OPTIMISING THE QUALITY OF DONOR ORGANS FOR TRANSPLANTATION:

STUDIES OF HORMONE RESUSCITATION OF THE BRAIN-DEAD MULTI-ORGAN DONOR AND THE DEVELOPMENT OF A LONG-TERM PRESERVATION STRATEGY TO OPTIMISE FUNCTION OF THE TRANSPLANTED IN A PORCINE MODEL

By Alfred J. Hing, BSc(med)(hons), MB BS

Supervisor: Professor Peter S. Macdonald

A thesis submitted to the University of New South Wales in fulfilment of the requirements for the degree of Doctor of Philosophy

Transplant Program The Victor Chang Cardiac Research Institute, Darlinghurst, NSW, 2010

The Department of Heart and St. Vincent’s Hospital, Darlinghurst, NSW 2010

June, 2009 ORIGINALITY STATEMENT ‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no material previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis.

I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project’s design and conception or in style, presentation and linguistic expression is acknowledged.’

Alfred J. Hing BSc(med)(hons), MB BS June, 2009

ii COPYRIGHT STATEMENT

‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International.

I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.’

Alfred Hing April, 2010

AUTHENTICITY STATEMENT

‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’

Alfred Hing April, 2010

iii

“It was an unforgettable sight. We had taken the old heart and we needed to move damn fast to fill that huge hole with a new heart. No wonder they were frightened of us then. No wonder they thought we were out of our minds.”

Adrian Kantrowitz, Detroit Michigan, October 2003

“Brilliant as the technique of the surgeons and the skill of the anaesthetists may already be, advanced as the knowledge of the immunologists may be, we feel sure they would agree with the suggestion that the current work is, relatively speaking, crude fumbling. Everybody is still learning. Everybody’s experience is still worth communicating”

Anonymous (1969). The Medical Journal of Australia, 1(13):691

“Nothing can take the place of a well supported and well planned research effort for solving the most difficult of medical problems.”

Daniel diBardino (1999). The History and Development of Cardiac Transplantation. Texas Heart Institute Journal, 26(3): 198-205

“…. how can you mend a broken heart ….”

Barry Gibb, Robin Gibb (c. 1970)

iv

Dedication

To my parents, Winsome (Wan Sat) Hing (1936- 1995) and Robert Joe Hing (1919-2006)

This thesis and the work that it embodies is dedicated to my mother, Winsome Hing, and to my father, Robert Hing, who was ill throughout the period of my PhD research and who passed away during the preparation of this thesis.

It is to both my parents that I owe so much, including my interest in science and thirst for knowledge, reasoning and wisdom. It is also due to them that I have become the person that I am today. I am where I am today because of their support, love and encouragement, and because of their willingness to allow me to discover things for myself, to choose what I wanted to do in life and to fully support me in those endeavours.

v ACKNOWLEDGEMENTS

There are many people to whom I am very grateful for their assistance throughout the course of this project. First and foremost, my sincere thanks and appreciation goes to my supervisor, Professor Peter Macdonald, for giving me the opportunity to undertake this project in his laboratory. His advice, support, guidance, and encouragement throughout this project were invaluable and very much appreciated. He has helped me to develop the skills and gain the experience on my journey to becoming a clinician scientist.

The experiments carried out in this thesis were dependant on many people, without whom the research that underlies this thesis would not have been possible, and to whom I am very grateful. Staff from The Victor Chang Cardiac Research Institute, St. Vincent’s Hospital and the Collaborative Transplant Research Group at Sydney University who were involved at various times in the project included: Dr Mark Hicks, Dr Ling Gao, Ms Sarah Garlick, Mr Steve Faddy, Dr Aisling McMahon, Mr Scott Kesteven, Dr Graham Stewart, Mr Peter Tran, Dr Jerome Laurence, Mr Jair Kwan, Mr Andrew Dinale and Mr Jonathan Cropper. My thanks also go to Dr Michael Wilson for teaching me the skills of porcine , and for his assistance and advice. Later in the project, Dr Paul Jansz provided assistance and advice in the transplant study, and I am also grateful to him for his support and for arranging the leave from my job as the senior cardiothoracic surgical registrar at St. Vincent’s Hospital for me to complete this thesis. I am grateful to Dr Alasdair Watson for performing the porcine transplants in the GTN group of animals for the study described in Chapter 5. My predecessor in the laboratory, Dr Jon Ryan was a great source of advice and help, particularly in the early days of the project and I am especially grateful for the time he spent in helping me to establish myself in the project, and for teaching me about the intricacies of the Preload Recruitable Stroke Work relationship (PRSW) and multiple linear regression analysis. Professor Michael Feneley was also a great source of assistance in reviewing and discussing ventricular contractility data, and for helping me to develop a better understanding of the PRSW. Mr Scott Kesteven, whom I shared an office with throughout my research, was a great source of technical expertise, advice

vi and discussion, and I am grateful for his assistance during my candidature. Dr Mark Hicks was also a great source of technical expertise, assistance and discussion.

The studies reported in this thesis were funded by the National Health and Medical Research Council (NHMRC), and scholarship funding was provided by the NHMRC, the Royal Australasian College of Surgeons and the Cardiac Society of Australia and New Zealand. I would also like to acknowledge the institutional support provided by The Victor Chang Cardiac Research Institute and St. Vincent’s Hospital. Valuable assistance was provided by the Biological Resources Centre at the University of NSW in supplying animals for this study and by St. Vincent’s Pathology (SydPath), St. Vincent’s Hospital for the blood and urine assays in Chapter 4. I would also like to acknowledge the Department of Clinical Perfusion, the Department of Anaesthetics and the operating theatres at St. Vincent’s Hospital for providing some of the equipment for the experiments in this thesis.

I would also like to take this opportunity to acknowledge and thank my professional mentors. In particular, I would like to thank the late Dr Mark Shanahan for his friendship, advice and encouragement for over 20 years. He, along with the late Dr Victor Chang, supported me and ignited my interest in cardiothoracic surgery and heart transplantation. I would also like to acknowledge the support of Dr Phil Spratt for the St. Vincent’s cardiothoracic surgeon-scientist program and for his support of my training in cardiothoracic surgery. I would also like to thank Dr Alan Farnsworth for his patronage and support of my training in cardiothoracic surgery.

My personal thanks and gratitude go to Dr Aisling McMahon for her support, friendship, advice, assistance and encouragement, both professionally and personally. She has also provided invaluable assistance in the preparation and proofreading of this thesis, and in my development as a scientist.

Finally to my parents, the late Winsome and Robert Hing, I wish to express my most heartfelt thanks and gratitude for their many years of self-sacrifice, support, caring, encouragement and understanding.

vii TABLE OF CONTENTS

Originality Statement ...... ii Copyright Statement and Authenticity Statement...... iii Quotes...... iv Dedication ...... v Acknowledgements...... vi Table Of Contents ...... viii Presentations...... xvi Publications...... xx Prizes ...... xxii Research Scholarships ...... xxiv Travel Scholarships...... xxiv List Of Figures...... xxv List Of Tables ...... xxxvi List Of Abbreviations ...... xxxviii Abstract...... xlii

CHAPTER 1 AN INTRODUCTION AND REVIEW OF THE LITERATURE IN CARDIAC TRANSPLANTATION AND THE MANAGEMENT OF THE BRAIN-DEAD ORGAN DONOR...... 1 1.1 THE CHALLENGES IN CARDIAC TRANSPLANTATION...... 2 1.2 A BRIEF HISTORY OF CARDIAC TRANSPLANTATION ...... 4 1.2.1 Experimental Cardiac Transplantation...... 4 1.2.2 Human Cardiac Transplantation...... 6 1.3 THE MODERN ERA OF CARDIAC TRANSPLANTATION IN AUSTRALASIA AND OVERSEAS...... 8 1.3.1 Indications For Cardiac Transplantation ...... 8 1.3.2 Outcomes Following Cardiac Transplantation...... 11 1.3.3 Determinants Of Outcome In Cardiac Transplantation...... 14 1.3.4 Demographics Of Cardiac Transplantation And ...... 17 1.3.5 The Problem Of Supply Versus Demand ...... 20

viii 1.4 THE BRAIN-DEAD ORGAN DONOR...... 21 1.4.1 Defining Brain Death In The Organ Donor And Its Diagnosis...... 21 1.4.2 The Causes Of Brain Death...... 23 1.4.3 The Negative Effects Of Brain Death On Transplantation Outcomes ...... 23 1.4.4 The Cardiovascular, Neurohormonal And Immunological Consequences Of Brain Death ...... 24 1.4.4.1 The Autonomic Storm And Cardiovascular Changes In Brain Death...24 1.4.4.2 Neurohormonal Changes In Brain Death ...... 27 1.4.4.3 Immunological And Inflammatory Changes In Brain Death ...... 31 1.4.4.4 Impaired Oxidative Metabolism In Brain Death ...... 32 1.4.4.5 Effects Of Brain Death On Cardiac Contractile Function...... 33 1.4.4.6 Effects Of Brain Death On The Kidneys, Liver, Pancreas And Lungs.34 1.4.5 Management Of The Brain-Dead Organ Donor In The Intensive Care Unit...... 38 1.4.5.1 The Principles Of Donor Management And The Need For Aggressive Management Of The Organ Donor ...... 38 1.4.5.2 The Use Of Inotropes/Vasopressors To Support The Haemodynamically Unstable Donor, And Its Impact On Transplantation Outcomes...... 43 1.4.5.3 Hormone Resuscitation Of The Brain-Dead Donor ...... 47 1.4.5.4 Anti-Inflammatory Treatment Of The Brain-Dead Donor...... 56 1.5 CARDIAC ALLOGRAFT PRESERVATION ...... 57 1.5.1 Excision Of The Donor Heart And Preservation Techniques ...... 57 1.5.2 Injury To The Heart During Storage And Transplantation ...... 59 1.5.2.1 Static Hypothermic Ischaemic Storage Of The Donor Heart...... 59 1.5.2.2 Reperfusion Injury...... 63 1.5.2.3 Oxygen-Derived Free Radical Injury ...... 63 1.5.2.4 Calcium Overload During Reperfusion...... 64 1.5.2.5 The Role Of White Blood Cells In Ischaemic Storage Injury...... 65 1.5.2.6 Endothelial Injury During Ischaemia And Reperfusion...... 65 1.5.2.7 The Impact Of Ischaemic Storage Time On Heart Transplantation Outcomes ...... 67 1.5.3 Organ Preservation Solutions...... 69 1.5.3.1 Formulation Of Preservation Solutions ...... 69

ix 1.5.3.2 Electrolyte Composition Of Preservation Solutions...... 71 1.5.3.3 Chemical Additives To Preservation Solutions...... 74 1.5.3.4 The Development Of Celsior As A Heart Preservation Solution...... 74 1.5.4 Novel Approaches To Myocardial Protection...... 76 1.5.4.1 Sodium-Hydrogen Exchange Inhibition To Prevent Intracellular Calcium Overload And Protect The Heart...... 76 1.5.4.2 The Anti-Ischaemic Effects Of Glyceryl Trinitrate...... 80 1.6 EXPERIMENTAL MODELS IN CARDIAC TRANSPLANTATION RESEARCH ...... 83 1.6.1 The Use Of A Porcine Model Of The Brain-Dead Donor And Orthotopic Cardiac Transplantation ...... 83 1.6.2 The Use Of The Preload Recruitable Stroke Work (PRSW) Relationship To Analyse Cardiac Function...... 85 1.7 AIMS OF THE RESEARCH REPORTED IN THIS THESIS ...... 86

CHAPTER 2 EXPERIMENTAL METHODS...... 89 2.1 INTRODUCTION ...... 90 2.2 EXPERIMENTAL DESIGN ...... 90 2.3 ANIMALS...... 91 2.4 PERSONNEL...... 92 2.5 DONOR MANAGEMENT...... 93 2.5.1 Donor Anaesthetic Management And Monitoring...... 93 2.5.2 Preparation For The Induction Of Brain Death In The Donor Animal ...... 98 2.5.3 Donor Animal Surgery And Cardiac Instrumentation For Data Acquisition...... 101 2.5.4 Induction Of Brain Death ...... 106 2.5.5 Donor Management...... 107 2.5.6 Hormone Resuscitation Protocol...... 107 2.5.7 Cardiac Allograft Explantation And Preservation...... 108 2.6 RECIPIENT MANAGEMENT...... 111 2.6.1 Recipient Anaesthetic Management And Monitoring...... 111

x 2.6.2 Recipient Animal Surgery And Institution Of Cardiopulmonary Bypass Support...... 111 2.6.3 Orthotopic Cardiac Allograft Transplantation...... 114 2.6.4 Management Of The Cardiac Transplant Recipient After Reperfusion And Weaning From Cardiopulmonary Bypass Support ...... 117 2.6.5 Left Ventricular Wall Volume Assessment...... 119 2.7 DATA ACQUISITION AND ANALYSIS...... 119 2.7.1 Haemodynamic Data...... 119 2.7.2 Left Ventricular Pressure-Volume Loops ...... 120 2.7.3 Normalisation Of Pressure-Volume Loop Data ...... 122 2.7.4 Assessment Of Left Ventricular Contractility Utilising The Preload Recruitable Stroke Work (PRSW) Relationship...... 123 2.7.5 Cardiac Troponin I ...... 125 2.7.6 Pulmonary Function...... 125 2.7.7 Data Reporting And Statistical Analyses ...... 126

CHAPTER 3 THE EFFECTS OF HORMONE RESUSCITATION ON CARDIAC FUNCTION AND HAEMODYNAMICS IN THE BRAIN-DEAD ORGAN DONOR ...... 128 3.1 INTRODUCTION ...... 129 3.2 METHODS...... 131 3.2.1 Porcine Model Of The Brain-Dead Organ Donor ...... 131 3.2.2 Donor Management...... 131 3.2.3 Data Acquisition And Study Outcomes ...... 132 3.2.4 Power Calculation And Statistical Analyses ...... 133 3.3 RESULTS ...... 134 3.3.1 Experimental Animals...... 134 3.3.2 Inotrope Requirements...... 135 3.3.3 Cardiac Contractility...... 136 3.3.4 Haemodynamic Changes...... 142 3.3.5 Troponin I, Blood Glucose, Pulmonary Function And Acid-Base Balance...... 144

xi 3.4 DISCUSSION...... 146 3.4.1 Study Rationale And Experimental Groups ...... 146 3.4.2 The Impact Of Hormone Resuscitation On Inotrope Usage To Maintain Blood Pressure In The Donor...... 147 3.4.3 The Effects Of Brain Death On Cardiac Contractility ...... 148 3.4.4 The Use Of Hormone Resuscitation To Ameliorate The Negative Effects Of Brain Death On Cardiac Contractility...... 148 3.4.5 The Effects of Donor Brain Death And Subsequent Donor Management On Troponin I Release...... 150 3.4.6 The Effects Of Hormone Resuscitation On the Haemodynamics Of The Brain-Dead Cardiac Donor ...... 151 3.4.7 The Effects Of Steroids On Pulmonary Function And Blood Glucose Control ...... 152 3.4.8 Study Limitations...... 152 3.5 CONCLUSION ...... 153

CHAPTER 4 THE EFFECTS OF HORMONE RESUSCITATION ON THE LUNGS AND ABDOMINAL ORGANS FOR TRANSPLANTATION IN THE BRAIN-DEAD MULTI-ORGAN DONOR...... 155 4.1 INTRODUCTION ...... 156 4.2 METHODS...... 158 4.2.1 Porcine Model Of The Brain-Dead Organ Donor ...... 158 4.2.2 Donor Animal Anaesthesia, Surgery And Preparation For Brain Death Induction And Cardiac Data Acquisition...... 158 4.2.3 Abdominal Surgery And Instrumentation Of The Abdominal Organs For Data Acquisition ...... 158 4.2.4 Cardiothoracic And Abdominal Tissue Biopsy Protocol ...... 164 4.2.5 Experimental Protocol...... 164 4.2.6 Study Outcomes...... 169 4.2.7 Power Calculation And Statistical Analyses ...... 169 4.3 RESULTS ...... 170 4.3.1 Experimental Animals...... 170

xii 4.3.2 Inotrope Requirements, Haemodynamic Indices And Cardiac Function..172 4.3.3 Coronary Blood Flow And Troponin I Release ...... 177 4.3.4 Pulmonary Function...... 179 4.3.5 The Liver: Blood Flow, Injury And Function ...... 179 4.3.6 Blood Glucose Control And Exogenous Insulin Requirements...... 183 4.3.7 Pancreatic Injury...... 187 4.3.8 Renal Function...... 187 4.3.9 Acid-Base Balance...... 190 4.4 DISCUSSION...... 193 4.4.1 Study Rationale And Experimental Groups ...... 193 4.4.2 Hormone Resuscitation To Improve Haemodynamics And Cardiac Function ...... 194 4.4.3 Myocardial Injury...... 196 4.4.4 The Negative Effects Of Noradrenaline On Pulmonary Function ...... 197 4.4.5 Does Hormone Resuscitation Affect The Liver Any Differently To Conventional Treatments Used In Donor Management?...... 200 4.4.6 Blood Glucose Control And The Management Of The Donor Pancreas ..203 4.4.7 Superior Renal Function With Hormone Resuscitation ...... 205 4.4.8 The Decline In pH In The Organ Donor ...... 208 4.4.9 Study Limitations...... 209 4.5 CONCLUSION ...... 210

CHAPTER 5 THE ROLE OF CARIPORIDE AND GLYCERYL TRINITRATE IN IMPROVING LONG-TERM PRESERVATION OF THE DONOR HEART FOR TRANSPLANTATION ...... 212 5.1 INTRODUCTION ...... 213 5.2 METHODS...... 215 5.2.1 Porcine Model Of Orthotopic Heart Transplantation...... 215 5.2.2 Preparation Of The Donor Animal And Subsequent Management...... 216 5.2.3 Experimental Groups...... 217 5.2.4 Preparation Of Cariporide ...... 218

xiii 5.2.5 Preparation Of The Recipient Animal And Orthotopic Cardiac Allograft Transplantation...... 219 5.2.6 Management Of The Transplant Recipient And Weaning From Cardiopulmonary Bypass Support ...... 219 5.2.7 Study Outcomes...... 220 5.2.8 Data Acquisition And Analysis...... 220 5.2.9 Power Calculation And Statistical Analyses ...... 220 5.3 RESULTS ...... 222 5.3.1 Experimental Animals...... 222 5.3.2 Weaning From Cardiopulmonary Bypass ...... 223 5.3.3 Cardiac Contractility...... 224 5.3.4 Haemodynamic Changes In The Donor And The Recipient Post- Transplantation...... 228 5.3.5 Left Anterior Descending Coronary Artery (LAD) Flow ...... 231 5.3.6 Troponin I...... 234 5.4 DISCUSSION...... 235 5.4.1 Study Rationale And Experimental Groups ...... 235 5.4.2 The Ability To Wean Transplanted Off Cardiopulmonary Bypass Support After 14 Hours Static Hypothermic Ischaemic Storage ...... 237 5.4.3 The Effects Of Long-Term Storage On Left Ventricular Function And Haemodynamics After Transplantation ...... 237 5.4.4 Troponin I Release As A Marker Of Myocardial Injury...... 239 5.4.5 Increases In Left Anterior Descending Coronary Arterial Flow After Transplantation...... 239 5.4.6 The Role Of Cariporide To Ameliorate Ischaemia-Reperfusion Injury In Transplantation...... 240 5.4.7 The Role Of Glyceryl Trinitrate To Ameliorate Ischaemia-Reperfusion Injury In Transplantation And Its Synergistic Interaction With Cariporide ...... 243 5.4.8 Study Limitations...... 245 5.5 CONCLUSION ...... 246

xiv CHAPTER 6 FINAL DISCUSSION: A SUMMARY OF THE KEY FINDINGS, THEIR RELEVANCE TO CLINICAL PRACTICE AND THE SCOPE FOR FUTURE RESEARCH ...... 248 6.1 CARDIAC TRANSPLANTATION AS A TREATMENT FOR END- STAGE HEART FAILURE...... 249 6.2 SUMMARY OF THE KEY FINDINGS IN THIS THESIS...... 250 6.3 THE RELEVANCE TO CLINICAL TRANSPLANTATION...... 253 6.4 STUDY LIMITATIONS...... 254 6.5 FUTURE DIRECTIONS...... 255

REFERENCES...... 258

xv PRESENTATIONS

Oral Presentations

“Optimising the quality of donor organs for transplantation: From donor management to organ preservation” (invited talk). The Transplantation Society New Key Opinion Leader Meeting, Sydney, July 2007

“Hormone resuscitation of the brain-dead donor to improve organ quality for transplantation”. The Transplantation Society of Australia & New Zealand Annual Scientific Meeting, Canberra, March 2007

“The Effects of Hormone Resuscitation on Transplantable Organs in the Brain Dead Donor”. The International Society for Heart and Lung Transplantation 27th Annual Meeting and Scientific Sessions, San Francisco USA, April 2007 (this abstract was chosen as a finalist in the Phillip K. Caves Award Session)

“The Effects of Donor Pre-Treatment on the Heart and Other Solid Organs for Transplantation: A Comparison Between Hormone Resuscitation, Noradrenaline and Intravenous Fluids to Improve Organ Quality”. Combined meeting of The Australasian Society of Cardiac and Thoracic Surgeons and The International Society for Minimally Invasive Cardiac Surgery, Cairns October 2006.

“Hormone Resuscitation of the Brain Dead Donor and its Effects On Transplantable Organs”. World Transplant Congress, Boston USA, July 2006

“The Use of a Porcine Model of the Brain-Dead Multi-Organ Donor to Assess the Effects of Hormone Resuscitation to Improve Donor Organ Quality”. The Surgical Research Society of Australasia 43rd Annual Scientific Meeting, Sydney, May 2006

“Preserving The Donor Heart For Transplantation Using Cariporide and Glyceryl Trinitrate”. The Transplantation Society of Australia & New Zealand Annual Scientific Meeting, Canberra, April 2006

xvi “Hormone Resuscitation of the Brain Dead Donor and its Effects On Transplantable Organs: Early Results”. The Transplantation Society of Australia & New Zealand Annual Scientific Meeting, Canberra, April 2006

“Improving Long-Term Preservation of the Transplanted Heart Using Cariporide and Glyceryl Trinitrate”. The Australasian Society of Cardiac and Thoracic Surgeons Annual Scientific Meeting, Noosa October 2005

“Improving Long-Term Preservation of the Transplanted Heart Using Cariporide and Glyceryl Trinitrate”. The 15th St. Vincent’s & Mater Health Sydney Research Symposium, Sydney, October 2005

“Cariporide In Long Term Cardiac Preservation In A Porcine Heart Transplantation Model”. The Transplantation Society of Australia & New Zealand Annual Scientific Meeting, Canberra, May 2005

“Assessment of Cariporide In Long Term Hypothermic Storage Using a Porcine Model of Heart Transplantation”. The Australasian Society of Cardiac and Thoracic Surgeons Annual Scientific Meeting, Noosa October 2004

“Assessment of Hormone Resuscitation in the Brain Dead Donor Using a Porcine Model”. The Transplantation Society of Australia & New Zealand Annual Scientific Meeting, Canberra April 2004

Poster Presentations

“Preserving the Donor Heart for Transplantation Using Cariporide and Glyceryl Trinitrate”. The University of NSW Faculty of Medicine Research Day, Sydney, October 2006 (same poster as for CSANZ ASM 2006)

“Hormone Resuscitation of the Brain-Dead Organ Donor and Its Effects On Transplantable Organs”. The 16th St. Vincent’s & Mater Health Sydney Research Symposium, Sydney, September 2006

xvii “Preserving the Donor Heart For Transplantation Using Cariporide and Glyceryl Trinitrate”. The Cardiac Society of Australia and New Zealand Annual Scientific Meeting, Canberra, August 2006

Tran P, Hing A, Hicks M, Gao L, Faddy S, Kesteven S, Laurence J, Stewart G, Macdonald P, Sharland A. “Expression of NKG2D Ligands Following Brain-Death”. World Transplant Congress, Boston USA, July 2006

“Minimizing Ischemia Reperfusion Injury With Cariporide and Glyceryl Trinitrate During Long-Term Preservation of the Donor Heart for Transplantation”. World Transplant Congress, Boston USA, July 2006

“Optimising Long-Term Preservation of the Transplanted Heart Using Glyceryl Trinitrate and Cariporide”. The International Society for Heart and Lung Transplantation 26th Annual Meeting and Scientific Sessions, Madrid Spain, April 2006

“Sodium Hydrogen Exchange Inhibition In Long-Term Hypothermic Preservation Of The Transplanted Porcine Heart”. The University of NSW Faculty of Medicine Merck Sharp and Dohme (Aust) Research Student Poster Awards, Sydney, October 2005 (same poster as for CSANZ ASM 2005)

“Sodium Hydrogen Exchange Inhibition In Long-Term Hypothermic Preservation Of The Transplanted Porcine Heart”. The Cardiac Society of Australia and New Zealand Annual Scientific Meeting, Perth, August 2005

“Resuscitation of the Brain Dead Donor Heart Using Hormones in a Porcine Model”. The International Society for Heart and Lung Transplantation 25th Annual Meeting and Scientific Sessions, Philadelphia PA, April 2005

“Cariporide Improves Long Term Cardiac Preservation in a Porcine Model of Heart Transplantation”. The International Society for Heart and Lung Transplantation 25th Annual Meeting and Scientific Sessions, Philadelphia PA, April 2005

xviii “Assessment of Hormone Resuscitation in the Brain Dead Donor Using a Porcine Model”. The 14th St. Vincent’s & Mater Health Sydney Research Symposium, Sydney, September 2004 (same poster as for CSANZ ASM 2004).

“Assessment of Hormone Resuscitation in the Brain Dead Donor Using a Porcine Model”. The Cardiac Society of Australia and New Zealand Annual Scientific Meeting, Brisbane August 2004

xix PUBLICATIONS

Peer Reviewed Research Publications

Hing AJ, Watson A, Hicks M, Gao L, Faddy SC, McMahon AC, Kesteven SH, Wilson MK, Jansz P, Feneley MP, Macdonald PS. Combining cariporide with glyceryl trinitrate optimises cardiac preservation during porcine heart transplantation. American Journal of Transplantation. 2009; 9:2048-2056. Impact factor: 6.559.

Hing AJ, Hicks M, Garlick S, Gao L, Kesteven SH, Faddy SC, Wilson MK, Feneley MP, Macdonald PS. The effects of hormone resuscitation on cardiac function and hemodynamics in a porcine brain-dead organ donor model. American Journal of Transplantation. 2007; 7:809-817. Impact factor: 6.559.

Hing AJ, Hicks M, Gao L, Faddy SC, Tran P, Kesteven SH, Sharland AF, Stewart GJ, Macdonald PS. The effects of hormone resuscitation on the lungs and abdominal organs for transplantation in a porcine brain-dead organ donor model. (To be submitted).

Reviews

Hicks M, Hing A, Gao L, Ryan J, Macdonald PS. Organ Preservation. In: Methods in Molecular Biology: Transplantation Immunology Methods and Protocols. Vol. 333, pp. 331-373; Eds: P Hornick and M Rose (2006). Totowa, NJ: Humana Press Inc. (Invited review).

Hing A, Hicks M, Gao L, Wilson M, Mackie F, Macdonald PS. The case for a standardised protocol that includes hormone resuscitation for management of the cadaveric multi-organ donor. Critical Care and Resuscitation. 2005; 7:43-50.

Abstracts

Hing A, Tran P, Hicks M, Gao L, Faddy SC, Kesteven SH, Stewart G, Sharland A, Macdonald PS. “The effects of hormone resuscitation on transplantable organs in the brain dead donor”. The Journal of Heart and Lung Transplantation. 2007; 26(2):s214.

xx Hing A, Hicks M, Gao L, Faddy S, McMahon A, Kesteven S, Feneley M, Wilson M, Macdonald P. “Preserving the Donor Heart For Transplantation Using Cariporide and Glyceryl Trinitrate”. Heart Lung and Circulation. 2006; 15S:S15

Hing A, Hicks M, Gao L, Faddy SC, McMahon AC, Kesteven SH, Feneley MP, Wilson MK, Macdonald PS. “Optimising Long-Term Preservation of the Transplanted Heart Using Glyceryl Trinitrate and Cariporide”. The Journal of Heart and Lung Transplantation. 2006; 25(2S): S160.

Hing A, Hicks M, Gao L, Faddy S, McMahon A, Kesteven S, Feneley M, Wilson M, Macdonald P. “Assessment of Cariporide In Long Term Hypothermic Storage Using a Porcine Model of Heart Transplantation”. Heart Lung and Circulation. 2005; 14(3):219

Hing A, Hicks M, Garlick S, Gao L, Kesteven S, Wilson M, Feneley M, Macdonald P. “Resuscitation of the Brain Dead Donor Heart Using Hormones in a Porcine Model”. The Journal of Heart and Lung Transplantation. 2005; 24(2S): S152-3.

Hing A, Hicks M, Gao L, Faddy S, McMahon A, Kesteven S, Feneley M, Wilson M, Macdonald P. “Cariporide Improves Long Term Cardiac Preservation in a Porcine Model of Heart Transplantation”. The Journal of Heart and Lung Transplantation. 2005; 24(2S): S153.

Hing A, Hicks M, Garlick S, Gao L, Kesteven S, Wilson M, Feneley M, Macdonald P. Assessment of hormone resuscitation in the brain dead donor using a porcine model. Heart Lung and Circulation. 2004; 13 (Supplement 2):S53.

Hing A, Hicks M, Garlick S, Gao L, Kesteven S, Wilson M, Feneley M, Macdonald P. Assessment of hormone resuscitation in the brain dead donor using a porcine model. Immunology and Cell Biology. 2004; 82 (Supplement):A9-10.

xxi PRIZES

Finalist in the Phillip K. Caves Award at The International Society for Heart and Lung Transplantation 27th Annual Meeting and Scientific Sessions, San Francisco USA, April 2007

The Transplantation Society of Australia and New Zealand Young Investigator Award: TSANZ Annual Scientific Meeting, 2007

The Merck Sharp and Dohme (Aust) Research Student Poster Certificate of Merit, University of NSW Faculty of Medicine Research Day, October 2006

The Surgical Research Society of Australasia Award for outstanding scientific paper presented at the Annual Scientific Meeting 2006.

The Transplantation Society of Australia and New Zealand Young Investigator Award: TSANZ Annual Scientific Meeting, 2006

Australasian Society of Cardiac and Thoracic Surgeons TAG Medical Young Achievers Award for 2005, ASCTS Annual Scientific Meeting 2005

The Merck Sharp and Dohme (Aust) Research Student Poster Award, University of NSW Faculty of Medicine October 2005

The Transplantation Society of Australia and New Zealand: Elsevier Prize, TSANZ Annual Scientific Meeting 2005

The Transplantation Society of Australia and New Zealand Novartis Young Investigator Award: TSANZ Annual Scientific Meeting, 2005

Australasian Society of Cardiac and Thoracic Surgeons TAG Medical Young Achievers Award for 2004, ASCTS Annual Scientific Meeting 2004

xxii The Transplantation Society of Australia and New Zealand: Award for the Best Laboratory Based Presentation, TSANZ Annual Scientific Meeting 2004 (sponsored by Kidney Health Australia)

The Transplantation Society of Australia and New Zealand Novartis Young Investigator Award: TSANZ Annual Scientific Meeting, 2004

xxiii RESEARCH SCHOLARSHIPS

Royal Australasian College of Surgeons: Surgeon Scientist Scholarship, 2004-2006

National Health and Medical Research Council: Medical Postgraduate Scholarship, 2004-2006

National Heart Foundation Postgraduate Research Scholarship 2004-2006: Scholarship declined due to award of RACS and NHMRC Scholarships

The Cardiac Society of Australia and New Zealand: Research Scholarship, 2003

TRAVEL SCHOLARSHIPS

Heart Foundation of Australia Travel Grant 2006: travel grant to attend and present at the World Transplant Congress, 2006

The Transplantation Society of Australia and New Zealand Travel Award 2006: travel award to attend and present at the World Transplant Congress, 2006.

Heart Foundation of Australia Travel Grant 2005: travel grant to attend and present at the International Society for Heart and Lung Transplantation 25th Anniversary Meeting and Scientific Sessions, 2005

The Transplantation Society of Australia and New Zealand Travel Award 2005: travel award to attend and present at the International Society for Heart and Lung Transplantation 25th Anniversary Meeting and Scientific Sessions, 2005

The Cardiac Society of Australia and New Zealand Annual Scientific Meeting Scholarship: travel scholarship to attend and present at the CSANZ ASM, 2004

xxiv LIST OF FIGURES

Figure Title Page 1.1 International Society for Heart and Lung Transplantation Registry 13 Kaplan-Meier survival data for adult and paediatric heart transplants performed between January 1982 and June 2006 (with 95% confidence limits). Conditional half-life = time to 50% survival for those recipients surviving the first year post-transplantation (Taylor et al. 2008).

1.2 Actuarial survival for all heart transplants performed in Australia 13 and New Zealand (from the Australia and New Zealand Cardiothoracic Organ Transplant Registry – ANZCOTR) from February 1984 to December 2007 (Keogh et al. 2008) compared with the International Society for Heart and Lung Transplantation Registry (ISHLTR) survival data for all heart transplants performed from January 1982 to June 2006 with 95% confidence limits (from Figure 1.1; Taylor et al. 2008).

1.3 Relative incidence of the leading causes of death for post-transplant 14 deaths reported from January 1992 through June 2008. CAV, coronary allograft vasculopathy; lymph/PTLD, lymphoma or post- transplant lymphoproliferative disease; CMV, cytomegalovirus (Taylor et al. 2009).

1.4 An analysis of the Cardiac Transplant Research Database (CTRD) 16 of post-transplant outcomes for 7 283 heart transplants performed between 1990 and 1999 illustrates the impact of donor age and recipient age on death due to coronary artery disease in the transplanted heart beyond the first year post transplant (Kirklin et al. 2004).

1.5 Number of heart transplants reported to the ISHLT Transplant 17 Registry by year (1982-2006). It should be noted that this figure includes only the heart transplants that are reported to the Registry

xxv Figure Title Page and hence does not represent the worldwide total number of transplants performed (Taylor et al. 2008).

1.6 Number of heart and heart-lung transplants reported to the Australia 18 and New Zealand Cardiothoracic Organ Transplant Registry by year (February 1984 to December 2007) (Keogh et al. 2008).

1.7 International donor statistics for 2007 - donors per million head of 19 population. Donor rates in Australia and New Zealand compares very unfavourably with the rest of the world (Source: International Registry of Organ Donation and Transplantation (Excell et al. 2008))

1.8 Effects of donor brain death, organ retrieval and the transplantation 32 process on donor organs and their impact on the immune response. Activation of the inflammatory and immune systems can determine the occurrence of acute and chronic rejection. Diagram reproduced from Gasser et al (2001).

1.9 The cardiac donor management algorithm developed by the Crystal 41 City Consensus Conference Heart Work Group, which have been adopted into the United Network for Organ Sharing (UNOS) Critical Pathway (Zaroff et al. 2002).

1.10 The effects of thyroid hormone on cardiovascular haemodynamics. 50 Triiodothyronine increases cardiac output by affecting tissue oxygen consumption (thermogenesis), vascular resistance, blood volume, cardiac contractility and heart rate. RAA=renin- angiotensin-aldosterone. Diagram reproduced from Klein and Ojamaa (2001).

1.11 Activity of the Na+-H+ exchanger (NHE) under a) normal 62 conditions; and b) during ischaemic conditions. During non- ischaemic conditions, ATP supply is non-limiting. Intracellular sodium is extruded by the Na+-K+ ATPase, calcium is extruded by

xxvi Figure Title Page the Na+-Ca2+ exchanger (NCE) and the NHE is quiescent. During ischaemia, ATP is depleted and the Na+-K+ ATPase becomes inactive. The accumulation of hydrogen ions (H+) as a result of glycolysis (anaerobic metabolism) activates the NHE to eliminate H+ from the cell, resulting in a large influx of sodium ions. Sodium is cleared from the cell by the NCE in reverse mode, resulting in a dangerous accumulation of intracellular calcium that may cause electrical instability, contractile dysfunction and myocyte death. Diagram reproduced from Hicks et al (2006).

1.12 The opposing relationships between the vasoconstrictor endothelin 66 and the vasodilators nitric oxide and prostacyclin pathways in the regulation of coronary artery tone (adapted from Humbert et al. 2004).

1.13 The effects of donor heart ischaemic time on a) 1-year mortality; b) 68 5-year mortality; and c) 10-year mortality. 95% confidence limits are shown. (Taylor et al. 2008; www.ishlt.org/registries/slides.asp?slides=heartLungRegistry).

2.1 Animal transportation crate. a) External; b) Internal. 92

2.2 Intramuscular injection site for pre-medication (circled) in: a) the 94 trapezius muscle (posterior neck), and b) the gluteal muscle.

2.3 Intravenous cannulation of the porcine dorsal ear vein with 20 G 94 cannula.

2.4 Laryngoscope with blade extension for porcine intubation. 95

2.5 Porcine laryngoscopy: visualising the vocal cords for intubation. 96

2.6 Animal secured onto the operating theatre table and theatre set-up, 97

including ventilator (a); monitoring equipment (including O2 pulse oximeter, Datex capnograph and SpaceLabs medical monitor) (b); infusion pumps delivering various treatments such as hormone

xxvii Figure Title Page resuscitation (c); and data acquisition equipment to capture data such as intra-ventricular pressures and cardiac dimensions from the ultrasonic pressure transducers and micromanometer-tipped catheters (d) (see text for further details regarding monitoring equipment).

2.7 Surgery for induction of brain death. a) Burr hole in the right fronto- 100 parietal region of the skull; b) underlying dura mater (arrow); and c) Foley catheter in situ.

2.8 The porcine sternum. a) Median sternotomy, and b) superior part of 102 sternum. Note the shape and thickness of the sternum (arrow) – often 3-5 cm in maximal depth at the superior aspect of the sternum.

2.9 The instrumented porcine heart. The heart has been exposed by silk 103 stay sutures to the pericardial edges, creating a pericardial well (a); a right internal mammary vein central venous catheter is in situ (b); ultrasonic dimension transducers have been sewn to the epicardium (c); micromanometer-tipped catheters have been placed in both the left and right ventricles (d) and flow probe around the left anterior descending coronary artery (e). Ao = aorta; RVOT= right ventricular outflow tract; RV = right ventricle; LV = left ventricle; LAD = left anterior descending coronary artery; LAA = left atrial appendage; and RAA = right atrial appendage.

2.10 Ultrasonic dimension transducer (a), flow probe (b) and 105 micromanometer-tipped catheter (c).

2.11 Left anterior descending coronary artery flow probe in situ (a). Note 105 also the ultrasonic dimension transducer sewn in situ (b).

2.12 The left azygos vein (a). 106

2.13 Explanted donor heart stored in a plastic bag of preservation 110 solution, then placed in an esky filled with crushed ice.

2.14 The cardiopulmonary bypass machine – roller pumps (a); 112

xxviii Figure Title Page oxygenator (b); and heater/cooler (c).

2.15 The pericardial cavity after recipient cardiectomy with atrial sewing 114 cuffs prepared. SVC = superior vena cava venous cannula; IVC = inferior vena cava venous cannula; AoC = aortic arterial cannula; Ao = aorta; AXC = aortic cross clamp; PA = pulmonary artery; RA = right atrial cuff; LA = left atrial cuff.

2.16 Acquisition of ECG, LAD flow, and cardiac pressure and dimension 121 data using SonoSoft – data traces shown on computer monitor during data acquisition.

2.17 The geometric model for estimation of left ventricular volume: a) 122 Diagrammatic representation of the ultrasonic dimension transducer positions on the short (minor; anterior-posterior) and long (major; base-apex) axes of the left ventricle and the position of the left ventricular pressure catheter; b) Diagrammatic representation of the prolate ellipsoid model used to calculate epicardial left ventricular volume. This model is the assumed shape of the left ventricle – the prolate ellipse is formed by an ellipse rotated around its long axis (Z) and has a circular cross-section in the short axis (X-Y plane). Therefore to calculate its volume, the short axis and the long axis needs to be measured.

3.1 Noradrenaline Usage in Individual Animals: a) Control and b) 138 Hormone Resuscitation. *Noradrenaline dose at 6 hours was 9.333 μg/kg/min; #Noradrenaline dose at 6 hours was 6.279 μg/kg/min. At one hour post-brain death, all animals required noradrenaline to support blood pressure. By three hours post-brain death, 8/8 hormone resuscitation and 8/9 control animals required noradrenaline. By six hours, 2/8 hormone resuscitation and 9/9 control animals required noradrenaline.

xxix Figure Title Page 3.2 Left Ventricular (LV) Pressure-Volume (PV) Loops. Representative 139 loops obtained during transient vena caval occlusion and taken from the same hearts at baseline and at 6.25 hours post-brain death in Control and Hormone Resuscitation groups. Note the decline in PV loop area (stroke work) in the control animal compared with the hormone resuscitation animal between baseline and 6.25 hours.

3.3 Preload Recruitable Stroke Work (PRSW) Relationship. 140 Representative PRSW relationships from the Control and Hormone Resuscitation groups are compared. Note the decrease in slope in the control animal compared with an increase in slope in the hormone-resuscitated animal between baseline and 6.25 hours. Additionally, there is an increase in the x-axis intercept of the

PRSW relationship (nVw,epi) in the control animal compared with a decrease in the hormone-resuscitated animal. These changes indicate a decline in contractility in the control animal compared with the hormone-treated animal.

4.1 (a) Left kidney with renal artery flow probe in situ. (b) Magnified 160 view of hilum of the left kidney demonstrating the left renal artery (RA) with a flow probe around it, renal vein (RV) and ureter (Ur).

4.2 Mobilised porcine pancreas (Ps). St = stomach (reflected 161 superiorly); Spl = spleen; Duo = duodenum; SI = small intestine.

4.3 a) Mobilising the portal triad with the liver reflected superiorly; and 162 b) The portal triad demonstrating the cannulated common bile duct (CBD) for bile collection, the portal vein (PV) and the hepatic artery (HA).

4.4 Placement of flow probes around the portal vein (PV) and the 163 hepatic artery (HA). The liver is reflected superiorly and the common bile duct (CBD) catheter is in situ.

xxx Figure Title Page 4.5 Study design and the donor management protocol for each of three 166 experimental groups. Intravenous (IV) saline was given in each group to maintain CVP 0-5 mmHg. Saline was also given in the control group to maintain mean arterial pressure (MAP). Data and tissue sampling time-points are also shown in the figure. Tissue biopsies were taken from the heart, lung, liver, kidney and pancreas. Arterial blood gas and blood glucose analyses were performed at baseline and at hourly intervals post-brain death induction. Further details can be found in Section 4.2.5.

4.6 Haemodynamics. a) Heart Rate (bpm; beats per minute); b) Mean 174 Arterial Blood Pressure (MAP); and c) Central Venous Pressure (CVP). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. See Section 4.3.2 for further details.

4.7 Cardiac Function. a) Stroke Work (SW); b) Stroke Volume (SV); 176 and c) Cardiac Output (CO). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. See Section 4.3.2 for further details

4.8 Left Anterior Descending Coronary Artery (LAD) Blood Flow. IV 178 Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between FL and HR groups. See Section 4.3.3 for further details.

4.9 Pulmonary Function. a) PaO2/FiO2 Ratio; b) Alveolar-arterial (Aa) 180

Gradient (mmHg); and c) PaCO2 (mmHg). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. See Section 4.3.4 for further details.

4.10 Blood Flow To The Liver. a) Hepatic Artery Blood Flow (mL/min); 181 and b) Hepatic Portal Vein Blood Flow (mL/min). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR;

xxxi Figure Title Page n=9) Groups. There were no significant differences between groups at any time-point for both hepatic arterial and portal venous flow.

4.11 Bile Production (mL/hr). IV Fluids (FL; n=9), Noradrenaline (NA; 182 n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between groups at any time-point.

4.12 Liver Function Tests. a) Alanine Aminotransferase (ALT) (U/L); b) 184 Aspartate Aminotransferase (AST) (U/L); and c) Total Bilirubin (μmol/L). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between groups at any time-point.

4.13 International Normalised Ratio (INR). IV Fluids (FL; n=9), 185 Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between groups at any time-point.

4.14 Blood Glucose Levels (mmol/L). IV Fluids (FL; n=9), 186 Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. See Section 4.3.6 for further details.

4.15 Pancreatic Enzymes. a) Amylase (Units/L); and b) Lipase (Units/L). 188 IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between groups at any time-point.

4.16 Renal Function. a) Left Renal Artery Blood Flow (mL/min); and b) 189 Creatinine Clearance (mL/min). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. See Section 4.3.8 for further details.

4.17 Renal Function. a) Plasma Creatinine (μmol/L); and b) Plasma Urea 191 (mmol/L). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between groups at any time-point.

xxxii Figure Title Page 4.18 Acid-Base Balance. a) pH; and b) Base Excess (BE). IV Fluids (FL; 192 n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. See Section 4.3.9 for further details.

5.1 Left Ventricular (LV) Pressure-Volume Loops. Representative 225 loops from an animal from the cariporide pre-treatment group (CAR1) are shown. These loops were obtained from the donor heart prior to brain death, prior to explantation (approximately six hours after brain death induction) and after transplantation into the recipient animal (approximately two hours after weaning from cardiopulmonary bypass support). Data for these loops were acquired during transient occlusion of the inferior vena cava. Volumes were normalised to the baseline (pre-brain death) steady state end-diastolic volume.

5.2 Preload Recruitable Stroke Work (PRSW) Relationship. Derived 226 from the left ventricular pressure-volume loops shown in Figure 5.1. The PRSW relationships at pre-brain death (open circles), pre- explantation (open diamonds) and post-transplant (two hours post- weaning from cardiopulmonary bypass) (closed triangles) are shown. Data obtained during transient occlusion of the inferior vena cava. Stroke work and volume were normalised to the baseline (pre- brain death) steady state stroke work and end-diastolic volume.

xxxiii Figure Title Page 5.3 Left Ventricular Stroke Work (mean±standard deviation): CON = 229 control group (n=5); CAR1 = cariporide pre-treatment (n=5); CAR2 = cariporide + Celsior (n=5); GTN = glyceryl trinitrate + Celsior group (n=5); COMB = cariporide + glyceryl trinitrate + Celsior (n=6). Baseline data taken pre-brain death and post-weaning data taken after successful weaning from cardiopulmonary bypass post- heart transplant. As only one CON, one CAR2 and no GTN transplant/s were weaned from bypass, there are no CON, CAR2 and GTN results post-transplant. There were no significant differences in stroke work between groups post-transplantation. See text for further details of statistical analyses.

5.4 Mean Arterial Blood Pressure (mean±standard deviation): CON = 230 control group (n=5); CAR1 = cariporide pre-treatment (n=5); CAR2 = cariporide + Celsior (n=5); GTN = glyceryl trinitrate + Celsior group (n=5); COMB = cariporide + glyceryl trinitrate + Celsior (n=6). Baseline data taken pre-brain death and post-weaning data taken after successful weaning from cardiopulmonary bypass post- heart transplant. As only one CON, one CAR2 and no GTN transplant/s were weaned from bypass, there are no CON, CAR2 and GTN results post-transplant. *p<0.020 (compared with baseline). See text for further details of statistical analyses.

5.5 Cardiac Output (mean±standard deviation): CON = control group 232 (n=5); CAR1 = cariporide pre-treatment (n=5); CAR2 = cariporide + Celsior (n=5); GTN = glyceryl trinitrate + Celsior group (n=5); COMB = cariporide + glyceryl trinitrate + Celsior (n=6). Baseline data taken pre-brain death and post-weaning data taken after successful weaning from cardiopulmonary bypass post-heart transplant. As only one CON, one CAR2 and no GTN transplant/s were weaned from bypass, there are no CON, CAR2 and GTN results post-transplant. See text for details of statistical analyses.

xxxiv Figure Title Page 5.6 Left Anterior Descending Coronary Artery (LAD) Flow 233 (mean±standard deviation): CON = control group (n=5); CAR1 = cariporide pre-treatment (n=5); CAR2 = cariporide + Celsior (n=5); GTN = glyceryl trinitrate + Celsior group (n=5); COMB = cariporide + glyceryl trinitrate + Celsior (n=6). Baseline data taken pre-brain death and post-weaning data taken after successful weaning from cardiopulmonary bypass post-heart transplant. As only one CON, one CAR2 and no GTN transplant/s were weaned from bypass, there are no CON, CAR2 and GTN results post- transplant. See text for details of statistical analyses.

xxxv LIST OF TABLES

Table Title Page 1.1 Indications and contraindications for cardiac transplantation 10 (Kasper et al. 2002; Macdonald 2008). 1.2 Risk factors for mortality within one year after heart 15 transplantation for the era January 2002 to June 2007 in 10 705 recipients. Table includes the relative risk for mortality for each independent variable. Donor and recipient age, and donor heart ischaemic time continue to be powerful risk factors for death at one year post transplant in the more recent era (Taylor et al. 2009). 1.3 Composition of various preservation solutions used in organ 72 transplantation (from Michel et al. 2002). Concentrations are given as millimoles per litre unless otherwise stated. LYPS: Lyon preservation solution; STS-1: St. Thomas Hospital No. 1 cardioplegic solution; STS-2: St. Thomas Hospital No. 2 cardioplegic solution; UW-1: University of Wisconsin modified solution; UW: Standard University of Wisconsin solution; HTK: Histidine-Tryptophane-Ketoglutarate (Bretschneider) solution; STF: Stanford solution; EC: Euro-Collins solution; NaCl: Normal saline; Extra: Extracellular-type solution; Intra: Intracellular-type solution; PEG: Polyethyleneglycol; HES: hydroxyethyl starch.

2.1 Hormone resuscitation protocol. 108

3.1 Characteristics of the Control (n=9) and Hormone Resuscitation 135 (n=8) Groups. 3.2 Noradrenaline doses (μg/kg/min) required to maintain blood 136 pressure in the experimental groups. 3.3 Preload Recruitable Stroke Work (PRSW) Relationship: Control 141 (CON; n=9) and Hormone Resuscitation (HR; n=8) Groups. 3.4 Haemodynamics: Control (CON; n=9) and Hormone Resuscitation 143 (HR; n=8) Groups.

xxxvi Table Title Page 3.5 Troponin I (μg/L; median and range): Control (n=7) and Hormone 144 Resuscitation (n=7) Groups. 3.6 Blood glucose (mmol/L; mean±standard deviation): Control (n=9) 145 and Hormone Resuscitation (n=8) Groups. 3.7 Arterial blood gas results (mean±standard deviation): Control 146 (CON; n=9) and Hormone Resuscitation (HR; n=8) Groups.

4.1 Characteristics of the IV Fluids (control) (n=9), Noradrenaline 171 (n=9) and Hormone Resuscitation (n=9) Groups. 4.2 Troponin I (μg/L; median and range). IV Fluids (FL; n=9), 177 Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups.

5.1 Characteristics of Experimental Groups. 223 5.2 Preload Recruitable Stroke Work (PRSW) Relationship. 227 5.3 Troponin I (μg/L; median and range). 234

xxxvii LIST OF ABBREVIATIONS

Aa gradient Alveolar-arterial oxygen gradient ACT Activated clotting time ACTH Adrenocorticotrophic hormone ADH Antidiuretic hormone (also known as vasopressin) ALT Alanine aminotransferase ANOVA Analysis of variance ANZICS Australian and New Zealand Intensive Care Society ANZOD Australia and New Zealand Organ Donation Registry ANZCOTR Australia and New Zealand Cardiothoracic Organ Transplant Registry AST Aspartate aminotransferase ATP Adenosine 5’-triphosphate AV Atrioventricular AVP Arginine vasopressin BD Brain Death bpm Beats per minute (heart rate) Bypass Cardiopulmonary bypass CABG Coronary artery bypass CAR1 Cariporide pre-treatment study group (Chapter 5) CAR2 Cariporide plus Celsior treated study group (Chapter 5) CAV Cardiac allograft vasculopathy cGMP Guanosine-3’,5’-cyclic monophosphate CK Creatine kinase CK-MB MB isoenzyme of creatine kinase CMV Cytomegalovirus CO Cardiac output

CO2 Carbon dioxide COMB Cariporide plus glyceryl trinitrate plus Celsior treated study group (Chapter 5) CON Control study group CVA Cerebrovascular accident CVP Central venous pressure

xxxviii DC Direct current DCD [Organ] donation after cardiac death DDAVP 1-desamino-8-D-arginine vasopressin (also known as desmopressin) ECG Electrocardiogram EDP End-diastolic pressure

EDVepi End-diastolic volume (epicardial) EVLWI Extravascular lung water index

FiO2 Fractional concentration of oxygen in inspired gas FL Group IV fluids experimental group (Chapter 5) Fr French Ga Gauge GTN Glyceryl trinitrate or GTN plus Celsior treated study group (Chapter 5) h or hr Hour hrs Hours HR Hormone resuscitation HR Group Hormone resuscitation experimental group (Chapters 3, 4 and 5) iCa2+ Intracellular calcium ICAM-1 Intercellular adhesion molecule-1 ICP Intracranial pressure IFN- Interferon-gamma IL-1 Interleukin-1 IL-2 Interleukin-2 IL-6 Interleukin-6 INR International normalised ratio iNa+ Intracellular sodium ISHLT International Society For Heart And Lung Transplantation IV Intravenous IVC Inferior vena cava k The single-beat constant

+ K ATP ATP-sensitive potassium channel LAD Left anterior descending coronary artery LDH Lactate dehydrogenase LV Left ventricle/ventricular

xxxix LVP Left ventricular pressure

LVVepi Left ventricular volume (epicardial) MAP Mean arterial blood pressure MCP-1 Monocyte chemoattractant molecule-1 MIP-1 Macrophage inflammatory protein-1 alpha MLR Multiple linear regression

Mw Slope of PRSW relationship NA Group Noradrenaline treated experimental group (Chapters 5) Na+/K+ ATPase ATP-dependant sodium potassium transporter NCE Sodium calcium (Na+-Ca2+) ion exchanger or antiporter nEDVepi Normalised epicardial end-diastolic volume NHE Sodium hydrogen ion (Na+-H+) exchanger or antiporter NHMRC National Health and Medical Research Council, Australia nLVVepi Normalised epicardial left ventricular volume nMw Slope of the normalised PRSW relationship NO Nitric oxide nSW Normalised stroke work nVw,epi nEDVepi-axis intercept of the normalised PRSW relationship NYHA New York Heart Association OPTN The and Transplantation Network (USA)

O2 Oxygen

PaCO2 Arterial partial pressure of carbon dioxide

PaO2 Arterial partial pressure of oxygen PDA Posterior descending branch of the right coronary artery PFO Patent foramen ovale pHi Intracellular pH PRSW Preload recruitable stroke work PV Loop Pressure-volume loop RV Right ventricle/ventricular ssSW Steady state stroke work ssEDVepi Steady state end-diastolic volume (epicardial) SV Stroke volume SVC Superior vena cava

xl SW Stroke work SWI Stroke work index T3 Triiodothyronine T4 Thyroxine TnI Troponin I TNF- Tumour necrosis factor-alpha TSH Thyroid stimulating hormone UNOS The United Network for Organ Sharing (USA) VCAM-1 Vascular cell adhesion molecule-1 VF Ventricular fibrillation

Vw,,epi End diastolic epicardial volume-axis intercept of the PRSW relationship

xli ABSTRACT

Brain death has adverse effects on the organ donor, increasing organ dysfunction and affecting transplantation outcomes. It can also render organs unsuitable for transplantation. Another determinant of organ quality is ischaemia-reperfusion injury, which limits ischaemic storage time for hearts to six hours.

The aim of this thesis was to investigate the effectiveness of hormone resuscitation (HR) of the donor to ameliorate the effects of brain death. Another aim was to develop a donor management and organ preservation strategy to ameliorate the effects of ischaemia-reperfusion injury on the heart, thereby extending ischaemic preservation times. A porcine model of the brain-dead multi-organ donor with orthotopic cardiac transplantation was utilised.

Donor HR was shown to improve cardiac contractility and haemodynamics, thereby reducing inotrope requirements. A follow-up study investigating the effects of three different donor management protocols demonstrated that donor haemodynamics, renal arterial flow and creatinine clearance were superior in HR animals compared with animals treated with noradrenaline or intravenous fluid alone. Noradrenaline was associated with a significant deterioration in pulmonary function (PaO2 and alveolar- arterial oxygen gradient) and a decline in donor pH. HR was not associated with any detrimental effects on the lungs, liver or pancreas compared with the other two groups.

Preservation strategies incorporating glyceryl trinitrate (GTN) and cariporide, a Na+-H+ exchange inhibitor, were investigated to safely extend cardiac ischaemic preservation times. Pre-treatment with intravenous cariporide prior to heart explantation (donor) and reperfusion of the transplanted heart (recipient) was shown to effectively extend ischaemic time to 14 hours, evidenced by weaning off cardiopulmonary bypass. GTN and cariporide-supplemented Celsior, used as a cardioplegic/storage solution, was also effective in extending preservation time to 14 hours, with superior cardiac contractility compared with cariporide pre-treated hearts. Both treatments also ameliorated reperfusion injury, stabilising haemodynamics for up to three hours post-bypass.

xlii This thesis has demonstrated the effectiveness of HR to ameliorate the negative effects of donor brain death. It also provides evidence that combined GTN and cariporide- supplemented Celsior improves long-term preservation of the donor heart. These strategies offer the potential to increase the proportion of transplantable organs, to improve donor organ quality, and thereby improve transplantation outcomes.

xliii

CHAPTER 1

AN INTRODUCTION AND REVIEW OF THE LITERATURE IN CARDIAC TRANSPLANTATION AND THE MANAGEMENT OF THE BRAIN- DEAD ORGAN DONOR

- 1 - CHAPTER 1

AN INTRODUCTION AND REVIEW OF THE LITERATURE IN CARDIAC TRANSPLANTATION AND THE MANAGEMENT OF THE BRAIN-DEAD ORGAN DONOR

1.1 THE CHALLENGES IN CARDIAC TRANSPLANTATION

The first experimental cardiac transplant was reported in 1905 in a canine model (Carrel et al. 1905) and the first human to human cardiac transplant was performed in 1967 (Barnard 1967). Since that time, cardiac transplantation has become an established treatment for patients with severe end-stage cardiac failure refractory to optimal medical therapies. In a relatively short period of time, over 80 000 heart transplants have been performed throughout the world (Taylor et al. 2008), of which more than 1 900 were carried out in Australia and New Zealand (Keogh et al. 2008). However despite its successes, there are still many challenges faced by this treatment that limit its availability and also limit its success, in terms of morbidity and mortality.

The first hurdle faced by heart transplantation is the issue of supply versus demand. It is well recognised that the ever-growing demand for cardiac transplantation is not matched by the supply of suitable donor hearts available for transplantation (Copeland 2001; Conte et al. 2002; Kasper et al. 2002). At the same time, the mortality rate for those on the transplant waiting list has been reported to be as high as 17-22% per year (Copeland 2001; Zaroff et al. 2002). Clearly, there is a need to increase the number of donor organs available for transplantation, either by increasing the number of organ donors or by maximising the usage of organs that are available from the currently available pool of organ donors. An increase in suitable donor organ availability would improve the quality of life of those suffering from refractory end-stage cardiac failure and prevent them from dying on waiting lists.

