenance organizations paying for pancreas transplantation was 43 per cent in 1986 and 55 per cent in 1992 [25].

279

280

Table 20.1 Pancreas transplant charges Center (reference)

Time period

Evans et al. [25] Eggers [4] Robertson [26] Sollinger et al. [27]

1988 1988 1992 1991–92

133 284 – –

SPK; SPK; SPK; SPK;

S-B S-B S-B S-B

Jones et al. [28]

1987–92 1993–95 1991 1995 1992–95

27 32 10 10 –

SPK; SPK; SPK; SPK; SPK;

S-B S-B S-B S-B S-B

1997 1997 1997 1996–98 1998–99 1991–97

236 101 104 14 20 42 20 23 23 26 16 16 16 28 27 27 – – 3 342

Stratta et al. [31] Holohan [29] Gruessner et al. [36]

Reddy et al. [35] Stratta et al. [39] Kuo et al. [44] Douzdjian et al. [45] Stratta et al. [47] Stratta et al. [48] Stratta et al. [50] Douzdjian et al. [19] Douzdjian et al. [52] Whiting et al. [54]

1994–95 1995 1990–95 1995–96 1997–98 1997–98 1996–98 1998–99 1998–99 1992–96 1996 1991–97

No. of transplants

Type/technique of transplant

Length of stay (days)

SPK; S-B PAK; S-B PA; S-B SPK; S-B SPK; S-E PA; S-B PAK; S-B SPK; S-B SPK; S-E SPK; S-B SPK; S-E SPK; S-B SPK; P-E SPK; P-E SPK; S-E SPK; P-E SPK; S-B PAK; S-B SPK

P-E, portal-enteric (drainage); S-B, systemic-bladder (drainage); S-E, systemic-enteric (drainage).

Hospital charges (US$)

Comments

21 23 – 29

55 65 90 67

27 15 16 13 –

96 096 121 000 112 261 110 950 68 000–110 000

Professional fees excluded Professional fees excluded Estimated Excludes organ acquisition and professional fees Includes 10 solitary PTXs Includes three solitary PTXs Professional fees excluded Professional fees excluded Professional fees excluded, multiple centers surveyed Excludes surgeon’s fees Excludes surgeon’s fees Excludes surgeon’s fees Excludes organ acquisition and professional fees Professional fees excluded

– – – 16 8 17 19 20 21 20 13 14 13 12 12 13 – – –

888 308 000 694

112 730 86 700 84 74 64 555 36 582 100 398 118 228 110 000 125 000 107 193 73 458 100 215 94 083 99 517 102 255 105 789 185 721 130 780 189 019

Professional fees excluded Professional fees excluded Professional fees excluded Professional fees excluded Professional fees excluded; includes nine solitary PTX Professional fees excluded Professional fees excluded Professional fees excluded Total 1-year charges Total 1-year charges Total 1-year charges

R.J. STRATTA

Based on a review of 284 Medicare kidney transplant stays coded as SPK, Dr Paul Eggers of the HCFA reported that the median Medicare charge in 1988 for SPK (less professional fees) was $65 308, with a median length of stay of 23 days as compared to $43 110 for KTA [4]. In 1992, Robertson reported that the estimated total inpatient charges for the initial hospital admission for SPK at the University of Minnesota were approximately $90 000 [26]. In 1993, Sollinger et al. analysed hospital charges, including professional fees, for a subset of SPK recipients transplanted in fiscal year 1991 to 1992 [27]. Average first admission charges for SPK were $67 694 versus $41 791 for KTA. With outliers excluded, the mean SPK charge fell to $60 754. Mean length of stay was 29 days after SPK versus 23 days after KTA. In 1996, Jones et al. reported on the experience with pancreas transplantation at the University of Louisville [28]. Results were divided into era 1 (March 1987 to December 1992) and era 2 (January 1993 to October 1995). In era 1, mean transplant admission charges were $96 096 with a mean length of stay of 27 days. In era 2, mean hospital charges were $121 000 with a mean length of hospital stay of 15 days. As part of a report by Holohan to the Office of Health Technology Assessment, the University of Rochester stated that they received a total payment for SPK of approximately $73 000 from private insurers, including $8000 for professional (surgeon) fees [29]. The Ohio Solid Organ Transplant Consortium reported average SPK charges from transplant admission to discharge of $76 032 between 1990 and 1993. The Mayo Clinic reported that pancreas transplant (costs) were approximately $100 000 for the transplant and first hospital stay. In summarizing the available data, Holohan reported that charges for the transplant hospital admission after SPK ranged from $68 000 to $110 000, exclusive of professional fees. Total charges for SPK and 1 year of follow-up care ranged from approximately $97 000 to $189 000. Douzdjian et al. estimated total costs in the first year after SPK (including the transplant event) to be $185 271 [19]. Magee et al. studied resource utilization in 67 consecutive SPKs with bladder drainage performed from October 1992 to November 1996 (minimum 6 months follow-up) [30]. Resource utilization, facility costs, and professional fees (excluding the transplant hospital admission) were retrospectively studied using an institutional cost accounting system. Excluding the transplant hospital admission, total inpatient plus outpatient first year facility costs averaged $42 489 and total first year professional fees were $10 489. The authors noted that after the first 6 months, utilization declined sharply with the exception of the need for enteric conversion in 24 per cent of patients.

Factors influencing simultaneous kidney and pancreas transplantation charges In 1997, Stratta et al. reported on initial hospital admission charges in 10 SPK patients transplanted in 1991 versus 10 transplanted in 1995 [31]. The two groups were well matched, outliers were excluded, and professional fees were not considered. No attempt was made to convert 1991 to 1995 dollars. Total hospital charges for SPK were no different in 1991 and 1995 and averaged $111 605. Pharmacy, organ acquisition, and clinical laboratory services accounted for nearly 80 per cent of charges in each group. During the initial transplant hospital admission, the 1995 group experienced significant reductions in total number of laboratory tests, clinical laboratory charges, and total inpatient charges (with organ acquisition charges excluded). However, these potential savings were offset by a nearly 47 per cent increase in organ acquisition charges and a 38 per cent increase in medical/surgical supplies. Consequently, total hospital charges for SPK were no different in 1991 and 1995. The stabilization of hospital charges over time was attributed to implementation of a critical pathway that was successful

281

282

ECONOMICS OF PANCREAS TRANPLANTATION

in significantly reducing the mean length of hospital stay from 16.3 to 13.5 days [32]. In virtually all of these economic studies of pancreas transplantation, hospital charges are directly related to the length of hospital stay. In a follow-up study by Stratta, 10 SPKs performed in 1995 with standard hospital charges were compared to 10 SPK patients transplanted between 1990 and 1996 with ‘outlier’ hospital charges [33]. Outliers were defined as patients with initial transplant hospital admission charges exceeding 2 standard deviations above the mean hospital charge ($127 000 ± $38 500) generated during the previous 5 years. Therefore, patients in the outlier group had initial hospital charges in excess of $204 000. The two groups were well matched for demographic, immunological, and perioperative clinical characteristics. A number of specific cost centres were analysed in order to determine differential effects on outcome and resource utilization. Mean hospital charges were nearly $260 000 in the outlier group as compared to $111 000 in the standard group. With the exception of organ acquisition, each cost centre was significantly increased in the outlier group. The outlier group was characterized by: (a) a four-fold increase in laboratory testing and radiographic procedures; (b) a 3.5-fold increase in length of stay; and (c) a significant (P < 0.05) increase in the incidence of surgical complications and infections usually related to the pancreas allograft. Based on this analysis, it was concluded that surgical complications have a major impact on inpatient charges and outcome after SPK, suggesting that their prevention is critical to improving outcomes and competing effectively in the setting of managed care. In addition to prolonged length of stay and surgical complications, other markers of fiscal inefficiency include infection, rejection, and readmissions [34]. In 2000, Reddy et al. analysed initial hospital charges in sequential cohorts of SPK recipients with either bladder or enteric exocrine drainage [35]. Hospital charges were analysed according to the following categories: pharmacy, inpatient room, laboratory, operating room, medical-surgical supply, radiology/nuclear medicine, and miscellaneous. Organ acquisition charges and professional fees were not included in the analysis. The mean hospital stay for patients with enteric drainage was 7.8 ± 2.2 days (range 5 to 12, median 7.5 days) compared with 15.9 ± 7 days (range 8 to 38, median 15 days) for patients with bladder drainage (P = 0.002). The mean hospital charges during initial hospital admission for the enteric drainage group were $36 582 ± $11 424 compared to $64 555 ± $29 054 for the bladder drainage group (P = 0.005). There was a significant decrease in charges related to pharmacy, inpatient room, laboratory, radiology/nuclear medicine, and miscellaneous category in the enteric drainage group compared with the bladder drainage group, while the charges related to the operating room and medical/surgical supply were no different between groups. Patient and graft survival rates were comparable between groups. The authors concluded that SPK patients with enteric drainage had a 43 per cent reduction in hospital charges due to a shorter hospital stay and a reduction in pharmacy, radiology/nuclear medicine, and laboratory charges.

Effect of surgical complications In 1997, Gruessner et al. reported that median hospital charges were $112 730 for SPK, $86 700 for sequential pancreas after kidney transplant (PAK), and $84 740 for pancreas alone (PA) transplant [36]. Median surgeon fees were $10 840 for SPK, $8010 for PAK, and $7300 for PA. Univariate and multivariate analysis of donor and recipient risk factors revealed that older recipient age, relaparotomy, and older donor age each significantly increased the costs of transplantation. Recipient age (per 10 year increments) increased median hospital charges by $15 000; relaparotomy (number) by $36 000; and donor age (per 10 year increments) by $5000. When separate analyses were performed for SPK, PAK, and PA, relaparotomy was consistently the major risk factor affecting costs in each

R.J. STRATTA

recipient category. Relaparotomy increased median hospital charges by $35 000 after SPK, $42 900 after PAK, and $25 000 after PA. In 1998, Troppmann et al. analysed surgical complications after pancreas transplantation in the ciclosporin era [37]. The overall rate of relaparotomy in the first 3 months after transplant was 32 per cent (36 per cent after SPK, 25 per cent after PAK, 16 per cent after PA). Median total hospital charges during the first 3 months after transplant were significantly higher in all three recipient categories for those with versus those without relaparotomy (SPK $169 000 with versus $101 700 without relaparotomy; PAK $108 500 with versus $73 000 without relaparotomy; PA $126 900 with versus $78 200 without relaparotomy). Multiple relaparotomies were performed in 28 per cent of patients with surgical complications. Relaparotomy resulted in an increase in median hospital charges of $67 300 for SPK, $35 500 for PAK, and $48 700 for PA. Therefore, the median cost of pancreas transplants with (versus without) surgical complications was augmented by 67 per cent in SPK, 49 per cent in PAK, and 62 per cent in PA recipients. In a follow-up study from the University of Minnesota, Humar et al. correlated prolonged preservation (over 20 h) with an increased incidence of surgical complications [38].

Costs of solitary pancreas transplantation The charges associated with PA transplants are comparable to those for kidney transplantation and less than those for other solid transplant procedures. Additional charges for adding a pancreas to a kidney transplant are slightly lower than for KTA. In 1993, estimated total first-year charges for PA were $65 000 (versus $87 700 for KTA) by actuarial analysis [25]. Projected total 5-year charges were $70 300 for PA versus $124 900 for KTA based on 1993 actuary data. These data are actuarial and may not accurately reflect the long-term charges associated with PA. In addition, many of the charges for SPK may be covered under the diagnosis-related group (DRG) for kidney transplantation, making it difficult to estimate accurately the charges for pancreas transplantation. Late hospital admission or reoperations for pancreas-related problems could also add to the total costs of pancreas transplantation. In 1997, Stratta et al. analysed 62 consecutive solitary transplant recipients, including 42 PAs and 20 sequential PAKs [39]. Mean hospital charges (excluding professional fees) for the initial transplant hospital admission were $100 398 after PA and $118 228 following PAK. In a follow-up study, the use of antilymphocyte induction therapy in PAK recipients was studied [40]. In PAK patients receiving OKT3 induction, mean hospital charges for the initial hospital admission were $121 493. In patients receiving ciclosporin, prednisone, and azathioprine without OKT3 induction, mean initial hospital charges were $112 446.

Bladder versus enteric drainage Due to a favourable experience with enteric conversion after pancreas transplantation with bladder drainage coupled with advances in preservation, immunosuppression, and diagnostic technology, a resurgence of interest has occurred in primary enteric drainage. In the United States, the number of pancreas transplants performed with enteric drainage has increased from 15 per cent in 1995 to over 60 per cent in 1999. Both single-centre and Registry reports have noted comparable results with less morbidity associated with primary enteric drainage [41–45]. In 1997, Kuo et al. compared sequential cohorts of 23 SPK recipients with bladder drainage versus 23 with enteric drainage [44]. Because of the differing lengths of follow-up, data were analysed with respect to the first 6 months after transplant. Length of stay (mean 20 days) and total hospital charges (mean $118 000) were similar in the

283

284

ECONOMICS OF PANCREAS TRANPLANTATION

two groups. Charges accrued following the initial hospital admission were, on average $55 000 and $44 000, respectively, in the bladder and the enteric drainage groups. The mean number of readmissions in the first 6 months were likewise comparable, including 1.9 per patient in the bladder and 1.7 per patient in the enteric drainage groups. Consequently, total hospital charges for the first 6 months after SPK were equivalent (mean $165 000 in the bladder and $169 000 in the enteric groups). Douzdjian and Rajagopalan studied 42 consecutive SPK recipients, including 26 with bladder drainage followed by a sequential cohort of 16 with enteric drainage [45]. Length of initial hospital stay was significantly shorter with enteric drainage (mean 13 days with enteric versus 20 days with bladder drainage) resulting in a reduction in total hospital charges for the initial admission (mean $73 458 for enteric versus $107 193 for bladder drainage). Patients with bladder drainage were cared for exclusively in the inpatient unit, while patients with enteric drainage were cared for both in inpatient and outpatient units. The total number of readmissions (mean 1.7 per patient with bladder versus 1.2 per patient with enteric drainage) and the length of additional hospital stay (mean 14 days with bladder versus 10 days with enteric drainage) in the first 6 months after transplant were similar between groups. Douzdjian et al. subsequently reported that establishment of an outpatient transplant unit resulted in significant cost savings after SPK without jeopardizing outcomes [46]. In a prospective evaluation of pancreas transplantation with systemic-bladder versus portal-enteric drainage, mean hospital charges for the initial transplant hospital admission (excluding professional fees) were $100 215 and $94 083, respectively [47]. Morbidity and length of stay (mean 13 days) were comparable between the two groups. In a parallel study of 28 SPK recipients with portal-enteric drainage and no antilymphocyte induction transplanted between September 1996 and November 1998, mean initial hospital charges were $99 517 with a mean length of stay with 12.5 days [48]. In another study of 30 consecutive SPK recipients performed with either systemic-bladder or portalenteric drainage and no antilymphocyte induction, median initial hospital charges were $93 602 with a median length of stay of 11.5 days [49]. The median number of readmissions and days hospital in hospital were 1 and 2, respectively, in the first 3 months after SPK. The median hospital charge per readmission was $5462, with charges totaling $110 072 in the first 3 months. In 2001, Stratta et al. performed a prospective evaluation of SPK with systemic-enteric versus portal-enteric drainage [50]. Mean initial hospital charges for the transplant event (excluding professional fees) were $102 255 and $105 789, respectively. Morbidity and length of stay (mean 12.5 days) were comparable between the two groups. In each of these studies, immunosuppression consisted of tacrolimus, mycophenolate mofetil, and steroids without antilymphocyte induction. The rates of acute rejection ranged from 21 to 35 per cent, and the rate of relaparotomy was 25 per cent.

Cost-effectiveness analyses In his report to HCFA, Holohan was unable to complete a formal cost-effectiveness analysis for SPK because reliable cost data were not available and only isolated information on charges and payments were accessible [51]. Therefore, a model was constructed to assess the relative cost-effectiveness of SPK over a range of quality of life estimates by taking into account quality-adjusted life years. Within the limitations of this model, Holohan concluded that SPK may be comparable to KTA on the basis of cost-effectiveness only when provided to complicated diabetic patients who have incurred large annual expenses for care secondary to hyperlability, and when a significant improvement in quality of life due to the pancreas transplant has been clearly demonstrated. The treatment choices for a patient with uraemia and IDDM are the following: treatment with chronic dialysis while continuing to receive insulin; a living or cadaver donor KTA without a subse-

R.J. STRATTA

quent pancreas transplant; sequential kidney and pancreas transplant from living and/or cadaveric donors; or an SPK from a cadaver or living donor. Many patients with uraemia and diabetes who opt for a kidney transplant are also interested in a pancreas transplant so that they may be insulin independent as well as dialysis free. The most common methods to achieve this goal are a cadaver donor SPK or a living donor KTA followed by a cadaver donor PAK. Douzdjian et al. applied clinical decision analysis to examine the cost-effectiveness of dialysis, cadaver donor KTA, living donor KTA, and SPK [19]. The analysis was based on a 5-year model, and the measures of outcome used were costs and cost adjusted for quality of life. Cost figures for dialysis ($45 205 per year) and KTA ($108 983 in the first year) were based on 1990 Medicare expenditures. Median charges for the first year after SPK ($185 271) were calculated to be 1.7 times higher than those for KTA according to the National Cooperative Transplantation Study. Diabetes management ($13 509 per year) and transplant follow-up ($9447 per year) were also included in the model. The measure of preference for quality of life was obtained by analysing 17 SPK recipients transplanted between January 1992 and June 1996 at the Medical University of South Carolina. Utilities were weighted as follows: dialysis-free/insulin-independent = 1.0; dialysis-free/insulin-dependent = 0.5; dialysis-dependent/insulin-dependent = 0.4; and death = 0. The expected 5-year costs for the treatment strategies of dialysis, cadaver donor KTA, and living donor KTA were $216 068, $214 678, and $210 872, respectively, whereas the costs for SPK were highest at $241 207. However, the expected cost per quality-adjusted year for each of the treatment strategies was highest for dialysis ($317 746) intermediate for cadaver donor KTA ($156 042) and for living donor KTA ($123 923), and lowest for SPK ($102 422). SPK remained the most cost-effective strategy after varying survival probabilities, costs, and utilities over plausible ranges by means of one-way sensitivity analysis. In a follow-up study, Douzdjian et al. performed a cost–utility analysis of living donor KTA followed by sequential PAK versus SPK [52]. Similar assumptions for costs and utilities were made in the model, and the costs of PAK ($130 780 in the first year) were estimated to be 1.2 times higher than those for KTA as derived from previous reports. Also included in this analysis were the costs associated with waiting for SPK and PAK based on median waiting times obtained from the 1996 UNOS Annual Report. The expected 5-year costs were $277 638 for living donor KTA followed by PAK versus $288 466 for SPK. When adjusted for utilities, living donor KTA followed by PAK cost $153 911, which was less cost effective than SPK at $110 828 per quality-adjusted year. Two-way sensitivity analysis showed that in order for living donor KTA followed by PAK to be at least as cost-effective as SPK, 5-year patient and pancreas graft survival rates following PAK would need to surpass 86 and 80 per cent, respectively. Based on these 5-year decision analysis models, the authors concluded that SPK is the most cost-effective treatment for the uraemic diabetic patient. In 2000, Kiberd and Larson performed a theoretical analysis by constructing a Markov model to compare outcomes for patients with type 1 diabetes mellitus and early overt nephropathy assigned to either standard insulin therapy or PA transplantation [53]. Probabilities for the development of ESRD, blindness, mortality, and direct health-care costs were taken from the literature. Utility scores for the relevant health states were determined by the standard Gamble method on 16 type 1 diabetic patients suitable for PA transplantation. Assuming a baseline graft life expectancy for the pancreas of 10 years, early pancreas transplantation could provide 0.42 more life years and 2.2 more qualityadjusted life years (discounted at 3 per cent) to patients above standard insulin therapy. The incremental costs (charges) for early pancreas transplantation over standard therapy were estimated to be modestly high (about $56 600/quality-adjusted life year for the baseline case). Pancreas transplant costs were also a very sensitive parameter in the cost–utility analysis. Costs per life year gained were in excess of $300 000. The authors concluded that early PA transplantation is an attractive option and

285

286

ECONOMICS OF PANCREAS TRANPLANTATION

may well be at the stage to consider a trial in selected type 1 diabetic patients at risk for renal and retinal disease. In 2001, Whiting et al. merged Medicare claims information with UNOS and USRDS Registry Data to identify all American cadaveric donor renal transplants between 1991 and 1997 [54]. Clinical and economic data for 75 per cent of all transplants were available through this database, including 3342 SPK, 5178 diabetic KTA, and 29 009 non-diabetic KTA recipients. A 5 per cent discount rate was used, monies were adjusted to 1998 dollars, and a $25 000 organ acquisition charge was added to all payments. The 1-, 3-, and 5-year total Medicare charges for SPK were $189 019, $263 108, and $310 109, respectively. The corresponding numbers for diabetic KTA were $145 421, $216 329, and $276 827, respectively. For non-diabetic KTA, the corresponding charges were $134 771, $184 804, and $229 400, respectively. The 5-year total Medicare payments were $125 947 for SPK, $125 757 for diabetic KTA, and $107 643 for non-diabetic KTA (P = NS). Sensitivity analysis conducted by varying the discount rate, pass-through payment, or conducting the analysis on only transplants performed since 1995 had no qualitative effect on the results. The authors concluded that SPK is initially more expensive to Medicare than KTA. However, by 5 years post-transplant, the costs to Medicare is equivalent between SPK and diabetic KTA although both remain more expensive than non-diabetic KTA. At 5 years, charges for SPK remain higher than for KTA.

Future prospects At present, most pancreas transplant centres are reporting average charges ranging from $80 000 to $120 000 for SPK, which includes hospital, professional, and organ acquisition charges for the initial hospital admission. In order to control costs, transplant centres have become more selective in donors and recipients chosen for transplantation, more aggressive in reducing inpatient stays by implementing critical pathways and rapidly shifting to outpatient therapies, and more aware of using the most cost-effective treatment modalities. The health-care system continues to evolve and be redefined towards end points such as efficacy, cost-containment, equitable allocation of resources, and satisfied health-care customers. Because health-care reform has become an economic-based reality, health-care providers are increasingly being asked to factor economics into their clinical decision-making process. Due to its resource-intensive and expensive nature, transplantation has come under great scrutiny by managed care providers. With the implementation of critical pathways, mean length of stay has decreased between 8 and 12 days [32,49]. With improvements in surgical techniques, the relaparotomy rate has decreased form from 30 to 35 per cent to 15 to 20 per cent [55]. The evolution from bladder to enteric drainage not only has decreased readmissions, but also has reduced the risk of infection and eliminated the need for enteric conversion [41–45,47,50]. With advances in immunosuppression, the incidence of rejection after SPK has decreased from between 50 to 80 per cent to 15 to 30 per cent, resulting in lower utilization of expensive antilymphocyte agents either for induction or rejection therapy [48,49,56]. The aggregate effect of these developments has been lower overall morbidity and hospital admission after SPK leading to reductions in resource utilization and overall costs. Currently, the average annual costs of immunosuppressant medications and post-transplant monitoring after SPK are approximately $15 000 and $10 000, respectively. Therefore, total 5-year costs after SPK are estimated to range from $205 000 to $245 000, which represents a cost savings compared to previous analyses. In the last decade, overwhelming clinical evidence has accumulated to support the contention that pancreas transplantation is no longer investigational and indeed is regarded as the treatment of choice for uraemic diabetic patients. The results of pancreas transplantation are at least equivalent if not

R.J. STRATTA

superior to other Medicare-funded solid organ transplants. Recent data suggests that SPK may not only be life-enhancing but life-saving compared to alternative treatment options [57,58]. With the current tide of financial data documenting the cost-effectiveness of SPK and PAK compared to alternative treatment options, it is not surprising that pancreas transplantation will become an increasingly important treatment option in the management of diabetic patients with or without renal failure.

Acknowledgement The assistance and expertise of Jo Lariviere in the preparation of this manuscript is gratefully acknowledged.

References 1 American Diabetes Association. Economic consequences of diabetes mellitus in the US in 1997. Diabetes Care 1998;28:296–309. 2 American Diabetes Association. Diabetic nephropathy. Diabetes Care 1998;21(Suppl. 1):S50–3. 3 US Renal Data System. USRDS 2000 Annual Data Report: incidence and prevalence of ESRD. Am J Kidney Dis 2000;36(Suppl. 2):S37–54. 4 Eggers PW. Comparison of treatment costs between dialysis and transplantation. Semin Nephrol 1992;12:284. 5 Garner TI, Dardis R. Cost-effectiveness analysis of end-stage renal disease treatments. Med Care 1987;25:25. 6 Karlberg I. Cost analysis of alternative treatments in end-stage renal disease. Transplant Proc 1992;24:335. 7 Port FK, Wolfe RA, Mauger EA, et al. Comparison of survival probabilities for dialysis patients versus cadaveric renal transplant recipients. JAMA 1993;270:1339–43. 8 Laupacis A, Keown P, Pus N, et al. A study of the quality of life and cost–utility of renal transplantation. Kidney Int 1996;50:235. 9 Evans RW. Organ transplantation and the inevitable debate as to what constitutes a basic health care benefit. In Terasaki PI, Cecka JM, ed. Clinical transplants 1993. Los Angeles: UCLA Tissue Typing Laboratory, 1994:359–91. 10 Roberts ST, Maxwell DR, Gross TL. Cost-effectiveness care of end-stage renal disease: a billion-dollar question. Ann Intern Med 1980;92:243–8. 11 Robinson R. Cost–utility analysis. Br Med J 1993;307:924–6. 12 USRDS 2000 Annual Data Report. Economic costs of ESRD. Am J Kidney Dis 2000;36(Suppl. 2):S163–76. 13 Eggers PW, Kucken LE. Cost issues in transplantation. Surg Clin N Am 1994;74:1259–67. 14 Schweitzer EJ, Wiland A, Evans D, et al. The shrinking renal replacement therapy ‘break-even’ point. Transplantation 1998;66:1702–8. 15 Javor KA, Kotsanos JG, McDonald RC, et al. Diabetic ketoacidosis charges relative to medical charges of adult patients with type I diabetes. Diabetes Care 1997;20(3):349–54. 16 The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–86. 17 The Diabetes Control and Complications Trial Research Group. Resource utilization and costs of care in the diabetes control and complications trial. Diabetes Care 1995;18:1468–78.

287

288

ECONOMICS OF PANCREAS TRANPLANTATION

18 The Diabetes Control and Complications Trial Research Group. Lifetime benefits and costs of intensive therapy as practiced in the diabetes control and complications trial. JAMA 1996;276:1409–15. 19 Douzdjian V, Ferrara D, Silvestri G. Treatment strategies for insulin-dependent diabetics with ESRD: a cost-effectiveness decision analysis model. Am J Kidney Dis 1998;31:794–802. 20 Klonoff DC, Schwartz DM. An economic analysis of interventions for diabetes. Diabetes Care 2000;23:390–404. 21 Pastan S, Bailey J. Dialysis Therapy. N Engl J Med 1998;338:1428–37. 22 Eggers PW. Effect of transplantation on the Medicare end-stage renal disease program. N Engl J Med 1988;318:223. 23 Eggers PW. Cost-effectiveness of kidney transplantation. Presented at the Immunosuppression Conference in Organ Transplantation: patient access to long-term care. Philadelphia: 4 December 1998. 24 Smith CR, Woodward RS, Cohen DS, et al. Cadaveric versus living donor kidney transplantation: A Medicare payment analysis. Transplantation 2000;69:311–14. 25 Evans RW, Manninen DL, Dong FB. An economic analysis of pancreas transplantation: costs, insurance coverage, and reimbursement. Clin Transplant 1993;7:166–74. 26 Robertson RP. Pancreatic and islet transplantation for diabetes: cures or curiosities? N Engl J Med 1992;327:861–8. 27 Sollinger HW, Ploeg RJ, Eckhoff DE, et al. Two hundred consecutive simultaneous pancreas–kidney transplants with bladder drainage. Surgery 1993;114:736–44. 28 Jones JW, Mizrahi SS, Bentley FR. Success and complications of pancreatic transplantation at one institution. Ann Surg 1996;223:757–64. 29 Holohan TV. Simultaneous pancreas–kidney and sequential pancreas after kidney transplantation. Report to the Office of Health Technology Assessment, 1995. 30 Magee JC, Bromberg JS, Punch JD, et al. Long-term resource utilization and economic outcomes analysis after bladder-drainage pancreas-kidney transplantation. Transplantation 1998;65(12):S132(A215). 31 Stratta RJ, Cushing KA, Frisbie K, et al. Analysis of hospital charges after simultaneous pancreaskidney transplantation in the era of managed care. Transplantation 1997;64:287–92. 32 Cushing KA, Stratta RJ. Design, development and implementation of a critical pathway in simultaneous pancreas-kidney transplant recipients. J Transplant Coord 1997;7:164–72. 33 Stratta RJ. Outcome analysis of hospital charges after simultaneous kidney–pancreas transplantation: influence of outliers on resource utilization. Transplant Proc 1998;30:261. 34 Stratta RJ, Taylor RJ, Sindhi R, et al. Analysis of early readmissions after combined pancreas–kidney transplantation. Am J Kidney Dis 1996;28:867–77. 35 Reddy KS, Johnston TD, Karounas D, Ranjan D. Hospital charges following simultaneous kidney–pancreas transplantation: Enteric drainage versus bladder drainage. Clin Transplant 2000;14:375–9. 36 Gruessner AC, Troppmann C, Sutherland DER, et al. Donor and recipient risk factors significantly affect cost of pancreas transplants. Transplant Proc 1997;29:656–7. 37 Troppmann C, Gruessner AC, Dunn DL, et al. Surgical complications requiring early relaparotomy after pancreas transplantation: a multivariate risk factor and economic impact analysis of the cyclosporine era. Ann Surg 1998;227:255–68. 38 Humar A, Kandaswamy R, Drangstveit MB, et al. Prolonged preservation increases surgical complications after pancreas transplants. Surgery 2000;127:545–51. 39 Stratta RJ, Weide LG, Sindhi R, et al. Solitary pancreas transplantation: experience with 62 consecutive cases. Diabetes Care 1997;20:362–8.

R.J. STRATTA

40 Stratta RJ. Sequential pancreas after kidney transplantation: is anti-lymphocyte induction therapy needed? Transplant Proc 1998;30:1549–51. 41 Gruessner AC, Sutherland DER. Analysis of United States (US) and non-US pancreas transplants as reported to the International Pancreas Transplant Registry (IPTR) and to the United Network for Organ Sharing (UNOS). In: Cecka JM, Terasaki PI, ed. Clinical transplants 1998. Los Angeles: UCLA Tissue-Typing Laboratory, 1999:53–71. 42 Sollinger HW, Odorico JS, Knechtle SJ, et al. Experience with 500 simultaneous pancreas–kidney transplants. Ann Surg 1998;228:284–96. 43 Pirsch JD, Odorico JS, D’Allessandro AM, et al. Post-transplant infection in enteric versus bladderdrained simultaneous pancreas–kidney transplant recipients. Transplantation 1998;66:1746–50. 44 Kuo PC, Johnson LB, Schweitzer EJ, et al. Simultaneous kidney–pancreas transplantation: a comparison of enteric and bladder drainage of exocrine pancreatic secretions. Transplantation 1997;63:238–43. 45 Douzdjian V, Rajagopalan PR. Primary enteric drainage of the pancreas allograft revisited. J Am Coll Surg 1997;185:471–5. 46 Douzdjian V, Lanza KT, Uber L, et al. The effectiveness of a transplant outpatient unit as a costreducing strategy following pancreas transplantation. Transplant Proc 1998;30:272. 47 Stratta RJ, Gaber AO, Shokouh-Amiri MH, et al. A prospective a comparison of systemic-bladder versus portal-enteric drainage in vascularized pancreas transplantation. Surgery 2000;127:217–26. 48 Stratta RJ, Gaber AO, Shokouh-Amiri MH, et al. Evolution in pancreas transplantation techniques: simultaneous kidney–pancreas transplantation using portal-enteric drainage without anti-lymphocyte induction. Ann Surg 1999;229:701–12. 49 Reddy KS, Stratta RJ, Shokouh-Amiri MH, et al. Simultaneous kidney–pancreas transplantation without anti-lymphocyte induction. Transplantation 2000;69:45–54. 50 Stratta RJ, Shokouh-Amiri MH, Egidi MF, et al. A prospective comparison of simultaneous kidney–pancreas transplantation with systemic-enteric versus portal-enteric drainage. Ann Surg 2001;233:740–51. 51 Holohan TV. Cost-effectiveness modeling of simultaneous pancreas–kidney transplantation. Int J Tech Assess Health Care 1996;12(3):416–24. 52 Douzdjian V, Escobar F, Kupin WL, et al. Cost–utility analysis of living-donor kidney transplantation followed by pancreas transplantation versus simultaneous pancreas–kidney transplantation. Clin Transplant 1999;13:51–8. 53 Kiberd BA, Larson T. Estimating the benefits of solitary pancreas transplantation in non-uraemic patients with type 1 diabetes mellitus: A theoretical analysis. Transplantation 200;70:1121–7. 54 Whiting JF, Martin JE, Cohen DS, et al. Economic outcomes of simultaneous kidney–pancreas transplantation as compared to kidney alone. Transplantation 2001 (in press). 55 Humar A, Kandaswamy R, Granger D, et al. Decreased surgical risks of pancreas transplantation in the modern era. Ann Surg 2000;231:269–75. 56 Stratta RJ. Review of immunosuppressive usage in pancreas transplantation. Clin Transplant 1999,13:1–12. 57 Becker BN, Pintar TJ, Becker YT, et al. Increasing expected lifespan: the impact of kidney–pancreas transplantation. Transplantation 1999;67(7):S7(A4). 58 Tyden G, Bolinder J, Solders G, et al. Improved survival in patients with insulin-dependent diabetes mellitus and end-stage diabetic nephropathy ten years after combined pancreas and kidney transplantation. Transplantation 1999;67:645–8.

289

This page intentionally left blank

Chapter 21

Gene therapy for diabetes Muralidhar Karanam, Z. Song, and Rahul M. Jindal

Historical aspects of gene therapy in relationship to diabetes During the mid-1980s, gene therapy was investigated as a possible therapeutic regimen for diseases caused by single gene defects, for example haemophilia, Duchenne’s muscular dystrophy, and sickle cell anaemia. In the late 1980s and early 1990s, the concept of gene therapy expanded into a number of acquired diseases. Gene therapy can be targeted to somatic (body) or germ (egg and sperm) cells. In somatic gene therapy the recipient genome is changed, but the change is not passed along to the next generation. This form of gene therapy is contrasted with germline gene therapy where the goal is to pass change on to the offspring [1]. The aetiology of type I diabetes is under investigation and a unified hypothesis has not yet emerged. In the non-obese diabetic (NOD) mouse model, loci on at least 15 chromosomes have been implicated in determining susceptibility to diabetes [2]. At present, no single mutant type I diabetes gene is known. Due to these factors, it is currently not feasible to treat diabetes by merely replacing a defective gene with a normal allele. Genotyping of the DQ and DR class II major histocompatibility complex (MHC) genes has some predictive value as certain alleles at these loci are highly associated with type I diabetes [3–5]. Germline gene therapy is not being actively investigated in larger animals and humans. Somatic gene therapy for diabetes can be broadly defined as any therapeutic modality that uses gene transfer technology to improve the clinical status of a patient with diabetes [6]. Gene therapy for diabetes can be divided into four major approaches as follows (Table 20.1).

Expansion of ␤-cells or ␤-cell precursors An increasing number of genes that are involved in the process of ␤-cell growth and differentiation are being discovered. Induction of differentiation in precursor cells is an attractive, although difficult approach for ␤-cell expansion [6, 7].

Engineering of glucose-responsive insulin secretion Many different methods of gene transfer, including lipofection, electroporation, biolistic projectile targeting and viral vector-mediated gene transfer have been employed in these studies [8–11].

Altering peripheral insulin resistance in type II diabetes As the major target for insulin action is the muscle, effective gene therapy for endorgan insulin unresponsiveness might require transduction of a high percentage of muscle cells [12]. Manipulation of pathways, which are involved in the regulation of adiposity such as those involving the hormone

292

GENE THERAPY FOR DIABETES

Table 21.1 Summary of gene therapy approaches to diabetes (from [52]) Therapeutic intervention

Prevention

Direct treatment

Symptomatic treatment

Engineering of immune X system cells or ␤-cells to prevent autoimmune ␤-cell destruction

␤-cell modification to prevent apoptotic destruction

Relevant to type II diabetes

X

X

X

X

Enhancement of ␤-cell neogenesis and/or regeneration

X

X

X

Genetic engineering of glucose-responsive insulin secretion into non-␤-cells

X

X

X

Genetic engineering of allogeneic or xenogeneic ␤-cells

X

X

X

Modification of end organs to improve insulin sensitivity Engineering of end organs to reduce complications, e.g. antiangiogenesis strategies for neovascularization

X

Relevant to type I diabetes

X

X

X X

X

X

leptin, holds particular promise because it is likely that a reduction in adiposity will result in dramatic improvement in insulin resistance found in type II diabetes. A reduction in adiposity reduces the toxic effects of fatty acids on ␤-cells in type II diabetes [13], high levels of free fatty acids produced by insulin-unresponsive adipocytes have been shown to trigger ␤-cell apoptosis [14]. It might also be possible to prevent ␤-cell loss by inhibiting apoptosis or by manipulating the leptin pathway [15].

Immune modulation Immune modulation is used to prevent autoimmune destruction of pancreatic ␤-cells during early stages of type I diabetes and to protect islet grafts from autoimmune attack. Genetic engineering of ␤-cells or immune cells to circumvent allograft rejection of islets have had encouraging results in recent years [16–25]. Gene therapy should not be confused with cloning, which has been the subject of much media attention recently. Cloning is defined as creating an individual with essentially the same genetic makeup and is very different from gene therapy.

VECTORS USED FOR THE GENE THERAPY OF DIABETES The basic challenge in gene therapy is to develop methods for delivering genetic material to the appropriate cells in a way that is specific, efficient, and safe. For this purpose, gene delivery vehicles (vectors) are used to introduce therapeutic genes into targeted cells, in particular, to the nucleus of the cell. If genes are appropriately delivered, they can persist for the life of the cell and potentially lead to a cure. In principle, gene transfer can be carried out in vivo, by introducing the gene of interest into the patient using a viral vector or a non-viral agent or by ex-vivo modification of cells explanted from the patient. These manipulated cells can subsequently be injected back using an appropriate site into the patient.

M. KARANAM ET Al.

For a vector to be of use several inherent characteristics are required. 1. Ability to carry the intact gene (a genome or cDNA) in a stable form without disrupting or mutating other endogenous genes. 2. Ability to introduce the gene into a specific cell or organ of interest. 3. Ability to express the gene of interest in sufficiently high levels in a regulated manner over sufficient length of time to be clinically useful. 4. The gene product, as well as proteins derived from the vectors should not elicit an adverse immune response or toxicity in the recipient, which may over time diminish the expression of introduced gene. To characterize the ability to infect human pancreatic endocrine cells under several conditions, nonviral techniques, including lipofection, electroporation, biolistic transfection, ballistic gene delivery (gene gun), calcium phosphate coprecipitation, mono- and polycationic liposomes, and lipofectamine have been used to transfer genes into primary pancreatic islets [6]. All these methods, however, have resulted in low integration efficiency and transient gene expression. Several animal viruses have been tested as potential vectors, but none has proven to have all the desired properties. Retroviruses and RNA containing viruses are difficult to propagate in sufficient titres, they do not integrate into non-dividing cells [26], and are of concern because of their oncogenic properties in some hosts. Furthermore, these viruses are not site-specific; genes introduced by retroviral vectors are frequently expressed for relatively short periods of time. Another virus used as a vector in various model systems has been the human adenovirus. Successful adenovirus-mediated gene transfer to rodent islets has been described in the literature [27–29]. The drawbacks of adenoviral vectors are the lack of integration into the host cell genome and the potent immune response directed against adenoviral structural proteins, leading to transient expression of the transgenes [30]. A third virus of interest as a possible vector is adeno-associated virus (AAV), a single-stranded DNA-containing human parvovirus [31,32]. It has a number of advantages over other viral vectors: between 80 and 90 per cent of the human population has been exposed to AAV and no symptoms or pathology have been attributed to this infection [33]. It can be constructed in such a manner that it will not express protein on a cell surface and is able to establish a stable latent infection with high frequency [34] where the viral genome has been shown to integrate into the chromosomal DNA in a sitespecific manner [35]. In non-human primate studies, these vectors have been reported to cause T-cell lymphomas [36]. Conversely, parvoviruses have not been known to cause malignant disease [31,32,37]. In our laboratory the efficacy of rAAV vectors was investigated to introduce the insulin gene into malignant and non-malignant cell lines [31], as was the transduction of haematopoietic stem cells with the insulin gene using rAAV in vitro [32]. When injected in vivo, we could reverse streptozotocin (STZ)-induced diabetes in the rat for the short term and normal animals remained euglycaemic despite the identification of the rat insulin gene in various tissues [31,32]. Another class of vectors derived more recently from lentiviruses such as those belonging to the human immunodeficiency virus (HIV) family, can infect both dividing and non-dividing cells, including pancreatic ␤-cells [38–41]. Human immunodeficiency viruses belong to a subclass of retroviruses known as lentivirus. These viruses have several advantages over Moloney murine leukaemia retroviral vectors. Lentivirus vectors are able to transduce non-dividing cells, as well as those that are actively dividing, thereby considerably broadening their usefulness as gene transfer vehicles. As lentiviruses are able to integrate their genetic material into the genome of the host cell, they have the potential to result in the long-term, stable gene expression of transgenes [42]. Since lentiviruses have

293

294

GENE THERAPY FOR DIABETES

inherent tropism for CD4+ T cells, macrophages, and haematopoietic stem cells, the prospects as vectors for immunological purposes are exciting. Genetic modifications, such as the introduction of vesticular somatitis virus G protein into the lentiviral envelope have widened the tropism of this vector for potential clinical application. Gallichan et al. [19] demonstrated that the HIV-based lentiviral vector was capable of stably transducing immunoregulatory genes into whole islets. Grafts containing insulin-positive ␤-islet cells expressing foreign protein (immunoregulatory molecules) indicated that transduction did not interfere with glucose regulation. When islets transduced with an HIV vector expressing interleukin (IL)-4 were transplanted into diabetes-prone mice, animals were protected from autoimmune insulitis and islet destruction for 12 weeks. The absence of inflammatory infiltrates in grafts suggested that transduction did not activate the immune system. This study showed that HIV-based lentiviral vectors could effectively transduce islet grafts and, through expression of IL-4, increase the graft survival by preventing insulitis and promoting the development of non-pathogenic autoimmune T cells. However, the potential of lentivirus vector-mediated transduction of islets by IL-4, other cytokines, or immune modulation to block or deviate a mature diabetic immune system has yet to be determined. Muller et al. [43] showed that IL-4 expressed transgenically in islets failed to protect allografts from rejection. Although this development is exciting, in practice, the use of an HIV-based vector for therapies targeting diseases other than HIV might be extremely difficult to introduce clinically and ethically, until it has been established that these vectors are safe to use. Many of the viruses that are used as vectors lack replicating genes, and therefore, cannot replicate in normal cells. The recombinant virus with its transgene must be grown up to higher titres in a packaging cell line. This is a cell line that contains all of the complementary genes that the virus requires to replicate. The recombinant viral particles can then be purified as live infectious virus from the packaging cell line and used to transduce cells or tissues in vivo and ex vivo. Gene transfer ex vivo has the advantage of expressing only a small number of the patient’s cells to genetic manipulation, requires lower viral titres for infection, and allows careful observation of cell function and genetic susceptibility before reintroduction into the patient. Conversely, this approach is limited to those cell types that can be easily explanted. For example, to obtain hepatocytes, patients have to be subjected to partial hepatectomy, a major surgical procedure. Leibowtiz et al. [9] studied the factors that influence the efficiency of infection of human fetal and adult pancreatic cells. Adenovirus appeared to be the most potent vector for ex vivo expression of foreign genes in adult endocrine pancreatic cells and may be the ideal vector for application where high level but transient expression is desired. Under optimal conditions, infection with murine retroviral and lentiviral vectors may be reasonably efficient and stable but only lentiviral vectors efficiently infect pancreatic ␤-cells. Preliminary studies in our laboratory have shown that AAV vectors can be used to transfer preproinsulin gene into haematopoietic stem cells. Further work is in progress to understand the mechanism of glucose regulation in the infected stem cells.

Current status of gene therapy for diabetes Prevention of type I diabetes Prevention of islet damage is obviously the best approach for treatment of type I diabetes. However, this depends upon the development of highly reliable techniques for diagnosis of prediabetic people. At present, no single mutant type I diabetes gene is known whose replacement with a normal allele through gene therapy would reverse the course of the disease. It has been shown that certain alleles of the class II MHC are strongly associated with the development of type I diabetes [44]. In addition,

M. KARANAM ET Al.

numerous environmental factors are also likely to play an important role in the pathogenesis of diabetes [44]. Since it is difficult to identify with certainty patients with type I diabetes, in vivo gene therapy would not be feasible in the near future. In type I diabetes, cytokines produced by T cells, such as tumour necrosis factor ␣ (TNF-␣), interferon (IFN-␥), and particularly IL-1 ␤, can trigger ␤-cell apoptosis through a nitric oxide dependent process [45]. In addition to chronic immunosuppression, a number of agents have been used to prevent damage to islets, such as nicotinomide to promote ␤-cell regeneration/neogenesis and oral and subcutaneous insulin therapy to induce tolerance [46,47]. However, none of these approaches were shown to be clinically effective. Gene therapy directed at the autoimmune response could involve genetic modification of the ␤-cell or of the immune cells [20]. However, it is difficult to make efficient genetic modification of immune cells because the immune system has an immense number of target cells. Thus, selective modification of islet cells is probably a more attractive approach. Cytokine induced ␤ -cell apoptosis is mediated through raised expression of inducible nitric oxide synthesis (iNOS) and an intracellular increase in nitric oxide (NO) free radicals [48,49]. Recent studies have shown that overexpression of antiapoptosis genes, bcl-2, by gene delivery into prediabetic islet in vivo could increase ␤-cell resistance to T-helper type I (Th1) cytokines [50,51]. Modification of islet cells to secrete molecules that downregulate the immune response locally is also promising. Recent studies have shown that expression of the immunomodulatory molecule IL-4, downregulation of the expression of MHC, blockade of the costimulation pathway of T cells with anti-CD40L and CTLA4 immunoglobulin could prolong islet graft survival in animal models [23,24]. However, therapies that prevent diabetes in rodent models have not been efficacious in humans.

Gene therapies specific to type II diabetes In type II diabetes, there are defects both in insulin action and ␤-cell function. To deal with the problem of end-organ unresponsiveness, the exact nature of the defect must be understood in order to find specific sites which could be targeted for gene transfer studies. However, a definitive link between the molecular and population genetics of type II diabetes has yet to be made [52]. As muscle and liver cells are the major target for insulin action, effective gene therapy for endorgan insulin unresponsiveness might require stable transduction of a rather high percentage of muscle and hepatic cells. Although the insulin gene could be delivered into muscle and hepatic cells by various gene transfer means, this problem has not yet been solved [12]. A recent study has shown that fatty acid can induce ␤-cell apoptosis suggesting a link between obesity and diabetes; reduction in adioposity and ␤-cell apoptosis with gene therapy might be effective [14].

Gene therapies potentially applicable to both type I and II diabetes Type I and II diabetes share the characteristics of ␤-cell dysfunction and failure. Inhibition of apoptosis, promotion of ␤-cell regeneration, ␤-cell replacement, engineering of non ␤-cells, and engineering of ␤-cells are the major strategies for gene therapy studies.

Inhibition of apoptosis It has recently been suggested that apoptosis is the final common pathway by which islet cells are destroyed in both type I and II diabetes [53,54]. As mentioned above, the strategies to prevent apoptosis of ␤-cells holds promise for clinical application. One reason for the poor outcome of islet cell

295

296

GENE THERAPY FOR DIABETES

transplantation in humans seems to be that a high percentage of islets are destroyed in the immediate post-transplant period. Apoptosis and incompatibility between human blood and isolated islets may be important; in vitro transfer of genes into allogeneic islets before transplantation to protect them from apoptosis could be beneficial [55].

Promotion of ␤-cell regeneration The pancreatic ductal epithelial cells — the putative precursor islet cells — are not damaged by the diabetic process. These cells may potentially be targeted for regeneration with an appropriate signal. Several molecules, such as islet neogenesis-associated protein (INGAP), pancreatic duodenum homeobox 1 (PDX-1), most notably members of the reg gene family, have been proposed to play roles in stimulating ␤-cell regeneration [56–59]. These genes may theoretically be used to promote the conversion of precursor cells into islets.

␤-cell replacement strategies Transplantation of intact human islets would be ideal. A major problem with this approach is the lack of donor material and the need for chronic immunosuppression. Xenotransplantation could overcome the shortage of donor organs but there are still considerable difficulties in isolating and purifying sufficient islets and inhibiting xenograft rejection. Genetically modified animals, such as transgenic pig expressing human genes, bred in recent years, could offer a better source of islet, however, this approach is still at a preclinical stage. Genetic manipulation for tolerance induction and encapsulation of allogeneic or xenogeneic islets in vitro before the transplantation may also be effective approaches for successful ␤-cell replacement in human diabetes [60].

Engineering of ␤-cells Possible methods for generating immortalized ␤-cell lines include irradiation or the expression of oncogenes, particularly SV40 large T antigen, are readily available. The problem with this approach has been that these cells have a tendency to lose their glucose responsiveness after short period in culture [61]. This is probably due to dedifferentiation that results in a decrease in the level of the essential glucose-sensing proteins, GLUT2 and glucokinase, as well as diminished insulin gene expression [62]. A cell line was isolated and immortalized from a patient with persistent hyperinsulinaemia of infancy in which islet stem cells released insulin in the absence of glucose. Recently, these cells (named NES2Y) were triple-transfected with cDNAs encoding the two components of the K(ATP) channel (SUR1 and Kir6.2) and PDX-1. One selected clonal cell line (NISK9) had normal K(ATP) channel activity to give near normal insulin secretory responses to glucose suggesting that this approach might provide the solution to the problem of developing an unlimited source of human pancreatic ␤-cells [63].

Engineering of non-␤-cells A potential advantage of using non-␤-cells is that these cells may not be recognized and destroyed by the autoimmune response generally seen in patients with type I diabetes. Furthermore, these may be readily taken from the patient eliminating the complications of graft rejection [64]. However, ␤-cells are unique not only in expressing the insulin gene, but also because they possess additional characteristics which are exquisitely suited for their physiological role. To mimic ␤-cell function successfully, three major obstacles will have to be overcome: proinsulin synthesis, proinsulin process, and mature insulin storage and regulated secretion [65,66]. Proinsulin biosynthesis has been achieved in a variety

M. KARANAM ET Al.

of cell lines, for example those derived from pituitary and liver [67,68]. Engineering cells similar to ␤-cells, such as neuroendocrine cells, to secrete insulin has been tried on the mouse corticotrophic cell line, AtT20. The cells did not secrete insulin in response to glucose following stable transfection with the insulin cDNA, despite the fact that enzymes that process proinsulin to insulin and the correct secretory pathway were present. Furthermore, cells secreted ACTH, which led to insulin resistance, an undesirable property [69]. Non-neuroendocrine cells have the problem that they do not contain the proinsulin processing enzymes, PC2 and PC3, a limitation that may be overcome by transfection with mutant insulin in which the C-peptide cleavage sites are altered to allow cleavage by the ubiquitous endoprotease, furin. This system worked in cultured liver cells, myoblasts, and fibroblasts and would allow the cells to release a constant low level of insulin [70]. More recently, a hepatocyte cell line (HEP G2ins/g) has been developed with the transfection of full-length insulin cDNA and the human islet glucose transporter, GLUT2. These cells exhibit insulin synthesis, storage, and glucose-stimulated insulin secretion properties [70]. A recent report has shown that the transfer of the preproinsulin II gene into rat haematopoietic stem cells by rAAV can be achieved [71]. The major problem to this approach is obtaining the physiological glucose-responsive insulin secretion and long-term gene expression. Development of a promoter that is appropriately activated and repressed by high and low glucose, respectively, has been proposed as an approach to gene therapy for diabetes [72]. But the transcriptional response has a relatively long lag time compared with the extremely rapid response of the regulated secretory pathway, an approach which could lead to hypoglycaemia.

Work done in the authors’ laboratory We have investigated a possible delivery system for the rat preproinsulin II gene (rI2) utilizing a rAAV vector system, with the long-term goal of engineering stably transfected insulin-producing cell lines. The rAAV vector was chosen because it has been shown to be a safe and non-pathogenic method of gene transfer. The plasmid pBC12BI (ATCC) was purified and digested with restriction enzymes Ssp-1 and Stu-1 to release a fragment containing the Rous sarcoma virus long terminal repeat (RSV-LTR) promoter-driven rat preproinsulin II gene (rI2). Subsequently, the RSV-rI2 gene fragment was cloned into the BamH1 site of rAAV vector plasmid pWP-19 to produce the rI2 recombinant plasmid designated pLP-1. PWP-19 also encodes the AAV inverted terminal repeats for integration and replication, and the herpes virus thymidine kinase promoter-driven gene for resistance to neomycin (neor). The cell line 293 (ATCC) was then cotransfected with pLP-1 and helper plasmid pAAV\AD, which is required for viral replication. The rAAV genome, now containing rI2, was rescued using adenovirus and packaged into mature AAV virions termed vLP-1. Finally, human pancreatic adenocarcinoma cells (HPAC, ATCC) were exposed to vLP-1 and selected for G418 resistance. Successful rescue was confirmed by Southern blot analysis using the rI2 gene probe derived from the original plasmid. The final titre of 1.25 × 109 particles/ml was determined by DNA slot blots using pLP-1 as the standard. HPAC cells were infected with vLP-1 (termed HPAC/rI2). Integration of the rI2 genome in G418-resistant clones was confirmed by southern blot analysis. We have developed a rAAV-mediated gene transfer system for the rat preproinsulin gene. In addition, we have successfully transfected HPAC with rI2 utilizing this system [34]. We then investigated the ability of rAAV, to mediate the transfer of rI2 gene into rat haematopoietic stem cells in vitro and expression of rI2 following intravenous injection of infected stem cells into syngeneic rats. Bone marrow from female Wistar-Furth rats was enriched for stem cells by using plastic adherence and negative selection with monoclonal anti-rat CD3 and CD45RA to deplete T and B cells. The remaining cells were exposed to vLP-1 (moi = 50 : 1) for 2 h [34]. Transfection was

297

298

GENE THERAPY FOR DIABETES

Fig. 21.1 Construction of rAAV containing rI2 gene.

M. KARANAM ET Al.

confirmed by polymerase chain reaction (PCR) of neomycin-resistance gene (neoR) after 8 days in culture. For in vivo studies, 10 million exposed stem cells were injected intravenously into syngeneic rats (n = 3). The results represent three identical experiments. Expression of neoR and rI2 was analysed by RT-PCR. At week 1, neoR and rI2 were expressed in liver, spleen, thymus, peripheral blood lymphocytes, and bone marrow. At week 2, neoR was expressed in spleen and brain, while at week 6, thymus, lymph nodes, bone marrow, liver, spleen, and brain expressed neoR. rI2 was not detected after week 1. In summary, we showed that rAAV was efficient for transferring neoR and rI2 into rat haematopoietic stem cells. To study the effect of injecting haematopoietic stem cells containing rI2 via rAAV into normal and STZ-induced diabetic rats, rI2 was transfected into rat haematopoietic stem cells using rAAV vector. Stem cells were injected by intravenous route into normal and STZ-induced diabetic rats to study blood sugar and expression of rI2 in various tissues. The pLP-1 recombinant plasmid containing rI2 (vLP-1) was engineered as previously described [34]. Approximately 10 million exposed stem cells were injected by intravenous route into each animal; there were four groups: (1) normal animals at moi 50 : 1 or (6) moi 100 : 1. Groups 3 (n = 9) were STZ-induced diabetic animals at moi 100 : 1. Animals that showed reversal of diabetes from group 3 were sacrificed for the study of gene expression at weeks 1, 2, and 6 respectively. Control diabetic animals did not receive stem cells or virus constituted group 4. Expression of rI2 was analysed by RT-PCR and Southern analyses. Despite introduction of insulin gene, groups 1 and 2 had blood sugar that remained within normal levels, while three of nine animals in group 3 showed reversal of diabetes; using RT-PCR, group 1 expressed rI2 in the liver, spleen, thymus, brain, and heart at week 1 only. In group 2, rI2 was seen in the thymus up to 6 weeks; in diabetic animals (group 3) rI2 was seen in the liver, bone marrow, spleen, thymus, and peripheral blood lymphocytes at week 2 and in the thymus and lymphocytes at week 6. We have shown that (a) rAAV is a useful vector for transferring rI2 into rat haematopoietic stem cells; (b) normal animals remained euglycaemic after injection of stem cells containing rI2 despite identification in various tissues suggesting autoregulation; and (c) short-term reversal of diabetes was achieved in some animals by injection of stem cells containing rI2. In another ongoing experiment, we used the NOD mouse model; an experimental group (n = 10) of animals were intramuscularly injected with 107 RAAV virions containing the insulin gene and compared to a mock-injected control group (n = 10). Blood glucose was then measured weekly for 16 weeks. Data showed that the experimental group contained 70 per cent euglycaemic animals (defined as glucose of less than 200 mg/dl) versus 10 per cent of the control animals (P < 0.05) at 14 weeks. Mean weight in the treated group was greater than the untreated group. Insulin mRNA was detected at the injection site of all of the treated animals, but not controls. Complete destruction of all pancreata was confirmed by histology ruling out the possibility of spontaneous reversal of insulinitis. We concluded that intramuscular delivery of the insulin gene in the NOD mouse was able to prevent clinical diabetes up to 14 weeks in a majority of treated animals.

Conclusion Genetic modification of islets in vitro prior to transplantation has the potential to improve survival and function. One approach would be the generation of ␤-cell lines. Use of rodent ␤-cell lines in diabetic animals has shown that pure ␤-cells can function for long periods of time [73]. These cell lines could potentially provide a stable reproducible source of cells and they could be modified by gene therapy in culture. The use of oncogenes to transform ␤-cells, both rodent [74,75] and human [76,77]

299

300

GENE THERAPY FOR DIABETES

has been disappointing so far due to phenotypic instability and dedifferentiation. Transformed ␤-cells tend to lose the capability to perform insulin biosynthesis and regulate secretion, as well as the release of insulin at subphysiological glucose concentrations [78]. More encouraging is the strategy of conditional oncogene expression in ␤-cells relying on a regulatory system for gene expression based on the bacterial tetracycline (tet) operon [73]. Using this a tight and reversible expression of the SV40 T antigen (Tag) oncoprotein in ␤-cells was possible. Transplantation into syngeneic STZ diabetic mice showed them to be functional for months in the growth arrested stage. They were capable of replication once the tet block was removed [73]. Insulin secretion in vivo was regulated by hyperglycaemia as shown by hyperglycaemic clamp studies [79]. Concerns about using transformed cells in humans could be addressed by designing ways to eliminate the oncogenes from cells as with the Cre-loxP DNA recombination approach, or by introducing suicide mechanisms into the engineered cells such as the herpes simplex thymidine kinase gene [78]. In the short term, it is likely that a better outcome in the treatment of diabetes will come from the improvement in conventional therapies. Despite the many obstacles that remain to be overcome, gene therapy has become a realistic approach through the increased understanding of the molecular and biochemical events involved in insulin production, its release and subsequent action, as well as by the improvement in gene transfer technology. Very little is known about the factors involved in the physiological renewal of islet cells in the adult. There is evidence for the existence of a small compartment of islet cells which maintain replicative capacity [80]. Hepatocyte growth factor has been shown to induce proliferation in cultured human islets [81]. A variety of growth factors have been shown to induce proliferation in fetal/neonatal islets. Identification of specific factors that induce proliferation and differentiation in this pool of stem cells will generate sufficient number of human islets in culture for transplantation. In addition, inducible gene expression systems encoding for such factors could be targeted to the pancreas/islet cells to overcome the current problems of poor islet yield [78]. The above approaches provide the hope that those alternative methods for generating insulin-secreting tissue in large quantities will one day be available. If this tissue can be made human in source, and immunologically and clinically compatible with individual recipients, then exciting possibilities are obvious. However, for the present the experiments described above are still preliminary and many remain to be done in the development of safe efficient vectors [82]. Recent advances in the isolation and characterization of islet stem cells, and the ability of these cells to reverse diabetes in animal models of diabetes, may provide an abundant islet source for the treatment of patients with diabetes [83]. However, these stem cells are present in ducts of the pancreas and obtaining these in the human would pose considerable problems requiring major surgery. Isolating islet stem cells from cadaver pancreas is always a possibility, however, transplantation of allogeneic islet stem cells would require immunosuppression and we will have to overcome difficulties similar to those encountered with transplantation of human allogeneic islets. Therefore, it would seem that research into various aspects of gene therapy for diabetes, especially using autologous cells, should continue.

References 1 Benedum J. The early history of endocrine cell transplantation. J Mol Med 1999;77:30–5. 2 Wicker LS, Todd JA, Peterson LB. Genetic control of autoimmune diabetes in the NOD mouse. Ann Rev Immunol 1995;13:179–200. 3 Honeyman MC, Harrison LC, Drummand B, Colman PC, Tait BC. Analysis of families at risk for IDDM reveals that HLA antigens influence progression to clinical disease. Mol Med 1995;1:576–82. 4 Tisch R, McDevitt H. Insulin-dependent diabetes mellitus. Cell 1996;85:291–7.

M. KARANAM ET Al.

5 Gottlieb PA, Eiseubarth GS. Diagnosis and treatment of pre-insulin dependent diabetes. Ann Rev Med 1998;49:391–405. 6 Levine F. Gene therapy for diabetes: Strategies for beta-cell modification and replacement. Diabetes Metabol Rev 1997;13:209–46. 7 Beattie GM, Itkin-Ausari P, Cirulli V, et al. Sustained proliferation of PDX-1 cells derived from human islets. Diabetes 1999;48:1013–19. 8 Saldeen J, Curiel DT, Eizirik DL, et al. Efficient gene transfer to dispersed human pancreatic Islet cells in vitro using adenovirus-polylysine/DNA compressed or polycationic liposomes. Diabetes 1996;45:1197–203. 9 Leibowitz G. Beattie GM, Kafri T, et al. Gene transfer to human pancreatic endocrine cells using viral vectors. Diabetes 1999;48:745–53. 10 Shah R, Sidner RA, Bochan MR, Jindal RM. Reversal of diabetes in streptozotocin-treated rats by intramuscular injection of recombinant adeno-associated virus containing rat preproinsulin II gene. Transplant Proc 1999;31:641–2. 11 Jindal RM, Sidner R, Shah R. Autoregulation of blood glucose after injection of haematopoietic stem cells exposed to recombinant adeno-associated virus containing preproinsulin II gene. Surg Forum 1998;XLIX:355–6. 12 Millar AD. Putting muscle to work for gene therapy. Nat Med 1997;3:278–9. 13 Unger RH. How obesity causes diabetes in Zucker diabetic fatty rats. Trends Endocrinol Metabol 1997;8:276–82. 14 Shimabukuro M, Zhou YT, Levi M, et al. Fatty acid induced beta cell apoptosis: a link between obesity and diabetes. Proc Natl Acad Sci USA 1998;95:2498–502. 15 Shimabukuro M, Wang MY, Zhou YT, et al. Protection against lipoapoptosis of ␤-cells through leptindependent maintenance of Bcl-2 expression. Proc Natl Acad Sci USA 1998;95:9558–61. 16 Efrat S, Fejer G, Brownlee M, Horwitz M. Prolonged survival of pancreatic islet allografts mediated by adenovirus immunoregulatory transgenes. Proc Natl Acad Sci USA 1995;92:6947–51. 17 Lau HT, Yu M, Fohrana A, Stoeckert CJ Jr. Prevention of islet allograft rejection with engineered myoblasts expressing Fas-L in mice. Science 1996;273:109–12. 18 Bellgarau D, Gold D, Selawry H, Moore J, Franzusoff A, Duke RC. A role for CD95 ligand in preventing grafts rejection. Nature 1995;377:630–2. 19 Gallichan WS, Kafri T, Krahl T, Verma IM, Sarvetnick N. Lentivirus-mediated transduction of islet grafts with interleukin 4 results in sustained gene expression and protection from insulitis. Hum Gen Ther 1998;9:2717–26. 20 Moritani M, Yoshimoto K, Ii S, et al. Prevention of adoptively transferred diabetes in nonobese diabetic mice with IL-10 transduced islet specificTh1-lymphocytes. A gene therapy model for autoimmune diabetes. J Clin Invest 1996;98:1851–9. 21 Muller R, Khral T, Sarvetnick N. Pancreatic expression on interleukin-4 abrogates insulitis and autoimmune diabetes in nonobese diabetic (NOD) mice. J Exp Med 1996;184:1093–9. 22 Von Herring MG, Efrat S. Oldstone MB, Horwitz MS. Expression of adenoviral E3 transgenes in beta cells prevents autoimmune diabetes. Proc Natl Acad Sci USA 1997;94:9808–12. 23 Balasa B, Karhl T, Pastone G, et al. CD40 ligand-CD40 interactions are necessary for the initiation of insulitis and diabetes in nonobese diabetic mice. J Immunol 1997;159:4620–7. 24 Rossini AA, Parker DC, Phillips NE, et al. Induction of immunological tolerance to islet allograft, Cell Transplant 1996;5:49–52. 25 Lau HT, Stoeckert CJ. FasL — too much of a good thing? Transplanted grafts of pancreatic islet cells engineered to express Fas ligand are destroyed not protected by the immune system. Nat Med 1997;3:727–8.

301

302

GENE THERAPY FOR DIABETES

26 Miller DG, Adam MA, Miller AD. Gene transfer by retroviruses occurs only in cells that are actively replicating at the time of infection. Mol Cell Biol 1992;10:4239–242. 27 Becker TC, Beltrandel rio H, Noel RJ, Johnson JH, Newgard CB. Over expression of hexokinase I in isolated islets of langerhans via recombinant adenovirus. Enhancement of glucose metabolism and insulin secretion at basal but not stimulatory glucose levels. J Biol Chem 1994;269:21234–8. 28 Csete ME, Benhamou PY, Drazan KE, et al. Efficient gene transfer to pancreatic islets mediated by adenoviral vectors. Transplantation 1995;59:263–8. 29 Sigalla J, David A, Anegon I, et al. Adenovirus-mediated gene transfers into isolated mouse adult pancreatic islets: normal beta-cell function despite induction of an anti-adenovirus immune response. Hum Gen Ther 1997;8:1625–34. 30 Yang Y, Nunes FA, Berencsi K, Furth EE, Gönczöl E, Wilson JM. Cellular immunity to viral antigens limits E-1 deleted adenoviruses for gene therapy. Proc Natl Acad Sci USA 1994;91:4407–11. 31 Jindal RM, Sidner RA, Bochan MR, Srivastava A. Adeno-associated virus vectors potential for gene therapy. Graft 1998;1:147–53. 32 Srivastava A. Parvovirus-based vectors for human gene therapy. Blood Cells 1994;20:531–8. 33 Blacklow NR, Hoggan MD, Sereno MS, et al. A seroepidemiological study of adeno-associated virus infection in infants and children. Am J Epidemiol 1971;94: 359–66. 34 Peng L, Sidner RA, Bochan MR, Burton MM, Cooper TS, Jindal RM. Construction of recombinant adeno-associated containing the rat preproinsulin II gene. J Surg Res 1997;69:193–8. 35 Kotin RM, Siniscalco M, Samulski RJ, et al. Site-specific integration by adeno-association virus. Proc Natl Acad Sci USA 1990;87:2211–15. 36 Donahue RE, Kessler SW, Bodine D, et al. Helper virus induced T cell lymphoma in non human primates after retroviral medicated gene transfer. J Exp Med 1992;176:1125–35. 37 Carter BJ. Adeno-associated virus vectors. Curr Opin Biotechnol 1993;3:533–9. 38 Naldini L, Blömer U, Gallay P, et al. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 1996;272:263–7. 39 Blömer U, Naldini L, Kafri T, Trono D, Verma IM, Gage FH. Highly efficient and sustained gene transfer in adult neurons with lentivirus vectors. J Virol 1997;71:6641–49. 40 Kafri T, Blömer U, Peterson DA, Gage FH, Verma IM. Sustained expression of genes delivered directly into liver and muscle by lentiviral vectors. Nat Genet 1997;17:314–17. 41 Ju Q, Edelstein D, Brendel MD, et al. Transduction of nondividing adult human pancreatic beta cells by an integrating lentiviral vectors. Diabetologia 1998;41:736–9. 42 Sadaie MR, Zamani M, Wang S, et al. Towards developing HIV-2 lentivirus-based retroviral vectors for gene therapy: dual gene expression in the context of HIV-2 LTR and Tat. J Med Virol 1998;54:118–28. 43 Muller R, Davies JD, Krahl T, Sarvetnick N. IL-4 expression by graphs from transgenic mice fails to prevent allograft rejection. J Immunol 1997;159:1599–603. 44 Barnett AH, Eff C, Leslie RDG, Pyke DA. Diabetes in identical twins: a study of 200 pairs. Diabetologia 1981;20:87–93. 45 Eizirik DL, Pavlovic D. Is there a role for nitric oxide in beta-cell dysfunction and damage in IDDM? Diabetes Metab Rev 1997;13:293–307. 46 Muir A, Ramiya V. New strategies in oral immunotherapy for diabetes prevention. Diabetes Metab Rev 1996;12:1–14. 47 Reddy S, Stefaovic N, Karanam M. Prevention of autoimmune diabetes by oral administration of syngeneic pancreatic extract to young NOD mice. Pancreas 1999;20:55–60. 48 Corbett JA, Wang JL, Hughes JH, et al. Nitric oxide and cyclic GMP formation induced by interleukin 1 beta in islets of Langerhans. Evidence for an effector role of nitric oxide in islet dysfunction. Biochem J 1992;287:229–35.

M. KARANAM ET Al.

49 Karlsen AE, Andersen HU, Vissing H, et al. Cloning and expression of cytokine-inducible nitric oxide synthase cDNA from rat islets of Langerhans. Diabetes 1995;44:753–8. 50 Iwahashi H, Hanafusa T, Eguchi Y, et al. Cytokine-induced apoptotic cell death in a mouse pancreatic beta-cell line: inhibition by Bcl-2. Diabetologia 1996;39:530–6. 51 Rabinovitch A, Suzrez-Pinzon W, Strynadka, et al. Transfection of human pancreatic islets with an anti-apoptotic gene (bcl-2) protects beta-cells from cytokine-induced destruction. Diabetes 1999;48:1223–9. 52 Levine F, Leibowitz G. Towards gene therapy of diabetes mellitus. Mol Med Today 1999;5:165–71. 53 Mauricio D, Mandrup-Poulsen T. Apoptosis and the pathogenesis of IDDM: a question of life and death. Diabetes 1998;47:1537–43. 54 Davalli AM, Scaglia L, Zangen DH, Hollister J, Bonner-Weir S, Weir GC. Vulnerability of islets in the immediate posttransplantation period. Dynamic changes in structure and function. Diabetes 1996;45:1161–7. 55 Bennet W, Sundberg B, Groth CG, et al. Incompatibility between human blood and isolated islets of Langerhans: a finding with implications for clinical intraportal islet transplantation? Diabetes 1999;48:1907–14. 56 Rafaeloff R, Pittenger GL, Barlow SW, et al. Cloning and sequencing of the pancreatic islet neogenesis associated protein (INGAP) gene and its expression in islet neogenesis in hamsters. J Clin Invest 1997;99:2100–9. 57 Stoffers DA, Stanojevic V, Habener JF. Insulin promoter factor-1 gene mutation linked to early-onset type 2 diabetes mellitus directs expression of a dominant negative isoprotein. J Clin Invest 1998;102:232–41. 58 Waeber G, Thompson N, Nicod P, Bonny C. Transcriptional activation of the GLUT2 gene by the IPF1/STF-1/IDX-1 homeobox factor. Mol Endocrinol 1996;10:1327–34. 59 Bone AJ, Banister SH, Zhang S. The REG gene and islet cell repair and renewal in type 1 diabetes. Adv Exp Med Biol 1997;426:321–7. 60 Lacy PE. Pancreatic islet cell transplant. Mt Sinai J Med 1994;61:23–31. 61 Docherty K. Gene therapy for diabetes mellitus. Clin Sci 1997;92:321–30. 62 Freeman DJ, Leclerc I, Rutter GA. Present and potential future use of gene therapy for the treatment of non-insulin dependent diabetes mellitus. Int J Mol Med 1999;4:585–92. 63 MacFarlane WM, Chapman JC, Shepherd RM, et al. Engineering a glucose-responsive human insulinsecreting cell line from islets of Langerhans isolated from a patient with persistent hyperinsulinemic hypoglycemia of infancy. J Biol Chem 1999;274:34059–66. 64 Lipes MA, Davalli AM, Cooper EM. Genetic engineering of insulin expression in nonislet cells: implications for beta-cell replacement therapy for insulin-dependent diabetes mellitus. Acta Diabetol 1997;34:2–5. 65 Nielsen DA, Welsh M, Casadaban MJ, Steiner DF. Control of insulin gene expression in pancreatic beta cells and in an insulin-producing cell line, RIN-5F cells. I. Effects of glucose and cyclic AMP on the transcription of insulin mRNA. J Biol Chem 1985;260:13585–9. 66 Docherty K, Clark AR. Nutrient regulation of insulin gene expression. FASEB J 1994;8:20–7. 67 Moore HP, Walker MD, Lee F, Kelly RB. Expressing a human proinsulin cDNA in a mouse ACTHsecreting cell. Intracellular storage, proteolytic processing, and secretion on stimulation. Cell 1983;35:531–8. 68 Vollenweider F, Irminger JC, Gross DJ, Villa-Komaroff L, Halban PA. Processing of proinsulin by transfected hepatoma (FAO) cells. J Biol Chem 1992;267:14629–36. 69 Newgard CB. Cellular engineering and gene therapy strategies for insulin replacement in diabetes. Diabetes 1994;43:341–50.

303

304

GENE THERAPY FOR DIABETES

70 Simpson AM, Marshall GM, Tuch BE, et al. Gene therapy of diabetes: glucose-stimulated insulin secretion in a human hepatoma cell line (HEP G2ins/g). Gen Ther 1997;4:1202–5. 71 Shah R, Jindal RM. Stable transfection of rat preproinsulin II gene into rat hematopoietic stem cells via recombinant adeno-associated virus. Life Sci 1999;65:2041–7. 72 Mitanchez D, Doiron B, Chen R, Kahn A. Glucose-stimulated genes and prospects of gene therapy for type I diabetes. Endocr Rev 1997;18:520–40. 73 Efrat S, Fusco-DeMane D, Lemberg H, al Emran O, Wang X. Conditional transformation of pancreatic beta cell line derived from transgenic mice expressing a tetracycline-regulated oncogene. Proc Natl Acad Sci USA 1995;92:3576–80. 74 Efrat S, Leiser M, Surana M, Tal M, Fusco-Demane D. Fleischer N. Murine insulinoma cell line with normal glucose-regulated insulin secretion. Diabetes 1993;42:901–7. 75 Knaack D, Fiore DM, Surana M, et al. Clonal insulinoma cell line that stably maintains correct glucose responsiveness. Diabetes 1994;43:1413–17. 76 Soldevila G, Buscema M, Marini V, et al. Transfection with SV40 gene of human pancreatic endocrine cells. J Autoimmun 1991;4:381–96. 77 Wang S, Beattie GM, Mally MI, Lopez AD, Hayek A, Levine F. Analysis of a human fetal pancreatic islet cell lines. Transplant Proc 1997;29:2219. 78 Efrat S. Prospects for gene therapy of insulin-dependent diabetes mellitus. Diabetologia 1998;41:1401–9. 79 Fleischer N, Chen C, Surana M, et al. Functional analysis of a conditionally transformed pancreatic beta-cell line. Diabetes 1998;47:1419–25. 80 Swenne I. Effects of aging on the regenerative capacity of the pancreatic beta cell of the rat. Diabetes 1983;32:14–19. 81 Hayek A, Beattie GM, Cirulli V, Lopez AD, Ricordi C, Rubin J. Growth factor/matrix-induced proliferation of human adult beta cells. Diabetes 1995;44:1458–60. 82 Anderson W. Human gene therapy. Nature 1998;392(Suppl.):25–30. 83 Ramiya VK, Maraist M, Arfors KE, et al. Reversal of insulin-dependent diabetes using islets generated in vitro from pancreatic stem cells. Nat Med 2000;6:278–82.