- 2 - A significant factor that also requires consideration is the issue of donor organ quality. The success of is obviously critically dependant on the quality of the donor organ. This is determined by a number of factors, including donor age and pre-existing disease, the mechanism of brain death, donor management prior to organ procurement, the duration and manner of hypothermic storage, and the circumstances of reperfusion. Whilst transplant clinicians cannot influence the donor-related factors such as age, pre-existing disease and the mechanism of brain death, they can intervene at other stages of the transplantation process to improve the quality of outcomes in transplantation. For example, interventions can be implemented during the management of the donor in the intensive care unit, at the time of organ explantation, during organ storage, and at the time of transplantation and subsequent organ reperfusion.

Treatments aimed at improving transplantation outcomes need to be based on an understanding of the pathophysiology of the transplantation process and how interventions can minimise (and potentially reverse) the detrimental events that occurs during organ transplantation. Brain death in the organ donor results in a series of haemodynamic, neurohormonal, and pro-inflammatory/immune system perturbations that contribute to a high incidence of complications in the donor and donor organ dysfunction (Smith 2004; Hicks et al. 2006). These changes have the potential to render donor organs, especially the heart, unsuitable for transplantation, and in those organs that are transplanted, can lead to increased morbidity and mortality in the recipient (Cooper et al. 1989; Bittner et al. 1996b; Novitzky 1997a; Wilhelm et al. 2000b; Seguin et al. 2001; Hicks et al. 2006). At the same time, the process of organ transplantation exposes the donor organ to ischaemia-reperfusion injury. This injury causes cell damage, leading to dysfunction and, in some cases, cell death by both necrotic and apoptotic pathways (Bittner et al. 1996b; Allen et al. 2000; Fleischer et al. 2002; Avkiran 2003; Teshima et al. 2003; Hicks et al. 2006). As a result, ischaemia- reperfusion injury has an impact on donor organ quality and thereby directly influences outcomes in transplant recipients, with longer ischaemic times associated with increased ischaemia-reperfusion injury and increased recipient mortality risk (Fleischer et al. 2002; Hicks et al. 2006; Taylor et al. 2006; Taylor et al. 2008).

There are many factors that influence the quality of donor organs and the subsequent outcomes in those recipients who receive these organs. Therefore, treatments designed

- 3 - to improve organ quality should be aimed at multiple molecular and cellular targets at various stages of the transplantation process. Interventions that have been previously investigated include the use of hormonal treatments in the brain-dead donor (Novitzky et al. 1987b; Novitzky et al. 1988c; Wheeldon et al. 1995; Rosengard et al. 2002; Zaroff et al. 2002; Rosendale et al. 2003a; Hicks et al. 2006), the use of myocardial sodium hydrogen ion (Na+-H+) exchange inhibitors such as cariporide (Martin et al. 1998; Kim et al. 1998a; Kim et al. 1998b; Martin et al. 1999; Scheule et al. 2003; Ryan et al. 2003a; Martin et al. 2003b; Ryan et al. 2003b), and the use of glyceryl trinitrate (GTN) (Oz et al. 1993; Pinsky et al. 1994b; Bhabra et al. 1996; Du et al. 1998; Baxter et al. 1999a; Baxter et al. 1999b; Baxter et al. 2001; Maczewski et al. 2003; Gao et al. 2005). These interventions will be discussed further later in this Chapter.

1.2 A BRIEF HISTORY OF CARDIAC TRANSPLANTATION

1.2.1 Experimental Cardiac Transplantation

The first cardiac transplant was reported in 1905 by and Charles Guthrie at the University of Chicago, where the heart of a small dog was explanted and then transplanted into the neck of a larger dog by anastomosing the cut ends of the jugular vein and the carotid artery to the aorta, pulmonary artery, one of the vena cavae and a pulmonary vein (Carrel et al. 1905). This heterotopic transplant re-established active coronary circulation after an ischaemic time of 1 hour and 15 minutes, and the heart developed ventricular contractions at a rate of 88/min (normal being 100/min), acting as a non-auxiliary pump. Whilst the exact arrangement of the anastomotic configuration of Carrel and Guthrie’s heterotopic cardiac transplant was not described in their report (Carrel et al. 1905), the most likely possibility has been illustrated and outlined by Najarian and Simmons (Najarian et al. 1972; Griepp et al. 1984)

The next published report of experimental cardiac transplantation was made by Frank Mann and James Priestly at the Mayo Clinic in 1933 (Mann et al. 1933). Utilising a canine heterotopic cardiac transplantation model, they first identified the pathological changes of allograft rejection, in which they speculated that the “failure of the homotransplanted heart to survive is not due to the technique of transplantation but to

- 4 - some biologic factor which is probably identical to that which prevents survival of homotransplanted tissues and organs”.

In the mid 1940’s to early 1960’s, Vladimir Demikhov performed a large number of surgical transplantation experiments in canine models. However this work was not published in the West until 1962 due to the political constraints of the Iron Curtain. Demikhov’s reports included descriptions of multiple technical variations of heterotopic cardiac transplantation (including inguinal and 24 intrathoracic variants); the first heart transplant whereby the transplanted heart acted as an auxiliary pump in 1946; the first successful heart-lung transplant, also in 1946; the first successful orthotopic cardiac transplantation with the transplanted heart acting as the principal pump in 1951 (in an era pre-dating hypothermic preservation and cardiopulmonary bypass machines); “transplantation of the head”; “transplantation of halves of the body”; and “the surgical combination of two animals with the creation of a single circulation” (Demikhov 1962).

Meanwhile, in the Western world, the next report of cardiac transplantation after Mann et al was in 1951 by Marcus, Wong and Luisada (Marcus et al. 1953). Thereafter followed a steady stream of experimental reports throughout the 1950’s and into the 1960’s, which paralleled the advances that were being made in non-transplant cardiac surgery (described in further detail by Reitz (Reitz 2002) and DiBardino (DiBardino 1999)). Experimental orthotopic cardiac transplantation continued to develop with the adoption of techniques such as hypothermia (Neptune et al. 1953) and cardiopulmonary bypass (Webb et al. 1957). The first British contribution to cardiac transplantation was made by Cass and Brock who, utilising a canine transplantation model, introduced the innovative technique of combining pulmonary venous and vena caval anastomoses into two atrial anastomoses (Cass et al. 1959).

One of the major advances made in cardiac transplantation was reported in 1960 by the cardiac transplantation research program at Stanford University. Utilising a technique that involved suturing the atrial cuffs, aorta and pulmonary artery, and reported a series of eight canine orthotopic cardiac transplants in which five of the transplanted animals had survived 6 to 21 days (Lower et al. 1960). These animals behaved normally, eating, drinking and exercising, and died as a result of allograft rejection. Despite being described more than 45 years ago, the transplantation

- 5 - technique described by Lower and Shumway is still used today by many cardiac transplantation surgeons and researchers (DiBardino 1999; Baumgartner 2002).

1.2.2 Human Cardiac Transplantation

The first human heart transplant was a xenograft transplant from a chimpanzee in 1964 by at the University of Mississippi (Hardy et al. 1964). After reperfusion, the transplanted heart contracted steadily whilst on cardiopulmonary bypass. However, it could not cope with the large venous return in the human recipient and the patient died about 1.5 hours after discontinuation of bypass.

The first and arguably most famous human to human cardiac transplant was performed in Cape Town, South Africa by on 3 December 1967 in a 54 year old man with ischaemic cardiomyopathy utilising a non-beating heart donor (Barnard 1967). The donor heart came from a 24 year old woman who had sustained massive head injuries after being hit by a speeding car. Once the donor was pronounced brain dead, she was removed from an artificial respirator and once her heart stopped beating (i.e. cardiac death), she was placed on cardiopulmonary bypass and hypothermia induced. The heart was then explanted, cooled topically and transplanted orthotopically (utilising the Lower and Shumway technique) with a donor ischaemic time of about 30 minutes. The recipient died 18 days later from Pseudomonas pneumonia and the complete report of the transplant was published in record time on 30 December 1967.

Three days after Barnard’s transplant, the second human to human heart transplant was performed in Brooklyn, New York by Adrian Kantrowitz who transplanted the heart of a two day old anencephalic neonate into a 17 day old baby with Ebstein’s anomaly (Kantrowicz et al. 1968). However, the baby died within seven hours of the transplant due to metabolic and respiratory acidosis.

The next transplant performed by Barnard on 2 January 1968 was to become the first truly successful transplant, with the recipient being discharged from hospital and resuming some physical activity. This patient survived for approximately 20 months, dying later of chronic rejection (Reitz 2002).

- 6 - After ten years of transplantation research at Stanford University, Shumway and colleagues performed the fourth (and their first) human to human heart transplant on 6 January 1968 (Stinson et al. 1968). Despite the recipient succumbing to gram-negative sepsis by the 15th post-operative day, the Stanford group became the most active transplant group for the following 15 years, and was responsible for many important developments in clinical transplantation. These included the transvenous technique and the histologic grading system for serial monitoring for cardiac allograft rejection that was developed by an English Surgeon working in Stanford, Philip Caves and a pathologist, Margaret Billingham (Caves et al. 1974; Reitz 2002).

The first heart transplant in Australia was performed on 23 October 1968 by Harry Windsor at St. Vincent’s Hospital Sydney (Windsor 1969). The recipient was a 57 year old maintenance foreman on the Sydney Harbour Bridge with ischaemic cardiomyopathy and complete occlusion of all three major coronary arteries (Gibson et al. 1969). He was to survive for 45 days post-transplantation and died suddenly as a result of haemorrhage from an anterior 2 mm perforation in his native aorta secondary to infection, 2 mm from the aortic anastomosis (and the nearest needle perforation) (Windsor 1969).

With these early transplants in rapid succession, there was a worldwide explosion in the number of centres performing transplants, and the number of transplants performed. By the end of 1968, 102 transplants had been performed at 52 centres across 17 countries (Kirklin et al. 2002; Reitz 2002). However, results were generally poor with 60% mortality by the 8th post-operative day and a mean survival of 29 days (Kirklin et al. 2002). By the end of 1969 there were less than 50 transplants performed, followed by less than 20 transplants by the end of 1970 (Reitz 2002). During the 1970’s, fewer than 50 transplants were performed per year in only a few centres worldwide, with the majority being performed by the Stanford group (Reitz 2002). In this 10 year period, this group improved their one-year transplant survival rate from 22% to 65% (Griepp 1979). The second heart transplant in Australia was performed in 1974, again at St. Vincent’s Hospital Sydney, with the recipient surviving for 62 days before succumbing to infection. The St. Vincent’s cardiothoracic surgical unit subsequently voluntarily suspended their transplantation activity in 1975 (Chang 1984).

- 7 -

The 1980’s saw a re-ignition of cardiac transplantation activity. In 1980, fewer than 360 transplants were performed in 17 centres worldwide, but by 1993 there were 229 active transplant centres, performing 22 400 total heart transplants (in addition to 58 heart-lung transplant centres performing 1 349 procedures) (Kaye 1993; Lansman et al. 1995). This renewed enthusiasm in transplantation was principally due to the incorporation of the cyclosporin A into clinical practice, resulting in an exponential growth in the number of transplant centres around the world and the number of transplants performed (DiBardino 1999; Reitz 2002). Along with this growth, the survival rates of transplant recipients also improved over the course of the 1980’s (DiBardino 1999).

In Australia, cardiac transplantation was recommenced in 1984 with the establishment of the Australian Cardiac Transplantation Programme at St. Vincent’s Hospital Sydney, under the chairmanship of Victor Chang (Chang 1984). Since that time, a further five heart transplant programmes have been established across Australia and New Zealand: Auckland City Hospital (New Zealand) in 1987, a paediatric programme at The Royal Children’s Hospital Melbourne in 1988, The Alfred Hospital Melbourne in 1989, The Prince Charles Hospital Brisbane in 1990, and Royal Perth Hospital in 1995 (Keogh et al. 2008).

1.3 THE MODERN ERA OF CARDIAC TRANSPLANTATION IN AUSTRALASIA AND OVERSEAS

1.3.1 Indications For Cardiac Transplantation

It has been over 40 years since the first human to human cardiac transplant, and in that time cardiac transplantation has become an established and widely used therapeutic option for the treatment of severe end-stage heart failure that is refractory to all other medical and non-transplant surgical therapies (Bolling et al. 2001; Magliato et al. 2001; Kasper et al. 2002; Al-khaldi et al. 2006). It has become a therapeutic option that offers improvements in both survival and quality of life with 80-90% of cardiac recipients returning to New York Heart Association (NYHA) functional class 1 (Copeland 2001). It should also be noted that in recent years with the development and success of

- 8 - mechanical circulatory support devices, these devices have become a therapeutic option for destination therapy in those patients with severe end-stage heart failure who are not suitable candidates or marginal candidates for cardiac transplantation. These devices have also been used as a bridge to cardiac transplantation (whilst awaiting a suitable donor organ to become available) or as a bridge to recovery of the native heart.

The generally accepted indications and contraindications for cardiac transplantation are listed in Table 1.1. It should be noted though, that the setting of an upper age limit in potential recipients is a controversial issue and there is a growing body of evidence to support transplantation in carefully selected patients older than 60 years (Blanche et al. 2001; Kasper et al. 2002; Morgan et al. 2003; Al-khaldi et al. 2006). Data from the Organ Procurement and Transplantation Network (OPTN; optn.transplant.hrsa.gov) reported that the percentage of cardiac recipients 65 years old in the USA had increased from 1.4% in 1988 to 11.7% in 2008. Today, recipients aged 60 years and over account for nearly 25% of all transplants reported to the International Society for Heart and Lung Transplantation (ISHLT) Registry (Taylor et al. 2008). Despite this increase in transplantation in the older age group, it is still well recognised that increased recipient age is a risk factor for 1, 5 and 10 year mortality rates (Taylor et al. 2008). In addition, the liberalisation of upper age limits would also be expected to exacerbate the current donor organ supply shortage.

As the worldwide experience in cardiac transplantation has grown, there has also been a liberalisation of transplant recipient selection criteria, with more high-risk candidates being considered, who previously would not have been listed for transplantation (Al- khaldi et al. 2006). As mentioned earlier, older candidates are being transplanted, but also patients with conditions such as diabetes mellitus, amyloidosis, positive serology for human immunodeficiency virus (HIV), hepatitis B and C virus, fixed pulmonary hypertension, irreversible renal dysfunction, and irreversible liver dysfunction have been successfully transplanted (Al-khaldi et al. 2006; Macdonald 2008).

- 9 - Table 1.1: Indications and contraindications for cardiac transplantation (Kasper et al. 2002; Macdonald 2008).

Indications: End-stage heart disease not amenable to other medical or surgical therapy NHYA class III-IV symptoms on optimal medical therapy and prognosis for one-year survival less than 50% Intractable angina Repeated shocks from automatic implantable cardioverter defibrillator Age generally 60-65 years or younger Healthy apart from heart disease Emotionally stable, well motivated to resume active lifestyle Compliant with medical advice Supportive family/companions willing and able to make similar long-term commitments

Contraindications:

Psychiatric or cognitive illness likely to result in non-compliance with therapy

Continued abuse of alcohol, tobacco or other substances

Chronic systemic infection

Malignancy

Morbid obesity (body mass index >30 kg/m2)

Severe peripheral or cerebrovascular disease not amenable to revascularisation

Irreversible secondary organ failure unless considering combined organ transplant

Chronic neurodegenerative or musculoskeletal diseases that prevent rehabilitation

Fixed pulmonary hypertension

Diabetes mellitus with established microvascular target organ damage

Cardiac amyloidosis (most forms)

According to the latest ISHLT Registry report, the top two indications for cardiac transplantation worldwide over the past five years has shifted from an equal split between coronary heart failure (42% in 2006) and non-coronary cardiomyopathy (46% in 2006) to a significantly greater proportion of patients with non-coronary

- 10 - cardiomyopathy (50% in 2008) compared with coronary heart failure (34% in 2008) (Taylor et al. 2006; Taylor et al. 2008). Adult congenital heart disease (3%), re- transplantation (2%), and valvular heart disease (2%) account for many of the remaining transplants (Taylor et al. 2008). By comparison, in the Australia and New Zealand Cardiothoracic Organ Transplant Registry (ANZCOTR) 2007 Report, 37% of cardiac transplant recipients had ischaemic heart disease and 40% idiopathic cardiomyopathy, with the remainder having congenital heart disease (5%), viral cardiomyopathy/myocarditis (3%), familial cardiomyopathy (2%), peripartum cardiomyopathy (1%) and other causes (12%; e.g. valvular heart disease, adriamycin toxicity) (Keogh et al. 2008). With respect to infants, the most common indication for transplantation has been congenital heart disease (63%) followed by cardiomyopathy (30%), whereas in older children this is reversed with cardiomyopathy the reason for transplantation in 57% and congenital heart disease in 31% (Kirk et al. 2008).

1.3.2 Outcomes Following Cardiac Transplantation

In the latest ISHLT Registry Report, over 80 000 heart transplants performed worldwide have been analysed (Taylor et al. 2008). The Report found that between January 1982 and June 2006, the combined group of adult and paediatric cardiac transplant recipients had a graft half-life (i.e. the time at which 50% of those transplanted were still alive, or median survival) of 10 years, and of those who survived the first year, the graft half-life was 13 years. After an initial steep decline in survival in the first six months post- transplantation, survival declined at a relatively linear rate of approximately 3.5% per year, even beyond 15 years post-transplant (Figure 1.1). At the same time, transplant recipient survival continued to improve with each successive 5 to 10 year era, with the major survival gains occurring in the first 6 to 12 months post-transplantation. In the most recent era (2002-6/2006), the projected graft half-life is approximately 11 years, whereas in 1992-2001 it was 10.5 years and in 1982-1991 it was 8.8 years (Taylor et al. 2008).

In Australia and New Zealand, graft half-life was approximately 14 years for all heart transplants performed between 1984 to 2007, with a one-year survival of 86.9%, followed by 5 and 10 year survival of 77.1% and 61.0% respectively (Keogh et al.

- 11 - 2008). Comparison of the Australian/New Zealand survival rates with international (ISHLT Registry) data is shown in Figure 1.2.

Primary, non-specific graft failure accounts for the majority of deaths (42%) within the first 30 days after transplantation, followed by multi-organ failure (12%) and non- cytomegalovirus (CMV) infection (13%). By 31 to 365 days post-transplant, the distribution of cause of death changes, with non-CMV infection (33%) becoming the primary cause of death, followed by primary, non-specific graft failure (18%) and acute rejection (12%). After five years the distribution changes again, with cardiac allograft vasculopathy (CAV) and late graft failure (likely due to CAV) causing 33% of deaths, followed by malignancies (including lymphoma) (23%) and non-CMV infections (11%) (Taylor et al. 2008). Figure 1.3 depicts the relative incidence of the leading causes of death for post-transplant deaths reported from January 1992 through June 2008.

In Australia and New Zealand, infection (21%) is the most common cause of death in recipients less than five years post-transplantation, followed by rejection (18%), non- specific graft failure (14%) and CAV (13%). For deaths greater than five years post- transplant, CAV accounts for 24% of deaths, followed by non-lymphoid malignancy (21%), lymphoproliferative disease (6%), infection (6%) and rejection (5%) (Keogh et al. 2008).

- 12 - Figure 1.1: International Society for Heart and Lung Transplantation Registry Kaplan-Meier survival data for adult and paediatric heart transplants performed between January 1982 and June 2006 (with 95% confidence limits). Conditional half-life = time to 50% survival for those recipients surviving the first year post- transplantation (Taylor et al. 2008).

ANZCOTR

ISHLTR

Figure 1.2: Actuarial survival for all heart transplants performed in Australia and New Zealand (from the Australia and New Zealand Cardiothoracic Organ Transplant Registry – ANZCOTR) from February 1984 to December 2007 (Keogh et al. 2008) compared with the International Society for Heart and Lung Transplantation Registry (ISHLTR) survival data for all heart transplants performed from January 1982 to June 2006 with 95% confidence limits (from Figure 1.1; Taylor et al. 2008).

- 13 -

Figure 1.3: Relative incidence of the leading causes of death for post-transplant deaths reported from January 1992 through June 2008. CAV, coronary allograft vasculopathy; lymph/PTLD, lymphoma or post-transplant lymphoproliferative disease; CMV, cytomegalovirus (Taylor et al. 2009).

In terms of functional status, according to the ISHLT Registry data (1995-June 2006), 89.7-91.8% of cardiac transplant recipients reported no limitations in activity at one, three, five and seven years post-transplantation, with only 0.5-0.8% requiring total assistance with activities of daily living in the corresponding period. At the same time, the re-hospitalisation rate in the first year post-transplant was 44.8% but improved to 22.2% between the second and third years, 22.3% between fourth and fifth years, and 23.8% between the sixth and seventh years. However, despite the improved quality of life, only 30.4-34.7% of recipients were working part- or full-time at one, three, five and seven years post-transplant, with 19.2-30.5% retiring from work. (ISHLT Registry 2008: http://www.ishlt.org/registries/slides.asp?slides=heartLungRegistry).

1.3.3 Determinants Of Outcome In Cardiac Transplantation

The ISHLT registry has identified a large number of risk factors for mortality at 1, 5 and 10 years post-transplantation (Taylor et al. 2008). Of note is that mortality in the first year is greater than that of the next four years combined and therefore the one-year mortality risk factors are also powerful predictors of outcome at five years. In an effort to adjust for the high first year mortality, the ISHLT Registry identified five-year

- 14 - “conditional” survival risk factors (i.e. risk factors in those who survived the first year post-transplant). With respect to the one-year risk factors, the most significant risk factors were temporary mechanical circulatory support and having adult congenital heart disease as the indication for transplantation, although these patients represent less than 3% of the total transplant recipients each. Other one-year risk factors included: dialysis or mechanical ventilation at time of transplant, infection requiring intravenous antibiotics within two weeks prior to transplant, compatible but non-identical ABO matching and prior blood transfusions. In terms of significant risk factors imparting a continuous rather than categorical impact on mortality, these have remained relatively unchanged from previous years. These include recipient age, donor age, transplant centre volume (inverse relationship), donor heart ischaemic time, donor body mass index (BMI) (inverse relationship), pre-transplant recipient bilirubin and creatinine, and recipient pulmonary artery diastolic pressure (Taylor et al. 2008). Table 1.2 outlines the risk factors for one-year mortality after heart transplantation based on ISHLT Registry data.

Table 1.2: Risk factors for mortality within one year after heart transplantation for the era January 2002 to June 2007 in 10 705 recipients. Table includes the relative risk for mortality for each independent variable. Donor and recipient age, and donor heart ischaemic time continue to be powerful risk factors for death at one year post transplant in the more recent era (Taylor et al. 2009).

- 15 - In terms of the five-year “conditional” risk factors for mortality, these include: ventilatory support at time of transplant; CAV identified in the first post-transplant year; diagnosis other than cardiomyopathy, valvular or coronary artery disease; pre-transplant history of diabetes; and treated rejection episodes. Some of the continuous five-year risk factors are similar to the one-year factors, such as donor and recipient age, pulmonary vascular resistance and donor BMI. At 10 years, some of the mortality risk factors at one- and five-years remained and included repeat transplantation, recipients on ventilators or ventricular assist devices or inotropes pre-transplant, HLA mismatches, pre-transplant congenital heart disease or coronary artery disease, recipient and donor age, donor heart ischaemic time and recipient BMI (Taylor et al. 2008).

More detailed analyses of the impact of various risk factors on post-transplantation outcomes have been performed based on data from the Cardiac Transplant Research Database (CTRD) and has been reported by Kirklin and colleagues (Kirklin et al. 2002; Kirklin et al. 2003; Kirklin et al. 2004). An example of the analysis of the CTRD illustrating the impact of donor age and recipient age on CAV-related deaths post- transplantation is shown in Figure 1.4.

Figure 1.4: An analysis of the Cardiac Transplant Research Database (CTRD) of post- transplant outcomes for 7 283 heart transplants performed between 1990 and 1999 illustrates the impact of donor age and recipient age on death due to coronary artery disease in the transplanted heart beyond the first year post transplant (Kirklin et al. 2004).

- 16 -

1.3.4 Demographics Of Cardiac Transplantation And Organ Donation

There were 3 205 heart transplants reported to the ISHLT registry in 2006, which has increased slightly in the last two years having slowly declined in the preceding 10 years after a peak of 4 429 in 1994 (Figure 1.5) (Taylor et al. 2008). The previous downward trend was due to both decreased reporting and decreased numbers of transplants performed (Taylor et al. 2006). When other transplant registries that do not report to the ISHLT Registry were analysed, it was estimated that more than 2 000 heart transplants were performed annually which were not reported to the Registry (Taylor et al. 2004). This would therefore put the annual worldwide number of heart transplants at over 5 000.

Figure 1.5: Number of heart transplants reported to the ISHLT Transplant Registry by year (1982-2006). It should be noted that this figure includes only the heart transplants that are reported to the Registry and hence does not represent the worldwide total number of transplants performed (Taylor et al. 2008).

In Australia and New Zealand, there were 68 heart transplants reported to the ANZCOTR in 2007 (80 in 2006, 89 in 2005) (Keogh et al. 2008). Whilst there was not the same consistent decline in reported transplants as reported to the ISHLT registry,

- 17 - there have been fewer transplants in recent years compared with much of the 1990’s (Figure 1.6).

Figure 1.6: Number of heart and heart-lung transplants reported to the Australia and New Zealand Cardiothoracic Organ Transplant Registry by year (February 1984 to December 2007) (Keogh et al. 2008).

According to data from the OPTN website, the number of heart transplants performed in the USA from 1995 to 2008 has ranged from 2015 in 2004 to 2 363 in 1995, and as of 15 May 2009, there were approximately 2 785 patients on the cardiac transplant waiting list (and 87 patients on the heart/lung transplant waiting list). In 2008, there were 7 984 deceased donors in the USA, from whom 2 225 hearts were recovered and 2 163 transplanted. In Australia and New Zealand, according to data from the Australian and New Zealand Organ Donation Registry (ANZOD; www.anzdata.org.au/anzod/v1/indexanzod.html), there were 236 brain-dead donors in 2007. Of these, there were 181 eligible cardiac donors (i.e. those in whom hearts were requested and who had consented), of whom only 87 were retrieved and 74 transplanted (including six heart-lung transplants) (Excell et al. 2008). Clearly there is a significant proportion of donor hearts that are not recovered and later transplanted, often due to poor organ function and an unwillingness to use sub-optimal donors (Rosendale et al. 2002; Zaroff et al. 2002; Hicks et al. 2006).

In terms of the waiting list in Australia and New Zealand, there were 45 patients on the active waiting list as of 1 January 2007, with a further 109 patients added during the

- 18 - year. The mean waiting time for a heart transplant in 2007 was 163±180 days and the median waiting time was 125 days (range: 2-974). These waiting times have increased from the downward trend between 1999 and 2004 (mean waiting time = 236±260 days in 1999 to 134±176 days in 2004) (Keogh et al. 2008).

At the same time, the transplant waiting list mortality rate has been reported to be as high as 17-22% per year (Copeland 2001; Zaroff et al. 2002). In Australia and New Zealand, 10 patients died on the waiting list in 2007, equating to about 6.5% of all patients who had been on the transplant waiting list in 2007 (Keogh et al. 2008).

Whilst Australia and New Zealand have a significantly smaller population compared with countries such as the USA, the organ donation rate per million population (dpmp) is very poor (9.0 dpmp in Australia and 9.0 dpmp in New Zealand in 2007) compared with the rest of the world. For example, Spain has a donation rate of 34.3 dpmp, USA 24.6 dpmp and Canada 14.8 dpmp (Figure 1.7) (Excell et al. 2008).

Figure 1.7: International donor statistics for 2007. Donors per million population in Australia and New Zealand compares very unfavourably with the rest of the world. Source: International Registry for Organ Donation and Transplantation (Excell et al., 2008).

- 19 - 1.3.5 The Problem Of Supply Versus Demand

Heart failure is a rapidly growing health problem in many countries and it has been estimated that between 25 000 to 65 000 patients in the USA could benefit from a cardiac transplant (Copeland 2001; Zaroff et al. 2002). In Australia, there are no national data on the incidence of heart failure. However, approximately 263 000 Australians (1.3% of the population) self reported heart failure or symptoms of heart failure in the 2004-05 National Health Survey and it has been estimated that 30 000 new cases are diagnosed per year (AIHW 2008). Heart failure accounted for 2 225 Australian deaths in 2005 (AIHW 2008), and given the numbers of Australians with heart failure, it is likely that a significant number of these patients would benefit from a heart transplant.

However, as described above and in Section 1.3.4, there is a significant disparity between the number of patients who might benefit from a cardiac transplant, those on the waiting list for a transplant and the number of transplants actually performed (i.e. the demand for transplantation outweighs the supply of suitable organs) (Copeland 2001; Conte et al. 2002; Kasper et al. 2002; Rosendale et al. 2002; Zaroff et al. 2002). There are many reasons for this disparity. These include the lack of donors and the lack of suitable organs from the currently available donor pool, with organs being rejected for transplantation due to factors such as advanced donor age, pre-existing disease in the donor such as heart disease, trauma, infection, cardiac arrest, poor organ function and the damaging effects of brain death on the donor. In addition, the liberalisation of transplant recipient criteria (as discussed in Section 1.3.1) has also increased the number of potential recipients on the waiting list.

In an effort to reduce the disparity between the demand for transplantation and the supply of suitable organs, there has been an increasing use of marginal donors, otherwise known as extended criteria organ donors (e.g. older donors or donors with evidence of chronic organ disease or dysfunction prior to brain death) (Vigneswaran et al. 1993; Zembala et al. 1996; Mullen et al. 2001; Rosendale et al. 2002; Lima et al. 2006). Whilst there has been an increase in the potential donor pool with the use of these marginal donors, the discard rate of cadaveric organs offered for transplantation has also increased, with more than 15% of kidneys, 25% of livers and 60% of hearts and

- 20 - lungs discarded due to poor donor organ quality (Rosendale et al. 2002; Rosengard et al. 2002; Zaroff et al. 2002). Therefore, whilst marginal donors have provided some additional organs for transplantation, the use of marginal donors alone is insufficient and other strategies must be developed to help bridge the gap between supply and demand. These include programmes to increase the numbers of donors within the population, and treatments to optimise the organ quality and hence utilisation rate from the currently available pool of donors.

1.4 THE BRAIN-DEAD ORGAN DONOR

1.4.1 Defining Brain Death In The Organ Donor And Its Diagnosis

Although the first human to human cardiac transplant utilised a brain-dead donor after cardiac death (i.e. donation after cardiac death, DCD) (Barnard 1967), the majority of transplants are performed using brain-dead beating heart donors. More recently however, there have been moves to increase the numbers of transplantable organs by utilising non-heart beating donors (i.e. DCD) for organs such as kidneys (Fung 2000; Gerstenkorn et al. 2000; Light et al. 2000), livers (Reich et al. 2000; Manzarbeitia et al. 2004), pancreatic islets (Clayton et al. 2000), lungs (Steen et al. 2001; Corris 2002; Gamez et al. 2005) and hearts (Martin et al. 1999; Martin et al. 2003a; Singhal et al. 2005).

The concept of brain death was first introduced in 1965, and was later formally defined in 1968 by the Ad Hoc Committee of the Harvard Medical School as unresponsiveness and lack of receptivity, the absence of movement and breathing, the absence of brain- stem reflexes and coma whose cause has been identified (Anonymous 1968; Randell 2004). Since that time, there have been numerous descriptions and statements regarding the definition and diagnosis of brain death, including an evidence-based review and report from the American Academy of Neurology (Wijdicks 2001).

The determination of brain death is based primarily on the clinical neurological examination (supplemented with neuroimaging and confirmatory laboratory tests as necessary) and requires the absence of brain-stem reflexes, motor response, and respiratory drive in a normothermic, non-drugged, comatose patient with a known

- 21 - irreversible brain lesion and no contributing metabolic derangements. Confirmatory investigations may also be used in determining brain death such as cerebral angiography, electroencephalography, transcranial doppler ultrasonography and cerebral scintigraphy (Wijdicks 2001; ANZICS 2008).

Legislation defining and determining death (whether it be the irreversible cessation of all brain function - brain death- or the irreversible cessation of blood circulation and heart beat - cardiac death), particularly in relation to transplantation, varies from country to country, and even from state to state within countries. Whilst most countries provide a legislative framework and practice guidelines with respect to brain death and transplantation, there is considerable variability with the procedures in the diagnosis of brain death, such as the number of physicians required and their level of expertise, the time required for observation, and the use of confirmatory tests (Wijdicks 2002; Randell 2004). In the USA, the American Academy of Neurology has published practice guidelines and the Uniform Determination of Death Act has been accepted by most states. However, whilst most states have comparable statutes, there are still many notable differences (Wijdicks 2002). In Australia and New Zealand, the Australian and New Zealand Intensive Care Society (ANZICS) has published a statement on death and organ donation (ANZICS 2008). The Statement provides guidelines for the legal determination of brain death by clinical examination in adults, infants and children, along with imaging guidelines to demonstrate the absence of intracranial blood flow. In order to ascertain brain death, clinical testing of brain-stem function must be carried out by two medical practitioners with specific experience and qualifications. In addition, the patient must be observed for a minimum of four hours in order to confirm brain death. In the United Kingdom, brain death is defined by brain-stem death – an irreversible loss of the capacity of consciousness combined with an irreversible loss of the capacity to breathe (even in the presence of cerebral blood flow). Whilst this definition is used in some Commonwealth countries, it is not used in Australia and New Zealand, whereby whole brain death is required for the legal determination of death (Wijdicks 2002; ANZICS 2008).

- 22 - 1.4.2 The Causes Of Brain Death

In most situations, brain death is caused by a sudden or gradual increase in intracranial pressure (ICP), which causes a progressive mass effect in the brain and cerebral ischaemia. This in turn leads to venous engorgement and brain swelling, thereby forcing the brain stem through the foramen magnum, causing arterial compression and further cerebral ischaemia and infarction. This further exacerbates brain swelling, increasing ICP until the pressure exceeds the arterial blood pressure and intracranial circulation is stopped (Smith 2004).

In general, the most common causes of brain death in adults are traumatic brain injury and subarachnoid haemorrhage, whereas in children, abuse is a more common cause than motor vehicle accidents or asphyxia (Wijdicks 2001). In Australia and New Zealand, over the period from 2001 to 2007, cerebrovascular accidents (CVA; including cerebral infarction, intracerebral haemorrhage and subarachnoid haemorrhage) were the most common cause of death amongst donors – 53% in Australia and 56% in New Zealand (Excell et al. 2008). This was followed by road trauma - 18% in Australia and 21% in New Zealand; non-road trauma (falls, assaults, gunshots and accidents) - 10% in Australia and 12% in New Zealand; and hypoxia-anoxia (such as cardiac arrest, asthma, drowning and hanging) - 15% in Australia and 6% in New Zealand. Road trauma has become a declining cause of death amongst donors, whereas the incidence CVA has been steadily increasing in Australia but remained steady in New Zealand.

1.4.3 The Negative Effects Of Brain Death On Transplantation Outcomes

The pathophysiological changes of brain death in the organ donor lead to haemodynamic instability and organ dysfunction (Mertes et al. 1994b; Novitzky 1997a; Pratschke et al. 1999; Seguin et al. 2001; Smith 2004). This contributes to the rejection of organs for transplantation and to post-transplant complications (Pennefather et al. 1995; Pratschke et al. 1999; Smith 2004).

In the early 1980’s, whilst developing a continuous, hypothermic perfusion storage system for the donor heart, Cooper and colleagues observed that hearts transplanted

- 23 - from healthy anaesthetised baboons had far superior function to those taken from brain- dead pigs and humans (Wicomb et al. 1984; Cooper et al. 1989). They concluded that a loss of myocardial energy stores and/or other reversible damage sustained during and after brain death caused the reduction in graft function. It has also been observed that outcomes in cadaveric organ transplantation, such as , are significantly inferior to transplants utilising organs from a living donor, whether they be related or unrelated to the recipient (Terasaki et al. 1995; Pratschke et al. 1999). Since that time, it has become clear that brain death is a significant risk factor for both short- and long-term outcomes and is dependent upon donor characteristics, as well as the aetiology of the central injury leading to brain death (Pratschke et al. 1999; Gasser et al. 2000).

Brain death in the organ donor results in a series of complex pathophysiological changes in the donor and alters their physiological, cellular and biochemical functions, leading to organ dysfunction (Tuttle-Newhall et al. 2003). These changes also activate the inflammatory and immune systems, which in turn, increase the risk of acute rejection by stimulating allogeneic recognition and response (Wilhelm et al. 2000a; Gasser et al. 2001; Tuttle-Newhall et al. 2003). However, not all of these pathophysiological changes are seen in every potential donor. The incidence, time course and severity of these changes vary according to the tempo and aetiology of the neurological insult leading to brain death, and increases with time after the onset of brain death (Smith 2004). Therefore, a thorough understanding of the pathophysiological changes that occur in brain death is needed in order to devise management strategies to prevent and/or reverse the deleterious effects of brain death on organ function and thus improve transplant outcomes.

1.4.4 The Cardiovascular, Neurohormonal And Immunological Consequences Of Brain Death

1.4.4.1 The Autonomic Storm And Cardiovascular Changes In Brain Death

The effects of brain death on the haemodynamic system can be divided into two phases: an early hypertensive phase that occurs at the onset of brain death, followed by a later hypotensive (or occasionally normotensive) phase. The sequence of events occurs as

- 24 - different areas of the brain stem become ischaemic. Initial pontine ischaemia causes mixed vagal and sympathetic stimulation, resulting in the classical Cushing Response characterised by bradycardia, hypertension and irregular breathing patterns (Cushing 1902; Smith 2004). Later, as ischaemia progresses through the brain and brain stem affecting centres such as the vagal and cardiomotor nuclei, there is massive unopposed sympathetic outflow. These changes have been termed the “autonomic storm”, but have also been referred to as the “sympathetic” or “catecholamine” storm (Cooper et al. 1989; Pratschke et al. 1999; Smith 2004). This storm is characterised by a massive increase in circulating catecholamine levels such as adrenaline, noradrenaline and dopamine (Novitzky et al. 1984; Cooper et al. 1989; Powner et al. 1992; Mertes et al. 1994a; Bittner et al. 1995a; Bittner et al. 1995b; Chen et al. 1996; Chiari et al. 2000). The magnitude of increase in catecholamine levels has varied between reports (and different animal models and human studies), but there have been reported increases as high as 1100% in adrenaline (Novitzky et al. 1984), 800% in dopamine (Chen et al. 1996) and 310% in noradrenaline (Chiari et al. 2000). Organs are therefore exposed to intense sympathetic stimulation, either from direct neural activity or from elevated endogenous catecholamines.

The autonomic storm also alters organ perfusion due to intense vasoconstriction and elevated vascular resistance. This substantially decreases blood flow to the organs, despite increased perfusion pressure from the sympathetic-stimulated hypertension and tachycardia, thereby causing tissue ischaemia. The increased sympathetic stimulation and peripheral resistance also contribute to a sudden and massive increase in myocardial work and oxygen consumption, which is further complicated by coronary vasoconstriction (again from sympathetic stimulation), leading to sub-endocardial ischaemia (Pratschke et al. 1999; Seguin et al. 2001; Smith 2004). As a result, cardiac function is impaired by this brain death-induced ischaemia.

Further cardiac dysfunction is caused by direct injury from the increased catecholamines, of which the sympathetic nerves supplying the myocardium are a major source of pathological catecholamine exposure (Mertes et al. 1994a). This catecholamine myocytotoxicity results in myocardial structural damage in the form of myocytolysis, contraction band necrosis, subendocardial haemorrhage, oedema and interstitial mononuclear cell infiltration. There is also damage to the coronary arteries in

- 25 - many donors, often with no pre-existing heart disease (Okereke et al. 1987; Novitzky et al. 1988f; Cooper et al. 1989; Shivalkar et al. 1993; Baroldi et al. 1997; Pratschke et al. 1999; Smith 2004).

The autonomic storm is also associated with sudden and large increases in intracellular calcium, leading to the activation of various enzymes such as lipases, proteases, endonucleases, xanthine oxidase and nitric oxide synthase (Novitzky 1997a). These enzymes then cause impaired adenosine 5’-triphosphate (ATP) production and also increase the production of oxygen free radicals, which cause further cell damage and hence organ dysfunction (Novitzky et al. 1987a; Pratschke et al. 1999; Smith 2004). Furthermore, there have been reports of altered myocardial gene expression in animal models of brain death (Yeh et al. 1999), which may also contribute to the myocardial dysfunction following brain death.

Following the massive sympathetic discharge of the autonomic storm, there is a profound reduction in sympathetic outflow (Gramm et al. 1992b), leading to a loss of sympathetic tone and a massive reduction in systemic vascular resistance due to vasodilatation (Smith 2004) and hence can cause significant hypotension. This results in impaired vascular autoregulation with diminished blood and oxygen delivery to organs (Pratschke et al. 1999). Furthermore, there is reduced cardiac output due to the impaired inotropic and chronotropic state of the heart (Pratschke et al. 1999), decreased preload due to peripheral vasodilatation, and a loss of intravascular volume due to complications such as diabetes insipidus (discussed later) or haemorrhage (from the initial event leading to brain death). All of these contribute to the severe haemodynamic instability, hypoperfusion and ischaemic injury to the organs seen in brain death.

As a result of these autonomic and cardiovascular changes, brain-dead donors require proactive management to support their haemodynamics, tissue perfusion and organ function. Failure to do so often leads to cardiac arrest in a relatively short time and a subsequent loss of the donor organs.

- 26 - 1.4.4.2 Neurohormonal Changes In Brain Death Additional consequences of brain death are the endocrine changes that lead to a loss of the neurohormonal support mechanisms and impaired metabolism in organs such as the heart (Novitzky et al. 1987a; Cooper et al. 1989; Novitzky et al. 1989a; Mertes et al. 1994b; Bittner et al. 1995b; Chen et al. 1996; Novitzky 1997a; Pratschke et al. 1999; Wilhelm et al. 2000b). These changes can be divided into two categories: those associated with the autonomic storm and the subsequent changes in circulating catecholamines (as discussed previously), and those associated with the hypothalamus and pituitary gland (and hence the thyroid and adrenal glands). However, whilst it has been generally accepted that brain death causes a disruption of the hypothalamic- pituitary axis and that there is failure of the anterior and posterior pituitary, there have been many controversial and conflicting results regarding the endocrine changes associated with brain death and their influence on haemodynamic parameters and organ quality (Chen et al. 1996; Pratschke et al. 1999; Smith 2004; Hicks et al. 2006). There have been differences in endocrine studies in both animal models of brain death (Novitzky et al. 1984; Cooper et al. 1989; Bittner et al. 1995b) and in human subjects (Howlett et al. 1989; Powner et al. 1990; Powner et al. 1992; Gramm et al. 1992a; Mariot et al. 1995), with much of the evidence supporting hormonal depletion more convincing in animal models than in humans.

In the original experimental studies by Novitzky, Wicomb, Cooper and colleagues into the physiological changes occurring in brain death, an animal (Chacma baboon) model of experimental brain death was used to investigate changes in the electrocardiogram (ECG), haemodynamic parameters, blood chemistry and hormone levels (Novitzky et al. 1984). In a series of ten baboons, animals were observed prior to brain death, and then during and after brain death. Brain death was induced by the inflation of a subdural Foley catheter balloon and animals were monitored for 24 hours after brain death induction. This study demonstrated impressive initial increases in catecholamine levels (described above), which then fell to subnormal levels, along with the decline to undetectable levels of thyroid hormones (triiodothyronine – T3 and thyroxine – T4), as well as decreased levels of cortisol, insulin and antidiuretic hormone (ADH; also known as vasopressin). There were however, no changes detected in thyroid stimulating hormone (TSH), glucagon or ionised calcium. Novitzky, Cooper and colleagues went

- 27 - on to produce an impressive body of work investigating the endocrine, physiological, functional and morphological changes associated with brain death (Novitzky et al. 1986; Novitzky et al. 1987a; Novitzky et al. 1987b; Cooper et al. 1988a; Novitzky et al. 1988a; Cooper et al. 1988b; Novitzky et al. 1988b; Novitzky et al. 1988c; Novitzky et al. 1988e; Cooper et al. 1989; Novitzky et al. 1989a; Wicomb et al. 1989b; Novitzky et al. 1990; Novitzky 1991; Novitzky 1996a; Novitzky 1997a). In a porcine model of brain death, they replicated their findings of decreased plasma T3 levels, although cortisol and insulin levels remained unchanged (Cooper et al. 1989).

Diabetes insipidus is present in up to 87% of brain-dead patients and is due to a deficiency in ADH caused by ischaemic damage to the posterior pituitary, a phenomenon that has been consistently demonstrated in numerous animal and human studies (Howlett et al. 1989; Sazontseva et al. 1991; Gramm et al. 1992a; Bittner et al. 1995a; Bittner et al. 1995b; Chen et al. 1996). This causes inappropriate diuresis leading to excessive free water loss, and results in severe hypovolaemia, hyperosmolarity, haemodynamic instability and electrolyte abnormalities (e.g. hypernatraemia, hypokalaemia, hypocalcaemia, hypophosphataemia and hypomagnesaemia) (Smith 2004; Wood et al. 2004). These changes can also lead to hypoperfusion of the lungs by exacerbating neurogenic pulmonary oedema and the hypernatraemia seen in these patients can often be worsened by the infusion of sodium containing fluids during volume resuscitation of the donor (Smith 2004).

Hormonal disturbances involving thyroid hormones have also been of major interest, particularly because of their role in cell metabolism and mitochondrial function (Novitzky et al. 1987b; Novitzky et al. 1988c; Novitzky et al. 1989b). Rapid declines in free T3 have been reported almost universally in baboon (Novitzky et al. 1984), porcine (Wicomb et al. 1986a; Novitzky et al. 1987d), canine (Bittner et al. 1995a) and human studies (Howlett et al. 1989; Powner et al. 1990; Gramm et al. 1992a; Mariot et al. 1995). However, changes in T4 and TSH have been more variable with animal studies commonly reporting declines in T4 (Novitzky et al. 1984; Wicomb et al. 1986a; Novitzky et al. 1987d; Bittner et al. 1995a), whereas in humans there have been no changes in T4 in some reports (Gramm et al. 1992a) or falls in only 29-38% of patients (Howlett et al. 1989; Mariot et al. 1995). TSH levels have also been variable with some human studies reporting increases from baseline values (Gramm et al. 1992a) and others

- 28 - reporting decreases (Howlett et al. 1989; Sazontseva et al. 1991; Mariot et al. 1995). In a study of 31 brain-dead donors, Howlett et al found that T3 levels were below normal levels in 81%, serum free T4 index was subnormal in 29% and TSH was subnormal in 23% (Howlett et al. 1989). None of the donors was found to have below normal T4 and TSH, which would be consistent with sick euthyroid syndrome rather than TSH deficiency. Gramm et al examined 32 organ donors and found that total T3 was subnormal or immeasurable in 62% and free T3 was subnormal or immeasurable in 76% (Gramm et al. 1992a). This study also demonstrated normal free T4 levels in 91% of donors and within the first 24 hours of brain death, TSH levels were normal in 52% of donors (although there was an increase from baseline).

In a canine model of the brain-dead donor, cortisol levels have been shown to decline significantly after the induction of brain death (Bittner et al. 1995a). This was associated with a significant decrease in adrenocorticotrophic hormone (ACTH) levels after 45 minutes to zero towards the end of the experiment. In contrast, most human studies have shown no changes in ACTH levels (Howlett et al. 1989; Sazontseva et al. 1991; Gramm et al. 1992a). Cortisol levels were also reported to decline significantly in Novitzky’s baboon model (Novitzky et al. 1984). In humans however, there have been mixed reports of no change in cortisol levels in brain death (Keogh et al. 1988; Robertson et al. 1989; Powner et al. 1990; Gramm et al. 1992a), decreases in only a small proportion of patients (Howlett et al. 1989; Mariot et al. 1995) or significant decreases (Taniguchi et al. 1992; Dimopoulou et al. 2003). Gramm et al demonstrated cortisol and ACTH levels within normal ranges in 32 potential brain-dead organ donors (Gramm et al. 1992a). In a study of 31 consecutive brain stem-dead donors, only one of 21 donors not receiving corticosteroids had undetectable cortisol levels to indicate complete ACTH deficiency (Howlett et al. 1989). In five of these donors, cortisol levels were elevated in the range of 710-880 nmol/L, which would be compatible with stress- induced elevations. However, there were 13 donors who had cortisol in the normal resting range, which would be lower than expected in the context of a “stressful” illness of brain death. Despite these reports, it has been suggested that cortisol levels are depressed in the brain-dead human donor, which may impair the donor stress response and contribute to cardiovascular instability, and that treatment of the donor with exogenous corticosteroids can stabilise organ function (Smith 2004).

- 29 - Variable changes in insulin levels after brain death have been reported in both animal models and in humans. In a baboon study, insulin levels fell within the first five minutes of brain death induction and continued to fall to less than half baseline levels (Novitzky et al. 1984). A reduction in insulin levels was also reported in a brain-dead canine model (Finkelstein et al. 1987). In contrast, Bittner and colleagues demonstrated moderate increases in insulin levels and a significant decrease in glucagon levels in a study of 10 brain-dead canines (Bittner et al. 1995a). Most human studies have demonstrated increases in insulin levels. A prospective study of 25 consecutive brain- dead patients in France demonstrated elevated insulin and C-peptide levels during donor management, which were associated with hyperglycaemia (Masson et al. 1993). Similarly, in a study of 14 human heart-beating brain-dead donors compared with eight normal controls, insulin and C-peptide levels were also elevated in brain-dead donors in association with hyperglycaemia, with glucagon and pancreatic polypeptide levels unchanged (Brunicardi et al. 2000). In another study of 16 brain-dead and 14 severely brain-injured patients, insulin and cortisol levels were either normal or elevated (Powner et al. 1990), and in a Japanese study of 16 patients, insulin levels were significantly higher just prior to brain death, and then dropped to normal levels after brain death (Taniguchi et al. 1992).

Equally variable are reports of changes in other hormone levels such as growth hormone, prolactin, follicle stimulating hormone, luteinizing hormone (Howlett et al. 1989; Powner et al. 1990; Sazontseva et al. 1991; Sugimoto et al. 1992; Gramm et al. 1992a; Arita et al. 1993). The correlations between hormonal levels, and metabolic and haemodynamic parameters as well as post-transplant allograft function are also diverse (Pratschke et al. 1999). Given the diversity of the histological changes in the pituitary gland and the changes in hormone levels, it is likely that there is still partial cerebral flow to some areas in selected brain-dead patients and that there is partial preservation of the hypothalamic-pituitary axis with suppressed hypothalamic function (Fiser et al. 1987; Sugimoto et al. 1992; Arita et al. 1993; Pratschke et al. 1999; Gasser et al. 2001). Differences in hormonal changes in experimental and clinical reports may also be due to the fact that experimental models have a highly reproducible, uniform and consistent mechanism of brain death, as opposed to the highly variable mechanisms of brain death in donors in whom both the mechanism and tempo of brain death is heterogeneous.

- 30 - 1.4.4.3 Immunological And Inflammatory Changes In Brain Death It has become increasingly recognised that brain death and excessive catecholamine release results in an activation of endothelial cells, platelets and leukocytes, which in turn activate the immune system and up-regulate pro-inflammatory cytokines within peripheral organs and the blood. These changes may contribute to organ dysfunction and have been hypothesised to make organs more susceptible to post-transplantation host inflammatory and immunological responses by stimulating the recipient’s immune system, thereby accelerating acute allograft rejection, as well as affecting graft function over the short to long term (Follette et al. 1998; Takada et al. 1998; Pratschke et al. 1999; Pratschke et al. 2000; Wilhelm et al. 2000a; Birks et al. 2001; Pratschke et al. 2001a; Pratschke et al. 2001b; Smith 2004).

Up-regulation of Major Histocompatibility Complex (MHC) class II antigens, and pro- inflammatory mediators such as cytokines/lymphokines (e.g. TNF-, IFN-, IL-1 and IL-6), chemokines (e.g. MIP-1 and MCP-1) and adhesion molecules (e.g. ICAM-1 and VCAM-1) (Amado et al. 1995; Takada et al. 1998; Wilhelm et al. 2000a; Gasser et al. 2001; Pratschke et al. 2001a; Plenz et al. 2002; Smith 2004) have been seen in both animal models and human donors. The up-regulation of MHC class II molecules has been seen in the liver (van der Hoeven et al. 1999b), kidneys (van der Hoeven et al. 1999b; Pratschke et al. 2001a; Pratschke et al. 2002) and hearts (Wilhelm et al. 2000a) of rats, which increases the immunogenicity of the organ via T-cell recognition and increases leukocyte (such as T-cells, macrophages and lymphocytes) infiltration in the allograft. As a result, there is accelerated allograft rejection and decreased survival (van der Hoeven et al. 1999b; Wilhelm et al. 2000a; Pratschke et al. 2002). The up- regulation of cell adhesion molecules such as ICAM-1 and VCAM-1 in the liver, kidney and heart also facilitate the adhesion and rolling of leukocytes onto endothelial cells and their infiltration into the organ (van der Hoeven et al. 1999b; Wilhelm et al. 2000a).

Further contributing to the immune system activation, ischaemia-reperfusion injury during brain death and the subsequent transplantation process can also up-regulate the expression of cytokines, chemokines and adhesion molecules (Pratschke et al. 1999; Smith 2004). Non-specific endothelial injury can also have immunogenic consequences leading to acute rejection (Pratschke et al. 1999). In addition to the non-specific injuries

- 31 - that occur before organ retrieval, perfusion, storage, transplantation and reperfusion can further stimulate an inflammatory response that can trigger and amplify acute host immunologic activity – as shown in Figure 1.8 (Gasser et al. 2001). These changes impact on the quality of organs used for transplantation and may also affect transplantation outcomes in terms of rejection. An understanding of these processes however, can provide therapeutic targets for treatments to improve transplantation outcomes.

Figure 1.8: Effects of donor brain death, organ retrieval and the transplantation process on donor organs and their impact on the immune response. Activation of the inflammatory and immune systems can determine the occurrence of acute and chronic rejection. Diagram reproduced from Gasser et al (2001).

1.4.4.4 Impaired Oxidative Metabolism In Brain Death Normal metabolic activity is considerably disturbed as a result of brain death and is associated with the hormonal perturbations seen in brain death (Hicks et al. 2006). Impaired aerobic metabolism despite normal oxygen delivery has been demonstrated

- 32 - both globally (Novitzky et al. 1988c; Depret et al. 1995) and within specific organs such as the heart (Novitzky et al. 1987d) and kidney (Wicomb et al. 1986b). As a result, there is an increased reliance on anaerobic metabolism resulting in lactic acidosis (Wicomb et al. 1986b; Novitzky et al. 1987d; Depret et al. 1995), in addition to a rapid depletion of high-energy substrates such as ATP, leading to contractile dysfunction in the heart (Novitzky et al. 1987b; Novitzky et al. 1987d; Smith 2004). Many of these metabolic and functional disturbances have been reversed with the use of hormonal treatments such as T3, cortisol and insulin, suggesting that hormonal changes are a major cause of mitochondrial dysfunction with impaired energy production at a cellular level (Novitzky et al. 1986; Novitzky et al. 1987b; Novitzky et al. 1987d; Novitzky et al. 1988c; Novitzky 1991; Novitzky 1997a). However, there have also been some studies to dispute the effectiveness of T3 administration (Howlett et al. 1989; Randell et al. 1993; Goarin et al. 1996). The catecholamine exposure seen in brain death also contributes to these metabolic changes, whereby pre-treatment of experimental animals prior to brain death induction with the - and -adrenergic blocker labetolol has been shown to prevent the shift to anaerobic metabolism seen in brain death, along with preventing myocardial ischaemia and contractile dysfunction (Seguin et al. 2001).

1.4.4.5 Effects Of Brain Death On Cardiac Contractile Function Haemodynamic instability is common in brain-dead donors, resulting from deterioration in cardiac function and loss of autonomic control of the cardiovascular system. Consequently, a significant number of hearts from previously young healthy donors may become unsuitable for transplantation (Novitzky et al. 1987b). Several investigators have demonstrated experimentally that cardiac contractility as assessed by the preload recruitable stroke work (PRSW) relationship deteriorates in untreated animals after brain death (Bittner et al. 1996a; Ryan et al. 2003c; Lyons et al. 2005). The reasons for this contractile dysfunction are controversial and likely to be multifactorial, many of which have been discussed in Section 1.4.4.1.

The reduction in cardiac contractility is biphasic with an initial transient decrease which may be caused by myocardial ischaemia (Ryan et al. 2003c) or rapid desensitisation of the -adrenergic signalling pathway (White et al. 1995) during the “autonomic storm”. This is followed by a sustained reduction in contractility which may be due to ongoing

- 33 - catecholamine-induced myocardial ischaemia (Seguin et al. 2001; Lyons et al. 2005), impaired -adrenergic signalling (D'Amico et al. 1995; White et al. 1995), impaired myocardial metabolism associated with the altered neurohormonal environment after brain death (Novitzky et al. 1988c; Bittner et al. 1996b; Lyons et al. 2005) and an up- regulation of pro-inflammatory mediators (Pratschke et al. 1999; Lyons et al. 2005). In addition, electrolyte imbalances can produce myocardial oedema and cardiac arrhythmias, further impairing cardiac function (Gasser et al. 2001).

1.4.4.6 Effects Of Brain Death On The Kidneys, Liver, Pancreas And Lungs In addition to the effects of brain death on the heart discussed earlier, other transplantable solid organs are also negatively affected by brain death. Brain death causes tissue ischaemia via a number of different mechanisms, such as hypotension and the autonomic storm-induced vasoconstriction. It is well recognised that kidneys transplanted from living donors, whether they be related or unrelated, have consistently better outcomes compared with those from brain-dead donors (Nagareda et al. 1993; Koo et al. 1999; van der Hoeven et al. 1999a; Pratschke et al. 2000; Pratschke et al. 2001a; Pratschke et al. 2001b; Blasi-Ibanez et al. 2009).

Brain death is associated with immunological and non-immunological damage to the kidney, which results in delayed allograft function (Kutsogiannis et al. 2006). Both the extensive histopathological changes and the metabolic disturbances in kidneys have been well described as a result of brain death (Wicomb et al. 1986b; Nagareda et al. 1993; Lagiewska et al. 1996). Renal damage due to ischaemia has been shown to affect post-transplant function with a high incidence of acute tubular necrosis in kidneys from haemodynamically unstable donors (Lagiewska et al. 1996; Pratschke et al. 1999). It has also been shown that if donor systolic blood pressure is below 80-90 mmHg, the incidence of allograft failure increases due to the loss of renal blood flow autoregulation and reduced glomerular filtration leading to acute tubular necrosis (Szostek et al. 1997; Kutsogiannis et al. 2006). These changes mediated by brain death in the donor result in delayed allograft function, which is associated with reduced recipient survival, increased rejection rates and increased rates of renal allograft nephropathy (Nagareda et al. 1993; van der Hoeven et al. 1999a; Pratschke et al. 2000; Kutsogiannis et al. 2006).