Chapter 22

An historical view of the development of cellular and islet transplantation Derek W.R. Gray

In most research fields advancement comes in hesitant steps; periods of slow progress punctuated by occasional leaps forward. Each advance is built on a foundation that mixes the contributions of numerous individuals. The record of this advancement is the published literature and this is the basis on which this historical account will be largely based, each key historical reference being distinguished by superscript numbering. Which publication should be chosen for distinction? Some studies are of such quality that they stand out, but most are either flawed or inconsequential. Only those studies that are repeatable can be considered as true contributions to the development of the field, and many remarkable claims that cannot be independently confirmed must be consigned to the historical discard bin. This is a particular problem for some areas of islet transplantation research, which have suffered (the term is used deliberately) from a surfeit of commercial interest and secrecy. The field of islet transplantation is still evolving. In some cases the value of recent contributions will become clear-cut only with further development of the field. Who should be credited with the multiple advances that form the foundations of such an emerging field? Inevitably a single author account such as is presented in this chapter will reflect one viewpoint, and usually what is done on such occasions is for the author to produce a story-line, with a meandering thread that mentions the names of ‘the great and the good’ and hopefully makes a decent read. But this form of ‘cronyism’ is often grossly unfair for many reasons, and the account presented here will try to develop an alternative approach, in which the text will contain only references and not names. In this account of the historical development of the broad field of cellular transplantation I will concentrate chiefly on islet transplantation, as it is undoubtedly the most highly developed form of cellular transplantation, and because it is the field where most of the original advances have been made. The major contributors will then be identified within the list of authors by making liberal use of different fonts and print styles to provide greater, and hopefully more accurate, identification of significant contributions (see footnote). Islet transplantation is a particularly fascinating area of research because it has substantial links to other disciplines, and often these links have provided valuable impetus to development in other areas. The fields which are substantial enough to warrant separate treatment from the main topic of islet transplantation include islet/␤-cell physiology, diabetes research, and general immunology. A final category can be reserved for advances in non-islet cellular transplantation.

The early development of islet transplantation Following the introduction of insulin for treatment of diabetes in the 1920s it took many years for the initial euphoria surrounding the remarkable discovery to wear off, and the clinical syndromes that we

306

HISTORICAL VIEW OF THE DEVELOPMENT OF CELLULAR AND ISLET TRANSPLANTATION

now know as the complications of diabetes only became manifest after 15 to 20 years of persistent hyperglycaemia. It was therefore not until the late 1940s that it became clear that insulin-treated diabetics not only suffered the inconvenience of repeated measurement and injections with continued dietary restriction, but were often afflicted with devastating complications such as blindness, renal failure, and vascular disease. During the 1950s the scale and severity of the problem was documented. The necessity of finding a way to prevent these complications and the early success of kidney transplantation in the early 1960s parented the birth of transplantation for diabetes as a serious concept. Subsequently, extensive experimental and clinical interest led to the early (1965 to 1970), largely disastrous, trials of vascularized pancreas transplantation, which are described in detail elsewhere in this volume. The fact that the exocrine component of the pancreas gland is not needed to treat diabetes by transplantation was recognized even before insulin was discovered, and proposals to treat diabetes by using separated islet cells were also made remarkably early. However, these were not realistic aims as there was no practical means of separating islet from endocrine tissue. In the late 1950s there was increasing interest in obtaining islet tissue to allow physiological studies of insulin secretion in vivo [1], and this was initially achieved by introduction of a microdissection technique that was slow but produced sufficient islet tissue from suitable murine strains for the purpose of physiological studies [2]. However, the technique was so laborious that it led others to attempt to speed up the process of digestion of chopped (guinea-pig) pancreas tissue using a crude collagenase preparation to digest the tissue [3], and this latter technique was the crucial first technical advance that established the field of islet transplantation. The potential of this technique was recognized in at least two centers (Philadelphia and St Louis), both of which moved to the rat model, but it was the latter group that described the addition of pancreatic duct cannulation and distension with cold balanced salt solution prior to excision of the gland followed by further scissor mincing. These manoeuvres improved the efficiency of the subsequent collagenase digestion process, allowing separation of up to 200 islets per pancreas, sufficient to allow consideration of transplantation as a realistic option [4].

Islet transplantation studies in the isogeneic rodent model The availability of inbred strains of rats and mice was a particular boon to the development of islet transplantation, since it allowed transplantation experiments using multiple donors and so was able to overcome the early problems of poor yield of islet tissue. To test the viability of the islets, a rat model of experimental diabetes was used, derived from work on the mechanism of induction of diabetes by the agent streptozotocin [5,6]. The first successful islet transplant was therefore in the syngeneic rat model, using intraperitoneal islets to cure experimental diabetes [7]. In fact the function achieved in these first islet isografts was relatively poor despite the use of multiple donors (three or more). It has since become clear that the intraperitoneal site requires approximately twice the number of islets to induce reversal of diabetes than it is possible to use in other implantation sites. However, at the time (1972) the first report of successful islet transplantation was sensational, and a number of other laboratories rapidly repeated the technique. There followed a series of experiments from various laboratories, all based on the rat model, which explored ways of improving the islet yield, using various sources of insulin-secreting tissue sources, seeking the best site for experimental and clinical islet transplantation. An important technical advance was the description of the use of density gradients such as Ficoll to separate islets from the digested exocrine tissue [8], after showing that islets were less dense than exocrine tissue. Ficoll, a polysaccharide polymer, remains useful for this purpose to this day. In the search for the ideal site for islet implantation an obvious choice would be the pancreas itself. However, this is much harder to

D.W.R.T GRAY

achieve in practice than in theory, especially in the rat where the pancreas is a more diffuse organ. Pancreatic venous drainage is normally to the liver, which is also the main site of metabolic action by insulin, and therefore the liver was an early candidate site for transplantation of islets. Direct injection into the liver parenchyma was not found to be very efficacious, but by adapting the technique so that the islets were injected into the portal vein [9] rapid and complete reversal of experimental diabetes by islet transplantation was achieved for the first time.

Islet transplantation studies in rodent models of diabetes Type 1 diabetes is now proven beyond doubt to be an autoimmune disease, and therefore disease recurrence may be expected to be a problem unless steps are taken to prevent it damaging the islets. Islet transplantation has been used extensively as a research tool to investigate the aetiology of autoimmune diabetes in murine models such as the non-obese diabetic (NOD) mouse [10] and the BB rat [11], and to investigate the likelihood and mechanism of diabetes recurrence after islet transplantation, for example to examine the role of T cells [10].

Fetal pancreas transplantation (see also Chapter 23) The fortunate ability of explanted fetal murine pancreas to undergo atrophy of the exocrine tissue during tissue culture yet allow growth and development of the endocrine tissue was first described in 1960 [12]. This meant that fetal pancreatic tissue could be processed to particles of transplantable size by simple mechanical mincing, without the need for collagenase digestion. The model used originally was transplantation to the anterior chamber of the eye, a site which has a degree of immune protection. Fetal pancreatic tissue was subsequently shown to survive in tissue culture and produce insulin, both for rat [13] and human tissue [14]. These early in vitro culture studies demonstrated that fetal pancreatic islets could mature and function normally, but the use of fetal tissue for human clinical transplantation became clear (see above). In an isogeneic rat model, a novel transplantation site, namely the kidney capsule, was used to demonstrate implantation of cultured fetal pancreatic tissue, and subsequently to demonstrate actual graft function [15]. It was also shown that reversal of diabetes required prolonged periods in a normoglycaemic environment before graft function was established, but after waiting for maturation then reversal of diabetes could be achieved using a single donor [16]. Several centres then took up the technique, and confirmed these findings, documenting the metabolic efficiency and exploring other sites of implantation, but no implantation site proved better for fetal pancreas in the experimental animal than the kidney capsule site.

Early studies of human fetal pancreas transplantation As was the case with trials of clinical isolated adult islet transplantation (see below), a number of centres progressed directly from rodent-based studies into phase I clinical trials, and again the results were essentially negative or at best could not be proven by repetition. The scale of some of these trials was tremendous: in China and the then Soviet Union many thousands of fetal pancreas implants were performed, and yet still no repeatable technique has emerged. It remains uncertain what exactly the cause of failure was in these early trials, but there was no proven method for prevention of rejection in most cases. It is worth recognizing that these clinical trials were conducted without the benefit of an intermediate experimental model in a large animal, largely because there is a barrier in that fetal tissue is inevitably a half match with the mother, and until the advent of truly inbred large animal herds neither isograft nor autograft models were possible. The first repeatable test of in vivo function of

307

308

HISTORICAL VIEW OF THE DEVELOPMENT OF CELLULAR AND ISLET TRANSPLANTATION

human fetal pancreatic tissue was by transplantation, making use of immunocompromised animals as hosts, where glucose regulation to the normal human range (as opposed to the higher rodent range) was demonstrated [17].

Neonatal pancreas transplantation (see also Chapter 23) It was shown that some of the difficulties that surrounded the timing of pregnancy and dissection necessary to obtain fetal pancreatic tissue could be obviated by using neonatal rather than fetal tissue, and there was also the advantage that more tissue was retrieved. The exocrine tissue was still relatively undeveloped at this stage, but the islet tissue could potentially function immediately. However, to break up the neonatal pancreas efficiently it was necessary to reintroduce a collagenase digestion step, but further purification other than by tissue culture was unnecessary [18].

General studies, methods, and techniques that made valuable contributions to islet transplantation A number of ‘baseline’ studies, resources, and laboratory techniques are regularly used in islet transplantation as useful tools, and are difficult to allot to any one particular area of islet transplantation above others. These include the introduction of immunodeficient murine strains, beginning with the nude, athymic mutant mouse [19], the use of neutral red as a viability stain and also as a specific stain for islets (but only when given intravenously) [20], which allowed the earliest (and arguably still the best) studies of the number and volume of islets within the human pancreas and subsequently other species too [21,22].

Techniques for identification of islets The ability to stain islet tissue rapidly to distinguish it from exocrine tissue was an important requirement for developing any new modification to islet isolation techniques, and the description of the zinc chelator dithizone as a specific rapid stain for islets must be seen as a major contribution from the point of view of the development of islet transplantation. It is the author’s understanding that the technique was originally applied to islet tissue for the purpose of islet identification by a group who were generous in passing on the idea to others but slow to publish [23], the method being more rapidly published by a group who did not originally make the discovery but perhaps better understood the potential value of the technique to transplantation and diabetes research [24,25].

Techniques for assessment of islet viability The development of tissue culture systems was applied to short-term and subsequently long-term culture of isolated islets [26], and perfusion systems measuring oxygen uptake and the physiological response to glucose [27] became the gold standard for in vitro measurement of isolated islet viability. However, these techniques were technically demanding and give results only days or weeks later. The use of supravital stains to assess viability is commonplace for the assessment of single cells growing in tissue culture, and many studies used the long-established technique of trypan blue dye exclusion for islets, sometimes combined with neutral red to improve live–dead discrimination. However, the use of these dyes was not formally tested for islets and it was common experience that intermediate degrees of reduced islet viability were difficult to detect or interpret with these techniques.

D.W.R.T GRAY

Fluorescent dyes were developed as more accurate discriminators of single cell viability, and it was subsequently shown that the combination of fluorescein diacetate and ethidium bromide was an accurate, rapid, and cheap method for assessing islet viability [28] which allowed the assessment of intermediate degrees of islet viability. Other fluorescence dyes (which are more stable and therefore easier to handle) can also be used with similar results for direct microscopy [29], and have the further advantage of also allowing electronic signal detection and computerized image analysis to assess viability objectively [30]. Isolated islets of Langerhans certainly survive in tissue culture and are small enough to be potentially storable by making use of cryopreservation techniques. However, the tissue is very fragile and very easily damaged by unfavourable conditions during the freezing/thawing process, and the definition of the exact conditions that will allow most efficient freezing, storage, and restoration of complete viability after thawing has been a major undertaking [31–33]. Although the claimed results of islet freezing and retrieval are impressive from certain centres, the consensus view is that still approximately 50 per cent of the tissue is destroyed during the freezing process, but it is likely that recent improvements have advanced that figure considerably.

Prevention of contamination with lymph nodes A particular problem that mainly afflicts murine models of experimental islet transplantation, is that contaminating lymph nodes have a similar density to islets, which can certainly give very unusual results if the islets are included in any immunological studies. A technique for enhancing the subtle differences between islet endocrine and exocrine tissue using transmitted green light and side-lighting with white light [34] has been described and this has been used in many rodent and human studies as a way of excluding lymph node contamination.

The role of exocrine contamination and hyperglycaemia in islet implantation Histological examination of islet grafts implanted beneath the kidney capsule made it clear that acinar exocrine tissue did not survive the transplantation process, although ductular elements were often found. Since the acinar tissue failed to survive it could result in an inflammatory process that would impair islet implantation, and impaired implantation of islets due to contaminating exocrine tissue to a confined implantation site such as the kidney capsule was first demonstrated in a rodent isograft model [35]. Hyperglycaemia has long been identified as being detrimental to islet function and longterm hyperglycaemia was shown to be detrimental to islet graft survival [36].

Attempts to quantify the outcome of islet isolation An important barrier to advancement in the field of islet transplantation has been the question of quantification and comparison between laboratories. To try and overcome this problem an important contribution was an attempt to produce a consensus view on which methods should be used to quantify yield, purity, and viability of islet preparations [37]. This did not entirely overcome the problem of comparability between laboratories, which continues to be a concern today, but it did make some progress towards the aim, and remains a useful standard.

The anatomy and physiology of islet transplantation Murine pancreatic islets were used for many years for numerous in vitro studies of insulin secretion but the relevance for humans was unknown until the sudden availability of human islets meant that a

309

310

HISTORICAL VIEW OF THE DEVELOPMENT OF CELLULAR AND ISLET TRANSPLANTATION

variety of characteristics of insulin secretion could be confirmed as being similar for human-derived islets [38,39]. Other important studies examined the revascularization of islets using direct microscopy [40], or measurement of islet blood flow [41] and documented poor perfusion of islets in a hyperglycaemic environment [42,43]. Reinnervation of the transplanted tissue was also documented [44].

Allotransplantation of islets and fetal pancreas in murine models The easy availability of inbred strains meant that most early studies of islet allotransplantation were carried out in murine species, first documenting the occurrence of rejection [45] and many centres then studied the rejection process in a series of publications, in great detail [46,47]. Several centres undertook studies of islet allograft rejection and attempted immunosuppression with a wide variety of agents. The burgeoning interest in islet and fetal pancreas transplantation coincided with the introduction of ciclosporin, and naturally this agent was extensively investigated for its effect alone and combined with other agents. The outcome of these studies was recorded and analysed in an influential series of summary papers [48,49]. Most, if not all immunosuppressive agents, including ciclosporin, were relatively ineffective for the prevention of islet allograft rejection, when compared to the effect on vascularized grafts. Partly for this reason, approaches to islet transplantation that relied more on the manipulation of the rejection process began to be studied in islet and fetal pancreas transplantation models. These studies contributed to immunobiology as a whole rather than just to the particular cause of islet transplantation. Examples of innovative contributions include the demonstration of a lower limit or threshold to the allograft rejection response, using reduced size islet allografts (of 50 islets) showing that it was possible for the immune system to fail to detect the transplantation of such small grafts [50]. The remarkable specificity of the allograft rejection response was also demonstrated using mixed islet allograft/isografts [51].

Migratory antigen-presenting cells The concept of what are now called migratory antigen-presenting cells (APCs) bearing costimulatory molecules was originally termed as ‘passenger leucocytes’ bearing a ‘second signal’. Their presence was inferred by the reduction in rejection of cultured free allografts, subjected to conditions designed to reduce the survival of leucocytes. Originally described for thyroid allografts, fetal pancreatic allotransplantation in the murine model proved to be an ideal test system [52,53]. The concept was extended to adult islet transplantation in a variety of approaches based on passenger leucocyte depletion in a mouse allograft model [54,55] and in a rat model [56], demonstrating eventual development of tolerance induction [55]. The routine addition of tissue culture to collagenase dispersed preparations of neonatal or fetal pancreas in specific culture conditions led to the development of islet-like cell clusters, termed by some workers as ‘ICCs’ or ‘proislets’ by others. This was shown to be an efficient and reproducible way of reducing immunogenicity for murine allograft models, including grafts into outbred recipients [57]. However, the phenomenon was shown to exhibit strain variability [58], but could me made effective in all strains by addition of anti-CD4 antibody treatment and/or the use of low-dose immunosuppression [59].

Immunologically privileged implantation sites An interesting feature of free graft rejection discovered in the early days of transplantation was that certain implantation sites such as the brain or vitreous humour of the eye exhibited a varying degree

D.W.R.T GRAY

of protection against immune attack, which depended on the exact nature of the tissue and method of transplantation. For islets, virtually every site physically possible has been investigated and confirmed as an implantation site, and immune privilege claimed for the eye and brain ventricles, but few of these claims have led to a useful technique based on repeatable studies. One exception has been the investigation of the intratesticular site, which can allow function if placed intra-abdominally [60]. A long series of investigations has demonstrated that this effect is due to a factor associated with Sertoli cells [62]. This work has given fresh impetus to the approach of cotransplanting cells that will turn off the immune response, or getting the graft itself to express the inhibitory molecules, both approaches being currently under active investigation by a number of groups.

Alternative approaches to prevention of rejection Thymic tolerance induction An implantation site investigated in murine models for islet transplantation, but which turned out to have far wider implications for immunobiology, was the intrathymic implantation site [63]. The exciting finding was that not only did islet allografts transplanted to the thymus function (it is now clear that the intrathymic site is metabolically quite inefficient), but they also did not reject if even minimal doses of antilymphocyte antibody was given. The potential of the site for islet transportation was overshadowed by the demonstration that this approach led to tolerance for subsequently transplanted tissue. Many immunobiologists describe this as a ‘rediscovery’, because the effect had been described for non-cellular antigen two decades before. However, there is no doubt that the resurgence in interest was due to recent findings with islet allografts. The basis of this tolerance has been extensively investigated by at least a dozen major centres, and although there remain controversies, it is agreed that at least some of the effect is due to intrathymic deletion of alloreactive cells due to the presence of the immunologically reactive graft [64]. The potential for clinical application was exciting, and encouraged attempts to develop large animal models of intrathymic allotransplantation. Initial reports were not encouraging but recent results may change that view.

Encapsulation As early as 1954 it was shown that simply enclosing cultured cells in a membrane with limited pore size was sufficient to prevent rejection of allografted tissue [65]. This approach was given a significant boost for islet transplantation by the description of microencapsulation as a way of wrapping each islet in a barrier membrane [66]. Treatment of experimentally induced diabetes has been demonstrated and confirmed in rodent allograft models. However, doubts remain over the ability of such membranes to prevent xenograft rejection, where most of the immune response is likely to be driven by the indirect presentation pathway, a fact originally pointed out much earlier [65] and reconfirmed in elegant experiments more recently [67]. Despite intensive research by a number of privately financed centres, no repeatable method for transplanting encapsulated xenogeneic islets into humans to cure diabetes has emerged so far.

Costimulatory molecule blockade Recently much interest has focused on the role of the blockade of costimulatory molecules as a way of abrogating the alloresponse. Much of this work is too recent to be included in an historical account such

311

312

HISTORICAL VIEW OF THE DEVELOPMENT OF CELLULAR AND ISLET TRANSPLANTATION

as this, since its true value and place in the developing scheme is not yet possible to assess. Similar comments apply to the use of potent depleting immunotoxins which have shown great potential for prevention of rejection. However, it is worth noting that once again islet transplantation models have been at the forefront of these approaches, and undoubtedly will appear in subsequent historical treatises.

Early attempts at clinical islet isolation and transplantation In terms of writing a history of an academic subject it is always difficult to know how much credit must be given to those who perform essentially negative studies, whose value is essentially to show the direction that a field should not take. The early attempts at clinical islet transplantation come into this category. Looking back, these early trials were excessively optimistic, applying a method for islet isolation that was developed for the rodent pancreas to the obviously different human pancreatic gland, with little thought or care as to the content or viability of the tissue being transplanted. Deaths were recorded from portal hypertension and disseminated intravascular coagulation, and it is likely these reports were only the tip of an unreported iceberg. A number of claims of successful human islet transplantation were made, and one or two of these claims were even backed up by reasonable evidence, but none led to a repeatable technique. This author is of the opinion that these hasty experiments contributed little except by acting as a warning and gave islet transplantation a reputation from which it has struggled to recover. The advances that did lead to eventual clinical isolated adult islet transplantation were mainly through development of islet isolation techniques in large animal models.

Islet transplantation in the large animal model In the 1990s it became increasingly politically incorrect to support large animal experimentation. Yet if islet transplantation or a derivative ever becomes clinically reality for routine treatment of diabetes, it is a field where the contribution of large animal experimentation must never be underestimated. To evaluate the functional consequences of various islet preparation methods, techniques for reliable induction of experimental diabetes in a large animal model had to be developed, by far the most reliable being that of total pancreatectomy. The lack of inbred strains meant that the excised pancreas itself became the only reliable source of islet tissue for the original autotransplantation studies. To perform total pancreatectomy whilst keeping the pancreatic tissue viable for islet isolation was a technical triumph of surgery that has been little recognized. The techniques used were either developed for, or subsequently used for, studies of vascularized pancreas transplantation techniques in animal models, and also used for islet transplantation models. The first repeatable technique for islet transplantation in the dog abandoned the attempt at islet purification completely. The islet preparation technique was crude mincing of the tissue with carefully controlled collagenase digestion, but the viability of the preparation was sufficient to produce fasting normoglycaemia for the first time in a fully diabetic, total pancreatectomy canine autotransplantation model. A key point in the success of the studies in the dog model was the use of the spleen as the implantation site, initially using direct injection into the pulp [68], later by retrograde infusion into the splenic venous system [69]. This model was repeated and intensively studied by others, and extended to porcine allograft model, where claims of allograft function were made, but were not repeatable. The next advance was to move to infusion of collagenase directly into the duct [70], modified to a ductal injection [71] with a two-stage digestion process, which was subsequently adapted a singlestage digestion process that allowed studies of islet autotransplantation after total pancreatectomy in

D.W.R.T GRAY

the dog [72] and monkey [73]. These studies proved beyond doubt in a preclinical model that islet grafts could be performed repeatably and safely, that islets could function in the intraportal implantation site [72,73], and that intraportal islet grafts were capable of prolonged graft function [74]. These studies also showed that the intraportal site was functionally better than the intrasplenic and kidney capsule sites in preclinical primate models, although functional grafts at these alternative sites were later shown to be possible [74,75]. It is notable that the above advances do not include studies which utilize porcine pancreas. This was not because of any lack of interest in isolating porcine islets: in fact numerous isolation methods were published up to the late 1980s but none were repeatable. Although the pancreatic tissue of the pig appears similar to that of the human in terms of gross anatomy, the islets are surprisingly fragile, lacking a robust collagen framework, which means that porcine islet isolation is a particularly demanding procedure. In fact it only became possible to isolate porcine islets reproducibly by applying the lessons already learned in human islet isolation [76] adapting that technique by adoption of a previously described digestion/filtration chamber [77], with modifications to produce slower digestion such as lowering the incubation temperature [78] and further refinements leading to a repeatable method for porcine islet isolation [79]. Successful islet autotransplantation has now been achieved in the pig model, but it remains a demanding procedure, which means there is little enthusiasm generally for using the porcine model to investigate islet allotransplantation.

Studies of islet allotransplantation in large animal models The repeatable reports of islet allotransplantation were all in the dog model, and used ciclosporin to prevent rejection [80]. These studies were remarkable for emphasizing the high levels of ciclosporin necessary to prevent islet allograft rejection, later confirmed and further defined by others [81]. The number of islets necessary to reverse pancreatectomy induced diabetes was defined at between 3 and 5000 per kilogram of body weight [82].

The further development of human islet isolation into clinical application The human pancreas is a relatively dense, fibrous gland, varying in thickness to between 0.2 and 4.5 cm. Application of the original method for separation of murine islets of Langerhans involves a ductal injection of cold Hanks solution prior to mincing the tissue with sharp scissors, then digestion with collagenase. In the rodent pancreas the ductal injection produces massive distension of the gland, and mechanically separates at least some of the islets allowing introduction of collagenase to further cleave the exocrine tissue away from the islets. However, mechanical distension usually has little effect on the adult human pancreas, and virtually no islets are isolated. The advance, which allowed isolation of human islets in large numbers for the first time, was the delivery of the collagenase by retrograde injection into the biliary duct in a relatively small volume, with the aim of selectively permeating the exocrine tissue with the relative sparing of the islets, relying on the collagenase alone to produce the cleavage of exocrine from islet tissue. This produced optimal conditions to ensure collagenase enzyme activity [76] and made use of twin mesh screens to ensure only digested tissue was released. Purification then followed on a discontinuous Ficoll density gradient identical to that used in the rat. This method produced a yield of approximately 1000 islets per gram of pancreatic tissue, which was

313

314

HISTORICAL VIEW OF THE DEVELOPMENT OF CELLULAR AND ISLET TRANSPLANTATION

calculated to be around 10 to 20 per cent of the available islet tissue, probably insufficient to undertake human islet transplantation with any expectation of producing insulin independence. Several groups repeated these results and began to seek improvements. An early study from the St Louis group reported the introduction of the first digestion/filtration chamber to semi-automate the digestion process, allowing egress of islets only once the digestion process had progressed to a point at which the tissue particles had reduced to such a size that passage was allowed through a filtration mesh. The appropriately named digestion/filtration chamber was originally applied to rhesus monkey pancreas using a modification of the rat isolation method, with modest success [77], but was later applied to digestion of the pancreas of various species in combination with the intraductal collagenase injection technique [76]. The result was a semi-automated method for preparation of islets from first the porcine [78] and then human pancreas [83], with further modifications to improve the method for the porcine pancreas [79] made by the same group. The technique has been widely adopted and is often given the eponym of only the one author, which is hardly fair recognition of the input of the others involved in this advance. It would be fairer to refer to it as the ‘automated digestion chamber technique’ and drop eponyms. Although the addition of a digestion/filtration chamber represented a real advance in terms of both the yield and speed of processing human and porcine pancreatic tissue, the large quantities of digest produced still had to be processed by separating the islets from exocrine tissue using discontinuous density gradients, mostly based on Ficoll as the dense molecule. This was highly labour intensive and time-consuming, so the introduction of an automated machine to do the job, namely the Cobe 2991 [84], was welcome news to those working in the field and represented a real advance. The machine was originally designed for processing large volumes of blood to separate blood products, but adapts well for the purpose of islet isolation. Apart from the speed of processing the machine also allows simple preparation of a continuous density gradient, which is certainly convenient and which may lessen the variability of yield between pancreases.

Clinical islet transplantation: the quest for insulin independence The goal of islet transplantation is to produce control of glucose metabolism sufficient to allow withdrawal of the insulin therapy. To achieve this aim it is not only necessary to transplant sufficient viable islet tissue, but also to avoid two destructive processes: graft rejection and recurrent autoimmune disease. Although there were a number of claims of successful islet transplants with reversal of diabetes in the 1970s and early 1980s, none of these were repeatable by other groups, the basic problem being the lack of viable islet tissue in the grafts performed. The situation remained unchanged until the mid-1980s, when the improvements in the yield and purity of the islet preparations and the use of several donors combined with moderately heavy immunosuppression resulted in the first fully documented and repeatable example of insulin independence following islet transplantation in a type 1 diabetic recipient with proven absence of insulin or C-peptide production pretransplant [85]. Although the duration of insulin independence achieved was little more than 2 weeks, subsequently other groups using a similar approach obtained periods of insulin independence that were longer [86,87]. A period of intensive effort to improve the islet yield followed resulting in reduction of the number of donor pancreas necessary for reversal of diabetes, eventually down to just a single donor [88], although this was not achievable routinely (10 to 20 per cent of cases achieved insulin independence). The importance of autoimmunity as a factor in the relatively poor results of islet transplanta-

D.W.R.T GRAY

tion at this time was demonstrated by the high proportion of patients who achieved insulin independence when islet allotransplantation was performed in association with liver transplantation in patients made diabetic by total pancreatectomy and performed in conjunction with liver resection prior to combined liver and islet transplantation [88]. Once it was demonstrated that insulin independence in type 1 diabetes could be achieved by islet allotransplantation, the drive was then for documentation of long-term function [89] and slow but steady improvement followed. Analyses of the parameters affecting the success of islet isolation for transplantation were performed, and the most important factors were found to be a variety of donor characteristics and prolonged ischaemia [90]. Reports of increasing success were confirmed and extended by several laboratories [91,92], often collaborating to analyse data [93]. The relevant data and results of islet transplantation were initially recorded alongside the vascularized organ data in the pancreas transplantation registry [94]. Subsequently a database dedicated to islet transplantation alone was set up to accumulate data, which was fed back to the research community in the form of regular newsletters containing analyses and tables. Coming up to the late 1990s, which is the limit of this historical review, the data showed that there was a modest but definite improvement in the number of islet transplants achieving graft function as defined by C-peptide production, but little change in the proportion of patients achieving insulin independence, which remained around 10 per cent overall, although some centres were performing slightly better [95].

Islet autotransplantation following total pancreatectomy Further confirmation that autoimmunity was an important factor in type 1 diabetic patients given islet transplants was the finding that surprisingly few islets were required for reversal of diabetes in patients who required total pancreatectomy (usually for the treatment of pain due to chronic pancreatitis in patients who then underwent transplantation of autologous islets prepared from the excised pancreas [96]. The minimum number of islets required to prevent the patient becoming diabetic after total pancreatectomy is difficult to calculate precisely, because the autotransplanted tissue is often not purified beyond the initial digestion step, being already relatively depleted of exocrine tissue, but it is of the order of half the islet mass. Furthermore continuing graft function beyond 7 years has been documented [9], giving an optimistic outlook for islet allotransplantation if rejection could be prevented.

Xenotransplantation of islet tissue Given the rejection problems which make islet allotransplantation so difficult to achieve, it may seem to observers of the research process that the idea of moving towards xenotransplantation must surely be a humorous gesture. But that is not the case. By moving to a xenogeneic model the problems of islet supply become solvable by processing sufficient numbers of donors. It is this potentially inexhaustible supply as well as the potential for the manipulation of both the donor and the recipient within a planned xenograft transplantation protocol, which is the major attraction of using xenogeneic tissue. However, there are many problems to be overcome, including animal rights issues and the potential for transmission of ‘unknown’ viruses. But by far the biggest scientific problem is that of rejection. The most rapid form of rejection of vascularized grafts is known as hyperacute rejection,

315

316

HISTORICAL VIEW OF THE DEVELOPMENT OF CELLULAR AND ISLET TRANSPLANTATION

which results in loss of the graft within minutes or hours. It is now known that hyperacute rejection is based on incompatibilities within the ‘blood group’, complement, and clotting cascades, and the relative proportion of each depends on exactly the species combination under consideration. Isolated islets are not primarily vascularized, so isolated islet xenografts were assumed to be free of problems from hyperacute rejection, and certainly in the rodent recipient model, this appears to be the case. However, extrapolation direct from rodent to human is always a dangerous undertaking. Primate recipient studies have shown that both rabbit and porcine islets are subject to an early destruction process that has all the hallmarks of hyperacute rejection [98], and this finding has now been repeated and confirmed by an independent laboratory (Korsgren et al., data-on-file, personal communication). The basis for this incompatibility is the subject of recent research, but beyond the scope of this historical review. Interestingly, it may well not be the same for fetal pancreatic xenografts, since cultured fetal porcine proislets transplanted into primate recipients did not appear to undergo the same process. If hyperacute rejection is avoided, as is normally the case for islets or fetal pancreas transplanted into murine recipients, then the rejection process is based upon cellular infiltration, initially demonstrated in the earliest islet xenograft experiments [99], and if the species difference is wide, most of the infiltrate appears to be macrophage-derived lineages [100], with just a scattering of T cells. However, the T-cell content is vital, because if they are removed no rejection occurs, as was demonstrated initially by xenotransplantation into nude athymic rodents [101]. Immunosuppression is remarkably powerful for rodent recipients of isolated islet xenografts, particularly mice receiving islets from rats, and most of the powerful immunosuppressive drugs have been shown to prevent xenografted islet rejection, again usually using the rat to mouse donor situation. However, moving model to the large animal or primate recipient, the rejection of vascularized xenografts is vigorous, and what signs there are suggest that xenogeneic islet transplants across wide species barriers elicit rapid and powerful rejection responses in the primate recipient. There are now many approaches under active investigation with the aim of modifying the fate of xenogeneic tissue in the primate recipient, but these are outside the remit of a historical review.

Transplantation of other non-haemopoietic tissues and cells Up to this point this chapter has concentrated on islet transplantation, but the transplantation of other tissues will now be briefly considered, using two examples. These emerging fields deserve, and will doubtless one day get, their own historical review. In fact, attempted transplantation of tissues such as the parathyroids and thyroid predate attempts at islet transplantation by many years, but no repeatable or generally accepted techniques evolved. It was probably the emerging success of islet transplantation which spurred a revival of interest in grafts of other tissues [102]. Interestingly, autotransplantation of parathyroid tissue from patients undergoing total parathyroidectomy proved to be a technique that was simple and regularly successful [103]. This tissue can overcome barriers to implantation that seem insuperable for islets, for example transplantation to sites such as the forearm muscle pouch. Unfortunately it has become clear that the success of implantation is based on the fact that the tissue is semineoplastic, and a considerable number of these grafts have had to be re-explored and removed for continuous uncontrolled parathyroid secretion. Another major interest in free tissue transplantation has been transplantation of fetal medullary brain tissue or adult adrenal tissue used as treatment for Parkinson’s disease. Clinical trials followed

D.W.R.T GRAY

some 5 years after rodent and primate experiments first suggested survival and function of allogeneic and even xenogeneic tissue in the brain [104–106]. Despite some skepticism expressed regarding the initial reports of clinical transplantation of fetal neural tissue and adrenal tissue [107–109], it has become clear that at least some of the tissue survives after transplantation into the brain, both experimentally [110] and clinically [111,112]. These intracranial transplants may have advantages because the brain has special characteristics which limit the immune response, even without the use of long-term immunosuppression. Clinical trials of this technique, concentrating particularly on clinical improvement, are underway in more than one centre, and it remains to be seen if the early reports will be repeated and lead to a repeatable method and eventually a useable clinical therapy. Finally, this historical review cannot end with a true conclusion, since the story of islet transplantation and indeed cell transplantation in general is not only incomplete, it has really only just started. There are many clinicians who believe that cellular transplantation is going to revolutionize transplantation therapy and will be the key that unlocks the door to tolerance induction. That door may be about to creak open, hopefully sooner than many people think.

Note on references Most of the important contributions to a particular field of research must be credited to several sorts of individuals. The main benchwork is often undertaken by one or several relatively junior scientists, either as part of a research degree or a postdoctoral study, and naturally enough these persons are often the first authors on the papers reporting advances. These could be termed principal workers. However, behind each study there is usually a mentor, who is essentially the planner behind the project, setting the direction of research, perhaps not actually doing the work personally, but always directing the research path. Such persons are the shapers of the field, and the narrative text presented here will concentrate on these contributions. The other important contributor to be recognized is the facilitator, often a head of department, who provides the influence, environment, and obtains much of the money to allow the project to go ahead. Such individuals must have considerable long-term vision to maintain their commitment. Many individuals combine two of the above roles. Some are remarkable enough to fulfil all the above roles but usually this is not the case. The reference list will use differing fonts and styles to indicate the differing roles of the authors, as shown below.

Key to references Names in bold are often (but not always) the first authors on publications, and represent the principal workers. Names that are underlined can be considered mentors. Names in CAPITALS represent the facilitators. It will sometimes be possible to identify some individuals who take on two or even all three roles: this will be signified as in the following example: MEDAWAR (1962). Usually the reference given will be the first publication on that subject from the authors, but if the first publication is markedly inferior to a later one (perhaps because the latter is the more thorough article), this will be used instead. Occasionally a lightweight but earlier publication by one group may be deliberately overlooked in favour of a later but more substantial contribution by another group.

317

318

HISTORICAL VIEW OF THE DEVELOPMENT OF CELLULAR AND ISLET TRANSPLANTATION

A most difficult point is how the author should view the work of his own research group. The author will include publications from his own group and distinguish them by use of all authors names in italics, leaving the reader to judge the relative place of the contributors and their work.