- 34 - Hypotension and ischaemia can also have deleterious effects on the liver, although it has been demonstrated that the liver has a large physiological reserve and can tolerate marked hypotension without significant deleterious effects on its function (Scudamore et al. 1984; Lin et al. 1989a; Lin et al. 1989b; Pratschke et al. 1999; Gasser et al. 2001). Of more significance to the liver are the effects of brain death and donor management on its morphology with venous congestion, piecemeal necrosis and periportal necrosis being reported (Nagareda et al. 1989; Lin et al. 1989a). Whilst hepatic energy status has been reported to be well maintained in animal models of brain death (Lin et al. 1989a; Kitai et al. 1993), there is still controversy as to whether decreased T3 impairs hepatocyte metabolism (Gasser et al. 2001).

Brain death in the liver donor has also been recognised to provoke and sustain inflammatory changes that have detrimental effects on the outcome of in both humans and animal models (Pratschke et al. 1999; van der Hoeven et al. 2001; Weiss et al. 2007). Donor brain death has been shown to cause up- regulation of inflammatory cytokines, increased cellular infiltrates and an increased rate of apoptosis in humans and in animal models (Takada et al. 1998; van der Hoeven et al. 1999b; Jassem et al. 2003; Weiss et al. 2007). Brain death has also been shown to reduce allograft survival of rat livers after prolonged ischaemia compared with grafts from living donors (van der Hoeven et al. 2003). In a prospective study of liver biopsies taken from 32 brain-dead and 26 living liver donors, brain death was associated with significantly higher expression of inflammatory cytokines such as IL-6, IL-10, TNF-, TGF- and MIP-1 and increased cellular infiltrates (Weiss et al. 2007). This study also demonstrated that livers from brain-dead donors had significantly worse ischaemia- reperfusion injury (as reflected by elevated levels of alanine aminotransferase, aspartate aminotransferase and bilirubin in the recipient), increased rates of acute rejection and primary non-function, and significantly reduced production and excretion of bilirubin in recipients. Golling et al demonstrated in a porcine model of the brain-dead liver donor that brain death caused deterioration in portal venous flow, microperfusion (assessed with an intrahepatic thermal diffusion probe), aspartate aminotransferase and hepatic oxidative stress, which was independent of haemodynamic stability (Golling et al. 2003). Deterioration in hepatic arterial flow was also seen in this study, but only in hypotensive brain-dead donors.

- 35 -

Little is known about the direct effects of brain death on the endocrine pancreas. Its effects on the exocrine pancreas seem to be more important as a trigger of the inflammatory system than the loss of function of the endocrine pancreas (Obermaier et al. 2004). Insulin levels have been reported to fall after brain death in a baboon model (Novitzky et al. 1984), whereas in a canine model it has been shown to increase moderately after brain death (Bittner et al. 1995a). As discussed in Section 1.4.4.2, most studies of insulin levels in the human brain-dead donor have generally shown elevated levels (Powner et al. 1990; Taniguchi et al. 1992; Masson et al. 1993; Brunicardi et al. 2000).

In a murine model of the organ donor, brain death causes significant pathophysiological changes in the pancreas (Obermaier et al. 2004). This study demonstrated that there was deterioration in microcirculation, elevated inflammatory tissue response (as evidenced by increased leukocyte adherence) and histological evidence of pancreatic damage, consistent with mild pancreatitis. Interestingly, despite this damage, there was no significant change in lipase activity and whilst brain-dead animals had an elevated amylase level, this did not reach pathological levels. In a prospective study of 25 brain- dead patients, it has also been reported that provided haemodynamic stability is maintained, pancreatic function does not change markedly following brain death (Masson et al. 1993). Masson et al (1993) also reported normal histology and immunohistochemistry in the pancreas of brain-dead patients.

Hyperglycaemia is often seen in brain-dead donors and is believed to be secondary to the effects of central nervous system trauma, massive levels of catecholamines, infusions of glucose-containing fluids, peripheral insulin resistance and the effects of steroids (Gores et al. 1990; Gores et al. 1992; Masson et al. 1993; Brunicardi et al. 2000; Gasser et al. 2001; Smith 2004). Insulin resistance is likely to be a major contributor to the hyperglycaemia rather than a lack of insulin secretion, given the human studies already discussed. This is thought to be due to two main mechanisms: impaired insulin receptor binding (decreased insulin sensitivity) and alterations in intracellular metabolism (decreased insulin responsiveness) (Masson et al. 1993).

- 36 - The lungs are known to be particularly susceptible to the haemodynamic and neurohormonal changes seen in brain death. The damaging effects of brain death on the lungs directly are a major factor determining early deaths in lung transplantation from primary graft failure (Fisher et al. 1998; Fisher et al. 2001; Avlonitis et al. 2003; Avlonitis et al. 2007) and have been reported to cause 28.2% of deaths in the first 30 days post-transplantation (Christie et al. 2008). Primary graft failure is seen in about 15- 50% of transplant recipients and has the features of acute lung injury (Christie et al. 1998; Thabut et al. 2002). It is an important complication in lung transplantation as it can prolong post-operative ventilation and is associated with a mortality of up to 60% (Christie et al. 1998). Acute lung injury is also a risk factor for bronchiolitis obliterans syndrome – a major complication determining long-term transplant survival (Avlonitis et al. 2003).

Brain death causes acute lung injury by neurogenic pulmonary oedema and by inflammatory acute lung injury (Avlonitis et al. 2003). The catecholamine storm of brain death mediates pulmonary oedema by haemodynamic and sympathetic (e.g. - adrenergic stimulation by noradrenaline) mechanisms. This initial insult can also prime the lungs for further ischaemia-reperfusion injury (Avlonitis et al. 2003). As a result of brain death, significant structural injury and pulmonary oedema secondary to abnormal capillary permeability and diffuse interstitial and intra-alveolar haemorrhage has been seen (Pratschke et al. 1999; Gasser et al. 2001). These changes lead to impaired gas exchange and a progressive reduction in the volume of ventilated lung tissue. In addition, the lungs are exposed to environmental contamination such as microbiological pathogens, environmental toxins, trauma, mechanical ventilation and aspiration of gastric and oropharyngeal contents, which can further impair its functional capabilities and contribute to acute lung injury (Gasser et al. 2001; Avlonitis et al. 2003).

As discussed in Section 1.4.4.3, brain death causes an activation of the immunological and inflammatory systems. As a result, there is an up-regulation of inflammatory mediators such as cytokines, lymphokines and adhesion molecules. Cytokines such as TNF-, IL1 and IL-8, and adhesion molecules such as ICAM-1 and the selectins have been recognised as playing a role in the pathogenesis of acute lung injury, further contributing to primary graft failure (Avlonitis et al. 2003).

- 37 -

1.4.5 Management Of The Brain-Dead Organ Donor In The Intensive Care Unit

1.4.5.1 The Principles Of Donor Management And The Need For Aggressive Management Of The Organ Donor The management of the brain-dead organ donor has been recognised as “the most neglected area of transplant medicine”, resulting in a significant number of transplant recipients with poor graft function due to inadequate donor management or organ preservation (Wheeldon et al. 1995; Valero 2002). It has been estimated that failure to provide adequate physiological support to potential donors accounts for at least 25% of lost donor organs (Nygaard et al. 1990; Mackersie et al. 1991). As previously discussed, brain death in the organ donor causes significant haemodynamic instability and organ dysfunction via a number of different mechanisms, and as the time between brain death to organ procurement increases, so too does the incidence and severity of the instability (Nygaard et al. 1990). Therefore, timely treatment of the donor is essential, whereby the use of standardised donor treatments and algorithms to maintain the haemodynamic status of the donor can lead to more stable organ donors (Wood et al. 2004). In fact, aggressive donor management strategies may transform a significant number of organs that may be initially thought to be unsuitable for transplantation into acceptable and successful allografts. This would minimise the loss of donors during the maintenance phase and increase the yield of organs per donor that could be transplanted successfully. It seems intuitive that the optimisation of cardiac function and the stabilisation of the haemodynamic status of the donor will improve the function of all other transplantable organs (Wheeldon et al. 1995; Wood et al. 2004). An aggressive and standardised approach to donor management has been shown to be effective not only in increasing the yield of donor hearts, but in other transplantable solid organs such as the kidney, liver, lung and pancreas (Wheeldon et al. 1995; Rosendale et al. 2002; Rosendale et al. 2003a; Rosendale et al. 2003b; Salim et al. 2005).

In an effort to optimise the recovery and transplantation of organs from the currently available donor pool and to increase the quality of donor organs, a consensus conference entitled “Maximising Use of Organs Recovered From the Cadaver Donor”

- 38 - was held on 28 to 29 March 2001 in Crystal City, Virginia USA (Rosengard et al. 2002; Zaroff et al. 2002). This meeting brought together nearly 100 participants from key stakeholders in organ transplantation and included physician and surgeon members of the American Society of Transplantation (AST), American Society of Transplant Surgeons (ASTS), medical and executive directors of Organ Procurement Organisations (OPO), representatives from the Department of Health and Human Services (DHHS), the National Kidney Foundation (NKF), the International Society for Heart and Lung Transplantation (ISHLT), the Scientific Registry for Transplant Recipients University Renal Research and Education Association (URREA), and the United Network for Organ Sharing (UNOS). The conference assessed the current evidence regarding evaluation and management of potential organ donors and five work groups (heart, lungs, livers, kidneys and the expanded donor - cadaver donors with a history of malignancy or serology testing positive for hepatitis C or B virus) were established to provide, amongst other objectives, a donor management protocol and algorithm to maximise organ usage. The Heart Work Group proposed modified donor criteria to expand the available pool of cardiac donors in criteria such as older donors, smaller donors, hepatitis C or B virus-positive donors, mild left ventricular hypertrophy, mild or moderate valvular and congenital cardiac abnormalities amenable to ‘bench’ repair prior to transplantation, and acceptance of donors with mild coronary artery disease (Zaroff et al. 2002). The Heart Work Group also made recommendations for improving donor management and included advice on conventional management (including metabolic management), the use of echocardiography, hormonal resuscitation (discussed in Section 1.4.5.3) and aggressive haemodynamic management. These recommendations were incorporated into a standardised donor management algorithm (summarised in Figure 1.9), and have since been incorporated into the UNOS Critical Pathway algorithm (Zaroff et al. 2002).

- 39 -

Figure Overleaf

- 40 -

Figure 1.9: The cardiac donor management algorithm developed by the Crystal City Consensus Conference Heart Work Group, which have been adopted into the United Network for Organ Sharing (UNOS) Critical Pathway (Zaroff et al. 2002).

CVP: Central Venous Pressure; Sat.: Saturation; Hct: Haematocrit; Hb: Haemoglobin; MAP: Mean Arterial Blood Pressure; Echo: Echocardiogram; LVEF: Left Ventricular Ejection Fraction; LVH: Left Ventricular Hypertrophy; T3: Triiodothyronine; SVR: Systemic Vascular Resistance; BG: Blood Glucose; PCWP: Pulmonary Capillary Wedge Pressure.

• Adjust volume status: target CVP = 6 – 10 mmHg • Correct acidosis: target pH = 7.40 – 7.45 Conventional • Correct hypoxaemia: target PaO2 >80 mm Hg; O2 sat. >95% Management • Correct anaemia: target Hct 30%; Hb 10 g/dL • Adjust inotropes to keep MAP 60 mmHg (target dopamine or dobutamine <10 μg/kg/min)

• Rule out structural abnormalities:

Obtain - substantial LVH Initial Echo - valvular dysfunction - congenital lesions

LVEF 45% LVEF  45%

Proceed With HORMONAL RESUSCITATION • Recovery For T3: 4 μg bolus + infusion at 3 μg/hour Transplantation • Vasopressin: 1 U bolus + infusion at 0.5-4 U/hr (titrate to SVR of 800 - 1200 dyne/sec/cm5) • Methylprednisolone: 15 mg/kg bolus • Insulin: 1 U/hr minimum (titrate to BG of 120 – 180 mg/dL

Haemodynamic Management (duration 2 hours) • Place pulmonary artery catheter • Adjust fluids, inotropes and pressors q15 minutes to reduce use of - agonists and meet the following target criteria: - MAP >60 mmHg - CVP 4 – 12 mmHg - PCWP 8 – 12 mmHg - SVR 800 – 1200 dyne/sec/cm5

- Cardiac Index >2.4 L/min/m2 - Dopamine or dobutamine <10 μg/kg/min

CRITERIA MET CRITERIA NOT MET

PROCEED WITH RECOVERY DO NOT RECOVER HEART FOR TRANSPLANTATION - 41 - In general, the goals of haemodynamic management of the donor are to: 1) achieve normovolaemia; 2) adjust vasoconstrictors and vasodilators to maintain a normal afterload; 3) optimise cardiac output to achieve gradients of perfusion pressure and blood flow that promote organ function, without relying on high doses of -agonists or other inotropes and vasopressors which can increase myocardial oxygen demand and deplete myocardial energy stores; and 4) maintain blood pressure, targeting a MAP 60 mmHg, and in some cases, prevent hypertension and keeping MAP <90 mmHg (Powner et al. 2000a; Zaroff et al. 2002; Tuttle-Newhall et al. 2003; Powner et al. 2004; Smith 2004; Wood et al. 2004). Other goals of donor management are to: 1) maintain normal acid-base balance; 2) correct hormonal perturbations; 3) prevent hypoxaemia and hypercarbia; 4) correct anaemia (target haematocrit 30% and haemoglobin 10 g/dL); 5) correct electrolyte and osmotic imbalances, caused by, inter alia, diabetes insipidus (as discussed in Section 1.4.4.2); 6) maintain normoglycaemia, with insulin and/or glucose infusions if necessary; 7) prevent hypothermia and maintain core temperature >35°C (95°F), as brain death causes loss of hypothalamic thermoregulation resulting in a poikilothermic donor; 8) prevent and treat cardiac arrhythmias; and 9) correct coagulopathy and thrombocytopaenia with blood product replacement (Powner et al. 2000c; Powner et al. 2000d; Zaroff et al. 2002; Tuttle-Newhall et al. 2003; Powner et al. 2004; Smith 2004; Wood et al. 2004).

It is also vital to maintain judicious and optimal ventilation, not only to optimise oxygenation and achieve normocarbia, but also to prevent lung injury (e.g. barotrauma, atelectasis and pulmonary oedema) and inflammation (which includes increased cytokine release), as well as assisting in acid-base management (Powner et al. 2000b; Tuttle-Newhall et al. 2003; Powner et al. 2004; Smith 2004; Wood et al. 2004). In addition, there are competing requirements with regard to fluid resuscitation and its effects on the lungs, kidneys and other organs that need to be considered. On the one hand, minimally positive fluid balance reduces the risk of pulmonary oedema and is associated with higher rates of lung procurement, whereas on the other hand, aggressive fluid resuscitation facilitates organ perfusion and the maintenance of renal function (Wood et al. 2004).

- 42 - 1.4.5.2 The Use Of Inotropes/Vasopressors To Support The Haemodynamically Unstable Donor, And Its Impact On Transplantation Outcomes According to data from the Australia and New Zealand Organ Donation Registry (www.anzdata.org.au/index_ANZOD.htm), more than 90% of brain-dead donors develop hypotension and are treated with some form of inotrope or pressor support. The choice of agents used to support haemodynamics is highly variable, reflecting local practices, particularly as there is little evidence to support the use of any single agent over others. The most common treatment is noradrenaline (81% of donors in 2007), but other inotropes/vasopressors such as adrenaline, dopamine, dobutamine and metaraminol have also been used, with many donors receiving multiple agents (Excell et al. 2008). In addition, agents such as desmopressin and vasopressin have been commonly used (57% in 2007). Interestingly, in 2007, only 20 of 236 donors (8.5%) – of whom 12 were DCD donors - did not require any inotropic support, a proportion that has remained similar over several years. The duration of inotropic and vasopressor support has also varied considerably, with at least 90% of donors receiving between 6 and 24 hours of support between brain death declaration and organ retrieval.

Whilst some believe that high requirements for vasoactive support in the donor does not preclude successful donation (Finfer et al. 1996; Wood et al. 2004), the impact of inotrope and vasopressor treatments in the brain-dead donor and their effects on subsequent transplantation outcomes are unclear and vary depending on the organ transplanted. It is thought that myocardial ATP stores are rapidly depleted by exogenous catecholamine administration, leading to adverse effects on post-transplantation outcomes (Smith 2004). There are reports of poorer outcome in human cardiac transplantation utilising donor hearts exposed to noradrenaline with initial non-function post transplantation (hazard ratio 1.66, 95% confidence interval: 1.14-2.43) (Schnuelle et al. 2001). There is also experimental evidence of contraction band lesions and myocardial necrosis as a result of catecholamine toxicity (Todd et al. 1985a; Todd et al. 1985b) and cardiac dysfunction (such as left ventricular dilatation, reduced ejection phase indices and decreased contractility) with cardiac necrosis and fibrosis as a result of noradrenaline toxicity (Movahed et al. 1994).

- 43 - Both vasopressin and its synthetic analogue desmopressin have been used to treat diabetes insipidus (Howlett et al. 1989; Chen et al. 1999; Keck et al. 2001; Wood et al. 2004). Arginine vasopressin (AVP), also know as antidiuretic hormone (ADH), acts on the V1- and V2-vasopressin receptors to produce vasoconstriction and antidiuretic effects. Desmopressin, also known as 1-desamino-8-D-arginine vasopressin (or

DDAVP), is specific for the V2-vasopressin receptor and has predominately antidiuretic effects (Wood et al. 2004). Desmopressin does not have the same vasoconstrictor effects as vasopressin and hence has not been associated with changes in inotrope requirements as in other studies with vasopressin (Guesde et al. 1998; Chen et al. 1999; Keck et al. 2001). Of concern however, are reports that desmopressin may be associated with a higher incidence of human pancreatic graft thrombosis (Marques et al. 2004) and that it has been shown to impair pancreatic graft microcirculation in a murine model (Keck et al. 2001).

In a study of brain-dead adults, Yoshioka et al demonstrated that the use of a low dose vasopressin infusion with adrenaline (<0.5 mg/hour) maintained blood pressure, reduced inotrope requirements and led to long-term stabilisation of the donor for an average of 23 days (Yoshioka et al. 1986). Similarly, in other studies of brain-dead donors, it has been shown that the use of vasopressin infusions (300 μU/kg/min in one study and 0.04 to 0.1 U/min in another) significantly improves blood pressure and facilitates the reduction of inotrope requirements (and in many cases, a cessation of inotropes) (Pennefather et al. 1995; Wheeldon et al. 1995; Chen et al. 1999). Pennefather et al also demonstrated that graft function of the kidneys, livers and hearts transplanted were no different between donors treated with vasopressin and those with saline (Pennefather et al. 1995). The use of low dose vasopressin has also been associated with good renal, hepatic and cardiac graft function after transplantation in other studies (Wheeldon et al. 1995; Rosendale et al. 2003a).

In the field of renal transplantation, research into the effects of donor vasopressor treatment on transplant outcomes has been inconclusive with uncertain long-term effects (Schnuelle et al. 2001). In a review of 2 415 kidney, 755 liver and 720 heart transplants from the International Foundation registry, Schnuelle et al identified poorer outcomes in recipients of catecholamine-exposed donor hearts (hazard ratio 1.66, 95% confidence interval 1.14-2.43) (Schnuelle et al. 2001). On the other

- 44 - hand, they found better four-year graft survival in the recipients of catecholamine- treated kidneys (hazard ratio 0.85, 95% confidence interval 0.74-0.98), whose benefit was dose-dependent, but no effect was seen on post-transplant outcomes in the recipients of catecholamine-treated livers (hazard ratio 0.94, 95% confidence interval 0.67-1.32). Conversely, renal grafts from donors treated with dopamine with and without vasopressin have been reported to have a higher incidence of acute tubular necrosis (Schneider et al. 1983), whereas other reports have shown that haemodynamic support with vasopressin and adrenaline is effective for long-term renal preservation (Kinoshita et al. 1990; Nagareda et al. 1993). Other reports have found that combination vasopressor therapies in the donor have been associated with reduced acute rejection rates and improved graft survival in renal recipients (Schnuelle et al. 1999), and that catecholamines and dopamine may have immunomodulatory effects, such as inhibiting adhesion molecule up-regulation, that could mitigate the inflammatory changes seen in brain death (Amado et al. 1995; Schnuelle et al. 2001; Tilney et al. 2001). However, there are still some reports that demonstrate worse outcomes in recipients of donor organs treated with catecholamines (Schneider et al. 1983; Pienaar et al. 1990; Nakatani et al. 1991; Hicks et al. 2006).

The use of catecholamines such as noradrenaline, adrenaline and dopamine has been associated with worse gas exchange after clinical lung transplantation (Mukadam et al. 2005). In a retrospective analysis of 60 lung transplant donors and recipients, Mukadam et al (2005) demonstrated that there was a greater reduction in the PaO2/FiO2 ratio in catecholamine treated donors, that was independent of ischaemic time, preservation technique and recipient diagnosis. In a randomised controlled trial investigating hormone treatments of the donor and aggressive donor management (including ventilation and haemodynamic optimisation and bronchoscopy), the use of noradrenaline was associated with deteriorating PaO2/FiO2 ratio and extravascular lung water index (EVLWI) (Venkateswaran et al. 2008). It has also been suggested that noradrenaline released from the nerve endings of the sympathetic nervous system may be a mediator of neurogenic pulmonary oedema seen in brain death (Avlonitis et al. 2003). In contrast to the aforementioned studies, it has been hypothesised that catecholamines may reduce the impact of ischaemia-reperfusion injury by modulation of leukocyte adhesion mechanisms (Schnuelle et al. 1999), and may promote alveolar fluid clearance, thereby promoting recovery from pulmonary oedema (Maron et al.

- 45 - 1994; Berthiaume et al. 2002; Matthay et al. 2002; Azzam et al. 2004). In an animal model, it has also been shown that noradrenaline improves oxygenation and limits pulmonary oedema after brain death (Avlonitis et al. 2005; Rostron et al. 2008).

In liver transplantation, data from the literature on the effects of donor vasopressor use on transplant outcomes are conflicting (Schnuelle et al. 2001). It is generally accepted that haemodynamically compromised donors should be resuscitated adequately with intravenous fluids and electrolyte imbalances corrected prior to the use of vasoactive and inotropic agents. However, there have been concerns raised in the literature that the use of vasoconstrictors and inotropes may decrease mesenteric and renal perfusion.

With its powerful, non-selective vasoconstrictor effects, vasopressin may impair hepatic perfusion (Pennefather et al. 1995) and can potentiate the vasoconstrictor effects of many other agents, including noradrenaline (Holmes et al. 2003). Experimental animal studies have demonstrated that, at physiological concentrations, vasopressin has selective effects affecting predominately the cutaneous and skeletal circulations, rather than the hepatic circulation (Liard et al. 1982; Goldsmith 1987). Additionally, there is some evidence to suggest that vasopressin and adrenaline act synergistically to improve haemodynamics and maintain hepatic energy metabolism in brain-dead dogs (Manaka et al. 1990; Manaka et al. 1992) and that hepatic function is well preserved in human donors supported with vasopressin (Yoshioka et al. 1986; Nagareda et al. 1989). Investigation into the effects of noradrenaline in a conscious merino ewe model demonstrated that, at 0.4 μg/kg/min, noradrenaline did not alter superior mesenteric arterial blood flow, although there was mesenteric vasoconstriction (as demonstrated by a 20% increase in mesenteric vascular resistance) (Di Giantomasso et al. 2002). Dopamine has been implicated in the reduction of hepatic mitochondrial redox state (Nakatani et al. 1991) and the use of high-dose vasopressors in the donor has been associated with adverse hepatic transplantation outcomes (Yamaoka et al. 1990).

As with the liver, there have been concerns raised that vasopressin may impair pancreatic perfusion, function and preservation due to its vasoconstrictor effects (Papp et al. 1983; Beijer et al. 1984), and as previously mentioned, it can also potentiate the vasoconstrictor effects of agents such as noradrenaline (Holmes et al. 2003). Brain death has also been shown to cause significant pathophysiological alterations in the

- 46 - pancreas in an experimental rat model, manifested as deterioration of pancreatic microcirculation, elevated inflammatory tissue response and histological damage (Obermaier et al. 2004).

It appears that the use of catecholamines in the brain-dead organ donor has different effects on various organs and their transplantation outcomes. At present, it is unknown whether these differences in transplanted organ outcomes reflect differences in the type of donors that receive catecholamines, the direct effects of catecholamines on different donor organs or indirect effects (such as better maintenance of blood flow to the kidney in donors receiving catecholamine infusions). This potentially leads to the problem of deciding what balance of treatments should be used in the brain-dead organ donor, whereby one treatment may be beneficial for a particular organ but may be harmful to another.

In a large review of the OPTN database, it was found that 20% more kidneys were transplanted from donors whose heart was transplanted compared with those who did not have hearts transplanted (91.2% vs. 75.9%, p<0.001). Furthermore, kidneys from donors who had their heart transplanted also had a significantly lower incidence of delayed renal graft function (18.0% in those who had their heart transplanted vs. 24.8% when the heart wasn’t transplanted; p<0.001) and better one-year survival (90.9% vs. 87.3%; p<0.001) (Rosendale et al. 2002). These findings support the notion that treatment of the brain-dead organ donor to improve cardiac function, and hence cardiac output and peripheral organ perfusion, could also benefit renal function and possibly other organs also.

1.4.5.3 Hormone Resuscitation Of The Brain-Dead Donor Brain death has been associated with various hormonal depletions along with an up- regulation of the inflammatory/immunological system and a disturbance in oxidative metabolism. It has been proposed that hormone replacement therapies with treatments such as thyroid hormone, insulin, vasopressin and corticosteroids may reverse many of these disturbances. Studies of multi-hormone cocktail resuscitation of the brain-dead donor have been shown to improve cardiac function and donor haemodynamic status in both animal models and in humans (Novitzky et al. 1987b; Novitzky et al. 1987c;

- 47 - Novitzky et al. 1988a; Cooper et al. 1988b; Novitzky et al. 1988c; Jeevanandam et al. 1993; Wheeldon et al. 1995).

Following on from their work into the endocrine, physiological, functional and morphological effects of brain death (Novitzky et al. 1984; Wicomb et al. 1986a; Cooper et al. 1989), Novitzky and colleagues went on to investigate the potential role of hormone replacement therapy – initially in a porcine model of the brain-dead donor (Novitzky et al. 1987d), and then in human brain-dead organ donors (Novitzky et al. 1987b). Four hours after the induction of ischaemic brain death in a porcine model, hormone replacement therapy in the form of T3 (2 μg), cortisol (100 mg) and insulin (5- 10 units) was given intravenously, with repeat dosing one hour later (Novitzky et al. 1987d). When compared with brain-dead non-treated animals, hormone-treated animals were observed to have a return of cardiac output and stroke volume to pre-brain death control values, but left ventricular pressure remained significantly elevated. Hormone treatment was also associated with replenishment of myocardial energy and glycogen reserves, lost as a consequence of brain death, and a reduction in lactate levels. Following on from this porcine study, the first human study of hormone replacement therapy in the brain-dead donor was conducted, whereby 21 consecutive donors were given T3 (2 μg), cortisol (100 mg) and insulin (20 units) at hourly or two-hourly intervals (in addition to conventional therapy of intravenous fluids, inotropes and bicarbonate) and were managed for an average of 5.5 hours (Novitzky et al. 1987b). These were compared with 26 historical controls. Hormone-treated donors demonstrated significant increases in body temperature, improved MAP (53% increase from initial observation), a fall in central venous pressure (by 35%), increases in heart rate (by 35%), increases in cardiac output (measured in six cases only, increased from 3.6 to 7.5 L/min), and a reduction in lactate and pyruvate levels from supranormal levels by 52% and 45% respectively following hormonal therapy. At the same time, bicarbonate requirements to treat acidosis fell by 95% and the level of inotropic support also fell by 88%. T3 levels, which were initially low, were restored to normal or high levels after treatment, and cortisol and insulin levels (measured in only 10 patients) also rose from low or low normal levels to well above normal levels after treatment. In all cases, ECG abnormalities improved or disappeared, and all 21 donors were used for transplantation (heart, heart and lungs, and kidneys) with good immediate function, compared with 80% donor heart usage in the non-treated control group.

- 48 -

After demonstrating impairment of aerobic metabolism and an increase in anaerobic metabolism following brain death, Novitzky et al showed that T3 therapy alone could reverse metabolism from anaerobic back to aerobic (Wicomb et al. 1986b; Novitzky et al. 1988c). Following induction of brain death in a baboon model, the rate of utilisation of glucose, pyruvate and palmitate was markedly reduced, and was associated with an accumulation of lactate and free fatty acids, thereby indicating a switch from aerobic to anaerobic metabolism (Novitzky et al. 1988c). Following the administration of hourly boluses of T3 (2 μg), there was a substantial increase in the rate of utilisation of glucose, pyruvate and palmitate, and a reduction in plasma lactate and free fatty acids, indicating a reversal from anaerobic to aerobic metabolism. As mentioned above, a reduction in serum lactate and pyruvate levels after hormone treatment has also been reported in human donors (Novitzky et al. 1987b). These results support the notion that T3 is an important hormone to replace and is required for normal mitochondrial function and aerobic respiration. It has been hypothesised that mitochondrial oxidative activity is activated by T3, possibly by increasing intracellular calcium and activating key enzymes in the Kreb’s cycle, as well as stimulating ATPase systems.

As the active cellular form of thyroid hormone, T3 has many effects on cardiovascular physiology, which are shown in Figure 1.10. Cardiac functions including heart rate and cardiac output, and also systemic vascular resistance are closely linked to thyroid status. Thyroid hormone is known to increase peripheral oxygen consumption and substrate requirements, thereby increasing cardiac output, whilst at the same time, directly increasing cardiac contractility. T3 also acts directly on smooth muscle cells to dilate arterioles in the peripheral circulation, thereby decreasing systemic vascular resistance. This decrease in systemic vascular resistance subsequently activates the renin- angiotensin-aldosterone system, which leads to an increase in plasma volume and preload, that also increases cardiac output (Klein et al. 2001). T3 has been shown to have direct positive inotropic effects in isolated hearts in vitro (Snow et al. 1992; Ririe et al. 1995), and in normal and cardiomyopathic hearts in vivo (Jamall et al. 1997). The positive inotropic action of T3 appears to be mediated via up-regulation of sarco- endoplasmic reticulum Ca2+ (calcium ion) ATPase (SERCA) (Sayen et al. 1992; Holt et

- 49 - al. 1999; Trivieri et al. 2006) and is independent of the -adrenergic signalling pathway (Ririe et al. 1995).

In the UNOS Critical pathway algorithm (Rosengard et al. 2002; Zaroff et al. 2002), T3 has been recommended over T4 because the onset of action of T3 is rapid and it is not affected by the exogenous factors that can affect T4 (Novitzky et al. 2006). Whilst approximately 93% of donors have received T4 according to UNOS data, there have been no differences in effectiveness reported. Therefore although the onset of action of T4 is slower than T3, T4 is still thought to be effective, especially at large doses (Salim et al. 2001).

Figure 1.10: The effects of thyroid hormone on cardiovascular haemodynamics. Triiodothyronine increases cardiac output by affecting tissue oxygen consumption (thermogenesis), vascular resistance, blood volume, cardiac contractility and heart rate. RAA=renin-angiotensin-aldosterone. Diagram reproduced from Klein and Ojamaa (2001).

Hepatic energy status has been reported to be well maintained in animal models of brain death (Lin et al. 1989a; Kitai et al. 1993). However, there is still controversy as to whether decreased T3 impairs hepatocyte metabolism (Gasser et al. 2001). Some studies have demonstrated no benefit to T3 replacement alone in the brain-dead donor (Garcia-Fages et al. 1991), whereas when given in combination with vasopressin,

- 50 - improved hepatic mitochondrial reduction oxidation potential has been seen (Washida et al. 1992).

Since Novitzky’s publications investigating hormone resuscitation of the brain-dead donor, at least four other groups have reported the beneficial effects of hormone resuscitation (Jeevanandam et al. 1994; Wheeldon et al. 1995; Jeevanandam 1997; Salim et al. 2001; Zuppa et al. 2004; Salim et al. 2007). After initial pilot studies demonstrating improved cardiac function with hormone replacement therapy for initial compromised function (Wheeldon et al. 1994), the Transplant Unit at Papworth Hospital in Cambridge, United Kingdom reported encouraging results of their donor management regimen incorporating a hormone resuscitation protocol that was inspired by the Cape Town group (Novitzky et al. 1987b) and a Japanese group’s work on vasopressin (Yoshioka et al. 1986). Of 150 multi-organ donors, the Papworth group identified 52 donors who failed to meet minimum criteria for acceptance for heart donation due to low MAP (<55 mmHg), high central venous pressure (CVP>15 mmHg), elevated pulmonary capillary wedge pressure (PCWP>15 mmHg), low left ventricular stroke work index (<15 gm) and high inotrope usage (>20 μg/kg/min) (Wheeldon et al. 1995). After optimal management, which included hormone resuscitation (methylprednisolone, 15 mg/kg; T3, 4 μg bolus followed by 3 μg/hour infusion; arginine vasopressin, 1 U bolus followed by 1.5 U/hour infusion; insulin plus dextrose of minimum 1 U/hour to maintain normoglycaemia), 44 of the 52 donors yielded transplantable organs (29 hearts, 15 heart and lung blocs). Thirty-seven of 44 patients (84%) were alive and well from 13 to 48 months after transplantation with none of the five early deaths seen due to cardiac failure.

At the Temple University Health Sciences Center in Philadelphia, USA, Jeevanandam and colleagues reported on six donors who were receiving high dose inotropes with depressed left ventricular function and haemodynamic instability (Jeevanandam et al. 1994). After treatment with hourly T3 boluses of 0.2 μg/kg up to a total of 0.6 μg/kg, the haemodynamic condition stabilised and pressor requirements decreased, and all six hearts were transplanted. One week after transplantation, ejection fraction (EF) was >50% in all patients (on echocardiogram) and all were discharged alive an average of 13.6 days post-transplantation. In a follow-up to this study, 22 donors with elevated left atrial pressures (LAP>20.8±3.9 mmHg), low EFs (39.2±5.5%) and high inotrope

- 51 - requirements (22.3±5.2 μg/kg/min dopamine) were resuscitated with T3 (in a series of hourly 0.4 μg/kg bolus injections up to 1.2 μg/kg) (Jeevanandam 1997). Seventeen of the 22 donor hearts were successfully resuscitated, and were recovered and transplanted. All 17 patients survived transplant with no significant differences in cardiac function at one week compared with a control group of donor hearts that did not have any donor dysfunction (and hence did not receive T3), and all 17 were discharged from hospital.

At the Los Angeles County-University of Southern California Medical Centre in Los Angeles, USA, 19 haemodynamically unstable patients (MAP<70 mmHg despite fluid resuscitation and inotrope/vasopressor use, and utilising >10 μg/kg/min of adrenaline and/or dopamine) were resuscitated with one ampoule of 50% dextrose, 2 g of methylprednisolone, 20 U of insulin and 20 μg of levothyroxine, followed by a 10 μg/hour infusion (Salim et al. 2001). Following treatment, 11 of 19 donors had decreased inotrope requirements by two hours, and within four hours all donors had decreased requirements, with 10 donors completely weaned off vasopressors. Ten of the 19 patients underwent successful transplantation for a total of 33 organs, with the remaining nine not transplanted because of family refusal. In a review of 878 consecutive patients referred for possible organ donation, the policy of aggressive donor management which included hormone resuscitation with dextrose, solumedrol, insulin and T4, was associated with an increase in referrals and a reduction in the number of organ donors lost due to cardiovascular collapse when compared with the period before aggressive donor management (Salim et al. 2005).

In a retrospective cohort study, 91 brain-dead paediatric patients at the Children’s Hospital of Philadelphia, USA who were potential organ donors were given a bolus of T4 followed by an infusion based on weight (Zuppa et al. 2004). In addition, 57% of these patients received vasopressin and 40% received steroids. The use of T4 was associated with a significant reduction in vasopressor requirements, even after adjusting for steroid use, baseline vasopressor use and fluid balance. No information was given on organ usage or outcomes.

Whilst there have been many animal and human studies confirming the beneficial effects of hormone resuscitation in the haemodynamically unstable organ donor, there have also been a number of reports that failed to demonstrate any benefit from hormone

- 52 - treatment. In a porcine study of the brain-dead donor, animals were assigned to a vasopressin (2 U/hour), T3 (0.05μg/kg/hour) or control group (Meyers et al. 1993). However, neither the vasopressin nor the T3 group demonstrated any benefit over the control group in terms of preserving or prolonging left ventricular function after brain death. In a French study, 37 brain-dead patients were randomised to a T3 group (who received a single T3 bolus of 0.2 μg/kg) or a saline placebo group (Goarin et al. 1996). There were no significant differences seen in the T3 group compared with the placebo group with respect to haemodynamic status or myocardial function (as measured by echocardiography). At the Helsinki University Hospital in Finland, 25 consecutive donors were randomly allocated to either a control group or a treatment group, where a 2 μg/hour T3 infusion was given (Randell et al. 1992). Donors were haemodynamically stable in both groups and inotrope use was similar in both groups. After T3 treatment, arterial blood pressure and heart rate were no different to the control group and the inotropic support remained similar in both groups. At the same time metabolic acidosis was worse in the T3 group. Based on these results, the authors concluded that T3 therapy was not warranted in donor management and that it was harmful to the donor and their organs. Similarly, in a Spanish study of 52 consecutive donors randomised to either a control group or a T3 group (1 μg/kg bolus, followed by a 0.06 μg/kg/hour infusion), there was no difference demonstrated in haemodynamics or inotrope use between the two groups (Perez-Blanco et al. 2005). Lactate levels were lower in the T3 group and there were no differences between groups with respect to adenine nucleotide levels (a measure of cell viability). The authors concluded that T3 added no benefit to the organ donor. The failure to detect any benefit from T3 administration in the above studies may have been due to a number of reasons. Not all donors have a complete absence of anterior pituitary function and may have some thyroid hormone function, not all donors tested were haemodynamically unstable and hence the benefit from exogenous thyroid hormone may not be seen (particularly as many of the studies demonstrating a benefit from thyroid hormone treatment were in donors who were unstable), and an inadequate dose of thyroid hormone may have been given.

Despite the beneficial effects of hormone resuscitation demonstrated in both humans and animal models on cardiac function, haemodynamics and outcomes in heart and renal transplantation (Rosendale et al. 2003a; Rosendale et al. 2003b; Novitzky et al. 2006), published and unpublished data from the Medical University of South Carolina

- 53 - liver transplant program has suggested that hormone resuscitation has adverse effects on liver transplant outcomes, particularly when steatotic livers (where there is hepatocyte lipid accumulation) are used (Ellett et al. 2008). They reported that hormone resuscitation (including steroid and levothyroxine) caused higher peak transaminases, a greater rate of graft dysfunction and failure, and a greater need for postoperative plasmapheresis (Skerrett et al. 1996; Mandal et al. 2000; Ellett et al. 2008). In a mouse model of steatotic livers exposed to ischaemia-reperfusion injury, treatment with levothyroxine and methylprednisolone 48 hours prior to ischaemia was associated with decreased survival and a dramatic increase in liver necrosis (Ellett et al. 2008). This study also found a dramatic increase in the expression of uncoupling protein-2, a protein that has been implicated as a major mediator of ischaemia-reperfusion injury in steatotic animals, accompanied by a decrease in ATP levels after reperfusion. These findings however, should be interpreted with caution. The model used was neither a transplantation model nor did it incorporate brain death. Given that hormone resuscitation has been designed to ameliorate the negative effects of brain death, the model used in the study was not an ideal model for testing treatments in brain-dead donor management or transplant outcomes. In addition, as Novitzky and colleagues have pointed out, extrapolation of results in a rodent model to the clinical situation should be approached cautiously without confirmatory evidence from a large animal model (Novitzky et al. 2008).

In terms of hormone resuscitation in lung transplantation, T3 has been shown to increase alveolar fluid clearance (Folkesson et al. 2000). Conversely, in a randomised controlled trial, combined methylprednisolone and T3 or T3 treatment alone was shown not to affect lung yield, PaO2/FiO2 ratio or EVLWI (Venkateswaran et al. 2008). This trial also showed that methylprednisolone did reduce EVLWI. However, by the author’s own admission, these results should be interpreted with caution given the small numbers and the potential for the study to be underpowered to accurately detect differences between the groups. Other studies have also demonstrated the benefit of corticosteroids in the management of the donor. In a retrospective analysis of 118 consecutive organ donors, 80 donors received high dose methylprednisolone (14.5±0.06 mg/kg) and were managed for approximately 23.5 hours (Follette et al. 1998). Steroid-treated donors had a significant and progressive increase in PaO2/FiO2 ratio (by 16±14) and also had an increased yield of transplantable lungs from the available donors. In a retrospective

- 54 - review of organ donors, corticosteroid usage was an independent predictor of lung utilisation for transplantation (McElhinney et al. 2001) and in a UNOS retrospective review, three hormone resuscitation (methylprednisolone, thyroid hormone and vasopressin) was associated with a 2.8% increased probability of lungs being transplanted from a donor (Rosendale et al. 2003b).

After the development of the UNOS Critical Pathway for the Organ Donor (Rosengard et al. 2002; Zaroff et al. 2002), a retrospective analysis of all brain-dead donors in the United States (from the OPTN/UNOS database) from 1 January 2000 to 30 September 2001 was performed (Rosendale et al. 2003b). Donors who received hormone resuscitation (methylprednisolone bolus, vasopressin infusion and either T3 or T4) were compared with donors who received no hormones. In a cohort of 10 292 brain-dead donors, 701 received three-drug hormone resuscitation. The mean number of organs transplanted from this group (3.8) was 22.5% greater than that from non-hormone- treated donors (3.1). Multivariate analyses demonstrated that hormone resuscitation was associated with the following increased probabilities of an organ being transplanted from a donor: kidney 7.3%, heart 4.7%, liver 4.9%, lung 2.8% and pancreas 6.0%. Extrapolation of these probabilities to the 5 921 brain-dead donors recovered in the United States in 2001 would equate to an increase of 2 053 organs. In 218 donors in Australia in 2004, this would equate to an increase of about 76 organs. An increase of 4.8% in kidney utilisation was also seen in a pilot study the previous year, examining the use of a standardised, structured donor management algorithm, with no significant difference in either one-year graft survival or in the rate of delayed renal allograft function (Rosendale et al. 2002).

A further retrospective analysis of 4 543 heart recipients was undertaken using the UNOS/OPTN database (Rosendale et al. 2003a). Donors who received three-drug hormonal resuscitation (similar to that described above; n=394) were compared with those who did not receive all three drugs (n=4 149). Patients who received organs from donors with three-drug hormone resuscitation had significantly greater one month survival rate (96.2% vs. 92.1%) and one year survival rate (89.9% vs. 83.9%). One month graft loss was 3.8% for the three-drug group compared with 7.9% for hearts in the other group. Early graft dysfunction occurred in 5.6% of three-drug treated hearts compared with 11.6% of hearts that did not receive all three drugs. There was also a

- 55 - 46% reduced odds of death within 30 days and a 48% reduced odds of early graft dysfunction. Steroids alone and steroids plus T3 or T4 also significantly reduced the period of early graft dysfunction.

Unpublished OPTN/UNOS data has also shown that three-drug hormone resuscitation of the organ donor significantly improved one-year renal graft survival from both conventional and expanded criteria donors, compared with donors who received no hormone treatments (Novitzky et al. 2006). Similarly, in the same study, heart recipients also had significantly improved one-year survival. Multivariate analysis also demonstrated that three-drug hormone resuscitation was associated with a significantly reduced risk of one-year graft loss in both renal and heart recipients. On the other hand, three-drug hormone resuscitation had no effect on one-year liver transplantation survival.

With the increasing reports of the benefits of hormone resuscitation in the donor and the lack of any reports of adverse effects of hormone resuscitation on the donor or the organs to be transplanted, three-drug hormone treatment of the donor is becoming more common (Novitzky et al. 2006). There was an increase in three-drug hormone resuscitation from 8.8% in 2000 to 19.9% of donors reported to the OPTN in 2004.

1.4.5.4 Anti-Inflammatory Treatment Of The Brain-Dead Donor As discussed in Section 1.4.4.3, brain death is associated with the activation of the immune system and the up-regulation of pro-inflammatory mediators. These changes contribute to organ dysfunction and prime donor organs to be more susceptible to allograft dysfunction and rejection. Compounding these changes are the effects of ischaemia-reperfusion injury, which further activates the immunological and inflammatory systems, to also affect organ quality. Therefore, a potential approach to improving donor organs involves the use of specific and non-specific anti-inflammatory treatments aimed at reducing and even reversing the up-regulation of pro-inflammatory cytokines, adhesion molecules and donor specific antigens.

The use of high-dose steroids in donor management has been shown to improve the survival of renal and cardiac allografts in animal models of transplantation (Segel et al.

- 56 - 1997; Pratschke et al. 2001a). In clinical transplantation, the administration of steroids to brain-dead donors has been associated with improved yield of donor hearts and reduced early graft dysfunction after cardiac transplantation, possibly due to a reduction of intra-cardiac cytokines (Rosendale et al. 2003a). High-dose steroid administration in human donors has also been observed to improve oxygenation in donor lungs and improve the yield of lungs for transplantation from potential donors, as previously described (Follette et al. 1998; McElhinney et al. 2001; Rosendale et al. 2003b). The use of steroids as an anti-inflammatory treatment has also been demonstrated in the amelioration of neurogenic pulmonary oedema seen in acute lung injury in donors (Minnear et al. 1982; Edmonds et al. 1986) and has been shown to reduce lung injury and improve post transplant graft function in a rodent model (Wigfield et al. 2002).

The use of steroids as part of a hormone resuscitation management protocol of potential organ donors and its benefits has already been discussed extensively (Sections 1.4.5.1 and 1.4.5.3). Of note is the fact that the doses of steroids used in the studies described above and in the hormone resuscitation protocols discussed earlier were very high, indicating that the hormone treatments used in donor management is likely to be playing an anti-inflammatory as well as a hormone replacement role (Wheeldon et al. 1995; Salim et al. 2001; Rosendale et al. 2003b).

1.5 CARDIAC ALLOGRAFT PRESERVATION

1.5.1 Excision Of The Donor Heart And Preservation Techniques

More than 80% of brain-dead donors are multi-organ donors. In 2007, an average of 3.3 recipients were transplanted per donor in Australia and New Zealand (Excell et al. 2008). For donors who are having both thoracic and abdominal organs retrieved, a continuous median sternotomy and midline laparotomy is performed. After inspection and mobilisation of the organs to be retrieved has occurred, and the organs deemed suitable for transplantation, the donor is systemically heparinised. The ascending aorta is then cross-clamped, and cold preservation solution is rapidly infused into the aortic root and down the coronary arteries to rapidly cool the heart and to achieve electromechanical arrest of the heart. The venous circulation is vented, usually by incising the inferior vena cava, to decompress the right heart and to facilitate flushing of

- 57 - the abdominal organs, and the left heart is vented and decompressed, usually by incising the left atrial appendage or one of the left pulmonary veins (depending on whether the heart-lung bloc, lungs or the heart only is retrieved). At the same time, the abdominal organs are also flushed with cold preservation solution to rapidly cool the organs and achieve optimal preservation.

Once the organs have been flushed with preservation solution and have been cooled, they are individually excised (heart or heart-lung bloc, kidneys, liver and pancreas) and then submerged into cold preservation solution in plastic bags. The bags are then sealed (and usually multi-bagged into further bags of cold solution and crushed ice), and placed into an insulated container packed with ice to be stored at 0 to 8°C until implantation (Jahania et al. 1999; Hicks et al. 2006).

An alternative to static hypothermic ischaemic storage is continuous ex vivo perfusion of the donor heart. This involves the continuous infusion of oxygenated preservation fluid through the coronary circulation (Wicomb et al. 1982a). Continuous ex vivo perfusion techniques have been used to preserve donor kidneys since the 1960’s. Some experimental studies of continuous perfusion of the donor heart have demonstrated superior preservation over hypothermic ischaemic storage (Nutt et al. 1991; Qayumi et al. 1991), whilst others have demonstrated poor preservation with myocardial oedema and early graft dysfunction (Wicomb et al. 1982b; Wicomb et al. 1989a; Nickless et al. 1998). In a variation of this technique, Hassanein and colleagues reported on a portable perfusion device that allowed continuous, sanguinous perfusion of the donor heart in the beating working state, and found that it extended effective preservation time and reduced ischaemic injury (Hassanein et al. 1998). More recently, there has been the development of a commercially available portable warm blood perfusion system, which enables donor hearts to be stored ex vivo in a warm functioning state (www.transmedics.com). This system reportedly enables the heart to be perfused in a non-working state for resuscitation or in a working-heart mode to enable assessment of the function of the heart. According to the website for the company that produces this system, it has undergone clinical trials and has been cleared for marketing in Europe.

- 58 - 1.5.2 Injury To The Heart During Storage And Transplantation

1.5.2.1 Static Hypothermic Ischaemic Storage Of The Donor Heart

Hypothermia markedly reduces myocardial metabolism and energy consumption, and slows the loss of high-energy substrates. At the same time the ionic constituents of the preservation fluid facilitate rapid cessation of the heart’s electrical activity, which also reduces its energy consumption.

Hypothermia does not stop metabolism. It slows down biochemical reaction rates and decreases the rate at which intracellular enzymes degrade essential cellular components necessary for organ viability. At the same time myocardial energy consumption is also slowed but not stopped completely. According to Vant Hoff’s rule, enzyme activity decreases up to 50% for every 10°C drop in temperature. Hypothermia also retards lysis of organelles such as lysosomes, which would ordinarily release autolytic enzymes to cause cell death (Jahania et al. 1999).

Whilst hypothermia delays cell death, various processes are activated that can be deleterious to the stored organ. These include cell swelling, extracellular oedema, cellular acidosis, depletion of metabolic substrates, calcium overload and endothelial injury (Jahania et al. 1999). Normally, the ionic composition of the intracellular (high potassium, low sodium) and extracellular (high sodium, low potassium) fluid is maintained by the membrane Na+-K+ ATPase pump. During periods of ischaemia and hypothermia, there is a depletion of the energy stores (i.e. ATP, derived from oxidative phosphorylation in the mitochondria) that is required for the pump to function, as well as a suppression of the Na+-K+ ATPase pump activity (as per Vant Hoff’s rule). As a result, sodium and chloride diffuse across the cell membrane down their ionic concentration gradient, and the water that follows these ions causes cell swelling. In order to counter this cell swelling, colloids (high molecular weight vascular impermeants) and other impermeable substances can be added to preservation solutions to generate an osmotic pressure equivalent to the intracellular osmotic pressure to counter the water shift and hence prevent cell swelling. Examples of some of the substances used for this purpose include colloids such as hydroxyethyl starch, polyethylene glycol and dextran 40, and impermeants including saccharides such as

- 59 - lactobionate, raffinose, glucose and mannitol, and anions such as citrate, phosphate, sulphate and gluconate (Hicks et al. 2006).

Another consequence of hypothermic ischaemic storage is intracellular calcium overload. Normal calcium homeostasis in the cardiac myocyte under normothermic conditions is maintained by a number of sarcolemmal transport enzymes, and other exchangers and pumps. Many of these mechanisms are energy-dependant processes, whereby calcium is removed from the cytoplasm (either directly or indirectly) by ATPases. Inactivation of these ATPases (due to hypothermia and ischaemia, among other processes) along with activation of the Na+-H+ (sodium ion-hydrogen ion) exchanger during hypothermic storage causes calcium to accumulate in the cytoplasm leading to calcium overload.

Whilst hypothermic ischaemic storage slows down myocardial energy consumption and many enzymatic processes, some energy-dependant processes are still required to maintain cell viability. Therefore anaerobic glycolysis, converting glucose to pyruvate, occurs to provide energy for these processes. Under these conditions, this results in the rapid depletion of high-energy substrates, and pyruvate is metabolised to lactic acid by lactate dehydrogenase (LDH), thereby causing intracellular acidosis. Intracellular lactic acidosis injures intracellular organelles and activates macrophages, which then up- regulates cytokine production and stimulates the immune/inflammatory system (Jahania et al. 1999).

The accumulation of intracellular H+ (from increased lactic acid production) during hypothermic ischaemic storage of the heart activates the Na+-H+ exchanger or antiporter (NHE). The NHEs are a family of membrane proteins that transport Na+ and H+ in opposite directions on a one-to-one (i.e. electroneutral) basis and are widely distributed throughout the body (Avkiran 2003). There are seven isoforms of the NHE, of which the NHE-1 is the ubiquitous “housekeeping” isoform expressed throughout every cell type in the body. NHE-1 is one of four proteins involved in intracellular pH (pHi) regulation in cardiac myocytes and is relatively quiescent at normal pHi (as shown in

Figure 1.11a). However, its expression and activity is stimulated by a reduction in pHi, where it removes hydrogen in exchange for sodium in an energy-independent process. The subsequent influx of sodium is then removed by the Na+-K+ ATPase. During

- 60 - ischaemia, pHi falls and the NHE causes an influx of sodium, in exchange for extruding H+ from the cell (as shown in Figure 1.11b). However, ischaemia also results in ATP depletion and the Na+-K+ ATPase becomes inactive, which is further compounded by the effects of hypothermia. The increased sodium is therefore removed by activating the Na+-Ca2+ exchanger (NCE) in reverse mode, resulting in an increase in intracellular Ca2+ (iCa2+). Reperfusion also increases the activity of the NHE as it removes extracellular H+, increasing the H+ gradient across the sarcolemma. It is the increase in iCa2+ that has been implicated in the pathogenesis of ischaemia-reperfusion injury in the myocardium, leading to contractile dysfunction, ventricular arrhythmias and cell death (by both necrotic and apoptotic pathways) (Karmazyn et al. 1999; Allen et al. 2000; Avkiran 2003; Teshima et al. 2003).

It is well recognised that cellular and tissue acidosis is deleterious to normal cell function and can occur as a result of anaerobic metabolism producing lactic acid (as described above). One method to prevent acidosis during hypothermic ischaemic storage of organs is to add large concentrations of glucose or H+ buffers to the preservation solution. Buffers in cardiac preservation solutions have included potassium phosphate, sodium bicarbonate, magnesium sulphate and histidine (Jahania et al. 1999). One of the features of Bretschneider (HTK - Histidine Tryptophan ketoglutarate) solution, for example, is its extremely high concentration of histidine in comparison to other preservation solutions (See Table 1.3).

Another contributor to injury to the heart during ischaemic storage is extracellular oedema, resulting from the accumulation of interstitial fluid. This may occur as a result of the excessive hydrostatic pressure exerted on the organ vasculature during flushing of the organs with preservation solution during procurement. Extracellular oedema can impede flow of the flush solution by collapsing tissue capillaries, resulting in uneven perfusion and preservation of the organ. As with methods to prevent cell swelling, the addition of colloids and impermeants (as described above) to preservation solutions are utilised to prevent interstitial fluid accumulation (Jahania et al. 1999).

- 61 -

Figure 1.11: Activity of the Na+-H+ exchanger (NHE) under a) normal conditions; and b) during ischaemic conditions. During non-ischaemic conditions, ATP supply is non- limiting. Intracellular sodium is extruded by the Na+-K+ ATPase, calcium is extruded by the Na+-Ca2+ exchanger (NCE) and the NHE is quiescent. During ischaemia, ATP is depleted and the Na+-K+ ATPase becomes inactive. The accumulation of H+ as a result of glycolysis (anaerobic metabolism) activates the NHE to eliminate H+ from the cell, resulting in a large influx of Na+. Sodium is cleared from the cell by the NCE in reverse mode, resulting in a dangerous accumulation of intracellular calcium that may cause electrical instability, contractile dysfunction and myocyte death. Diagram reproduced from Hicks et al (2006).

- 62 - 1.5.2.2 Reperfusion Injury Although the restoration of oxygenated blood flow to organs is vital for the survival and function of ischaemic cells, the process of reperfusion can paradoxically lead to further tissue injury (Verma et al. 2002). The severity of this reperfusion injury is directly related to the severity and duration of the ischaemic insult that preceded it, which itself is related to the severity of cardiac myocyte damage. Reperfusion injury results in myocyte damage via myocardial stunning, microvascular and endothelial injury, and irreversible cell damage or necrosis (lethal reperfusion injury).

Several mechanisms and mediators have been identified in producing reperfusion injury. The most frequently reported include oxygen-derived free radicals, intracellular calcium overload, endothelial and microvascular dysfunction, platelet and complement activation, and altered myocardial metabolism (Verma et al. 2002). In addition, there is evidence that white blood cells also contribute directly to reperfusion injury after prolonged periods of ischaemia (Jordan et al. 1999).

1.5.2.3 Oxygen-Derived Free Radical Injury Restoration of oxygen delivery during reperfusion to tissues that have accumulated metabolites of anaerobic metabolism leads to a burst in the production of oxygen- derived free radicals and oxidants (Hicks et al. 2006). These include superoxide anion, hydrogen peroxide, hypochlorous acid, hydroxyl radical and peroxynitrite. Small amounts of oxygen-derived free radicals are produced as a normal by-product of a number of essential cellular processes (e.g. mitochondrial energy production and cell- to-cell signalling) but are prevented from causing cell injury by a variety of cellular antioxidant mechanisms. The abrupt increase in cellular levels of oxygen-derived free radicals that occurs during reperfusion after prolonged ischaemia is due in part to excess production of free radicals via the xanthine-hypoxanthine oxidase reaction.

Oxygen-derived free radicals cause cellular injury via a wide variety of mechanisms. These include lipid peroxidation, mucopolysaccharide polymerisation, membrane protein cross-linking, protein cleavage, alteration of glycosaminoglycan function and DNA disruption (Jahania et al. 1999). Compounding the problem of oxygen-derived

- 63 - free radical injury, prolonged ischaemia causes a depletion of the cell’s protective anti- oxidant reserves, thereby making it less capable of scavenging any excess free radicals generated during reperfusion (Hicks et al. 2006).

Potential approaches to the prevention of the burst in oxygen-derived free radical accumulation during organ storage and reperfusion include the addition to preservation solutions of pharmacological inhibitors of xanthine oxidase such as allopurinol, and the addition of exogenous antioxidant free radical scavengers (Hicks et al. 2006). These include reduced glutathione, mannitol, superoxide dismutase, prostaglandin synthesis inhibitors, vitamin E, desferrioxamine and 21-aminosteroids (Jahania et al. 1999; Hicks et al. 2006).

1.5.2.4 Calcium Overload During Reperfusion As discussed in Section 1.5.2.1, during hypothermic ischaemic storage of organs, there is an increase in intracellular calcium due to the coupled activation of the NHE and the NCE (in reverse mode). Reperfusion leads to further activation of the NHE to cause further calcium ion influx (Karmazyn et al. 1999). To further exacerbate the increase in intracellular calcium, calcium re-uptake into the sarcoplasmic reticulum is reduced due to the reduced activity of the ATP-dependent sarcoplasmic reticulum calcium pump and the fall in ATP levels during hypothermia and ischaemia. The sustained increase in intracellular calcium in diastole has two potentially lethal consequences for the cardiac myocyte. The first is the sustained contraction of actin-myosin proteins and the second is the sustained activation of calcium-dependant enzymes within the mitochondria resulting in mitochondrial failure (Hicks et al. 2006). Several approaches have been proposed to prevent calcium overload. These include the reduction of calcium concentration in preservation fluids, supplementation of preservation fluids with magnesium (which competes with calcium for calcium exchangers and pumps), and the addition of pharmacological agents that inhibit calcium influx, such as calcium channel blockers and the NHE inhibitors (Hicks et al. 2006). The addition of adenosine into cardiac preservation solutions has also been proposed because of its broad-spectrum cardioprotective properties and its action on targets such as the mitochondria and sarcoplasmic reticulum (Jahania et al. 1999; Dobson et al. 2004). Amongst other actions, adenosine is a potent systemic and coronary vasodilator, it preserves the

- 64 - myocardial phosphorylation potential, it is a scavenger of oxygen-derived free radicals and it may inhibit platelet aggregation.

1.5.2.5 The Role Of White Blood Cells In Ischaemic Storage Injury White blood cells have been identified as a potential source of oxygen-derived free radicals (Jahania et al. 1999). As a result of ischaemic injury to the vascular endothelium, there is an up-regulation of adhesion molecules, which in turn initiates adhesion and activation of circulating white blood cells and platelets (Jordan et al. 1999). In addition, the white blood cells may physically plug the lumens of microscopic vessels within the reperfused organ, leading to the no-reflow phenomenon (Rezkalla et al. 2002). In order to counteract these events, the use of white blood cell filters at the time of reperfusion have been shown to reduce reperfusion injury and graft dysfunction, both experimentally and clinically (Breda et al. 1989; Pearl et al. 1992a; Pearl et al. 1992b).

1.5.2.6 Endothelial Injury During Ischaemia And Reperfusion Under normal physiological conditions, the vascular endothelium produces a number of vasoactive substances that induce vascular smooth muscle relaxation and inhibit white blood cell and platelet adherence to the vessel wall (Hicks et al. 2006). These include nitric oxide (NO), endothelium-dependant hyperpolarisation factor and prostacyclin. Ischaemic injury to the endothelium inhibits the production of these compounds, and as a result, there is an up-regulation of pro-thrombotic and pro-inflammatory adhesion molecules, leading to platelet and neutrophil activation. This is further exacerbated by reperfusion, whereby oxygen-derived free radicals and endothelium-dependant vasoconstrictors such as endothelin-1 produced by reperfusion may further damage the vascular endothelium, increasing coronary vasoconstriction and reducing blood flow (Verma et al. 2002; Hicks et al. 2006). The major pathways involved in proliferation and contraction of the smooth muscle cells of the coronary vasculature is illustrated in Figure 1.12. This figure demonstrates the effects of endothelin, NO and prostacyclin on smooth muscle cells.

- 65 - Figure 1.12: The opposing relationships between the vasoconstrictor endothelin and the vasodilators nitric oxide and prostacyclin pathways in the regulation of coronary artery tone (adapted from Humbert et al) (Humbert et al. 2004).

NO and prostacyclins are potent vasodilators and have cytoprotective properties that may be beneficial for preservation of allograft function during and after cold ischaemic storage. Prostacyclin and related prostanoids have been used to produce vascular bed vasodilatation in the donor organ either by prior intravenous administration (Wallwork et al. 1987; Hirt et al. 1992; Nawata et al. 1996; Kishida et al. 1997) or by addition to preservation solution (Wallwork et al. 1987; Sanchez-Urdazpal et al. 1991; Changani et al. 1999). NO donors such as GTN and diazenium diolates (NONOates) have also been added to preservation solutions to counter the loss of endogenous nitric oxide that occurs during hypothermia and ischaemic reperfusion injury (Bhabra et al. 1996; Du et al. 1998; Baxter et al. 1999a; Baxter et al. 2001). NO may also scavenge oxygen- derived free radicals, and may prevent platelet aggregation and vascular smooth muscle proliferation (Radomski et al. 1987a; Radomski et al. 1987b).

- 66 - Endothelial injury caused by ischaemia-reperfusion has been implicated as a factor in the development of both acute allograft dysfunction and chronic allograft vasculopathy (Hicks et al. 2006). Another potential source of injury to the vascular endothelium is the high potassium concentration of some intracellular preservation solutions such as the University of Wisconsin (UW) solution, although this remains controversial (Jahania et al. 1999). Hyperkalaemia results in the release of tissue plasminogen factor, fibronectin, interleukin-1, NO and endothelin, all of which may be involved at various stages of hypothermic ischaemic storage-mediated cellular injury. A number of studies have suggested that the development of coronary allograft vasculopathy may differ according to the type of preservation solution used to preserve the organs, with two studies reporting higher rates of vasculopathy when the heart was stored in UW solution (Drinkwater et al. 1995; Garlicki 2003).

1.5.2.7 The Impact Of Ischaemic Storage Time On Heart Transplantation Outcomes As discussed earlier, both ischaemia and reperfusion can cause significant injury to the donor heart, which affects the recipient following heart transplantation. The severity of the injury caused by both ischaemia and reperfusion is, in part, determined by the duration of ischaemia. Donor heart ischaemic time has been identified as a risk factor for mortality at 1, 5, 10 and 15 years in the latest ISHLT Registry Report on heart transplantation (Taylor et al. 2008), as it has been in previous reports (Taylor et al. 2004; Taylor et al. 2005; Taylor et al. 2006; Taylor et al. 2007). The relative risk of mortality at 1, 5 and 10 years related to donor ischaemic time are shown in Figure 1.13. Based on these Registry data, heart preservation for transplantation is currently limited to four to six hours of hypothermic ischaemic storage (Jahania et al. 1999; Hicks et al. 2006; Taylor et al. 2008). This is in contrast to preservation of the liver, kidneys and pancreas, which have been successfully preserved and transplanted after 24 to 36 hours storage, although graft function may be transiently compromised (Jahania et al. 1999).

- 67 - a) Relative Risk Of 1-Year Mortality 1/2002-6/2006 (n=8 823)

b) Relative Risk Of 5-Year Mortality 1/2000-6/2002 (n=5 131)

c) Relative Risk Of 10-Year Mortality 1993-6/1996 (n=8 298)

Figure 1.13: The effects of donor heart ischaemic time on a) 1-year mortality; b) 5-year mortality; and c) 10-year mortality. 95% confidence limits are shown. (Taylor et al. 2008; www.ishlt.org/registries/slides.asp?slides=heartLungRegistry). - 68 - Several investigators have highlighted the significant interaction between donor heart ischaemic time and donor age in determining the risk of death after heart transplantation. In a single centre study, Del Rizzo and colleagues reported that ischaemic time only had a significant effect on post-transplant survival when the donor age was greater than 50 years suggesting that the older donor was more susceptible to ischaemic injury (Del Rizzo et al. 1999). More recent analyses of large multi-centre databases suggest that the interaction between ischaemic time and donor age may be evident at a younger donor age. An analysis of the UNOS database by Russo and colleagues reported an adverse interaction between increasing ischaemic time and increasing donor age with regard to post transplant survival (Russo et al. 2007). This interaction was observed in donors over the age of 20 years, but not in younger donors. Similarly, in the latest Registry report of the International Society for Heart and Lung Transplantation, donor heart ischaemic time was a powerful predictor of post-transplant mortality in recipients of hearts from donors over the age of 30 years but did not predict mortality in younger donors (Taylor et al. 2009). These analyses highlight the susceptibility of the older donor heart to ischaemic injury. As the utilisation of older donors increases, it can be expected that prolonged ischaemic times will continue to be associated with increased post-transplant mortality in the future unless more effective measures to combat ischaemia-reperfusion injury are developed.

1.5.3 Organ Preservation Solutions

1.5.3.1 Formulation Of Preservation Solutions

Given the complexity of the molecular and cellular mechanisms underlying ischaemia- reperfusion injury, it is unlikely that any single approach or treatment will provide optimal protection of the donor organ during ischaemic storage and reperfusion. A combination of therapeutic approaches targeting multiple steps of the pathophysiology of ischaemia-reperfusion injury is likely to provide maximal protection. One of the cornerstones of effective hypothermic ischaemic storage of the donor heart is the chemical composition of the storage and preservation solution. In the context of myocardial preservation, three general principles have guided the formulation of cardioplegic and preservation solutions: 1) rapid reduction of tissue metabolic rate by profound hypothermia and electromechanical arrest of the heart; 2) provision of a

- 69 - biochemical medium that maintains tissue viability and structural integrity; and 3) prevention of reperfusion injury (Hicks et al. 2006).

Many different myocardial preservation solutions have been developed and are in use in heart transplantation. However, there is no one recognised ideal solution that is used by a majority of heart transplant units, reflecting the uncertainty that exists over the optimal preservation strategy. In a study on the patterns of usage of preservation solutions and their associated survival, Demmy et al found that over 167 different types of preservation solutions were used in the United States (Demmy et al. 1997). The authors examined the survival benefits related to each solution, comparing solution used with data from the UNOS Heart Transplant Registry. They found that the choice of fluid was not necessarily based on any good evidence and that the related survival benefits were uncertain. Across Australia and New Zealand, there is also no universal preservation solution used by all units. However, in a survey of these units in 2002, all units were using extracellular preservation solutions to store donor hearts (McCrystal et al. 2004).

In terms of the preservation solutions available, many are commercially available solutions, such as Bretschneider (HTK, Custodiol); Celsior; St. Thomas solution (STS, Plegisol) and University of Wisconsin solution (UW, Viaspan), but many others are locally produced non-commercial solutions (Demmy et al. 1997; Richens et al. 2001). In some centres the same solution is used for both flush (cardioplegia) and storage, whereas other centres have elected to use separate solutions for initial cardioplegia and subsequent hypothermic storage and transportation. This strategy may not be optimal depending on the solution used, as some solutions have been designed for brief perfusion of the moderately hypothermic heart in situ and not for prolonged hypothermic ischaemic storage of cardiac allografts (Menasche et al. 1994). In a survey examining donor heart preservation techniques and outcomes across heart transplantation centres worldwide, the use of cardioplegic solution as a storage medium was associated with a 2.5 times increase in deaths compared with cold saline (Wheeldon et al. 1992). In other centres, donor hearts have been stored in solutions that have been developed for non-cardiac organ preservation such as Euro-Collins and UW solutions (Menasche et al. 1994). Whilst these solutions are better designed for static hypothermic ischaemic storage compared with cardioplegic solutions, they do not

- 70 - necessarily meet the specific metabolic requirements of ischaemic myocardium nor are they designed to prevent ischaemia-reperfusion injury in the cardiac allograft. Table 1.3 lists a selection of preservation solutions and their electrolyte composition and chemical additives.

1.5.3.2 Electrolyte Composition Of Preservation Solutions

Preservation solutions differ in terms of both electrolyte composition and chemical additives. Most solutions can be divided into two broad categories based on the sodium and potassium ion content: extracellular and intracellular solutions. Solutions that mimic extracellular fluid have a high sodium concentration (70 mmol/L) and a potassium concentration of 5-30mmol/L. Intracellular solutions contain a low sodium concentration (70 mmol/L) and a potassium concentration of 30-125 mmol/L. A list of some of the different types of solutions available and their constituents are shown in Table 1.3 and include Celsior and UW solutions, examples of an extracellular and an intracellular solution (Drinkwater et al. 1995; Wildhirt et al. 2000; Vega et al. 2001; Remadi et al. 2002).

The primary rationale for the use of intracellular preservation solutions is that the presence of similar concentrations of sodium and chloride in the intra-and extra-cellular compartments minimises the passive fluxes of these ions into the cell to cause cell swelling during hypothermia and ischaemia. Another advantage of intracellular solutions is that the high potassium concentration facilitates cardiac arrest, while the low sodium concentration reduces the drive for the NHE. However, despite these advantages, there is a significant concern over the effects of the hyperkalaemia in intracellular solutions to cause coronary endothelial cell injury and impair myocardial preservation. This is a controversial issue with contradictory experimental evidence (Sorajja et al. 1997; Chambers et al. 1999; Suleiman et al. 2001; Yang et al. 2004). The damaging effects of hyperkalaemia on endothelial cells may also be temperature- dependant. Several investigators have reported that UW solution provided excellent endothelial cell preservation at 4°C but caused endothelial injury at higher temperatures (von Oppell et al. 1990; Ou et al. 1999). These results suggest that if UW solution is used for preservation, it must be completely flushed out of the heart prior to rewarming at time of implantation.

- 71 - Table 1.3: Composition of various preservation solutions used in heart transplantation (from Michel et al. 2002).

Components LYPS Celsior STS- STS- UW-1 UW HTK STF EC NaCl 1 2

Type Extra Extra Extra Extra Extra Intra Intra Intra Intra Extra Electrolytes Na+ 110 100 147 120 125 30 15 20 15 154 K+ 20 15 20 16 30 125 9 27 115 - Ca2+ 1 0.25 2 1.2 - - 0.015 - - - Cl- 150 41.5 203 160 - - 32.03 27 15 154 Mg2+ 4 13 16 16 5 5 4 - - - 2- SO4 - - - - 5 5 - - - - 3- HPO4 ------42 - 2- H2PO4 - - - - 25 25 - - 15 - - HCO3 - - - 10 - - - 20 15 - Metabolic Agents Glucose 20 ------250 194 - Aspartate 2 ------Glutamate 1.4 20 -------Ketoglutarate ------1 - - - Tryptophan ------2 - - - Insulin (IU/L) 250 ------Pyruvate 2.5 ------Adenosine - - - - 5 5 - - - - Impermeants Lactobionate - 80 - - 100 100 - - - - Mannitol - 60 - - - - 30 60 - - PEG 2 ------D+ raffinose - - - - 35.36 35.36 - - - - HES (g/L) - - - - 50 50 - - - - Antioxidants Allopurinol - - - - 1 1 - - - - Reduced 0.3 3 - - 3 3 - - - - Glutathione Buffers Histidine - 30 - - - - 180 - - - Histidine-HCl ------18 - - - HEPES 20 ------Miscellaneous Chlorpromazine- 0.7 ------HCl Procaine-HCl - - 1 1 ------

Concentrations are given as millimoles per litre (mmol/L) unless otherwise stated. LYPS: Lyon preservation solution; STS-1: St. Thomas Hospital No. 1 cardioplegic solution; STS-2: St. Thomas Hospital No. 2 cardioplegic solution; UW-1: University of Wisconsin modified solution; UW: Standard University of Wisconsin solution; HTK: Histidine-Tryptophane- Ketoglutarate (Bretschneider) solution; STF: Stanford solution; EC: Euro-Collins solution; NaCl: Normal saline; Extra: Extracellular-type solution; Intra: Intracellular-type solution; PEG: Polyethyleneglycol; HES: hydroxyethyl starch.

- 72 - A further limitation of hyperkalaemic solutions relates to their depolarising action which results in continuing transmembrane fluxes and the consequent maintenance of high-energy phosphate metabolism, even during hypothermic ischaemia (Chambers et al. 1999). A potentially beneficial alternative to hyperkalaemic cardioplegia is to arrest the heart in a hyperpolarised or polarised state. This maintains the membrane potential of the arrested myocardium at or near to the resting membrane potential. At these potentials, transmembrane fluxes are minimised and metabolic demand is minimised as a result, hence improving myocardial protection. Recent studies have explored these alternative concepts for myocardial protection (Chambers et al. 1999; Dobson et al. 2004). The use of compounds such as adenosine or ATP-sensitive potassium channel openers, which are thought to induce hyperpolarized arrest, have demonstrated improved protection after normothermic, or short periods of hypothermic, ischaemia when compared to hyperkalaemic (depolarized) arrest. Similarly, studies in which the sodium channel blockers tetrodotoxin and lignocaine were used to induce polarised arrest (demonstrated by direct measurement of membrane potential during ischaemia) also showed better recovery of function after long-term hypothermic storage (Chambers et al. 1999; Dobson et al. 2004). Indeed, the combination of adenosine with lignocaine in the same cardioplegic solution, as proposed by Dobson and colleagues, has been shown to dramatically enhance myocardial protection during both normothermic and hypothermic ischaemia (Dobson et al. 2004).

Other important electrolyte components of myocardial preservation solutions are calcium and magnesium. As mentioned earlier, inactivation of Ca2+ ATPases together with activation of the NHE during hypothermic ischaemic storage allows calcium to accumulate within the cytoplasm, resulting in calcium overload during storage. Further activation of the NHE on reperfusion exacerbates calcium overload, causing activation of calcium-dependent enzymes, which cause cell injury through a variety of actions, contracture of the myofilaments and irreversible mitochondrial damage, culminating in cell death. Experimental studies indicate that complete omission of calcium from preservation solution is detrimental to myocardial recovery (Donnelly et al. 1991; Michel et al. 2000; Michel et al. 2002). On the other hand, similar experimental studies demonstrate that calcium concentrations equivalent to that found in the extracellular fluid are also detrimental to myocardial recovery after ischaemia (Chen 1996; Fukuhiro et al. 2000). Normocalcaemic concentrations appear to facilitate calcium overload

- 73 - during ischaemia and reperfusion. Low concentrations of calcium in the preservation solution together with high concentrations of magnesium, however, have been shown to limit calcium overload and improve myocardial preservation (Fukuhiro et al. 2000; McCully et al. 2003). Of the currently available commercial solutions, Celsior and STS No. 2 solution contain a low calcium concentration in combination with a high magnesium concentration.

1.5.3.3 Chemical Additives To Preservation Solutions In addition to differences in electrolyte composition, preservation solutions differ with respect to chemical additives. The additives used in commercially available preservation solutions fall into one of four broad categories, although some chemicals such as adenosine may belong to more than one category. The major categories are metabolic substrates, osmotic and oncotic impermeants, anti-oxidants and free-radical scavengers, and acid-base buffers. Examples of chemical additives within each category are shown in Table 1.3. Experimentally at least, these additives can be shown to enhance donor organ recovery using the preservation solutions to which they have been added (Wicomb et al. 1989a; Biguzas et al. 1990; Southard et al. 1990).

1.5.3.4 The Development Of Celsior As A Heart Preservation Solution Celsior is an extracellular preservation solution (i.e. high sodium and low potassium ion concentration) that was developed in France in the early to mid 1990’s. It was developed as a specific heart preservation solution that was designed to meet two major requirements: 1) to combine the general principles of cold organ storage with those specific for the preservation of ischaemic-reperfused myocardium, and 2) to allow a single solution to be used throughout all phases of the transplantation procedure, from initial cardioplegic arrest of the donor heart and subsequent hypothermic ischaemic storage, through to graft re-implantation and reperfusion (Menasche et al. 1994). It was also designed such that it could be used in its crystalloid form or as part of a blood- based medium.

The composition of Celsior is listed in Table 1.3. Its formulation was designed to prevent cell swelling, prevent oxygen-derived free radical injury, and prevent myocyte

- 74 - contracture by enhancing energy production and limiting intracellular calcium overload (Menasche et al. 1994). In order to prevent cell swelling, Celsior contains two impermeants in the form of mannitol and lactobionate at a total concentration of 140 mmol/L. This counteracts the intracellular osmotically active molecules responsible for water entry into the cells during hypothermic anaerobic inhibition of the Na+-K+ ATPase (Belzer et al. 1988).

Reduced glutathione is used as a powerful antioxidant that acts via several different mechanisms to directly and indirectly scavenge free radicals (Menasche et al. 1991). Reduced glutathione has also been shown to protect mitochondria from calcium influx, thereby allowing it to preferentially produce high-energy phosphate compounds (Menasche et al. 1994). The addition of the amino acid glutamate is used to yield high- energy phosphates under anaerobic conditions (Pisarenko et al. 1985) and acts synergistically with reduced glutathione to enhance the maintenance of energy stores, thereby limiting myocardial contracture related to the loss of ATP (Steenbergen et al. 1990).

A relatively high concentration of histidine is used in Celsior for its excellent hydrogen ion buffering capacities at low temperatures and also as an effective scavenger of singlet oxygen species (Menasche et al. 1994). As a result, histidine may play a role in limiting calcium overload and the resulting calcium-mediated myocardial contracture. In this setting, mannitol can provide additional protection via its hydroxyl radical scavenging properties. Furthermore, the low calcium content, high magnesium content and slight degree of acidosis (7.30 at 20°C) of Celsior further limits cellular calcium overload, thereby further inhibiting myocardial contracture (Menasche et al. 1994).

Since its development, Celsior has been tested in a variety of animal models and in clinical transplantation. It has been found to be a safe and effective cardiac preservation solution for transplantation (Garlicki et al. 1999; Wieselthaler et al. 1999; Michel et al. 2000; Vega et al. 2001; Michel et al. 2002; Remadi et al. 2002; Garlicki 2003). In a multicentre, randomised, control trial enrolling 131 heart transplant recipients (Celsior, n=64; control, n=67), Celsior was found to be as safe and effective compared with conventional solutions (Vega et al. 2001). In another study comparing HTK, UW and Celsior solutions, Celsior was shown to have better post transplant recovery and less

- 75 - vasculopathy and chronic rejection in the mid-term follow-up (Garlicki 2003). Celsior has also been demonstrated to be an effective preservation solution in clinical lung transplantation (D'Armini A et al. 2001; Thabut et al. 2001), kidney transplantation (Baldan et al. 1997; Faenza et al. 2001; Karam 2003; Pedotti et al. 2004), (Karam 2003; Boggi et al. 2004; Boggi et al. 2005) and liver transplantation (Nardo et al. 2001; Cavallari et al. 2003; Pedotti et al. 2004).

1.5.4 Novel Approaches To Myocardial Protection

1.5.4.1 Sodium-Hydrogen Exchange Inhibition To Prevent Intracellular Calcium Overload And Protect The Heart Research into myocardial ischaemia-reperfusion injury has led to the development of a number of potential interventions to reduce injury in an effort to increase ischaemic storage times and improve the quality of transplanted hearts and subsequent outcomes (Theroux 1999). One such development has been the inhibition of the NHEs. As discussed earlier, activation of the NHE during both ischaemia and reperfusion has been implicated in the pathogenesis of ischaemia-reperfusion injury in the myocardium. Its activation leads to an overload of cytosolic and mitochondrial calcium, and subsequently leads to ventricular arrhythmias, contractile dysfunction and cardiac myocyte death (Avkiran 1999; Allen et al. 2000; Avkiran 2003; Teshima et al. 2003). Selective inhibition of the NHE by cariporide (4-isopropyl-3-methylsulphonylbenzoyl- guanidine methanesulphonate, HOE642) and other related compounds has been shown to reduce the severity of ischaemia-reperfusion injury in a number of clinical (Buerke et al. 1999; Rupprecht et al. 2000; Theroux et al. 2000; Boyce et al. 2003) and animal model settings (Allen et al. 2000; Avkiran et al. 2002; Castella et al. 2002; Fedalen et al. 2003; Ruiz-Meana et al. 2003; Teshima et al. 2003; Ryan et al. 2003a; Kristo et al. 2004; Stevens et al. 2004).

Cariporide is one of the most extensively evaluated inhibitors of the NHE. It is highly specific for the NHE and is also highly selective for the NHE-1 isoform, which is found in the cardiac sarcolemma (Avkiran 2003). Studies of cariporide in animal models have demonstrated reduced iCa2+ overload and reduced damage from ischaemia-reperfusion

- 76 - injury (Allen et al. 2000; Avkiran et al. 2002; Teshima et al. 2003; Cropper et al. 2003a; Ryan et al. 2003a; Cropper et al. 2003b).

In one of the first human studies into NHE inhibition in the clinical setting of acute myocardial ischaemia and reperfusion, Rupprecht et al investigated the effects of cariporide on limiting infarct size and improving myocardial function in patients with acute anterior myocardial infarction and subsequent reperfusion by direct percutaneous transluminal coronary angioplasty (Rupprecht et al. 2000). In a multicentre, randomised, placebo-controlled clinical trial of 100 patients, they found that a 40 mg intravenous bolus of cariporide prior to reperfusion was associated with a higher EF (50% vs. 40%, p<0.05), a lower end-systolic volume (69 vs. 97 mL, p<0.05), significant improvements in regional wall motion abnormalities, and significantly reduced myocardial enzymes as markers of myocardial tissue injury (i.e. creatine kinase (CK), CK-MB and LDH).

Shortly after Rupprecht’s report, the first large-scale trial to assess the potential protective effect of NHE inhibition in humans was published and was known as the GUARDIAN trial (GUARd During Ischaemia Against Necrosis) (Theroux et al. 2000). In this trial, a total of 11 590 patients who were hospitalised for unstable angina (UA) or non-ST elevation myocardial infarction (NSTEMI) or undergoing high-risk percutaneous coronary intervention (PCI) or coronary artery bypass surgery (CABG) were randomised to receive placebo or one of three doses of cariporide for the period of risk. This consisted of cariporide doses of 20, 80 or 120 mg infusions every eight hours for two to seven days, and was initiated as soon as possible after admission with UA/NSTEMI and between 15 minutes to two hours before PCI or CABG. The trial failed to demonstrate any benefit of cariporide over placebo on the primary end-point of death or myocardial infarction (MI) assessed at 36 days. The 20 mg and 80 mg dosing regimen had no effect, whereas the 120 mg regime was associated with a 10% risk reduction (98% CI 5.5% to 23.4%, p=0.12). The benefit of this dose was restricted to the CABG group (risk reduction 25%, 95% CI 3.1% to 41.5%, p=0.03) and was maintained after six months. There was no effect on mortality. The rate of Q-wave MI was reduced by 32% across all groups but the rate of non-Q-wave MI was only reduced in the CABG group. Whilst the GUARDIAN trial failed to meet the study endpoint, it

- 77 - did demonstrate the safety of the drug with no increases in clinically serious adverse events, and that it was cardioprotective in patients undergoing CABG.

As a follow-up to the GUARDIAN trial, the first phase III myocardial protection trial known as the EXPEDITION (Na+/H+ Exchange inhibition to Prevent coronary Events in acute cardiac condition) study was devised to investigate the safety and efficacy of cariporide in the prevention of death or non-fatal MI inpatients undergoing CABG (Mentzer Jr et al. 2008). In a multicentre, international, randomised, placebo-controlled trial, 5 761 high-risk CABG patients were randomly allocated to receive either placebo or intravenous cariporide (180 mg pre-operative loading dose, followed by 40 mg per hour for 24 hours, and then 20 mg per hour for a further 24 hours – a total of 1620 mg over 49 hours). At five days, the combined incidence of death or MI was higher in the placebo group compared with the cariporide group (20.3% vs. 16.6%, p=0.0002) and MI alone was also higher in the placebo group compared with the cariporide group (18.9% vs. 14.4%, p=0.000005). However, mortality alone was higher in the cariporide group compared with the placebo group (2.2% vs. 1.5%, p=0.02) and was associated with an increase in cerebrovascular events. The benefits of cariporide on the combined endpoint of death or MI and of MI alone was maintained at six months, whereas the difference in mortality between groups at six months was not significant. As a result of these findings, the study authors concluded that cariporide was unlikely to be used clinically but that other NHE inhibitors still held promise as a treatment for ischaemia-reperfusion injury.

There have been a number of studies investigating the use of cariporide in cardiac transplantation. In a porcine model of orthotopic heart transplantation, Martin et al investigated the use of adenosine and cariporide supplemented, leukocyte-depleted blood cardioplegia to perfuse the hearts of non heart beating donor (NHBD) animals 30 minutes after circulatory arrest induced by exsanguination (i.e. 30 minutes of normothermic ischaemia) (Martin et al. 2003a). These hearts were stored for three hours, transplanted orthotopically, and then reperfused with adenosine and cariporide supplemented blood cardioplegia. They demonstrated that by using the above described preservation technique, the recovery of donor hearts from NHBD were comparable to organs used from beating heart donors. After showing benefit on myocardial compliance with cariporide using four hour storage (Kim et al. 1998b), Kim et al found

- 78 - that pre-treatment of donors and recipients with cariporide in a canine model of transplantation with 24 hour storage improved myocardial compliance, post weaning cardiac index, myocardial ultrastructure and weaning potential (Kim et al. 1998a). They also found no benefit of adding cariporide to cardioplegia or in pre-treating the recipient only. However, this study did not incorporate brain death or prolonged donor management. Other studies with cariporide in isolated perfused heart experiments have found improved function and reduced injury in pig (Scheule et al. 2003), rabbit (Myers et al. 1996) and rat (Kevelaitis et al. 2001; Cropper et al. 2003a; Cropper et al. 2003b) hearts, after 2.5 to 12 hours ischaemic storage.

The use of ischaemic preconditioning combined with the use of cariporide supplemented Celsior used to arrest and store hearts has been demonstrated to improve donor heart preservation in an isolated perfused rat heart model (Kevelaitis et al. 2001). This study utilised storage times of three hours. In another study utilising an isolated working rat heart model and six hours storage time, Cropper et al demonstrated that pre- treating hearts with cariporide prior to cardioplegic arrest and during reperfusion protected the heart from ischaemic injury and increased cardiac function after reperfusion (Cropper et al. 2003a). They also demonstrated that storage of hearts in cariporide-supplemented Celsior did not result in better recovery compared with cariporide pre-treatment. In a follow-up to this study, when combined with potassium channel openers (diazoxide and BMS-180488), cariporide was found to have significantly better cardioprotective effects than when used alone (Cropper et al. 2003b). The authors also found that pre-treatment (prior to storage and prior to reperfusion) with cariporide provided better cardioprotection than using either diazoxide or BMS-180488 alone as a pre-treatment prior to storage. As with their earlier study, this study utilised hearts that were stored for six hours.

Utilising a porcine model of donor brain death, hypothermic ischaemic preservation and orthotopic heart transplantation, Ryan et al investigated the use of pre-treatment of both the donor and recipient animal with cariporide (2 mg/kg bolus) prior to organ retrieval and again prior to reperfusion (Ryan et al. 2003a). Utilising an ischaemic storage time of four hours, they demonstrated that the use of cariporide reduced myocardial injury and improved contractile function compared with control animals. They also found that cariporide was associated with a lower release of troponin I (a marker of myocyte

- 79 - injury). Further studies from Ryan et al demonstrated that the use of cariporide pre- treatment enabled successful weaning from cardiopulmonary bypass after 14 hours ischaemic hypothermic storage and that cardiac contractility was higher and troponin I release was lower in hearts treated with cariporide after both 4 and 14 hours storage (Ryan et al. 2003d). Ryan et al also investigated the use of BMS-180448 (a pharmacologic ischaemic preconditioning agent) as a potential cardioprotective agent in cardiac allograft preservation in the porcine model of orthotopic heart transplantation described above utilising 14 hours storage time (Ryan et al. 2003b). In this study, it was demonstrated that pre-treatment of the donor and recipient with cariporide was more effective than with BMS-180448, and that there was no additional benefit from the combination of both treatments. It should be noted that none of the studies conducted by Ryan had prolonged donor management periods.

1.5.4.2 The Anti-Ischaemic Effects Of Glyceryl Trinitrate Organic nitrates such as GTN have been recognised to have anti-ischaemic effects and their use clinically has been well established since the 19th century. They are used predominately for the treatment and prevention of cardiovascular diseases, including coronary artery disease (stable, unstable and vasospastic angina), acute MI, hypertension and congestive heart failure (Ignarro et al. 2002; Csont et al. 2005). The anti-ischaemic and cardioprotective effects of organic nitrates and NO occur via a number of pathways, and include potent vasodilatation (both venous and arterial), anti- platelet activity, and inhibition of both neutrophil aggregation and adhesion (Tanoue et al. 1999; Csont et al. 2005).

Endogenous NO can have both cardioprotective and harmful effects during ischaemia and reperfusion. During myocardial ischaemia and reperfusion, endogenous NO release from the coronary endothelium decreases (Tanoue et al. 1999). However, in the presence of superoxide, NO reacts with superoxide quickly to form peroxynitrite, which is a highly reactive species that can induce lipid peroxidation, thereby damaging the coronary endothelium and myocardium (Beckman et al. 1990).

The vasodilatory effects of GTN occur via its conversion to NO or a NO-containing metabolite, which diffuses through plasma membranes to directly activate guanylate

- 80 - cyclase. This in turn catalyses the formation of guanosine-3’5’-cyclic monophosphate (cGMP). The second messenger cGMP activates the cGMP-dependant protein kinase, protein kinase G, which then alters the phosphorylation state of various proteins. This ultimately leads to alterations to calcium ion levels in smooth muscle cells, dephosphorylation of the myosin light chain and the activation of ATP-sensitive

+ potassium channels (K ATP). These changes cause vascular smooth muscle relaxation, resulting in vasodilatation, which then causes a decrease in cardiac pre-load and after- load, improves coronary collateral flow and dilates stenotic coronary arteries. As a result, there is a reduced myocardial workload and therefore reduced oxygen and energy demand, whilst at the same time, increasing oxygen and energy supply (Harrison et al. 1993; Csont et al. 2005). In addition to these effects, there is evidence that nitrates have

+ direct myocardial anti-ischaemic effects that may be mediated by the activation of K ATP by NO and are independent of cGMP (Ferdinandy et al. 1995; Csont et al. 1999; Csont et al. 2005). As a result of these anti-ischaemic effects there have been a number of reports using exogenous NO donors, mainly GTN, in small and large animal models as an effective adjunct to hypothermic cardioplegia and preservation in heart transplantation (Oz et al. 1993; Pinsky et al. 1994b; Du et al. 1998; Tanoue et al. 1999; Baxter et al. 1999a; Baxter et al. 1999b; Baxter et al. 2001; Duranski et al. 2005; Gao et al. 2005). Supplementation of preservation solution with GTN and the use of GTN in reperfusion of transplanted lungs have also been reported to enhance preservation and function of lungs in transplantation (Wada et al. 1099; Naka et al. 1995; Bhabra et al. 1996; Bando et al. 2000; Kawashima et al. 2000; Loehe et al. 2004).

Utilising a rat model of heterotopic heart transplantation, Baxter et al demonstrated that hearts preserved in UW solution supplemented with GTN enabled extended effective preservation of the heart to 16 hours, with grafted hearts continuing to beat at seven days after transplantation (Baxter et al. 1999a). This study also showed that left ventricular function (as reflected by left ventricular developed pressure, LVDP; maximum rate of left ventricular pressure rise, dP/dt; and peak rate of left ventricular pressure fall, -dP/dt) was significantly improved after 12 and 16 hours preservation in GTN-supplemented UW solution compared with storage in UW alone. In a later study also utilising a rat model of heterotopic heart transplantation, Baxter et al investigated GTN supplementation of Celsior solution in addition to UW (Baxter et al. 2001). In this study, they showed that the addition of GTN to Celsior also extended effective

- 81 - preservation of the heart to 16 hours, with similar survival and function to grafts stored in GTN-supplemented UW. However, in contrast to other studies (Drummond et al. 1997; Mohara et al. 1999), they demonstrated that Celsior was not as effective as UW for rat heart preservation as reflected in graft survival at seven days post-transplantation. In another study using a heterotopic rat cardiac transplant model, Pinsky et al at Columbia University demonstrated that the addition of GTN to a balanced salt preservation solution (Ringer’s lactate) also enhanced graft survival in a time- and dose- dependant manner, with 92% of GTN-supplemented hearts surviving 12 hours of preservation compared with 17% in hearts without GTN (Pinsky et al. 1994b). Similar preservation results were seen in this study when UW was used.

Further work at Columbia University investigated the use of a novel storage solution supplemented with a cAMP analogue (dibutyryl cAMP) and GTN, and was called Columbia University Solution (CU) (Oz et al. 1993). In both a heterotopic rat heart transplant model and an orthotopic baboon cardiac transplantation model, Oz et al demonstrated that the use of CU to preserve hearts for 24 to 28 hours facilitated superior preservation of the heart compared with UW or Ringer’s lactate solution.

In a large animal (canine) model of orthotopic heart transplantation, donor hearts preserved for 24 hours in UW solution supplemented with GTN had significantly superior parameters of systolic function (as evaluated by pressure-volume relations measured with a conductance catheter) compared with those without GTN (Tanoue et al. 1999). Whilst the use of GTN was associated with a higher serum lipid peroxide level compared with controls, reflecting an increase in the detrimental effects of lipid peroxidation, the overall effect of GTN was cardioprotective as demonstrated by the ventricular functional parameters. Interestingly, all transplanted canine hearts (with or without GTN supplemented preservation) were successfully weaned off cardiopulmonary bypass support.

As a result of the studies into the role of cariporide and GTN in myocardial preservation in transplantation, Gao et al investigated the use of GTN and cariporide alone and in combination in myocardial preservation utilising an isolated working rat heart model (Gao et al. 2005). Rat hearts were preserved under hypothermic conditions for either 6 hours or 10 hours, and were treated with various combinations of GTN, cariporide and

- 82 - + glibenclamide (a specific inhibitor of mitochondrial K ATP) in Celsior solution in a total of 12 experimental groups. In summary, the findings from this study were: 1) the use of either GTN or cariporide in Celsior significantly improved preservation of hearts for six hours but not for 10 hours; 2) the protective effects of GTN were inhibited by pre- treatment of the heart with glibenclamide; 3) Celsior supplemented with both GTN and cariporide resulted in viable recovery of cardiac function after 10 hours storage; 4) pre- treatment of isolated rat hearts with cariporide produced only a small incremental improvement in preservation compared with cariporide alone- or combined cariporide and GTN-supplementation of Celsior preservation solution; and 5) pre-treatment of rat hearts with GTN did not improve recovery of cardiac function beyond what was achieved by adding GTN to the storage solution. In essence the authors concluded that GTN and/or cariporide supplemented Celsior provided superior recovery of cardiac function after hypothermic storage compared with Celsior alone, and that combined supplementation of Celsior with GTN and cariporide extended safe and effective preservation of the rat heart to 10 hours.

1.6 EXPERIMENTAL MODELS IN CARDIAC TRANSPLANTATION RESEARCH

1.6.1 The Use Of A Porcine Model Of The Brain-Dead Donor And Orthotopic Cardiac Transplantation

The porcine model used in the studies reported in this thesis was developed by researchers in the Transplant Program at The Victor Chang Cardiac Research Institute (Ryan et al. 2000). Traditionally, dogs have been the most commonly used laboratory animal in large animal surgical research and teaching (Swindle 1984). In the last few decades, however, this tradition has changed with increasing use of swine (Sus scrofa) in experimental surgical protocols. This increase has been due to a variety of factors, including decreased availability of dogs due to legislative changes, public concern with the use of companion animals in research, economic factors, and the identification of anatomical and physiological similarities between pigs and humans (Swindle et al. 1988).

- 83 - The cardiovascular system of swine is very similar to that of humans. Anatomically, the heart of the pig is similar to humans with a notable exception being the presence of the left azygos (hemiazygos) vein, which drains the intercostal system directly into the coronary sinus (Swindle et al. 1986). The distribution of the coronary blood supply, the blood supply to the conduction system, the histological appearance of the myocardium and the wound-healing characteristics of the heart are almost analogous to the situation in humans (Swindle et al. 1988). The porcine coronary system is similar to 90% of the human population in terms of anatomy and function (Swindle et al. 1998). There are no pre-existing collateral vessels in the porcine myocardium, although it does have the ability to develop collateral circulation following infarction, similar to what is seen in humans but not seen in dogs or ruminants. As a result, porcine myocardium is relatively intolerant of ischaemic preservation as is human myocardium (White et al. 1986; Swindle et al. 1988; Swindle et al. 1998). Unlike pigs, dogs already have a collateral circulation in place and ruminants do not develop a collateral supply at all (Swindle et al. 1986). This extensive canine collateral blood supply therefore enables canine hearts to be much more resistant to myocardial ischaemia compared with humans, thereby making it a poor model for studies into ischaemia-reperfusion injury. This is evidenced by the fact that in a study by Tanoue et al, eight out of eight canine hearts that were preserved in unsupplemented UW solution for 24 hours were able to be transplanted orthotopically and weaned off cardiopulmonary bypass (Tanoue et al. 1999). Similarly, in a study by Kim et al, three out of five canine donor hearts stored in an unsupplemented hyperkalaemic crystalloid solution for 24 hours were transplanted orthotopically and were able to be weaned off cardiopulmonary bypass support (Kim et al. 1998a). In a study assessing left ventricular contractile dysfunction in a porcine model of orthotopic heart transplantation, Ryan et al also concluded that the porcine heart was more susceptible to ischaemia-reperfusion injury than the canine heart (Ryan et al. 2002).

Haemodynamically, swine have been demonstrated to have similar cardiac function to humans, although there are variations between breeds and age that need to be taken into consideration when interpreting results (Swindle et al. 1998). In terms of the abdominal organs, the swine liver and pancreas are functionally similar to humans, as are the splanchnic blood flow characteristics. The kidneys of the pig are also very similar to

- 84 - humans in terms of anatomy and function, more so than most other animal species including primates (Swindle et al. 1988; Swindle et al. 1998).

Whilst there are many anatomical and functional similarities between swine and humans, there are a number of physiological differences between the species (Hannon et al. 1990). Despite this, the porcine model has much more significance and applicability to human transplantation compared with other animal models and isolated working heart models. This is evidenced by the extensive use of pigs in research into areas such as ischaemic reperfusion injury (Fedalen et al. 2003; Kristo et al. 2004), the pathophysiology of the brain-dead organ donor and the testing of hormone-based treatments (Novitzky et al. 1987d; Cooper et al. 1989; Meyers et al. 1993), and orthotopic heart transplantation (Wicomb et al. 1986a; Ryan et al. 2000; Ryan et al. 2003a; Martin et al. 2003b; Ryan et al. 2003b; Ryan et al. 2003c; Ryan et al. 2003d; Ryan et al. 2003e).

1.6.2 The Use Of The Preload Recruitable Stroke Work (PRSW) Relationship To Analyse Cardiac Function

The PRSW relationship was originally described in 1985 by Glower and colleagues, as a result of re-examining the Frank-Starling relationship between left ventricular stroke work and preload (Glower et al. 1985). At that time, quantification of cardiac function was difficult, with pre-existing indices (such as EF, cardiac output, end systolic pressure-volume relationship and the maximum of the time derivative of ventricular pressure, dP/dtmax) being non-linear, difficult to quantify, preload- or afterload- dependant, and reliant on extensive assumptions. The PRSW relationship is the relationship between ventricular stroke work and end-diastolic volume, examples of which can be seen in Figure 3.3. This was found to be a highly linear and reproducible index of contractile state and has since been validated in a number of studies (Glower et al. 1985; Feneley et al. 1992; Karunanithi et al. 1992; Takeuchi et al. 1992; Karunanithi et al. 2000). In addition, the slope of the relationship was found to be sensitive to alterations of the contractile state but was insensitive to changes in heart rate, preload and afterload.

- 85 - Acute ischaemia causes a significant rightward shift in the PRSW relationship and is manifested by an increase in the volume (x) axis intercept of the relationship (Vw,epi) (Ryan et al. 2002). This is known as the “creep” phenomenon and may persist after reperfusion (Glower et al. 1988). In contrast, the slope of the relationship (Mw) is reduced during ischaemia but may return to normal or even be increased after reperfusion (Glower et al. 1988). The simultaneous changes in Mw and Vw,epi, especially in opposing directions, make the interpretation of changes in the PRSW relationship difficult and necessitates the use of multiple linear regression techniques (described in further detail in Section 2.7.4) (Glantz et al. 2001).

1.7 AIMS OF THE RESEARCH REPORTED IN THIS THESIS

The overall objective of the research described in this thesis was to develop a donor management and organ preservation strategy that optimises the quality and recovery of the transplanted heart after prolonged (14 hours) hypothermic ischaemic storage. The treatments tested as part of this strategy were designed to ameliorate the negative effects of brain death on the donor and their organs, and to also prevent, and possibly reverse, the damage caused by ischaemia-reperfusion injury that occurs during the transplantation process. Experiments were conducted to examine the effectiveness of hormone treatments in the brain-dead organ donor and the potential roles for cariporide and GTN to prevent ischaemia-reperfusion injury during organ preservation and subsequent transplantation. It was anticipated that the results of this research would have direct applicability to human solid organ transplantation, and would provide the experimental evidence to support human clinical trials of the donor management and organ preservation strategy developed and investigated in this thesis.

The specific aims of the project were:

1) To examine the effects of hormone resuscitation on the brain-dead cardiac donor. Of specific interest were the effects of hormonal treatments on the: a. haemodynamic status of the donor, b. inotrope requirements to maintain haemodynamics in the donor, and c. contractile function of the heart.

- 86 - 2) To examine the effects of noradrenaline on the brain-dead cardiac donor. The dose of noradrenaline required to maintain haemodynamics was examined, as were the effects of noradrenaline on the: a. haemodynamic status of the donor, and b. contractile function of the heart.

3) To examine the effects of hormone resuscitation in the brain-dead multi- organ donor on the quality of transplantable solid organs: the lungs, liver, kidneys and pancreas.

4) To compare the effects of hormone resuscitation with a noradrenaline-based protocol and with an intravenous fluid-based protocol (without inotropic/vasopressor support) on haemodynamics in the brain-dead multi- organ donor and on the quality of transplantable solid organs (i.e. the lungs, liver, kidneys and pancreas).

5) To assess the effects of long-term (14 hours) preservation strategies incorporating cariporide and GTN in heart transplantation on: a. the ability to facilitate weaning from cardiopulmonary bypass support after transplantation, b. contractile function of the heart after transplantation, c. the haemodynamic status in the recipient after transplantation, and d. coronary arterial blood flow (left anterior descending coronary artery) and troponin I release.

The aims described above were addressed in a series of three studies described in Chapters 3, 4 and 5. The first study (Chapter 3) was designed to address aims 1 and 2, and compared the use of hormone resuscitation versus noradrenaline-only resuscitation in the brain-dead cardiac donor. Once the effects of hormone resuscitation and noradrenaline resuscitation on the donor and its heart were established, the second study (Chapter 4) was conducted to address aims 3 and 4. This study compared the effects of hormone resuscitation with noradrenaline resuscitation and with intravenous fluid resuscitation on the brain-dead donor. It specifically investigated the effects on the heart, lungs, liver, kidneys and pancreas. The optimal donor management protocol

- 87 - determined from the results of the first and second studies was then employed in the orthotopic cardiac transplantation model utilised in the third study. This study (Chapter 5) was designed to address aim 5, and examined the use of cariporide and GTN as part of a long-term preservation strategy used in cardiac transplantation.

- 88 -

CHAPTER 2

EXPERIMENTAL METHODS

- 89 - CHAPTER 2

EXPERIMENTAL METHODS

2.1 INTRODUCTION

The animal models, experimental procedures and protocols described in this thesis were approved by the Garvan Institute of Medical Research / St. Vincent’s Hospital Animal Ethics Committee (formerly known as the Animal Experimentation Ethics Committee). Experiments were conducted in accordance with the National Health and Medical Research Council (NHMRC) Australian Code of Practice for the Care and Use of Animals for Scientific Purposes (NHMRC 2004). The porcine model of the brain-dead cardiac donor and orthotopic cardiac allograft transplantation used in this research was based on the methods previously developed in the Transplant Program at The Victor Chang Cardiac Research Institute (Ryan et al. 2000; Ryan et al. 2002). The porcine model for the brain-dead multi-organ (thoracic and abdominal) donor was developed during the project for the study described in Chapter 4 and was based on the cardiac donor model previously mentioned.

2.2 EXPERIMENTAL DESIGN

Three studies are reported in this thesis. The first two studies (Chapters 3 and 4) examine the use of hormone resuscitation in the management of the brain-dead organ donor, and the effects these treatments have on the donor and on their heart, lungs, liver, kidney and pancreas. The experiments for these studies utilised a porcine model of the brain-dead cardiac and multi-organ donor. The final study (Chapter 5) examines the development of a novel cardiac allograft preservation strategy utilising cariporide and glyceryl trinitrate (GTN). These studies utilised the porcine model of orthotopic cardiac allograft transplantation. Power calculations were undertaken to determine the number of animals required for each experimental group in these studies. These numbers were similar to previous experiments undertaken at the Transplant Program at The Victor Chang Cardiac Research Institute (Ryan et al. 2002; Ryan et al. 2003a; Ryan et al. 2003b; Ryan et al. 2003c; Ryan et al. 2003d; Ryan et al. 2003e).

- 90 -

For the donor management studies, donor animals were anaesthetised, baseline data were acquired and then brain death was induced. These donor animals were managed for a total of six hours. Depending on the study, treatments to be tested were commenced at either one hour after brain death induction or at three hours after brain death induction. Data were collected at regular intervals during the experiment and later analysed to assess the effects of various treatments on the donor animal and their organs.

For the transplant study examining organ preservation strategies, animals were obtained in pairs with the heavier of the pair designated as the donor animal. After induction of anaesthesia, baseline data were collected and brain death induced. These donors were also managed for a total of six hours and were treated with the optimal management regimen determined from the earlier donor management studies (Chapter 3 and 4). Data were collected at regular intervals. After six hours management, the donor heart was arrested and explanted. It was then stored under conditions according to the study group. The heart was later transplanted orthotopically into the recipient pig and was exposed to a total ischaemic time of approximately 14 hours. The transplanted heart was then reperfused and further data were collected post-reperfusion. Attempts were made to wean the recipient off cardiopulmonary bypass support and if successful, animals were monitored for a further three hours and additional data collected. If initial weaning was unsuccessful, further attempts at weaning were made at hourly intervals up to three hours post-reperfusion. If animals could not be weaned, they were deemed as unsuccessful transplants. The primary outcome measures in the transplantation studies were successful weaning from cardiopulmonary bypass, left ventricular contractile function and haemodynamic indices. Left anterior descending coronary artery (LAD) flow and troponin I were also assessed.

2.3 ANIMALS

Three different breeds of pigs (Sus scrofa) were used in the experiments described in this thesis: 1) Australian Minipigs; 2) Landrace Pigs; and 3) Westran Pigs. Animals were purchased from the Biological Resources Centre at the University of New South Wales (BRC) and most were 40-50 kg in body weight. Animals were obtained by the

- 91 - BRC from commercial suppliers at least seven days prior to the experiments and were housed at their facilities in Kensington. Whilst at the BRC, animals were weighed (providing the baseline weights for the experiments) and given prophylactic antibiotics (Kefvet (cephalexin) 1000 mg per orally daily) to minimise the risk of illnesses, including transport stress-induced illnesses. Westran pigs were utilised for the transplantation studies as these animals were from a highly inbred colony in which graft immunotolerance has been demonstrated (O'Connell et al. 2005). Landrace pigs used for transplant studies were also from a single inbred colony.

On the day of experiments, individual animals were transported to the Cardiac Mechanics Research Laboratory large animal operating facility at the Victor Chang Cardiac Research Institute by the BRC. Animals were transported in a custom built cage (Figure 2.1) in an air-conditioned van with a padded floor to minimise transport stress.

Figure 2.1: Animal transportation crate. a) External; b) Internal.

2.4 PERSONNEL

The experiments carried out for the studies described in this thesis were dependent on numerous people. The roles of surgeon, surgical assistant, porcine anaesthetist, perfusionist, physiologist and data collector were undertaken by a variety of clinicians and scientists at various times during the project. A number of staff fulfilled roles that were unfamiliar to them and had to be trained up for these roles. They acquired these skills rapidly and proficiently. Staff from The Victor Chang Cardiac Research Institute, St. Vincent’s Hospital and the Collaborative Transplant Research Group at Sydney

- 92 - University who were involved in the project at various times over the four years of the project were: Prof Peter Macdonald, Dr Michael Wilson, Dr Mark Hicks, Dr Ling Gao, Ms Sarah Garlick, Mr Steve Faddy, Dr Aisling McMahon, Mr Scott Kesteven, Dr Graham Stewart, Mr Peter Tran, Dr Paul Jansz, Dr Jon Ryan, Dr Jerome Laurence, Mr Jair Kwan, Mr Andrew Dinale and Mr Jonathan Cropper.

2.5 DONOR MANAGEMENT

2.5.1 Donor Anaesthetic Management And Monitoring

Within 10 minutes of delivery to the Cardiac Mechanics Research Laboratory large animal operating facility, animals were pre-medicated with an intramuscular injection of ketamine (Parnell Laboratories (Aust) Pty Ltd, Australia) 10 mg/kg, midazolam (Pfizer Australia Pty Ltd, Australia) 1 mg/kg and atropine (Pharmacia Australia Pty Ltd, Australia) 0.05 mg/kg. The site of injection was either in the trapezius muscle mass in the dorsal aspect of the neck, caudal to the ear, or in the gluteal muscle mass (Figure 2.2). In the majority of cases, animals were pre-medicated immediately after arrival at the facility. After the pre-medication was given, the room was darkened, a sheet was placed over the cage and the noise in the room minimised to reduce sensory stimulation, in order to minimise animal stress. In a small number of cases where the experiment was not commenced immediately after animal arrival, half doses of ketamine and midazolam were administered if animals were distressed. Sensory stimulation of the animals was minimised as described above and the subsequent doses of pre-general anaesthesia ketamine and midazolam were adjusted according to the previous doses and level of consciousness.

- 93 -

Figure 2.2: Intramuscular injection site for pre-medication (circled) in: a) the trapezius muscle (posterior neck); and b) the gluteal muscle.

Once adequately sedated, animals were removed from the holding cage and placed prone on the pre-anaesthetic room table. A pulse oximeter (Nellcor N180 Pulse

Oximeter, J. Morita Corporation ) was attached to the ear, and O2 saturation and heart rate monitored. The hair on the dorsum of the ear was shaved and an 18 Ga or 20 Ga intravenous (IV) cannula (BD Insyte, Becton Dickinson Infusion Therapy Systems Inc, USA) was inserted (Figure 2.3). If veins were not obvious in the dorsum of the ear, a towel soaked in warm water was wrapped around the ear to cause vasodilatation to facilitate cannulation.

Figure 2.3: Intravenous cannulation of the porcine dorsal ear vein with a 20 G cannula.

Once IV access was obtained, general anaesthesia was induced using thiopentone (Jurox Pty Ltd, Australia) 2-10 mg/kg IV given in boluses until adequately anaesthetised for intubation. Animals were pre-oxygenated with 100% oxygen via a nose cone and were

- 94 - then intubated with a standard endotracheal tube (Hi-Contour Tracheal Tube, Mallinckrodt Medical, Ireland) size 6.0-8.0. Unlike humans, pigs have a relatively straight but long pharynx from the snout to the vocal cords. As a result, intubation was achieved using a laryngoscope with a straight blade that had an extension welded to it giving the blade a total length of 24-30 cm (Figure 2.4). To facilitate visualisation of the vocal cords the snout was elevated and traction applied to the tongue (Figure 2.5). Once the cords were visualised, intubation was achieved by “rail-roading” an endotracheal tube over a flexible introducer (Portex Tracheal Tube Guide 10 Ch, Portex UK) that had been passed through the vocal cords. Alternatively, an endotracheal tube was placed directly through the cords with a stiff malleable 4.7 mm intubating stylet placed in the tube, which was then removed after successful intubation. Appropriate tube placement was then confirmed by auscultation of both lung fields. Once successfully intubated, the animal was manually ventilated with 100% oxygen and isoflurane (Forthane, Abbott Australasia Pty Ltd Australia) 2-4% inhaled gas.

Figure 2.4: Laryngoscope with blade extension for porcine intubation.

- 95 -

Figure 2.5: Porcine laryngoscopy: visualising the vocal cords for intubation.

Whilst the pig was still prone, hair was shaved from the contralateral ear (for further IV cannulation and to optimise pulse oximetry signal detection), the right fronto-parietal region of the head (burr-hole site), the lower back (diathermy return electrode site), the midline of the chest (median sternotomy site), the femoral regions (femoral arterial catheter site), the midline of the abdomen (laparotomy site) for studies involving the abdominal organs and all four legs and the left flank (for ECG electrodes). Additional 18 Ga or 20 Ga IV cannulae were placed in the ears so that there were at least two peripheral IV cannulae in situ, preferably one in each ear.

After intubation, IV cannulation and shaving, animals were transferred onto the operating theatre table (Figure 2.6) and placed in a supine position. A warming pad and diathermy return electrode plate was positioned underneath the animal. The animal was stabilised with 1 L bags of IV fluids placed on either side of the animal’s chest and it was secured to the operating table with cloth tape by tying the legs to the table.

- 96 -

Figure 2.6: Animal secured onto the operating theatre table and theatre set-up, including ventilator (a); monitoring equipment (including O2 pulse oximeter, Datex capnograph and SpaceLabs medical monitor) (b); infusion pumps delivering various treatments such as hormone resuscitation (c); and data acquisition equipment to capture data such as intra-ventricular pressures and cardiac dimensions from the ultrasonic pressure transducers and micromanometer-tipped catheters (d) (see text for further details regarding monitoring equipment).

Once secured to the operating table, endotracheal tube placement was re-checked by auscultation of the chest and the animal was placed on mechanical ventilation with an Ohio V5 Anaesthesia Ventilator (Ohio Medical Products, USA) or an ULCO Campbell Ventilator EV500 (ULCO Engineering Pty Ltd, Australia). Mechanical ventilation with 100% oxygen was commenced at a tidal volume of 100-150 mL/kg/min and at a respiratory rate of 10–15 breaths/min. These ventilatory parameters were adjusted according to arterial blood gas analyses and airway pressures. Depth of general anaesthesia was assessed at regular intervals by heart rate, blood pressure, jaw muscle tone and response to painful stimuli. Anaesthesia was maintained with isoflurane 1-4% inhaled gas and fentanyl (David Bull Laboratories, Australia) 100-150 μg IV boluses, titrated to the depth of anaesthesia required.

- 97 - Continuous physiological monitoring was maintained throughout each experiment. A three-limb electrocardiogram (ECG) was attached to monitor cardiac rhythm and heart rate. A femoral arterial catheter (20 Ga BD InSyte Cannula) was inserted (either directly percutaneously or via the Seldinger technique, or an open cut-down technique) for invasive arterial blood pressure monitoring and to obtain arterial blood for analysis. A

CO2 analyser (Datex Cardiocap or Datex Capnomac Ultima, Datex Instrumentation

Corp Finland) was connected to the ventilation tubing to monitor end-tidal expired CO2, and O2 saturation and heart rate were monitored with continuous pulse oximetry (as described above). An oesophageal temperature probe was inserted to monitor core body temperature. In addition, central venous pressure (CVP) was monitored continuously via an internal mammary central venous triple lumen catheter inserted after median sternotomy (as described in Section 2.5.3). Blood pressure, ECG, temperature and CVP were monitored using a SpaceLabs Medical Monitor (SpaceLabs Medical Inc, USA).

- Arterial blood analyses (consisting of pH, PaCO2, PaO2, HCO3 , base excess, O2 saturation, potassium, sodium, haematocrit, haemoglobin, TCO2 and glucose) were performed using an iSTAT Portable Clinical Analyzer with EG6+ cartridges (i-STAT Corporation, USA) and an Accu-Chek Active Blood Glucose Monitor (Roche Diagnostics GmbH, Germany). Analyses were performed after insertion of the femoral arterial catheter, immediately prior to induction of brain death, at hourly intervals after brain death induction and when clinically indicated.