References 1 LACY PE, Williamson JR. Quantitative histochemistry of the islets of Langerhans: II. Insulin content of dissected beta cells. Diabetes 1962;11:101–4. 2 HELLERSTROM C. A method for the microdissection of intact pancreatic islets of mammals. Acta Endocrinol 1964;45:122–32. 3 Moskalewski S. Isolation and culture of the islets of Langerhans of the guinea pig. Gen Comp Endocrinol 1965;5:342–53. 4 LACY PE, Kostianovsky M. Method for the isolation of intact islets of Langerhans from the rat pancreas. Diabetes 1967;16:35–9. 5 Junod A, Lambert AE, Stauffacher W, Renold AE. Diabetogenic action of streptozotocin: relationship of dose to metabolic response. J Clin Invest 1969;48:2129–39. 6 Junod A, Lambert AE, Orci L, Pictet R, Gonet AE, Renold AE. Studies of the diabetogenic action of streptozotocin. Proc Soc Exp Biol Med 1967;126:201–5. 7 Ballinger WF, LACY PE. Transplantation of intact pancreatic islets in rats. Surgery 1972;72:175–86. 8 SCHARP DW, Kemp CB, Knight MJ, Ballinger WF, LACY PE. The use of ficoll in the preparation of viable islets of Langerhans from the rat pancreas. Transplantation 1973;16:686–9. 9 Kemp CB, Knight MJ, Scharp DW, LACY PE, Ballinger WF. Transplantation of isolated pancreatic islets into the portal vein of diabetic rats. Nature 1973;244:447. 10 Wang Y, Hao L, Gill RG, LAFFERTY KJ. Autoimmune diabetes in NOD mouse is L3T4 T-lymphocyte dependent. Diabetes 1987;36(4):535–8. 11 Naji A, Silvers WK, Plotkin SA, Dafoe D, BARKER CF. Successful islet transplantation in spontaneous diabetes. Surgery 1979;86:218–26. 12 Coupland RE. The survival and growth of pancreatic tissue in the anterior chamber of the eye of the albino rat. J Endocrinol 1960;20:69–77. 13 Erlandsen SL, Wells LJ, LAZAROW A. Organ culture of pancreases of fetuses from normal and diabetic rats: effect of glucose on the insulin content of the media. Metabolism 1968;17:638–43. 14 Monteros MAE, Driscoll SG, Steinke J. Insulin release from isolated human fetal pancreatic islets. Science 1970;168:1111–12. 15 BROWN J, Molnar IG, Clark W, Mullen Y. Control of experimental diabetes mellitus in rats by transplantation of fetal pancreases. Science 1974;184:1377–9. 16 Mullen YS, Clark WR, Molnar IG, BROWN J. Complete reversal of experimental diabetes mellitus in rats by a single fetal pancreas. Science 1977;195:68–70. 17 Tuch BE, Osgerby KJ, Turtle JR. Normalization of blood glucose levels in nondiabetic nude mice by human fetal pancreas after induction of diabetes. Transplantation 1988;46:608–11. 18 Hegre OD, Leonard RJ, Schmitt RV, LAZAROW A. Isotransplantation of organ-cultured neonatal pancreas: reversal of alloxan diabetes in the rat. Diabetes 1976;25:180–9. 19 Rygaard J. Immunobiology of the mouse mutant ‘nude’. Acta Pathol Microbiol Scand 1969;77:761–2. 20 Bensley RR. Studies on the pancreas of the guinea pig. Am J Anat 1911;12:297–388. 21 Clark E. The number of islands of Langerhans in the human pancreas. Anat Anz 1913;43:81–94. 22 Hellman B. Actual distribution of the number and volume of the islets of Langerhans in different size classes in non-diabetic humans of varying ages. Nature 1959;184:1498–9.

D.W.R.T GRAY

23 Hansen WA, Christie MR, Kahn R, Norgaard A, Abel I, Petersen AM, Jorgensen DW, Baekkeskov S, Nielsen JH, Lernmark A, Egeberg J, Olesen HR, Grainger T, Kristensen JK, Brynitz S, Bilde T. Supravital dithizone staining in the isolation of human and rat pancreatic islets. Diabetes Res 1989;10:53–7. 24 Latif ZA, Noel J, Alejandro R. A simple method of staining fresh and cultured islets. Transplantation 1988;45:827–30. 25 Noel J, Latif Z, Alejandro R. A rapid staining method for the identification of mammalian islets of Langerhans. Diabetes 1987;36(Suppl.1):69. 26 Andersson A. Tissue culture of isolated pancreatic islets. Acta Endocrinol 1976;205(Suppl.1):283–94. 27 Andersson A, HELLERSTROM C. Metabolic characteristics of isolated pancreatic islets in tissue culture. Diabetes 1972;21(Suppl.2):546–54. 28 Gray DW, Morris PJ. The use of fluorescein diacetate and ethidium bromide as a viability stain for isolated islets of Langerhans. Stain Technol 1987;62:373–81. 29 Bank HL. Rapid assessment of islet viability with acridine orange and propidium iodide. In Vitro Cell Dev Biol 1988;24:266–73. 30 London NJ, Contractor H, Lake SP, Aucott GC, BELL PR, James RF. A microfluorometric viability assay for isolated human and rat islets of Langerhans. Diabetes Res 1989;12(3):141–9. 31 Rajotte RV, Warnock GL, Kneteman NM. Cryopreservation of insulin-producing tissue in rats and dogs. World J Surg 1984;8:179–86. 32 Rajotte RV, Stewart HL, Voss WA, Shnitka TK, Dossetor JB. Viability studies on frozen-thawed rat islets of Langerhans. Cryobiology 1977;14:116–20. 33 Rajotte RV, Scharp DW, Downing R, Preston R, Molnar GD, Ballinger WF, Greider MH. Pancreatic islet banking: the transplantation of frozen-thawed rat islets transported between centers. Cryobiology 1981;18:357–69. 34 Finke EH, LACY PE, Ono J. Use of reflected green light for specific identification of islets in vitro after collagenase isolation. Diabetes 1979;28:612–13. 35 Gray DW, Sutton R, McShane P, Peters M, Morris PJ. Exocrine contamination impairs implantation of pancreatic islets transplanted beneath the kidney capsule. J Surg Res 1988;45:432–42. 36 Gray DWR, Cranston D, McShane P, Sutton R, Morris PJ. The effect of hyperglycaemia on pancreatic islets transplanted into rats beneath the kidney capsule. Diabetologia 1989;32:663–7. 37 Ricordi C, Gray DWR, Hering BJ, Kaufman DB, Warnock GL, Kneteman NM, Lake SP, London NJM, Socci C, Alejandro R. Islet isolation assessment in man and large animals. Acta Diabetol Lat 1990;27:185–95. 38 Grant AM, Christie MR, ASHCROFT SJ. Insulin release from human pancreatic islets in vitro. Diabetologia 1980;19:114–17. 39 Harrison DE, Christie MR, Gray DW. Properties of isolated human islets of Langerhans: insulin secretion, glucose oxidation and protein phosphorylation. Diabetologia 1985;28:99–103. 40 Menger MD, Jager S, Walter P, Hammersen F, MESSMER K. A novel technique for studies on the microvasculature of transplanted islets of Langerhans in vivo. Int J Microcir Clin Exp 1990;9:103–17. 41 Jansson L. Flow distribution between the endocrine and exocrine parts of the isolated rat pancreas during perfusion in vitro with different glucose concentrations. Acta Physiol Scand 1986;126:533–8. 42 Sandler S, Jansson L. Blood flow measurements in autotransplanted pancreatic islets of the rat: impairment of the blood perfusion of the graft during hyperglycemia. J Clin Invest 1987;80:17–21. 43 Pipeleers D. The biosociology of pancreatic B cells. Diabetologia 1987;30:277–91. 44 Korsgren O, Jansson L, ANDERSSON R. Reinnervation of transplanted pancreatic islets: a comparison among islets implanted into the kidney, spleen, and liver. Transplantation 1993;56:138–43.

319

320

HISTORICAL VIEW OF THE DEVELOPMENT OF CELLULAR AND ISLET TRANSPLANTATION

45 Naji A, Reckard CR, Ziegler MM, BARKER CF. Vulnerability of pancreatic islets to immune cells and serum. Surg Forum 1975;26:459–61. 46 Slijepcevic M, Helmke K, FEDERLIN K. Islet transplantation in experimental diabetes in the rat. III. Studies in allogeneic streptozotocin-treated rats. Horm Metab Res 1975;7:456–61. 47 FEDERLIN K, Slijepcevic M, Helmke K. Islet transplantation in experimental diabetes of the rat. IV. The influence of transplantation site and of histocompatibility on islet function. Horm Metab Res 1976;8:97–101. 48 Sutherland DE. Pancreas and islet transplantation. II. Clinical trials. Diabetologia 1981;20:435–50. 49 Sutherland DE. Pancreas and islet transplantation. I. Experimental studies. Diabetologia 1981;20:161–85. 50 Gotoh M, MAKI T, Porter J, MONACO AP. Augmented survival of purified islet xeno- and allografts with reduced numbers. Transplant Proc 1987;19:984. 51 Sutton R, Gray DW, Peters M, McShane P, Dallman M, Morris PJ. Specificity of pancreatic islet allograft rejection in mixed strain rat islet transplants. Transplant Proc 1989;21:2680–1. 52 LAFFERTY KJ, Woolnough J. The origin and mechanism of the allograft reaction. Immunol Rev 1977;35:231–62. 53 Prowse SJ, LAFFERTY KJ, Simeonovic CJ, Agostino M, Bowen KM, Steele EJ. The reversal of diabetes by pancreatic islet transplantation. Diabetes 1982;31(Suppl.4):30–7. 54 LACY PE, Davie JM, Finke EH. Prolongation of islet allograft survival following in vitro culture (24 degrees C) and a single injection of ALS. Science 1979;204:312–13. 55 Faustman D, Lacy P, Davie JM. Transplantation without immunosuppression. Diabetes 1982;31(Suppl.4):11–14. 56 Hardy MA, Lau H, REEMTSMA K. Prolongation of rat islet allografts with the use of ultraviolet irradiation, without immunosuppression. Transplant Proc 1984;16:865–9. 57 Simeonovic CJ, LAFFERTY KJ. Immunogenicity of isolated fetal mouse proislets. Aust J Exp Biol Med Sci 1982;60:391–5. 58 Simeonovic CJ, LAFFERTY KJ. Immunogenicity of mouse fetal pancreas and proislets: a comparison. Transplantation 1988;45:824–7. 59 Burkhardt K, Charlton B, Mandel TE. An increase in the survival of murine H-2-mismatched cultured fetal pancreas allografts using depleting or nondepleting anti-CD4 monoclonal antibodies, and a further increase with the addition of cyclosporine. Transplantation 1989;47:771–5. 60 Selawry HP, Whittington K. Extended allograft survival of islets grafted into intra-abdominally placed testis. Diabetes 1984;33:405–6. 61 Selawry HP, Cameron DF. Sertoli cell-enriched fractions in successful islet cell transplantation. Cell Transplant 1993;2(2):123–9. 62 Bellgrau D, Gold D, Selawry H, Moore J, Franzusoff A, Duke RC. A role for CD95 ligand in preventing graft rejection (see comments) (published erratum appears in Nature 1998;394(6689):133). Nature 1995;377(6550):630–2. 63 Posselt AM, BARKER CF, Tomaszewski JE, Markmann JF, Choti MA, NAJI A. Induction of donorspecific unresponsiveness by intrathymic islet transplantation. Science 1990;249:1293–5. 64 Campos L, Posselt AM, Deli BC, Mayo GL, Pete K, BARKER CF, NAJI A. The failure of intrathymic transplantation of nonimmunogenic islet allografts to promote induction of donor-specific unresponsiveness. Transplantation 1994;57:950–3. 65 Algire GH, Weaver JM, Prehn RT. Growth of cells in vivo in diffusion chambers. I. Survival of homografts in immunized mice. J Natl Cancer Inst 1954;15:493–507. 66 SUN AM, Shea GMO, Goosen MF. Injectable microencapsulated islet cells as a bioartificial pancreas. Appl Biochem Biotechnol 1984;10:87–99.

D.W.R.T GRAY

67 Loudovaris T, MANDEL TE, Charlton B. CD4+ T cell mediated destruction of xenografts within cellimpermeable membranes in the absence of CD8+ T cells and B cells. Transplantation 1996;61(12):1678–84. 68 Mirkovitch V, Campiche M. Absence of diabetes in dogs after total pancreatectomy and intrasplenic autotransplantation of pancreatic tissue. Transplant Proc 1977;9:321–3. 69 Evans MG, Warnock GL, RAJOTTE RV. Comparison of sites for transplantation of canine pancreatic microfragments. Diabetes Res 1989;10:35–41. 70 Horaguchi A, MERRELL RC. Preparation of viable islet cells from dogs by a new method. Diabetes 1981;30:455–8. 71 Noel J, Rabinovitch A, Olson L, Kyriakides G, MILLER J, MINTZ DH. A method for large-scale highyield isolation of canine pancreatic islets of Langerhans. Metabolism 1982;31:184–7. 72 Alejandro E, Cutfield RG, Shienvold FL, Polonsky KS, Noel J, Olson L, Dillberger J, MILLER J, MINTZ DH. Natural history of intrahepatic canine islet cell autografts. J Clin Invest 1986;78:1339–48. 73 Gray DWR, Warnock GL, Sutton R, Peters M, McShane P, Morris PJ. Successful autotransplantation of isolated islets of Langerhans in the cynomolgus monkey. Br J Surg 1986;73:850–3. 74 Sutton R, Gray DW, Burnett M, McShane P, Turner RC, Morris PJ. Metabolic function of intraportal and intrasplenic islet autografts in cynomolgus monkeys. Diabetes 1989;38(Suppl.1):182–4. 75 Leow CK, Shimizu S, Gray DWR, Morris PJ. Successful pancreatic islet autotransplantation to the renal subcapsule in the cynomolgus monkey. Transplantation 1994;57:161–4. 76 Gray DWR, McShane P, Grant A, Morris PJ. A method for isolation of islets of Langerhans from the human pancreas. Diabetes 1984;33:1055–61. 77 Scharp DW, Murphy JJ, Newton WT, Ballinger WF, LACY PE. Transplantation of islets of Langerhans in diabetic rhesus monkeys. Surgery 1975;77:100–5. 78 Ricordi C, Finke EH, LACY PE. A method for the mass isolation of islets from the adult pig pancreas. Diabetes 1986;35:649–53. 79 Marchetti P, Finke EH, Vazeou AG, Falqui L, Scharp DW, LACY PE. Automated large-scale isolation, in vitro function and xenotransplantation of porcine islets of Langerhans. Transplantation 1991;52:209–13. 80 Alejandro R, Cutfield R, Shienvold FL, Latif Z, MINTZ DH. Successful long-term survival of pancreatic islet allografts in spontaneous or pancreatectomy-induced diabetes in dogs: Cyclosporineinduced immune unresponsiveness. Diabetes 1985;34:825–8. 81 Cattrall MS, Warnock GL, Kneteman NM, RAJOTTE RV. Transplantation of purified single-donor canine islet allografts with cyclosporine. Transplantation 1989;47:583–7. 82 Warnock GL, RAJOTTE RV. Critical mass of purified islets that induce normoglycemia after implantation into dogs. Diabetes 1988;37:467–70. 83 Ricordi C, LACY PE, Finke EH, Olack BJ, Scharp DW. Automated method for isolation of human pancreatic islets. Diabetes 1988;37:413–20. 84 Lake SP, Bassett PD, Larkins A, Revell J, Walczak K, Chamberlain J, Rumford GM, London NJM, Veitch PS, BELL PRF, James RFL. Large-scale purification of human islets utilizing discontinuous albumin gradient on IBM 2991 cell separator. Diabetes 1989;38(Suppl.1):143–5. 85 Scharp DW, LACY PE, Santiago JV, McCullough CS, Weide LG, Falqui L, Marchetti P, Gingerich RL, Jaffe AS, Cryer PE. Insulin independence after islet transplantation into type I diabetic patient. Diabetes 1990;39:515–18. 86 Warnock GL, Kneteman NM, Ryan EA, Evans MG, Seelis RE, Halloran PF, Rabinovitch A, RAJOTTE RV. Continued function of pancreatic islets after transplantation in type I diabetes. Lancet 1989;2:570–2.

321

322

HISTORICAL VIEW OF THE DEVELOPMENT OF CELLULAR AND ISLET TRANSPLANTATION

87 Socci C, Falqui L, Davalli AM, Ricordi C, Braghi S, Bertuzzi F, Maffi P, Secchi A, Gavazzi F, Freschi M, Magistretti P, Socci S, Vignali A, Carlo VD, POZZA G. Fresh human islet transplantation to replace pancreatic endocrine function in type 1 diabetic patients: report of six cases. Acta Diabetol 1991;28:151–7. 88 BRETZEL RG, Brandhorst D, Brandhorst H, Eckhard M, Ernst W, Friemann S, Rau W, Weimar B, Rauber K, Hering BJ, Brendel M. Improved survival of intraportal pancreatic islet cell allografts in patients with type-1 diabetes mellitus by refined peritransplant management. J Mol Med 1999;77(1):140–3. 89 Ricordi C, Tzakis AG, Carroll PB, Zeng Y, Rilo HLR, Alejandro R, Shapiro R, Fung JJ, Demetris AJ, Mintz DH, STARZL TE. Human islet isolation and allotransplantation in 22 consecutive cases. Transplantation 1992;53:407–14. 90 Warnock GL, KNETEMAN NM, Ryan EA, Rabinovitch A, RAJOTTE RV. Long-term follow-up after transplantation of insulin-producing pancreatic islets into patients with type 1 (insulin-dependent) diabetes mellitus. Diabetologia 1992;35:89–95. 91 Lakey JR, Warnock GL, RAJOTTE RV, Suarez Alamazor ME, Ao Z, Shapiro AM, KNETEMAN NM. Variables in organ donors that affect the recovery of human islets of Langerhans. Transplantation 1996;61(7):1047–53. 92 Bretzel RG, Hering BJ, Federlin KF. Islet cell transplantation in diabetes mellitus — from bench to bedside. Exp Clin Endocrinol Diabetes 1995;103(Suppl.2):143–59. 93 Secchi A, Socci C, Maffi P, Taglietti MV, Falqui L, Bertuzzi F, De Nittis P, Piemonti L, Scopsi L, Di Carlo V, Pozza G. Islet transplantation in IDDM patients. Diabetologia 1997;40(2):225–31. 94 Luzi L, Hering BJ, Socci C, Raptis G, Battezzati A, Terruzzi I, Falqui L, Brandhorst H, Brandhorst D, Regalia E, Brambilla E, Secchi A, Perseghin G, Maffi P, Bianchi E, Mazzaferro V, Gennari L, Di Carlo V, FEDERLIN K, POZZA G, Bretzel RG. Metabolic effects of successful intraportal islet transplantation in insulin-dependent diabetes mellitus. J Clin Invest 1996;97(11):2611–8. 95 Sutherland DE. Pancreas and islet transplant registry statistics. Transplant Proc 1984;16:593–8. 96 Hering BJ, Browatzki CC, Schultz AO, Bretzel RG, FEDERLIN K. Islet Transplant Registry report on adult and fetal islet allografts. Transplant Proc 1994;26(2):565–8. 97 Najarian JS, Sutherland DE, Baumgartner D, Burke B, Rynasiewicz J, Matas AJ, Goetz FC. Total or near total pancreatectomy and islet transplantation for treatment of chronic pancreatitis. Ann Surg 1980;192:526–42. 98 Farney AC, NAJARIAN JS, Nakhleh RE, Lloveras G, Field MJ, Gores PF, SUTHERLAND DER. Autotransplantation of dispersed pancreatic islet tissue combined with total or near-total pancreatectomy for treatment of chronic pancreatitis. Surgery 1991;110:427–39. 99 Hamelmann W, Gray DW, Cairns TD, Ozasa T, Ferguson DJ, Cahill A, Welsh KI, Morris PJ. Immediate destruction of xenogeneic islets in a primate model. Transplantation 1994;58(10):1109–14. 100 Weber C, Weil RD, McIntosh R, Hogle H, Warden G, REEMTSMA K. Xenotransplantation of piscine islets into hyperglycemic rats. Surgery 1975;77(2):208–15. 101 Wallgren AC, Karlsson Parra A, Korsgren O. The main infiltrating cell in xenograft rejection is a CD4+ macrophage and not a T lymphocyte. Transplantation 1995;60(6):594–601. 102 Usadel KH, Bastert G, Schwedes U, Obert I, Fortmeyer HP, Schoffling K. Human fetal pancreas transplants in nu/nu mice. Lancet 1977;1:365. 103 Fisher B, Fisher ER, Feduska N, Sakai A. Thyroid and parathyroid implantation: an experimental re-evaluation. Surgery 1967;62(6):1025–38. 104 Wells SA, Jr., Gunnells JC, Shelburne JD, Schneider AB, Sherwood LM. Transplantation of the parathyroid glands in man: clinical indications and results. Surgery 1975;78(1):34–44. 105 Bjorklund A, Stenevi U, Dunnett SB, Gage FH. Cross-species neural grafting in a rat model of Parkinson’s disease. Nature 1982;298(5875):652–4.

D.W.R.T GRAY

106 Freed WJ, Cannon Spoor HE, Wyatt RJ. Embryonic brain grafts in an animal model of Parkinson’s disease. Criteria for human application. Appl Neurophysiol 1984;47(1–2):16–22. 107 Morihisa JM, Nakamura RK, Freed WJ, Mishkin M, Wyatt RJ. Adrenal medulla grafts survive and exhibit catecholamine-specific fluorescence in the primate brain. Exp Neurol 1984;84(3):643–53. 108 Lindvall O, Backlund EO, Farde L, Sedvall G, Freedman R, Hoffer B, Nobin A, Seiger A, Olson L. Transplantation in Parkinson’s disease: two cases of adrenal medullary grafts to the putamen. Ann Neurol 1987;22(4):457–68. 109 Madrazo I, Drucker CR, Diaz V, Martinez MJ, Torres C, Becerril JJ. Open microsurgical autograft of adrenal medulla to the right caudate nucleus in two patients with intractable Parkinson’s disease. N Engl J Med 1987;316(14):831–4. 110 Madrazo I, Leon V, Torres C, Aguilera MC, Varela G, Alvarez F, Fraga A, Drucker Colin R, Ostrosky F, Skurovich M, et al. Transplantation of fetal substantia nigra and adrenal medulla to the caudate nucleus in two patients with Parkinson’s disease (letter). N Engl J Med 1988;318(1):51. 111 Bankiewicz KL, Plunkett RJ, Jacobowitz DM, Porrino L, di Porzio U, London WT, Kopin IJ, Oldfield EH. The effect of fetal mesencephalon implants on primate MPTP-induced parkinsonism. Histochemical and behavioral studies (see comments). J Neurosurg 1990;72(2):231–44. 112 Lindvall O, Brundin P, Widner H, Rehncrona S, Gustavii B, Frackowiak R, Leenders KL, Sawle G, Rothwell JC, Marsden CD, et al. Grafts of fetal dopamine neurons survive and improve motor function in Parkinson’s disease (see comments). Science 1990;247(4942):574–7.

323

This page intentionally left blank

Chapter 23

Fetal and neonatal pancreatic tissue transplantation Bernard E. Tuch

Human allografts The first record of transplanting fetal pancreatic tissue into humans was by Fischera in 1928. He allografted the tissue in three sites, beneath the tunica vaginalis, into muscle, and preperitoneally (reported in [1]). More systematic transplants of fetal pancreas did not occur until six decades later [2–4]. The systematic desire to examine the use of immature pancreatic tissue occurred in the 1970s at a time when all types of pancreatic tissue were being examined as possible sources for replacing the ␤-cells destroyed in type 1 diabetes. Whole pancreas, islets, and fetal tissue were identified as possible sources of viable ␤-cells. At the dawn of the new millennium, the role of the whole pancreas as a therapy for type 1 diabetes is known [5]. Problems that had to be overcome to achieve this goal have included optimizing methods to retrieve the pancreas from brain-dead donors, which immunosuppressive drugs are needed to prevent rejection, and the best site to transplant the pancreas. These issues are discussed elsewhere in this book. Whole pancreas transplants are reserved for those people with advanced complications, especially those with renal failure requiring a kidney transplant. Success rates are on a par with other whole organ transplants. The percentage of grafts still functioning 1 year after being transplanted is 83 per cent [5]. Adult islets are capable of normalizing blood glucose levels when allografted into humans but there is a considerable way to go before the place of this form of transplantation is known. Of the 353 type 1 diabetic recipients of this tissue, 33 became insulin-independent with 20 maintaining this state for 1 year [6] and four for 3 years [7]. Further trials using alternatives to conventional immunosuppressive drug treatments are underway and these offer hope for improved outcomes. These trials include monoclonal antibody therapy on a transient basis [8,9] and the combination of rapamycin, FK506 and interleukin 2 (IL-2) receptor antibodies [10] to prevent rejection. Transplantation of fetal human pancreas into diabetic humans has yet to result in normalization of blood glucose levels without the need to inject supplementary insulin. There are claims that this goal has been achieved but there are no reports in refereed journals supporting them. Insulin production from the grafted tissue has been achieved in recipients who were previously C-peptide negative [2,11], that is, produced no endogenous insulin, with reduction in exogenous insulin requirements in one series [11]. C-peptide is a biologically inactive hormone formed by cleavage of proinsulin, and is secreted from the ␤-cell on an equimolar basis with insulin. Consistent with the evidence of function of allografted fetal pancreatic tissue is the histological survival of transplanted tissue up to 1 year after transplantation, with best results obtained in the recipient grafted with six fetal pancreases [3]. The number of surviving ␤-cells was small suggesting that either larger numbers of fetal pancreases will

326

FETAL AND NEONATAL PANCREATIC TISSUE TRANSPLANTATION

need to be transplanted and/or better methods are required to prevent rejection of the graft. In Western countries supplies of human fetal pancreas during the second trimester are not generally available in numbers that would encourage further allografts with this tissue.

Xenografts of fetal and neonatal pancreas into humans There is a relative lack of human tissues and organs for allografting into humans. This has resulted in long waiting times for organs, for example, the average waiting time for a pancreas in Australia is 1 to 2 years and in the United States 6 months. This need has resulted in serious consideration by the transplant community of the use of animal organs and tissues. The pig is the animal most favoured as a donor. Non-human primates are less generally acceptable [12] because of the greater fear of infection. Pigs are in plentiful supply, and have a number of physiological and anatomical similarities to humans [13,14]. Porcine insulin, for example, has the same sequence as human insulin except for one amino acid difference. Islet-like cell clusters prepared from fetal [15] and neonatal pigs [16] have been transplanted into humans. Between 1991 and 1994, 10 people with type 1 diabetes and renal failure were transplanted with fetal pancreatic cells, either injected into the liver or placed beneath the renal capsule [15]. These recipients were immunosuppressed with conventional antirejection drugs, namely ciclosporin, azathioprine, prednisone, antilymphocyte globulin, and deoxyspergualin. The grafts functioned for many months, as shown by the presence of porcine C-peptide in urine 7 to 13 months after the transplant. No porcine C-peptide was detected in blood and there was no reduction in exogenous insulin requirements. Renal biopsy of one recipient showed the presence of ␤-cells and other porcine endocrine cells. In two diabetic recipients of neonatal porcine islet-like cell clusters, porcine C-peptide was detected in urine for up to 14 months after the cells were grafted in the mid 1990s [16]. One of the recipients was taking immunosuppressive drugs to prevent rejection of a previously transplanted kidney; the other was not immunosuppressed. In both recipients the cells were placed inside microcapsules to prevent them from being attacked by the body’s immune system. The recent fear that pig endogenous retroviruses (PERV) might cause a pandemic has delayed further human trials using fetal and neonatal pig pancreatic tissue; PERV constitutes 1 per cent of the pig genome just as human endogenous retroviruses constitute a similar percentage of the human genome. Experiments carried out in the laboratory of Robin Weiss in London showed that a human cell line could be infected by a PERV when cocultured with a pig cell line [17]. This prompted the warning that pig tissue when transplanted might result in PERV being transmitted to the human genome, and cause active infection in the recipient. Another retrovirus, the human immunodeficiency virus, which was probably transmitted to humans from the monkey, had already caused a major public health problem being transmitted from one human to another. So, the spectre of transmitting PERV from the recipient to his/her partner and thence to the general public was raised. As a consequence, the United Kingdom but not the United States placed a moratorium on xenotransplantation in early 1997. Two years later the moratorium was lifted. No long-term active infection from these viruses was observed in either humans [18] or rodents transplanted with pig tissue [19,20]. This was despite migration of pig cells containing PERV from the site of transplantation to other tissues of the recipient.

Types of fetal and neonatal pancreatic tissue transplanted Human and porcine fetal pancreatic tissue are transplanted in one of four forms. 1. Explants 1 mm3 created by dicing the fetal pancreas.

BERNARD E. TUCH

Fig. 23.1 Islet-like cell clusters (ICCs) obtained by collagenase digestion of pig fetal pancreas of gestational age 80 days. The ICCs have rounded up after 3 days in culture. Beta-cells are few in number and can be identified by the dark stain (bar 50 ␮m).

2. Islet-like cell clusters (ICC), which are formed by partial digestion of the pancreas with the enzyme, collagenase or liberase [21,22]. These clusters of cells round up during the 1 to 3 days they are kept in culture after their creation [23]. The most common cell type in these clusters is the precursor duct cell [24] from which endocrine and exocrine cells develop. ␤-cells constitute 6 to 9 per cent of cells (Fig. 23.1) in contrast to the 60 to 70 per cent of cells present in adult islets. Once the ICCs are transplanted, the precursor cells differentiate mostly into ␤-cells. It is for this reason that ICCs are also called proislets. 3. Fetal pancreatic cell (re)aggregates. Complete digestion of the pancreas with collagenase/liberase will produce single cells, which can be purified by gradient centrifugation [25] or expanded in tissue culture [26]. Reaggregation of these cells can be achieved by gently shaking the cells for 1 or more hours. The aggregates are then ready to be transplanted. 4. Islets. There are small number of islets in the human fetal pancreas [27] and none in the pig fetal pancreas. Islets differ from ICCs in having a much greater percentage of ␤-cells. Regardless of which of the above four forms fetal pancreas is transplanted, the tissue becomes vascularized and the majority of the grafted cells eventually develop into ␤-cells (Fig. 23.2). If single fetal pancreatic cells are transplanted alone, they do not survive. Unlike most other primary cells, pancreatic cells need to be clustered together to develop, suggesting an important role for adhesion molecules and extracellular matrix [26] such as laminin [28]. Neonatal pancreatic tissue has been transplanted only as ICCs [29] or islets [30]. The large amount of exocrine tissue prevents use of explants since the enzymes, especially proteases, produced by dicing the pancreas are toxic to the cells.

Efficacy of transplanting fetal and neonatal pancreatic tissue in diabetic animals Both fetal and neonatal pancreatic tissue is capable of normalizing blood glucose levels of diabetic recipient animals. This has been achieved with the following pancreatic tissue types.

327

328

FETAL AND NEONATAL PANCREATIC TISSUE TRANSPLANTATION

Fig. 23.2 Pig fetal pancreatic explants of 83 days gestational age 16 weeks after being transplanted beneath the renal capsule of a diabetic SCID mouse. The majority of cells in the graft are ␤-cells (dark stain) (the bar, located in the renal cortex, is 50 ␮m).

1. Human fetal pancreatic tissue transplanted into the immunoincompetent mouse [31,32] and immunoincompetent rat [33]. 2. Pig fetal pancreatic tissue transplanted into the immunoincompetent mouse [22,34–36]; normal mouse immunosuppressed with anti-CD4 antibodies [37]; the non-obese diabetic mouse, that naturally develop type 1 diabetes, immunosuppressed with anti-CD4 and anti-CD3 antibodies [38]; the immunoincompetent rat [39]; and the pig immunosuppressed with ciclosporin and deoxyspergualin [40]. 3. Pig neonatal pancreatic tissue transplanted into the immunoincompetent mouse [29,41], and in microcapsules into the non-obese diabetic mouse [16]. 4. Rat fetal pancreatic tissue transplanted into a syngeneic recipient [42–44]. 5. Rat neonatal pancreatic tissue transplanted into a syngeneic recipient [30]. 6. Mouse fetal pancreatic tissue transplanted into a syngeneic recipient [45]. Experiments to demonstrate the efficacy of fetal pancreatic tissue when transplanted into large animals are continuing. Pig fetal pancreatic tissue has been transplanted into normoglycaemic nonhuman primates, with survival of the tissue for at least 1 month when the animals were immunosuppressed with ciclosporin, cyclophosphamide, and prednisone [46], but not when deoxyspergualin and ciclosporin were used [47]. Pig fetal pancreatic tissue has yet to be transplanted into diabetic nonhuman primates, although adult pig islets have, with normalization of blood glucose levels [48].

Fetal pancreatic tissue from ruminants If fetal pancreatic tissue is to be used as a treatment of type 1 diabetes, supplies of tissue will need to be considerable. Estimates of people with this disorder in the world are between 3.5 and 6 million. A large animal available in plentiful numbers is the logical source of donor tissue, with pigs the most likely source. However, there are large numbers of other domestic animals available, especially the

BERNARD E. TUCH

sheep and the cow. The major difference between these animals and the pig is that they are ruminants and ferment food in the foregut. Pigs like humans ferment food in the hindgut. Fetal sheep but not bovine pancreas has been transplanted into diabetic immunoincompetent mice with no effect on the elevated blood glucose levels [49]. This is for two reasons. Firstly, the ␣-cell, which secretes glucagon rather than the ␤-cell is the dominant endocrine cell in the transplanted tissue. Secondly, fatty acids rather than glucose are the major stimulus for secretion of insulin from the adult sheep ␤-cell [50]. Even if the fetal sheep pancreas was effective in the diabetic mouse, it is likely that the resulting blood glucose levels would be too low for the mouse. The blood glucose level of the fetal sheep is 1.5 mmol/l and that of the adult sheep, 3 mmol/l, in contrast to the higher levels of 5 to 8 mmol/l for the mouse. Blood glucose levels that result from ␤-cells being transplanted are those of the donor species and not those of the recipient [51]. Thus, the ruminant is not a suitable animal to source pancreatic tissue.

Time required to normalize blood glucose levels Fetal and neonatal pancreatic tissue takes 1 to 5 months to normalize blood glucose levels when transplanted (Fig. 23.3). In contrast, adult ␤-cells achieve their effect within days of being grafted. The reason for the delay with fetal and neonatal pancreatic tissue is to allow the precursor cells to differentiate into ␤-cells, and for these ␤-cells to become responsive to glucose. There are four factors that influence how long fetal and neonatal pancreatic tissue will take to normalize blood glucose levels. 1. The amount of tissue transplanted. The greater the amount, the shorter the time required [29,41,52]. Thus, 1000 ICCs take longer than 2000, and 2000 longer than 4000. This suggests that a critical amount of insulin needs to be produced for blood glucose levels to be normalized. 2. The gestational age of the tissue transplanted. The older the tissue, the shorter the time to normoglycaemia [52]. This applies to fetal rather than neonatal pancreatic tissue. 3. The growth rate of the tissue. The faster the tissue grows, the quicker normoglycaemia will be achieved [52]. 4. Presence of nicotinamide in the culture medium in which ICCs form. Nicotinamide acts to enhance the differentiation of ␤-cells [53]. Fetal pig ICCs normalized blood glucose levels in half the time if they are cultured for 3 days in the presence of 10 mmol nicotinamide [54].

Monitoring the function of transplanted immature fetal ␤-cells The fetal ␤-cell is able to secrete insulin when exposed to agents that increase levels of cyclic adenosine monophosphate (AMP) and calcium [55], and activate protein kinase C [56], but not to glucose [57,58]. Because of the immaturity of the fetal ␤-cell, it is not possible to monitor its function immediately after it is transplanted by measuring levels of insulin or C-peptide. The viability of fetal pancreas can, however, be monitored during this time by measuring the levels of another pancreatic hormone, pancreatic polypeptide (PP). The PP cell, especially in response to cholinergic stimuli, secretes this hormone. Levels of PP are maximal during the first 3 weeks after transplantation and decline thereafter as the number of PP cells in the graft diminish [59]. Levels of this hormone fall earlier than this if rejection of the graft commences.

329

330

FETAL AND NEONATAL PANCREATIC TISSUE TRANSPLANTATION

Fig. 23.3 Blood glucose levels of a diabetic SCID mouse transplanted with explants of pig fetal pancreas obtained at 73 days gestation. The time taken for normalization of blood glucose levels was 5 months. Note the weight of the diabetic animal gradually increases before the blood glucose levels are normalized. Removal of the graft results in recurrence of hyperglycaemia.

Between 1 and 5 months after the fetal pancreas is grafted into a diabetic recipient, blood glucose levels are normalized. At this time the ␤-cell is still not fully mature especially in its ability to secrete insulin in response to glucose [22,32,40]. The first phase of insulin secretion, which occurs during the first 10 min after exposure to glucose, is absent. Only the second phase, which happens after the 10 min, occurs. Several months later the first phase of insulin secretion is established [22]. In contrast to the delay in response to glucose, the transplanted fetal ␤-cell is able to secrete insulin in response to glucagon, which acts by increasing levels of cyclic AMP, and arginine [40].

BERNARD E. TUCH

Differentiation of fetal ␤-cells The fetal pancreas develops from the endoderm. The most primitive pancreatic cell is undifferentiated and under the influence of different transcription factors will develop into either endocrine or exocrine cells. Neur Do and other transcription factors are required for endocrine [60] and pancreatic transcription factor 1-p48 for exocrine development [61]. Development of endocrine tissue commences before exocrine tissue. In humans, cells containing hormones including insulin are present from 8 weeks of fetal life [62], while cells containing exocrine enzymes are not present until 12 weeks or later [63]. Beta and other endocrine cells form from undifferentiated cells and initially remain in their midst together with developing exocrine cells. This mixture of the three cell types is apparent in both fetal pancreatic explants and ICCs. In the human fetal pancreas obtained early in the second trimester, 29 per cent of cells are undifferentiated, 48 per cent contain exocrine enzymes, and 16 per cent hormones (7 per cent insulin) [64]. In ICCs formed from the fetal pig pancreas in late gestation, 36 per cent of the cells are undifferentiated, 32 per cent contain exocrine enzymes, and 13 per cent insulin or glucagon [24]. Beta and other endocrine cells eventually form buds on the periphery of the mixed cell clusters and break away to coalesce and form islets [65]. After transplantation the fetal pancreatic tissue differentiates and proliferates (Fig. 23.4). This is best observed when tissue from species with a long period of gestation, such as humans and pigs, is transplanted [27,64,66]. Undifferentiated cells develop into endocrine cells, especially ␤-cells, with

Fig. 23.4 Human fetal pancreatic tissue of 13 weeks gestational age 97 weeks after being transplanted beneath the renal capsule of a nude mouse. The graft weighed 356 mg, as compared to 16 mg when transplanted. It has been removed from the kidney, which can be seen adjacent to it. On the other side of the kidney is the pancreas of the mouse.