Animals were kept hydrated with intravenous saline (0.9%) infused at 10 mL/kg in the first hour, followed by 5mL/kg/hr titrated to a CVP of 0-5 mmHg.

2.5.2 Preparation For The Induction Of Brain Death In The Donor Animal

The head and upper body of the animal was turned to the left to expose the right fronto- parietal region of the head and the right ear was taped down (Figure 2.7). A 4 cm incision was made lateral and parallel to the temporal ridge (a palpable bony ridge that marks the junction between the lateral and superior aspects of the skull), and the underlying tissue and muscle dissected down to the periosteum. A 15 mm burr hole was

- 98 - made in the skull using a bit and brace drill with perforator and burr drill bits to expose the dura mater (Figure 2.7). A cruciate incision was then made in the dura mater, exposing the cerebral cortex. A 14 Ch/Fr 4.7 mm Foley catheter (Bard SDN. BHD., Malaysia) was then inserted into the subdural space until both the catheter tip and balloon were completely within the skull. Blood pressure and heart rate were monitored during this procedure and if either increased, advancement of the catheter was stopped, withdrawn and re-positioned to ensure that there was no significant, if any, injury to the brain during catheter insertion.

Once the Foley catheter was in place, a cotton gauze swab was packed tightly in and around the catheter in the burr-hole to prevent herniation through the burr-hole, as herniation would cause a subsequent fall in intracranial pressure during induction of brain death. The skin incision was then closed with a 2-0 silk suture (Ethicon Inc, USA) and the catheter was secured to the skin with the suture to prevent dislodgement (Figure 2.7). An artery forcep was used to clamp the lumen of the catheter proximal to the balloon injection port to minimise herniation of the cerebrum through the catheter during and after brain death induction. The catheter balloon was left deflated at this stage. The animal’s head and upper body was then returned to the supine position and secured as before in preparation for median sternotomy.

- 99 -

Figure 2.7: Surgery for induction of brain death. a) Burr hole in the right fronto-parietal region of the skull; b) underlying dura mater (arrow); and c) Foley catheter in situ.

- 100 - 2.5.3 Donor Animal Surgery And Cardiac Instrumentation For Data Acquisition

The porcine heart is very sensitive to mechanical stimulation from sources such as the oscillating saw, suturing and instrumentation, all of which can induce ventricular fibrillation. Therefore, lignocaine (AstraZeneca Pty Ltd, Australia) 1 mg/kg IV was given to the animal prior to sternotomy for arrhythmia prophylaxis. Any further ventricular arrhythmias (such as ventricular fibrillation or low output ventricular tachycardia) encountered during surgery were treated with internal DC defibrillation (10-30 J) and additional boluses of lignocaine 0.5-1.0 mg/kg IV.

An electric oscillating saw was used for the median sternotomy. This was used instead of a reciprocating saw because of the shape and thickness of the superior part of the porcine sternum (Figure 2.8), and because the pericardium was frequently closely adherent to the posterior surface of the sternum. In the pig, the innominate vein is located immediately posterior to the superior extension of the sternum. This vein was mobilised and isolated to prevent injury during median sternotomy.

After sternotomy, the right, left and inferior pleural spaces were opened. In many animals, the thymus was prominent and had to be excised to improve exposure of the pericardium and heart. The pericardium was then opened with an inverted ‘T’ incision and the edges of the pericardium held open with 2-0 silk stay sutures to expose the heart. To improve access to the posterior aspects of the heart during cardiac instrumentation, traction was applied to the stay sutures to elevate the heart partially out of the pericardial well created by the sutures (Figure 2.9). The heart was inspected for any anatomical abnormalities and was palpated to exclude any thrills or coronary arterial abnormalities.

The internal mammary vein (usually the right) was mobilised and a 30 cm 7 Fr triple lumen central venous catheter (Arrow International Inc, USA) was inserted (Figure 2.9). This catheter was for central venous access for IV fluids and drug infusions, and was connected to the SpaceLabs Medical Monitor to continuously monitor CVP.

- 101 -

Figure 2.8: The porcine sternum. a) Median sternotomy, and b) superior part of sternum. Note the shape and thickness of the sternum (arrow), often 3-5 cm in maximal depth at the superior aspect of the sternum.

- 102 - b

Ao RAA RVOT

c a a dd LAA RVV e

c a a LVLV

d

c

Figure 2.9: The instrumented porcine heart. The heart has been exposed by silk stay sutures to the pericardial edges, creating a pericardial well (a); a right internal mammary vein central venous catheter is in situ (b); ultrasonic dimension transducers have been sewn to the epicardium (c); micromanometer-tipped catheters have been placed in both the left and right ventricles (d) and flow probe around the left anterior descending coronary artery (e). Ao = aorta; RVOT= right ventricular outflow tract; RV = right ventricle; LV = left ventricle; LAD = left anterior descending coronary artery; LAA = left atrial appendage; and RAA = right atrial appendage.

-103- The inferior vena cava (IVC) in the inferior pleural space (between the diaphragm and the pericardium) was mobilised, and a 0.318 cm wide 75 cm long polyester tape (Sherwood Davis & Geck, USA) was placed around it and drawn through a rubber snare.

Six two- or three-millimetre ultrasonic dimension transducers (Sonometrics Corp, Canada) (Figures 2.9 and 2.10) were sewn onto the epicardium with 4-0 silk sutures (Ethicon Inc, USA) to measure the base-apex major axis, anterior-posterior minor axis, and the left ventricular free wall-right ventricular free wall minor axis diameters of the heart. These transducers were sewn on: 1) the anterior aspect of the heart mid-way along the left anterior descending coronary artery (LAD) on the left ventricular surface; 2) the posterior aspect of the heart midway along the posterior descending branch of the right coronary artery (PDA) directly opposite the anterior transducer on the left ventricular surface; 3) the left ventricular free wall, midway between the LAD and the PDA, and midway between the base and apex; 4) the right ventricular free wall, directly opposite the left ventricular free wall transducer; 5) the apex of heart; and 6) the base of the heart, posterior to the aorta on the roof of the left atrium. The transducers were then connected to a Sonometrics Digital Ultrasonic Measurement System TRX Series 10 (Sonometrics Corp, Canada).

Next, a 3-0 Prolene (Ethicon Inc, USA) purse-string was placed in the anterior surface of the left ventricle near the apex and a full thickness puncture incision was made in the middle of the purse-string. A 5 Fr 70 cm micromanometer-tipped (pressure) catheter (Millar MPC 500, Millar Instruments Inc, USA) was placed through the incision into the left ventricular cavity and secured with the purse-string (Figures 2.9 and 2.10). This procedure was repeated with a pressure catheter placed in the right ventricular cavity. These catheters were connected to a Millar Transducer Control Unit Model TCB-600 (Millar Instruments Inc, USA) to measure intraventricular pressures.

The LAD was dissected and mobilised distal to the bifurcation of the left main coronary artery. A 2 mm flow probe was then placed around the LAD and connected to a T208 Transonic Volume Flow Meter (Transonic Systems Inc, USA) (Figures 2.9, 2.10 and 2.11). Data from the dimension transducers, pressure catheters and the flow probe were digitised at 200 Hz.

- 104 -

Figure 2.10: Ultrasonic dimension transducer (a), flow probe (b) and micromanometer- tipped catheter (c).

Figure 2.11: Left anterior descending coronary artery flow probe in situ (a). Note also the ultrasonic dimension transducer sewn in situ (b).

In donor animals used for transplant studies (as described in Chapter 5), further surgery and mobilisation of the heart was required prior to attachment of the ultrasonic dimension transducers, insertion of the pressure catheters and the attachment of the LAD flow probe. The left azygos vein is a constant tributary of the coronary sinus in the pig (Figure 2.12) and was ligated outside the pericardium in the left pleural space at the

- 105 - pulmonary hilum. This was done to prevent blood flooding into the pericardial well during allograft explantation. Next, the ascending aorta was dissected off the pulmonary trunk and mobilised, and a polyester tape was placed around the aorta. The superior vena cava (SVC) was mobilised below the level of the right azygos vein and two 2-0 silk ties were placed around SVC, but not tied.

In the transplant studies, the dimension transducers and the pressure catheters were left in situ during hypothermic storage. The flow probe was removed prior to allograft explantation and subsequent storage, and re-positioned around the LAD after transplantation.

Figure 2.12: The left azygos vein (a).

2.5.4 Induction Of Brain Death

After acquisition of baseline (pre-brain death) data, brain death was induced by inflating the Foley catheter balloon in the subdural space. The balloon was inflated with water in 3 mL increments every 3 seconds up to a total of 24 mL. This increased the intra-cranial pressure, thereby causing an autonomic storm similar to that seen clinically. The changes in heart rate and blood pressure observed in the porcine model was similar to that seen in humans during brain death as reported by Ryan and colleagues (Ryan et al. 2000).

Anaesthesia was ceased 15 minutes after commencing balloon inflation to allow clinical confirmation of brain death. Brain death was confirmed by the typical haemodynamic

- 106 - changes of brain death, the absence of responses to painful stimuli, the absence of jaw muscle tone, and the absence of pupillary and corneal reflexes.

2.5.5 Donor Management

Donor animals were managed according to the individual protocols described in the relevant chapters. Following induction of brain death all donor animals were managed for six hours. Donor management procedures common to all donor animals are described below.

The target mean arterial blood pressure (MAP) for donor animals was 60-70 mmHg. Depending on the study and experimental group, MAP was maintained with IV fluids, a noradrenaline infusion and/or hormone resuscitation. Noradrenaline infusions for supporting blood pressure were prepared using 2 mg noradrenaline (Levophed, Abbott Australasia Pty Ltd Australia) mixed in 100 mL 5% glucose in sterile water. Intravenous noradrenaline infusions were generally commenced at a rate of 10 mL/hour and then titrated to MAP 60-70 mmHg. In animals dependant on high doses of noradrenaline to support MAP, infusions were prepared at double, triple and sometimes quadruple strength (i.e. 4, 6 and 8 mg noradrenaline in 100 mL 5% glucose in sterile water). At all times during donor management, attempts were made to reduce noradrenaline doses whilst at the same time maintaining MAP. In some animals, the noradrenaline infusion was weaned off.

2.5.6 Hormone Resuscitation Protocol

The hormone resuscitation (HR) protocol used in the studies described in this thesis is shown in Table 2.1. This protocol was based on the cardiac recommendations made from the consensus conference on “Maximising Use of Organs Recovered From the Cadaver Donor” (Rosengard et al. 2002; Zaroff et al. 2002).

- 107 - Table 2.1: Hormone resuscitation protocol

Hormone Dose

Methylprednisolone 15 mg/kg intravenous bolus

Triiodothyronine 4 μg intravenous bolus, followed by 4 μg/hour intravenous infusion

Vasopressin 0.5–4.0 units/hour titrated to mean arterial blood pressure 60–70 mmHg

Insulin Titrated to blood sugar level 6–10 mmol/L at minimum 1 unit/hr intravenously

As part of the HR protocol, vasopressin was increased incrementally to a maximum of 4 units/hr to maintain MAP. If MAP was stable at 60-70 mmHg, attempts were made to wean the animals off vasopressin.

2.5.7 Cardiac Allograft Explantation And Preservation

Donor animals were prepared for cardiac explantation after final post-brain death data were acquired at six hours after brain death induction. Approximately 15 minutes before aortic cross clamping and subsequent explantation, 10 000 IU heparin was given intravenously. A 3-0 Prolene purse-string was placed in the adventitia of the ascending aorta approximately 1-2 cm proximal to the brachiocephalic trunk and then drawn through a rubber snare. A 12 Ga (9 Fr) cardioplegia aortic root cannula with an introducer needle (Medtronic Inc, USA) was inserted into the aortic lumen through the centre of the purse-string and secured with the snare. The cannula was then de-aired, clamped and connected to an IV solution infusion set which was connected to a one litre bag of pre-prepared cold cardioplegia.

The cardioplegia used in the transplantation experiments was Celsior (Genzyme- SangStat, France), with or without supplementation depending on the study protocol (as described in Chapter 5), and was prepared whilst the animal was being prepared for

- 108 - cardiac explantation. Once prepared, the bag of cardioplegia was connected to an IV solution infusion set and primed, and was then placed in a pressure infuser bag, which was inflated to 250-300 mmHg.

Once the cardioplegia was in place and ready to be infused into the heart, the internal mammary vein central venous catheter was withdrawn. The SVC was then ligated with the previously placed 2-0 silk ties, and the IVC was snared down and occluded by the previously placed polyester tape and rubber snare. An aortic cross clamp was applied between the cardioplegia aortic root cannula and the brachiocephalic trunk, and the heart was arrested with 900 mL of cold (1-4°C) cardioplegia delivered under pressure. Perfusion of the coronary vessels was confirmed by palpation of the aortic root pressure, and by the heart becoming cold and pale. The bag of cardioplegia was hung on a set of scales and weighed during infusion to ascertain when 900 mL (i.e. 900 gm) had been infused. Whilst the cardioplegia was being infused, the IVC in the inferior pleural space and one of the left pulmonary veins in the left pleural space were incised to decompress the right and left sides of the heart, respectively. Blood from these incised vessels was kept out of the pericardial space to prevent it from limiting the cooling of the heart and crushed ice was placed in the three pleural spaces to aid the cooling process.

Once the cardioplegia had been infused and the heart arrested, the aortic root cannula was removed and the purse-string suture tied. Explantation of the heart was commenced by lifting the heart anteriorly to expose the pericardial reflections around the inferior pulmonary veins and the IVC. The IVC was transected at the pericardial reflection, followed by the inferior pulmonary veins. The dissection was continued superiorly along the posterior aspect of the left atrium, and the superior pulmonary veins and the pulmonary arteries were divided as they were encountered. The heart was then placed back into its anatomical position and the SVC was divided between the two previously tied 2-0 silk ties. The superior aspect of the right atrium was then mobilised, followed by transection of the distal aortic arch beyond the origin of left subclavian artery. Finally, the pulmonary trunk immediately before its bifurcation was transected to complete explantation of the heart.

- 109 - The explanted heart was placed into a sealable plastic bag containing 100 mL of Celsior (supplemented or unsupplemented, depending on the experimental protocol), which was then placed into a plastic tray filled with crushed ice. The heart was inspected for any abnormalities and prepared for later transplantation. The pulmonary trunk was dissected from the aorta towards its origin from the heart and any extraneous tissue was excised. The IVC was identified and the right atrium was partially opened from the IVC to inspect the atrial septum for a patent foramen ovale (PFO). If present, the PFO was closed with a continuous 6-0 Prolene (Ethicon Inc, USA) suture. Next, the right atrium was checked to ensure that it was intact and the SVC silk tie was inspected to ensure it was securely in place. If the right atrium was not intact, it was either repaired with a continuous 3-0 Prolene suture or was incorporated into the right atrial sewing cuff at the time of transplantation. If the SVC was not securely closed, another 2-0 silk ligature was applied or a continuous 3-0 Prolene suture was used to close it. Finally, the pulmonary veins were identified and the left atrium was opened by cutting between the four pulmonary veins, thereby creating a left atrial sewing cuff. Once preparation of the heart for transplantation was completed, the plastic bag containing the heart was de- aired, sealed and submerged into an esky filled with crushed ice (Figure 2.13). The esky was then stored in a refrigerated cool room at 1-4°C.

Figure 2.13: Explanted donor heart stored in a plastic bag of preservation solution, then placed in an esky filled with crushed ice.

- 110 - 2.6 RECIPIENT MANAGEMENT

2.6.1 Recipient Anaesthetic Management And Monitoring

Recipient animals were anaesthetised and monitored as described for the donor animal (Section 2.5.1). In addition, methylprednisolone 500 mg IV was given at anaesthetic induction and again at 15 minutes prior to cardiac reperfusion after transplantation. The recipient animal was given heparin 10 000 IU IV during surgery to prepare for institution of cardiopulmonary bypass support and additional bolus doses (5 000–10 000 IU IV) were given to achieve an activated clotting time (ACT) greater than 500 seconds. Additional medications were given to the recipient according to the experimental protocol.

Once the animal was placed on cardiopulmonary bypass support, mechanical ventilation was ceased. Isoflurane 1-3% was delivered to the animal to maintain anaesthesia via the bypass oxygenator. As with the donor animal, the level of anaesthesia was assessed clinically at regular intervals, and isoflurane and IV boluses of fentanyl were titrated to the depth of anaesthesia required. Arterial blood analysis was performed as described in Section 2.5.1 after arterial catheter insertion and at regular intervals whilst on cardiopulmonary bypass and after reperfusion. Blood analysis was also performed as clinically indicated.

2.6.2 Recipient Animal Surgery And Institution Of Cardiopulmonary Bypass Support

As previously described, lignocaine 1 mg/kg IV was given for ventricular arrhythmia prophylaxis prior to sternotomy and any further arrhythmias were treated with internal DC defibrillation and additional boluses of lignocaine. The recipient’s heart was exposed via a median sternotomy and prepared as described for the donor animal in Section 2.5.3. An internal mammary vein central venous catheter was inserted, the pleural spaces were opened, a pericardial well was created and the left azygos vein was ligated. The IVC was snared with polyester tape and a rubber tube in the inferior pleural space. In addition, the SVC was mobilised below the level of the right azygos vein and a polyester tape was passed around it and snared with a rubber tube. The ascending

- 111 - aorta was separated from the pulmonary trunk and a polyester tape placed around the aorta.

A cardiopulmonary bypass machine (Figure 2.14) was prepared, and the arterio-venous circuit (Cardio Research Pty Ltd, Australia) and oxygenator (Capiox SX18 Hollow Fibre Oxygenator, Terumo Corp USA) were primed with 2 000 mL of Hartmann’s solution, 500 mL of Haemaccel (Hoechst Marion Roussel Australia Pty Ltd, Australia) and 15 000 IU heparin. The arterio-venous line was then clamped and divided once it was de-aired. A plastic Yankuer suction instrument (sucker) was connected to the bypass machine suction line (pump sucker) and a plastic tipped vent was attached to the vent suction line of the bypass machine. Once the ACT was greater than 500 seconds, the pump sucker and the vent were used to scavenge blood from the operative field throughout the experiment, which was then pumped through a filter and into the oxygenator reservoir to be autotransfused back to the animal. Additional Hartmann’s solution was added to the bypass circuit if required, to maintain adequate volume in the oxygenator reservoir.

Figure 2.14: The cardiopulmonary bypass machine – roller pumps (a); oxygenator (b); and heater/cooler (c).

- 112 - The heart was prepared for connection to the cardiopulmonary bypass machine with cannulae in the aorta, SVC and IVC. A 3-0 Prolene purse-string was placed in the adventitia of the distal ascending aorta at the junction with the brachiocephalic trunk and was drawn through a rubber snare. A second 3-0 Prolene aortic purse-string was placed circumferentially around the first purse-string and snared. The aorta was then cannulated through the purse-strings with a 22 Fr wire-reinforced aortic cannula (EOPA Cannula, Medtronic Inc USA) and secured with both snares. The cannula was then de- aired, connected to the arterial line of the bypass circuit and secured to the animal with 2-0 silk sutures. Next, a 3-0 Prolene purse-string was placed in the SVC and drawn through a rubber snare. A 26 Fr single-stage straight venous cannula (Medtronic Inc USA) was inserted through an incision made in the middle of the SVC purse-string, secured with the snare, de-aired and clamped. A clamped 24 Fr single-stage right angle venous cannula (Medtronic Inc USA), along with the clamped SVC cannula, was connected to a Y-connector, which was then connected to the venous line of the bypass circuit. The clamp was then removed from the SVC cannula, to enable the animal to be placed on partial cardiopulmonary bypass support if necessary whilst inserting the IVC cannula. The IVC cannulation site was situated deep in the inferior pleural space (between the diaphragm and the pericardium), which made it more technically difficult to cannulate compared with the SVC cannula and at times was associated with significant bleeding and hypotension during the cannulation procedure. If the animal became hypotensive during the IVC cannulation (due to bleeding or manipulation of the IVC, leading to decreased venous return), fluid was infused into the animal via the aortic cannula or partial bypass was instituted. As with the SVC cannulation, a 3-0 Prolene purse-string was placed in the IVC and snared. The IVC was then cannulated with the 24 Fr right angle venous cannula. This was then secured and the clamp removed.

Once the aorta, SVC and IVC were cannulated, cardiopulmonary bypass support of the animal was commenced and mechanical ventilation was ceased. The polyester tapes around the SVC and IVC were snared down onto the cannulae, and the aorta was cross- clamped immediately below the aortic cannula to institute total cardiopulmonary bypass. The animal was actively cooled to 32°C via the bypass circuit.

- 113 - 2.6.3 Orthotopic Cardiac Allograft Transplantation

The donor heart was orthotopically transplanted into the recipient animal using the technique described by Lower and Shumway (Lower et al. 1960). The recipient cardiectomy was commenced by opening the right atrial appendage in the atrioventricular (AV) groove. The right ventricle was detached from the right atrium by extending the incision superiorly and inferiorly in the AV groove. The left atrium was opened at the superior aspect of the inter-atrial septum in the AV groove. The incision was then extended inferiorly and laterally along the AV groove to complete the separation of the left ventricle from the left atrium. The recipient cardiectomy was completed by transecting the aortic root and pulmonary trunk as close as possible to their emergence from the heart, in order to maximise their length for the transplant (Figure 2.15).

Figure 2.15: The pericardial cavity after recipient cardiectomy with atrial sewing cuffs prepared. SVC = superior vena cava venous cannula; IVC = inferior vena cava venous cannula; AoC = aortic arterial cannula; Ao = aorta; AXC = aortic cross clamp; PA = pulmonary artery; RA = right atrial cuff; LA = left atrial cuff.

- 114 - The recipient right and left atrial sewing cuffs were prepared next, with the left atrial appendage being excised and the right atrial appendage partially excised, and ensuring adequate atrial tissue to suture to for the anastomoses (Figure 2.15). The pericardium on the left side was then incised to improve exposure and to form a cradle for the donor heart to sit in to commence the left atrial anastomosis.

The donor heart was removed from storage, oriented and placed in the pericardial cavity in its normal anatomical position. It was then rolled out 180° towards the left pleural space onto a bed of gauze swabs soaked in ice slush. The left atrial anastomosis was performed using a continuous 3-0 double-ended Prolene suture (Ethicon Inc, USA), commencing midway along the lateral wall of the left atrium and continued inferiorly to the inferior aspect of the interatrial septum. The anastomosis was completed by changing to the other needle, and suturing superiorly along the lateral left atrial wall, across the left atrial roof to the interatrial septum, then inferiorly to meet the other suture end and tied. As the anastomosis was completed, the heart was rolled back into the pericardial cavity into its normal anatomical position. Once the left atrial anastomosis was completed, 200 mL of cold (1-4°C) Celsior solution was infused under pressure (250-300 mmHg) into the coronary arterial system via the aortic root.

The donor right atrial sewing cuff was prepared by opening the right atrium from the posterior aspect of the IVC, extending along the septum, then passing up towards the tip of the right atrial appendage. This technique of preparing the right atrial sewing cuff was used to preserve the sino-atrial node and the incision was extended to create a cuff to match the size of the recipient right atrial sewing cuff. The right atrial anastomosis was performed using a continuous 3-0 double-ended Prolene suture, commencing at the superior aspect of the atrial septum, and continued inferiorly down the septum and around the IVC. This anastomosis was completed by changing to the other needle, and suturing superiorly, around the top of the atrial cuff, then inferiorly to meet the other suture end and tied. At this point, another 200 mL of cold (1-4°C) Celsior solution was infused under pressure (250-300 mmHg) into the coronary arterial system via the aortic root.

Next, the donor and recipient pulmonary arteries were trimmed, and their edges lined up to ensure that there was adequate length of both donor and recipient arteries and that

- 115 - there was no kinking. The pulmonary arterial anastomosis was performed using a continuous 4-0 double-ended Prolene suture, commencing at the left lateral aspect of the pulmonary artery. The anastomosis was then continued towards and across the posterior wall, before progressing anteriorly to the anterior aspect of the artery. The anastomosis was completed by changing needles, and suturing anteriorly to meet the other suture end and tied. At this point, rewarming of the animal was commenced and the 15 minutes pre-reperfusion methylprednisolone dose (500 mg IV) was given. Finally, the donor and recipient aortas were trimmed and matched as with the pulmonary arteries, with the donor aorta trimmed to remove the 3-0 Prolene repair at the cardioplegia cannulation site. The aortic anastomosis was then performed in a similar manner to the pulmonary arterial anastomosis using a continuous 4-0 double-ended Prolene suture, except that once the aortic anastomosis was completed, the two suture ends were left untied. Once the donor heart had been exposed to a total (cold and warm) ischaemic time of 14 hours, the aortic cross-clamp was removed and the heart was reperfused. The aortic root was then de-aired through the suture line at the site of the aortic anastomosis, and the suture ends then tightened and tied.

At this point, the polyester snares around the SVC and the IVC were released and the right side of the heart was vented through the venous cannulae. Next, a stab incision was made in the apex of the left ventricle and a paediatric vent (connected to the vent suction line of the bypass machine) was inserted into the left ventricle to vent the left side of the heart.

Whilst the heart was being vented and re-warming was continuing, the bi-caval cannulation for bypass support was changed to a single right atrial cannula. A 3-0 Prolene purse-string was placed at the tip of the right atrial appendage, and a 30 Fr single-stage straight venous cannula (Medtronic Inc, USA) was inserted into the appendage and secured with a rubber snare. The cannula was then de-aired and clamped. Next, the SVC cannula was clamped and then removed, with the 3-0 Prolene purse-string tied. Any bleeding from the SVC cannulation site was then oversewn with a 3-0 Prolene suture. Once haemostasis was achieved, the cardiopulmonary bypass machine was stopped. The IVC cannula was then clamped and removed, and the IVC purse-string was snared down to close the cannulation site. The Y-connector was disconnected from the venous line of the bypass circuit and the right atrial cannula was

- 116 - connected in its place. Cardiopulmonary bypass support was then re-instituted, the IVC purse-string was tied, and any bleeding from the cannulation site was oversewn with 3-0 Prolene. Mechanical ventilation of the animal was then recommenced, the flow probe was re-positioned around the LAD and connected to the flow meter, the pressure catheters were re-connected and the ultrasonic dimension transducers re-connected to the Sonometrics system.

2.6.4 Management Of The Cardiac Transplant Recipient After Reperfusion And Weaning From Cardiopulmonary Bypass Support

Once the animal was at normal physiological temperature and the arterial blood gases were checked, the heart was defibrillated with internal paddles and ventricular demand pacing was commenced at 120 beats per minute using Medtronic temporary epicardial pacing wires (Medtronic Inc, USA). This usually occurred approximately 20 minutes after commencing reperfusion of the heart, at which stage all hearts were fibrillating spontaneously thereby necessitating defibrillation. Any further episodes of reperfusion ventricular arrhythmias were treated with lignocaine (0.5-1.0 mg/kg IV) and DC defibrillation as required. When the cardiac rhythm was stable and the heart was ejecting blood, the left ventricular vent was removed and the vent site was oversewn with a 3-0 Prolene suture. Arterial blood gas analysis was initially performed approximately 15 minutes after commencing reperfusion, and then at hourly intervals after reperfusion and also when clinically indicated. Any blood loss occurring during the experiment was collected with the pump sucker and autotransfused back into the animal via the bypass circuit. Additional Hartmann’s solution was added to the bypass circuit to optimise haemodynamic performance.

A dobutamine infusion (Dobutrex, Aspen Pharmacare Australia Pty Ltd, Australia) was prepared at a concentration of 250 mg in 100 mL of 5% glucose in sterile water, and was commenced approximately 45 minutes after commencing reperfusion at a rate of 10 μg/kg/min. This dose of dobutamine was chosen as the minimum inotrope support required by most porcine allografts based on previous porcine transplant studies performed in our lab (Ryan et al. 2003a). No other inotropic or vasoactive therapies

- 117 - were used in the recipient. Furthermore, no attempts were made to wean animals off dobutamine support at any point during the experiment.

One hour after commencing reperfusion, an attempt was made to wean the animal from cardiopulmonary bypass. The degree of ventricular filling and the concomitant rate of withdrawal from bypass support were determined by clinical assessment, and were idiosyncratic for each recipient. Weaning from cardiopulmonary bypass support was deemed successful if the heart generated a self-sustaining aortic root pressure and cardiac output after complete cessation of bypass support for at least 20 minutes, irrespective of systemic arterial blood pressure or the level of dobutamine support required. In general, transplanted hearts that were successfully weaned off bypass were able to generate systolic left ventricular pressures of at least 80 mmHg. Systemic arterial blood pressure was not used as a criterion for successful weaning off bypass because it is dependent on afterload, which is influenced by non-allograft (i.e. recipient) factors.

If weaning off bypass was unsuccessful at one hour post-reperfusion, full cardiopulmonary bypass support was re-instituted and reperfusion in the non-working heart was continued. At two hours post-reperfusion, another attempt at weaning off bypass was made. If still unsuccessful, the animal was returned onto full bypass support and the dobutamine infusion was increased to 20 μg/kg/min. A third, and final, attempt at weaning off bypass was next made at three hours post-reperfusion. If still unsuccessful, the animal was deemed to have failed weaning from cardiopulmonary bypass support. The animal was then euthanased (whilst still under general anaesthesia) by terminating ventricular demand pacing and withdrawing cardiopulmonary bypass support. This resulted in spontaneous cardiac arrest in most animals and if not, an intracardiac injection of potassium chloride (20 mmol) was given into the left ventricle to arrest the heart.

Animals that were successfully weaned off cardiopulmonary bypass were monitored for a further three hours. Once off bypass, general anaesthesia was maintained with isoflurane delivered via the ventilator. After three hours of monitoring, animals were euthanased by terminating ventricular demand pacing and injecting 20 mmol of potassium chloride into the left ventricle to arrest the heart.

- 118 -

2.6.5 Left Ventricular Wall Volume Assessment

Measurement of the left ventricular wall volume for each transplanted heart was required to derive left ventricular pressure-volume (PV) loops and the preload recruitable stroke work relationship. Following termination of each experiment, the heart was explanted from the animal. Both atria and all valvular tissue, along with the free wall of the right ventricle were dissected off the left ventricle. The isolated left ventricular wall was then placed in a measuring cylinder filled with water and the volume displaced by the ventricular tissue was used to determine the left ventricular wall volume.

2.7 DATA ACQUISITION AND ANALYSIS

2.7.1 Haemodynamic Data

Haemodynamic data were collected at regular intervals in both donor and recipient animals during the experiments. These consisted of pulse rate, systolic, diastolic and mean arterial blood pressure, and LAD flow. Data were initially recorded immediately prior to brain death induction (baseline) in the donor animal. It was then recorded at intervals of 30 seconds for 10 minutes after commencing inflation of the Foley catheter balloon to induce brain death, followed by five minute intervals up to one hour, and then at hourly intervals up to six hours after brain death induction.

In the recipient animal, haemodynamic data were recorded immediately after weaning from cardiopulmonary bypass and then at hourly intervals for three hours post-weaning. Haemodynamic data were monitored and recorded in the period after transplantation, whilst on bypass support. However, heart rate and blood pressure were not fully dependant on allograft function because the heart rate was artificially generated by the external pacemaker and the blood pressure was determined by the bypass pump flow. Bearing this in mind, these data were not further analysed.

- 119 - Cardiac output, stroke work and stroke volume were also measured in the donor and recipient animals. These parameters were calculated as part of the left ventricular PV loop analyses (described in Section 2.7.2).

2.7.2 Left Ventricular Pressure-Volume Loops

Cardiac dimension data from the ultrasonic dimension transducers and ventricular pressure data from the pressure catheters were recorded at baseline (immediately prior to brain death induction in the donor animal), and then at designated time-points in the donor and recipient animals during each experiment. The dimension and pressure data were recorded at a sampling rate of 200 Hz and digitised. ECG data from the SpaceLabs Medical Monitor were also recorded, and the combined data (i.e. dimension, pressure and ECG data) were fed directly to a personal computer and acquired using SonoSOFT software version 3.1.6 (Sonometrics Corp, Canada) (Figure 2.16). The dimension and pressure data were used to derive left ventricular PV loops. The ECG data were used to assist in correlating the pressure trace with the various stages of the cardiac cycle (e.g. end-diastole and end-systole).

At the designated time-point, as specified by the study protocol, data were recorded immediately before (steady state, ‘ss’, for five seconds) and during transient occlusion of the IVC (vena caval occlusion, ‘VCO’, for 10 seconds). This occlusion was achieved by pulling up on the snared polyester tape around the IVC. This reduced venous return to the heart, thereby progressively reducing the left ventricular volume with a concomitant reduction in left ventricular pressure. Once data acquisition was completed, the IVC snare was released and the heart returned to its steady state function. Once steady state was achieved (approximately one minute after releasing the snare), a second set of data was acquired. This second data set was recorded as a ‘back-up’, and used if data in the first set was not analysable due to technical problems. Mechanical ventilation was stopped temporarily during the recording of each data set.

- 120 -

Figure 2.16: Acquisition of ECG, LAD flow, and cardiac pressure and dimension data using SonoSoft – data traces shown on computer monitor during data acquisition.

SonoSOFT software version 3.3.2 (Sonometrics Corp, Canada) was used to analyse the pressure and dimension data. This software used the prolate ellipsoid model (Figure

2.17) to calculate epicardial left ventricular volume (LVVepi) from the cardiac dimension  2 data (LVVepi = .a.b /6, where a = major axis length and b = minor axis length. Axes defined in Section 2.5.3). LVVepi data were then correlated with the left ventricular pressure data to construct left ventricular PV loops (example shown in Figure 3.2). End- diastole was determined automatically by the SonoSOFT software using the left ventricular pressure trace, and the end-diastolic LVVepi (EDVepi) and end-diastolic left ventricular pressure (EDP) were recorded. Stroke work (SW) was calculated as the area bounded by the PV loop for each heart beat. The linear relationship between SW and

EDVepi was then determined using linear regression analysis of the data acquired during vena caval occlusion. This relationship is known as the preload recruitable stroke work (PRSW) relationship and is a load independent index of myocardial contractility (Glower et al. 1985). An example of this relationship is shown in Figure 3.3.

- 121 -

Figure 2.17: The geometric model for estimation of left ventricular volume: a) Diagrammatic representation of the ultrasonic dimension transducer positions on the short (minor; anterior-posterior) and long (major; base-apex) axes of the left ventricle and the position of the left ventricular pressure catheter; b) Diagrammatic representation of the prolate ellipsoid model used to calculate epicardial left ventricular volume. This model is the assumed shape of the left ventricle – the prolate ellipse is formed by an ellipse rotated around its long axis (Z) and has a circular cross-section in the short axis (X-Y plane). Therefore to calculate its volume, the short axis and the long axis must be measured.

2.7.3 Normalisation Of Pressure-Volume Loop Data

Prior to analysing the pressure and volume data by linear regression analysis, SW and

EDVepi data were normalised in each individual animal to their baseline (pre-brain death) steady state values. The normalised SW (nSW) and normalised EDVepi (nEDVepi) were then used to derive the PRSW relationship, and the subsequent analyses of the relationship and changes in the relationship (between study groups or between different time-points) were based on the normalised values. As part of the process of

- 122 - normalisation of the data, the LVVepi were also normalised. Therefore, the PV loops generated in the experiments described in Chapters 3 and 5 were also normalised.

Normalisation of the data was done mainly to minimise the potential confounding effects of variable heart sizes in the animals and any potential errors in the methodology used to estimate the epicardial volume of the left ventricle from the dimension data. An added advantage of normalising the data was that the baseline data for the animals (and subsequently the derived group data) were relatively homogeneous, which aided interpretation of subsequent changes. Therefore, changes in SW and the volume-axis intercept of the PRSW relationship (Vw,epi) were assessed as relative changes compared with baseline rather than changes in absolute terms. Therefore, a 10 mL increase in

Vw,epi in a heart with a baseline ssEDVepi of 100 mL was equivalent to a 20 mL change in Vw,epi in a heart with a baseline ssEDVepi of 200 mL.

2.7.4 Assessment Of Left Ventricular Contractility Utilising The Preload Recruitable Stroke Work (PRSW) Relationship

As described in Section 1.6.2, changes in contractile state (for example from acute ischaemia) can cause changes in the slope of the PRSW relationship (Mw) and/or changes in the volume-axis intercept (Vw,epi). In using the PRSW relationship to describe contractile function of the heart and to assess potential changes in contractile function, the relationship (as described by a mathematical model) must first be derived. Once derived, comparisons between relationships could be performed.

After normalising the SW and EDVepi data, the normalised PRSW relationship for each study group at each time-point was determined using multiple linear regression (MLR) analysis (Slinker et al. 1988; Glantz et al. 2001; Ryan et al. 2003a). As a result, the data are described by the general linear model given by Equation 2.1:

nSW = b0 + i=1 to n-1 pi.Pi + b4.nEDVepi Equation 2.1

where b0, pi and b4 are regression coefficients, Pi are the individual animal dummy variables (coded using the effects method (Slinker et al. 1988; Glantz et al. 2001)) and

- 123 - the other terms are as defined earlier. The y-axis intercept of the relationship is b0, the mean slope of the normalised relationship (nMw) is b4 and i=1 to n-1 pi.Pi accounts for individual animal variability. The normalised EDVepi-axis intercept (nVw,epi) is given by

-b0/b4 and the stroke work index (SWI) is given by b0+b4. SWI is the regression estimate of the group’s mean nSW at baseline steady state nEDVepi (i.e. nEDVepi=1). It represents the net effect of the interaction between changes in slope (nMw) and volume axis intercept (nVw,epi) on SW at the normal operating volume of the heart (which is the most physiologically relevant end-diastolic volume). The position of each PRSW relationship is thus defined by the nVw,epi and SWI.

Comparisons between normalised PRSW relationships (either between groups or between different time-points) to determine if they were significantly different were undertaken utilising MLR implementation of two-way analysis of covariance with repeated measures (ANCOVA-RM) of the nSW and nEDVepi data (Slinker et al. 1988; Glantz et al. 2001; Ryan et al. 2003a).

The general linear model used to compare PRSW relationships of the same study group at two different time-points (for example between baseline and post-intervention) was:

nSW = b0 + i=1 to n-1 pi.Pi + b1.TP + b4.nEDVepi Equation 2.2 where TP was the time-point dummy variable (baseline: -1; post-intervention: +1) and the other terms as defined earlier. If b1>0, then nSW was greater post-intervention for any given nEDVepi. If b1<0, then nSW was less post-intervention for any given nEDVepi. The difference between time-points was considered to be statistically significant if the p value for b1 was less than 0.05.

The general linear model used to compare PRSW relationships of two different study groups at the same time-point was:

nSW = b0 + i=1 to n-1 pi.Pi + b2.SG + b4.nEDVepi Equation 2.3 where SG was the study group dummy variable (group A: -1; group B: +1) and the other terms as defined earlier. If b2>0, then nSW was greater in group B than group A - 124 - for any given nEDVepi. If b2<0, then nSW was greater in group A than group B for any given nEDVepi. Again, the difference between study groups was considered to be statistically significant if the p value for b2 was less than 0.05.

2.7.5 Cardiac Troponin I

The plasma concentration of troponin I (TnI) was measured in the donor prior to induction of brain death (baseline) and then at various time-points, depending on the study protocol, in both the donor and recipient animals. Arterial blood (approximately 10 mL) for the TnI assay was taken from the femoral arterial catheter and stored in a lithium heparin tube (Becton-Dickson, Australia) for later analysis. Dr Mark Hicks (Department of Clinical Pharmacology, St. Vincent’s Hospital Sydney) performed the TnI assays for the studies described in Chapters 3 and 5, and the Biochemistry Department (St. Vincent’s Pathology (SydPath), St. Vincent’s Hospital) performed the assays for the study described in Chapter 4.

Plasma was extracted from blood samples and the TnI concentration was assessed using one of two assay systems. Initially, the AxSYM microparticle enzyme immunoassay platform (Abbott Laboratories, USA) was used. This was later replaced with the Bayer- Centaur Automated Chemiluminescence System (Bayer Healthcare Diagnostics, UK).

2.7.6 Pulmonary Function

Arterial blood gas analysis was performed at hourly intervals in both donor and recipient animals, and when clinically indicated, as described in Section 2.5.1.

Pulmonary function was assessed using the PaO2/FiO2 ratio, PaCO2 and the alveolar- arterial oxygen (Aa) gradient. As FiO2=1.0 in all animals, PaO2/FiO2=PaO2, and hence

PaO2 was comparable between animals. The Aa gradient was calculated using the following formula (Mellemgaard 1966) (http://www.globalrph.com/aagradient.htm):

Aa gradient = PAO2 - PaO2 which can be simplified to:

Aa gradient = (713-PaCO2/0.8) - PaO2

- 125 -

where PaO2 = partial pressure of arterial O2 and PaCO2 = partial pressure of arterial CO2

(both obtained from arterial blood gas analysis) and PAO2 = partial pressure of alveolar

O2 (obtained from the alveolar gas equation). Aa gradient, PAO2, PaCO2 and PaO2 are expressed in mmHg.

The alveolar gas equation is:

PAO2 = PiO2 - (PaCO2 / R)

where PiO2 = FiO2 (PB - PH2O) and is the partial pressure of O2 in the central airways.

Using common values, the alveolar gas equation can be expressed as:

PAO2 = (FiO2 x (760 - 47)) - (PaCO2 / 0.8)

where FiO2 is the fraction of inspired oxygen, which in this study, FiO2 = 1.0; PaCO2 is obtained from the arterial blood gas analysis; and PB = barometric pressure (760 mmHg at sea level).

PB is calculated by:

PB = PN2 + PO2 + PCO2 + PH2O

where PH2O = water vapour pressure (47 mm Hg at 37°C) and R = Respiratory quotient

= VCO2 / VO2 = 0.8. R is the ratio of carbon dioxide production to oxygen consumption.

2.7.7 Data Reporting And Statistical Analyses

Continuous variables measured directly were reported as mean ± standard deviation and categorical variables as a percentage or actual incidence/number of hearts in the study group. Non-normally distributed data were reported as median (range). Regression coefficients were reported as mean ± standard error of the mean. Derived regression

- 126 - estimates (e.g. SWI and nVw,epi) were reported as absolute values based on the mean value of the regression coefficient.

Statistical analyses were performed using SPSS for Windows 12.0.1 (SPSS Inc, USA) and StatView for Windows 4.57 (Abacus Concepts Inc, USA). Differences were considered statistically significant if p<0.05. The statistical methods used to test for significant differences between groups or between time-points were determined by the type of data to be compared (for example, normal versus non-normal distribution of data and continuous versus categorical variables). These methods included Student’s t test for independent samples, Fisher’s exact test, analysis of variance (with or without repeated measures: ANOVA or ANOVA-RM) and Mann-Whitney U test. Significant differences in the ANOVA and ANOVA-RM were investigated using Student’s t test for paired samples with a Bonferroni correction for multiple comparisons. All tests were applied as two-tailed tests. The specific methods used are detailed in the methods section of Chapters 3, 4 and 5.

- 127 -

CHAPTER 3

THE EFFECTS OF HORMONE RESUSCITATION ON CARDIAC FUNCTION AND HAEMODYNAMICS IN THE BRAIN-DEAD ORGAN DONOR

- 128 - CHAPTER 3

THE EFFECTS OF HORMONE RESUSCITATION ON CARDIAC FUNCTION AND HAEMODYNAMICS IN THE BRAIN-DEAD ORGAN DONOR

3.1 INTRODUCTION

Despite the advances made in the medical management of congestive heart failure, cardiac transplantation remains the only potentially definitive treatment for patients with end-stage cardiac disease (Copeland 2001). The ever-growing demand for transplantation, however, is not matched by the supply of suitable donor hearts (Conte et al. 2002). This supply shortage is due to the relatively small pool of organ donors, and is compounded by the fact that a significant number of hearts are not recovered and transplanted, often due to organ dysfunction and an unwillingness to use sub-optimal donors (Rosendale et al. 2002; Zaroff et al. 2002; Hicks et al. 2006; Excell et al. 2008). Additionally, the transplant waiting list mortality rate has been reported to be as high as 17-22% per year (Copeland 2001; Zaroff et al. 2002). Clearly, there is a need to optimise organ usage from the currently available donor pool to reduce transplant waiting lists and waiting list mortality.

The success of cardiac transplantation is critically dependant upon the quality of the donor organ and is significantly influenced by the process of brain death in the organ donor. Reported consequences of brain death include haemodynamic instability and donor organ dysfunction (Mertes et al. 1994b; Novitzky 1997a; Pratschke et al. 1999; Seguin et al. 2001; Smith 2004). This can lead to the rejection of organs for transplantation due to sub-optimal function and/or poor donor haemodynamics, leading to high inotrope use and in some cases, cardiac arrest prior to organ procurement. There can also be an increase in post-transplant complications resulting from the effects of brain death, such as initial non-function (or sub-optimal function) of the transplanted heart and early acute rejection (Pennefather et al. 1995; Pratschke et al. 1999; Smith 2004; Hicks et al. 2006).

- 129 -

Brain death in the organ donor results in a series of complex pathophysiological changes in the donor and alters physiological, cellular and biochemical functions, leading to organ dysfunction (Tuttle-Newhall et al. 2003). These changes affect the autonomic nervous system and the levels of circulating endogenous catecholamines, the hypothalamic-pituitary axis and the immune system. The autonomic storm and the associated increase in endogenous catecholamines (Powner et al. 1992) leads to haemodynamic instability, organ ischaemia, disturbed ATP production and metabolism, cell injury and death (Novitzky et al. 1987a; Novitzky et al. 1988c; Shivalkar et al. 1993; Pratschke et al. 1999; Wilhelm et al. 2000b; Seguin et al. 2001; Smith 2004). Disruption of the hypothalamic-pituitary axis results in decreased circulating levels of triiodothyronine (T3) (Novitzky et al. 1984; Gifford et al. 1986; Novitzky et al. 1987a; Novitzky et al. 1988c), arginine vasopressin (Novitzky et al. 1984; Mertes et al. 1994b; Chen et al. 1999), cortisol/adrenocorticotrophic hormone (Novitzky et al. 1984; Smith 2004) and insulin (Novitzky et al. 1984; Smith 2004). These changes also affect high energy phosphate levels and metabolism (Novitzky et al. 1984; Novitzky et al. 1987b; Novitzky et al. 1988c; Smith 2004). Finally, immune system activation such as the activation of the endothelium, leukocytes and platelets leads to the up-regulation of various pro-inflammatory mediators such as cytokines, chemokines and adhesion molecules (Amado et al. 1995; Takada et al. 1998; Wilhelm et al. 2000a; Pratschke et al. 2001a; Plenz et al. 2002; Smith 2004). This enhances donor organ immunogenicity, stimulating the recipient’s immune system and potentially precipitating graft rejection, as well as affecting medium- to long-term graft function (Follette et al. 1998; Wilhelm et al. 2000a; Pratschke et al. 2001a; Smith 2004).

There have been reports that a structured donor management algorithm incorporating hormone replacement may ameliorate many of the negative effects of brain death, and improve cardiac function and donor haemodynamics in both animal models and humans (Novitzky et al. 1987b; Novitzky et al. 1988c; Wheeldon et al. 1995; Zaroff et al. 2002; Hicks et al. 2006). Based on these reports, a cadaveric donor management protocol for haemodynamically unstable donors was developed in the USA at a meeting of key transplantation stakeholders in 2001 (discussed in Section 1.4.5.1), incorporating invasive haemodynamic monitoring and hormone resuscitation (Rosengard et al. 2002; Zaroff et al. 2002). However, there have been no prospective, randomised controlled

- 130 - studies of the complete hormone resuscitation protocol advocated in this donor management protocol, which has since been incorporated in the United Network for Organ Sharing (UNOS) Critical Pathway.

The study described in this chapter was a randomised controlled study designed to examine the impact of the hormone resuscitation protocol in donor management compared with a donor management protocol based on noradrenaline, utilising a porcine model of the brain-dead cardiac donor.

The primary hypothesis of this study was that hormone resuscitation of the brain-dead cardiac donor decreases, if not eliminates, the inotropic requirements (specifically noradrenaline) in the donor to support arterial blood pressure. The hormone resuscitation protocol tested consisted of methylprednisolone, T3, vasopressin and insulin (Table 2.1).

The secondary hypotheses of this study were that hormone resuscitation in the cardiac donor improved the haemodynamic status of the donor and improved left ventricular contractile function.

3.2 METHODS

3.2.1 Porcine Model Of The Brain-Dead Organ Donor

A porcine model of the brain-dead organ donor was used, as described in Section 2.5. This model closely mimics the cadaveric human donor and their management in the intensive care unit setting. Twelve Landrace and 10 Westran pigs (31.85–60.00 kg) were used in this study.

3.2.2 Donor Management

Animals were anaesthetised and surgically prepared as described in Sections 2.5.1 to 2.5.6. Shortly after the completion of cardiac instrumentation, baseline data were acquired (as described in Sections 2.7.1 and 2.7.2). Once the heart returned to its steady state function after vena caval occlusion for data acquisition, brain death was induced

- 131 - by inflation of a Foley catheter balloon situated in the subdural space (as described in Sections 2.5.2 and 2.5.4). Animals were monitored and managed for a total of 6.25 hours after brain death induction. When the mean arterial blood pressure (MAP) fell below 60 mmHg after brain death, boluses of IV saline were given to maintain blood pressure. If this was unsuccessful, an IV noradrenaline infusion (20 μg/mL) was commenced and titrated to achieve a MAP of 60–70 mmHg. Three hours after brain death induction, animals were randomly assigned to commence on a hormone resuscitation (HR) protocol consisting of methylprednisolone, T3, vasopressin and insulin (Table 2.1) (in addition to noradrenaline) or remain on noradrenaline only (control) for a further three hours.

In HR animals, vasopressin was increased incrementally to a maximum of 4 U/hr to maintain MAP between 60-70 mmHg. If MAP was still below 60 mmHg despite being on maximal vasopressin, the noradrenaline infusion rate was increased to maintain MAP. Attempts were made to wean animals off noradrenaline in both groups whilst maintaining MAP. If successful, attempts were then made to wean animals off vasopressin, again whilst maintaining MAP between 60–70 mmHg.

After data were recorded at six hours post-brain death induction, noradrenaline was infused at a fixed low dose of 3.3 μg/min for 15 minutes in both control and HR animals. A data set was then recorded at 6.25 hours post-brain death induction. This was done to exclude noradrenaline dose as a confounder. Each experiment was concluded at 6.25 hours post-brain death induction and the animal euthanased with an intracardiac injection of potassium chloride (20 mmol) into the left ventricle to arrest the heart.

3.2.3 Data Acquisition And Study Outcomes

Haemodynamic data, and left ventricular pressure and cardiac dimension data (to generate left ventricular pressure-volume loops) were recorded prior to brain death induction (baseline), and then at hourly intervals for six hours after brain death induction. The final data set was taken at 6.25 hours post-brain death induction.

Primary outcomes for the study were noradrenaline requirements to maintain blood pressure, left ventricular (LV) contractile function (as assessed by the preload

- 132 - recruitable stroke work - PRSW - relationship) and haemodynamic indices. Secondary outcomes in the study were left anterior descending coronary artery (LAD) flow, arterial blood gases, blood sugar and troponin I. Troponin I was assayed using the AxSYM microparticle enzyme immunoassay platform (Abbott Laboratories, USA). Pulmonary function was assessed as described in Section 2.7.6.

3.2.4 Power Calculation And Statistical Analyses

The primary endpoint of the study was the ability to wean the donor animal from noradrenaline support while maintaining MAP between 60-70 mmHg. It was hypothesised that the utilisation of hormone resuscitation would enable noradrenaline to be weaned in at least 80% of animals compared with 10% of animals that were maintained on noradrenaline throughout the protocol. A study with 6 animals in each group had 90% power (1 – beta) to detect this difference at a significance level (alpha) of 0.05. Because the breed of pig available for the study varied and because of uncertainty regarding the impact that this might have on the primary outcome measure, the number of animals in each group was increased to 8.

The power calculation was based on the formula:

2 n = f(a, ) [p1 (100 p1) + p2 (100 p2)] / (p2 p1)

where p1 and p2 are the percent 'success' in the control and experimental group respectively and f(, ) = [-1(/2) + -1()]2. -1 is the cumulative distribution function of a standardised normal deviate (Pocock 1983; Julious 2004).

Statistical analyses were performed using SPSS for Windows 12.0.1 (SPSS Inc., USA). Continuous variables were reported as mean±standard deviation and categorical variables as actual prevalence in the group. Differences between groups with normally distributed data were compared using Student’s t test for independent samples. The proportion of animals in each group weaned off noradrenaline was compared using Fisher’s exact test. Non-parametrically distributed data were reported as median (range) and compared using the Mann-Whitney U test. Differences were considered statistically significant at p<0.05.

- 133 -

The significance of differences in the PRSW relationship was determined using a multiple linear regression (MLR) implementation of two-way analysis of covariance with repeated measures (Slinker et al. 1988; Glantz et al. 2001; Ryan et al. 2002; Ryan et al. 2003a). To reduce potential confounders, epicardial left ventricular volume

(LVVepi) and stroke work (SW) were normalised within individual animals to their baseline steady state values. The normalised epicardial end diastolic volume axis intercept (nVw,epi) and stroke work index (SWI) were derived from the MLR model of the PRSW relationship. Overall change in LV contractility as reflected in the PRSW relationship was determined by changes in slope of the relationship (nMw) and nVw,epi.

Ischaemia causes a decrease in slope and an increase in nVw,epi. SWI is the regression estimate of the group’s mean normalised SW at baseline steady state end-diastolic volume. It represents the net effect of the interaction between changes in slope and nVw,epi on normalised SW at the normal operating volume of the heart.

3.3 RESULTS

3.3.1 Experimental Animals

Twenty-two animals were used in this study: nine in the control group (five Landrace pigs, two Australian minipigs and two Westran pigs) and eight in the HR group (five Landrace pigs, two Australian minipigs and one Westran pig). Five animals were excluded from the study for technical reasons: one had a large atrial septal defect (measuring approximately 2 x 1 cm), one had cardiomyopathy, one had an anaesthetic- related death, one developed malignant hyperthermia and one sustained a significant LAD injury resulting in a large anterior myocardial infarct. Australian minipigs were initially used in the study but later Landrace pigs were utilised due to an outbreak of a respiratory illness in the minipig colony. Later, Westran pigs were used in the study due to supply issues.

Characteristics of the experimental groups are shown in Table 3.1. There were no significant differences in any of these characteristics between groups.

- 134 - Table 3.1: Characteristics of the Control (n=9) and Hormone Resuscitation (n=8) Groups.

Characteristic Control Hormone p value Group Resuscitation Group

Animal Weight (kg) 48.8±5.5 45.7±6.9 0.33

Left Ventricular Volume (mL) 116.1±11.4 111.8±17.3 0.56

Left Ventricular Volume to 2.4±0.2 2.5±0.2 0.58 Body Weight Ratio

Values are expressed as mean±standard deviation.

One hormone-treated and two control animals had ventricular fibrillation (VF) arrests prior to brain death induction. These arrests occurred during cardiac instrumentation for data acquisition and all were successfully cardioverted to sinus rhythm. The hormone- treated animal had two VF arrests, one whilst inserting the RV Prolene purse-string, requiring three 10 Joule shocks, and the other occurred whilst inserting the RV pressure catheter, requiring three 10 Joule and one 20 Joule shocks. The VF arrests in the control animals occurred during the sewing in of a dimension transducer in one animal, requiring two 10 Joule shocks, and in the second animal, occurred during the insertion of the LV pressure catheter, requiring two 10 Joule shocks. Additional doses of lignocaine were administered to animals that had arrested, and to one additional control and two hormone-treated animals for ventricular ectopic beats post-brain death induction.

3.3.2 Inotrope Requirements

Noradrenaline doses required to maintain MAP were non-parametrically distributed and are shown in Table 3.2. At one hour post-brain death induction, all animals required noradrenaline to support blood pressure. By three hours post-brain death, eight of nine control and eight of eight HR animals were still requiring noradrenaline.

- 135 - Table 3.2: Noradrenaline doses (μg/kg/min) required to maintain blood pressure in the experimental groups.

Time Post-Brain Death Control Hormone p value Induction Group Resuscitation Group

1 hr 0.066 0.111 0.124 (0.033-0.146) (0.053-0.209) 3 hrs 0.065 0.219 0.093 (0.000-1.083) (0.060-1.560) 6 hrs 0.563 0 <0.005 (0.017-9.333) (0-0.314) Values are expressed as median (range).

Noradrenaline was weaned off within six hours of brain death in none of the nine control animals (with the control animal off noradrenaline at three hours re- commencing noradrenaline) and in six of eight HR animals (p<0.005). Figure 3.1 demonstrates the noradrenaline use in each individual animal in both experimental groups. All animals weaned off noradrenaline were also weaned off vasopressin. The two HR animals still requiring noradrenaline were on a lower dose at six hours post- brain death, compared with three hours post-brain death (0.314 vs. 0.576 μg/kg/min and 0.071 vs. 0.284 μg/kg/min, respectively). These two animals were also receiving vasopressin infusions (0.0005 and 0.0012 units/kg/min).

3.3.3 Cardiac Contractility

Representative left ventricular pressure-volume (PV) loops from the control and HR groups are shown in Figure 3.2. This figure demonstrates the decline in stroke work (PV loop area) in a control animal compared with a hormone-treated animal between baseline and 6.25 hours. Representative PRSW relationships from the control and HR groups are shown in Figure 3.3. This demonstrates a decrease in slope in a control animal and an increase in slope in a hormone-treated animal between baseline and 6.25 hours. At the same time, in this particular animal, there was an increase in nVw,epi in the control animal compared with a decrease in the hormone-treated animal.

- 136 - Results of the MLR analysis of the PRSW relationship are shown in Table 3.3. There was no difference in LV contractility pre-brain death between control (SWI=0.974) and HR (SWI=1.012) groups (p=0.08). By six hours post-brain death induction, SWI had increased 53% to 1.494 in the control group and 76% to 1.777 in the HR group. At a fixed noradrenaline dose at 6.25 hours, the SWI was maintained in the hormone-treated group at 1.770, but declined in the control group to 0.918. MLR analysis confirmed that LV contractility was significantly higher in the HR group at both six and 6.25 hours post-brain death induction (p<0.0001).

- 137 - a) Control Animals (n=9) b) Hormone Resuscitation * # Animals (n=8) 2.0 1.8  1.6 1.4 1.2 1.0 0.8 0.6

Noradrenaline Dose (μg/kg/min) 0.4 0.2 0.0 3 6 3 6 Time Post-Brain Death (hours)

Figure 3.1: Noradrenaline Usage in Individual Animals: a) Control and b) Hormone Resuscitation. *Noradrenaline dose at 6 hours was 9.333 μg/kg/min; #Noradrenaline dose at 6 hours was 6.279 μg/kg/min. At one hour post-brain death, all animals required noradrenaline to support blood pressure. By three hours post-brain death, 8/8 hormone resuscitation and 8/9 control animals required noradrenaline. By six hours, 2/8 hormone resuscitation and 9/9 control animals required noradrenaline.