331

332

FETAL AND NEONATAL PANCREATIC TISSUE TRANSPLANTATION

islets forming [31,67], and the percentage of ␤-cells in the graft increases [68,69]. Few ␤-cells proliferate although undifferentiated cells do [64] as might be expected. Exocrine cells diminish in number at least partly because of dedifferentiation into precursor cells [64]. The rate of formation of ␤-cells varies depending on the gestational age of the tissue. When tissue is obtained at 40 per cent of the gestational period and transplanted, it takes more than 3 months for ␤-cells to be the predominant cell in the graft [64,68]. The time required for ␤-cell dominance is 1 to 2 months when pancreatic tissue is obtained between 75 per cent of the gestational period [22,70] and 3 days after birth [41]. Associated with this histological differentiation is biochemical maturation of the ␤-cells in the graft with glucoseinduced insulin secretion [71].

Rejection of fetal pancreatic tissue Fetal tissue is thought to be less immunogenic than tissue derived from an adult. This is true for tissue taken at a very early stage of gestation [72], but it is not so by the time the pancreas is of a size that makes it available for transplantation [73]. In humans this is after 14 weeks gestation (length of gestation is 40 weeks), and in mice after 18 days (length of gestation is 21 days). Indeed the immunogenicity of fetal pancreatic tissue is equal to if not more than that of adult tissue [73]. Rejection of allografted fetal pancreatic tissue occurs by the same process as that which causes rejection of allografted adult islets. It is a cellular process involving T lymphocytes, especially of the CD4 variety, and commences 4 or more days after the tissue is grafted. Conventional antirejection drugs [3,70] and monoclonal antibodies to CD4 lymphocytes [74] will delay or prevent this. Treatment of the tissue prior to transplantation, for example, by culturing it in the presence of high concentrations of oxygen, will reduce tissue antigenicity by killing antigen-presenting cells. This will improve survival and function of the tissue when grafted [75]. The process that rejects xenografted adult islets also destroys xenografted fetal pancreatic tissue. As with allografted tissue, the main form of rejection is cellular and this begins 4 or more days after the tissue is transplanted. The types of immune cell involved in rejection are CD8 lymphocytes and macrophages [76]. Since the cells are created in response to the transplant, they are part of a cognate immune response. Rejection can be prevented with conventional antirejection drugs [77] and monoclonal antibodies to CD4 [37] and CD3 [78]. Cellular rejection of a different form might also occur. This begins during the 24 h after tissue is transplanted, and is due to an innate response from natural killer cells and macrophages [76]. Xenografted organs are also subject to hyperacute rejection by preformed antibodies that react against an antigen on endothelial cells, called the galactosyl ␣-1,3-galactose epitope. This effect occurs within minutes of tissue being transplanted and results in loss of blood supply to the organ and its death within the hour. Fetal pancreatic tissue, like adult islets, may not be rejected in this manner [46,47,79] perhaps because they have few endothelial cells. However, the galactosyl ␣-1,3-galactose epitope is present on endocrine and non-endocrine cells of the pancreas [80,81], but perhaps in too little amount to cause a significant problem. If so, normal pigs rather than the transgenic pigs being bred to overcome hyperacute rejection will be adequate as a donor source of fetal pancreatic tissue. Autoimmune rejection of ␤-cells is responsible for the development of type 1 diabetes. Reactivation of this process can cause rejection of human allografts of adult islets [82]. Fetal pancreatic tissue may not be affected by this process. Fetal ␤-cells are resistant to the cytokines involved in the autoimmune process [33,83]. Further, when fetal pancreatic tissue is transplanted into mice that naturally develop type 1 diabetes, they are rejected by a process different from that which destroys syngeneically transplanted islets. Both the type of immune cells and the cytokines expressed differed between the two groups [84]. Finally, fetal ␤-cells, which express the antigen Fas, are resistant to destruction from Fas ligand, which is expressed on autoimmune cells that attack and kill pancreatic ␤-cells.[85]

BERNARD E. TUCH

Summary The use of fetal and neonatal pancreatic tissue to normalize blood glucose levels has a number of theoretical advantages over the use of adult islets for this purpose. 1. Ability to proliferate. Fetal cells have a higher proliferative capacity than do adult cells. 2. Longer lifespan. The genetic machinery in a fetal cell programmes this cell for a longer life than an adult cell. It has been suggested that the lifespan of Dolly the sheep, recently cloned using adult somatic DNA, may be less than that of a sheep bred by natural means. 3. Possible resistance to autoimmune rejection. 4. Technically easier to produce, at least as regards porcine pancreatic tissue. There are a number of relative disadvantages of using fetal and neonatal pancreatic tissue compared to adult islets. 1. Immaturity of the ␤-cell. It takes several months for the tissue to mature in its ability to secrete insulin in response to physiological stimuli, especially glucose. 2. Small size of the fetal pancreas. The pancreatic tissue proliferates after it is transplanted with the size of the tissue increasing over a period of months. The actual number of ␤-cells also increases as they develop from undifferentiated precursor cells of the pancreas. Fetal and neonatal pancreatic tissue has the ability to reverse diabetes when transplanted into diabetic recipients. Whether it will be useful for this purpose in diabetic humans remains to be established. Further human trials with fetal and neonatal pancreatic tissue are warranted in the immediate future. In the longer term, consideration also will need to be given to the use of insulin-secreting cells derived from embryonic stem cells. Protocols for developing these ␤-cells have been developed in mice and transplantation of them into diabetic mice results in normalization of blood glucose levels [86]. Protocols for developing insulin-secreting cells from human embryonic precursors will require passage of appropriate legislation by governments.

Acknowledgements I want to thank Dr Joe Ciccotosto for critically reviewing this chapter. Support from the following organizations is gratefully acknowledged: National Health and Medical Research Council of Australia, Australian Research Council Collaborative Grant, Juvenile Diabetes Foundation International, Rebecca L. Cooper Medical Research Foundation, Diabetes Australia Research Trust, Autogen Research Pty Ltd, Bunge Meat Industries Ltd, and Amrad Pharmaceuticals Pty Ltd.

References 1 Downing R. Historical review of pancreatic islet transplantation. World J Surg 1984;8:137–42. 2 Groth GC, Andersson A, Björken C, Gunnarsson R, Hellerström C, Lundgren G, et al. Attempts at transplantation of fetal pancreas to diabetic patients. Transplant Proc 1980;12(Suppl.2):208–12.

333

334

FETAL AND NEONATAL PANCREATIC TISSUE TRANSPLANTATION

3 Tuch BE, Sheil ARG, Ng ABP, Trent RJ, Turtle JR. Recovery of human fetal pancreas after one year of implantation in the diabetic patient. Transplantation 1988;46:865–70. 4 Voss F, Brewin A, Dawidson I, Lafferty K, Spees E, Collins G, et al. Transplantation of proliferated human pre-islets into diabetic patients with renal transplants. Transplant Proc 1989;21:2751–6. 5 Bland BJ (ed). International Pancreas Transplant Registry Newsletter. Minneapolis: University of Minnesota, 2001:13. 6 Brendel MD, Hering BJ, Schultz AO, Bretzel RG. International Islet Transplant Registry. Giessen, Germany: University Hospital, 2001:8. 7 Hering BJ, Ricordi C. Islet transplantation for patients with type 1 diabetes. Graft 1999;2:12–27. 8 Kenyon NS, Chatzipetrou M, Masetti M, Ranuncoli A, Oliveira M, Wagner JL, et al. Long-term survival and function of intrahepatic islet allografts in rhesus monkeys treated with humanized anti-CD154. Proc Natl Acad Sci USA 1999;96:8132–7. 9 Kenyon NS, Fernandez LA, Lehmann R, Masetti M, Ranuncoli A, Chatzipetrou M, et al. Long-term survival and function of intrahepatic islet allografts in baboons treated with humanized anti-CD154. Diabetes 1999;48:1473–81. 10 Shapiro AMJ, Lakey JRT, Ryan EA, Korbutt GS, Toth E, Warnock GL, et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regime. N Engl J Med 2000;343:230–8. 11 Farkas G, Karácsonyi S. Clinical transplantation of fetal human pancreatic islets. Biomed Biochem Acta 1985;1:155–9. 12 Advisory Group on the Ethics of Xenotransplantation. Animal tissue into humans. Her Majesty’s Stationery Office, Norwich, United Kingdom, 1997. 13 Miller ER, Ullrey DE. The pig as a model for human nutrition. Ann Rev Nutr 1987;7:361–82. 14 Hannon JP, Bossone CA, Wade CE. Normal physiological values for conscious pigs used in biomedical research. Lab Animal Sci 1990;40:293–8. 15 Groth CG, Korsgren O, Tibell A, Tollemar J, Möller E, Bolinder J, et al. Transplantation of porcine fetal pancreas to diabetic patients. Lancet 1994;344:1402–4. 16 Elliott RB, Escobar L, Garkavenko O, Crozson MC, Schroeder BA, McGregor M, et al. No evidence of infection with porcine endogenous retrovirus in recipients of encapsulated porcine islet xenografts. Cell Transplant 2000;9:895–901. 17 Patience C, Takeuchi Y, Weiss RA. Infection of human cells by an endogenous retrovirus of pigs. Nature Med 1997;3:282–6. 18 Heneine W, Tibell A, Switzer WM, Sandstrom P, Rosales GV, Mathews A, et al. No evidence of infection with porcine endogenous retrovirus in recipients of porcine islet-cell xenografts. Lancet 1998;352:695–8. 19 Deng Y-M, Tuch BE, Rawlinson WD. Transmission of porcine endogenous retroviruses in severe combined immunodeficient mice xenotransplanted with fetal porcine pancreatre cells. Transplantation 2000;70:1010–6. 20 Van der Laan LJW, Lockey C, Griffeth BC, Frasier FS, Wilson CA, Onions DE, et al. Infection by porcine endogenous retrovirus after islet xenotransplantation in SCID mice. Science 2000;407:90–4. 21 Sandler S, Andersson A, Schnell A, Mellgren A, Tollemar J, Borg H, et al. Tissue culture of human fetal pancreas: development and function of ␤-cells in vitro and transplantation of explants to nude mice. Diabetes 1985;34:1113–9. 22 Korsgren O, Jansson L, Eizirik D, Andersson A. Functional and morphological differentiation of fetal porcine islet-like cell clusters after transplantation into nude mice. Diabetologia 1991;34:379–86. 23 Tuch BE, Simpson AM, Smith MNS, Waugh P, Weinhaus AJ, Tu J, et al. Basic biology of fetal pig pancreas and its use as an allograft. In Peterson CM, ed. Fetal islet transplantation. New York: Plenum Pub Corp, 1995, 51–68.

BERNARD E. TUCH

24 Humphrey RK, Smith MS, Kwok J, Si Z, Tuch BE, Simpson AM. In vitro de differentiation of fetal porcine pancreatic tissue prior to transplantation as islet-like cell clusters. Cells Tissues Organs 2001;168:158–69. 25 Tu J, Tuch BE, Si Z. Purification of porcine fetal pancreatic ␤-cell. World Congr Int Pancreas Islet Transplant Assoc 1999;7:128. 26 Beattie GM, Rubin JS, Mally MI, Otonkoski T, Hayek A. Regulation of proliferation and differentiation of human fetal pancreatic islet cells by extracellular matrix, hepatocyte growth factor, and cell-cell contact. Diabetes 1996;45:1223–8. 27 Beattie GM, Otonkoski T, Lopez AD, Hayek A. Functional ␤-cell mass after transplantation of human fetal pancreatic cells: Differentiation or proliferation? Diabetes 1997;46:244–8. 28 Jiang FX, Cram DS, DeAizpurua HJ, Harrison LC. Laminin-1 promotes differentiation of fetal mouse pancreatic beta-cells. Diabetes 1999;48:722–30. 29 Korbutt GS, Elliott JF, Ao Z, Smith DK, Warnock GL, Rajotte RV. Large scale isolation, growth and function of porcine neonatal islet cells. J Clin Invest 1996;97:2119–29. 30 Hayek A, Guardian C. Hormone release, islet yield, and transplantation of fetal and neonatal rat dorsal and ventral pancreatic islets. Diabetes 1986;35:1189–92. 31 Hullett DA, Falany JL, Love RB, Burlingham WJ, Pan M, Sollinger HW. Human fetal pancreas — a potential source for transplantation. Transplantation 1997;43:18–22. 32 Tuch BE, Osgerby KJ, Turtle JR. Normalization of blood glucose levels in nondiabetic nude mice by human fetal pancreas after induction of diabetes. Transplantation 1988;46:608–11. 33 Tuch BE, Monk RS, Beretov J. Reversal of diabetes in athymic rats by transplantation of human fetal pancreas. Transplantation 1991;52:172–5. 34 Thompson SC, Mandel TE. Fetal pig pancreas. Transplantation 1990;49:571–81. 35 Liu X, Federlin KF, Bretzel RG, Hering BJ, Brendel MD. Persistent reversal of diabetes by transplantation of fetal pig proislets into nude mice. Diabetes 1991;40:858–66. 36 Tuch BE, Casamento FM. Outcome of xenografted fetal porcine pancreatic tissue is superior in inbred scid (C.B-17/lcr-scid/scid) compared to outbred nude (CD-1-nu/nu) mice. Cell Transplant 1999;8:259–64. 37 Simeonovic CJ, Ceredig R, Wilson JD. Effect of GK1.5 monoclonal antibody dosage on survival of pig proislet xenografts in CD4+ T cell-depleted mice. Transplantation 1990;49:849–56. 38 Mandel TE, Koulmanda M. Xenotransplantation of fetal pig pancreas and reversal of diabetes in spontaneously diabetic NOD mice. Transplant Proc 1995;27:2179–80. 39 Korsgren O, Jansson L. Porcine islet-like cell clusters cure diabetic nude rats when transplanted under the kidney capsule, but not when implanted into the liver or spleen. Cell Transplant 1994;3:49–54. 40 Vo L, Tuch BE, Wright DC, Keogh GW, Roberts S, Simpson AM, et al. Lowering of blood glucose to non-diabetic levels in a hyperglycaemic pig by allografting of fetal pig islet-like cell clusters. Transplantation 2001;71:1671–7. 41 Yoon K-H, Quickel RR, Tatarkiewicz K, Ulrich TR, Hollister-Lock J, Trivedi N, et al. Differentiation and expansion of beta cell mass in porcine neonatal pancreatic cell clusters transplanted into nude mice. Cell Transplant 1999;8:673–89. 42 Brown J, Molnar IG, Clark W, Mullen Y. Control of experimental diabetes mellitus in rats by transplantation of fetal pancreases. Science 1974;184:1377–9. 43 Mullen Y, Clark WR, Molnar IG, Brown J. Complete reversal of experimental diabetes mellitus in rats by a single fetal pancreas. Science 1977;195:68–70. 44 McEvoy RC, Hegre OD. Syngeneic transplantation of fetal rat pancreas. III. Effect of insulin treatment on the growth and differentiation of the pancreatic implants after reversal of diabetes. Diabetes 1979;28:141–6. 45 Mandel TE, Higginbotham L. Organ culture and transplantation of fetal mouse pancreatic islets. Transplant Proc 1979;11:1505–6.

335

336

FETAL AND NEONATAL PANCREATIC TISSUE TRANSPLANTATION

46 Mandel TE, Koulmanda M, Cozzi E, Waterworth P, Toland M, Langford G, et al. Transplantation of normal and DAF-transgenic fetal pig pancreas into cynomolgus monkeys. Transplant Proc 1997;29:940. 47 Söderlund J, Wennberg L, Casan˘os-Velez E, Biberfeld P, Zhu S, Tibell A, et al. Fetal porcine islet-like cell clusters transplanted to cynomolgus monkeys. Transplantation 1999;67:784–91. 48 Sun Y, Ma X, Zhou D, Vacek I, Sun AM. Normalization of diabetes in spontaneously diabetic cynomologus monkeys by xenografts of microencapsulated porcine islets without immunosuppression. J Clin Invest 1996;98:1417–22. 49 Tuch BE, Madrid JC. Development of fetal sheep pancreas after transplantation into athymic mice. Cell Transplant 1996;5:483–9. 50 Jordan HN, Phillips RW. Effect of fatty acids on isolated ovine pancreatic islets. Am J Physiol 1978;2234:E162–7. 51 Tuch BE, Monk RS. Regulation of blood glucose to human levels by human fetal pancreatic xenografts. Transplantation 1991;51:1156–60. 52 Tuch BE. Reversal of diabetes by human fetal pancreas: Optimization of requirements in the hyperglycaemic nude mouse. Transplantation 1991;51:557–62. 53 Otonkoski T, Beattie GM, Mally MI, Ricordi C, Hayek A. Nicotinamide is a potent inducer of endocrine differentiation in cultured human fetal pancreatic cells. J Clin Invest 1993;92:1459–66. 54 Korsgren A, Andersson A, Sandler S. Pretreatment of fetal porcine pancreas in culture with nicotinamide accelerates reversal of diabetes after transplantation to nude mice. Surgery 1993;113:205–14. 55 Tuch BE, Osgerby KJ, Turtle JR. The role of calcium in insulin release from the human fetal pancreas. Cell Calcium 1990;11:1–9. 56 Tuch BE, Palavidis Z, Turtle JR. Activators of protein kinase C stimulate insulin secretion from the human fetal pancreas. Pancreas 1988;3:675–80. 57 Ågren A, Andersson A, Björken C, Groth C-G, Gunnarsson R, Hellerström C, et al. Human fetal pancreas: culture and function in vitro. Diabetes 1980;29(Suppl.1):64–9. 58 Hoffman L, Mandel TE, Carter WM, Koulmanda M, Martin FIR. Insulin secretion by human fetal pancreas in organ culture. Diabetologia 1982;23:426–30. 59 Tuch BE, Tabiin MT, Casamento FM, Yao M, Amaratunga A, Pinto AN. Role of pancreatic polypeptide as a marker of transplanted insulin-producing fetal pig cells. Cell Transplant 2001;10:285–93. 60 Naya FJ, Stellrecht CM, Tsai MJ. Tissue-specific regulation of the insulin gene by a novel basic helixloop-helix transcription factor. Genes Dev 1995;9:1009–19. 61 Krapp A, Knofler M, Ledermann B, Burki K, Berney C, Zoerkler N, et al. The bHLH protein PTF1-p48 is essential for the formation of the exocrine and the correct spatial organization of the endocrine pancreas. Genes Dev 1998;12:37452–63. 62 Stefan Y, Grasso S, Perrelet A, Orci L. A quantitative immunofluorescent study of the endocrine cell populations in the developing human pancreas. Diabetes 1983;32:292–301. 63 Falin LI. The development and cytodifferentiation of the islets of Langerhans in human embryos and foetuses. Acta Anat (Basel) 1967;68:147–68. 64 Si Z, Tuch BE, Walsh DA. Development of human fetal pancreas after transplantation into SCID mice. Cells Tissues Organs 2001;168:147–57. 65 Pictet R, Rutter WJ. Development of the embryonic endocrine pancreas. In Steiner DF, Freinkel N, eds. Handbook of physiology, Section 7: Endocrinology Volume 1. Washington DC: American Physiological Society, 1972;25–66. 66 Tuch BE, Ng ABP, Jones A, Turtle JR. Histologic differentiation of human fetal pancreatic explants transplanted into nude mice. Diabetes 1984;33:1180–7.

BERNARD E. TUCH

67 Usadel KH, Schwedes U, Bastert G, Steinau U, Klempa I, Fassbinder W, et al. Transplantation of human fetal pancreas: experience in thymusaplastic mice and rats and in a diabetic patient. Diabetes 1980;29(Suppl.1):74–9. 68 Beattie GM, Butler C, Hayek A. Morphology and function of cultured human fetal pancreatic cells transplanted into athymic mice: a longitudinal study. Cell Transplant 1994;3:421–5. 69 Tuch BE, Grigoriou S, Turtle JR. Growth and hormonal content of human fetal pancreas passaged in athymic mice. Diabetes 1986;35:464–9. 70 Tuch BE, Wright DC, Martin TE, Keogh GW, Deol HS, Simpson AM, et al. Development of fetal pig endocrine cells after allografting into the thymus gland. Transplantation 1999;67:1184–7. 71 Tuch BE, Jones A, Turtle JR. Maturation of the response of human fetal pancreatic explants to glucose. Diabetologia 1985;28:28–31. 72 Zanjani ED, Pallavicini MG, Ascensao JL, Flake AW, Langlois RG, Reitsma M, et al. Engraftment and long-term expression of human fetal hemopoietic stem cells in sheep following transplantation in utero. J Clin Invest 1992;89:1178–88. 73 Simeonovic CJ, Bowen KM, Kotlarski I, Lafferty KJ. Modulation of tissue immunogenicity by organ culture: comparison of adult islets and fetal pancreas. Transplantation 1980;30:174–9. 74 Burkhardt K, Charlton B, Mandel TE. An increase in the survival of murine H-2-mismatched cultured fetal pancreas allografts using depleting or nondepleting anti-CD4 monoclonal antibodies, and a further increase with the addition of cyclosporine. Transplantation 1989;47:771–5. 75 Bowen KM, Andrus L, Lafferty KJ. Successful allotransplantation of mouse pancreatic islets to nonimmunosuppressed recipients. Diabetes 1980;29(Suppl.1):98–104. 76 Kirchhof N, Shibata S, Kulick DM, Salerno CT, Dalmasso AP, Clemmings S, et al. Pig to primate islet xenografts reverse diabetes before undergoing acute cellular rejection. World Congr Int Pancreas Islet Transpl Assoc 1999;7:142. 77 Wennberg L, Karlsson-Parra A, Sundberg B, Rafael E, Liu J, Zhu S, et al. Efficacy of immunosuppressive drugs in islet xenotransplantation: leflunomide in combination with cyclosporine and mycophenolate mofetil prevents islet xenograft rejection in the pig-to-rat model. Transplantation 1997;63:1234–42. 78 Mandel TE, Koulmanda M. Effect of immunosuppression with anti-T cell monoclonal antibodies on the survival of organ-cultured fetal pig pancreas xenografts in nonobese diabetic mice. Transplant Proc 1993;25:2926–7. 79 McKenzie IC, Koulmanda M, Mandel TE, Sandrin MS. Pig islet xenografts are susceptible to ‘anti-pig’ but not gal␣(1,3)gal antibody plus complement in gal o/o mice. J Immunol 1998;161:5116–9. 80 Koulmanda M, McKenzie I, Sandrin M, Mandel T. Fetal pig xenografts in NOD/Lt mice: lack of expression of Gal(alpha 1–3)Gal on endocrine cells and the effect of peritransplant anti-CD4 monoclonal antibody and graft immunomodification on graft survival. Transplant Proc 1995;25:3570. 81 Rayat GR, Rajotte RV, Elliott JF, Korbutt GS. Expression of gal␣(1,3)gal on neonatal porcine islet ␤-cells and susceptibility to human antibody/complement lysis. Diabetes 1998;47:1406–11. 82 Stegall MD, Lafferty KJ, Kam I, Gill RG. Evidence of recurrent autoimmunity in human allogeneic islet transplantation. Transplantation 1996;61:1272–4. 83 Bai L, Tuch BE, Hering B, Simpson AM. Fetal pig ␤ cells are resistant to toxic effects of cytokines. Transplantation 2002, in press. 84 Simeonovic CJ, Townsend MJ, Morris CF, Hapel AJ, Fung M-C, Mann DA, et al. Immune mechanisms associated with the rejection of fetal murine proislet allografts and pig proislet xenografts: comparison of intragraft cytokine mRNA profiles. Transplantation 1999;67:963–71. 85 Bai L. Effect of human cytokines on fetal pig pancreatic ␤ cells. PhD thesis. Sydney: University of New South Wales, 2001, 127–34. 86 Soria B, Roche E, Berná G, León-Quinto T, Reig JA, Martin F. Insulin-secreting cells from embryonic stem cells normalize glycemia in streptozotocin-induced diabetic mice. Diabetes 2000;49:157–62.

337

This page intentionally left blank

Chapter 24

Experimental approaches to the prevention of islet rejection Norma Sue Kenyon

Introduction Various animal models of pancreatic islet cell transplantation have been utilized to test novel approaches for the prevention of graft rejection. In the setting of type 1 diabetes, prevention of recurrent autoimmunity has also been a critical focus. Production of proinflammatory mediators by macrophages and endothelial cells that are activated subsequent to intrahepatic islet infusion can be detrimental to islet survival and function, thus leading to early islet loss or primary non-function (PNF), and can also serve to augment specific immune responses to the graft. Experimental approaches to the prevention of rejection must, therefore, target rejection, recurrent autoimmunity, and events associated with PNF. Furthermore, agents must be identified that do not have adverse effects on islet cell function. With the advent of monoclonal antibodies, our understanding of the immune system has grown exponentially. The availability of antibodies, as well as recombinant DNA technology and newer generations of generalized immunosuppressants, such as tacrolimus and sirolimus, has dramatically expanded the repertoire of reagents available for the prevention of the immune response to transplanted cells and tissues. With steady improvements in the management of the technical aspects of islet isolation and transplantation, clinical outcomes continue to improve. Experimental efforts are primarily directed at identification of immunointervention agents that effectively prevent rejection without the toxicity associated with generalized immunosuppression. The ultimate goal is to allow for permanent engraftment of insulin-producing tissue without the need for life long immunosuppression of the recipient (tolerance). Currently, the requirement for continued administration of potent, generalized immunosuppression necessitates that only those patients with advanced diabetes, or with endstage complications, are eligible for an islet cell transplant. Initial attempts at testing of novel reagents for the prevention of rejection have generally been undertaken in rat and murine models. Chemical induction of diabetes (for example streptozotocin) is followed by transplantation of islets into the liver or underneath the kidney capsule. With the kidney capsule model, confirmation that the islet graft is responsible for normoglycaemia is readily verified by a return to hyperglycaemia subsequent to nephrectomy of the islet-bearing kidney. The successful prolongation of, or induction of tolerance to, allogeneic and xenogeneic islets in these models has highlighted several of the critical steps in the immune response that can lead to graft rejection. In addition, rodent models of immunologically-mediated diabetes [BB rat, non-obese diabetic (NOD) mouse] allow investigators to assess the impact of various strategies on both rejection and recurrent autoimmunity. The fact that these approaches have not translated well to larger animal, preclinical models is most likely primarily attributable to the greater complexity of the immune system of larger

340

EXPERIMENTAL APPROACHES TO THE PREVENTION OF ISLET REJECTION

animals. In addition, the availability of an unlimited supply of inbred rodent pancreas donors allows the investigators reproducibly to isolate and transplant an exact number of viable islets for each experiment. Preclinical investigators working with larger animals models (such as dogs, non-human primates) must contend with either the use of multiple donors or of large donors and small recipients to achieve an adequate functional islet mass to reverse hyperglycaemia reproducibly. The outbred nature of the donor-recipient pairs further adds to the complexity of the system, but the advantage of such models is that they mimic the biological disparity encountered in the clinical setting. The lack of a spontaneously diabetic, preclinical model with biological proximity to type 1 diabetes in humans necessitates that the ultimate proof of concept can only be obtained from clinical studies. To highlight these points, the primary focus of this chapter will be on experimental studies of allogeneic islet cell transplantation, with an emphasis on recent advances in the prevention of rejection, as exemplified by studies in rodents and non-human primates.

Manipulation of the recipient Antibodies or recombinant molecules Prevention of rejection via manipulation of the recipient immune response has been attempted with various combinations of conventional, generalized immunosuppressive drugs, polyclonal or monoclonal antibodies, and chimeric molecules that target key components of the immune system, by donor antigen administration, and by transplantation of haematopoietic cells to induce chimerism (reviewed in [1]). Using a variety of approaches, it is now possible to prevent rejection of allogeneic islets in rodents, dogs, monkeys, and humans, but reports of donor-specific tolerance in larger animals and humans have remained more sporadic. To initiate a rejection response, CD4+ T cells must first encounter donor class II antigen via the T-cell receptor (TCR) complex, which consists of the donor-specific TCR itself and the associated CD3 complex [2,3]. Donor antigen must be presented by an antigen-presenting cell (APC), in the context of a major histocompatability (MHC) molecule, in order for the T cell to recognize it. If a donor APC provides the signal, then the class II molecules of the donor interact directly with the TCR (direct pathway); if a recipient APC is the stimulus, processed donor antigen is represented as a peptide in the cleft of the recipient class II molecules (indirect pathway) [4,5]. The direct pathway of antigen presentation appears to be primarily responsible for acute rejection responses, while the indirect pathway is the primarily mediator of chronic rejection [6,7]. This interaction of the TCR complex with antigen/MHC on the APC surface provides the first activation signal (signal 1) to the T cell. The CD3 complex is integrally involved in the process T-cell activation. Additional molecules on the T-cell surface are involved in signal 1, including the leucocyte common antigens (CD45). Signal 1 alone, however, will not result in the generation of an immune response. A second signal, known as costimulation, provided by interaction of additional molecules on the T-cell and APC surfaces must be received [8]. Blockade of costimulation can result in functional inactivation, or anergy, of the responding T-cell [9], as well as apoptosis [10]. Additional accessory molecules have been defined on the surface of immune cells, including CD2 and cell adhesion molecules (CAM) that are essential to T-cell activation. The cytokines produced by the activated cells are required for further application of an immune response. Experimental approaches to the prevention of rejection have been designed to block one or more of the myriad interactions that occur between T-cells and APC. Blockade of this afferent arm of the immune response can prevent the generation of efferent responses, such as the development of cyto-

N.S. KENYON

toxic T-cells and antibodies, thus enhancing graft survival. Approaches that solely target molecules involved in the efferent (effector) immune response have generally not been as effective at prevention of rejection or induction of tolerance. Such agents can be effective when used in combination with other drugs or biological agents that act earlier in the immune cascade by targeting cells that escape blockade at earlier time points. A variety of strategies have been used to block receptor-ligand interactions, thus suppressing or altering T-cell signaling and activation and leading to T-cell clonal deletion, anergy, or regulation. Polyclonal and monoclonal antibodies, recombinant molecules, generalized immunosuppressive drugs, and strategies incorporating administration of donor antigen, have been utilized to achieve these effects (Table 24.1), with prolongation of graft survival a desired effect and induction of tolerance the ultimate goal.

Table 24.1 Manipulation of the recipient Polyclonal antibodies Antithymocyte globulin Antilymphocyte globulin Monoclonal antibodies specific for MHC class I MHC class II TCR CD2 CD3 CD4 CD8 CD11a (LFA-1) CD25 CD45RB CD54 (ICAM) CD80 (B7.1) CD86 (B7.2) CD154 Chimeric molecules CTLA4 immunoglobulin IL-10Fc Intrathymic administration of donor antigen Islets Spleen cells Bone marrow cells Soluble MHC antigens MHC peptides Donor class I peptide pulsed host dendritic cells Donor bone marrow infusion and chimerism Irradiation based High-dose marrow infusion Costimulatory blockade

341

342

EXPERIMENTAL APPROACHES TO THE PREVENTION OF ISLET REJECTION

Several monoclonal antibodies specific for T-cells and T-cell subsets have been tested in islet transplant models. With regards to pan-T-cell specific reagents, both anti-CD2 and CD3 specific monoclonal antibodies have been tested in animal models. Peritransplant administration of anti-CD2 has been reported to prolong the survival of murine pancreatic islet allografts [11] and of rat islets in murine recipients [12]. Administration of anti-CD3 led to prolonged graft survival in murine islet allograft recipients, with some experiencing permanent engraftment [13], although tolerance was not achieved, and injection of donor strain leucocytes resulted in rejection in recipients with long-term surviving grafts. Taken together with data that demonstrates permanent remission of diabetes in antiCD3 treated NOD mice with recent disease onset [14], anti-CD3 treatment appears to have significant potential as therapy for patients with type 1 diabetes who undergo an islet cell transplant. Further support for the critical role of CD3 directed therapies is the demonstration that peritransplant treatment of non-human primates with an anti-CD3-immunotoxin conjugate, plus a short course of cyclosporin A and methylprednisolone, resulted in long-term islet survival in a concordant xenograft model [15]. With regards to T-cell subsets, a single course of a depleting anti-CD4 monoclonal antibody was shown to result in indefinite islet allograft survival in mice [16,17]. In another study, allografted mouse islets were permanently accepted in anti-CD4 or anti-CD4 plus CD8 treated mice, but not in recipients treated with anti-CD8 alone or with the pan-T-cell reagent, anti-Thy 1.2 [18]. Mice treated with an anti-donor class I monoclonal antibody exhibited modest prolongation of islet-allograft survival [19], and mice deficient in ␤2-microglobulin expression (and therefore, CD8 cell deficient) have been demonstrated to experience prolonged allograft survival [20,21]. CD45, leucocyte common antigen, is a family of transmembrane protein tyrosine phosphatases involved T-cell activation and function [22]. These molecules participate in the delivery of signal 1 to T-cells, and the various isoforms of the CD45 antigen appear to functionally distinguish CD4+ T-cell populations [23]. Treatment of mice with antibody specific for the CD45R isoform has been shown to result in indefinite islet allograft survival and donor-specific tolerance [24]. A shift in CD45 isoform expression, from CD45RBhi, CD45ROlo to CD45RBlo, CD45ROhi was observed and was correlated with an increase in the intragraft expression of interleukin (IL)-4 and IL-10 mRNA, consistent with the induction of immunoregulatory cell populations. A great deal of effort has been focused on antirejection approaches that block signal 2 (costimulation). Elucidation of the CD28–CD80/86 pathway of costimulation and its critical role in the generation of an immune response led to intensive study of the potential graft promoting effect of immunointervention with agents that affect these molecules. Treatment with CTLA 4 immunoglobulin has proven successful as a means to prevent graft rejection [25–27], as well as graft-versus-host disease (BVHD) [28,29], a finding with particular significance to tolerogenic protocols that involve donor haematopoietic cell infusion. Administration of CTLA4 immunoglobulin in combination with donor splenocytes can interrupt the progression of chronic allograft rejection in presensitized hosts [30]. Depending on the conditions and the agent used (CTLA4immunoglobulin versus antibodies specific for CD80 and/or CD86), blockade of this pathway can have variable effects on autoimmunity [31,32] in rodent models. NOD mice treated with anti-CD86 (B7.2) or CTLA4immunoglobulin at the onset of insulitis (2 to 4 weeks) did not develop diabetes, although the treatment did not alter the severity of insulitis and treatment at greater than 10 weeks of age did not prevent disease onset [33]. In contrast, treatment with anti-CD80 (B7.1) actually accelerated the course of disease [34]. With regards to islets, treatment with CTLA4-Fc resulted in indefinite islet allograft survival and donor specific tolerance that was CD4 dependent [35]. When CTLA4 immunoglobulin was used in combination with donor-specific transfusion (DST) pretransplant, there was a minimal delay in the rejection of islet allografts but no significant prolongation of survival as compared to controls [36].

N.S. KENYON

Treatment of mice with anti-CD86 (B7.2) but not anti-CD80 prolonged murine islet allograft survival, and a combination of the two monoclonal antibodies further enhanced extended survival [33,34]. CTLA4immunoglobulin has not proven to be highly effective for prolongation of nonhuman primate islet allograft survival, with three of five Cynomolgus monkeys rejecting at the same time as controls and two of five demonstrating some graft function for 30 and 50 days [37]. Another avenue to prevent rejection, with the potential for lending an aspect of donor specificity to the immunointervention, was discovered subsequent to the identification of the CD40-CD154 pathway. Prior to costimulation via the CD28-CD80/86 pathway, the CD40-CD154 pathway is operative [38–44]. As T-cells become activated, one of the earliest effects is expression and upregulation of CD154 on activated T-cells, followed by interaction with the CD40 antigen expressed on APC. CD40-CD154 mediated events then lead to interaction of CD28 on T-cells with the B7 molecule on APC. Unlike reagents that affect entire cell (for example, CD2 or CD3) or cell subset (for example, CD4 or CD8) populations, interference with CD40-CD154 may enable specific alteration of antidonor reactions as a consequence of the specific upregulation of CD154 on T-cells that encounter donor antigen in the form of transplanted tissue. While it can be argued that activation of pathogen-specific responses could also be altered if the host experiences infection concomitant with transplantation, published data suggests that, for example, viral infection can bypass the requirement for CD40-CD154 interactions [45]. Manipulation of the CD40-CD154 constimulatory pathway has been shown to prolong the survival of allogeneic islet transplants in non-autoimmune rodent models [46,47]. Addition of pretransplant DST to anti-CD154specific monoclonal antibody therapy led to donor-specific tolerance and indefinite islet allograft survival; evidence for negative signalling through the CTLA4 molecule in this tolerance model was observed [48]. In NOD mice, anti-CD154 therapy can prevent, or halt the progression of, autoimmune diabetes [49], delay recurrence of diabetes in recipients of syngeneic islets, and significantly prolong islet allograft survival [50]. Similarly, recurrence of diabetes was prevented in BB rat recipients of islet isografts that were treated with anti-CD154 [51]. In addition to the reported effects on islet allograft survival and autoimmune diabetes, interference with this key pathway has been demonstrated to prevent the production of non-specific mediators of inflammation, such as nitric oxide and cytokines [52,53], thus suggesting that anti-CD154 may prevent early islet loss (mediated by non-specific inflammatory events that occur as a result of intrahepatic islet transplantation). In striking contrast to other monotherapies, proven successful in rodent ICT models but not in larger animal, preclinical models, anti-CD154 monotherapy has resulted in significant prolongation (more than 1 year) of non-human primate islet allograft survival [54–56]. Additional therapy with anti-CD154 to treat suspected rejection episodes resulted in reversal of hyperglycaemia, thus enabling rescue of partial graft function or a return to insulin independence. Interference with additional accessory molecules, such as those involving adhesive interactions, has also been studied as a means to enhance islet allograft survival. Donor-specific tolerance to islet allografts has been obtained in several histoincompatible mouse strain combinations subsequent to antiLFA-1 monoclonal antibody therapy; neither clonal deletion nor anergy could account for the tolerogenic effect [57,58]. Studies involving cytokine and cytokine receptor directed therapies have resulted in prolongation of murine islet allograft survival, for example subsequent to treatment with an IL-2-diphtheria toxin fusion protein [59] or with anti-␥c specific monoclonal antibody (binds the common ␥ chain of IL-2, IL-4, IL-7, IL-9, and IL-15) [60].