- 138 - Pre-Brain Death Induction (Baseline)

100 Control 100 Hormone Resuscitation  80 80

60 60

40 40

20 20 LV Pressure (mmHg) LV

0 0 0.6 0.7 0.8 0.9 1.0 0.6 0.7 0.8 0.9 1.0 Normalised Epicardial LV Volume Normalised Epicardial LV Volume

6.25 Hours Post Brain Death Induction

Control Hormone Resuscitation 100 100  80 80

60 60

40 40

20 20 LV Pressure (mmHg) LV

0 0 0.6 0.7 0.8 0.9 1.0 0.6 0.7 0.8 0.9 1.0 Normalised Epicardial LV Volume Normalised Epicardial LV Volume

Figure 3.2: Left Ventricular (LV) Pressure-Volume (PV) Loops. Representative loops obtained during transient vena caval occlusion and taken from the same hearts at baseline and at 6.25 hours post-brain death in Control and Hormone Resuscitation groups. Note the decline in PV loop area (stroke work) in the control animal compared with the hormone resuscitation animal between baseline and 6.25 hours.

- 139 - 2.0 Control 1.8

1.6  1.4 Pre-Brain Death 1.2 (Baseline) 1.0

0.8

Normalised Stroke Work Normalised Stroke 0.6

0.4 6.25 Hours Post- 0.2 Brain Death

0 0.5 0.6 0.7 0.8 0.9 1.0 1.1

2.0 Hormone Resuscitation 6.25 Hours Post- 1.8 Brain Death 1.6  1.4

1.2 Pre-Brain Death (Baseline) 1.0 0.8

0.6 Normalised Stroke Work Normalised Stroke 0.4 0.2

0 0.5 0.6 0.7 0.8 0.9 1.0 1.1 Normalised Epicardial End Diastolic Volume

Figure 3.3: Preload Recruitable Stroke Work (PRSW) Relationship. Representative PRSW relationships from the Control and Hormone Resuscitation groups are compared. Note the decrease in slope in the control animal compared with an increase in slope in the hormone- resuscitated animal between baseline and 6.25 hours. Additionally, there is an increase in the x-axis intercept of the PRSW relationship (nVw,epi) in the control animal compared with a decrease in the hormone-resuscitated animal. These changes indicate a decline in contractility in the control animal compared with the hormone-treated animal.

- 140 -

Table 3.3: Preload Recruitable Stroke Work (PRSW) Relationship: Control (CON; n=9) and Hormone Resuscitation (HR; n=8) Groups.

Group nMw y-Axis Intercept nVw,epi SWI

Pre-Brain Death 2.727±0.088 -1.753±0.076 0.643 0.974 (CON) Pre-Brain Death (HR) 2.849±0.065 -1.837±0.056 0.645 1.012

3 hrs Post-Brain 4.464±0.177 -2.704±0.136 0.606 1.760 Death (CON) 3 hrs Post-Brain 4.360±0.102 -2.715±0.075 0.623 1.645 Death (HR)

6 hrs Post-Brain 4.035±0.258 -2.541±0.207 0.630 1.494 Death (CON) 6 hrs Post-Brain 4.478±0.119 -2.701±0.086 0.603 1.777 Death (HR)

6.25 hrs Post-Brain 2.430±0.108 -1.512±0.081 0.622 0.918 Death* (CON) 6.25 hrs Post-Brain 4.833±0.146 -3.063±0.111 0.634 1.770 Death* (HR)

nMw=slope of PRSW Relationship, nVw,epi=normalised epicardial end diastolic volume axis intercept, SWI=stroke work index. *Noradrenaline infusion fixed at 3.3μg/min. nMw and y-axis intercept are expressed as the group mean±standard error of the normalised PRSW. nVw,epi and SWI calculated from nMw and y-axis intercept.

- 141 - 3.3.4 Haemodynamic Changes

Haemodynamic data are shown in Table 3.4. At baseline (pre-brain death induction) and three hours post-brain death induction (and before hormone resuscitation was commenced), there were no differences in MAP between groups. Once hormone resuscitation was commenced, MAP was higher at six and 6.25 hours compared with three hours post-brain death in the HR group, whereas MAP was lower in the control group at six hours and lower still at 6.25 hours, compared with three hours post-brain death. Consequently, MAP was higher in the HR group compared with the control group at six hours (74±17 vs. 54±15 mmHg; p<0.05) and 6.25 hours post-brain death (72±21 vs. 38±11 mmHg; p<0.005).

There were no differences in heart rate between groups at baseline (p=0.429) and after brain death induction (Table 3.4; p=0.698, 0.092 and 0.880 at three, six and 6.25 hours post-brain death, respectively). Cardiac output (CO) was higher in the HR group at 6.25 hours post-brain death (5.8±1.4 vs. 3.2±1.2 L/min; p<0.005), but not at any other time- points (Table 3.4; p=0.189, 0.119 and 0.858 at baseline, three and six hours post-brain death). Similar trends were seen in SW and stroke volume (SV) (Table 3.4): SW was higher in the HR group at 6.25 hours post-brain death (3540±1083 vs. 1536±702 mL.mmHg; p<0.005), as was SV (37.2±8.2 vs. 21.5±9.8 mL; p<0.01). There were no significant differences in SW or SV between groups at any other time-points. LAD flow was not significantly different between groups at any time-point (Table 3.4).

- 142 -

Table 3.4: Haemodynamics: Control (CON; n=9) and Hormone Resuscitation (HR; n=8) Groups.

MAP Heart Cardiac Stroke Work Stroke LAD Group (mmHg) Rate Output (mL.mmHg) Volume Flow (bpm) (L/min) (mL) (mL/min)

Pre-BD 64±18 99±26 4.7±1.1 3146±1087 48.6±9.5 23±10 (CON) Pre-BD 59±13 91±12 4.1±0.5 2916±473 46.4±9.0 28±5 (HR)

3 hrs Post- 63±9 161±25 5.6±0.9 3356±724 35.2±6.4 35±14 BD (CON) 3 hrs Post- 61±7 166±27 4.6±1.4 2634±903 28.5±10.0 44±17 BD (HR)

6 hrs Post- 54±15 186±41 5.1±1.9 2980±1528 28.1±8.4 47±23 BD (CON) 6 hrs Post- 74±17† 157±19 5.3±2.2 3224±1158 33.9±13.3 45±25 BD (HR)

6.25 hrs 38±11 155±25 3.2±1.2 1536±702 21.5±9.8 31±14 Post-BD* (CON) 6.25 hrs 72±21‡ 156±15 5.8±1.4‡ 3540±1083‡ 37.2±8.2† 51±15 Post-BD* (HR)

*Noradrenaline infusion fixed at 3.3μg/min. †p<0.05, ‡p<0.005. BD=Brain death, MAP=Mean arterial blood pressure, LAD=Left Anterior Descending Coronary Artery.

- 143 - 3.3.5 Troponin I, Blood Glucose, Pulmonary Function And Acid-Base Balance

Troponin I levels increased progressively in both groups over time, but no significant differences between groups were identified at any time-point (Table 3.5). Troponin I results from three animals (two control animals and one hormone-treated animal) were excluded from the analysis due to a change in the laboratory assay technique from the AxSYM microparticle enzyme immunoassay platform to the Bayer-Centaur Automated Chemiluminescence System. Results from the Bayer-Centaur method were not directly comparable with results from the AxSYM methodology.

Table 3.5: Troponin I (μg/L; median and range): Control (n=7) and Hormone Resuscitation (n=7) Groups.

Time Control Hormone Resuscitation p value

Pre-Surgery 0.0 (0.0-6.0) 0.0 (0.0-3.5) 0.777

Post-Instrumentation 1.3 (0.6-13.5) 2.4 (0.5-22.5) 0.749

1 h Post-Brain Death 5.1 (1.8-20.2) 6.4 (2.5-31.2) 0.456

3 h Post-Brain Death 12.9 (5.4-31.9) 14.1 (8.8-38.3) 0.259

6 h Post-Brain Death 22.7 (16.3-500) 30.5 (13.9-106.1) 0.731

Blood glucose data is shown in Table 3.6. There was no significant difference in blood glucose levels between groups at baseline. Three HR animals required IV insulin for hyperglycaemia: one requiring 10 units over the final three hours of management and the other two requiring 5 and 10 units. After correction with insulin, there were no significant differences in blood glucose levels at six hours post-brain death induction (7.5±2.4 mmol/L in HR animals vs. 10.6±4.5 mmol/L in controls; p=0.096), despite three control animals having levels >10 mmol/L at six hours.

- 144 - Table 3.6: Blood glucose (mmol/L; mean±standard deviation): Control (n=9) and Hormone Resuscitation (n=8) Groups.

Time Control Hormone p value Resuscitation

Pre-Brain Death 6.2±1.1 6.1±1.2 0.903

3 h Post-Brain Death 6.4±1.6 6.8±1.9 0.615

6 h Post-Brain Death 10.6±4.5 7.5±2.4 0.096

Arterial blood gas results are shown in Table 3.7. As FiO2=1.0 in all animals,

PaO2/FiO2=PaO2, and hence PaO2 was comparable between animals. There were no differences between groups with respect to PaO2, PaCO2 and alveolar-arterial (Aa) gradient at baseline or at three hours post-brain death (i.e. before commencing hormone resuscitation). By six hours post-brain death, there were still no significant differences between groups in PaO2 (346±131 mmHg in HR vs. 336±160 mmHg in control; p=0.896), PaCO2 (62.2±13.5 vs. 56.3±19.6 mmHg; p=0.481), or Aa gradient (289.3±133.3 vs. 306.3±168.2 mmHg; p=0.822).

Acid-base balance was also not significantly different at any time-point (Table 3.7). No statistically significant differences were detected between experimental groups in either pH or base excess at baseline, three hours or six hours post-brain death induction.

- 145 - Table 3.7: Arterial blood gas results (mean±standard deviation): Control (CON; n=9) and Hormone Resuscitation (HR; n=8) Groups.

Aa p O p CO Base Group a 2 a 2 Gradient pH (mmHg) (mmHg) Excess (mmHg)

Pre-Brain Death 440±77 47.3±11.2 213.4±71.1 7.45±0.07 8±2 (CON) Pre-Brain Death 471±93 44.5±6.2 186.2±92.1 7.47±0.08 9±4 (HR)

3 Hours Post-Brain 374±90 47.3±6.3 279.8±92.1 7.43±0.05 7±2 Death (CON) 3 Hours Post-Brain 443±86 46.3±4.7 212.6±83.6 7.44±0.08 7±4 Death (HR)

6 Hours Post-Brain 336±160 56.3±19.6 306.3±168.2 7.25±0.08 -3±8 Death (CON) 6 Hours Post-Brain 346±131 62.2±13.5 289.3±133.3 7.27±0.09 1±4 Death (HR)

Aa Gradient=Alveolar-arterial gradient. No statistically significant differences were found between groups (see Section 3.3.5).

3.4 DISCUSSION

3.4.1 Study Rationale And Experimental Groups

This study utilised a porcine model of the brain-dead organ donor to compare the combined hormone resuscitation protocol (consisting of T3, arginine vasopressin, methylprednisolone and insulin) endorsed in the UNOS Critical Pathway (Rosengard et al. 2002; Zaroff et al. 2002) with a noradrenaline-based protocol for cadaveric donor management. The hormone resuscitation protocol (Table 2.1) and many of the management principles employed in the model were based on the USA donor management algorithm described earlier (Zaroff et al. 2002). Noradrenaline was used

- 146 - because it is the most commonly used inotrope in donor management in Australia and New Zealand. In 2007, 92% of donors in Australia and New Zealand received inotropic support, with noradrenaline being used in 81% of cases (Excell et al. 2008). Similarly high rates of catecholamine administration have been reported in other registries such as the Eurotransplant registry, where 91% of brain-dead donors were maintained with catecholamines and 1/3 donors receiving more than one catecholamine infusion (Schnuelle et al. 2001).

Animals in the control and HR groups were comparable. There were no significant differences between groups with respect to body weight, LV volume and LV volume to body weight ratio (Table 3.1). In addition, there were no baseline differences between groups in MAP, heart rate, CO, SW, SV, LAD flow, troponin I, blood glucose level,

PaO2, PaCO2, Aa gradient, pH and base excess (Tables 3.4, 3.5, 3.6 and 3.7).

3.4.2 The Impact Of Hormone Resuscitation On Inotrope Usage To Maintain Blood Pressure In The Donor

One of the most striking results from this study was the impact of hormone resuscitation on the need for noradrenaline to maintain blood pressure (Table 3.2 and Figure 3.1). Six of eight (75%) hormone-treated animals were weaned off noradrenaline, with the remaining two animals requiring lower doses of noradrenaline at six hours compared with three hours post-brain death when hormone therapy was first commenced. In contrast, none of the control animals were weaned off noradrenaline. Rather, their dosage requirements increased significantly over the final three hours of donor management. Many of these control animals had declining responses to noradrenaline as the dose was escalated, indicating the development of tachyphylaxis. Indeed, several control animals were unable to maintain the target MAP specified in the study protocol despite rapid escalation of the noradrenaline dose. The haemodynamic status of these animals closely mimicked that of human donors that become haemodynamically unstable after brain death. Such patients are often managed with further catecholamines and additional intravenous fluids, both of which can further impair donor heart function (Pennefather et al. 1995; Schnuelle et al. 2001; Smith 2004). Administration of noradrenaline in particular has been associated with myocardial damage and initial non-

- 147 - function after cardiac transplantation (Todd et al. 1985a; Movahed et al. 1994; Schnuelle et al. 2001). As a result, hearts from donors receiving high doses of catecholamine, particularly in the presence of haemodynamic instability, are frequently rejected for use in transplantation (Wheeldon et al. 1995; Zaroff et al. 2002; Excell et al. 2008).

3.4.3 The Effects Of Brain Death On Cardiac Contractility

Several investigators have demonstrated experimentally that cardiac contractility as assessed by the PRSW relationship deteriorates in untreated animals after brain death (Bittner et al. 1996a; Ryan et al. 2003c; Lyons et al. 2005). The reasons for this decline are likely to be multifactorial. The reduction in cardiac contractility is biphasic with an initial transient decrease which may be caused by myocardial ischaemia (Ryan et al. 2003c) or rapid desensitisation of the -adrenergic signalling pathway (White et al. 1995) during the “autonomic storm”. This is followed by a sustained reduction in contractility which may be due to ongoing catecholamine-induced myocardial ischaemia (Seguin et al. 2001; Lyons et al. 2005), impaired -adrenergic signalling (White et al. 1995), impaired myocardial metabolism associated with the altered neurohormonal environment after brain death (Novitzky et al. 1988c; Bittner et al. 1996b; Lyons et al. 2005) or up-regulation of pro-inflammatory mediators (Pratschke et al. 1999; Lyons et al. 2005).

3.4.4 The Use Of Hormone Resuscitation To Ameliorate The Negative Effects Of Brain Death On Cardiac Contractility

Analysis of the PRSW relationship in this study demonstrated that hormone-treated animals had increased cardiac contractility at six hours post-brain death, both in comparison with baseline (pre-brain death) and with noradrenaline-treated animals. The difference in contractility between the two groups was further accentuated at 6.25 hours post-brain death when both groups were placed on same dose of noradrenaline (Table

3.3, Figure 3.3). Whilst there was no significant change in nVw,epi in either group between baseline and 6.25 hours (Table 3.3), there was a decline in slope (nMw) in the control group (2.727 to 2.430), indicating a reduction in contractility. In contrast, there was an increase in slope in the HR group between the same time-points (2.849 to

- 148 - 4.833), indicating an increase in contractility. The superior cardiac function in hormone- treated animals was confirmed by MLR analysis and SWI changes.

In a similarly designed study using cross-bred pigs, Lyons et al reported that bolus administration of methylprednisolone prior to or soon after brain death prevented any deterioration in cardiac contractility and preserved the PRSW at baseline levels for up to six hours post induction of brain death (Lyons et al. 2005). Considering this finding of preserved function rather than improved function in relation to the present study, it suggests that the methylprednisolone given as part of the hormone resuscitation protocol in this study may have contributed to but cannot account for the increased contractility observed in hormone-treated animals. Furthermore, as six of eight animals in the hormone-treated group were weaned off vasopressin by six hours post-brain death, one might speculate that infusion of T3 was responsible for the increase in contractility observed in these animals. T3 has been shown to have direct positive inotropic effects in isolated hearts in vitro (Snow et al. 1992; Ririe et al. 1995), and in normal and cardiomyopathic hearts in vivo (Jamall et al. 1997). The positive inotropic action of T3 appears to be mediated via up-regulation of sarco-endoplasmic reticulum Ca2+ ATPase (SERCA) (Sayen et al. 1992; Holt et al. 1999; Trivieri et al. 2006) and is independent of the -adrenergic signalling pathway (Ririe et al. 1995).

Noradrenaline treatment after brain death also increased cardiac contractility in this study. However, the positive inotropic response at six hours post-brain death was less than that observed at three hours despite up-titration of the noradrenaline dose, indicating tachyphylaxis to noradrenaline. This finding is consistent with other reports of rapid desensitization of the myocardial -adrenergic signalling pathway after brain death (D'Amico et al. 1995; White et al. 1995), probably triggered by the high circulating levels of endogenous noradrenaline post-brain death. Administration of noradrenaline as in this study is likely to further desensitize myocardial -adrenergic signalling. Another potential explanation for the loss of contractile responsiveness to noradrenaline with time post-brain death is the development of catecholamine-induced myocardial injury. This phenomenon has been demonstrated in a variety of experimental and clinical settings (Todd et al. 1985a; Movahed et al. 1994; Schnuelle et

- 149 - al. 2001) and appears to be mediated via -adrenergic stimulation, as it can be abrogated by pharmacological beta blockade (Cruickshank et al. 1987).

3.4.5 The Effects of Donor Brain Death And Subsequent Donor Management On Troponin I Release

Chiari and colleagues reported a small increase in troponin I levels after brain death in pigs, consistent with myocardial ischaemic injury (Chiari et al. 2000). Troponin I levels recorded three hours after brain death in their study were comparable to those recorded in this study, allowing for differences in assay techniques and in the degree of surgical preparation of the heart between the two studies. Serial measurements of plasma troponin I levels in this study revealed no significant differences between treatment groups at any time-point post-transplant (Table 3.5). This observation should be interpreted cautiously for several reasons. A small rise in troponin I levels was evident in both groups of animals prior to induction of brain death, indicating a contribution of surgical preparation of the heart to the rise in troponin I level after brain death. Other variables that could affect troponin I levels such as cardioversion occurred in some animals but not others (although only one hormone-treated animal received cardioversion compared with two control animals). Finally, the range in troponin I values was wide and non-parametrically distributed suggesting that the study was underpowered to detect differences in troponin I release between the two treatment protocols. Bearing in mind these caveats, it is noteworthy that the enhanced contractility observed in animals that received hormone resuscitation was not associated with increased myocardial injury as judged by troponin I release. The possibility that both treatment protocols increased myocardial troponin I release cannot be excluded. Troponin I release after brain death and the effects of IV fluids, hormone resuscitation and noradrenaline on myocardial injury are further investigated and discussed in Chapter 4.

- 150 - 3.4.6 The Effects Of Hormone Resuscitation On the Haemodynamics Of The Brain-Dead Cardiac Donor

This study showed that overall haemodynamic status was improved in hormone-treated animals compared with control animals. Cardiac output was well maintained by both treatment protocols for up to six hours (Table 3.4). However, at 6.25 hours (when the noradrenaline dose was identical in both groups), cardiac output was significantly higher in the hormone-treated group indicating the reliance of control animals on noradrenaline to maintain cardiac output.

In terms of arterial blood pressure, there were marked differences in blood pressure responses between treatment groups at six and 6.25 hours post-brain death (Table 3.4). By six hours, control animals had a significantly lower MAP despite up-titration of the noradrenaline dose. As the cardiac output was similar in both groups at six hours, the difference in MAP can only be explained by differing actions of the treatment protocols on systemic vascular resistance. This indicates the development of rapid desensitisation to noradrenaline at the vascular level, which occurred in parallel with the myocardial desensitisation described above. The cause for this vascular desensitisation is unclear but could be explained by high circulating levels of endogenous catecholamines after brain death (Chiari et al. 2000; Lyons et al. 2005) acting directly on the 1-adrenergic receptor (Carrier et al. 1978), or by the loss of other vasoconstrictor mechanisms after brain death (Pennefather et al. 1995; Chen et al. 1996; Chen et al. 1999). Regardless of the mechanism, the findings of this study indicate that noradrenaline alone is a poor choice to maintain blood pressure in the brain-dead donor.

Lyons et al reported that both CO and arterial blood pressure fell over six hours after brain death induction in pigs (Lyons et al. 2005). Interestingly, methylprednisolone did not prevent the reduction in cardiac output, even though it maintained cardiac contractility (Lyons et al. 2005). This finding indicates that the fall in cardiac output resulted from altered loading conditions on the heart, either reduced preload or increased afterload. Their findings suggest that methylprednisolone at least partly preserved systemic vascular resistance, although this was not measured directly. In comparison, the combined hormone resuscitation protocol used in this study maintained both blood pressure and CO at or above the levels measured prior to brain death.

- 151 -

3.4.7 The Effects Of Steroids On Pulmonary Function And Blood Glucose Control

High dose steroids have been shown to block up-regulation of inflammatory mediators (Wan et al. 1996), prevent ischaemic damage (Busuttil et al. 1975), and improve organ function and graft survival after transplantation (Follette et al. 1998; Pratschke et al. 2001a; Rosendale et al. 2003a). High dose steroids have also been associated with improved oxygenation and lung recovery in a large retrospective analysis of potential lung donors (Follette et al. 1998).

No statistically significant differences were detected in this study between the effects of combined hormonal therapy and noradrenaline treatment on lung function after brain death. Donor lung function, as assessed by serial measurements of PaO2, PaCO2 and Aa gradient, remained stable over the six hours after induction of brain death (Table 3.7). In the study by Follette et al, donors were managed for approximately 23.5 hours after receiving high dose steroid therapy (Follette et al. 1998). Possibly a longer period of observation post-brain death or a longer treatment period of hormonal therapy may have revealed treatment differences in this study.

The use of high dose steroids in the combined hormonal therapy did not adversely affect blood sugar control (Table 3.6). Indeed, blood sugar levels tended to be lower in animals that received hormone therapy (7.5±2.4 vs. 10.6±4.5 mmol/L; p=0.096).

3.4.8 Study Limitations

There are a number of limitations in this study to be considered. Clinically, the majority of multi-organ donors are managed for more than six hours after brain death prior to organ retrieval. In Australia, 98% of donors (95% in New Zealand) had a time of more than six hours between certification of brain death and aortic cross clamp, with a median time of 15.2 hours in Australia and 11.8 hours in New Zealand (Excell et al. 2008). However the duration of this study was limited to six hours after brain death, mainly due to research funding and staffing considerations. Despite this, in light of the

- 152 - trends that were apparent by six hours post-brain death, the differences between treatments may have been even greater had the animals been followed for longer.

Another limitation of this study was that it did not assess the impact of individual hormones on the donor, nor did it examine the molecular and cellular effects of these compounds on the organs. However, as the hormone resuscitation protocol had already been endorsed by key transplantation stakeholders in the USA (as well as being incorporated in the UNOS Critical Pathway for clinical use), the study was designed to examine the complete hormone cocktail rather than the individual components to provide clinically relevant results.

The primary objective of this study was to examine the effects of HR on the cardiac organ donor. As such, it did not assess the effects of the treatment regimens on intra- abdominal organs. There have been concerns expressed that vasopressin may impair function and preservation of abdominal organs such as the liver, pancreas and kidney because of its non-selective vasoconstrictive effects (Pennefather et al. 1995). Post- transplantation outcomes with these treatments were also not assessed. However these concerns may be unfounded, with observational data suggesting that the retrieval rate of all intra-abdominal organs is increased, and the risks of primary graft dysfunction and graft loss appear to be reduced when the donor is treated with combined hormone therapy after brain death (Rosendale et al. 2002; Rosendale et al. 2003a; Rosendale et al. 2003b; Rosendale et al. 2004). Nonetheless, the lack of a prospective randomised study limits any conclusions that can be made regarding the impact of combined hormone therapy on the quality of intra-abdominal donor organs.

3.5 CONCLUSION

To date, there are no prospective, randomised controlled studies testing combined hormone resuscitation, and examining its effects on donor cardiac function, haemodynamics and vasopressor requirements in the literature. Previous experimental studies have tested single hormonal replacement or a combination of hormonal therapies (Novitzky et al. 1987b; Rosendale et al. 2002) but not the complete hormone replacement protocol advocated in the UNOS Critical Pathway. Clinical studies of the hormone replacement cocktail have been observational, retrospective and used non-

- 153 - matched controls (Wheeldon et al. 1995; Rosendale et al. 2003a). In addition, the contribution of other changes in donor management such as invasive haemodynamic monitoring make it difficult to judge the relative contribution of combined hormonal resuscitation to the improvements in donor organ outcomes reported in these studies. The results of the present study support the use of combined hormone resuscitation to enhance donor heart function, however there is a clear need for the impact of combined hormone resuscitation on intra-abdominal donor organs to be assessed. This will be addressed in a randomised study reported in Chapter 4.

In conclusion, this study has demonstrated that combined hormone resuscitation of the brain-dead organ donor as advocated in the UNOS Critical Pathway increases cardiac contractility, stabilises haemodynamics and markedly reduces the need for catecholamine support. The study also highlights the limitations and potential adverse consequences of noradrenaline on cardiac function and haemodynamic status after brain death. Translation of these findings to clinical practice may be expected to improve the quality of donor hearts, including hearts currently classified as “sub-optimal” that would otherwise be rejected for transplantation.

- 154 -

CHAPTER 4

THE EFFECTS OF HORMONE RESUSCITATION ON THE LUNGS AND ABDOMINAL ORGANS FOR TRANSPLANTATION IN THE BRAIN- DEAD MULTI-ORGAN DONOR

- 155 - CHAPTER 4

THE EFFECTS OF HORMONE RESUSCITATION ON THE LUNGS AND ABDOMINAL ORGANS FOR TRANSPLANTATION IN THE BRAIN-DEAD MULTI- ORGAN DONOR

4.1 INTRODUCTION

As described in Section 1.4, brain death in the organ donor has multiple negative effects on the donor and their organs, leading to haemodynamic instability and donor organ dysfunction (Mertes et al. 1994b; Novitzky 1997a; Pratschke et al. 1999; Seguin et al. 2001; Smith 2004; Hicks et al. 2006). This leads to the loss of organs for transplantation due to sub-optimal function and/or poor donor haemodynamics. There is also an overall increase in post-transplantation complications, such as sub-optimal cardiac function that is unable to support the recipient’s circulation unassisted (either mechanically or pharmacologically), and early acute rejection (Pennefather et al. 1995; Pratschke et al. 1999; Smith 2004; Hicks et al. 2006).

The study described in Chapter 3 demonstrated that the hormone resuscitation protocol advocated in the United Network for Organ Sharing (UNOS) Critical Pathway donor management protocol (Rosengard et al. 2002; Zaroff et al. 2002) had beneficial effects on donor haemodynamics, inotrope usage and cardiac function (Hing et al. 2007). Hormone resuscitation in the porcine model of the brain-dead organ donor was found to significantly reduce the inotrope requirements to maintain arterial blood pressure, with 75% of animals completely weaned off all inotropic support. There were also improved haemodynamic parameters (e.g. blood pressure, cardiac output, stroke work and stroke volume) and superior cardiac function (as demonstrated by the preload recruitable stroke work – PRSW - relationship) in hormone-treated animals.

However, whilst this study, along with other published data, supports the clinical use of hormone resuscitation in the brain-dead organ donor, particularly in those who are

- 156 - haemodynamically unstable (Novitzky et al. 1987b; Novitzky et al. 1987d; Cooper et al. 1988b; Novitzky et al. 1988b; Novitzky et al. 1988c; Novitzky et al. 1990; Jeevanandam et al. 1993; Jeevanandam et al. 1994; Wheeldon et al. 1995; Salim et al. 2001; Hing et al. 2007), there is relatively little known about the effects that these treatments have on other transplantable organs, such as the lungs, liver, kidney and pancreas. In order to advocate the clinical use of hormone resuscitation during donor management for the specific purpose of improving haemodynamics and cardiac function, it must be proven that this treatment has minimal, if any, negative impact on non-cardiac organs. Intuitively, it might be predicted that any treatment to improve donor haemodynamics and cardiac function would be likely to also improve perfusion of peripheral organs such as the kidney and liver, and hence should improve their function and subsequent post-transplantation results.

In an effort to address the issue of the effects of hormone resuscitation on non-cardiac organs, the study described in this chapter was designed as a randomised controlled study to compare the impact of different donor management protocols based on intravenous fluids only, noradrenaline or hormone resuscitation, utilising a porcine model of the brain-dead multi-organ donor.

The primary hypothesis of this study was that hormone resuscitation (HR) in the brain- dead multi-organ donor did not have a negative impact on the quality of transplantable solid organs (specifically, the lungs, liver, kidneys and pancreas). Furthermore, it was hypothesised that HR would result in better preservation of donor haemodynamics, and consequently improve perfusion and function of transplantable organs than donor management protocols based on noradrenaline (NA) or intravenous (IV) fluids alone without inotropic/vasopressor support (FL).

The secondary hypothesis of this study was that the NA and FL donor management protocols would have either negative or no effects on the quality of transplantable solid organs (specifically, the lungs, liver, kidneys and pancreas) when compared with HR.

- 157 - 4.2 METHODS

4.2.1 Porcine Model Of The Brain-Dead Organ Donor

A porcine model of the brain-dead multi-organ donor was used as described in Section 2.5. This model closely mimics the cadaveric human donor and their management in the intensive care unit. A total of 30 pigs (29.50–58.00 kg) were used in the study. Of these, six were Westran pigs and 24 were Landrace pigs. The species of pigs used for the studies were changed from Westran to Landrace animals due to animal supply problems.

4.2.2 Donor Animal Anaesthesia, Surgery And Preparation For Brain Death Induction And Cardiac Data Acquisition

Animals were prepared, anaesthetised and monitored as described in Section 2.5.1. Once animals were anaesthetised, an orogastric tube (Salem Sump Tube 18 Fr, Sherwood Medical USA) was inserted. The gastric contents were aspirated and the tube then placed on free drainage into a drainage bag.

Next, a burr-hole was made in the skull and a Foley catheter placed in the sub-dural space for brain death induction (as described in Section 2.5.2). The heart was then exposed by median sternotomy and instrumented for data collection (i.e. placement of the ultrasonic cardiac dimension transducers, the left ventricular and right ventricular micromanometer-tipped pressure catheters and the left anterior descending coronary artery (LAD) flow probe) as described in Section 2.5.3.

4.2.3 Abdominal Surgery And Instrumentation Of The Abdominal Organs For Data Acquisition

Once cardiac instrumentation was completed, a midline laparotomy was performed. The urinary bladder was identified and a 2-0 silk purse-string was placed in the anterior surface of the bladder. A one cm incision was made in the centre of the purse-string and a 14 Ch/Fr 4.7 mm Foley catheter (Bard SDN. BHD., Malaysia) was inserted into the bladder. The purse-string was then tightened and tied around the catheter to secure it,

- 158 - followed by inflation of the catheter balloon with 10 mL of water. The catheter was connected to a urine measurement bag. The bladder was emptied of all urine and a timed collection of urine was commenced to calculate baseline (pre-brain death) creatinine clearance.

The left and right kidneys were next identified, with the right kidney mobilised to facilitate wedge biopsies to be performed later in the experiment. The left renal artery was identified and mobilised, taking care not to damage the renal vein or ureter. Once mobilised, a 4 or 6 mm flow probe (Transonic Systems Inc, USA) was placed around the renal artery (Figure 4.1).

Next, the tail and the distal part of the body of the pancreas were mobilised (Figure 4.2). This was done to enable serial biopsies to be taken, commencing distally and then progressing proximally through the body of the pancreas.

The portal triad was next identified, and the common bile duct, hepatic artery and portal vein were mobilised (Figure 4.3a and b). The cystic duct was mobilised and ligated. The common bile duct was then ligated distally and divided, with the proximal end cannulated with a ureteric catheter. The catheter was secured with a 4-0 silk tie to the common bile duct (Figure 4.3b) and the bile drained via the catheter into a 20 mL Falcon tube. The bile volume was measured at hourly intervals to calculate hourly bile production.

A 12 or 14 mm flow probe (Transonic Systems Inc, USA) was placed around the portal vein and a 4 or 6 mm flow probe was placed around the hepatic artery (Figure 4.4). The flow probes around the renal artery, portal vein and hepatic artery were then connected to a T208 Transonic Volume Flow Meter (Transonic Systems Inc, USA). Data collected from these devices were digitised at 200 Hz.

- 159 - a)

RV

RA

Ur

b)

Figure 4.1: (a) Left kidney with renal artery flow probe in situ. (b) Magnified view of hilum of the left kidney demonstrating the left renal artery (RA) with the flow probe around it, renal vein (RV) and ureter (Ur).

- 160 - St

Spl

Ps

SI Duo

Figure 4.2: Mobilised porcine pancreas (Ps). St = stomach (reflected superiorly); Spl = spleen; Duo = duodenum; SI = small intestine.

- 161 - a)

CBD

HA PV

b)

Figure 4.3: a) Mobilising the portal triad with the liver reflected superiorly; and b) The portal triad demonstrating the cannulated common bile duct (CBD) for bile collection, the portal vein (PV) and the hepatic artery (HA).

- 162 - CBD

PV HA

Figure 4.4: Placement of flow probes around the portal vein (PV) and the hepatic artery (HA). The liver is reflected superiorly and the common bile duct (CBD) catheter is in situ.

- 163 - 4.2.4 Cardiothoracic And Abdominal Tissue Biopsy Protocol

Tissue biopsies of the heart, lungs, liver, kidney and pancreas were collected in this study for two separate research projects to that reported in this thesis. Biopsies were collected prior to brain death induction (baseline), and at one, three and six hours post- brain death induction.

Two to three core biopsies of the left ventricle were taken using a Bard Max•Core disposable core biopsy instrument (C.R. Bard Inc, USA). Any bleeding from the biopsy site/s was oversewn using 4-0 Prolene. Wedge biopsies (approximately 1 x 2 cm) of the inferior lobe of the left lung were taken with the cut surface ligated with a 2-0 silk tie. Wedge biopsies of approximately 1 x 1 cm were cut from the liver edge and the cortex of the right kidney. The biopsy sites were then oversewn with 3-0 Prolene. Bleeding from the hepatic biopsy site was controlled using Prolene alone. Haemostasis was achieved at the renal biopsy site by oversewing with Prolene utilising the renal capsule and peritoneum as pledgets and in some cases, packing the kidney with moist gauze swabs. Pancreatic biopsies (approximately 1-2 cm in length) were cut from the distal end of the pancreas with the cut end ligated with a 2-0 silk tie. Tissue samples were processed and analysed for additional research projects not reported herein.

4.2.5 Experimental Protocol

The study design and the donor management protocol for each experimental group are shown in Figure 4.5. The timetable for the collection of tissue, blood, and urine samples, and for data collection (e.g. haemodynamic indices, blood flow through the LAD, renal artery, portal vein and hepatic artery, and bile production) in the study is also shown in Figure 4.5. Once the cardiac and abdominal instrumentation were completed, baseline (pre-brain death) haemodynamic and cardiac pressure-volume (PV) loop data were acquired (as described in Sections 2.7.1 and 2.7.2). Tissue biopsies of the heart, lungs, liver, kidney and pancreas were then taken.

- 164 -

Figure Overleaf

- 165 -

Figure 4.5: Study design and the donor management protocol for each of three experimental groups. Intravenous (IV) saline was given in each group to maintain CVP 0-5 mmHg. Saline was also given in the control group to maintain mean arterial pressure (MAP). Data and tissue sampling time-points are also shown in the figure. Tissue biopsies were taken from the heart, lung, liver, kidney and pancreas. Arterial blood gas and blood glucose analyses were performed at baseline and at hourly intervals post-brain death induction. Further details can be found in Section 4.2.5.

Study termination and animal General Anaesthetic Induction Brain Death Induction euthanased

IV Fluids (Control) Group (FL) ; n=9 Instrumentation Commence IV noradrenaline and titrate to MAP 60-70 mmHg

Noradrenaline Group (NA); n=9 Instrumentation Commence Hormone Resuscitation

Hormone Resuscitation

- 166- Group (HR) ; n=9 Instrumentation

Organ Tissue Biopsy (Bx): Bx Bx Bx Bx

Blood Collection (Bl):Bl Bl Bl Bl Bl Urine Collection (Ur): Ur Ur Ur Ur Ur

Organ Blood Flow (Fl): FlFl Fl Fl Fl Fl Fl Fl Fl Fl

Bile Production (Bi): Bi Bi Bi Bi Bi Bi Bi Cardiac Pressure-Volume Data (PV): PV PV PV PV Haemodynamic Data (HD): HDHH H HD HD HD HD HD HD

Time (minutes): -180 -120 -60 0 60 120 180 240 300 360 Once the heart returned to its steady state function, brain death was induced by inflation of the Foley catheter balloon situated in the subdural space as described in Sections 2.5.2 and 2.5.4. Animals were then monitored and managed for a total of six hours. Haemodynamic data were collected as described in Section 2.7.1. When the mean arterial blood pressure (MAP) fell below 60 mmHg after brain death, boluses of IV saline were given to support the blood pressure.

One hour after brain death induction, animals were randomly assigned to one of three experimental groups. Animals in the first group were managed using IV fluids (0.9% saline) only to maintain blood pressure and central venous pressure (CVP), and served as the control group (FL group). The second group of animals was managed using an IV noradrenaline infusion (20 μg/mL) to maintain MAP, which was commenced one hour post-brain death induction (NA group). The third group was commenced on a hormone resuscitation protocol consisting of methylprednisolone, triiodothyronine (T3), vasopressin and insulin (described in detail in Section 2.5.6 and Table 2.1) at one hour- post brain death (HR group). Methylprednisolone was administered as a single IV bolus of 15 mg/kg. T3 was administered as an initial bolus dose of 4 μg, followed by a continuous infusion of 4 μg/hr. Vasopressin was commenced at 0.5 Units/hr and increased incrementally to a maximum of 4 Units/hr to maintain the MAP between 60 to 70 mmHg. Animals in the NA and HR groups also received IV saline to maintain CVP between 0 to 5 mmHg. Each experimental group was managed for a further five hours.

The IV noradrenaline infusion was commenced in the NA group one hour post-brain death induction only when IV fluid resuscitation was unsuccessful in maintaining MAP. The infusion was titrated to maintain a MAP of 60–70 mmHg. In hormone resuscitation animals, vasopressin was increased incrementally to a maximum of 4 U/hr to maintain the MAP between 60-70 mmHg. Attempts were made to wean animals off vasopressin if MAP was maintained between 60-70 mmHg.

Arterial blood samples were collected from the femoral arterial catheter at baseline, one, three and six hours post-brain death induction. Blood was collected to measure troponin I, plasma urea and creatinine, liver function (serum bilirubin, alanine aminotransferase (ALT), aspartate aminotransferase (AST)), international normalised ratio (INR),

- 167 - amylase and lipase. These analyses (including urine creatinine, discussed below) were performed under contract by St. Vincent’s Pathology (SydPath), St. Vincent’s Hospital. Cardiac troponin I was assayed utilising the Bayer-Centaur Automated Chemiluminescence System (Bayer Healthcare Diagnostics, UK). Plasma urea and creatinine, bilirubin, ALT, AST, amylase and lipase were assayed utilising the Modular P 800 Consolidated Work Station (Roche Diagnostics, Australia), and INR was assayed utilising the STA-R Evolution Workstation (Diagnostica Stago, USA).

Arterial blood gas and blood glucose analyses were performed at hourly intervals, as described in Section 2.5.1. Pulmonary function was assessed utilising the arterial blood gas results as described in Section 2.7.6 and calculating the PaO2/FiO2 ratio and the alveolar-arterial oxygen (Aa) gradient. PaCO2 was also assessed.

Urine was collected from the urine collection bag immediately prior to brain death induction (baseline), and at one, three and six hours post-brain death induction. Urine creatinine was measured utilising the Modular P 800 Consolidated Work Station. The volume of urine production was also measured immediately prior to brain death induction (baseline), and at one, three and six hours post-brain death induction. The plasma creatinine, urinary creatinine, urine volume and time of urine production were used to calculate creatinine clearance utilising the following formula (McPherson 1997):

Creatinine Clearance = (Urinary Creatinine x Urine Volume) ÷ (Plasma Creatinine x Total Time of Urine Collection)

where creatinine clearance is expressed as mL/min, urinary and plasma creatinine are expressed as mmoles/L, urine volume is expressed in mL and time in minutes

Blood flow in the LAD, left renal artery, portal vein and hepatic artery were measured using flow probes and recorded at 15 minute intervals in the first hour after brain death induction, and then at hourly intervals up to six hours. Bile production was calculated at hourly intervals. Heart rate and MAP were also recorded at 15 minute intervals in the first hour after brain death induction, and then at hourly intervals up to six hours.

- 168 - Cardiac pressure and volume data was collected as described in Section 2.7.2 at baseline (pre-brain death), and at one, three and six hours post-brain death induction. These data were used to calculate stroke volume (SV), stroke work (SW) and cardiac output (CO).

Six hours after brain death induction, the final set of data, and tissue, bile, blood and urine were collected as described earlier. The experiment was then concluded and the animal euthanased with an intracardiac injection of potassium chloride (20 mmol) into the left ventricle to arrest the heart.

4.2.6 Study Outcomes

The outcome measurements for this study were donor haemodynamics, pulmonary function, hepatic function, pancreatic function and renal function. The haemodynamic indices examined were heart rate, MAP, SW, SV, CO and LAD blood flow. Troponin I was assessed as a marker of myocardial injury. Pulmonary function was assessed using arterial blood gas results as described in Section 2.7.6 (i.e. PaO2/FiO2 ratio, PaCO2 and the Aa gradient). Hepatic function and injury was assessed by bile production, hepatic artery, portal vein flow and INR, and liver function blood tests. Pancreatic injury was assessed by lipase and amylase levels in the blood, and renal function was assessed by renal arterial blood flow and creatinine clearance.

4.2.7 Power Calculation And Statistical Analyses

Apart from the study presented in Chapter 3 of this thesis, there was little information available for determining the potential impact of different strategies of donor management on transplantable organs other than the heart. Extending the findings of Chapter 3, it was hypothesised that the hormonal resuscitation protocol would result in better preservation of donor MAP, and consequently improve perfusion and function of transplantable organs compared with a noradrenaline-based protocol or one which involved fluid resuscitation alone. From Chapter 3, the difference in MAP between hormone resuscitation and noradrenaline-based treatments at 6 hours was 20 mmHg with a standard deviation of approximately 15 mmHg in each treatment group. A study with 10 animals in each treatment group had greater than 80% power (1 – beta) to detect

- 169 - a 20 mmHg difference in MAP between the hormone-treated and other treatment groups at a significance level (alpha) of 0.05.

The power calculation was based on the formula:

2 2 n = f(, ) 2 / (1 2)

where 1 and 2 are the mean outcome in the control and experimental group respectively, is the standard deviation and f(, ) = [-1(/2) + -1()]2. -1 is the cumulative distribution function of a standardised normal deviate (Pocock 1983; Julious 2004)

Statistical analyses were performed using SPSS for Windows 12.0.1 (SPSS Inc, USA) and StatView for Windows 4.57 (Abacus Concepts Inc, USA). Continuous variables were reported as mean±standard deviation. Non-normally distributed data were reported as median (range).

The statistical methods used to test for significant differences between groups or between time-points were determined by the type of data to be compared (for example, normal vs. non-normal distribution of data). These methods included analysis of variance (with or without repeated measures: ANOVA or ANOVA-RM). Non- parametrically distributed data were reported as median (range) and compared using the Kruskall-Wallis test. Significant differences were investigated further using post hoc Student’s t test for paired samples with a Bonferroni or Games-Howell correction for multiple comparisons, where applicable. All tests were applied as two-tailed tests. Differences were considered statistically significant at p<0.05.

4.3 RESULTS

4.3.1 Experimental Animals

Thirty animals were used in this study: ten in the FL group (seven Landrace pigs and three Westran pigs), ten in the NA group (eight Landrace pigs and two Westran pigs) and ten in the HR group (nine Landrace pigs and one Westran pig). Three animals, one

- 170 - in each experimental group, were excluded from the study for technical reasons: one had portal vein thrombosis, along with a burst Foley catheter balloon in the sub-dural space after inflation (which altered the behaviour and physiology of the brain-dead animal compared with the remaining animals with intact balloons due to changes in the intracranial pressure) and problems with the flow probe measuring LAD flow; one had an anaesthetic-related death prior to the commencement of surgery; and one had malfunctioning renal and hepatic artery flow probes preventing data collection, along with a malfunctioning syringe pump, which failed to deliver any vasopressin to an animal in the HR group. Therefore, there were nine animals in each experimental group included in the data analyses: FL group – two Westran and seven Landrace pigs; NA - two Westran and seven Landrace pigs; and HR group - one Westran and eight Landrace pigs.

Characteristics of the experimental groups are shown in Table 4.1. HR animals were significantly heavier than FL animals (p=0.024) and NA animals (p=0.020). However, there were no significant differences in left ventricular volume or in the left ventricular volume to body weight ratio between groups.

Table 4.1: Characteristics of the IV Fluids (control) (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups.

Characteristic FL Group NA Group HR Group p value

Animal Weight (kg) 41.5±5.7* 41.2±6.6# 50.4±7.5 0.009

Left Ventricular 119.4±22.7 111.7±15.4 126.1±16.9 0.276 Volume (mL)

Left Ventricular 2.9±0.4 2.7±0.3 2.5±0.3 0.123 Volume to Body Weight Ratio

Values are expressed as mean±standard deviation. *p=0.024 compared with HR; #p=0.020 compared with HR.

- 171 - Two hormone-treated animals had ventricular fibrillation arrests prior to brain death induction. These arrests occurred during cardiac instrumentation for data acquisition and all were successfully cardioverted to sinus rhythm. Additional doses of lignocaine were administered to animals that had arrested, and to one additional hormone-treated, one fluid-treated and two noradrenaline-treated animals for arrhythmias post-brain death induction.

4.3.2 Inotrope Requirements, Haemodynamic Indices And Cardiac Function

Noradrenaline doses required to maintain MAP in the NA group were non- parametrically distributed. At one hour post-brain death induction, all except one NA animal had MAP<60 mmHg. Noradrenaline was commenced in animals with MAP<60 mmHg to support blood pressure. At both three and six hours post-brain death, all NA animals required noradrenaline with a median dose of 0.476 (0.155-4.203) μg/kg/min at three hours and 3.101 (1.401-6.328) μg/kg/min at six hours.

The heart rate in each experimental group is shown in Figure 4.6a. At baseline (pre- brain death induction), the heart rate was higher in the NA group than in the HR group (92±11 vs. 76±12 bpm; p=0.037), although there were no significant differences at 15, 30 and 45 minutes post-brain death induction. Heart rate increased significantly after brain death induction compared with baseline (pre-brain death) in all three groups (p<0.0001). At one hour post-brain death induction, heart rate was higher in the NA group compared with the FL group (139±21 vs. 112±15 bpm; p=0.010). Heart rate remained consistently higher in the NA group compared with the FL and HR groups at two, three, four, five and six hours post-brain death: 149±26 vs. 107±10 and 112±15 bpm (p=0.002 and 0.007); 168±21 vs. 113±14 and 108±19 bpm (p<0.001 and 0.001); 164±17 vs. 110±14 and 115±24 bpm (p<0.001 and 0.001); 173±19 vs. 111±19 and 115±23 bpm (p<0.001 and 0.001); 167±28 vs. 116±25 and 113±21 bpm (p=0.001 and <0.001), respectively. There were no statistically significant differences in heart rate between FL and HR groups.

- 172 - MAP data are shown in Figure 4.6b. There were no significant differences between groups from baseline (pre-brain death) to two hours post-brain death. At three, four, five and six hours post-brain death induction, MAP was higher in the HR group compared with both the FL and NA groups: 66±4 vs. 45±8 and 50±9 mmHg (p<0.002); 64±6 vs. 42±12 and 48±17 mmHg (p=0.004 and 0.040); 66±5 vs. 41±13 and 43±15 mmHg (p=0.001 and 0.004); 64±5 vs. 38±12 and 36±14 mmHg (p<0.001), respectively. There were no significant differences in MAP between the FL and NA groups. Hormone resuscitation was commenced at one hour post-brain death induction in the HR group and at two hours after commencement of hormone resuscitation MAP was significantly higher in the HR group. There was a trend towards an increased MAP from baseline that was well maintained in the HR group, whereas there was a downward trend in MAP in both the FL and NA groups over time (as seen in Figure 4.6b).

CVP data are shown in Figure 4.6c. There were no significant differences in CVP between groups from baseline to six hours post-brain death induction. There were also no significant differences between groups with respect to the volume of IV saline infused over the six hours of brain death management. FL animals received 13594±3902 mL, NA animals received 14433±6514 mL and HR animals received 13511±4559 mL (p=0.915).

- 173 - # 200 a) Heart Rate * * * 190 * 180 * 170 ^ 160 150 140 130 120 110 Heart Rate (bpm) 100 90 80 #NA>HR, p<0.05;  70 ^NA>FL, p<0.05;  60 *NA>FL; NA>HR, p<0.05  50 0 60 120 180 240 300 360 80 b) Mean Arterial Blood Pressure * * * * 70

60

50

MAP (mmHg) MAP 40

30 *HR>FL; HR>NA, p<0.05  20 0 60 120 180 240 300 360 10 c) Central Venous Pressure 9 8 7 6 5 4

CVP (mmHg) CVP 3 2 1 0 -1 0 60 120 180 240 300 360 -2 Time Post-Brain Death (minutes) FL NA HR Figure 4.6: Haemodynamics. a) Heart Rate (bpm; beats per minute); b) Mean Arterial Blood Pressure (MAP); and c) Central Venous Pressure (CVP). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. See Section 4.3.2 for further details.

- 174 - SW data are shown in Figure 4.7a. SW was comparable between groups at baseline. One hour post-brain death induction SW was higher in HR animals than NA animals (4134±871 vs. 3076±957 mL.mmHg, p=0.015). SW was also higher in HR animals compared with NA animals at six hours post-brain death (3355±800 vs. 1343±872 mL.mmHg, p=0.003). There were no significant differences in SW between FL and NA or between FL and HR groups at any time-point. Over time, there was a significant decline in SW in the NA group from 2845±1132 at baseline to 1343±872 mL.mmHg at six hours (p=0.0021) and in the FL group, from 3030±678 to 2286±1585 mL.mmHg (p=0.0459) but not in the HR group (3873±1131 to 3355±800 mL.mmHg, p=0.4205).

SV data are shown in Figure 4.7b. Baseline SV was similar between groups. SV was higher in HR compared with NA animals at one hour (55±22 vs. 29±14 mL, p=0.013), three hours (53±20 vs. 23±16 mL, p=0.007) and six hours (50±17 vs. 12±10 mL, p<0.0001) post-brain death. SV was also higher in the FL group compared with the NA group at six hours post- brain death (36±22 vs. 12±17 mL, p=0.021). There were no significant differences between FL and HR groups. Of note, there was a 76% decline in SV from baseline to six hours in the NA group (from 49±21 mL at baseline to 29±14 mL at 1 hour, 23±16 mL at three hours and 12±10 mL at six hours post-brain death, p<0.0001). There were also statistically significant declines in SV in the HR and FL groups, although these were not as large as in the NA group. There was a 31% reduction in SV in the HR group from 72±22 mL at baseline to 50±17 mL at six hours (p=0.0006), and a 32% reduction in the FL group from 53±15 to 36±22 mL (p=0.0011).

CO data are shown in Figure 4.7c. There were no differences in CO at baseline. By one hour post-brain death, HR animals had a higher CO compared with NA animals (6186±2102 vs. 3880±1697 mL/min, p=0.048). At three hours there were no differences between groups but by six hours, CO was higher in the HR group compared with the NA group (5375±1404 vs. 1898±1480 mL/min, p<0.001). There were no differences between HR and FL or NA and FL groups at any time-point. Over time, there was a significant decline in CO in the NA group by 57% from 4807±1408 at baseline to 2085±1464 mL/min at six hours (p=0.0027) but not in the HR or FL groups (2% decline from 5826±1166 to 5718±1019 mL/min, p=0.1571 and 9% decline from 4703±1109 to 4297±2742 mL/min, p=0.2420, respectively).

- 175 - 5500 a) Stroke Work 5000 * 4500 * 4000 3500 3000 2500 2000 SW (mL.mmHg) SW 1500 1000 500 *HR>NA, p<0.02 0 0 60 120 180 240 300 360 100 b) Stroke Volume 90 80 * * * 70 # 60 50

SV (mL) SV 40 30 20 10 *HR>NA, p<0.02 #FL>NA, p<0.05  0 0 60 120 180 240 300 360 9000 * 8000 c) Cardiac Output * 7000 6000 5000 4000

CO (mL/min) 3000 2000 1000 *HR>NA, p<0.05 0 0 60 120 180 240 300 360 Time Post-Brain Death (minutes) FL NA HR Figure 4.7: Cardiac Function. a) Stroke Work (SW); b) Stroke Volume (SV); and c) Cardiac Output (CO). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. See Section 4.3.2 for further details.

- 176 - 4.3.3 Coronary Blood Flow And Troponin I Release

Mean blood flow in the left anterior descending coronary artery (LAD) in each group are shown in Figure 4.8. There were no differences in LAD flow between groups from baseline to one hour post-brain death induction. From two to six hours post-brain death induction LAD flow was higher in the NA group compared with the FL group: 28±10 vs. 16±7 mL/min (p=0.026), 29±9 vs. 17±8 mL/min (p=0.026), 38±21 vs. 18±10 mL/min (p=0.023), 41±13 vs. 20±11 mL/min (p=0.003) and 37±11 vs. 20±11 mL/min (p=0.008) at two, three, four, five and six hours post-brain death induction, respectively. At five and six hours post-brain death, LAD flow was also higher in the NA group compared with the HR group (41±13 vs. 21±11 mL/min, p=0.004 and 37±11 vs. 20±9 mL/min, p=0.009, respectively).

Troponin I levels increased progressively in all three groups over time. Whilst there were no significant differences between groups identified at any time-point (Table 4.2), there was a trend towards a higher troponin I release in the NA group compared with the FL and HR groups at six hours post-brain death (p=0.05).

Table 4.2: Troponin I (μg/L; median and range). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups.

Time FL NA HR p value

Pre-Surgery 0.02 (0-0.21) 0.08 (0-0.36) 0 (0-0.12) 0.173

Post- 0.76 0.59 1.1 0.350 Instrumentation (0.38-1.41) (0.38-1.56) (0.4-9.67)

1 h Post-Brain 1.36 1.1 1.92 0.322 Death (0.41-2.35) (0.64-2.51) (0.7-14.13)

3 h Post-Brain 1.9 3.36 2.69 0.300 Death (0.70-3.83) (1.16-8.32) (1.37-16.95)

6 h Post-Brain 3.14 10.18 3.99 0.050 Death (1.31-7.42) (3.48-55.69) (1.99-24.68)

- 177 - 60 * * # 55 * 50 # 45 40 * * 35 30 25 20 LAD Flow (mL/min) 15 10 5 *NA>FL, p<0.05 #NA>HR, p<0.01  0 0 60 120 180 240 300 360 Time Post-Brain Death (minutes) FL NA HR

Figure 4.8: Left Anterior Descending Coronary Artery (LAD) Blood Flow. IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between FL and HR groups. See Section 4.3.3 for further details. 

- 178 - 4.3.4 Pulmonary Function

Pulmonary function was assessed utilising arterial blood gas results. The results of each experimental group with respect to PaO2/FiO2 ratio, the Aa gradient and PaCO2 are shown in Figure 4.9. There were no significant differences in the PaO2/FiO2 ratio between groups from baseline to five hours post-brain death induction (Figure 4.9a). At six hours post-brain death, NA animals had a significantly lower PaO2/FiO2 ratio compared with the FL and HR animals (345±106 vs. 464±63 and 470±95, p=0.027 and

0.020). There was a significant decline in the PaO2/FiO2 ratio in the NA group over time by 29% from 486±79 at baseline to 345±106 at six hours post-brain death (p=0.0004), whereas the PaO2/FiO2 ratio was stable over time in the HR or FL groups (p=0.1083 and 0.954, respectively). Similarly, analysis of the Aa gradient demonstrated that there were no significant differences between groups from baseline to five hours post-brain death induction (Figure 4.9b). However, at six hours post-brain death, NA animals had a significantly higher Aa gradient compared with the FL and HR animals (311±101 vs. 199±62 and 192±91 mmHg, p=0.031 and 0.021). There was a significant increase in the Aa gradient over time in the NA group, which nearly doubled from 156±93 mmHg at baseline to 311±101 mmHg at six hours (p=0.0005). As with the PaO2/FiO2 ratio, there was no change in the Aa gradient over time in either the HR or FL groups (p=0.1343 and 0.9754, respectively). There were no significant differences between groups at any time-point with respect to PaCO2 (Figure 4.9c).

4.3.5 The Liver: Blood Flow, Injury And Function

Blood flow in the hepatic artery and the portal vein are shown in Figure 4.10. There were no significant differences in hepatic arterial blood flow between groups from baseline to six hours post-brain death induction. There were also no significant differences in portal venous blood flow between groups from baseline to six hours post- brain death.

Bile production in each experimental group is shown in Figure 4.11. There were no significant differences in bile production between groups from baseline to six hours post-brain death induction.

- 179 - 650 a) PaO2/FiO2 Ratio 600 * 550 500 450 400 350 PaO2/FiO2 Ratio 300 250 *FL>NA; HR>NA, p<0.05  200 0 60 120 180 240 300 360 450 b) Alveolar-Arterial (Aa) Gradient * 400 350 300 250 200 150 Aa Gradient (mmHg) 100 50 *NA>FL; NA>HR, p<0.05  0 0 60 120 180 240 300 360 60 c) PaCO2

50

40

PaCO2 (mmHg) 30

20 0 60 120 180 240 300 360 Time Post-Brain Death (minutes) FL NA HR

Figure 4.9: Pulmonary Function. a) PaO2/FiO2 Ratio; b) Alveolar-arterial (Aa) Gradient (mmHg); and c) PaCO2 (mmHg). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. See Section 4.3.4 for further details.  - 180 - 360 a) Hepatic Artery Flow 320

280

240

200

160

120 Hepatic Artery Flow (mL/min) Hepatic

80

40

0 0 60 120 180 240 300 360 1600 b) Portal Vein Flow

1400

1200

1000

800

600 Portal Vein Flow (mL/min) Vein Portal 400

200

0 0 60 120 180 240 300 360 Time Post-Brain Death (minutes) FL NA HR Figure 4.10: Blood Flow To The Liver. a) Hepatic Artery Blood Flow (mL/min); and b) Hepatic Portal Vein Blood Flow (mL/min). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between groups at any time-point for both hepatic arterial and portal venous flow.

- 181 - 40

35

30

25

20

15 Bile Production (mL/hr) Bile Production 10

5

0 0 60 120 180 240 300 360 Time Post-Brain Death (minutes)

FL NA HR

Figure 4.11: Bile Production (mL/hr). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between groups at any time-point.

- 182 - Liver function test results (ALT, AST and total bilirubin) are shown in Figure 4.12. There were no significant differences in any of the three components of the liver function test between groups at any time-point from baseline to six hours post-brain death induction. The INR for each experimental group is shown in Figure 4.13. There were also no differences in INR between groups at any time-point.

4.3.6 Blood Glucose Control And Exogenous Insulin Requirements

Blood glucose data for the experimental groups are shown in Figure 4.14. There was no significant difference in blood glucose levels between groups at baseline. There were also no differences between groups at one, two and three hours post-brain death induction. At four hours post-brain death, blood glucose was higher in the NA group compared with the FL group (7.4±4.2 mmol/L vs. 4.2±1.1 mmol/L; p=0.0164). At five hours post-brain death, blood glucose was higher in the NA group compared with the FL and HR groups (8.9±4.2 vs. 4.2±1.0 and 5.5±1.4; p=0.0009 and p=0.0117, respectively). Similarly, at six hours post-brain death, blood glucose was higher in the NA group compared with the FL and HR groups (10.4±4.2 vs. 5.0±1.5 and 5.9±1.4; p=0.0003 and p=0.0019, respectively). There were no significant differences between the FL and HR groups at either five or six hours after brain death induction.

In the HR group, no animals required insulin for hyperglycaemia (i.e. blood glucose over 10 mmol/L). Of all the animals in the three experimental groups, there were only three animals that had blood glucose levels above 10 mmol/L. All three animals were from the NA group.

- 183 - 45 a) Alanine Aminotransferase 40

35

30

25

20 ALT (U/L) ALT 15

10

5

0 -180 -120 -60 0 60 120 180 240 300 360 180 160 b) Aspartate Aminotransferase 140 120 100 80

AST (U/L) AST 60 40 20 0 -180 -120 -60 0 60 120 180 240 300 360 11 c) Total Bilirubin 10 9 8 7 6 5 4 3 Total Bilirubin (μmol/L) Total 2 1 0 -180 -120 -60 0 60 120 180 240 300 360 Time Post-Brain Death (minutes) FL NA HR Figure 4.12: Liver Function Tests. a) Alanine Aminotransferase (ALT) (U/L); b) Aspartate Aminotransferase (AST) (U/L); and c) Total Bilirubin (μmol/L). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between groups at any time-point.

- 184 - 1.5

1.4

1.3

1.2 INR 1.1

1

0.9

0.8 -180 -120 -60 0 60 120 180 240 300 360 Time Post-Brain Death (minutes) FL NA HR

Figure 4.13: International Normalised Ratio (INR). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between groups at any time-point.

- 185 - # 15 * 14 # * 13 12 * 11 10 9 8 7 6 5 4 3 Blood Glucose Level (mmol/L) 2 *NA>FL, p<0.02; 1 #NA>HR, p<0.002 0 0 60 120 180 240 300 360 Time Post-Brain Death (minutes) FL NA HR

Figure 4.14: Blood Glucose Levels (mmol/L). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. See Section 4.3.6 for further details. 

- 186 - 4.3.7 Pancreatic Injury

Pancreatic enzyme levels for each experimental group over time are shown in Figure 4.15. There were no significant differences in amylase levels between any of the experimental groups (Figure 4.15a). There was a trend towards declining levels of amylase over the duration of brain death management. Similarly, there were no differences in lipase levels between experimental groups over time (Figure 4.15b). In contrast however, there was a trend towards increasing levels of lipase over time.

4.3.8 Renal Function

Blood flow through the left renal artery is shown in Figure 4.16a. From baseline (pre- brain death induction) to three hours post-brain death there were no significant differences between groups. However, at four, five and six hours post-brain death induction, renal arterial blood flow was significantly higher in the HR group compared with the NA group: 210±91 vs. 101±64 mL/min (p=0.019), 209±93 vs. 73±63 mL/min (p=0.004) and 204±96 vs. 41±42 mL/min (p<0.001), respectively. In addition, at six hours post-brain death, renal arterial flow was higher in the HR group compared with the FL group (204±96 vs. 115±68 mL/min, p=0.048). Over time, there was a significant decline in renal arterial flow in the FL group by 41% (p<0.0001) and in the NA group by 80% (p<0.0001), whereas in the HR group there was no significant change in flow (p=0.7773).

There were no significant differences between groups with respect to creatinine clearance in the one hour prior to brain death induction and during the first hour after brain death induction. Creatinine clearance was significantly higher in the HR group compared with the NA group at both the one to three hour interval and at the three to six hour interval post-brain death induction (Figure 4.16b). At three hours, the creatinine clearance was 160.9±78.0 mL/min in the HR group compared with 84.2±46.0 mL/min in the NA group (p=0.034), and at six hours it was 132.5±54.4 vs. 40.4±34.3 mL/min (p=0.002). There was a significant decline in creatinine clearance from baseline to six hours in the FL group by 44% (p=0.0212) and in the NA group by 66% (p<0.0001), whereas there was no significant change over time in the HR group (decline by 36%, p=0.2820).

- 187 - 3000 a) Amylase 2800 2600 2400 2200 2000 1800 1600 1400 Amylase (U/L) 1200 1000 800 600 400 200 0 -180 -120 -60 0 60 120 180 240 300 360 14 13 b) Lipase 12 11 10 9 8 7

Lipase (U/L) 6 5 4 3 2 1 0 -180 -120 -60 0 60 120 180 240 300 360 Time Post-Brain Death (minutes) FL NA HR Figure 4.15: Pancreatic Enzymes. a) Amylase (Units/L); and b) Lipase (Units/L). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between groups at any time- point.

- 188 - 400 a) Left Renal Artery Blood Flow 360

320 # * * * 280

240

200

160

120 Left Renal Artery Blood Flow (mL/min) Left Renal 80

40 *HR>NA, p<0.02; #HR>FL, p<0.05 0 0 60 120 180 240 300 360 360 340 b) Creatinine Clearance 320 300 280 260 * 240 220 200 * 180 160 140 120 100

Creatinine Clearance (mL/min) Creatinine 80 60 40 20 *HR>NA, p<0.05 0 0 60 120 180 240 300 360 Time Post-Brain Death (minutes) FL NA HR Figure 4.16: Renal Function. a) Left Renal Artery Blood Flow (mL/min); and b) Creatinine Clearance (mL/min). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. See Section 4.3.8 for further details.

- 189 - Plasma creatinine and urea data are shown in Figure 4.17. There were no significant differences in plasma creatinine between experimental groups at any time-point. Similarly, there were no differences in plasma urea between groups.

4.3.9 Acid-Base Balance

Acid-base balance data (i.e. pH and base excess) for each experimental group is shown in Figure 4.18. There were no significant differences in pH between groups at baseline, one and two hours post-brain death (Figure 4.18a). At three hours post-brain death pH was higher in the FL group compared with the NA group (7.38±0.07 vs. 7.30±0.08; p=0.0071) and was also higher in the HR group compared with the NA group (7.39±0.04 vs. 7.30±0.08; p=0.0042). Similar trends were seen at four, five and six hours post-brain death with pH being higher in the FL and HR groups compared with the NA group: 7.32±0.07 and 7.33±0.06 vs. 7.21±0.10 (p=0.0104 and 0.0057) at four hours; 7.30±0.05 and 7.33±0.08 vs. 7.16±0.09 (p=0.001 and 0.0002) at five hours; and 7.27±0.03 and 7.31±0.08 vs. 7.08±0.10 (p<0.0001) at six hours, respectively.

There were no significant differences in base excess between groups at baseline, one and two hours post-brain death (Figure 4.18b). At three hours post-brain death, the base excess was higher in the HR group compared with the NA group (-1.44±3.09 vs. - 5.67±3.50; p=0.0136). Similarly at four hours, the base excess was higher in the HR group compared with the NA group (-3.67±3.20 vs. -9.33±4.24; p=0.0021). By the fifth and sixth hours post-brain death, the base excess was higher in both FL and HR groups compared with NA: -6.89±2.62 and -4.72±3.75 vs. -12.78±4.41 (p=0.0023 and p<0.0001) at five hours respectively, and -8.33±1.94 and -5.33±4.00 vs. -16.22±4.24 (p<0.0001) at six hours respectively.

- 190 - 140 a) Plasma Creatinine 130

120

110

100

90

80

70

60 Plasma Creatinine (μmol/L) Plasma Creatinine 50

40

30 -180 -120 -60 0 60 120 180 240 300 360

6.5 b) Plasma Urea 6

5.5

5

4.5

4

3.5

Urea (mmol/L) Urea 3

2.5

2

1.5

1 -180 -120 -60 0 60 120 180 240 300 360 Time Post-Brain Death (minutes) FL NA HR Figure 4.17: Renal Function. a) Plasma Creatinine (μmol/L); and b) Plasma Urea (mmol/L). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. There were no significant differences between groups at any time-point. 

- 191 - 7.60 a) pH 7.55 7.50 * 7.45 * 7.40 * * 7.35 7.30 7.25 pH 7.20 7.15 7.10 7.05 7.00 6.95 *FL>NA, p<0.02; HR>NA, p<0.005 6.90 0 60 120 180 240 300 360

6 b) Base Excess 4 2 * * # # 0 * * -2 -4 -6 -8 BE -10 -12 -14 -16 -18 *HR>NA, p<0.02; -20 #FL>NA, p<0.005 -22 0 60 120 180 240 300 360 Time Post-Brain Death (minutes) FL NA HR

Figure 4.18: Acid-Base Balance. a) pH; and b) Base Excess (BE). IV Fluids (FL; n=9), Noradrenaline (NA; n=9) and Hormone Resuscitation (HR; n=9) Groups. See Section 4.3.9 for further details.

- 192 - 4.4 DISCUSSION

4.4.1 Study Rationale And Experimental Groups

Hormone resuscitation of the brain-dead organ donor has been endorsed in the UNOS Critical Pathway donor management protocol (Rosengard et al. 2002; Zaroff et al. 2002). This recommendation was made based on reports describing the beneficial effects of hormone resuscitation in ameliorating the negative effects of brain death in the organ donor in both animal models and humans (Novitzky et al. 1986; Novitzky et al. 1987b; Novitzky et al. 1987c; Novitzky et al. 1987d; Novitzky et al. 1988c; Novitzky et al. 1988d; Novitzky et al. 1990; Jeevanandam et al. 1993; Jeevanandam et al. 1994; Wheeldon et al. 1995; Hicks et al. 2006). Subsequent retrospective analyses have shown that hormone resuscitation increases the yield of transplantable organs from donors without any reduction in organ quality and in some cases, better survival and function (Rosendale et al. 2002; Rosendale et al. 2003a; Rosendale et al. 2003b; Rosendale et al. 2004).

Chapter 3 of this thesis described a prospective, randomised controlled study investigating combined hormone resuscitation in the brain-dead donor. Utilising a porcine model of the brain-dead cardiac donor, it demonstrated that combined hormone resuscitation reduced inotropic requirements to support blood pressure, and improved both haemodynamics and cardiac function compared with conventional management with noradrenaline and intravenous fluids (Hing et al. 2007). However, despite the evidence supporting the beneficial effects of hormone resuscitation, there have been few prospective studies investigating the effects of hormone treatment on transplantable non-cardiac organs, such as the lungs, kidney, liver and pancreas. As such, the study described in this chapter was designed to investigate the effects of various donor management protocols on non-cardiac organs in a prospective, randomised controlled study.

Despite the fact that HR animals were heavier than animals from the FL and NA groups, there were no differences in left ventricular volume or in left ventricular volume to body weight ratio. There were also no significant differences in any of the baseline measurements of haemodynamic function such as blood pressure, CVP, SW and CO;

- 193 - pulmonary function such as Aa gradient and PaO2/FiO2 ratio; organ blood flows such as LAD, renal artery, hepatic artery and portal vein flows; creatinine clearance; plasma urea and creatinine; acid-base balance (i.e. pH and base excess); pancreatic injury (i.e. amylase and lipase); or hepatic injury (i.e. liver function tests).

4.4.2 Hormone Resuscitation To Improve Haemodynamics And Cardiac Function

Heart rate was significantly higher in the NA group compared with the HR group at baseline (mean difference of 16 beats per minute – bpm) and the FL group at one hour post-brain death induction (mean difference 27 bpm) (Figure 4.6a). However, the magnitude of the difference in heart rate in the NA group compared with the HR and FL groups was much higher from two to six hours post-brain death induction, ranging from 37 to 62 bpm. In addition, the heart rate after induction of brain death was significantly higher than at baseline (pre-brain death) in all three groups. This result is consistent with other studies, reflecting the impact of brain death (Novitzky et al. 1984; Ryan et al. 2003c). The higher heart rate was consistent with the known chronotropic effects of noradrenaline in the NA group and the escalating dose used – median dose of 0.476 μg/kg/min at three hours up to 3.101 μg/kg/min at six hours (Rudis et al. 1996; Di Giantomasso et al. 2002).

In concordance with the results in Chapter 3 and reported by Hing et al (2007), MAP was significantly higher in the hormone-treated animals compared with the FL and NA groups (Figure 4.6b). This difference was seen two hours after the commencement of hormone resuscitation (i.e. three hours after brain death induction) and remained elevated over the final three hours of the study (between 63±11 to 66±5 mmHg). Despite escalating doses of noradrenaline over time in the NA group, up to 3.101 μg/kg/min, MAP could not be maintained above 60 mmHg and in fact declined over time, from 59±6 mmHg at two hours post-brain death induction to 36±14 mmHg at six hours. Similarly, MAP declined over time in the FL group from 52±11 mmHg at two hours to 38±12 mmHg at six hours. As there were no significant differences in the total volume of intravenous fluids given to each of the three groups and considering that there were no other differences in treatment other than the experimental protocol, it can

- 194 - be concluded that the hormone resuscitation was responsible for the superior blood pressure in donor animals. Interestingly, the use of noradrenaline in the NA group did not produce any improvement in blood pressure than IV fluids alone (FL group) in the study, particularly as the dose of noradrenaline was escalated to exceedingly high levels. This suggests that the animals developed tachyphylaxis to noradrenaline, a phenomenon that was observed and discussed in the earlier study described in Chapter 3. In clinical practice however, noradrenaline remains one of the mainstays of treatment to support blood pressure in the brain-dead donor, with 81% of donors in Australia and New Zealand receiving noradrenaline in 2007 (Excell et al. 2008).

Whilst there was a trend towards a higher CVP in the NA group compared with the HR and FL groups, there were no statistically significant differences at any time-point during the study (Figure 4.6c). There were also no significant differences in the volume of intravenous fluid given to each of the three experimental groups: 13594±3902 mL in the FL group; 14433±6514 mL in the NA group; and 13511±4559 mL in the HR group.

Left ventricular contractility, as measured by the PRSW relationship, was not analysed in this study. This was because contractility had already been investigated in the earlier study described in Chapter 3, which demonstrated that hormone resuscitation in the donor produced superior left ventricular contractility compared with treatment with noradrenaline. However, other measures of cardiac function such as SW, SV and CO were examined. Six hours after induction of brain death, SW was 150% higher in the HR group compared with the NA group (3355±800 vs. 1343±872 mL.mmHg, p=0.003), SV was 317% higher (50±17 vs. 12±10 mL, p<0.00010), and CO was 183% higher (5375±1404 vs. 1898±1480 mL/min, p<0.001), respectively (Figure 4.7). Whilst CO was stable over time in the HR and FL groups, it declined significantly in the NA group by 57% from 4807±1408 mL/min at baseline to 2085±1464 mL/min at six hours (p=0.0027). Similarly SW did not change significantly in the HR group over time but declined significantly in the NA group by 53% from 2845±1132 mL.mmHg at baseline to 1343±8725 mL.mmHg at six hours (p<0.0001), and in the FL group by 25% from 3030±678 to 2286±1585 mL.mmHg (p=0.0459). These results demonstrate that hormone treatment of the donor provides superior preservation of cardiac haemodynamics compared with IV fluids or noradrenaline treatment. In addition, noradrenaline treatment was associated with significant declines in SW, SV and CO

- 195 - over time despite increasing noradrenaline doses, similar to that seen with MAP. When compared with the haemodynamic trends seen in Chapter 3, the results presented in this chapter further validate the results seen in the earlier study.

In contrast to the earlier study described in Chapter 3, there were significant differences seen in LAD flow (Figure 4.8). The differences in LAD results seen between the two studies (Chapter 3 and Chapter 4) may be due to the fact that animals in this study were treated for longer with hormone resuscitation (five hours rather than three hours in Chapter 3). Noradrenaline was associated with significantly higher LAD flows compared with FL animals from two hours post-brain death induction to six hours. At five and six hours, LAD flow was also higher in the NA group compared with HR animals. There was also a significant increase in LAD flow by 131% over time in the NA group from 16±6 mL/min to 37±11 mL/min (p<0.0001). These findings were consistent with the expected adrenergic effects of noradrenaline on  and adrenoceptors, which has been demonstrated to cause increases in coronary artery blood flow (Di Giantomasso et al. 2002; Tune et al. 2002).

4.4.3 Myocardial Injury

Troponin I has been previously demonstrated as a sensitive marker of myocardial injury in both humans and in a porcine transplantation model (Potapov et al. 2001; Ryan et al. 2003d), and was therefore examined in this study. Furthermore, troponin I has been reported to be elevated after brain death in pigs as a result of myocardial ischaemia (Chiari et al. 2000) and therefore this study investigated potential treatments to ameliorate the rise in troponin as a result of brain death.

Although there were no statistically significant differences in troponin release between experimental groups, there was a trend towards a higher troponin in the NA group compared with the HR and FL groups at both three and six hours after brain death induction (Table 4.2). Interestingly, the NA group had higher median troponin levels despite two HR animals (and no FL or NA animals) having ventricular fibrillation arrests requiring DC cardioversion. It might be expected that DC cardioversion would cause cardiac myocyte injury and hence elevate troponin levels. However this was not

- 196 - seen in this study, further reinforcing the significance of the troponin rise in the NA group.

By six hours, the median troponin I release in the NA group was 10.18 μg/L compared with 3.14 and 3.99 μg/L in the FL and HR groups (p=0.050). A small rise in troponin was seen in all groups after surgical instrumentation, which likely reflects myocyte injury from surgery. However, it is generally accepted that troponin is detectable on average approximately four to six hours after myocyte injury begins (McPherson 1997; Sarko et al. 2002). Therefore, animals in this study may not have been observed for long enough to detect significant differences in troponin levels, and hence assess the effects of the various treatments on myocardial injury. The wide range of troponin values in each group at each time-point also suggests that the study was underpowered to detect differences in troponin I release. This may also explain the differences in troponin results in this chapter compared with those seen in Chapter 3. In addition, animals in this study were treated with hormone resuscitation for an additional two hours compared with animals in Chapter 3 and hence had longer to develop differences.

Despite the limitations discussed in interpreting the troponin results, the NA group of animals had the lowest MAP, SW, SV and CO. These findings are consistent with the trend towards higher troponin release in NA animals and hence greater myocardial injury in this group, resulting in decreased cardiac function. More importantly, troponin levels in the donor have previously been shown to be a predictor of early graft failure in heart transplantation and thus donor troponin levels can be of importance in selecting suitable hearts for transplantation (Potapov et al. 2001; Potapov et al. 2003). Potapov et al (2001) demonstrated that troponin I levels greater than 1.6 μg/Lin the donor were associated with an odds ratio for the development of acute graft failure of 42.7 with a specificity of 94%. Therefore, any treatments that minimise the rise in donor troponin levels will potentially be important to improving transplantation outcomes.

4.4.4 The Negative Effects Of Noradrenaline On Pulmonary Function

Primary graft failure is one of the main causes of early deaths in lung transplantation, causing 28.2% of deaths in the first 30 days post-transplantation (Christie et al. 2008). A major factor behind these early deaths is ischaemia-reperfusion injury and the

- 197 - damaging effects of brain death on the lungs directly (Fisher et al. 1998; Fisher et al. 2001; Avlonitis et al. 2003; Avlonitis et al. 2007). Brain death causes acute lung injury by neurogenic pulmonary oedema and by inflammatory acute lung injury (Avlonitis et al. 2003). Therefore, a potential intervention to reduce the incidence of primary lung graft failure could be the development of treatments to ameliorate the injury caused by brain death and ischaemia-reperfusion injury.

The results of this study demonstrated that the treatment of brain-dead animals with either intravenous fluids or hormone resuscitation did not significantly change oxygenation. The PaO2/FiO2 ratio (Figure 4.9a) and the Aa gradient (Figure 4.9b) at six hours post-brain death induction were no different from baseline in the FL (p=0.954 and 0.9754, respectively) and HR groups (p=0.1083 and 0.1343, respectively). Noradrenaline treatment of the donor however, was associated with a significant deterioration in oxygenation. The PaO2/FiO2 ratio fell by 29% from 486±79 at baseline to 345±106 at six hours (p=0.0004) and the Aa gradient nearly doubled from 156±93 mmHg to 311±101 mmHg (p=0.0005). There were no significant differences in PaCO2 between groups at any time-point, suggesting that the treatments tested did not have any significant effect on CO2 levels and that ventilation was equivalent across the groups. In fact, the PaCO2 levels were essentially normal in all groups at all times, averaging from 37±4 to 46±6 mmHg.

These results were consistent with earlier clinical studies. In a clinical setting, the use of catecholamines such as noradrenaline, adrenaline and dopamine have been associated with worse gas exchange after lung transplantation (Mukadam et al. 2005). In a retrospective analysis of 60 lung transplant donors and recipients, Mukadam and colleagues demonstrated that there was a greater reduction in the PaO2/FiO2 ratio in catecholamine-treated donors that was independent of ischaemic time, preservation technique and recipient diagnosis. In a randomised controlled trial investigating hormone treatments of the donor and aggressive donor management (including ventilation and haemodynamic optimisation and bronchoscopy), the use of noradrenaline was associated with deteriorating PaO2/FiO2 ratio and extravascular lung water index (EVLWI) (Venkateswaran et al. 2008). It has also been suggested that noradrenaline released from the nerve endings of the sympathetic nervous system may

- 198 - be a mediator of neurogenic pulmonary oedema seen after brain death (Avlonitis et al. 2003).

In contrast to the aforementioned studies, it has been hypothesised that catecholamines may reduce the impact of ischaemia-reperfusion injury by modulation of leukocyte adhesion mechanisms (Schnuelle et al. 1999), and may promote alveolar fluid clearance, thereby promoting recovery from pulmonary oedema (Maron et al. 1994; Berthiaume et al. 2002; Matthay et al. 2002; Azzam et al. 2004). In an animal model, it has been shown that noradrenaline improves oxygenation and limits pulmonary oedema after brain death (Avlonitis et al. 2005; Rostron et al. 2008).

In terms of hormone resuscitation, T3 has been shown to increase alveolar fluid clearance (Folkesson et al. 2000). Conversely, in a randomised controlled trial, combined methylprednisolone and T3 or T3 treatment alone were shown not to affect lung yield, PaO2/FiO2 ratio or EVLWI (Venkateswaran et al. 2008). This study showed that methylprednisolone did reduce EVLWI. However, by the author’s own admission, the results from the study should be interpreted cautiously, given the small numbers involved and the potential for the study to be underpowered to detect differences between groups. Other studies have also demonstrated the benefit of corticosteroids in the management of the donor. In a retrospective analysis of 118 consecutive organ donors, 80 donors received high dose methylprednisolone (14.5±0.06 mg/kg), similar to the doses used in the study reported in this chapter, and were managed for approximately 23.5 hours (Follette et al. 1998). Steroid-treated donors had a significant and progressive increase in PaO2/FiO2 ratio (by 16±14) and also had an increased yield of transplantable lungs from the available donors. In a retrospective review of organ donors, corticosteroid usage was an independent predictor of lung utilisation for transplantation (McElhinney et al. 2001) and in a UNOS retrospective review, triple hormone resuscitation (methylprednisolone, thyroid hormone and vasopressin) was associated with a 2.8% increased probability of lungs being transplanted from a donor (Rosendale et al. 2003b). Despite these reports, the results of the study presented here did not demonstrate any benefit on oxygenation from hormone resuscitation, with no change in PaO2/FiO2 ratio or Aa gradient from baseline to six hours post-brain death, and no difference compared to the fluid-treated animals (FL). Perhaps a longer period of management and observation after brain death may have revealed differences between

- 199 - treatment groups. Irrespective of this, however, this study has demonstrated that hormone resuscitation does not have any detrimental effects on the lungs and their ability to oxygenate within the time frame investigated.

4.4.5 Does Hormone Resuscitation Affect The Liver Any Differently To Conventional Treatments Used In Donor Management?

Brain death in the organ donor is known to provoke and sustain inflammatory changes that have detrimental effects on the outcomes of liver transplantation in both humans and animal models (Pratschke et al. 1999; van der Hoeven et al. 2001; Weiss et al. 2007). Donor brain death has been shown to cause up-regulation of inflammatory cytokines, increased cellular infiltrates and an increased rate of apoptosis in humans and animal models (Takada et al. 1998; van der Hoeven et al. 1999b; Jassem et al. 2003; Weiss et al. 2007). Brain death has also been shown to reduce allograft survival of rat livers after prolonged ischaemia compared with grafts from living donors (van der Hoeven et al. 2003). In a prospective study of liver biopsies taken from 32 brain-dead and 26 living liver donors, brain death was associated with significantly higher expression of inflammatory cytokines such as IL-6, IL-10, TNF-, TGF- and MIP-1 and increased cellular infiltrates (Weiss et al. 2007). This study also demonstrated that livers from brain-dead donors experienced significantly greater ischaemia-reperfusion injury (as reflected by elevated levels of ALT, AST and bilirubin in the recipient), increased rates of acute rejection and primary non-function, and significantly reduced production and excretion of bilirubin in the recipient. Golling and colleagues demonstrated in a porcine model of the brain-dead liver donor that brain death caused deterioration in portal venous flow, microperfusion (assessed with an intrahepatic thermal diffusion probe), AST and hepatic oxidative stress, which were independent of haemodynamic stability (Golling et al. 2003). Deterioration in hepatic arterial flow was also seen in this study, but only in hypotensive brain-dead donors. Given these negative effects of brain death on the donor liver, it is important to examine treatments that may ameliorate these effects. Of interest is whether hormone resuscitation has any role in protecting donor livers, given its effects on the cardiac donor and their haemodynamics.

- 200 - Despite the beneficial effects of hormone resuscitation in the brain-dead donor, demonstrated in both humans and animal models, on cardiac function, haemodynamics and outcomes in heart and renal transplantation (Rosendale et al. 2003a; Rosendale et al. 2003b; Novitzky et al. 2006), there has been little published on the effects of hormone resuscitation on the donor liver. Published and unpublished data from the Medical University of South Carolina liver transplant program has suggested that hormone resuscitation has adverse effects on liver transplant outcomes, particularly when steatotic livers (where there is hepatocyte lipid accumulation) are used (Ellett et al. 2008). It was reported that hormone resuscitation (including steroid and levothyroxine (T4)) caused higher peak transaminases, a greater rate of graft dysfunction and failure, and a greater need for postoperative plasmapheresis (Skerrett et al. 1996; Mandal et al. 2000; Ellett et al. 2008). In a mouse model of steatotic livers exposed to ischaemia-reperfusion injury, treatment with T4 and methylprednisolone 48 hours prior to ischaemia was associated with decreased survival and a dramatic increase in liver necrosis (Ellett et al. 2008). This study also found a dramatic increase in the expression of uncoupling protein-2, a protein that has been implicated as a major mediator of ischaemia-reperfusion injury in steatotic animals, accompanied by a decrease in ATP levels after reperfusion. These findings however, should be interpreted with caution. The model used was neither a transplantation model nor did it incorporate brain death. Given that hormone resuscitation has been designed to ameliorate the negative effects of brain death, the model used in the study was not an ideal model for testing treatments in brain-dead donor management or transplant outcomes. In addition, as Novitzky et al have pointed out, extrapolation of results in a rodent model to the clinical situation should be approached cautiously without confirmatory evidence from a large animal model (Novitzky et al. 2008).

In the study presented in this chapter, there was no evidence to suggest that hormone resuscitation was associated with any deterioration in liver function compared with more traditional treatments consisting of intravenous fluids or noradrenaline. There were no significant differences in blood flow to the liver between any of the treatment groups, as measured by flow through the hepatic artery and the hepatic portal vein (Figure 4.10). This is of particular importance as there have been concerns raised in the literature that a disadvantage of vasopressin is its powerful, non-selective vasoconstrictive effects, which may impair hepatic perfusion (Pennefather et al. 1995).

- 201 - In addition, vasopressin potentiates the vasoconstrictor effects of many agents, including noradrenaline (Holmes et al. 2003). Despite using much higher noradrenaline doses (up to 3.101 μg/kg/min) in a brain–dead model, the findings in the present study are consistent with the report by Di Giantomasso et al. They investigated the effects of noradrenaline in a conscious merino ewe model and demonstrated that at 0.4 μg/kg/min, noradrenaline did not alter superior mesenteric arterial blood flow, although there was mesenteric vasoconstriction (as demonstrated by a 20% increase in mesenteric vascular resistance) (Di Giantomasso et al. 2002). Vasopressin has been reported to cause a biphasic response in hepatic artery blood flow that is manifested by an initial sharp decrease in flow followed by a more sustained increase within minutes (Bynum et al. 1980). This was not seen in the study reported here and, as previously mentioned, hormone-treated animals did not have significantly different hepatic arterial flow compared with any other group. This was in the context of HR animals receiving an average infusion of 0.6±1.3 to 1.7±1.5 U/hr of vasopressin. These flow results also coincide with other experimental animal studies that have demonstrated that, at physiological concentrations, vasopressin has selective effects affecting predominately the cutaneous and skeletal circulations, rather than the hepatic circulation (Liard et al. 1982; Goldsmith 1987). Additionally, there is some evidence to suggest that low-dose vasopressin is effective in maintaining hepatic energy metabolism in brain-dead dogs (Manaka et al. 1990; Manaka et al. 1992) and that hepatic function is well preserved in human donors supported with vasopressin (Yoshioka et al. 1986; Nagareda et al. 1989).

Hepatic injury, as assessed by the liver enzymes ALT and AST, and bilirubin was not significantly different between any of the treatment groups at any time-point (Figure 4.12). Similarly, measurements of hepatic synthetic function as measured by bile production and INR were also not significantly different between any of the treatment groups at any time-point (Figures 4.11 and 4.13, respectively). The INR results however, should be interpreted cautiously. Given that factor VII has the shortest half- life of all the clotting factors of approximately 3-5 hours (Beck 1991) and in the absence of any major bleeding that would increase consumption of clotting factors, any increase in INR in the relatively short observation time of six hours used in this study would not be expected.

- 202 - In summary, when compared with treatment with intravenous fluids alone or noradrenaline, hormone resuscitation was not associated with any significant differences in blood flow to the liver, bile production, liver enzymes or INR. This suggests that hormone resuscitation is not detrimental to the liver when compared with other more conventional treatments employed clinically to maintain the brain-dead organ donor.

4.4.6 Blood Glucose Control And The Management Of The Donor Pancreas

Little is known about the direct effects of brain death on the endocrine pancreas. Its effects on the exocrine pancreas seem to be more important as a trigger of the inflammatory system than the loss of function of the endocrine pancreas (Obermaier et al. 2004). Hyperglycaemia is often seen in brain-dead donors and is believed to be secondary to the effects of central nervous system trauma, massive levels of catecholamines, infusions of glucose-containing fluids, peripheral insulin resistance and the effects of steroids (Gores et al. 1990; Gores et al. 1992; Masson et al. 1993; Brunicardi et al. 2000). The insulin resistance seen is thought to be due to two main mechanisms: impaired insulin receptor binding (decreased insulin sensitivity) and alterations in intracellular metabolism (decreased insulin responsiveness) (Masson et al. 1993). Another potential contributor to the hyperglycaemia may be pancreatic endocrine insufficiency, which has been demonstrated in a brain-dead baboon experimental model (Novitzky et al. 1984). However this has not been seen in a number of human studies (Powner et al. 1990; Masson et al. 1993; Brunicardi et al. 2000), which have demonstrated elevated insulin levels. Reports of histologic immunohistochemical examination of the donor pancreas have also failed to identify any abnormalities (Steiner et al. 1988; Masson et al. 1993).

Hyperglycaemia is a potentially important factor in transplantation as there is evidence to suggest that donor hyperglycaemia is a significant risk factor for pancreatic allograft loss (Hesse et al. 1989; Gores et al. 1990; Gores et al. 1992) and it is also recognised that tight glycaemic control in critically ill patients has a survival benefit (Kutsogiannis et al. 2006). Tight control of blood glucose has also been shown to improve renal

- 203 - function in the organ donor (Blasi-Ibanez et al. 2009). Despite its prominence in donor management and its potential role in transplantation outcomes, there have been no randomised trials reported that evaluate glycaemic control in organ donors (Kutsogiannis et al. 2006).

Despite the reports of hyperglycaemia in the brain-dead donor, this study failed to demonstrate any significant hyperglycaemia (Figure 4.14). In fact, the highest recorded blood glucose level was 18.1 mmol/L and was seen in the NA group. Levels were above 10 mmol/L in only three animals, all of whom were in the NA group. The range of glucose levels seen in the FL group was 1.6 to 7.2 mmol/L, in the NA group 1.5 to 18.1 mmol/L and in the HR group 2.6 to 8 mmol/L. In addition, none of the HR animals had blood glucose levels necessitating the use of insulin and the use of steroids in this group was not associated with hyperglycaemia. There was a significantly higher blood glucose level in the NA group compared with the FL group at four, five and six hours post-brain death (p<0.02), and it was also higher than the HR group at the five and six hour time- points (p<0.002). Catecholamines such as adrenaline have been demonstrated to impair tissue sensitivity to insulin, resulting in a decrease in glucose clearance (Masson et al. 1993). Therefore the higher glucose levels seen in this study could be due to the effects of noradrenaline. Bearing this in mind however, the noradrenaline effects did not elevate blood glucose levels to excessively high levels, with a mean peak of 10.4±4.2 mmol/L

As with the liver, concerns have been raised that the use of vasopressin may impair pancreatic perfusion, function and preservation due to its powerful non-selective vasoconstrictor effects (Papp et al. 1983; Beijer et al. 1984). As previously mentioned, vasopressin can also potentiate the vasoconstrictor effects of agents such as noradrenaline (Holmes et al. 2003). Brain death has also been shown to cause significant pathophysiological alterations to the pancreas in an experimental rat model, manifested as deterioration of pancreatic microcirculation, elevated inflammatory tissue response and histological pancreas damage (Obermaier et al. 2004).

Pancreatic injury was assessed in the present study by examining the pancreatic enzyme levels of plasma amylase and lipase. In a rodent model, Obermaier et al demonstrated that brain death was associated with a significantly higher amylase level compared with

- 204 - controls, although these were not at pathological levels (Obermaier et al. 2004). They also demonstrated that there were no differences in lipase levels between brain-dead animals and controls. In the study reported here, there were no significant differences seen between any of the experimental groups at any time-point with respect to amylase and lipase levels. There were declining trends in the amylase levels in all experimental groups over time (Figure 4.15a), suggesting that there was no significant pancreatic injury to cause a release of amylase into the blood stream. In terms of the lipase results, there was a trend towards increasing levels over time across all experimental groups (Figure 4.15b). However lipase levels were within normal range in all animals at all time-points, with the highest value seen in the HR group at six hours at 9±4 U/L. Importantly, these results suggest that hormone resuscitation does not cause any significant injury to the donor pancreas (as reflected by significantly elevated pancreatic enzyme levels) compared with the more traditional treatments of the brain-dead organ donor such as noradrenaline and intravenous fluids.

4.4.7 Superior Renal Function With Hormone Resuscitation

It is well recognised that kidneys transplanted from living donors, whether they be related or unrelated, have consistently better outcomes compared with those from brain- dead donors (Nagareda et al. 1993; Koo et al. 1999; van der Hoeven et al. 1999a; Pratschke et al. 2000; Pratschke et al. 2001a; Pratschke et al. 2001b; Blasi-Ibanez et al. 2009). Brain death is associated with immunological and non-immunological damage to the kidney, which results in delayed allograft function (Kutsogiannis et al. 2006). This in turn leads to reduced recipient survival, increased rejection rates and increased rates of renal allograft nephropathy (Nagareda et al. 1993; van der Hoeven et al. 1999a; Pratschke et al. 2000; Kutsogiannis et al. 2006). It has also been shown that if donor systolic blood pressure is below 80-90 mmHg, the incidence of allograft failure increases, due to the loss of renal blood flow autoregulation and declines in glomerular filtration leading to acute tubular necrosis (Szostek et al. 1997; Kutsogiannis et al. 2006). Therefore the use of hormone resuscitation to improve haemodynamics and reduce inotrope use might reasonably be expected to improve renal function.

After the publication of the UNOS Critical Pathway for the Organ Donor (Rosengard et al. 2002; Zaroff et al. 2002), a retrospective analysis of brain-dead donors in the United

- 205 - States demonstrated a 22.5% increased usage of organs from donors who had received hormonal resuscitation compared with non-hormonal resuscitation donors (Rosendale et al. 2003b). This study also found that hormone resuscitation resulted in a 7.3% increased probability of the kidney being transplanted from a donor. An increase of 4.8% in kidney utilisation was also seen in a pilot study the previous year with no significant difference in either one year graft survival or in the rate of delayed renal allograft function (Rosendale et al. 2002). In a retrospective analysis two years after this pilot study, Rosendale and colleagues again demonstrated increased yield of kidneys used for transplantation in hormone-treated donors compared with non-hormone-treated donors, and also demonstrated that hormone resuscitation was associated with improved renal allograft survival (Rosendale et al. 2004).

As with the liver and pancreas, there have been concerns over the use of vasopressin because of its non-selective vasoconstrictor effects and the potential to reduce renal blood flow, thereby impairing renal function and preservation (Pennefather et al. 1995). However at low doses, there is experimental evidence in animals to suggest that the vasoconstriction is selective, affecting mainly the skeletal and cutaneous circulations, with the renal circulation relatively unaffected and hence has no detrimental effect on renal preservation (Liard et al. 1982; Goldsmith 1987). Preservation of renal function with vasopressin has also been shown in humans (Yoshioka et al. 1986; Kinoshita et al. 1990). The use of steroids has been demonstrated to improve survival in a rodent model of renal transplantation, whereby the intensity of ischaemia-reperfusion injury and of acute rejection was reduced (Pratschke et al. 2001a; Pratschke et al. 2002). These studies also showed that the expression of pro-inflammatory mediators and cellular infiltrates were diminished, presumably by the administration of steroids, thereby contributing to the improved survival. In a brain-dead porcine model of the donor, Pienaar et al showed that hypotensive animals who were treated with dopamine had reduced survival and progressive increases in creatinine (Pienaar et al. 1990). However, when these dopamine-treated animals were also treated with T3, post-transplant renal function was well preserved. Similarly in a porcine model of the brain-dead donor, Wicomb and colleagues studied the effects of brain death on renal function, which was measured in vitro in renal slices by the K+/Na+ ratio (Wicomb et al. 1986b). They found that a combination of T3, cortisol and insulin reversed the deterioration in renal slice function seen after brain death. A possible mechanism for the improvement in renal

- 206 - function is that T3 increases the levels of enzymes involved in cellular respiration and hence improve energy production (Pienaar et al. 1990).

Blood flow through the left renal artery in this study was found to be superior in hormone-treated animals compared with the noradrenaline-treated animals from four hours post-brain death induction to six hours (Figure 4.16a). In addition, at six hours, renal blood flow in the HR group was also higher than the FL group. Renal arterial flow was maintained throughout the period of brain death management in the HR group, whilst in the FL and NA groups renal flow declined significantly over time by 41% in the FL group and 80% in the NA group (baseline to six hours). These results suggest that hormone resuscitation is far superior in maintaining renal arterial flow comparable to pre-brain death levels compared with the more traditional treatments of noradrenaline and intravenous fluids. At doses of 0.4 μg/kg/min, noradrenaline has been shown to increase renal blood flow by 30%, increase urine output by 540% and increase creatinine clearance by 46% in a conscious merino ewe model (Di Giantomasso et al. 2002). There have also been a number of other experimental and clinical reports consistent with Di Giantomasso’s findings of improved renal flow and glomerular filtration rate with low to medium dose noradrenaline (Di Giantomasso et al. 2002). It should be noted though, that the study reported here differs from Di Giantomasso’s study in that it utilised a brain-dead model rather than a conscious model. In addition the peak doses of noradrenaline were significantly higher, up to 3.5 μg/kg/min. There is also literature to suggest that at high doses, renal flow can be compromised by noradrenaline (Di Giantomasso et al. 2002), as was seen in this study. In terms of the maintenance of renal arterial flow in the HR group, this was not surprising given the haemodynamic results demonstrating superior blood pressure in the HR group compared with the FL and NA groups (Figure 4.6b) and the superior cardiac output in the HR group at six hours compared with the NA group (Figure 4.7c). It seems intuitive that improved donor haemodynamics will contribute to improved regional blood flow and end organ perfusion.

Creatinine clearance was higher in the HR group at three and six hours after brain death compared with the NA group (Figure 4.16b). At the six hour time-point, the creatinine clearance in the HR group correlated with the superior renal arterial flow, blood pressure and cardiac output compared with the NA group. At the same time there were

- 207 - no significant differences between groups with respect to plasma creatinine and urea at any time-point. Despite the concerns over the effects of vasopressin on renal circulation, the results in this study have demonstrated that hormone resuscitation does not have any detrimental effect on the kidneys compared with treatments consisting of noradrenaline and/or intravenous fluid. In fact, hormone resuscitation was associated with superior renal blood flow and creatinine clearance compared with noradrenaline-treated animals.

4.4.8 The Decline In pH In The Organ Donor

The maintenance of acid-base balance in the blood and throughout body tissues is an important objective in the management of the brain-dead organ donor. Acid-base disturbances can have serious negative effects on cellular function and potentially render organs unsuitable for transplantation (Powner et al. 2000c). Arterial blood pH reflects the acid-base status of tissues, respiratory function via the elimination of carbon dioxide from the lungs, and the interaction of serum electrolytes and proteins. Alkalaemia or acidaemia can lead to electrolyte imbalances, coronary artery constriction, arrhythmias, decreased cardiac contractility, pulmonary artery constriction, heart failure, reduced hepatic and renal blood flow, insulin resistance, increased blood volume, catecholamine resistance and decreased cellular energy metabolism (Powner et al. 2000c). As a result, a potential donor can become haemodynamically unstable and rendered unsuitable as a donor, the organs can be damaged to the point of becoming unusable for transplantation, or at the very least may decrease the quality of the donor organ, thereby affecting transplant recipient outcomes.

There were significant declines in the pH in all of the experimental groups in this study (Figure 4.18a) from normal values (7.36-7.44) pre-brain death to acidotic values (7.31±0.08 in HR; 7.27±0.03 in FL and 7.08±0.10 in the NA group) at six hours post- brain death. Of note however, the pH was better preserved in the HR and FL groups compared with the NA group from three hours to six hours after brain death. In terms of base excess, NA animals had significantly worse base excesses compared with the HR animals from three to six hours, and by five hours it was also significantly worse than the FL group (Figure 4.18b). The exact cause of the metabolic acidosis seen in NA animals is not clear, particularly as lactate levels were not measured. The acidosis may be due to lactic acidosis (contributed from anaerobic metabolism) or acidosis from acute

- 208 - renal impairment. Despite this, these results demonstrate that of all the treatments examined, acid-base balance is best preserved with hormone resuscitation or intravenous fluids. Certainly, hormone treatment is not associated with worse acid-base balance compared with the more traditional treatments of noradrenaline or intravenous fluids.

4.4.9 Study Limitations

There are a number of study limitations to be considered when interpreting the results in this study. The first and main limitation is that donor animals were only managed for a relatively short period of time of six hours after induction of brain death. This is of particular importance given that in 2007, the median time between certification of brain death and aortic cross clamping in the donor leading to organ retrieval was 15.2 hours in Australia and 11.8 hours in New Zealand (Excell et al. 2008). Of note, only 2% of donors in Australia (and 5%in New Zealand) had aortic cross clamping within six hours of brain death certification (Excell et al. 2008). Therefore, whilst there were no significant differences identified between groups in some parameters, such as those affecting blood flow and function in the liver and pancreas, there may have been differences identified if the animals were observed for a longer period of time. Despite this, there were no results seen that would suggest that there would be a detrimental effect of hormone resuscitation on transplantable organs compared with the other two experimental groups had the donor animals been observed over a longer period of time.

Another limitation in this study is that the organs observed in the study were not used for transplantation. Ultimately the purpose of studies into the various treatments of the brain-dead organ donor is to determine how best to manage the donor to ensure optimal transplantation outcome, both short- and long-term. In an experimental model designed to test donor treatments, it would therefore be preferable to examine transplantation outcomes. However, this was beyond the scope and funding of the project reported here. There are however, future plans in the Transplant Program (and in collaboration with other research units) to test the various donor treatments in an animal model of transplantation, in both thoracic and abdominal organs.

- 209 - Other limitations include the lack of histological and immunohistochemical analysis of the various transplantable organs examined in this project. There was no assessment of cytokines or other inflammatory and protective mediators that have been identified as important in donor brain death and transplantation outcomes. Whilst these were not reported in this thesis, tissue and blood samples were collected and stored to examine these factors as part of ongoing studies.

Finally, there was no blinding in this study to the various treatment regimens tested and the pigs used in the study were juvenile (<12 months). However, it was believed that the study design and methodologies were rigorous enough that the lack of blinding was not likely to be critical and should not have biased the results. With respect to the use of juvenile pigs, this was done mainly due to the size of adult pigs and the logistical difficulties presented by the use of larger adult animals. Certainly the use of non-adult animals does somewhat limit the potential to extrapolate these results to an adult transplantation population, particularly in donors who are older with pre-existing co- morbidities.

4.5 CONCLUSION

There have been no other studies previously reported that have examined the effects of hormone resuscitation on both thoracic and abdominal organs in a prospective, randomised controlled study, as is presented here. In addition to examining the effects of hormone resuscitation, as advocated in the UNOS Critical Pathway for the Organ Donor (Rosengard et al. 2002; Zaroff et al. 2002), this study also compared hormone treatments with the more conventional management strategies involving noradrenaline and intravenous fluids. This has also not been previously examined and reported in the literature.

The main finding of this study was that hormone resuscitation of the brain-dead multi- organ donor was not associated with any detrimental effects on transplantable thoracic and abdominal organs compared with the traditional treatments of noradrenaline and intravenous fluids. Hormone resuscitation was found to produce better donor haemodynamics and cardiac function. It was also associated with improved preservation of renal blood flow and function, as measured by creatinine clearance. In addition,

- 210 - noradrenaline treatment was seen to be associated with worse pulmonary function (as measured by PaO2/FiO2 ratio and Aa gradient) and also with worse renal arterial flow and renal function (as reflected in creatinine clearance). With respect to the liver, this study was unable to detect any differences in blood flow, injury and function between any of the three treatment regimens tested. Similarly, there were no differences detected in the pancreas with respect to injury as reflected by amylase and lipase between any of the three treatment regimens.

Interestingly, this study demonstrated that the brain-dead donor could be maintained with intravenous fluids alone. Whilst MAP was below 50 mmHg from three to six hours post-brain death and reached as low as 38±12 mmHg at six hours, none of the FL animals arrested due to hypotension and all remained viable for the six hours of the study. With respect to blood pressure, SW and CO, animals in the FL group were no different from those in the NA group.

Within the limitations of the study previously discussed, this study supports the use of hormone resuscitation in donor management as advocated in the UNOS Critical Pathway for the Organ Donor and has shown that hormone resuscitation does not have any significant detrimental effects on transplantable solid organs with respect to the parameters examined. Despite these encouraging findings however, the ultimate test of these donor management regimens is to examine them in a transplantation model or in clinical transplantation. The main goal of donor management protocols is to optimise organ usage and to optimise transplantation outcomes, both short- and long-term. Therefore it is imperative that treatments such as hormone resuscitation be tested to examine their effects on the transplantability of donor organs and the effects on transplant outcomes. The results of the study presented here provide the basis for further experimental and clinical investigation into hormone treatments in donor management and their effects on transplantation outcomes.

- 211 -

CHAPTER 5

THE ROLE OF CARIPORIDE AND GLYCERYL TRINITRATE IN IMPROVING LONG-TERM PRESERVATION OF THE DONOR HEART FOR TRANSPLANTATION

- 212 - CHAPTER 5

THE ROLE OF CARIPORIDE AND GLYCERYL TRINITRATE IN IMPROVING LONG-TERM PRESERVATION OF THE DONOR HEART FOR TRANSPLANTATION

5.1 INTRODUCTION

The quality of the donor heart used for transplantation is a major determinant of both short- and long-term outcome. This is determined by a number of factors, including donor age and pre-existing disease, the mechanism of brain death, donor management prior to organ procurement, the duration and manner of hypothermic storage, and the circumstances of reperfusion. Whilst some of these factors cannot be modified clinically, there are others (such as donor management, organ storage and preservation, and reperfusion of the transplanted organ) that can be altered in order to optimise donor organ function.

As discussed in Section 1.4, the systemic effects of donor brain death play a significant role in causing donor organ dysfunction. The studies described in Chapters 3 and 4 demonstrated that some of these effects could be ameliorated by the use of hormone resuscitation during donor management prior to organ procurement without any significant detrimental effects.

Ischaemia-reperfusion injury is another major mediator of donor organ dysfunction, contributing to morbidity and mortality in heart transplant recipients (Fleischer et al. 2002; Huang et al. 2004; Hicks et al. 2006; Russo et al. 2007; Taylor et al. 2007). Current commercial organ preservation solutions incorporate a number of therapeutic approaches that are aimed at minimising ischaemia-reperfusion injury. However, the ability of these solutions to abrogate ischaemia-reperfusion injury is limited to the extent that donor heart ischaemic times in excess of four to six hours are associated with

- 213 - an exponential increase in the risk of primary graft failure and death after heart transplantation (Taylor et al. 2006; Taylor et al. 2007; Taylor et al. 2008).

The Na+-H+ exchanger (NHE) has been implicated in the pathogenesis of ischaemia- reperfusion injury in the myocardium and is a potential candidate that has not been ‘targeted’ by current organ preservation solutions. Activation of the NHE occurs during both ischaemia and subsequent reperfusion. Its activation causes cytosolic and mitochondrial calcium overload, leading to ventricular arrhythmias, contractile dysfunction and myocardial cell death (Allen et al. 2000; Avkiran 2003; Teshima et al. 2003). Selective inhibition of the NHE by cariporide (HOE642) and other related compounds have been shown to reduce the severity of ischaemia-reperfusion injury in a number of clinical (Rupprecht et al. 2000; Theroux et al. 2000) and animal model settings (Allen et al. 2000; Avkiran et al. 2002; Teshima et al. 2003; Ryan et al. 2003a). A more detailed discussion about the NHE, its role in ischaemia-reperfusion injury and its blockade with compounds such as cariporide can be found in Sections 1.5.2 and 1.5.4.

The anti-ischaemic effects of organic nitrates such as glyceryl trinitrate (GTN) are mediated via a number of pathways and their use clinically has been well established. The vasodilatory effect of GTN occurs via its conversion to nitric oxide (NO) or a NO- containing metabolite which in turn increases levels of guanosine-3’,5’-cyclic monophosphate (cGMP). This causes vascular smooth muscle relaxation, and improves coronary collateral flow and dilates stenotic coronary arteries, resulting in increased oxygen and energy supply (Csont et al. 2005). In addition, there is evidence that nitrates have direct myocardial anti-ischaemic effects that may be mediated by cGMP-

+ independent activation of ATP-sensitive potassium channels (K ATP) by NO (Ferdinandy et al. 1995; Sasaki et al. 2000; Csont et al. 2005). As a result, there have been a number of reports using GTN in small animal models as an effective adjunct to hypothermic cardioplegia and cardiac preservation (Pinsky et al. 1994b; Baxter et al. 1999a; Baxter et al. 2001; Gao et al. 2005).

This laboratory has previously reported that long-term myocardial preservation (10 hours) in an isolated working rat heart model is markedly enhanced when both cariporide and GTN are added to Celsior solution used for storage compared with either

- 214 - agent alone (Gao et al. 2005). Furthermore, the myocardial preservation achieved by combined supplementation of the storage solution was superior to that achieved by perfusion of the isolated heart with cariporide prior to cardioplegic arrest, storage (both short- and long-term) and subsequent reperfusion. The present study described in this chapter was undertaken to build on the results of this small animal study, by trialling these preservation strategies in a large animal model.

The aim of the study described in this chapter was to assess the potential of cariporide and GTN to ameliorate the harmful effects of ischaemia-reperfusion injury, and thus improve recovery of the transplanted heart after long-term hypothermic storage. The primary hypothesis of this study was that long-term (14 hours) preservation strategies incorporating cariporide and GTN in heart transplantation facilitate weaning from cardiopulmonary bypass support after transplantation.

The secondary hypotheses of this study were that these preservation strategies incorporating cariporide and GTN would improve contractile function of the heart after transplantation, improve the haemodynamic status in the recipient after transplantation, improve coronary arterial blood flow (left anterior descending coronary artery), and reduce troponin I release (as a marker of myocardial injury).

5.2 METHODS

5.2.1 Porcine Model Of Orthotopic Heart Transplantation

A porcine model of orthotopic heart transplantation was used in this study as described in Chapter 2. This model closely mimics human heart transplantation, incorporating donor brain death, donor management in an intensive care unit-style setting, organ retrieval, hypothermic ischaemic preservation, orthotopic transplantation, warm ischaemia during transplantation, reperfusion of the transplanted heart and weaning off cardiopulmonary bypass with the aid of dobutamine and ventricular demand pacing. Fifty-six pigs obtained in pairs (one donor and one recipient) were used in this study. Forty-six of these animals were highly inbred Westran pigs with the remaining 10 pigs being Landrace pigs. The species of pigs used for the studies were changed due to animal supply problems.

- 215 -

5.2.2 Preparation Of The Donor Animal And Subsequent Management

Animals were anaesthetised and surgically prepared as described in Chapter 2. Shortly after the completion of cardiac instrumentation, baseline data were acquired (as described in Sections 2.7.1 and 2.7.2). Once the heart returned to its steady state function after vena caval occlusion for data acquisition, brain death was induced by inflation of a Foley catheter balloon situated in the subdural space (as described in Sections 2.5.2 and 2.5.4).

Animals were monitored and managed for a total of six hours after brain death induction. When the mean arterial blood pressure (MAP) fell below 60 mmHg after brain death, boluses of IV saline were given to maintain the blood pressure. If this was unsuccessful, an IV noradrenaline infusion (20 μg/mL) was commenced and titrated to achieve MAP 60–70 mmHg. Three hours after brain death induction, animals were commenced on a hormone resuscitation protocol consisting of methylprednisolone, triiodothyronine, vasopressin and insulin (Table 2.1) (in addition to noradrenaline) for a further three hours. Methylprednisolone was administered as a single IV bolus of 15 mg/kg. Triiodothyronine was administered as an initial bolus dose of 4 μg, followed by a continuous infusion of 4 μg/hr. Vasopressin was commenced at 0.5 Units/hr and increased incrementally to a maximum of 4 Units/hr to maintain MAP 60-70 mmHg. If MAP was still below 60 mmHg despite being on maximal vasopressin, the noradrenaline infusion rate was increased to maintain MAP. Attempts were made to wean animals off noradrenaline whilst maintaining MAP. If successful, attempts were then made to wean animals off vasopressin, again whilst maintaining MAP 60–70 mmHg.

After data were recorded at six hours post-brain death induction, the donor animal was heparinised and prepared for cardiac allograft explantation, as described in Section 2.5.7. The donor heart was then arrested with Celsior solution, with or without supplementation depending on the study protocol. Once the cardioplegia was

- 216 - completely infused and the heart arrested, the heart was explanted, prepared and stored (as described in Section 2.5.7).

5.2.3 Experimental Groups

A total of twenty-eight orthotopic heart transplants were performed across five experimental groups. The author (AH), as operating surgeon, was not blinded to group allocation. Two transplants were excluded from analysis due to technical problems. In one experiment, the donor animal experienced a catecholamine storm just prior to donor cardiectomy and the arterial monitoring line was dislodged. At the same time this animal also had a period of approximately 20-30 minutes of hypoxia and hypercarbia of unknown cause. As a result, the donor heart was contracted and stone hard after ischaemic hypothermic preservation and remained so after transplantation and reperfusion. No other donor hearts in this study had this appearance. It was decided to exclude this transplant from the study because of the aforementioned events. The second transplant that was excluded was due to technical problems with cardiopulmonary bypass. After reperfusion of the transplanted heart, the venous reservoir was inadvertently run empty and as a result, air was pumped into the transplanted heart. This situation was irretrievable and it was decided to terminate the experiment and exclude it from the study.

Twenty-six transplants were included in the data analysis for this study. All donor hearts were exposed to a total ischaemic time (cold and warm) of approximately 14 hours. Animals in the first experimental group (CAR1; n=5) were pre-treated with a single intravenous dose of cariporide (2mg/kg) 15 minutes prior to donor cardiectomy and again at 15 minutes prior to reperfusion in the recipient. Celsior solution was used for cardioplegia and donor hearts were stored in Celsior alone. The dose of cariporide in this group was chosen based on clinical studies showing that 120 mg is both the maximum tolerated dose and the minimum therapeutic dose (Buerke et al. 1999; Theroux et al. 2000). The timing of cariporide administration was chosen based on experimental data demonstrating a role for NHE inhibition during periods of ischaemia and reperfusion post-ischaemia (Hurtado et al. 2001). Both the dose and the timing of cariporide administration have been used in previous studies in this laboratory (Ryan et al. 2003a; Ryan et al. 2003b).

- 217 -

The second experimental group (CAR2; n=5) utilised Celsior with cariporide added (10μmol/L), and the third experimental group (GTN; n=5) had GTN (Mayne Pharma, Australia) (100 mg/L) added to Celsior. The fourth experimental group (COMB; n=6) had both cariporide (10 μmol/L) and GTN (100 mg/L) added to Celsior. The supplemented Celsior solution in CAR2, GTN and COMB groups was used for both cardioplegia and storage. The final experimental group was a control group (CON; n=5), whereby animals were not exposed to either cariporide or GTN. Celsior alone was used as cardioplegia and for storage in the CON group. With the exception of the GTN group, all transplants (n=23) were performed by the author. Transplants in the GTN group were performed by Dr Alasdair Watson.

Gao and colleagues (2005) demonstrated that hearts stored for 10 hours in Celsior supplemented with GTN did not provide effective preservation and recovery of the rat heart in an isolated working rat heart model. As a result of these findings, the GTN group of experiments were not performed as part of the original study. It was postulated that if the rat heart did not recover after 10 hours of ischaemic preservation using GTN- supplemented Celsior, then the transplanted porcine heart would not recover after 14 hours of ischaemic preservation in GTN-supplemented Celsior. However, prior to acceptance for publication, an additional experimental group utilising GTN- supplemented Celsior was requested by the reviewing panel, which was subsequently undertaken (Hing et al. 2009).