Intrathymic transplantation Tolerogenic strategies may target central deletional mechanisms that are involved in the generation of tolerance to self antigens; T-cells emerging from the bone marrow migrate through the thymus. If self

343

344

EXPERIMENTAL APPROACHES TO THE PREVENTION OF ISLET REJECTION

antigen is encountered, the cells are deleted before they can leave for the periphery. Treatments that eliminate mature T-cells, used in combination with strategies that result in the expression of donor alloantigen in the thymus, can therefore allow for the induction of donor-specific tolerance via naturally occurring mechanisms as newly emerging T lymphocytes migrate through the thymus and are deleted subsequent to encounter with donor antigen. As an example, intrathymic transplantation of allogeneic islets can lead to permanent allograft acceptance in rats [61]. Indefinite survival of islet allografts was also achieved in a low responder rat strain combination using sublethal, total body irradiation (TBI) of the recipient and intrathymic inoculation of ultraviolet B (UVB) irradiated donor strain splenocytes [62]. In a high responder strain, intrathymic injection of donor splenocytes in recipients treated with sublethal TBI or ALS resulted in indefinite islet allograft survival and donorspecific tolerance in 50 per cent of recipients as compared to 80 to 100 per cent in recipients of intrathymic UVB-irradiated donor splenocytes [63]. In contrast, subcutaneous, intratesticular, or intravenous inoculation of UVB-irradiated donor splenocytes had no graft-promoting effect in either strain combination. Additional studies demonstrate that intrathymic transplantation of donor bone marrow cells [64] or spleen cell membrane antigens [65] also led to permanent islet allograft survival. Intrathymic inoculation of APC-depleted islets, however, led to survival of islets within the thymus but not in extrathymic sites [66]. The role of CD4+ T cells and the direct pathway of antigen presentation in inducing the clonal deletion of T cells capable of reacting to allogeneic islets was demonstrated utilizing a TCR transgenic mouse model [67]. Intrathymic inoculation with donor class I pulsed host dendritic cells, in combination with Antilymphecitic serum (ALS) treatment, led to 100 per cent donor-specific, permanent islet allograft survival, thus demonstrating the potential for tolerance induction to donor alloantigen via the indirect pathway of antigen presentation [68]. The intrathymic approach does not appear to be as effective for the prolongation of xenogeneic (rat to mouse) islet survival, even after profound T-cell depletion [69]. This route does, however, appear to be beneficial for the prevention of autoimmunity in the BB rat model [70,71].

Haematopoietic chimerism Bone marrow transplantation into cytoablated recipients is a well-established approach to the induction of donor-specific tolerance in many species [72,73]. Fully allogeneic chimeras, in which immune cells are entirely of donor origin, are problematic, in that engraftment is difficult to obtain when protocols such as T-cell depletion of the marrow prior to transplant are utilized to prevent the occurrence of lethal GVHD [73]. If engraftment does occur, the animals are relatively immunoincompetent due to a lack of host APC to present antigen to donor T cells that have been educated in the host thymus. These problems can be avoided via the production of mixed allogeneic chimeras, which do not suffer from lethal GVHD, are immunocompetent, and are also tolerant of subsequent cellular or solid organ allografts from animals of the same strain as the bone marrow donor [73]. Application of this approach to the field of human transplantation has been hampered by the requirement for cytoreductive treatment of the recipient, a measure that is not advantageous for individuals who are not suffering from malignancy. Recent demonstration of bone marrow engraftment in non-cytoablated animals, via multiple infusions of large numbers of donor bone marrow cells [74], has provided the impetus to continue to explore this approach as a means to induce tolerance to transplanted organs or islets. Infusion of haematopoietic cells (bone marrow, blood, splenocytes) at the time of transplantation has long been studied as a means to induce acceptance of grafted tissues in both animals and humans [75,76], and the success of such protocols may be due to the establishment of chimerism. In allogeneic canine islet transplant studies, high-dose donor bone marrow infusion in recipients treated with high-dose cyclosporine (CSA) (700 to 1000 ng/ml trough levels) actually enhanced islet

N.S. KENYON

rejection. Addition of an anti-thy 1 specific monoclonal antibody resulted in graft survival during the CSA treatment period but did not result in the induction of tolerance [77]. Under the cover of subimmuosuppressive CSA (100 to 300 ng/ml trough levels, beginning 5 days prior to transplant and continuing for 29 days thereafter), dogs rejected their islet allografts in 7 to 10 days. Addition of donor marrow infusion (administered on postoperative days 0, 5, 10, 15, and 20, with day 0 being the day of islet cell transplant) resulted in survival of the majority of the grafts during CSA therapy, with a mean survival time of 28 days. Although tolerance was not achieved, elimination of class II bright cells from the marrow prior to infusion resulted in significant prolongation of islet allograft survival beyond the discontinuation of CSA treatment, with a mean survival time of 72 days [78]. The ability to detect donor-derived haematopoietic cells in the recipient’s circulation correlated with islet allograft function. An additional impetus to continue to explore approaches to transplantation tolerance via the establishment of chimerism is the potential for prevention of recurrent diabetes. Studies in NOD mice have yielded evidence to support the concept that autoimmune diabetes is a disease that can be prevented by transplantation of bone marrow from diabetes-resistant donors into irradiated NOD mice [79]. Data have been published which demonstrate that mixed allogeneic chimerism, achieved with sublethal irradiation, can also prevent the occurrence of diabetes in NOD mice [80]. Published data from Mathieu and Waer suggests that a minimum of 5 per cent, and as much as 25 per cent, chimerism may be essential to prevent the development of diabetes in cases where diabetes-prone marrow is part of the haematopoietic system [81]. Recently published results demonstrate that blockade of the CD40–CD154 and CD28/B7 costimulatory pathways leads to donor marrow engraftment, chimerism, and donor-specific tolerance in murine recipients treated with sublethal TBI [82]. Deletion of peripheral, donor-specific host T cells, as well as central deletion of newly emerging donor-specific T cells, was observed. These results were achieved with one injection of an anti-CD154 specific monoclonal antibody (MR1) and CTLA4immunoglobulin, plus 3 Gy of TBI. In combination with irradiation, treatment with CTLA4immunoglobulin alone did not result in the establishment of chimerism. While MR1 treatment alone resulted in chimerism in sublethally irradiated mice, establishment of chimerism was not as consistent as observed for mice treated with both agents and was not stable. Chimeric mice accepted donor, but rejected third party, skin allografts. It has subsequently been demonstrated that costimulatory blockade consisting of anti-CD154 and CTLA4immunoglobulin, together with highdose donor marrow infusion, resulted in multilineage macrochimerism and donor-specific skin acceptance without the need for cytoreductive treatment of the recipient [83]. Similar results were achieved with anti-CD154 and high-dose marrow infusion alone [84]. Establishment of mixed allogeneic chimerism via lethal and non-lethal conditioning results in the induction of donor-specific tolerance to simultaneous islet allografts [85]. Similarly, lethal irradiation of NOD mice resulted in permanent islet allograft acceptance and prevention of recurrent diabetes when islets were transplanted under the kidney capsule within 24 h of donor bone marrow infusion [86].

Immunomodulation Both the direct and indirect pathways of antigen presentation contribute to allograft rejection. Direct recognition of graft APC by recipient T lymphocytes plays a role in the early rejection response. Indirect presentation of donor MHC antigens by recipient APC requires processing and presentation of the donor antigen and emerges at a later time point than the direct pathway. The cellular nature of islet transplantation offers the opportunity to alter the ability of islets to stimulate an immune response directly. Several approaches have been taken to achieve this goal (Table 24.2).

345

346

EXPERIMENTAL APPROACHES TO THE PREVENTION OF ISLET REJECTION

Table 24.2 Approaches to altering the capacity of transplanted islets to stimulate rejection Alteration of islet APC Culture in high oxygen (95%) Culture at low temperature (22°C) Cryopreservation Ultraviolet light (UVB) irradiation Anti-class II + complement Antidendritic cell + complement mCTLA4-Fc Alteration of islet class I or adhesion molecules Masking with anti-class I Anti-ICAM Transfection of islets to produce immunomodulators CTLA4 immunoglobulin Soluble FasL OX40 immunoglobulin Cotransplantation of islets with cells that have immunomodulatory properties FasL transfected myoblasts Testicular grafts/Sertoli cells

Elimination or functional inactivation of APC that reside within the graft has been achieved by treating islets prior to transplantation with UVB light irradiation, anti-class II or antidendritic cellspecific monoclonal antibodies and complement, cryopreservation, or by culture at low temperature or in high oxygen concentrations. Immunomodulation of islets prior to transplantation has led to prolongation of allograft survival, and in rodent models, donor-specific tolerance has been achieved with combinations of short course immunosuppression and transplantation of modulated islets [87, 88]. In the studies of Faustman, unresponsiveness to allogeneic islets (subsequent to islet immunomodulation with anti-class II or antidendritic cell-specific monoclonal antibodies) could be abrogated by injection of donor splenocytes or dendritic cells. With regards to larger animal models, prolongation of rabbit fetal islet allograft survival subsequent to culture in high oxygen and UVB irradiation was achieved only in haploidentical rabbits treated with a high peritransplant dose of CSA or with chronic, low-dose CSA [89]. In a canine model, transplantation of UVB-irradiated [90] or anticlass II [91] treated islets into recipients treated with subimmunosuppressive doses of CSA resulted in significant prolongation of allograft survival but did not lead to tolerance. Transplantation of UVBirradiated islets has also been performed in non-human primates; however, analysis of the graft-promoting effect in this study was hindered by the adverse effect of CSA on islet function [92]. Species differences in the expression of class II antigens on islet APC and endothelium may be important in the outcome of experiments aimed at alteration of APC function [93]. In addition, it has been demonstrated that APC-depleted murine islet allografts can be rejected if B7-1 is expressed on islet ␤cells [94]. In support of the role of the B7 pathway in islet allograft rejection, islets coated ex vivo with murine CTLA4/Fc were accepted in 42 per cent of murine recipients. One half of the mice that experienced graft survival beyond 150 days were formally proven to be specifically tolerant to donor tissue [95]. Tolerance to APC-depleted murine islet allografts has been demonstrated to be dependent on CD4 T cells in one study [96].

N.S. KENYON

Although tolerance was not achieved in dogs, as it has been for rodents, the observed prolongation of allograft survival in the canine model indicates that deletion or functional inactivation of APC significantly impacts the direct pathway of antigen presentation, thus enhancing survival in the early post-transplant period. Incorporation of immunomodulation strategies into protocols that target the indirect pathway may result in enhanced allograft survival or tolerance.

Masking, genetic modification, and cotransplantation approaches Other approaches to alter the ability of the immune system to be stimulated by islets include masking techniques and cotransplantation with cells that possess immunomodulatory properties. Faustman and colleagues demonstrated that masking of human islets with F(ab)2 fragments of monoclonal antibodies specific for donor class I antigens led to extended graft survival in the absence of immunosuppression in mice, the theory being that the antibody fragments prevented donor-specific CD8+ cells from interacting with islet class I molecules [97]. A CTLA4 immunoglobulin and/or OX40 immunoglobulin transfected insulinoma cell line, with the NOD genotype and expressing the immunogenic SV40 large T antigen, survived better in young NOD mice than a non-transfected control [98]. CTLA4 immunoglobulin transfected muscle cells have been demonstrated to enhance islet allograft survival when the muscle cells are syngeneic to the host, and prolonged graft survival was achieved with the addition of a short course of anti-LFA1 specific monoclonal antibody [99]. Cotransplantation of FasL expressing, syngeneic myoblasts with allogeneic islets was reported to prolong islet allograft survival in a mouse model [100]. In a separate study, transplantation of FasL expressing myoblast cell lines with syngeneic islets has been reported to induce a vigorous neutrophilmediated inflammatory response that led to islet destruction [101]. Cotransplantation of FasL expressing Sertoli cells demonstrated a dose-dependent enhancement of syngeneic islet survival in NOD mice at cell doses of 1 to 4 million cells. In contrast, a loss of the protective effect was observed at a dose of 8 million Sertoli cells, with immunohistochemical identification of neutrophil infiltration and graft destruction at the higher cell dose [102]. Cotransplantation of allogeneic islets and FasL expressing, allogeneic testicular tissue, but not FasL negative testicular tissue, resulted in prolongation of islet survival [103]. The protective effect of the testicular tissue was abrogated subsequent to treatment of the recipients with a FasL specific monoclonal antibody. Transfection of mouse islets with human CTLA4immunoglobulin or human soluble FasL resulted in prolongation of allograft survival in some animals [104]. One possible explanation for the variable results obtained in these studies may be a result of FasL expression at levels not normally encountered in vivo, thus leading to graft destruction in some models (overexpression) and graft prolongation in others.

Primary non-function As mentioned previously, in addition to prevention of rejection and recurrent autoimmunity, increasing effort is being placed on identification of approaches that prevent early, non-specific inflammatory events that occur subsequent to intrahepatic islet infusion. Treatment of mice to deplete macrophages in vivo, via administration of silica, or treatment with CSA abrogated PNF in a study that provided histological evidence of the role of macrophages and macrophage products in PNF [105]. Many new agents are emerging that have the potential to inhibit activation of macrophages and endothelial cells, as discussed previously with regards to the CD40–CD154 pathway. Incorporation of these agents into multifaceted transplant protocols should enhance the survival of islet allografts and decrease the number of islets required to achieve normoglycaemia.

347

348

EXPERIMENTAL APPROACHES TO THE PREVENTION OF ISLET REJECTION

Conclusion As our ability to prevent clinical islet allograft rejection continues to improve, the identification of alternative sources of insulin-producing tissues gains importance as a critical area of research. Currently, there are 2000 to 4000 organ donors per year in the United States, but there are 1.5 million individuals with type 1 diabetes. Potential sources of tissue for biological replacement are expanded islets (from adult, neonatal, or fetal tissues), islets from other species such as the pig, and genetically engineered glucose responsive, insulin-producing cell lines. The recent successes in clinical islet cell transplantation [106] have led to an increased demand for work in these areas of research. Successful application of biological replacement therapies for all patients with diabetes, therefore, will be the result of successful delineation of relatively non-toxic transplant protocols that result in the establishment of donor-specific tolerance and the identification of an as yet to be defined, abundant source of insulin-producing cells.

References 1 Rossini AA, Greiner DL, Morder JP. Induction of immunologic tolerance for transplantation. Physiol Rev 1999;79:99–141. 2 rensky AM, Weiss A, Crabtree G, Davis MM, Parham P. T-lymphocyte-antigen interactions in Transplant rejection. N Engl J Med 1990;322:510–17. 3 Krieger NR, Yin DP, Garrison Fathman C. CD4+ but not CD8+ cells are essential for allorejection. J Exp Med 1996;184:2013–18. 4 Sayegh MH, Watschinger B, Carpenter CB. Mechanisms of T cell recognition of alloantigen: the role of peptides. Transplantation 1994;57:1295–302. 5 Lechler RI, Lombardi G, Batchelor JR, Reinsmoen N, Bach FH. The molecular basis of alloreactivity. Immunol Today 1990;11:83–8. 6 Ciubotariu R, Liu Z, Colovai AI. Persistent allopeptide reactivity and epitope spreading in chronic rejection of organ allografts. J Clin Invest 1998;101:398–405. 7 Vella JP, Spadafora-Ferreira M, Murphy B. Indirect allorecognition of major histocomptability complex allopeptides in human renal transplant recipients with chronic graft dysfunction. Transplantation 1997;64:795–800. 8 Janeway CA Jr, Bottomly K. Signals and signs for lymphocyte responses. Cell 1994;76:275–85. 9 Gimmi CD, Freeman GJ, Gribben JG, Gray G, Nadler LM. Human T-cell clonal anergy is induced by antigen presentation in the absence of B7 constimulation. Proc Natl Acad Sci 1993;90:6586–90. 10 Noel PJ, Boise LH, Green JM, Thompson CB. CD28 costimulation prevents cell death during primary T cell activation. J Immunol 1996;157:636–42. 11 Kapur S, Khanna A, Sharma VK, Li B, Suthanthiran M. CD2 antigen targeting reduces intragraft expression of mRNA-encoding granzyme B and IL-10 and induces tolerance. Transplantation 1996;62(2):249–55. 12 Chavin KD, Lau HT, Bromberg JS. Prolongation of allograft and xenograft survival in mice by antiCD2 monoclonal antibodies. Transplantation 1992;54(2):286–91. 13 Mackie JD, Pankewycz OG, Bastos MG, Kelley VE, Strom TB. Dose-related mechanisms of immunosuppression mediated by murine anti-CD3 monoclonal antibody in pancreatic islet cell transplantation and delayed-type hypersensitivity. Transplantation 1990;49(6):1150–4. 14 Chatenoud L, Primo J, Bach JF. CD3 antibody-induced dominant self tolerance in overtly diabetic mice. J Immunol 1997;158(6):2947–54.

N.S. KENYON

15 Contreras JL, Eckhoff DE, Cartner S, Bilbao G, Ricordi C, Neville DM Jr, et al. Long-term functional islet mass and metabolic function after xenoislet transplantation in primates. Transplantation 2000;69(2):195–201. 16 Shizuru JA, Gregory AK, Chao CT, Fathman CG. Islet allograft survival after a single course of treatment of recipient with antibody to L3T4. Science 1987;237(4812):278–80. 17 Alters SE, Song HK, Fathman CG. Evidence that clonal energy is induced in thymic migrant cells after anti-CD4-mediated transplantation tolerance. Transplantation 1993;56(3):633–8. 18 Song HK, Alters SE, Fathman CG. Evidence that anti-CD8 abrogates anti-CD4-mediated clonal anergy but allows allograft survival in mice. Transplantation 1993;55(1):133–9. 19 Osorio RW, Ascher NL, Stock PG. Prolongation of in vivo mouse islet allograft survival by modulation of MHC class I antigen. Transplantation 1994;57(6):783–8. 20 Markmann JF, Bassiri H, Desai NM, Odorico JS, Kim JI, Koller BH, et al. Indefinite survival of MHC class I-deficient murine pancreatic islet allografts. Transplantation 1992;54(6):1085–9. 21 Desai NM, Bassiri H, Kim J, Koller BH, Smithies O, Barker CF, et al. Islet allograft, islet xenograft, and skin allograft survival in CD8+ T lymphocyte-deficient mice. Transplantation 1993;55(4):718–22. 22 Alexander DR. The CD45 tyrosine phosphatase: a positive and negative regulator of immune function. Sem Immunol 2000;12:349–59. 23 Zhong RZ, Lazarovits AI. Monoclonal antibody against CD45RB for the therapy of rejection and autoimmune disease. J Mol Med 1998;76:572–80. 24 Basadonna GP, Auersvald L, Khuong CQ, Zheng XX, Kashio N, Zekzer D, et al. Antibody-mediated targeting of CD45 isoforms: a novel immunotherpeutic strategy. Proc Natl Acad Sci 1998;95(7):3821–6. 25 Bolling SF, Liln H, Wei RQ, Linslely PL, Turka LA. The effect of combination cyclosporine and CTLA4immunoglobulin therapy on cardiac allograft survival. J Surg Res 1994;57(1):60–4. 26 Pearson TC, Alexander DZ, Winn KJ, Linsley PS, Lowry RP, Larsen CP. Transplantation tolerance induced by CTLA4immunoglobulin. Transplantation 1994;57(12):1701–6. 27 Pearson TC, Alexander DZ, Hendrix R, Elwood ET, Linslety PS, Winn KJ, et al. CTL4immunoglobulin plus bone marrow induces long-term allograft survival and donor-specific unresponsiveness in the murine model. Transplantation 1996;61(7):997–1004. 28 Blazar BR, Taylor PA, Gray GS, Vallera DA. The role of T cell subsets in regulating the in vivo efficacy of CTLA4immunoglobulin in preventing graft-versus-host disease in recipients of fully MHC or multiple minor histocompatibility-disparate donor inocula. Transplantation 1994;58(12):1422–5. 29 Hakim FT, Cepeda R, Gray GS, June CH, Abe R. Acute graft-versus-host reaction can be aborted by blockade of costimulatory molecules. J Immunol 1995;155:1757–66. 30 Chandraker A, Azuma H, Nadeau K, Carpenter CB, Tilney NL, Hancock WW, et al. Late blockade of T cell costimulation interrupts progression of experimental chronic allograft rejection. J Clin Invest 1998;101(11):2309–18. 31 Finck BK, Linsley PS, Wofsy D. Treatment of murine lupus with CTLA4immunoglobulin. Science 1994;265:1225–7. 32 Perrin PJ, Scott D, Davis TA, Gray GS, Doggett MJ, Abe R, et al. Opposing effects of CTLA4immunoglobulin and Anti-CD80 (B7-1) plus Anti-CD86 (B7-2) on experimental allergic encephalomyelitis. J Neuroimmunol 1996;65:31–9. 33 Lenschow DJ, Zeng Y, Hathcock KS, Zuckerman LA, Freeman G, Thistlethwaite JR, et al. Inhibition of transplant rejection following treatment with anti-B7-2 and anti-B7-1 antibodies. Transplantation 1995;60(10):1171–8.

349

350

EXPERIMENTAL APPROACHES TO THE PREVENTION OF ISLET REJECTION

34 Lenschow DJ, Ho SC, Sattar H, Rhee L, Gary G, Nabavi N, et al. Differential effects of anti-B7.1 and anti-B7.2 monoclonal antibody treatment on the development of diabetes in the nonobese diabetic mouse. J Exp Med 1995;181(3):1145–55. 35 Tran HM, Nickerson PW, Restifo AC, Ivis-Woodward MA, Patel A, Allen RD, et al. Distict mechanisms for the induction and maintenance of allograft tolerance with CTLA4-Fc treatment. J Immunol 1997;159(5):2232–9. 36 Roy-Chaudhury P, Nickerson PW, Manfro RC, Zhen XX, Steiger J, Li YS, et al. CTLA4immunoglobulin attentuates accelerated rejection (presensitization) in the mouse islet allograft model. Transplantation 1997;64(1):172–5. 37 Levisetti MG, Padrid PA, Szot GL, Mittal N, Meehan SM, Wardrip CL, et al. Immunosuppressive effects of human CTLA4immunoglobulin in a non-human primate model of allogeneic pancreatic islet transplantation. J Immunol 1997;159(11):5187–91. 38 Ranheim EA, Kipps TJ. Activated T cells induce expression of B7/BB1 on normal or leukemic B cells through a CD40 dependent signal. J Exp Med 1993;177:925–35. 39 Roy M, Aruffo A, Ledbetter J, Linsley P, Kehry M, Noelle R. Studies on the interdependence of gp39 and B7 expression and function during antigen specific immune responses. Eur J Immunol 1995;25:596–603. 40 Han S, Hathcock K, Zheng B, Kepler TB, Hodes R, Kelsoe G. Cellular interaction in germinal centers. Roles of CD40 ligand and B7-2 in established germinal centers. J Immunol 1995;155:556–67. 41 Shinde S, Wu Y, Guo Y, Niu Q, Xu J, Grewal IS, et al. CD40L is important for induction of, but not response to, costimulatory activity. ICAM-1 as the second costimulatory molecule rapidly upregulated by CD40L. J Immunol 1996;157:2764–8. 42 Yang Y, Wilson JM. CD40 ligand dependent T cell activation: requirement of B7-CD28 signaling through CD40. Science 1996;273:1862–4. 43 Grewal IS, Foellmer HG, Grewal KD, Xu J, Hardardottir F, Baron JL, et al. Requirement for CD40 ligand n costimulation induction, T cell activation, and experimental allergic encephalomyelitis. Science 1996;273:1864–7. 44 Lederman S, Yellin MK, Inghirami G, Lee JJ, Knowles DM, Chess I. Molecular interactions mediating T-B lymphocyte collaboration in human lymphoid follicles. Roles of T cell — B cell activating molecule (5c8 antigen) and CD40 in contact dependent help. J Immunol 1992;149:3817–26. 45 Ridge JP, DiRosa F, Matzinger P. A conditioned dendritic cell can be a temporal bridge between a CD4+ T-helper and a T-killer cell. Nature 1998;393(6684):413–14. 46 Parker DC, Greiner DL, Phillips NE, Appel MC, Steele AW, Durie FH, et al. Survival of mouse pancreatic islet allografts in recipients treated with allogeneic small lymphocytes and antibody to CD40 ligand. Proc Natl Acad Sci USA 1995;92:9560–4. 47 Rossini AA, Parker DC, Phillips NE, Durie FH, Noelle RJ, Mordes JP, et al. Induction of immunological tolerance to islet allografts. Cell Transplant 1996;5:49–52. 48 Zheng XX, Markees TG, Hancock WW, Li Y, Greiner DL, Li XC, et al. CTLA4 signals are required to optimally induce allograft tolerance with combined donor-specific transfusion and anti-CD154 monoclonal antibody treatment. J Immunol 1999;162(8):4983–90. 49 Balasa B, Krahl T, Patstone G, Lee J, Tisch R, McDevitt HO, et al. CD40 ligand-CD40 interactions are necessary for the initiation of insulitis and diabetes in nonobese diabetic mice. J Immunol 1997;159:4620–7. 50 Molano RD, Berney T, Li H, Cattan P, Pileggi A, Vizzardelli C, et al. Prolonged islet graft survival in NOD mice by blockade of the CD40–CD154 pathway of T-cell costimulation. Diabetes 2001;50(2):270–6. 51 Kover KL, Geng Z, Hess DM, Benjamin CD, Moore WV. Anti-CD154 (CD40L) prevents recurrence of diabetes in islet isografts in the DR-BB rat. Diabetes 2000;49(10):1666–70.

N.S. KENYON

52 Dechanet J, Grosset C, Taupin JL, Merville P, Banchereau J, Ripoche J, et al. CD40 ligand stimulates proinflammatory cytokine production by human endothelial cells. J Immunol 1997;159:5640–7. 53 Kiener PA, Moran-Davis P, Rankin BM, Wahl AF, Aruffo A, Hollenbaugh D. Stimulation of CD40 with purified soluble gp39 induces proinflammatory responses in human monocytes. J Immunol 1995;155:4917–25. 54 Kenyon NS, Chatzipetrou M, Masetti M, Ranuncoli A, Oliveira M, Wagner J, et al. Long-term survival and function of intrahepatic islet allografts in rhesus monkeys treated with humanized antiCD154. PNAS 1999;96(14):8132–7. 55 Kenyon NS, Fernandez LA, Masetti M, Lehmann R, Ranuncoli A, Chatzipetrou M, et al. Long-term survival and function of intrahepatic islet allografts in baboons treated with humanized anti-CD154. Diabetes 199;48(7):1473–81. 56 Kenyon NS, Inverardi L, Alejandro R, Ricordi, C. On the pre-clinical results of islets and anti-CD154. Graft 2000;3(5):230–4. 57 Gotoh M, Fukuzaki T, Monden M, Dono K, Kanai T, Yagita H, et al. A potential immunosuppressive effect of anti-lymphocyte function-associated antigen-1 monoclonal antibody on islet transplantation. Transplantation 1994;57(1):123–6. 58 Nicolls MR, Coulombe M, Yang H, Bolwerk A, Gill RG. Anti-LFA-1 therapy induces long-term islet allograft acceptance in the absence of IFN-gamma or IL-4. J Immunol 2000;164(7):3627–34. 59 Pankewycz O, Mackie J, Hassarjian R, Murphy JR, Strom TB, Kelley VE. Interleukin-2-diphtheria toxin fusion protein prolongs murine islet cell engraftment. Transplantation 1989;47(2):318–22. 60 Li XC, Ima A, Li Y, Zheng XX, Malek TR, Strom TB. Blocking the common gamma-chain of cytokine receptors induces T cell apoptosis and long-term islet allograft survival. J Immunol 2000;164(3):1193–9. 61 Posselt AM, Barker CF, Tomaszewski JE, Markmann JF, Choti MA, Naji A. Induction of donorspecific unresponsiveness by intrathymic islet transplantation. Science 1990;249(4974):1293–5. 62 Oluwole SF, Jin MX, Chowdhury NC, James T, Fawwaz RA. Induction of specific unresponsiveness to rat islet allografts by intrathymic UVB donor spleen cells. Transplantation 1993;56(5):1142–7. 63 James T, Jin MX, Chowdhury NC, Oluwole SF. Tolerance induction to rat islet allografts by intrathymic inoculation of donor spleen cells. Transplantation 1993;56(5):1148–52. 64 Posselt AM, Odorico JS, Barker CF, Naji A. Promotion of pancreatic islet allograft survival by intrathymic transplantation of bone marrow. Diabetes 1992;41(6):771–5. 65 Qian T, Ricordi C, Shachner R, Inverardi L, Alejandro R. Tolerance induction to multiple donor rat islet allografts by intrathymic inoculation of spleen cell membrane antigens. Transplantation 1995;60(2):208–9. 67 Campos L, Posselt AM, Deli BC, Mayo GL, Pete K, Barker CF, et al. The failure to intrathymic transplantation of nonimmunogenic islet allografts to promote induction of donor-specific unresponsiveness. Transplantation 1994;57(6):950–3. 68 Turvey SE, Hara M, Morris PJ, Wood KH. Mechanisms of tolerance induction after intrathymic islet injection: determination of the fate of alloreactive thymocytes. Transplantation 1999;68(1):30–9. 69 Ali A, Garrovillo M, Jin MX, Hardy MA, Oluwole SF. Major histocompatibility complex class I peptide-pulsed host dendritic cells induce antigen-specific acquired thymic tolerance to islet cells. Transplantation 2000;69(2):221–6. 70 Tran HM, Patel A, Allen RD, O’Connell PJ. Intrathymic inoculation of donor antigen: an ineffective strategy for prolonging xenograft survival. Xenotransplantation 1999;6:147–54 71 Koevary SB, Blomberg M. Prevention of diabetes in BB/Wor rats by intrathymic islet injection. J Clin Invest 1992;89(2):512–6.

351

352

EXPERIMENTAL APPROACHES TO THE PREVENTION OF ISLET REJECTION

72 Posselt AM, Barker CF, Friedman AL, Naji A. Prevention of autoimmune diabetes in the BB rat by intrathymic islet transplantation at birth. Sciences 1992;256(5061):1321–4. 73 Charlton B, Auchincloss Jr H, Fathman CG. Mechanisms of transplantation tolerance. Ann Rev Immunol 1994;12:707–34. 74 Kostecke RA, Ildstad ST. Chimerism and the facilitating cell. Transplant Rev 1995;9:97–110. 75 Stewart FM, Crittenden RB, Lowry PA, Pearson-White S, Quesenberry PJ. Long-term engraftment of normal and post-5-fluorouracil murine marrow into normal nonmyeloablated mice. Blood 1993;81:2566–71. 76 Blumberg N, Heal JM. Effects of transfusion on immune function. Cancer recurrence and infection. Arch Pathol Lab Med 1994;118:371–9. 77 Jensen LS. Immunosuppression and leukocytes. Curr Stud Hematol Blood Transf 1994;60:64–74. 78 Brendel MD, Kong SS, Schachner RD, Qian T, Selvaggi G, Alejandro R, et al. The influence of donor specific vertebral body derived bone marrow cell infusion on canine islet allograft survival without irradiation conditioning of the recipient. Exp Clin Endocrinol Diabetes 1995;103(2):129–32. 79 Kenyon NS, Selvaggi G, Fernandez L, Xu X-m, Knapp J, Montelongo J, et al. Infusion of class II dim donor bone marrow enhances islet allograft survival in low dose CSA treated dogs. Transplant Proc 1997;29:2189. 80 LaFace DW, Peck AB. Reciprocal allogeneic bone marrow transplantation between NOD mice and diabetes — nonsusceptible mice associated with transfer and prevention of autoimmune diabetes. Diabetes 1989;38:894. 81 Li H, Kaufman CL, Boggs SS, Johnson PC, Patrene KD, Ildstad ST. Mixed allogeneic chimerism induced by a sublethal approach prevents autoimmune diabetes and reverses insulitis in nonobese diabetic (NOD) mice. J Immunol 1996;156(1):380–8. 82 Mathieu C, Casteels K, Bouillon R, Waer M. Protection against autoimmune diabetes in mixed bone marrow chimeras. J Immunol 1997;158:1453–7. 83 Wekerle T, Sayegh MH, Hill J, Zhao Y, Chandraker A, Swenson KG, et al. Extrathymic T cell deletion and allogeneic stem cell engraftment induced with costimulatory blockade is followed by central T cell tolerance. J Exp Med 1998;187(12):2037–44. 84 Wekerle T, Kurtz J, Ito H, Ronquillo JV, Dong V, Zhao G, et al. Allogeneic bone marrow transplantation with costimulatory blockade induces macrochimerism and tolerance without cytoreductive host treatment. Nature Med 2000;6(4):464–9. 85 Durham MM, Bingaman AW, Adams AB, Ha J, Waitze SY, Pearson TC, et al. Cutting edge: administration of anti-CD40 ligand and donor bone marrow leads to hemopoietic chimerism and donor-specific tolerance without cytoreductive conditioning. J Immunol 2000;165(1):1–4. 86 Li H, Kaufman CL, Ildstad ST. Allogeneic chimerism induces donor-specific tolerance to simultaneous islet allografts in nonobese diabetic mice. Surgery 1995;118(2):192–7. 87 Li H, Colson YL, Ildstad ST. Mixed allogeneic chimerism achieved by lethal and nonlethal conditioning approaches induces donor-specific tolerance to simultaneous islet allografts. Transplantation 1995;60(6):523–9. 88 Faustman DL, Steinman RM, Gebel HM, Hauptfeld V, Davie JM, Lacy PE. Prevention of rejection of murine islet allografts by pretreatment with anti-dendritic cell antibody. Proc Natl Acad Sci 1984;81(12):3864–8. 89 Faustman D, Hauptfeld V, Lacy P, Davie J. Prolongation of murine islet allograft survival by pretreatment of islets with antibody directed to Ia determinants. Proc Natl Acad Sci 1981;78(8):5156–9. 90 Edelman G, Duke D, Renn E, Jonasson O. Fetal islet allotransplantation in rabbits. Transplantation 1988;46(5):660–4.

N.S. KENYON

91 Kenyon NS, Strasser S, Alejandro R. Ultraviolet light immunomodulation of canine islets for prolongation of allograft survival. Diabetes 1990;39(3):305–11. 92 Alejandro R, Latif Z, Noel J, Shienvold FL, Mintz DH. Effect of anti-Ia antibodies, culture, and cyclosporin on prolongation of canine islet allograft survival. Diabetes 1987;36(3):269–73. 94 Stegall MD, Chabot J, Weber C, Reemtsma K, Hardy MA. Pancreatic islet transplantation in cynomolgus monkeys. Initial studies and evidence that cyclosporin impairs glucose tolerance in normal monkeys. Transplantation 1989;48(6):944–50. 95 Shienvold FL, Alejandro R, Mintz DH. Identification of Ia-bearing cells in rat, dog, pig, and human islets of Langerhans. Transplantation 1986;41(3):364–72. 96 Coulombe M, Yang H, Guerder S, Flavell RA, Lafferty KJ, Gill RG. Tissue immunogenicity: the role of MHC antigen and the lymphocy costimulator B7-1. J Immunol 1996;157(11):4790–5. 97 Steurer W, Nickerson PW, Steele AW, Steiger J, Zheng XX, Strom TB. Ex vivo coating of islet cell allografts with murine CTLA4/Fc promotes graft tolerance. J Immunol 1995;155(3):1165–74. 98 Coulombe M, Yang H, Wolf LA, Gill RG. Tolerance to antigen-presenting cell-depleted islet allografts is CD4 T cell dependent. J Immunol 1999;162(5):2503–10. 99 Faustman D. Strategies for circumventing transplant rejection: modification of cells, tissues, and organs. Trends Biotechnol 1995;12(3):100–5. 100 Brady JL, Lew AM. Additive efficacy of CTLA4immunoglobulin and OX40immunoglobulin secreted by genetically modified grafts. Transplantation 2000;69(5):724–30. 102 Chahine AA, Yu M, McKernan MM, Stoeckert C, Lau HT. Immunomodulation of pancreatic islet allografts in mice with CTLA4immunoglobulin secreting muscle cells. Transplantation 1995;59(9):1313–18. 103 Lau HT, Yu M, Fontana A, Stoeckert CJ Jr. Prevention of islet allograft rejection with engineered myoblasts expressing FasL in mice. Science 1996;273(5271):109–12. 105 Turvey SE, Gonzalez-Nicolini V, Kingsley CI, Larregina AT, Morris PJ, Castro MG, et al. Fas ligandtransfected myoblasts and islet cell transplantation. Transplantation 2000;69(9):1972–6. 106 Korbutt GS, Suarez-Pinzon WL, Power RF, Rajotte RV, Rabinovitch A. Testicular Sertoli cells exert both protective and destructive effects on syngeneic islet grafts in non-obese diabetic mice. Diabetologia 2000;43(4):474–80. 107 Takeda Y, Gotoh M, Dono K, Nishihara M, Grochowiecki T, Kimura F, et al. Protection of islet allografts transplanted together with Fas ligand expressing testicular allografts. Diabetologia 1998;41(3):315–21. 108 Gainer AL, Suarez-Pinzon WL, Min WP, Swiston JR, Hancock-Friesen C, Korbutt GS, et al. Improved survival of biolistically transfected mouse islet allografts expressing CTLA4immunoglobulin or soluble Fas ligand. Transplantation 1998;66(2):194–9. 109 Kaufman DB, Platt JL, Rabe FL, Dunn DL, Bach FH, Sutherland DE. Differential roles of Mac-1+ cells, and CD4+ and CD8+ T lymphocytes in primary nonfunction and classic rejection of islet allografts. J Exp Med 1990;172(1):291–302. 110 Shapiro AM, Lakey JR, Ryan EA, Korbutt GS, Toth E, Warnock GL, et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoidfree immunosuppressive regimen. N Engl J Med 2000;343(4):289–90.

353

This page intentionally left blank

Chapter 25

Clinical effectiveness of islet transplantation Alberto M. Davalli, Federico Bertuzzi, and Antonio Secchi

Within the past 10 years, 329 new cases of adult islet allotransplantation and 145 cases of islet autotransplantation have been recorded by the International Islet Transplantation Registry [1]. Collectively, the results indicate that intrahepatic implantation of human islets can effectively replace the function of the endocrine pancreas and restore a near normal glucose homeostasis. However, clinical trials also show that the outcome of islet transplantation differs considerably depending on whether the graft is aimed at preventing or curing diabetes. Islet autotransplantation, performed to prevent diabetes in patients undergoing total pancreatectomy for painful chronic pancreatitis, restores normoglycaemia and independence from exogenous insulin administration in the majority of recipients [2]. Similarly, islet allotransplantation effectively prevents diabetes in patients treated with upper abdominal exenteration for extensive abdominal cancer, with 60 per cent remaining insulin independent after transplantation [3]. In sharp contrast with the impressive results obtained in the prevention of surgical diabetes, the success rate of islet allotransplantation aimed at treating spontaneous diabetes is dramatically lower. The different outcome of this procedure in these two clinical settings is likely to reflect the levels and complexity of the problems facing the islets after implantation. In this chapter, we briefly review the results of islet transplantation in humans to highlight the major findings and lessons clinical trials have so far provided. The aim is to outline the major outstanding questions that need to be answered to improve the clinical effectiveness and metabolic performance of islet transplantation in humans.

Islet autotransplantation to prevent surgical diabetes Islet autotransplantation in chronic pancreatitis Islet autotransplantation is currently performed in an attempt to prevent diabetes after extensive pancreatic resection for pain relief in patients with chronic pancreatitis. Since the earliest report [4], several studies have been published showing progressively improving results in parallel with improvements in islet isolation techniques [2,5–9]. Autotransplantation offers the opportunity to study the function of the implanted islets without most of the confounding factors common in other transplantation settings. Immune rejection of transplanted islets, recurrence of autoimmune destruction of transplanted ␤-cells, and diabetogenicity of immunosuppressive drugs cannot be blamed for impaired function of islet autografts. Therefore, the relevance of other variables, such as the quality, mass, and purity of the islet preparations, can be assessed more rigorously. Studies in autotransplantation have shown that islet purification is not a significant determinant of the graft success and that the

356

CLINICAL EFFECTIVENESS OF ISLET TRANSPLANTATION

infusion of up to 50 ml of digested tissue in the portal vein is not associated with major adverse effects if adequate precautions, such as the monitoring of the intraportal vein pressure, are taken [2,9]. Clinical effectiveness of islet autotransplantation is impressive: 50 per cent of recipients are insulin independent over a year after transplantation, and sustained insulin independence after 2 years is achieved in about 35 per cent of cases, and in some patients maintained for more than 7 years [2]. Islet autografts have shown that the number of transplanted islets is the most important factor predicting the success of the graft, with 74 per cent of the recipients insulin independent over than 2 years after transplantation, if more than 300 000 islet equivalents (IEO) were transplanted [2]. Islet autotransplantation has also been successfully performed in a paediatric recipient who became insulin independent after implantation of a relatively small islet mass [10]. However, about 2 years after implantation, likely due to a 65 per cent increase in the patient’s body weight, the graft started to fail and exogenous insulin was required, demonstrating that an initially adequate ␤-cell mass was incapable of meeting the increased demand of normal growth and development. Thus, even though the successful outcome of autotransplantation is generally reassuring about the value of the liver as an islet transplantation site, pancreatic ␤-cells appear to lose their normal replicative potential when etherotopically implanted into the liver. Alternatively, substantial apoptotic ␤-cell death may occur in intrahepatically transplanted islets without being compensated by a sufficient growth of new ␤-cells.

Insulin and glucagon secretion by autografted islets Autotransplantation has also provided important insights about the secretion patterns of intrahepatically implanted islets. In spite of normal fasting glucose concentrations, normal glycosylated haemoglobin A 1c (HbA 1c ) values and normal timing of insulin secretory responses to intravenously administered glucose, the magnitude of the insulin response of autografted patients was significantly lower than in normal subjects [9]. Similar results have been observed in healthy living donors submitted to hemipancreatectomy for pancreatic tissue transplant in type 1 diabetic relatives [11] and suggest a condition of suboptimal ␤-cell mass. Conversely, insulin and C-peptide responses to arginine were similar to those of normal controls. In successfully autotransplanted patients, glucose potentiation of arginine-induced insulin secretion (GPAIS) has also been studied to determine the graft insulin secretory reserve. Because GPAIS is the most sensitive indicator of subclinical alterations in ␤-cell function, as shown by GPAIS levels about 50 per cent of normal [12] in hemipancreatectomized patients, it was thought to be more informative than other insulin secretory tests. Indeed, recipients of islet autografts had a significantly reduced ␤-cell secretory reserve. However, responses to GPAIS were not superior to arginine- and glucose-induced insulin responses during fasting to evaluate engraftment [13]. All these three indicators correlated well with the number of transplanted islets, demonstrating their possible use in estimating the surviving islet mass. Intrahepatically autotransplanted islets characteristically show an alteration in the physiological regulation of glucagon secretion. Baseline plasma glucagon and pancreatic polypeptide secretion, as well as their enhancement by insulin-induced hypoglycaemia, were reduced in autotransplanted patients compared to normal subjects. None of these patients had a glucagon response to hypoglycaemia but all responded to arginine administration [14]. Lack of islet innervation, decreased total ␣cell mass and/or abnormal glucagon secretion due to the hepatic location, might all be potential causes for the abnormal ␣-cell physiology. Using a canine model of autotransplantation, it has been shown that islet transplanted into the peritoneum, but not into the liver, have an intact glucagon secretory response to hypoglycaemia [15], suggesting that a transplantation site other than the liver may be required to achieve normal glucagon secretion. However, defective controregulatory glucagon

A. M. DAVALLI ET AL.

secretion does not appear of particular concern in successfully islet autotransplanted patients, who are not treated with exogenous insulin and thus are not at risk of hypoglycaemia. Conversely, abnormal counter-regulation might represent a problem in insulin-treated recipients of failing islet autografts because severe hypoglycaemic events may occur. This potential risk, however, does not contraindicate autotransplantation because the majority of patients with chronic pancreatitis, eventually developing a form of secondary diabetes characterized by unregulated low glucagon levels, are also at high risk of severe hypoglycaemia on insulin therapy [16].

Future endeavours A still unexplored issue that deserves attention is whether or not the proinsulin to insulin ratio is increased in islet autotransplanted patients, as seen in hemipancreatectomized patients [12]. It has already been shown that a disproportionately high proinsulin secretion can result from sustained activation of human ␤-cells following prolonged in vitro exposure to elevated glucose levels [17]. Indeed, in the face of the reduced functional insulin secretory reserve, grafted islets may compensate the increased metabolic demand by recruiting immature secretory granules. We speculate that the higher the proinsulin to insulin ratio, the lower the insulin secretory reserve and that, above a certain threshold level, it might predict incipient graft failure due to ␤-cell functional exhaustion. The importance of measuring other proinsulin products in the plasma of islet-transplanted patients is indicated by the elevated split proinsulin levels in autograft recipients requiring exogenous insulin despite normal Cpeptide levels [18]. The reported cross-reactivity of high split proinsulin levels with C-peptide assay [18] suggests that the simple measurement of C-peptide levels might be inadequate for monitoring islet graft function. A significant percentage (30 per cent) of autografts is doomed to fail. These failures are only partially correlated with the mass and purity of the transplanted islets and suggest that other factors might be involved. They include poor islet viability, post-transplantation metabolic control before graft functioning, and ␤-cell loss due to hypoxic injury and non-specific inflammation in the transplantation site [19]. Islet autotransplantation thus appears the ideal setting to explore novel in vitro viability tests predictive of islet function, and strategies aimed at improving islet engraftment.

Islet allotransplantation to prevent surgical diabetes Islet allotransplantation in patients with abdominal cancer Allogeneic islets have been successfully transplanted, simultaneously with the liver, in patients undergoing pancreatectomy and hepatectomy for extensive abdominal cancer [3,20,21]. In contrast with autotransplantation, in this setting, transplanted islets have to face rejection and the toxicity of immunosuppressive drugs. Islets obtained from the same liver donor, from a third party donor or both, were cotransplanted with the liver in patients immunosuppressed only with tacrolimus. Even though the liver transplantation might have down-modulated islet rejection, the results have been impressive: islet function was detected in every recipient and nine patients out of 15 (60 per cent) remained insulin independent [20]. Eventually, all the recipients died from recurrent metastatic disease but a year after transplantation six of them were still normoglycaemic, 5 years being the longest insulin independence. Basal C-peptide responses to intravenous glucose challenge were detected in all recipients. In patients with normal basal glycaemic and HbA1c levels, C-peptide increased over three-fold in response to intravenous glucose, but blood glucose did not return to prestimulation levels until 180 min after glucose infusion. This abnormal tolerance to intravenous glucose

357

358

CLINICAL EFFECTIVENESS OF ISLET TRANSPLANTATION

has been related to altered C-peptide secretory kinetics, showing the lack of first-phase release and peak response delayed at 60 min [20]. These findings are somehow different from those reported in autotransplanted patients, who showed a normal timing of insulin release and demonstrated that the intrahepatic islet location does not determine per se abnormal insulin secretory dynamics. Conversely, the lack of first-phase insulin release observed in these patients is reminiscent of situations of substantial reduction of ␤-cell mass and of the prodrome of spontaneous diabetes [22]. However, the patients who remained insulin independent after islet allotransplantation received an islet mass particularly elevated in relation to their body weight and, obviously, recurrence of autoimmunity cannot be responsible for ␤-cell destruction. In addition, the metabolic demand placed on the transplanted islets could have been relatively scarce in these subjects who, like other anorectic patients, may be particularly sensitive to insulin [23]. Therefore, other causes of ␤-cell loss might be involved in these patients and normal liver function indicates tacrolimus-induced ␤-cell damage, rather than chronic rejection. Tacrolimus has direct toxic effects on human islets as shown by more frequent and severe islet damage, documented by biopsies of the transplanted pancreas, in recipients treated with FK506 compared to cyclosporin A [24]. However, the combination of low-dose tacrolimus with other immunosuppressive drugs is expected to minimize diabetogenicity, and is currently under investigation in different transplantation trials.

Islet allotransplantation to cure type 1 diabetic patients Clinical trials of islet allotransplantation in type 1 diabetic patients Allogeneic islet transplantation can restore insulin production in C-peptide negative patients with type 1 diabetes mellitus [25–29]. Islets are usually transplanted into the liver of long-lasting type 1 diabetic patients requiring immunosuppression for a prior (islets after kidney, IAK) or simultaneous (simultaneous islets and kidney, SIK) kidney graft performed for endstage diabetic nephropathy. Although the rate of success of IAK versus SIK differs among the different centres, in general, their clinical effectiveness appears similar [1]. Before analysing the outcome of islet allotransplantation in type 1 diabetic patients, it should be emphasized that in this case islet grafts are asked to cure, and not simply prevent, diabetes. This obvious distinction is not always adequately considered in the analysis of the results. Indeed, patients with long-lasting diabetes have metabolic and vascular abnormalities which may impair the engraftment and long-term survival of the implanted islets. These patients are usually insulin resistant as a consequence of long-lasting hyperglycaemia and chronic insulin treatment. Insulin resistance, by increasing the metabolic demand placed on transplanted ␤-cells, might facilitate their functional exhaustion. Diabetic microangiopathy might also interfere with islet revascularization. Moreover, early graft losses due to non-specific inflammatory reaction at the site of implantation [30] are likely to be augmented in the diabetic autoimmune environment [31]. Furthermore, the hyperglycaemic activity of plasma glucagon, usually increased in type 1 diabetic patients [32], represents an additional challenge for the transplanted islets. The function of islet transplants in type 1 diabetic patients is negatively affected by diabetogenic immunosuppressive regimens, in particular steroids, ciclosporin A, and tacrolimus. Recurrence of autoimmune-mediated ␤-cell destruction has been documented in type 1 diabetic patients receiving islet grafts [33] and islet allograft survival is significantly shorter in recipients with autoantibodies (anti-GAD65 or ICA) compared to autoantibody-negative recipients at the time of implantation [34]. It has been recently shown that a minority of patients receiving pancreas allografts under generalized immunosuppression show a stimulation of islet autoantibody reac-

A. M. DAVALLI ET AL.

tivity, which is almost invariably followed by graft function failure and resumption of insulin therapy [35]. This study suggests that post-transplantation increase of diabetes autoantibodies, rather than their presence before transplantation, might be the real predictor of pancreas and islet grafts failure caused by recurrence of autoimmunity. A detailed analysis of the 200 patients with type 1 diabetes mellitus transplanted with adult islets between 1990 and 1997 is available at the International Islet Transplantation Registry [1]. Clinical trials have shown that graft survival at 1 year, defined as basal C-peptide over 0.5 ng/ml, was 35 per cent, with 8 per cent of the recipients insulin independent. Insulin independence is facilitated by the following conditions: (a) more than 6000 IEQ per kilogram of recipient’s body weight are transplanted; (b) islets are obtained from pancreata with a cold ischaemia time of less than 8 h; (c) islets are transplanted into the liver; and (d) the induction of immunosuppression includes monoclonal or polyclonal T-cell antibodies. When all these conditions were met, 48 per cent of the recipients showed basal C-peptide levels of over 0.5 ng/ml, 73 per cent had HbA1c levels less than 7 per cent, and 22 per cent were insulin independent over 1 year after transplant. Basal C-peptide levels in insulinindependent patients were stably in the 2 to 2.5 ng/ml range 1 month and 1 year after transplantation. Conversely, in patients still requiring insulin, C-peptide levels declined to less than 1.0 ng/ml at 1 year. Return to normoglycaemia has also been achieved when the implanted tissue was mainly composed of cryopreserved islets [26], or unpurified islet preparations transplanted intrahepatically together with a kidney graft [28]. Patients receiving unpurified preparations were treated with 15-deoxyspergualin to induce immunosuppression, and azathioprine, prednisone, and ciclosporin to maintain it. Both recipients, grafted with 7800 and 10 300 IEQ per kilogram of body weight, respectively, showed clear evidence of graft function and the patient receiving the larger islet mass became insulin independent.

The Milan experience The first islet graft was performed in Milan in 1989, in this early series of islet allotransplantations in type 1 diabetic patients, evidence of graft function was obtained in five out of six recipients, with reduction of exogenous insulin requirement above 50 per cent of pretransplantation values in four patients, and complete insulin independence in one patient [27]. In a more recent experience seven out of 20 type 1 diabetic patients became insulin independent and two additional patients experienced a 50 per cent decreased exogenous insulin requirement with sustained C-peptide secretion; eight out of nine received islets after a prior kidney graft [29]. Three patients showed high C-peptide levels immediately after transplantation, followed by undetectable C-peptide levels presumably as a consequence of acute rejection. One patient in this cohort did not receive antilymphocyte globulin (ALG) due to previous cytomegalovirus uveitis, while antithymocyte globulin (ATG) was suspended in another patients for serum sickness. In this latter patient, C-peptide secretion was lost concurrently with a steroid-resistant kidney rejection. Functional exhaustion of the transplanted islets was considered responsible for graft failure in a third group of patients (40 per cent), where the post-transplantation increase in C-peptide levels was not accompanied by any metabolic improvement. In these patients, insulin requirement increased after transplantation and C-peptide disappeared in about 4 months. The number of islets transplanted was significantly higher in the successful cases than in patients with presumed functional exhaustion (IEQ 11 137 ± 907 versus 7487 ± 1025, P < 0.05). In our experience, the longest period of insulin independence was of 4 years, in a patient who received about 10 000 IEQ per kilogram of body weight and who died of myocardial infarction while still off exogenous insulin and with near-normal BhA1c levels. Histology showed the presence of well-granulated islets evenly spread in the portal spaces of the whole liver without evidence of lymphomononuclear cell infiltration (Fig. 25.1).

359

360

CLINICAL EFFECTIVENESS OF ISLET TRANSPLANTATION

Fig. 25.1 Histological analysis of the autopic liver of a type 1 diabetic patient who underwent islet allotransplantation. The recipient died, 50 months after islet transplantation, of myocardial infarction while she was still in good metabolic control (HbA1c < 7 per cent) in the absence of exogenous insulin administration. Serial sections are shown of representative intrahepatic islets stained for insulin (Ins) and glucagon (Gluc). Islets were localized in the portal spaces and did not show evidence of lymphomononuclear cellular infiltration. Insulin immunoreactivity was particularly intense, indicating a good hormone storage by the engrafted ␤-cells. Glucagon-positive cells were also easily detected while somatostatin and pancreatic polypeptide-secreting cells were never observed (not shown) (Magnification × 200). From Davalli A.M., Maffi P., Socci C., Sanvito F., Freschi M., Bertuzzi F., Falqui L., Di Carlo V., Pozza G., Secchi A. Insights from a successful case of intrahepatic islet transplantation into a type 1 diabetic patient, J Clin Endocrinol Metab; 2000; 85(10); 3847–52. © The Endocrine Society.

In addition to the islet mass, another parameter that positively influenced the success of the implantation was the strict metabolic control of the recipients in the early postsurgical period, achieved by intravenous insulin administration to keep glycaemic levels constantly below 150 mg/dl. The rationale for this intensive insulin treatment is based on the exquisite susceptibility of human islets to the toxic effects of high glucose concentrations [36].

Metabolic studies in successfully transplanted patients Insulin-independent patients showed near normal fasting glucose levels, mild postprandial hyperglycaemia in a 24-h metabolic profile (Fig. 25.2), and a prompt insulin release with a peak at 5 min in response to arginine [29]. Quantitatively, the insulin response to arginine in successfully transplanted patients was significantly lower than in whole pancreas recipients, but it was similar to that of patients

A. M. DAVALLI ET AL.

Fig. 25.2 Twenty-four hour metabolic profiles in type 1 diabetic patients 3 to 6 months after transplantation of whole (▲ 15 patients) and segmental (● 15 patients) pancreatic grafts and dispersed islets (■ 6 patients). Upper panel depicts blood glucose excursions before and after meals, in the lower panel are the peripheral insulin levels at the same time periods. Recipients of dispersed islet grafts show higher basal and postprandial glycaemic levels than whole and segmental pancreas recipients. Basal and postprandial insulin secretion were similar in segmental pancreas and islet graft recipients that, however, were both significantly reduced compared with insulin responses detected in whole pancreas recipients. B, breakfast; L, lunch; D, dinner.

who received a segmental pancreatic graft (Fig. 25.3) [37]. Continued insulin dependence, despite normal peripheral insulin sensitivity and normal C-peptide levels, has been reported in a diabetic patient transplanted simultaneously with kidney and islets [38], suggesting the presence of an abnormally elevated glucose production by the liver as a consequence of inadequate hepatic insulinization. Whether intrahepatic islets provide portal or systemic insulin delivery, or a proportion of insulin is retain in the liver, is unclear. To explore this issue, basal hepatic glucose production, whole body glucose homeostasis, and insulin action were measured in eight type 1 diabetic patients who became insulin independent after islet transplantation [39]. In spite of the abnormal insulin secretion pattern characterized by the impairment of both first- and second-phase insulin release during the hyperglycaemic clamp, successful intraportal islet grafts normalized basal hepatic glucose production and improved total tissue glucose disposal that, however, remained lower than in normal control subjects. Surprisingly, no correlation was found between fasting hepatic glucose production and fasting Cpeptide levels, and number of transplanted islets. The selective loss of glucose-induced insulin secretion in response to intravenous glucose, observed also by islet allografts performed to prevent surgical diabetes, is a fundamental characteristic of the hyperglycaemic state. This phenomenon has been rec-

361

362

CLINICAL EFFECTIVENESS OF ISLET TRANSPLANTATION

Fig. 25.3 Insulin secretory responses to L-arginine infusions (upper panel) and IVGTTs (lower panel) performed 3 to 6 months after transplantation of whole (▲) and segmental (●) pancreatic grafts and dispersed islets (■). Insulin responses to intravenous glucose were significantly higher in whole than in segmental pancreas recipients who, in turn, showed higher responses than islet graft recipients. Quantitative insulin responses to L-arginine were also higher in whole pancreas recipients than in the other groups, but were similar in segmental pancreas and islet grafts recipients. The selective loss of glucose-induced insulin secretion in response to intravenous glucose, observe in islet-transplanted patients, might be due to the mild chronic hyperglycaemia detected in these patients.

ognized in not only type 2 (non insulin-dependent) diabetes but also in early type 1 diabetes [22,40], in failing pancreas transplants [41,42], and in rodents with islet transplantation exposed to experimental hyperglycaemia [43 44]. Near post-transplantation normalization of glucose metabolism in successfully transplanted patients might not be sufficient to prevent the desensitization of the glucose sensor unit in transplanted ␤-cells. Perhaps a more complete normalization of glucose homeostasis (that is, lack of postprandial hyperglycaemia) is required to achieve this goal. The presence of normal hepatic glucose production in spite of elevated postabsorptive glucagon concentrations in successfully transplanted patients is also remarkable. Indeed, basal glucagon levels were found to be significantly higher in these patients and in patients with non-functioning islet grafts, compared to normal healthy subjects [39]. Type 1 diabetic patients are already hyperglucagonaemic but plasma glucagon levels may be further increased by islet transplantation because, along with ␤-cells, the recipient also receives a large mass of glucagon-secreting ␣-cells. An important decrease in ␤-cell mass of human islets transplanted into diabetic rodents occurs immediately after transplantation, with a dramatic decrease during the first 15 days but persisting even after revascularization has occurred [45]. Conversely, the endocrine non-␤-cell mass remains stable, indicating that

A. M. DAVALLI ET AL.

engrafted non-␤-cells are more resistant than ␤-cells. The non-␤/␤-cell ratio of a human islet graft can increase over time [45], suggesting that a change in islet structure could influence its function. In a failed graft, due to functional exhaustion of the ␤-cells or recurrence of autoimmunity, but not rejection, ␣-cells might be the only surviving islet cells. In this scenario, a failing islet graft might even aggravate the metabolic control of the recipient. The comparison between pre- and postransplantation glucagon levels could shed light on the mechanisms of graft failure and indicate autoimmune destruction and/or functional exhaustion versus rejection.

Analysis of the cellular composition of islet grafts The characterization of the islet graft in terms of its cellular composition [␣-cells, ␤-cells, ␦-cells, duct cells, major histocompatibility complex (MHC) class II positive cells, and percentage of damaged cells], in addition to ␤-cell number, and insulin synthesis, has provided interesting information [46]. In this study, seven type 1 diabetic patients received 1 to 2 million ␤-cells per kilogram of body weight, corresponding to 2600 to 5300 IEQ calculated by dithizone staining. Three out of seven patients remained C-peptide positive for more than 1 year and two become insulin independent with near normal fasting glycaemia and HbA1c levels. Interestingly, the two patients who become insulin independent received significantly lower IEQ per kilogram of body weight than usually transplanted in the other successful cases [1]. The prolonged period of culture utilized in this study might have selected the most efficient ␤-cells, thus explaining their performance. None of the pretransplantation characteristics of the islet preparations was predictive of the transplantation success indicating that graft failure was dependent on in vivo events. Two of the 4 patients who experienced graft failure were initially GAD65-antibody positive and exhibited a transient rise in their titre during graft destruction. Therefore, this study confirms that the analysis of GAD65 antibodies is a useful marker of the autoimmune response against the transplanted ␤-cells and may become a negative selection criterion for islet cell transplantation. Interestingly, the three patients who experienced long-term graft function had all been treated with ATG during the prior kidney transplantation, whereas all four patients with graft failure had not. The comparison of islet graft-specific alloreactive T-cell responses in the peripheral blood of successful and unsuccessful cases showed no sign of alloreactivity in the former group. Conversely, graft failure was accompanied by an enhanced frequency of graft-specific alloreactive T cells [47]. Interestingly, implantation of islet allografts sharing human leucocyte antigens (HLA) with the previous kidney graft, did not result in alloreactivity towards the repeated HLA mismatches, indicating that islet grafts do not jeopardize the function of established kidney grafts and suggesting a role for the kidney graft in the induction of tolerance to shared islet antigens [48].

Identification of novel predictors of islet integrity before transplantation As previously mentioned, none of the in vitro quality tests utilized to evaluate islet quality proved successful in predicting graft outcome so far. Proper islet function is an obvious prerequisite for successful transplantation and novel in vitro parameters that may predict islet function in vivo are urgently needed. An analysis performed in our institute in a small panel of type 1 diabetic patients submitted to islet allotransplantation has indicated that pretransplantation monitoring of cytosolic islet cell Ca2+ concentration ([Ca2+]i] might accomplish this goal. A statistical difference was found in basal [Ca2+]i values between the islets that conferred insulin independence and those that reduced insulin requirement after transplantation [49]. Furthermore, basal [Ca2+]i showed negative correlation with: (a) insulin peak after intravenous glucose tolerance test (IVGTT) at 3 months (n = 7, P = 0.04, r2 =

363

364

CLINICAL EFFECTIVENESS OF ISLET TRANSPLANTATION

0.581); (b) insulin ⌬ secretion area after IVGTT at 3 months (n = 7, P = 0.03, r2 = 0.571); and (c) reduction of insulin requirement (n = 8, P = 0.02, r2 = 0.60). We consider these preliminary data encouraging and we are currently validating the potentiality of [Ca2+]i measurements in a larger group of transplanted patients.

Adjuvant therapy to improve islet engraftment and survival A number of drugs aimed at promoting islet engraftment have been recently introduced in the post-transplantation therapeutic management of type 1 diabetic patients receiving islet grafts. Drugs currently utilized clinically as ‘adjuvant’ therapy are pentoxifylline, verapamil, vitamin E, and nicotinamide (vitamin PP). Pentoxifylline, a methylxanthine derivative with rheological properties that also modulates immune function by decreasing tumour necrosis factor (TNF)- ␣ release by human macrophages [50] and reducing serum TNF-␣ levels, should reduce macrophage-induced nonspecific inflammatory reaction at the site of islet implantation. Verapamil exerts a protective effect on transplanted islets by reducing ciclosporin-mediated toxicity [51]. Vitamin E and PP are thought to reduce the oxidative stress and the damage induced by free radicals on transplanted islets. Vitamin E has also been shown to promote the engraftment of islet xenografts by attenuating leucocyte-endothelial cell interactions and inhibiting microvascular rejection [52]. Nicotinamide, in association with desferrioxamine, protects islets from allograft rejection in mice with chemical [53] and autoimmune diabetes [31]. The real contribution of the adjuvant therapy on the outcome of islet allografts in humans is still uncertain. However, in consideration of their safety profiles, vitamin E and nicotinamide should not be negated to islet graft recipients. An additional strategy to improve long-term islet survival may be to reduce insulin resistance in the recipient to diminish the metabolic demand placed on the transplanted islets. To this end, we tested the effect of metformin administration on the function of islet allografts in type 1 diabetic patients. Metformin is an antidiabetic drug widely used in patients with type 2 diabetes and obesity, which acts by increasing the sensitivity of peripheral tissues to the action of insulin, thereby allowing the achievement of normal glucose homeostasis in spite of inadequate insulin secretion. Increasing the sensitivity to insulin of peripheral tissues in patients undergoing islet transplantation might put transplanted islets at partial rest, thereby reducing the risk of their functional exhaustion. Moreover, it has been recently shown that metformin restores normal secretory patterns in islets whose function has been impaired by chronic exposure to elevated free fatty acids or glucose levels [54]. Compared with historical data, our recent results suggest that metformin may contribute to the improvement of the outcome of islet grafts in diabetic recipients, mainly by decreasing the number of grafts lost to functional exhaustion (Table 25.1) [55]. However, the putative beneficial effect of metformin needs confirmation in future by specifically designed case-control studies.

Solitary islet transplantation in type 1 diabetic patients An important contribution to the field of islet transplantation has been recently given by the trial of ‘solitary’ islet transplantation in type 1 diabetic subjects performed in Edmonton. Graft recipients were characterized by metabolic instability and severe hypoglycaemic episodes. In these patients with brittle diabetes or reduced hypoglycaemia awareness, the benefits of achieving stable blood glucose control by islet transplantation was considered worth the immunosuppression required to prevent graft rejection. In the first series, seven patients were transplanted under the cover of a steroid-free immunosuppressive therapy consisting of sirolimus, tacrolimus, and daclizumab. The islet mass transplanted in these patients was 11 547 ± 1640 IEQ requiring the utilization of two or three

A. M. DAVALLI ET AL.

Table 25.1 Clinical outcome of islet allografts performed in type 1 diabetic patients: group A recipients treated with ‘adjuvant’ therapy; group B historical matched controls Graft outcome

Rejection Functional exhaustion Successful Insulin independence

Group A (n = 2 1 )

Group B (n = 1 3 )

3 (14%) 3 (14%) 17 (80%) 12 (57%)

1 (8%) 4 (31%) 9 (70%) 7 (54%)

Adjuvant therapy consisted of: pentoxifylline 400 mg twice a day, nicotinamide 250 mg three times a day, and metformin 400 mg twice a day. Recipients were matched for number of IEQ transplanted per kilogram of body weight. Successful grafts were considered those that induced a decrease of exogenous insulin requirement over 50 per cent of the pretransplantation values.

pancreases. Results were impressive, showing sustained insulin independence after transplantation in all recipients [56]. Along with complete insulin independence, all transplanted patients achieved normalization of HbA1c levels, reduction of the glycaemic excursions, and, most importantly, did not experience further episodes of hypoglycaemic coma. A more accurate study of the clinical outcome and insulin secretion of these grafts was presented in a second report that showed data obtained from 12 patients with a median follow-up of about 10 months [57]. Results showed the permanence of sustained insulin production and normalization of glycaemic and HbA1c levels. Intravenous glucose tolerance tests, however, showed that only four patients had normal glucose tolerance, while five had impaired glucose tolerance. Moreover, three patients had recurrence of diabetes and required lowdose exogenous insulin. Immunosuppressive regimen, most likely sirolimus, induced in five patients an increase in serum cholesterol levels and lipid-lowering therapy was required in three of them. Of certain concern was the report that in two patients, who had elevated serum creatinine levels before transplantation, a further and sustained increased in serum creatinine was observed after transplantation. In conclusion, the Edmonton trial has shown that in subjects with labile type 1 diabetes difficult to control, the risk-to-benefit ratio is in favour of a solitary islet transplantation. However, the complications of the steroid-free immunosuppressive protocol used in this trial contraindicate the applications of islet transplantation in people with stable diabetes without any significant complications.

Concluding remarks In conclusion, the evidence that islet allotransplantation in type 1 diabetic patients is substantially less successful than in the prevention of surgical diabetes should not be surprising. Actually, given the many hurdles involved, the achievement of even temporary insulin independence in 12 per cent of patients with type 1 diabetes should be considered a remarkable accomplishment. The accumulated experience will help to develop strategies aimed at solving the obstacles, that have been now clearly defined. In the near future, the appropriate selection of recipients, the availability of non-diabetogenic immunosuppression, the avoidance of steroids, and the intelligent use of drugs capable of reducing insulin resistance and non-specific inflammatory responses, are expected to improve substantially the outcome of islet transplantation in type 1 diabetic patients. Of course, the ultimate aim of islet transplantation remains to cure type 1 diabetic patients before the chronic degenerative complications of diabetes appear. At the time of diagnosis, most of these patients maintain a significant residual ␤-cell mass that can survive for a prolonged period of time if ␤-cells are kept at rest by intensive insulin treatment [58]. Similarly, the survival of residual endogenous ␤-cells should increase following a successful islet graft,

365

366

CLINICAL EFFECTIVENESS OF ISLET TRANSPLANTATION

early after diagnosis. Obviously the graft should be performed under the cover of an immunosuppressive strategy capable of inhibiting rejection of transplanted islets and autoimmune destruction of residual and transplanted ␤-cells. The goal of a solitary islet graft, performed at the diagnosis of type 1 diabetes, that is capable of inducing lifelong disease remission appears to be less distant.

References 1 Brendel MD, Hering BJ, Schulz AO, Bretzel RG, International Islet Transplantation Registry. Department of Medicine, Justus-Liebig University of Giessen, Germany, Newsletter 8, 1999. 2 Wahoff DC, Papalosi BE, Najarian JS, et al. Autologous islet transplantation to prevent diabetes after pancreatic resection. Ann Surg 1995;222:562–79. 3 Carrol PB, Rilo HLR, Alejandro R, et al. Long-term (> 3-year) insulin independence in a patient with pancreatic islet cell transplantation following upper abdominal exenteration and liver replacement for fibrolamellar hepatocellular carcinoma. Transplantation 1995;59:875–9. 4 Najarian JS, Sutherland DE, Matas AJ, et al. Human islet autotransplantation following pancreatectomy. Transplan Proc 1979;11:336–40. 5 Cameron JL, Mehigan DG, Harrington DP, Zuidema GD. Metabolic studies following intrahepatic autotransplantation of pancreatic islet grafts. Surgery 1980;87:397–400. 6 Traverso LW, Abou-Zamzam AM, Longmire WP Jr. Human pancreatic cell autotransplantation following total pancreatectomy. Ann Surg 1981;193:191–5. 7 Hinshaw DB, Jolley WB, Hinshaw DB, et al. Islet autotransplantation after pancreatetomy for chronic pancreatitis with a new method of islet isolation. Am J Surg 1981;142:118–22. 8 Farney AC, Najarian JS, Nakhleh RE, et al. Autotransplantation of dispersed pancreatic islet tissue combined with total or near-total pancreatectomy for treatment of chronic pancreatitis. Surgery 1991;110:427–37. 9 Pyzdrowski KL, Kendall DM, Halter JB, et al. Preserved insulin secretion and insulin independence in recipients of islet autografts. N Engl J Med 1992;327:220–6. 10 Wahoff DC, Papalois BE, Najarian JS, et al. Islet autotransplantation after total pancreatectomy in a child. J Pediat Surg 1996;31:132–5. 11 Kendall DM, Sutherland DE, Najarian JS, et al. Effects of hemipancreatectomy on insulin secretion and glucose tolerance in healthy humans. N Engl J Med 1990;322:898-903. 12 Seaquist ER, Robertson RP. Effects of hemipancreatectomy on pancreatic alpha and beta cell function in healthy human donors. J Clin Invest 1992;89:1761–6. 13 Teuscher AU, Kendall DM, Yves FC, et al. Successful islet autotransplantation in humans. Functional insulin secretory reserve as an estimate of surviving islet cell mass. Diabetes 1998;47:324–30. 14 Kendall DM, Teuscher AU, Robertson RP. Defective glucagon secretion during sustained hypoglycemia following successful islet allo- and autotransplantation in humans. Diabetes 1997;46:23–7. 15 Gupta V, Rooney DP, Kendall DM, et al. Defective glucagon responses from transplanted intrahepatic pancreatic islets during hypoglycemia is transplantation-site determined. Diabetes 1997;46:28–32. 16 Sjoberg RJ, Kidd GS. Pancreatic diabetes mellitus. Diabetes Care 1989;12:715–24. 17 Hostens K, Ling Z, Van Schravendijk C, Pipeleers D. Prolonged exposure of human beta-cells to high glucose increases their release of proinsulin during acute stimulation with glucose or arginine. J Clin Endocrinol Metab 1999;84:1386–90. 18 Johnson PR, White SA, Robertson GS, et al. Pancreatic islet autotransplantation combined with total pancreatectomy for the treatment of chronic pancreatitis. The Leicester experience. J Mol Med 1999;77:130–2. 19 Davalli AM, Seaglia L, Zangen DH, et al. Vulnerability of islets in the immediate posttransplantation period. Dynamic changes in structure and function. Diabetes 1996;45:1161–7.

A. M. DAVALLI ET AL.

20 Tzakis AG, Ricordi C, Alejandro R, et al. Pancreatic islet transplantation after upper abdominal exenteration and liver replacement. Lancet 1990;336:402–5. 21 Ricordi C, Tzakis AG, Carroll PB, et al. Human islet isolation and allotransplantation in 22 consecutive cases. Transplantation 1992;53:407-14. 22 McCulloch DK, Klaff LJ, Kahn SE, et al. Nonprogression of sublinical b-cell dysfunction among firstdegree relatives of IDDM patients: 5-yr follow up of the Seattle family study. Diabetes 1990;39:549-56. 23 Fukushima M, Nakai Y, Taniguchi A, et al. Insulin sensitivity, insulin secretion, and glucose effectiveness in anorexia nervosa: a minimal model analysis. Metabolism 1993;42:1164-8. 24 Drachenberg CB, Klassen DK, Weir MR, et al. Islet cell damage associated with tacrolimus and cyclosporine: morphological features in allograft biopsies and clinical correlation. Transplantation 1999;68:396-402. 25 Scharp DW, Lacy PE, Santiago JV, et al. Insulin independence after islet transplantation into type 1 diabetic patients. Diabetes 1990;39:515-18. 26 Warnock, GL, Kneteman NM, Ryan E, et al. Normoglycemia after transplantation of freshly isolated and cryopreserved pancreatic islets in type 1 (insulin independent) diabetes mellitus. Diabetologia 1991;34:55-8. 27 Socci C, Falqui L, Davalli AM, et al. Fresh human islet transplantation to replace pancreatic endocrine function in type 1 diabetic patients. Acta Diabetol 1991;28:151-7. 28 Gores PF, Najarian JS, Stephanian E, et al. Insulin independence in type 1 diabetes after transplantation of unpurified islets from single donor with 15-deoxyspergualin. Lancet 1993;341:19-21. 29 Secchi A, Socci C, Maffi P, et al. Islet transplantation in IDDM patients. Diabetologia 1997;40:225-31. 30 London NJM, Robertson GSM, Chadwick DR, et al. Human pancreatic islet isolation and transplantation. Clin Transplant 1994;8:421-59. 31 Nomikos IN, Prowse SJ, Carotenuto P, et al. Combined treatment with nicotinamide and desferrioxamine prevents islet allograft destruction in NOD mice. Diabetes 1986;35:1302-4. 32 Cryer PE. Hypoglycemia: the limiting factor in the management of IDDM. Diabetes 1994;43:1378-89. 33 Stegall MD, Lafferty KJ, Kam I, et al. Evidence of recurrent autoimmunity in human allogeneic islet transplantation. Transplantation 1996;61:1272-4. 34 Jaeger C, Brendel M, Hering B, et al. Progressive islet graft failure occurs significantly earlier in autoantibody-positive than in autoantibody-negative IDDM recipients of intrahepatic islet allografts. Diabetes 1997;46:1907-10. 35 Braghi S, Bonifacio E, Secchi A, et al. Modulation of humoral islet autoimmunity by pancreas allotransplantation influences allograft outcome in patients with type 1 diabetes. Diabetes 2000;49:218-24. 36 Davalli AM, Ricordi C, Socci C, et al. Abnormal sensitivity to glucose of human islets cultured in a high glucose medium: partial reversibility after an additional culture in a normal glucose medium. J Clin Endocrinol Metab 1991;72:202-8. 37 Secchi A, Taglietti MV, Socci C, et al. Insulin secretory patterns and blood glucose homeostasis after islet allotransplantation in IDDM patients: comparison with segmental- or whole-pancreas transplanted patients through a long term longitudinal study. Journal of Mol Med 1999;77:133-13. 38 Atkinson PR, Behme MT, Zucker P, et al. Continued insulin dependence despite normal range insulin sensitivity and insulin connecting peptide levels in a kidney/islet transplant patient. Diabetes Care 1996;19:236-40. 39 Luzi L, Hering B, Socci C, et al. Metabolic effects of successful intraportal islet transplantation in insulin-dependent diabetes-mellitus. J Clin Invest 1994;97:2611-18. 40 Eisenbarth GS. Type 1 diabetes mellitus: a chronic autoimmune disease. N Engl J Med 1986;314:1360–8.

367

368

CLINICAL EFFECTIVENESS OF ISLET TRANSPLANTATION

41 Tibel A, Linden R, Larsson M, et al. Long-term glucose control after pancreatic transplantation. Transplant Proc 1990;22:645-6. 42 Pyke D. Pancreas transplantation. Diabet Metab Rev 1991;7:3–14. 43 Gray DWR, Cranston D, McShane P, et al. The effect of hyperglycemia on pancreatic islets transplanted into rats beneath the kidney capsule. Diabetologia 1989;32:663-7. 44 Korsgren O, Jansson L, Andersson A. Effects of hyperglycemia on function of isolated mouse pancreatic islets transplanted under the kidney capsule. Diabetes 1989;38:510-15. 45 Davalli AM, Ogawa Y, Ricordi C, et al. A selective decrease in the beta cell mass of human islets transplanted into diabetic nude mice. Transplantation 1995;59:817-20. 46 Keymeulen B, Ling Z, Gorus FK, et al. Implantation of standardized beta-cell grafts in a liver segment of IDDM patients: graft and recipient characteristics in two cases of insulin-independence under maintenance immunosuppression for prior kidney graft. Diabetologia 1998;41:452-9. 47 Roep BO, Stobbe I, Duinkerken G, et al. Auto- and alloimmune reactivity to human islet allografts transplanted into type 1 diabetic patients. Diabetes 1999;48:484-90. 48 Stobbe I, Duinkerken G, van Rood JJ, et al. Tolerance to kidney allograft transplanted into type 1 diabetic patents persists after in vivo challenge with pancreatic islet allografts that express repeated mismatches. Diabetologia 1999;42:1379-80. 49 Bertuzzi F, Nano R, Maffi P, et al. In vitro parameters correlate with graft function. Acta Diabetol 1999;36:207. 50 Balibrea JL, Arias-Diaz J, Garcia C, Vara E, Effect of pentoxifylline and somatostatin on tumor necrosis factor production by human pulmonary macrophages. Circulation Shock 1994;43:51-6. 51 Shi CL, Rooth P, Taljedal IB. The protective effect of verapamil on mouse islets transplanted under the kidney capsule. Transplantation 1993;56:1491-5. 52 Vajkoczy P, Lehr HA, Hubner C, et al. Prevention of pancreatic islet xenograft rejection by dietary vitamin E. Am J Pathol 1997;150:1487-95. 53 Mendola J, Wright JR Jr, Lacy PE. Oxygen free-scavengers and immune destruction of murine islets in allograft rejection and multiple low-dose streptozotocin-induced insulitis. Diabetes 1989;38:379-85. 54 Patanè G, Piro S, Rabuazzo MA, et al. Metformin restores insulin secretion altered by chronic exposure to free fatty acids or high glucose. A direct metformin effect on pancreatic b-cells. Diabetes 2000;49:735–40. 55 Maffi P, Bertuzzi F, Aldrighetti L, et al. Islet transplantation in humans: effect of adjuvant therapy on C-peptide secretion. Acta Diabetol 1999;36:219. 56 Shapiro AMJ, Lakey JR, Ryan EA, Korbutt GS, Toth E, Warnock GL, et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med 2000;343:230–8. 57 Ryan EA, Lakey JR, Rajotte RV, Korbutt GS, Kin T, Imes S, et al. Clinical outcomes and insulin secretion after islet transplantation with the Edmonton prococol. Diabetes 2001;50:710–9. 58 Mirouze J, Selam JL, Pham TC, Mendoza E. Remission of diabetes during conventional insulin therapy or therapy controlled by use of an artificial pancreas. Semin Hop 1979;55:354-9.

Index

abdominal cancer, islet allotransplantation, preventing surgical diabetes 357–8 abdominal infections 159–61 anastomotic leak 161–2 see also infectious complications acid–base disturbances 167–70 acute abdomen, pancreatic rejection 192 adeno-associated virus, as vector 293 adrenal tissue, transplantation 316–17 amylase hyperamylasaemia 84–5, 161 urine 194–5 anaesthesia 147 anastomotic leak, complication of surgery 161–2 animal models 27–44 immunosuppressants, experimental studies 250–5 ruminants, fetal and neonatal pancreatic tissue transplantation 328–9 see also dog; pig; rat models antibodies OKT3 induction therapy 132, 198–9 recipient immune response in islet rejection 340–3 see also monoclonal antibodies; polyclonal antibodies anticoagulation therapy 149–50 risk of haemorrhage 158 antigen-presenting cells, islet allotransplantation 310, 340, 345 antilymphocyte induction therapy 132 apoptosis, gene therapy potentially applicable to types I/II DM 295–6 arterial reconstruction, pancreaticoduodenal Tx 119–20 autoimmune diabetes graft rejection 214 IDDM 264 immunosuppressants 263–4 islet-directed autoimmunity 7 autoimmune rejection of beta-cells, cause of type-1 DM 332 autonomic neuropathy, secondary complication of PTx in DM 224

basiliximab 132 beta-cells apoptosis, tumour necrosis factor-alpha 295, 364 autoimmune rejection, cause of type-1 DM 332 closed loop system 23–4 differentiation 331–2 engineering 296 function mimicking 296–7 monitoring 329–30

long-term survival 364 pancreatic rejection 192 precursors, expansion, gene therapy 291 regeneration gene therapy 291, 296 promotion by nicotinamide 295, 329 regulation of insulin 18 transplanted immature, monitoring function 329–30 see also islet allotransplantation; islet autotransplantation bicarbonate metabolic acidosis 167–70 reduced in SPK recipients, aetiologies 168 biopsy, diagnosis of pancreatic rejection 195–6 bladder drainage 116–17, 121, 162–3 conversion to ED 162–3 dysuria syndrome 163 vs ED, economic cost 283–4 infections 173 metabolic acidosis 167–70 related complications 5, 162–3 risk factors for technical failures 156, 159 segmental pancreas Tx 122 systemic–bladder (S–B) vs PE drainage 133–4, 172 volume depletion 169, 170 bladder infections, post-transplant period 151–3 bladder leak, complication of SPKTx 152 bladder/duodenal leak, complication of SPKTx 172 bone disease, post-transplant 175–8 brain death criteria 82–3 brain tissue, transplantation 316–17 brequinar animal models 255 phase I study 260 buffers, graft preservation 100–1

C-peptide secretion 357–8 cadaveric donor 69, 82–3 organ inspection 79–80, 81–2 pancreatectomy 89–90 calcitonins 178 calmodulin, and cyclosporin A 249 Candida, intra-abdominal infections 159–61 cardiac evaluation 53–5 cardiovascular disease in DM 224–5, 247–8 post-transplant 178–80 risk, immunosuppressants 266–7 Carolina rinse (CR) solution 109 carotid intima thickness 224–5 Carrel patch, vs iliac Y graft 158

370

INDEX

cellular and islet transplantation historical aspects 305–23 see also islet allotransplantation; islet autotransplantation chimerism, haematopoietic, prevention of islet rejection 344–5 cholesterol, dyslipidaemia 180–1 closed loop system 23–4 coagulation deficiencies 74–5 cold ischaemia injury 37, 104 colloid additives, graft preservation 99–100 compliance 76 complications of PTx in diabetes mellitus see diabetes mellitus (DM) complications of SPKTx see medical complications; post-transplant period computed tomography (CT), pancreatic rejection 197 coronary artery disease algorithm 55 evaluation 53–5, 74 costimulatory molecules, blockade, rejection prevention 311–12, 341, 342 costimulatory pathways CD28-CD80/86 343 CD40-CD154 343 cotransplantation, and genetic modification, islet rejection prevention 347 creatinine, diagnosis of pancreatic rejection 193 cyclosporin 5, 250–1 and calmodulin 249 diabetogenic effect 256 drug interactions 250 experimental studies 250–1 PTDM 214, 256 and sirolimus 253 studies in vitro and in rodents 250–1 studies in larger animal models 251 high dose 344–5 cytology, pancreatic exocrine drainage 196 cytomegalovirus (CMV) infection 128, 171, 174–5, 215 anti PTLD strategy 183

daclizumab 132 deoxyspergualin animal models 252–3 and post-transplant diabetes 258 diabetes mellitus 15–26 acute complications of PTx see diabetes mellitus, secondary complications of PTx autoimmune rejection of beta-cells, cause of type-1 DM 332 bone fracture 176–7 classification 46 clinical course 17 genetics 45–6 hyperlipidaemia 248 insulin 17–23 management, economic cost 277–8

metabolic profiles and monitoring 19–20 pathogenesis impact of PTx on understanding 6–7 types and causes 15–17, 45–7 post-transplant defined 248 see also post-transplant diabetes mellitus pretransplant medical evaluation for candidates 45–58 prevention, gene therapy 294–5 risk of cardiovascular disease 247–8 risk factor modification 48–50 secondary complications of PTx 219–27 diabetic nephropathy 32, 222–4 diabetic neuropathy 224 diabetic retinopathy 220–2 hypoglycaemia 219–20 macrovascular disease 178–81, 224–5 see also autoimmune diabetes; post-transplant diabetes mellitus diabetic animals, fetal and neonatal pancreatic tissue transplantation 327–8 diabetic nephropathy diagnosis 50 incidence 229 prevention, models 32–3 secondary complication of PTx in DM 222–4 diabetic neuropathy, secondary complication of PTx in DM 224 diabetic retinopathy, secondary complication of PTx in DM 220–2 dialysis, economic cost 278 dobutamine echocardiography, cardiac evaluation 54 dog kidney graft model, FTY720 255 dog model of islet allotransplantation, history 312–13 dog model of PTx 33–9 anatomy 34–5 metabolic complications, long-term aspects 238–9 studies 37–9 cyclosporin 251 early rejection vs graft pancreatitis 38 mycophenolate mofetil 254 prevention of cold ischaemia injury 37–8 prevention of diabetic complications 38–9 tacrolimus 252 technique 35–7 donor selection for PTx 79–80, 83–7 allograft assessment 81–2 cadaveric donors 69, 82–3 contraindications 83 loss 82 multiorgan procurement 89–90 donor mismanagement, complications 85 factors, analysed 80 graft procurement and preparation 89 liberalization 79 marginal donors 80 organ inspection 79–80 pancreas alone (PTA) 68–70 recipient factors 81 related pancreas donor 85–6

INDEX

resuscitation 68 SPK, living related donor 51, 63–4, 85–6 drainage in PTx see bladder drainage; enteric drainage; portal venous drainage; portal enteric drainage duct injection 122 duodenal jejunostomy 2–3 duodenal/pancreatic transplantation 117–21, 156 see also pancreaticoduodenal dyes, islet viability assessment techniques 308–9 dyslipidaemia 180–1, 247 dysuria syndrome 163

economic cost of PTx 15, 277–89 bladder vs enteric drainage 283–4 cost-effectiveness analyses 284–6 diabetes management 277–8 dialysis 278 future prospects 286–7 kidney transplantation 278–9 pancreas alone transplantation (PTA) 283 simultaneous kidney and pancreas transplantation (SPK) 279–82 surgical complications effects 282–3 elastase 194 electrolytes graft preservation 98–9 monitoring 152 energy substrates, graft preservation 100 enteric drainage 116, 121 vs BD 121 economic cost 283–4 conversion from BD 162–3 infections 173–4 PAK transplantation 86 risk factors for technical failures 156, 159 epinephrine, action 219–22 Epstein–Barr virus (EBV) infection anti PTLD strategy 183 PTLD 215–6 Euro Collins (EC) solution 37, 105 experimental studies immunosuppressants 250–5 newer and emerging agents 252–5 rejection prevention 339–53 see also animal models; specific substances

FasL, Sertoli cells 347 fetal brain tissue, transplantation 316–17 fetal and neonatal pancreatic tissue transplantation 325–37 differentiation of beta-cells 329, 331–2 efficacy in diabetic animals 327–8 history human 307–8

murine models 307, 310–11 human allografts 325–6 monitoring function of transplanted immature fetal beta-cells 329–30 rejection 332 time required to normalize blood glucose levels 329 tissue from ruminants 328–9 types transplanted 326–7 xenografts, into humans 326 fine needle aspiration biopsy, diagnosis of pancreatic rejection 196 FK-binding protein (FKBP) 249 fluids and electrolytes 167–70 metabolic acidosis 167–70 volume depletion 169, 170 FTY720, animal models 255, 264 fungal (intra-abdominal) infections 159–61

gastrointestinal complications, post-transplant period 152 gene therapy 291–304 altering peripheral insulin resistance in type II DM 291–2 author’s laboratory 297–9 current status 294–7 expansion of beta-cells/precursors 291 glucose-responsive insulin secretion 291 historical aspects 291–2 immune modulation 292 potentially applicable to DM types I and II 295–7 beta-cell engineering 296 beta-cell regeneration and replacement 296 inhibition of apoptosis 295–6 non-beta-cell engineering 296–7 prevention of types I/II DM 295 vectors used 292–4 genetic modification, and cotransplantation, islet rejection prevention 347 glucagon action 219–21 secretion by autografted islets 362 preventing surgical diabetes 356–7 glucagon-like peptide (GLP-1) 17 glucose control, DCCT results 230, 247 glucose potentiation of arginine-induced insulin secretion (GPAIS) 356 glucose-dependent insulinotrophic peptide (GIP) 17 glucose-responsive insulin secretion, gene therapy 291 glutamic acid decarboxylase (GAD) 15 anti-GAD65 358, 363 graft-associated malignancies 181–4 post-transplant lymphoproliferative disorder 182–3 skin cancers 183 graft non-function, causes 95–6 graft pancreatitis complication of surgery 161 vs pancreatic rejection, models 31, 38

371

372

INDEX

graft preservation 95–113 goals 95–6 management of donor 83–5 mismanagement 85 principles 98–103 buffers 100–1 colloids 99–100 electrolytes 98–9 energy substrates 100 impermeants 99 oxygen radical scavengers 100–1 techniques 96–8, 102 checklist 104 hypothermia 97–8 graft preserving solutions 105–9 Euro Collins (EC) solution 37, 105 flushout solution (Carolina rinse (CR) solution) 109 histidine-tryptophan-ketoglutarate (HTK) solution 31, 99, 107–8 two layer storage technique 108–9 University of Wisconsin (UW) solution 105–7 graft procurement and preparation 89–94 complications 103–5 cold preservation injury 104 harvesting injury 104 pre-preservation injury 103–4 reperfusion injury 105 rewarming injury 105 donor selection 89 graft assessment 81–2 multiorgan procurement of pancreas, liver and kidneys 89–90 organ allocation system 55–6 pancreas without liver procurement 91 preparation of graft for PTx 91–4 vascular variations, combined pancreas/liver 90–1 graft rejection see pancreatic rejection graft survival rates 67, 155, 191, 229–30, 230 see also outcomes graft-versus-host disease 342 growth hormone, and insulin, dawn phenomenon 20

haematopoietic chimerism, prevention of islet rejection 344–5 haematuria 171–2 haemorrhage anticoagulation therapy 158 complication of post-transplant period 150 histidine-tryptophan-ketoglutarate (HTK) solution 31, 99, 107–8 histology of cellular/islet transplantation 360 histology of PTx 205–17 acute rejection 206–10 chronic rejection 210–14 cytomegalovirus infection 171, 215 drug toxicity 214–15 PTLD in pancreas allografts 215–16 recurrent diabetes 214

historical background of PTx 1–13 antirejection treatment 5 diabetes and complications 6–7 portal–enteric (PE) drainage 126–8 recipient selection and programme development 7 surgical techniques 2–5, 115–16 history of cellular/islet transplantation 305–23 fetal murine allografts 307, 310–11 fetal and neonatal human allografts 307, 325–6 human islet isolation, clinical application 313–14 islet allotransplantation anatomy and physiology 309–10 early clinical attempts 312 early development 305–6 exocrine contamination and hyperglycaemia 309 fetal murine models 310–11 immunologically privileged implantation sites 310–11 insulin independence 314–15 large animal models 312–13 migratory antigen-presenting cells 310 Milan, first 359–60 prevention of contamination with lymph nodes 309 studies in isogeneic rodent model 306–7 islet autotransplantation, following total pancreatectomy 315 islet identification techniques 308 islet isolation, outcome quantifying 309 islet viability assessment techniques 308–9 islet xenotransplantation 315–16, 326, 332 neonatal PTx 308 notes on references 317 rejection prevention alternative approaches 311–12 costimulatory molecule blockade 311–12, 341, 342 encapsulation 311 thymic tolerance induction 311 transplantation of other non-haemopoietic tissues and cells 316–17 history of gene therapy for diabetes 291–2 HIV, as vector 293 HLA effect of type 250 genetics of diabetes mellitus 45–6 hormone replacement therapy 178 human adenovirus, as vector 293 hyperamylasaemia donor 84–5 simulating graft pancreatitis 161 hypercoagulation work-up 75 hyperglycaemia after PTx 248–9 pancreatic rejection 192 hyperinsulinaemia 170–1, 180 hyperlipidaemia 180–1, 248 hypertension 180 hypoamylasaemia, graft rejection 194–5 hypoglycaemia, secondary complication of PTx in DM 219–20 hypothermia, graft preservation 97–8

INDEX

iliac Y graft 119–20, 130, 158 immune modulation, gene therapy 292 immunologically privileged sites, islet allotransplantation 310–11 immunosuppressants ALI therapy 131 autoimmune diabetes, paradoxical effects 263–4 and cardiovascular risk 266–7 clinical studies 255–61 diabetogenesis 266 diabetogenicity 247–75 IPITA recommendations (1998) 265–6 experimental studies 250–5 and neoplasms 182 newer and emerging agents 132, 252–5, 257–60 OKT3 induction therapy 132, 198–9 post-transplant lymphoproliferative disorder (PTLD) 182–3 risk of cardiovascular disease 247–8 steroids 255–6 T-cell depletion 198–9 see also specific substances impermeants, graft preservation 99 ‘incretins’ 17 infectious complications 173–4 aetiology 173–4 anastomotic leak 161–2 cytomegalovirus infection 128, 171, 174–5, 215 incidence and types 173 intra-abdominal infections 159–61 mycotic pseudoaneurysms 159 opportunistic 173–4 post-transplant 150–2 prophylaxis 174 insulin action, and consequences of deficiency 17–19 autografted islets, preventing surgical diabetes 356–7 closed loop systems 23–4 control and supertight control 22–3 dawn phenomenon 20 hyperinsulinaemia 170–1, 180 regimens 20–2 insulin independence, islet allotransplantation 314–15 insulin resistance 171 Somoygi phenomenon 20 steroid-induced 249, 255–6 type II DM, gene therapy 291–2 interleukin-2 (IL-2) receptor blockade, PTDM 259–60 in rejection 194 International Pancreatic and Islet Transplantation Association, recommendations (1998), specific immunosuppression 265–6 International Pancreatic and Islet Transplantation Registry 2 data on islet transplantation 355 data on types and rates 59, 67, 115, 125, 231 intra-abdominal infections 159–61 anastomotic leak 161–2 see also infectious complications ischaemia injury 37, 100–1

islet allotransplantation altering capacity to stimulate rejection 346 curing type 1 DM 358–65 adjuvant therapy to improve islet engraftment and survival 364 cellular composition of islet grafts 363 clinical outcomes 365 clinical trials 358–9 metabolic studies in successfully transplanted patients 360–3 Milan experience 359–60 novel predictors of islet integrity 363–4 solitary islet transplantation 364–5 fetal pancreas in murine models 310–11 islets after kidney (IAK) 358 islets with kidney (SIK) 358 preventing surgical diabetes 357–8 abdominal cancer patients 357–8 with liver 357 islet autotransplantation preventing surgical diabetes 355–7 chronic pancreatitis 355–6 future endeavours 357 insulin and glucagon secretion by autografted islets 356–7 islet cell transplantation see segmental pancreas transplantation islet-directed autoimmunity 7 islet isolation, human clinical applications 313–14 islet-like cell clusters (ICC) 327 islet neogenesis-associated protein (INGAP) 296 islet rejection, experimental approaches to prevention 339–53 immunomodulation 345–7 masking, genetic modification and cotransplantation approaches 347 primary non-function 347 recipient immune response manipulation 340–5 antibodies or recombinant molecules 340–3 haematopoietic chimerism 344–5 intrathymic transplantation 343–4 islet xenotransplantation 315–16, 326, 332

kidney and pancreas transplantation complications 167–89 indications 59–66 KTA see kidney transplantation alone living related pancreas transplantation 51, 63–4, 85–6 multiorgan procurement of pancreas, liver and kidneys 89–90 PAK see pancreas after kidney (PAK) transplantation patient selection 59–66 criteria 59–61 evolution 7 history 7 pancreas alone (PTA) 51–2, 70–6 recipient groups 61–2

373

374

INDEX

kidney and pancreas transplantation (contd.) SPK see simultaneous kidney and pancreas transplantation see also medical complications of SPKTx; kidney transplantation alone (KTA) economic cost 278–9 graft survival rates 229–30 model, rejection 39–40 vs SPK, long-term aspects 235–8

leflunomide animal models 254–5 lack of diabetogenicity 260 lentivirus, as vector 293 leucine, steroid-induced insulin resistance 249 lipids, dyslipidaemia 180–1 liver graft following PTx, immunoprtection 32 procurement multiorgan (pancreas, liver and kidneys) 89–90 vascular variations, combined pancreas/liver 90–1 living related donor for PTx 51, 63–4, 85–6 long-term aspects post PTx 229–39, 243–5 metabolic aspects 238–9 outcomes single centre reports 231–5 SPK vs kidney Tx alone (KTA) 235–8 lymphocyte immune globulins 198 antilymphocyte induction therapy 132 lymphoproliferative disease, pancreas allografts, posttransplantation 215–16

macrovascular disease, secondary complication of PTx in DM 157–9, 178–81, 224–5 magnetic resonance imaging 197 malignancies skin cancers 183 transplant-associated 181–4 masking, islet rejection prevention 347 medical complications of SPKTx 167–89 fluids, electrolytes and acid-base disturbances 167–70 infectious complications 173–4 macrovascular disease 178–81, 224–5 metabolic complications 170–1 post-transplant bone disease 175–8 post-transplant period, complications 149–53 transplant-associated malignancies 181–4 urological complications 171–2 metabolic acidosis 167–70 BD conversion to ED 162–3, 168 metabolic complications of SPKTx 152, 170–1 hyperinsulinaemia 170–1, 180 insulin resistance 171 long-term aspects 238–9 post-transplant period 152

migratory antigen-presenting cells, islet allotransplantation 310 Milan, first islet allotransplantation 359–60 minerals, post-transplant bone disease 177 MODY 1–3 diabetes mellitus, classification 46 monoclonal antibodies 341 anti-B-cell 182–3 basiliximab, daclizumab 132 masking techniques 347 OKT3 induction therapy 132, 198–9 rituximab 183 T-cell-depleting MAbs 198–9 mouse models allotransplantation of islets and fetal pancreas 310–11 NOD 291, 342, 345 rAAV vector system 299 MRI, pancreatic rejection 197 mycophenolate mofetil 132 animal models 254 economic cost 279 patient and graft survival rates 234 reduction of post-transplant diabetes 258 mycotic pseudoaneurysms 159

neonatal PTx see fetal and neonatal pancreatic tissue transplantation newer and emerging agents, experimental studies 252–5 nicotinamide, promotion of beta-cell regeneration 295, 329 non-haemopoietic tissues and cells, history of transplantation 316–17

obesity 181 octreotide, inhibition of exocrine secretion 161 oestrogen replacement therapy 178 OKT3 induction therapy 132, 198–9 organ allocation system 55–6 organ inspection 79–80, 81–2 organ preservation see graft preservation osteonecrosis, post-transplant 176 osteoporosis 175–6 outcomes of islet allotransplantation 365 outcomes of PTx single centre reports 231–5 SPK vs kidney transplantation alone (KTA) 235–8 see also graft survival rates oxygen persufflation 100 oxygen radical scavengers, graft preservation 100–1

pancreas after kidney (PAK) transplantation 3, 63, 230 criteria for patient selection 61, 68 economic cost 282 enteric drainage 86

INDEX

evaluation 50–1 graft survival rates 191, 230 incidence 67, 86 indications 49 risk factors for technical failures 156, 159 survival rates 155 pancreas alone (PTA) transplantation 62–3, 67–78 criteria for patient selection 61 donor selection 68–70 economic cost 283 graft survival rates 67, 155, 191, 230 indications 51–2, 62–3, 67–78 recipient selection 51–2, 61, 70–6 risk factors for technical failures 156, 159 see also graft procurement pancreas-specific protein (PSP) 194 pancreatic polypeptide, monitoring function of beta-cells 329–30 pancreatic preservation see graft preservation pancreatic rejection 153, 191–203 acute rejection 198, 206–10 chronic rejection 210–14 clinical diagnosis 192 vs graft pancreatitis, model 38 histology acute rejection 206–10 chronic rejection 210–14 history of antirejection treatment 5 imaging techniques 196–7 scintigraphy 197 ultrasound, CT and MRI 197 immunosuppression 198–9 polyclonal T-cell-depleting Abs 198 T-cell-depleting MAbs 198–9 incidence 139, 191 laboratory diagnosis 192–5 other serum markers 194 serum anodal trypsinogen (SAT) 193–4 serum creatinine 193 urine markers 194–5 pathophysiology 192 reversibility, models 31–2 tissue and cell diagnosis 195–6 cytology 196 fine needle aspiration biopsy 196 needle core biopsy 195–6 pancreatic secretory trypsin inhibitor (PSTI) 194 pancreatic transplantation contraindications 52 indications 49 living related donor 51, 63–4, 85–6 perioperative management in PTx PTA see pancreas alone (PTA) transplantation with renal transplantation see kidney and pancreas transplantation risk factors 52–3 SPK see simultaneous kidney and pancreas transplantation see also islet; segmental pancreaticoduodenal transplantation 117–21, 156 arterial reconstruction 119–20

duct management 121 venous reconstruction 117–19 pancreatitis chronic, islet transplantation 355–6 mismanagement of donor 85 reflux 172 see also graft pancreatitis parathyroidectomy 177 patient selection 70–6 pancreas after kidney (PAK) transplantation 61 pancreas alone (PTA) 51–2, 61, 70–6 preoperative evaluation and management 145–6 simultaneous kidney and pancreas transplantation (SPK) 61 simultaneous transplantation (SPK) 62 perfluorochemical (PFC), two layer storage technique 108–9 perioperative management in PTx 145–53 anaesthesia/general procedure 147 preoperative evaluation and management 145–6 postoperative management 148–9 see also post transplant period, complications and surveillance peripheral vascular disease, post-transplant 178–80, 224 pig model of PTx 39–42 anatomy 39 fetal and neonatal xenografts, into humans 326, 332 leflunomide 254–5 rejection correlation following pancreaticoduodenal Tx 41 grading differences after simultaneous combined kidney/PTx 41–2 single vs combined kidney/PTx 39–40 technique 39 polyclonal ABs 341 polyclonal T-cell-depleting ABs, pancreatic rejection 198 portal venous drainage 4–5, 117, 119 portal–enteric (PE) drainage 125–43 historical background 126–8 immunosuppression 131–2 Memphis programme overview 128–30 results 133–40 demographic characteristics 136 systemic–bladder (SB) vs PE drainage 133–4, 172, 234 systemic–enteric (SE) vs PE drainage 134–6, 172 statistical analysis 132–3 post-transplant bone disease 175–8 bone fracture 176–7 management 177–8 antibone resorptive agents 177–8 hormone replacement 178 vitamins and minerals 177 osteonecrosis 176 osteoporosis 175–6 post-transplant diabetes mellitus (PTDM) 248–60 acute complications of PTx see diabetes mellitus, secondary complications of PTx correction 262–3 definition 248 hyperglycaemia 249

375

376

INDEX

post-transplant diabetes mellitus (PTDM) (contd.) International Pancreatic and Islet Transplantation Association, recommendations (1998) 265–6 natural history 263 proposed mechanisms 249–50 drug interactions 250 effect of HLA type 250 solid organ transplants clinical studies 255–60 cyclosporin 256–7 new and emerging agents 257–60 pre-cyclosporin era 255–6 tacrolimus 256–7 steroid-induced 249, 255–6 effect of steroid reduction 261–2 symptomatology 248 post-transplant lymphoproliferative disorder (PTLD) 182–3 histology, pancreas allografts 215–16 post-transplant macrovascular disease 178–81, 224–5 diabetes mellitus 181 dyslipidaemia 180–1 hyperinsulinaemia 170–1, 180 hypertension 180 obesity 181 post-transplant period algorithm 151 complications and surveillance 149–53 gastrointestinal 152 haemorrhage 150 infection 150–2 metabolic 152 rejection 153 thrombosis 140, 149–50, 158 urinary tract 152–3 ECG monitoring 149 management, very early 148–9 postoperative see post-transplant period preoperative evaluation and management 145–6 preproinsulin II gene (rI[U]2[u]) 297–9 preserving solutions 105–9 pretransplant medical evaluation for candidates 45–58 cardiac evaluation 53–5 organ allocation system 55–6 pancreas after kidney (PAK) 50–1 pancreas alone (PTA) 51–2, 61, 70–6 risk factors 52–3 simultaneous kidney and pancreas Tx (SPK) 47–50 procurement and preparation of graft see graft procurement proinsulin: insulin ratio 357 proinsulin biosynthesis 296–7 pseudoaneurysms 159

quality of life after PTx 239–45 quinoline carboxylic acid derivatives, brequinar 255, 260

rAAV vector system, rat preproinsulin II gene (rI[U]2[u]) 297–9 rapamycin see sirolimus rat model of islet transplantation, isogeneic model 306–7 rat model of PTx 27–33 anatomy 27–30 studies 31–3 cyclosporin 250–1 graft immunoprotection 32 leflunomide 254–5 prevention of diabetic nephropathy 32–3 prevention of post-transplant graft pancreatitis 31 reversal of vascular complications of diabetes 33 reversibility of rejection 31–2 tacrolimus 249, 251–2 technique 30–1 rat preproinsulin II gene (rI[U]2[u]) 297–9 receptor–ligand interactions, blocking 341 recipient selection see patient selection recombinant molecules, recipient immune response in islet rejection 340–3 references, notes on references 317 reflux pancreatitis 172 rejection of islets see islet rejection rejection of whole graft see pancreatic rejection renal failure dialysis, economic cost of PTx 278 SPK transplants 167 reperfusion injury, graft preservation 105 retroviruses, as vectors 293 rewarming injury, graft preservation 105 Ringer lactate 97 risk factors cardiovascular disease in DM 247–8 modification in DM 48–50 pancreatic transplantation (PTx), relative/absolute contraindiations 73–6 PTx 52–3, 70 surgical failures 156, 159 rituximab 183 rodent models see mouse; rat ruminants, fetal and neonatal pancreatic tissue transplantation 328–9

scintigraphy, pancreatic rejection 197 SDZ-RAD animal models 254 and post-transplant diabetes 259 segmental pancreas transplantation 121–2 living related donor 51, 63–4, 85–6 Sertoli cells, FasL 347 serum anodal trypsinogen (SAT) 193–4 serum markers, pancreatic rejection 192–3 simultaneous kidney and pancreas transplantation (SPK) 45–50 bladder vs enteric drainage, economic cost 283–4 cardiovascular disease 179–80

INDEX

complications 167–89 criteria for patient selection 61 drainage 118–19 switch to enteric drainage 162 economic cost 279–81 cost-effectiveness analyses 284–6 graft survival rates 155, 191, 230 indications 49, 62 vs KTA, long-term aspects 235–8 medical complications 167–89 outcomes of PTx, SPK vs KTA 235–8 patient and graft survival rates 234 peripheral vascular disease, vs KA 178–80 pig model 39, 41–2 preoperative evaluation and management 47–50, 145–6 rejection 139 risk factors for technical failures 156, 159 sirolimus (rapamycin) animal models 253–4 reduction of post-transplant diabetes 258–9 SDZ-RAD 254, 259 skin cancers 183 solitary PTA see pancreas alone (PTA) sphingosine, FTY720 255, 264 SPKTx see medical complications of SPKTx; simultaneous kidney and pancreas transplantation (SPK) Staphylococcus, intra-abdominal infections 159–61, 173 statistical background, portal–enteric (PE) drainage 132–3 steroids immunosupressants 255–6 induction of insulin resistance 249 reduction, effect on PTDM 261–2, 266 surgical techniques of PTx 115–24 current techniques 116–22 complications 155–65 anastomotic leak 161–2 bladder drainage 162–3 graft pancreatitis 161 haemorrhage 158–9 intra-abdominal infections 159–62 mycotic pseudoaneurysms 159 thrombosis 140, 149–50, 158 historical background 2–5, 115–16 risk factors for technical failures 156, 159 whole organ pancreaticoduodenal Tx 4, 117–21 venous reconstruction 117–19 systemic venous drainage 117–19 systemic–bladder (SB) drainage, vs PE drainage 133–4, 172 systemic–enteric (SE) drainage vs PE drainage 134–6 results 137 demographic characteristics 136

T-cell depletion, immunosuppressants 198–9 T-cells, recipient immune response, islet rejection 340–5

tacrolimus 5, 132, 251–2 costs 279 and cyclosporin 256–7 vs cyclosporin, post-transplant diabetes 214, 260–1 diabetogenic effect 256–7 experimental studies 251–2, 264–5 in vitro and in rodents 249, 251–2 in larger animal models 252 paradoxical effects 264–5 post-transplant diabetes mellitus 256–7 thymic tolerance induction, rejection prevention 311 thymic transplantation, prevention of islet rejection 343–4 transplant-associated malignancies see graft-associated malignancies tumour necrosis factor-alpha, beta-cell apoptosis 295, 364 two layer storage technique, preserving solutions 108–9

ultrasound, in pancreatic rejection 197 United Network for Organ Sharing (UNOS) Registry 230 report 238 University of Tennessee (UT) Memphis PTx programme 128–43 quality of life aspects 241–2 University of Wisconsin (UW) solution 105–7, 108–9 UNOS, economic cost of PTx 286 urethritis/urethral strictures 172 urinary drainage see bladder drainage urinary tract infections, post-transplant period 151–3, 162–3 urine, amylase 194–5 urine markers, pancreatic rejection 194–5 urological complications of SPKTx 171–2 bladder/duodenal leak 152, 172 dysuria syndrome 163 haematuria 171–2 reflux pancreatitis 172 urethritis/urethral strictures 172 urinary tract infections 151–3, 162–3 US Renal Data System economic cost of PTx 286 long-term outcomes 238 US Scientific Renal Transplant Registry, long-term outcomes 238

van’t Hoff coefficient, hypothermia in graft preservation 97–8 vascular complications of PTx see cardiovascular; macrovascular vascular variations, combined pancreas/liver graft 90–1 vectors, gene therapy 292–4 venous reconstruction, pancreaticoduodenal Tx 117–19 vesicular stomatitis virus, as vector 294

377

378

INDEX

viral vectors, gene therapy 292–4 vitamins, post-transplant bone disease 177 volume depletion 169, 170

whole organ pancreaticoduodenal transplantation see surgical techniques of PTx

wound infections 150–2

xenotransplantation fetal and neonatal pancreatic tissue transplantation 326–7 hyperacute rejection 332 islets 315–16