5.2.4 Preparation Of Cariporide

The cariporide used in these experiments was a gift from the manufacturer, Aventis Pharma (Strasbourg, France), under an independent external investigator agreement. Each dose of cariporide was individually prepared on the day of experiment, according to the weight of the animals for experimental group CAR1 or 10 μmol/10 mL for experimental groups CAR2 and COMB. Cariporide powder was weighed and then mixed with 0.1-0.2 mL of dimethyl sulfoxide (DMSO) to form a slurry. This was then dissolved in 10 mL of normal (0.9%) saline by vortexing.

- 218 - 5.2.5 Preparation Of The Recipient Animal And Orthotopic Cardiac Allograft Transplantation

Recipient animals were anaesthetised and surgically prepared as described in Section 2.5.1. Animals were given methylprednisolone (500 mg IV at anaesthetic induction and again at reperfusion post-transplantation) and heparinised as described in Section 2.6.1. They were placed on total cardiopulmonary bypass support and managed as described in Sections 2.6.1 and 2.6.2. Whilst on cardiopulmonary bypass, animals were actively cooled to 32°C during transplantation.

The recipient’s native heart was explanted and the donor heart was transplanted orthotopically utilising the bi-atrial technique described by Lower and Shumway (Lower et al. 1960). A detailed description of this procedure is given in Section 2.6.3. Celsior (200 mL) was infused into the coronary system via the aortic root after completion of the left atrial anastomosis, and again after the right atrial anastomosis. Re-warming was commenced during the aortic anastomosis. Once the transplant was completed and 14 hours ischaemic time had passed, the aortic cross-clamp was removed and the heart reperfused. The heart was de-aired with a left ventricular vent and once warm, was defibrillated and paced (ventricular demand) at 120 beats per minute. Dobutamine was commenced at 10 μg/kg/min 45 minutes after reperfusion. This dose was chosen based on previous porcine transplant studies (Ryan et al. 2003a).

5.2.6 Management Of The Transplant Recipient And Weaning From Cardiopulmonary Bypass Support

Management of the recipient animal post-transplant, the protocol for weaning off bypass and the criteria for successful weaning from bypass are described in detail in Section 2.6.4. In summary, after one hour of reperfusion, attempts were made to wean the animal from cardiopulmonary bypass. If unsuccessful, the animal was placed back onto full bypass support and another attempt at weaning from bypass was made at two hours post-reperfusion. If still unsuccessful, dobutamine was increased to 20 μg/kg/min and a third attempt at weaning was made at three hours. If still unsuccessful, the study was terminated. During all attempts at weaning, central venous pressure was monitored continuously to ensure adequate cardiac filling, in addition to visual inspection of the

- 219 - filling state of the transplanted heart. Animals that were successfully weaned from bypass were monitored for a further three hours with data acquired hourly (as described below and in Section 2.7). Blood loss during the study was autotransfused back into the animal via the bypass circuit and additional Hartmann’s solution was added to the bypass circuit to optimise haemodynamic performance.

5.2.7 Study Outcomes

The primary outcome measures for the study were successful weaning from cardiopulmonary bypass, left ventricular contractile function determined from the preload recruitable stroke work (PRSW) relationship derived from pressure-volume (PV) loop analysis (Ryan et al. 2002) and haemodynamic indices (i.e. stroke work, blood pressure and cardiac output). Left anterior descending coronary artery (LAD) flow and troponin I were also assessed. Troponin I was assayed as previously described (Section 2.7.5).

5.2.8 Data Acquisition And Analysis

Haemodynamic and left ventricular PV loop data were recorded prior to brain death induction (baseline), and then at hourly intervals for six hours prior to cardioplegia and explantation of the donor heart (as described in Sections 2.7.1 and 2.7.2). Data were also recorded one hour after weaning from cardiopulmonary bypass post-transplantation and hourly thereafter for an additional two hours.

Left ventricular PV loop data were acquired and analysed as described in Sections 2.7.2 and 2.7.3 utilising SonoSOFT Software (Sonometrics Corp.). Stroke work (SW) and end-diastolic volume were determined from these data. SW was determined by the area of the PV loop. The relationship between SW and end-diastolic volume, termed the PRSW relationship was then determined using linear regression analysis of data acquired during vena caval occlusion as described in Section 2.7.4 (Glower et al. 1985).

5.2.9 Power Calculation And Statistical Analyses

The primary endpoint of the study was successful weaning from cardiopulmonary bypass after orthotopic heart transplantation. It was hypothesised that >90% hearts that

- 220 - received cariporide as part of the preservation protocol (CAR1, CAR2 and COMB groups) would be weanable from bypass whereas < 10% of hearts in the control or GTN groups would be weanable. Using estimates of 90% versus 10% success in weaning for the different treatment groups, a study with 5 animals in each group had 80% power (1 – beta) to detect this difference at a significance level (alpha) of 0.01. The higher level of significance level was set to allow for multiple comparisons.

The power calculation was based on the formula:

2 n = f(, ) [p1 (100 p1) + p2 (100 p2)] / (p2 p1)

where p1 and p2 are the percent 'success' in the control and experimental group respectively and f(, ) = [-1(/2) + -1()]2. -1 is the cumulative distribution function of a standardised normal deviate (Pocock 1983; Julious 2004).

Statistical analyses were performed using SPSS for Windows 12.0.1 (SPSS Inc., Chicago IL USA) and StatView for Windows 4.57 (Abacus Concepts Inc, USA). Continuous variables were reported as mean±standard deviation and categorical variables as actual incidence/number of hearts in the study group. The characteristics of the study groups were compared using analysis of variance (ANOVA) and the proportion of hearts in each group that were successfully weaned from bypass was compared using Fisher’s Exact test for a 2 x 5 contingency table. MAP, cardiac output (CO), troponin I and LAD flow were compared with ANOVA or Student’s t test for independent samples between groups where applicable. Significant differences in ANOVA were investigated using Student’s t test for paired samples with a Bonferroni correction for multiple comparisons. Differences were considered statistically significant at p<0.05.

PRSW relationships were compared with a multiple linear regression (MLR) implementation of two-way analysis of covariance with repeated measures as described in Section 2.7.4 (Glantz et al. 2001). Overall changes in left ventricular contractility, as reflected in the PRSW relationship, are determined by changes in slope (nMw, calculated with normalised data as described in Sections 2.7.3 and 2.7.4) and nVw (the normalised volume axis intercept of the PRSW, also described in Sections 2.7.3 and

- 221 - 2.7.4). Ischaemia causes a decrease in slope and an increase in nVw. The Stroke Work Index (SWI, described further in Section 2.7.4) was determined, and represents the net effect of the interaction between changes in slope and nVw on normalised stroke work (nSW). Standard MLR models were used to determine whether changes in left ventricular contractility between pre- and post-transplant were significant.

5.3 RESULTS

5.3.1 Experimental Animals

Characteristics of the five experimental groups are shown in Table 5.1. There were no significant differences in body weights, left ventricular volumes, left ventricular volume to body weight ratios, brain death management times (i.e. time from induction of brain death to donor heart explantation), warm ischaemic and total ischaemic times. The warm ischaemic time was the total time taken to transplant the donor heart and was calculated as the time from application of the aortic cross clamp to the recipient just prior to recipient cardiectomy to the removal of the cross clamp at the conclusion of the final anastomosis (i.e. aortic anastomosis) of the transplant. The total ischaemic time was calculated as the time from application of the aortic cross clamp to the donor prior to donor cardiectomy to the removal of the cross clamp from the recipient to reperfuse the transplanted heart. The mean time between commencement of the Celsior infusion and cardiac asystole at the time of donor heart procurement was 1.1±0.4 minutes with no significant differences between treatment groups.

- 222 - Table 5.1: Characteristics of Experimental Groups

Characteristic CON CAR1 CAR2 GTN COMB p value*

Weights (kg) Donor 42.9±8.0 43.1±7.6 52.4±5.1 51.4±8.7 50.3±6.5 0.119

Recipient 40.5±6.9 42.1±4.4 48.2±2.3 47.6±7.2 47.0±6.7 0.161

Donor LV Volume 106±13.4 116±16.7 125±18.7 122±21.0 120±13.0 0.423 (mL) LVV/BW Ratio 2.5±0.2 2.7±0.2 2.4±0.4 2.4±0.2 2.4±0.4 0.550

Times (min) Brain Death to 380.0±5.5 388.6±18.8 378.6±2.8 374.6±3.9 383.2±6.3 0.210 Explant Warm Ischaemic 52.4±9.6 48.4±9.0 72.6±21.8 57.8±10.9 75.2±26.5 0.077 Time Total Ischaemic 845.6±11.9 835.8±12.9 856.0±25.9 861.4±13.8 851.3±24.7 0.250 Time

*ANOVA. LVV=Left ventricular volume; BW=body weight. Values are expressed as mean±standard deviation. CON = control group (n=5); CAR1 = cariporide pre-treatment group (n=5); CAR2 = cariporide + Celsior group (n=5); GTN = glyceryl trinitrate + Celsior group (n=5); COMB = cariporide + glyceryl trinitrate + Celsior group (n=6).

5.3.2 Weaning From Cardiopulmonary Bypass

All five CAR1 and five of six COMB animals were successfully weaned from cardiopulmonary bypass compared with only one of five CAR2, none of five GTN and one of five CON animals (p=0.001). All transplanted hearts weaned from bypass were on 10 μg/kg/min dobutamine. CAR1 animals were weaned from bypass between 83– 156 min after commencing reperfusion (122±28 min), COMB animals between 84–182 min (120±41 min) (p>0.05), and the sole CAR2 and CON animals at 87 and 103 min post-reperfusion.

- 223 - 5.3.3 Cardiac Contractility

Representative left ventricular PV loops from the cariporide pre-treatment group (CAR1) are shown in Figure 5.1 and corresponding PRSW relationships are shown in Figure 5.2. Results of the MLR analysis of the PRSW relationships are shown in Table 5.2. There were no significant differences in baseline left ventricular contractility between the groups. As only one heart was weaned from cardiopulmonary bypass in each of the CON and CAR2 groups, comparisons of the PRSW relationship between donor and recipient in these groups were not performed.

Both slope and nVw were increased post transplantation compared with baseline in CAR1 and COMB animals. SWI increased by 25% from baseline to three hours post-weaning in COMB animals compared with a reduction of 11% in CAR1 animals. Similar trends were seen at one and two hours post-weaning with COMB animals showing increased SWI (+17% and +24%) and CAR1 animals showing reduced SWI (-15% and -9%) respectively. MLR analysis confirmed superior contractility in COMB animals compared with CAR1 animals at one, two and three hours post-weaning (p<0.0001). Additionally, SWI was relatively stable for up to three hours post-weaning in both groups.

- 224 - 90 Pre-Brain Death 70

50

30

10 LV Pressure (mmHg) Pressure LV

-10 0.4 0.5 0.6 0.7 0.8 0.9 1.0 90 Pre-Explantation 70

50

30

LV Pressure (mmHg) Pressure LV 10

-10 0.4 0.5 0.6 0.7 0.8 0.9 1.0 110 Post-Transplantation 90

70

50

30

LV Pressure (mmHg) Pressure LV 10

-10 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Normalised Epicardial LV Volume

Figure 5.1: Left Ventricular (LV) Pressure-Volume Loops. Representative loops from an animal from the cariporide pre-treatment group (CAR1) are shown. These loops were obtained from the donor heart prior to brain death, prior to explantation (approximately six hours after brain death induction) and after transplantation into the recipient animal (approximately two hours after weaning from cardiopulmonary bypass support). Data for these loops were acquired during transient occlusion of the inferior vena cava. Volumes were normalised to the baseline (pre-brain death) steady state end-diastolic volume.

- 225 - 1.6 1.4 1.2 1.0 0.8 0.6 0.4

Normalised Stroke Work Normalised Stroke 0.2 0 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 Normalised Epicardial End Diastolic Volume

Figure 5.2: Preload Recruitable Stroke Work (PRSW) Relationship. Derived from the left ventricular pressure-volume loops shown in Figure 5.1. The PRSW relationships at pre-brain death (open circles), pre-explantation (open diamonds) and post-transplant (two hours post-weaning from cardiopulmonary bypass) (closed triangles) are shown. Data obtained during transient occlusion of the inferior vena cava. Stroke work and volume were normalised to the baseline (pre- brain death) steady state stroke work and end-diastolic volume.

- 226 - Table 5.2: Preload Recruitable Stroke Work (PRSW) Relationship.

Time Point Slope (nMw) y-Axis x-Axis Stroke Intercept Intercept Work (nSW) (nVw) Index

Donor

Baseline (CON) 2.481±0.040 -1.478±0.035 0.596 1.003

Baseline (CAR1) 2.740±0.050 -1.703±0.043 0.622 1.037

Baseline (CAR2) 2.747±0.065 -1.704±0.054 0.620 1.043

Baseline (GTN) 2.425±0.052 -1.432±0.044 0.591 0.993

Baseline (COMB) 2.863±0.049 -1.812±0.042 0.633 1.051

Recipient (Post Transplant)

1 h Post Wean From 5.165±0.157 -4.283±0.153 0.829 0.882 Bypass (CAR1) 1 h Post Wean From 5.408±0.231 -4.184±0.209 0.774 1.224 Bypass (COMB)

2 h Post Wean From 4.706±0.133 -3.765±0.129 0.800 0.941 Bypass (CAR1) 2 h Post Wean From 5.161±0.097 -3.855±0.085 0.747 1.306 Bypass (COMB)

3 h Post Wean From 4.808±0.142 -3.887±0.138 0.808 0.921 Bypass (CAR1) 3 h Post Wean From 4.785±0.132 -3.472±0.113 0.726 1.313 Bypass (COMB) y = normalised stroke work; x = normalised epicardial end-diastolic volume. Regression coefficients (nMw and y-axis intercept) are the group mean±standard error of the normalised PRSW. nVw and stroke work index are calculated from the mean value of the regression coefficients (i.e. nMw and y-axis intercept). CON = control group (n=5); CAR1 = cariporide pre-treatment group (n=5); CAR2 = cariporide + Celsior group (n=5); GTN = glyceryl trinitrate + Celsior group (n=5); COMB = cariporide + glyceryl trinitrate + Celsior group (n=6). Baseline = Pre-brain death. See Section 5.3.3 for further details and analysis.

- 227 - 5.3.4 Haemodynamic Changes In The Donor And The Recipient Post- Transplantation

Serial changes in left ventricular stroke work (LVSW) over the course of the study in each experimental group are shown in Figure 5.3. There were no significant differences in LVSW between the groups at baseline. LVSW increased significantly in all groups at three hours after brain death (p<0.001, vs. baseline), with no significant differences between groups. LVSW did not change significantly between three and six hours after brain death induction. As only one CON and one CAR2 animal were weaned off cardiopulmonary bypass after transplantation, LVSW was not assessed in these groups after transplantation. In the CAR1 and COMB groups, the post-weaning LVSW was lower compared with six hours post brain death. When compared with the baseline LVSW, post-weaning LVSW was not significantly different. The LVSW remained stable over the three hours post-weaning and there were no significant differences between the CAR1 and COMB groups at any time point.

MAP data are shown in Figure 5.4. There were no differences in MAP between groups during pre-explantation and post-transplantation phases of the study. MAP was lower post-transplantation compared with baseline in CAR1 animals (50±3, 50±4, 48±4 and 50±5 mmHg at zero, one, two and three hours post-weaning vs. 62±5 mmHg; p<0.020), whereas there were no differences in COMB animals at similar time-points (56±10, 54±8, 56±8 and 55±6 mmHg vs. 62±4 mmHg; p>0.05). MAP did not change significantly during the three hours after weaning off bypass in either group.

- 228 - 7000 CON CAR1 6000 CAR2 GTN 5000 COMB

4000

3000

Stroke Work (mmHg/mL) Work Stroke 2000 - 229 1000

0 Baseline 3 h Post-Brain Death 6 h Post-Brain Death 1 h Post-Wean 2 h Post-Wean 3 h Post-Wean

Pre-transplant Post-transplant Time Point Figure 5.3: Left Ventricular Stroke Work (mean±standard deviation): CON = control group (n=5); CAR1 = cariporide pre-treatment (n=5); CAR2 = cariporide + Celsior (n=5); GTN = glyceryl trinitrate + Celsior group (n=5); COMB = cariporide + glyceryl trinitrate + Celsior (n=6). Baseline data taken pre-brain death and post-weaning data taken after successful weaning from cardiopulmonary bypass post-heart transplant. As only one CON, one CAR2 and no GTN transplant/s were weaned from bypass, there are no CON, CAR2 and GTN results post-transplant. There were no significant differences in stroke work between groups post-transplantation. See text for further details of statistical analyses.  120 CON CAR1

CAR2 GTN 100 COMB

80

60 * * *

40 Mean Arterial Pressure (mmHg) Arterial Pressure Mean - 230 20

0 Baseline 3 h Post-Brain Death 6 h Post-Brain Death 1 h Post-Wean 2 h Post-Wean 3 h Post-Wean

Pre-transplant Post-transplant Time Point

Figure 5.4: Mean Arterial Blood Pressure (mean±standard deviation): CON = control group (n=5); CAR1 = cariporide pre-treatment (n=5); CAR2 = cariporide + Celsior (n=5); GTN = glyceryl trinitrate + Celsior group (n=5); COMB = cariporide + glyceryl trinitrate + Celsior (n=6). Baseline data taken pre-brain death and post-weaning data taken after successful weaning from cardiopulmonary bypass post-heart transplant. As only one CON, one CAR2 and no GTN transplant/s were weaned from bypass, there are no CON, CAR2 and GTN results post-transplant. *p<0.020 (compared with baseline). See text for further details of statistical analyses. CO data are shown in Figure 5.5. There were no significant differences in CO between the groups during pre-explantation and post-transplantation periods. There was no deterioration in CO in CAR1 animals post-transplantation compared with baseline, despite 14 hours of ischaemic storage (4174±1198, 4314±626 and 4498±743 mL/min at one, two and three hours post-weaning vs. 3219±997 mL/min at baseline). Similarly in COMB animals, CO was not significantly different post-transplantation compared with baseline (4831±1192, 4485±1081, 4593±1023 mL/min at one, two and three hours post- weaning vs. 4281±935 mL/min at baseline). Whilst there were no statistically significant differences in CO post-transplantation compared with baseline, there was a trend towards a higher CO post-transplantation. CO did not change significantly in animals weaned off bypass for up to three hours in either CAR1 or COMB.

5.3.5 Left Anterior Descending Coronary Artery (LAD) Flow

LAD flow data are shown in Figure 5.6. Post hoc testing demonstrated a significant difference in LAD flow between GTN and COMB groups at baseline (29±8 mL/min vs. 13±6 mL/min; p<0.05), but not between the other groups. Over the six hours of donor management, there was a trend towards higher LAD flows in the GTN group over time. This was significant at six hours post-brain death induction but not at three hours. There were no differences in LAD flow between groups pre-explantation, nor were there any differences between CAR1 and COMB post-weaning. Measurements of LAD flow post- weaning from cardiopulmonary bypass after heart transplantation were only obtainable in the CAR1 and COMB groups. Flow in CAR1 was initially significantly higher at one hour post-weaning (39±5 mL/min) compared with baseline (23±11 mL/min; p<0.03) but by two and three hours post-weaning (32±7 and 36±3 mL/min), flows were not significantly different to baseline. In contrast, when compared with six hours post-brain death (12±3 mL/min), LAD flows were higher at one, two and three hours post-weaning (39±5, 32±7 and 36±3 mL/min; p<0.004). Flows in COMB were higher at one, two and three hours post-weaning (34±8, 38±10 and 35±10 mL/min) compared with baseline (13±6 mL/min; p<0.01). When compared with six hours post-brain death (16±9 mL/min), flows were significantly higher at one and two hours post-weaning (34±8 and 38±10 mL/min; p<0.04) but not at three hours post-weaning (35±10 mL/min). Flows did not change significantly in either CAR1 or COMB post-weaning.

- 231 - 9 CON CAR1 8 CAR2 GTN 7 COMB 6

5

4

3 Cardiac Output (L/min) - 232 2

1

0 Baseline 3 h Post-Brain Death 6 h Post-Brain Death 1 h Post-Wean 2 h Post-Wean 3 h Post-Wean

Pre-transplant Post-transplant Time Point Figure 5.5: Cardiac Output (mean±standard deviation): CON = control group (n=5); CAR1 = cariporide pre-treatment (n=5); CAR2 = cariporide + Celsior (n=5); GTN = glyceryl trinitrate + Celsior group (n=5); COMB = cariporide + glyceryl trinitrate + Celsior (n=6). Baseline data taken pre-brain death and post-weaning data taken after successful weaning from cardiopulmonary bypass post-heart transplant. As only one CON, one CAR2 and no GTN transplant/s were weaned from bypass, there are no CON, CAR2 and GTN results post-transplant. See text for details of statistical analyses. 60 CON CAR1 CAR2

50 GTN COMB

40

30

LAD flow (mL/min) 20 - 233

10

0 Baseline 3 h Post-Brain Death 3 h Post-Brain Death 1 h Post-Wean 2 h Post-Wean 3 h Post-Wean

Pre-transplant Post-transplant Time Point Figure 5.6: Left Anterior Descending Coronary Artery (LAD) Flow (mean±standard deviation): CON = control group (n=5); CAR1 = cariporide pre-treatment (n=5); CAR2 = cariporide + Celsior (n=5); GTN = glyceryl trinitrate + Celsior group (n=5); COMB = cariporide + glyceryl trinitrate + Celsior (n=6). Baseline data taken pre-brain death and post-weaning data taken after successful weaning from cardiopulmonary bypass post-heart transplant. As only one CON, one CAR2 and no GTN transplant/s were weaned from bypass, there are no CON, CAR2 and GTN results post-transplant. See text for details of statistical analyses. 5.3.6 Troponin I

Troponin I data are shown in Table 5.3. There were no significant differences between groups in troponin I in the donor animals, nor at 15 min post-reperfusion. After reperfusion of the transplanted heart, the highest troponin I values were seen in CON animals, with the lowest values being seen in the COMB group. Troponin I was significantly higher in CON compared with COMB animals at one hour post- reperfusion (190.0±46.1 vs. 80.7±43.0 μg/L; p<0.02) and at three hours post-reperfusion (895.6±533.8 vs. 234.9±102.2; p<0.04).

Table 5.3: Troponin I (μg/L; median and range).

Time CON CAR1 CAR2 GTN COMB

Baseline 0 0 1.0 0.4 0.9 (0-1.2) (0-0.8) (0-1.5) (0.2-0.9) (0.6-1.5)

3 h Post- 3.6 3.8 2.5 N/A 3.5 Brain Death (2.4-5.2) (2.2-8.4) (1.0-3.1) (1.4-5.2)

6 h Post- 8.0 7.0 3.2 3.0 6.4 Brain Death (1.0-68.0) (5.6-51.1) (1.2-4.0) (1.5-8.4) (2.0-7.7)

15 min Post- 42.0 34.0 23.4 15.6 22.3 Reperfusion (25.0-70.0) (25.0-41.0) (15.0-35.6) (8.1-26.6) (11.2-48.2)

1 h Post- 196.0 181.0 101.3 100.4 71.6 Reperfusion (118.0-236.0) (100.0-244.0) (60.0-200.0) (60.8-260.9) (41.0-126.4)

3 h Post- 554.0 539.5 236.6 424.5 244.7 Reperfusion (460.0-1500.0) (219.0-870.0) (84.4-813.0) (239.0-604.4) (118.6-352.0)

CON (Control) (n=5); CAR1 (cariporide pre-treatment) (n=5); CAR2 (cariporide and Celsior) (n=5); GTN (glyceryl trinitrate + Celsior) (n=5); COMB (glyceryl trinitrate + cariporide + Celsior) (n=6) Groups. Baseline=Pre-Brain Death; N/A=Data not available.

- 234 - 5.4 DISCUSSION

5.4.1 Study Rationale And Experimental Groups

To date, there have been no reports in the literature of viable recovery of a transplanted heart after 14 hours ischaemic storage in a large animal model that incorporates brain death and prolonged management (six hours) of the brain-dead donor. As mentioned earlier, Gao et al demonstrated a potential role for cariporide and GTN in long-term myocardial preservation (Gao et al. 2005). They reported that supplementation of Celsior solution with both GTN and cariporide resulted in viable recovery of cardiac function after 10 hours hypothermic storage in an isolated working rat heart model. They also demonstrated that supplementation of Celsior with either GTN or cariporide alone did not afford adequate preservation to hearts stored for 10 hours. As a result, the study reported here was designed to investigate similar preservation strategies in a large animal model and to determine whether the strategies tested in an isolated small animal heart model could be extrapolated to a large animal model that incorporated many of the elements of clinical heart transplantation.

The duration of donor heart storage was considerably longer than that seen in clinical heart transplantation. This storage time was chosen based on previous findings in a similar orthotopic porcine heart transplantation study demonstrating viable recovery of the donor heart after 14 hours storage when both the donor and recipient were treated with intravenous cariporide (Ryan et al. 2003b). It has also been demonstrated that intravenous cariporide improves cardiac preservation utilising conventional ischaemic times in this model (Ryan et al. 2003a), and thus it is likely that the benefits observed in this study, such as superior cardiac contractility, would also apply to donor hearts subjected to conventional ischaemic times.

The orthotopic transplantation model used in this study was designed to mimic clinical transplantation. The Westran pigs utilised were sourced from a highly inbred colony in which graft immunotolerance has been demonstrated, thereby removing a potential confounder (O'Connell et al. 2005). The Landrace animals were also sourced from a single inbred colony. In addition, as discussed in Section 1.6.1, there are many anatomical and physiological similarities between pigs and humans, with both porcine

- 235 - and human myocardium being relatively intolerant of ischaemic preservation (White et al. 1986; Swindle et al. 1988; Swindle et al. 1998). There are however, some physiological differences between swine and humans (Hannon et al. 1990). Despite this, the porcine model has much more significance and applicability to human transplantation compared with other animal models and isolated working heart models.

All donor animals in the study were subjected to brain death and managed for six hours with a protocol incorporating combined hormone resuscitation. Hormone resuscitation has been shown to improve cardiac function and donor haemodynamics in both animal models and human subjects (Novitzky et al. 1987b; Cooper et al. 1988b; Novitzky et al. 1988c; Jeevanandam et al. 1993; Wheeldon et al. 1995; Hicks et al. 2006; Hing et al. 2007) and has been included in a US consensus statement on maximising organ recovery from cadaveric donors (Rosengard et al. 2002; Zaroff et al. 2002). The hormone resuscitation protocol described in the consensus statement has since been incorporated into the UNOS Critical Pathway algorithm for cadaveric donor management in the United States (Zaroff et al. 2002). Studies described earlier in this thesis (Chapters 3 and 4) have also demonstrated the beneficial effects of hormone resuscitation on donor haemodynamics and organ function (including the heart and kidneys), without having any significant adverse effects on other organs used for transplantation. On the basis of these studies, combined hormone resuscitation was incorporated in the donor management phase of this study.

The preservation protocols used in this study utilising Celsior, cariporide and GTN (i.e. CON, CAR1, CAR2, GTN and COMB groups) were designed based on the results from previous studies from this laboratory (Cropper et al. 2003a; Ryan et al. 2003a; Ryan et al. 2003b; Gao et al. 2005). All five experimental groups used in this study were comparable, with no significant differences between groups with respect to donor and recipient weights, donor left ventricular volume, donor management times and ischaemic times (Table 5.1). In addition, there were no significant differences between groups in the donor animals with respect to left ventricular contractility (Table 5.2), LVSW, MAP, CO and troponin I (Figures 5.3, 5.4 and 5.5, and Table 5.3). In terms of the LAD flow, there was a significant difference between the GTN and COMB groups at baseline. However, as no interventions had been commenced on any of the animals at baseline, it may be assumed that a type I statistical error occurred (i.e. a false positive

- 236 - result). Alternatively, the difference may have been due to species differences between the Westran and Landrace pigs, given that both animal size and heart size were matched between groups.

5.4.2 The Ability To Wean Transplanted Hearts Off Cardiopulmonary Bypass Support After 14 Hours Static Hypothermic Ischaemic Storage

The most significant and clinically relevant finding in this study was the successful weaning of transplanted hearts off cardiopulmonary bypass support after 14 hours of ischaemic storage of hearts in which both the donor and the recipient were pre-treated with IV cariporide (CAR1) and hearts that were arrested and stored in Celsior supplemented with GTN and cariporide (COMB). In contrast, donor hearts stored for the same period in Celsior solution alone or in Celsior solution supplemented with either cariporide or GTN showed poor recovery. Transplanted hearts in these groups (CON, CAR2 and GTN) could not be reliably weaned off cardiopulmonary bypass support after orthotopic transplantation.

5.4.3 The Effects Of Long-Term Storage On Left Ventricular Function And Haemodynamics After Transplantation

Cardiac contractility was assessed by PRSW analysis and haemodynamic indices (LVSW, CO and MAP). PRSW analysis of the groups weaned off cardiopulmonary bypass demonstrated that cardiac contractility in CAR1 animals was lower post- transplantation compared with baseline. In contrast, cardiac contractility was found to be higher post-transplantation in the COMB animals. Contractility was also found to be superior in COMB animals post-weaning compared with CAR1 animals, suggesting that cariporide and GTN in Celsior facilitates better cardiac preservation than pre-treatment with cariporide alone.

It has been previously reported that dobutamine influences the PRSW relationship and in the absence of ischaemic injury, dobutamine doses of 10 μg/kg/min have been shown to increase the SWI by 120% (Steendijk et al. 1998). In this study, all animals weaned

- 237 - off bypass received 10 μg/kg/min, thereby making the experimental groups comparable. The SWI had increased by no more than 25%, and in some cases had decreased by up to 15%. These changes in SWI reflect the complex changes occurring in the heart during the transplantation process and reflect the interaction between the effects of ischaemia- reperfusion injury (sustained during brain death and subsequent donor management, hypothermic storage of the heart, implantation and reperfusion) and dobutamine.

In the present study, LVSW after transplantation in the presence of dobutamine was similar to baseline LVSW measured prior to brain death and exposure to dobutamine. This indicates a markedly blunted response to dobutamine after transplantation relative to normal hearts, further demonstrating the negative effects of the transplantation process (and specifically of ischaemia-reperfusion injury) on the donor heart.

Whilst there was no statistically significant change in CO in both CAR1 and COMB groups post-transplantation compared with baseline, there was a trend towards higher CO after transplantation. In addition, CO was stable for up to three hours after weaning from cardiopulmonary bypass. Unlike CO, MAP was significantly lower post-transplant compared with baseline in CAR1 animals. There was also a trend towards a lower MAP in COMB animals post-transplant, although this was not statistically significant. The decline in blood pressure post-transplantation may be explained in part by the administration of dobutamine and isoflurane, both of which have vasodilatory properties. Despite the significant insults of brain death and 14 hours hypothermic storage, the transplanted hearts from the CAR1 and COMB groups generated sufficient CO to sustain stable circulatory haemodynamics in the transplanted animal. In addition, these parameters did not deteriorate for three hours post-weaning from cardiopulmonary bypass, indicating stability of the transplanted heart and the absence of worsening function due to reperfusion injury.

It should be noted that in terms of pure isolated contractile function, the PRSW relationship is an effective measure of cardiac function and its inotropic state because it is independent of preload, afterload and heart rate (Glower et al. 1985; Karunanithi et al. 2000). Other measures of haemodynamic function such as LVSW, MAP and CO are dependant on factors such as preload, afterload and heart rate, in addition to cardiac contractility. These can be influenced by factors such as systemic vascular resistance,

- 238 - pulmonary vascular resistance and circulating hormones. These are potential confounders in this study because measurements were made in different donor and recipient animals, and hence are not directly comparable. Therefore, PRSW is a more accurate measure of cardiac function because it excludes confounders such as loading conditions and heart rate. However, in clinical practice it is the haemodynamic indices that are most relevant and reflect the ability of the transplanted heart to support the circulation. In this study, the haemodynamic results support the weaning and cardiac contractility (PRSW) results.

5.4.4 Troponin I Release As A Marker Of Myocardial Injury

Troponin I has been shown to be a sensitive marker of myocardial injury in a porcine model of orthotopic heart transplantation (Ryan et al. 2003d). Troponin I release after transplantation was significantly higher in CON animals compared with COMB animals at both one and at three hours post-reperfusion. This was consistent with the transplantation results (i.e. weaning and haemodynamics post-transplantation), demonstrating less myocardial injury in the COMB animals. At similar time points, troponin I levels appeared to be higher in CON animals compared with CAR1 animals, however this was not statistically significant. Given the superior weaning, contractility and haemodynamic results, less cardiomyocyte injury as reflected in lower troponin I release might be expected in CAR1 and COMB groups compared with CON, CAR2 and GTN groups. As troponin I was a secondary end-point, the study may not have had sufficient statistical power to detect differences in troponin I release between groups.

5.4.5 Increases In Left Anterior Descending Coronary Arterial Flow After Transplantation

A striking finding in this study was the increase in LAD flows (in both CAR1 and COMB) after weaning from bypass post-transplantation compared with pre-explantation flows at six hours post-brain death. In addition, after weaning from bypass, there were no significant differences in LAD flow between CAR1 and COMB animals, and as with the haemodynamics and contractility, flows were stable for up to three hours after weaning from bypass. It was also noted that the LAD flow was higher in the GTN group compared with the COMB group at baseline, as discussed earlier in Section 5.4.1. It is

- 239 - unlikely that this difference in LAD flow at baseline had any significant impact on the primary end-point given that none of the GTN animals were weaned off bypass support. These groups of animals behaved as predicted, when considering the rat studies by Gao et al (2005).

The increased LAD flow post transplantation is most likely explained by reactive coronary hyperaemia after prolonged ischaemia (Budrikis et al. 1998). The significance of this finding is uncertain, but Budrikis and colleagues reported that the intensity and duration of early reactive hyperaemia after ischaemic storage were related to the extent of preservation injury in an isolated blood-perfused porcine heart model (Budrikis et al. 1998). Loss of normal regulatory control of LAD flow in the denervated transplanted heart could also contribute to the increased LAD flow observed post-weaning, but this seems unlikely as no consistent significant changes in LAD flow were observed during the six hour period following induction of brain death. Furthermore, other investigators have not observed any increase in coronary blood flow after cardiac denervation in a porcine model (Ootaki et al. 2007).

5.4.6 The Role Of Cariporide To Ameliorate Ischaemia-Reperfusion Injury In Transplantation

Many experimental studies have demonstrated a powerful cardio-protective effect of cariporide and other NHE inhibitors in the setting of hypothermic ischaemia and heart transplantation (Myers et al. 1996; Kim et al. 1998a; Kim et al. 1998b; Kevelaitis et al. 2001; Cropper et al. 2003a; Ryan et al. 2003a). However, these studies did not incorporate brain death, prolonged management of the donor, hormone resuscitation of the donor, prolonged ischaemic storage and have control comparisons as in the transplantation model described in this thesis. In a porcine model of orthotopic heart transplantation, Martin and colleagues investigated the use of adenosine and cariporide supplemented, leukocyte-depleted blood cardioplegia used to perfuse the hearts of non heart beating donor (NHBD) animals 30 minutes after circulatory arrest from exsanguination (equating to 30 minutes of normothermic ischaemia) (Martin et al. 2003a). These hearts were stored for three hours and then transplanted orthotopically, and then reperfused with adenosine and cariporide supplemented blood cardioplegia.

- 240 - They demonstrated that by using the above-described preservation technique, the recovery of donor hearts from NHBD were comparable to organs used from beating heart donors. Unlike the present study however, their model did not incorporate brain death and had only three hours storage. Kim et al demonstrated benefit with cariporide pre-treatment in a brain-dead donor transplantation model utilising 4 hours storage (Kim et al. 1998b). Following on from this, they went on to show that cariporide pre- treatment of donors and recipients in a non-brain-dead canine transplantation model with 24 hour storage improved myocardial compliance, post weaning cardiac index, myocardial ultrastructure and weaning potential (Kim et al. 1998a). However, this model did not have a prolonged donor management period nor did it utilise hormone resuscitation. They also found no benefit to adding cariporide to cardioplegia or in pre- treating the recipient only. In addition, as control animals were weaned off bypass after 24 hours, this would suggest that the canine heart, unlike the human heart, is relatively resistant to myocardial ischaemia and prolonged storage. Other studies with cariporide in isolated non-brain-dead perfused heart experiments have found improved function and reduced injury in pig (Scheule et al. 2003), rabbit (Myers et al. 1996) and rat models (Kevelaitis et al. 2001; Cropper et al. 2003a; Cropper et al. 2003b), utilising 2.5 to 12 hours storage times. However, as with the studies described above, these models did not incorporate brain death or donor management.

Work from this laboratory has previously demonstrated that donor and recipient pre- treatment with cariporide in a porcine transplant model facilitates weaning from cardiopulmonary bypass and better preservation of ventricular contractility after 4 and 14 hours storage (Ryan et al. 2003a; Ryan et al. 2003b). However, these studies did not incorporate prolonged donor management or hormone resuscitation. Hearts were explanted one hour after brain death and therefore were not exposed to the long-term damaging effects of brain death as in the studies presented in this thesis. Therefore, whilst there have been many studies demonstrating the benefits of cariporide in a transplantation setting, the model used here has been arguably the most clinically relevant, replicating much of what happens in human transplantation by incorporating brain death, hormone resuscitation and prolonged donor management.

Despite the studies that have been reported in the literature, the optimal mode of delivery and timing of administration of cariporide remains uncertain. Some studies

- 241 - have reported that the cardio-protective effect of cariporide is greater when the drug is administered intravenously to both donor and recipient, compared with when the drug is added to the donor preservation solution alone (Kim et al. 1998b; Cropper et al. 2003a). In addition, concerns have been raised regarding the safety of systemically administered cariporide (Mentzer Jr 2003; Mentzer Jr et al. 2008). Initially, the GUARDIAN trial demonstrated the safety of cariporide in humans and that in surgical ischaemia- reperfusion settings such as coronary artery bypass surgery (CABG), cariporide was cardioprotective (Theroux et al. 2000). These results formed the rationale for the EXPEDITION trial, which examined the safety and efficacy of cariporide in preventing death or myocardial infarction (MI) in patients undergoing high risk CABG (Mentzer Jr 2003; Mentzer Jr et al. 2008). Whilst cariporide was shown to reduce the composite primary endpoint of death or MI (20.3% in placebo vs. 16.6% in treatment group; p=0.0002), the mortality rate increased from 1.5% in the placebo group to 2.2% with cariporide (p=0.02). This increase in mortality was associated with an increase in cerebrovascular events and prompted the authors to conclude that cariporide was unlikely to be used clinically. It should be noted though, that the EXPEDITION trial used large doses of cariporide given as an IV infusion (1140 to 1620 mg over 49 hours), which may have contributed to the adverse outcomes. Despite these outcomes, the authors also concluded that NHE inhibitors still held promise as a new class of drugs to reduce myocardial injury associated with ischaemia-reperfusion injury.

In the study reported in this chapter, cariporide was administered intravenously in the CAR1 group at considerably lower doses to those used in the EXPEDITION trial. An average of 86 mg of cariporide was used in the donor (much of which may have been flushed out with Celsior solution during donor heart explantation and later transplantation) and 84 mg was given to the recipient in the CAR1 group. Despite these lower doses, the concern regarding the potential toxicity of systemically administered cariporide prompted a re-evaluation of the use of cariporide. It was therefore decided to examine the use of cariporide as a supplement to existing preservation solutions (CAR2 and COMB groups) in order to reduce the recipient exposure to cariporide. In these groups, 3.79 mg (10 μmol) of cariporide was mixed in 1000 mL of Celsior and used as a cardioplegia and storage solution (with no intravenous administration). Although plasma cariporide levels were not measured in the CAR2 and COMB groups, it is likely that exposure of the recipient animals in these groups to cariporide would have been

- 242 - negligible because there was no direct administration of cariporide to the recipient and any cariporide-containing preservation solution was flushed from the heart with unsupplemented Celsior solution during implantation.

Previous studies have reported that the cardioprotective capacity of cariporide when added to existing preservation solutions appears limited (Kim et al. 1998b; Cropper et al. 2003a), and this was confirmed in the present study in which only one of five animals in the CAR2 group could be weaned from cardiopulmonary bypass post- transplant. It is unlikely that the addition of a higher concentration of cariporide to Celsior solution would have enhanced its cardioprotective action because the concentration of cariporide added to Celsior solution in this study (10 μmol/L) was the same as that which we found produced maximal cardioprotection in an isolated working rat heart model (Cropper et al. 2003a).

5.4.7 The Role Of Glyceryl Trinitrate To Ameliorate Ischaemia- Reperfusion Injury In Transplantation And Its Synergistic Interaction With Cariporide

Several investigators have shown that the cardioprotective capacity of cariporide is markedly enhanced when combined with ischaemic preconditioning (Kevelaitis et al. 2001) or with pharmacological agents that may mimic this phenomenon, including NO donors (Gao et al. 2005), adenosine agonists (Kristo et al. 2004) and mitochondrial

+ K ATP agonists such as diazoxide and BMS-180448 (Kevelaitis et al. 2001; Cropper et al. 2003b). In this study, we chose to combine cariporide with GTN for a number of reasons. GTN is already widely used in clinical medicine. There have been numerous studies investigating the use of exogenous NO donors such as GTN in a variety of hypothermic preservation solutions to improve organ function in various animal models of heart and lung transplantation (Oz et al. 1993; Pinsky et al. 1994a; Pinsky et al. 1994b; Bhabra et al. 1996; Du et al. 1998; Baxter et al. 1999a; Bando et al. 2000; Kawashima et al. 2000; Baxter et al. 2001; Maczewski et al. 2003; Duranski et al. 2005; Gao et al. 2005). Many of these studies have demonstrated improvements in preservation and organ function. GTN exerts its protective anti-ischaemic effects on the donor heart via a number of pathways. These include activation of guanylate cyclase to

- 243 - increase cGMP production, which in turn dilates vascular beds during hypothermic storage (Pinsky et al. 1994a), and also via direct cardioprotective mechanisms mediated

+ by the activation of mitochondrial K ATP (Ferdinandy et al. 1995; Sasaki et al. 2000; Csont et al. 2005). Research in this laboratory utilising GTN and cariporide in Celsior in an isolated working rat heart model has already previously demonstrated superior organ preservation and recovery after ischaemic storage for up to 10 hours (Gao et al. 2005). No recovery was seen in the rat heart when Celsior was used with either agent alone. Furthermore, the myocardial preservation achieved by combined supplementation of Celsior solution was superior to that achieved by perfusion of the isolated rat heart with cariporide prior to storage and reperfusion. Building on this, the effects of GTN and cariporide in Celsior in the porcine model were investigated.

The results reported in this chapter closely matched the results seen in the isolated working rat heart model (Gao et al. 2005). GTN-supplemented (GTN group) or cariporide-supplemented Celsior (CAR2 group) did not facilitate recovery of the transplanted porcine heart to the extent of being consistently weaned off cardiopulmonary bypass support after transplantation. However, when both GTN and cariporide were added to Celsior (COMB group), viable recovery of the transplanted heart after 14 hours storage was achievable. This suggests an additive or synergistic interaction between cariporide and GTN. Whilst the exact mechanism of this synergy is unknown, there are a number of possible explanations. One possible explanation is the combined activation (phosphorylation) of the pro-survival kinase ERK1/2, which constitutes one limb of the Reperfusion Injury Salvage Kinase (RISK) Pathway (Hausenloy et al. 2004). ERK1/2 has been found to be weakly phosphorylated by either cariporide or GTN alone, but is strongly activated by the combination of the two agents (Kwan et al. 2008). This activation of ERK1/2 is associated with reduced apoptosis as assessed by cleaved caspase-3 expression on immunohistochemistry and Western blotting (Kwan et al. 2007). Another possible mechanism may be that in addition to their own independent anti-ischaemic effects (Ferdinandy et al. 1995; Sasaki et al. 2000; Csont et al. 2005), GTN may also improve delivery of cariporide to cardiac myocytes via its vasodilatory effects and hence improve the efficacy of cariporide. This may explain why cariporide-supplemented Celsior (CAR2 group) did not result in transplanted hearts that were weaned off cardiopulmonary bypass, as the cariporide may

- 244 - not have been able to be delivered effectively to the cardiac myocytes at time of cardioplegia via non-dilated coronary vessels.

It is possible that combining pre-treatment of the donor with intravenous cariporide prior to storage of the donor heart in Celsior supplemented with both cariporide and GTN may facilitate further enhancement of cardiac preservation. However, this was not investigated in this study because little is known about the impact of cariporide on other organs used for transplantation.

5.4.8 Study Limitations

The main limitation to this study is that transplanted animals were monitored for only three hours after weaning from cardiopulmonary bypass. It is possible that transplanted hearts that were successfully weaned from cardiopulmonary bypass may have failed at a later time-point. However, this is unlikely because cardiac function and circulatory haemodynamics remained stable throughout the three hour observation period and showed no evidence of deterioration of function. Longer follow-up of the transplanted animals would also have been of interest to determine short- to medium-term cardiac function and also the incidence of early rejection, and whether interventions such as hormone resuscitation and the various preservation strategies affect these outcomes.

Another limitation to this study includes the absence of histopathological examination of the heart post-transplantation compared with baseline and with pre-storage, and to assess whether the various preservation strategies affected myocardial structure and architecture. Assessment of the degree of cellular necrosis and apoptosis would have been complementary, as the use of NHE inhibitors has been linked with a decrease in ischaemia-reperfusion injury such as necrosis and apoptosis (Allen et al. 2000; Avkiran et al. 2002; Teshima et al. 2003).

Other measures of cardiac function such as left atrial pressures, ejection fraction and diastolic function were also not fully assessed in this study. These limitations, along with those already described, were mainly due to research funding and staffing considerations, which precluded more extensive cardiac analyses, and increased monitoring duration and follow-up of animals in this project. Despite this, important

- 245 - clinical parameters such as weaning from bypass support after transplantation, left ventricular contractile function and haemodynamics were investigated to provide evidence for the potential success of the treatment strategies tested in this clinically relevant porcine model.

Finally, other limitations of this study included the operating surgeon (AH) not being blinded to the treatment strategies and the pigs used in the study being healthy juvenile animals (<12 months). However, it was believed that the study design and methodologies were rigorous enough that the lack of blinding was not likely to be critical and should not have introduced bias into the results, particularly as all protocols were standardised in advance and strictly adhered to regardless of experimental group, . With respect to the use of juvenile pigs, this was done mainly due to the size of adult pigs and the logistical difficulties presented by using larger adult animals. Certainly the use of non-adult animals may perhaps limit the potential to extrapolate these results to an adult human heart transplantation population, particularly with respect to older donors with pre-existing co-morbidities.

5.5 CONCLUSION

This study has been one of the first studies to demonstrate viable recovery of a transplanted heart after 14 hours ischaemic storage in a large animal model that incorporates brain death and prolonged management (six hours) of the brain-dead donor that included hormone resuscitation. In summary, this study has demonstrated that supplementation of Celsior preservation solution with cariporide and GTN, or alternatively the pre-treatment of donors and recipients with cariporide, permits viable recovery of the transplanted heart after 14 hours ischaemic storage. This has been shown in a large animal model that was designed to mimic the sequence of events in human heart transplantation.

In addition, this study has shown that post-transplant myocardial contractility is best preserved and troponin ‘leak’ is lowest when donor hearts are stored in Celsior solution supplemented with both cariporide and GTN. It is believed that this preservation solution provides improved myocardial protection during prolonged hypothermic

- 246 - storage while avoiding any potential hazards that might result from systemic administration of cariporide to the recipient.

This work provides the basis for further investigation of cariporide and GTN in heart transplantation and its potential use in human transplantation involving long periods of ischaemic hypothermic storage. Its potential benefits might permit the extension of safe ischaemic preservation time of donor hearts, thereby providing many logistical advantages. These treatments may also improve the quality of donor hearts utilising “normal” ischaemic times, allow use of more marginal donors (thereby increasing the number of donor organs available for transplantation), and improve both short- and long-term outcomes in heart transplantation. Prior to its adoption into routine clinical transplantation, future work must be directed at determining the safety and efficacy of cariporide in transplantation, its effects on other donor organs and to also optimise cariporide dosage, whilst at the same time minimising the exposure of the recipient to cariporide, given the results from the EXPEDITION trial (Mentzer Jr et al. 2008).

- 247 -

CHAPTER 6

FINAL DISCUSSION: A SUMMARY OF THE KEY FINDINGS, THEIR RELEVANCE TO CLINICAL PRACTICE AND THE SCOPE FOR FUTURE RESEARCH

- 248 - CHAPTER 6

FINAL DISCUSSION: A SUMMARY OF THE KEY FINDINGS, THEIR RELEVANCE TO CLINICAL PRACTICE AND THE SCOPE FOR FUTURE RESEARCH

6.1 CARDIAC TRANSPLANTATION AS A TREATMENT FOR END-STAGE HEART FAILURE

Death from heart disease ranks amongst the top five causes of death in both males and females in Australia (AIHW 2008). Similar trends are seen in many countries around the world. At the same time, heart failure is a rapidly growing health problem in many countries. Approximately 263 000 Australians (1.3% of the population) reported symptoms of heart failure in the National Health Survey 2004-05 and 2 225 Australian deaths were attributable to heart failure (AIHW 2008).

Despite the advances made in the treatment of heart failure that have encompassed approaches such as pharmacological agents, biventricular pacing devices, various non- transplantation surgical procedures and mechanical circulatory support devices (e.g. the and ventricular assist devices), cardiac transplantation remains an established and widely used therapeutic option for the treatment of severe end-stage heart failure that is refractory to all other therapies (Bolling et al. 2001; Magliato et al. 2001; Kasper et al. 2002; Al-khaldi et al. 2006). In the United States alone, it has been estimated that between 25 000 to 65 000 patients might benefit from a cardiac transplant (Copeland 2001; Zaroff et al. 2002). However despite its successes, there are still many challenges faced by transplantation that limit its availability and also limit its success, in terms of morbidity and mortality in transplant recipients.

It is well recognised that the demand for cardiac transplantation is not matched by the supply of suitable donor hearts (Copeland 2001; Conte et al. 2002; Kasper et al. 2002). There is a need to increase the number of donor organs available for transplantation and to also optimise the usage of organs that are available from the currently available pool

- 249 - of organ donors. The process of brain death in the organ donor results in a series of haemodynamic, neurohormonal, and pro-inflammatory/immune system disturbances that have the potential to render donor organs, especially the heart, unsuitable for transplantation, and in those other organs that are transplanted, can lead to increased morbidity and mortality in the recipient (Cooper et al. 1989; Bittner et al. 1996b; Novitzky 1997a; Wilhelm et al. 2000b; Seguin et al. 2001; Hicks et al. 2006). Ischaemia-reperfusion injury that occurs during transplantation also has an impact on donor organ quality and outcomes in transplant recipients, with increased ischaemic times being associated with increased ischaemia-reperfusion injury and increased recipient mortality and morbidity risk (Fleischer et al. 2002; Hicks et al. 2006; Taylor et al. 2008).

The research described in this thesis examined some of the problems associated with organ transplantation, in particular examining treatments to reverse the negative effects of brain death on the multi-organ donor and their organs, and investigation into treatments to ameliorate ischaemia-reperfusion injury to the donor heart in an effort to improve donor heart quality, and effectively extend hypothermic ischaemic preservation times. More specifically, the hormone resuscitation protocol previously advocated in the UNOS Critical Pathway algorithm for donor management (Rosengard et al. 2002; Zaroff et al. 2002) was investigated and the role of cariporide and glyceryl trinitrate (GTN) in minimising cardiac ischaemia-reperfusion injury was examined.

6.2 SUMMARY OF THE KEY FINDINGS IN THIS THESIS

Utilising a porcine model of the brain-dead organ donor with orthotopic heart transplantation, the research described in this thesis investigated the effects of hormone resuscitation on the donor heart and on the haemodynamics of the multiorgan donor. Having demonstrated the benefits of hormone resuscitation on cardiac function and haemodynamics in Chapter 3, the question of what effects hormone resuscitation had on other donor organs, namely the lungs, kidneys, liver and pancreas, was addressed in Chapter 4. Once the safety and efficacy of donor hormone resuscitation was established, this treatment was then incorporated into the porcine model of orthotopic heart transplantation. Chapter 5 investigated the potential for cardiac myocyte Na+-H+ exchanger inhibition with cariporide and the use of the nitric oxide (NO) donor GTN to

- 250 - minimise ischaemia-reperfusion injury encountered during the transplantation process, and to safely extend the hypothermic ischaemic preservation time of the donor heart to 14 hours.

One of the key findings in Chapter 3 was that hormone resuscitation of the brain-dead organ donor enabled a high proportion of donors (75%) to be weaned off noradrenaline and vasopressin support. Of those not weaned off noradrenaline, all were on lower doses of noradrenaline support after hormone resuscitation was instituted. In comparison, none of the control animals were weaned off noradrenaline. Despite significant increases in noradrenaline (up to a median of 0.563 μg/kg/min), many donor animals did not achieve mean arterial blood pressures (MAP) over 60 mmHg. In fact, tachyphylaxis to noradrenaline was seen frequently. In terms of left ventricular contractility, hormone resuscitation was associated with significantly better contractile function, both over time and in comparison with the control group. The difference in contractility between groups was further accentuated when the noradrenaline dose was fixed in both hormone-treated and control groups at 3.3 μg/min. Similarly, at fixed noradrenaline doses, cardiac output, stroke work, stroke volume and blood pressure were significantly higher in hormone-treated donors compared with the control group. On high doses of noradrenaline towards the end of the study (six hours after brain death induction), cardiac output, stroke work and stroke volume were similar between both groups but blood pressure was significantly lower in the control group. In addition, in the series of experiments reported in Chapter 3, hormone resuscitation did not have any adverse effects on lung function or blood glucose levels despite the use of steroids.

The study described in Chapter 4 was designed to compare three different treatments of the brain-dead multi-organ donor. Hormone resuscitation was compared with more conventional treatments based on intravenous (IV) fluids only and on noradrenaline only. Blood pressure was found to be higher in the hormone-treated group compared with those treated with noradrenaline alone or IV fluids alone. Blood pressure also declined over time in both noradrenaline- and IV fluid-treated groups. Not surprisingly, heart rate was higher in noradrenaline-treated animals compared with hormone-treated and IV fluid-treated animals. Cardiac output, stroke work and stroke volume were higher in hormone-treated animals at six hours after brain death induction compared with the noradrenaline-treated animals. Both stroke work and cardiac output declined

- 251 - over time in the noradrenaline group. Interestingly, despite low MAP (38±12 mmHg at six hours after brain death induction), donor animals treated with IV fluids survived to six hours with none having a hypotensive cardiac arrest.

In terms of pulmonary function, noradrenaline-treated animals had significantly worse

PaO2 and Aa (alveolar-arterial) gradient compared with hormone-treated and IV fluid- treated animals. There were significant reductions in both PaO2 and Aa gradient over time in the noradrenaline-treated animals. Noradrenaline treatment was also found to be detrimental to acid-base balance. From three to six hours after brain death induction, pH was significantly lower in the noradrenaline group (falling to 7.08 at six hours) compared with the hormone-treated and IV fluid-treated animals. Similar trends were seen in the base excess, with more negative base excesses in noradrenaline animals, down to -16.

Renal function was well preserved by hormone treatment compared with noradrenaline and IV fluid treatments. Hormone resuscitation was associated with significantly higher renal arterial flow compared with noradrenaline-treated animals from four to six hours after brain death induction and by six hours it was superior to fluid-treated animals also. Renal arterial flow declined significantly over time in both noradrenaline- and IV fluid- treated animals. At the same time, creatinine clearance was significantly higher in hormone-treated animals compared with noradrenaline-treated animals. Creatinine clearance declined significantly over time in both the noradrenaline- and IV fluid- treated groups but not in the hormone-treated group.

With respect to the liver, hormone resuscitation did not have any detrimental effects on blood flow to the liver (portal vein and hepatic artery), bile production, liver function tests (alanine aminotransferase, aspartate aminotransferase and bilirubin) or hepatic synthetic function (bile production and INR-international normalised ratio), compared with noradrenaline and IV fluid treatments. Similarly, hormone resuscitation did not have any detrimental effects on the pancreas as reflected by the pancreatic enzymes, lipase and amylase.

The heart transplant study in Chapter 5 demonstrated that effective long-term preservation of the donor heart could be achieved with pre-treatment of donors and

- 252 - recipients with IV boluses of cariporide prior to ischaemia and reperfusion, respectively. The results in Chapter 5 also demonstrated that the addition of GTN and cariporide to Celsior was effective in ameliorating the effects of ischaemia-reperfusion injury and achieving effective long-term preservation of the donor heart. This supplemented Celsior solution was used as both a cardioplegic and storage solution.

Superior left ventricular function after transplantation following 14 hours hypothermic ischaemic storage was seen in GTN and cariporide-supplemented Celsior compared with the cariporide pre-treated animals. Hearts transplanted in the supplemented Celsior group had an increased stroke work index at three hours after weaning from cardiopulmonary bypass compared with donor baseline. This was in contrast to the cariporide pre-treatment group, which had a reduction in stroke work index at similar time-points. Importantly, after reperfusion, left ventricular contractility (as measured by the preload recruitable stroke work relationship), stroke work, blood pressure and cardiac output were stable for up to three hours after weaning off cardiopulmonary bypass support in both the cariporide pre-treatment group and the GTN plus cariporide- supplemented Celsior group. This suggests that these treatments provide effective protection from reperfusion injury. In addition, troponin I release was found to be highest in the control group and lowest in the combined GTN plus cariporide supplemented Celsior group. This finding implies that there was less cardiac myocyte injury in the combined group compared with controls, demonstrating that GTN plus cariporide protects the myocyte from ischaemia-reperfusion injury.

6.3 THE RELEVANCE TO CLINICAL TRANSPLANTATION

The combined findings from the studies described in this thesis support the use of hormone resuscitation in the brain-dead multi-organ donor. Hormone resuscitation was associated with improved haemodynamics, left ventricular function and renal function. It also significantly reduced inotrope (i.e. noradrenaline) requirements in the donor, which is beneficial for both the donor and transplant recipient. Importantly, hormone resuscitation did not have any demonstrable detrimental effects on any of the thoracic and abdominal transplantable solid organs investigated. Translation of these findings into clinical practice may be expected to improve the quality of donor hearts, including “sub-optimal” or marginal hearts that are currently rejected for transplantation. The

- 253 - improvement of cardiac function by these treatments might also be expected to improve perfusion of other transplantable organs and have the potential to improve the quality and outcomes of transplantation of these organs including the lungs, liver, kidney and pancreas.

Based in part on the results from Chapter 3 of this thesis, agreement from major stakeholders in organ donation in New South Wales (LifeLink Organ Donation Network NSW, and Australian and New Zealand Intensive Care Society (NSW State Liaison Group)) was obtained to conduct a pilot study utilising hormone resuscitation in unstable organ donors across NSW and ACT in Australia. Results from this study are yet to be published, although preliminary findings have indicated increased heart utilisation from unstable donors treated with hormone resuscitation compared with those who did not receive hormone resuscitation (P. Macdonald, personal communication).

The development of improved preservation strategies to ameliorate the negative effects of ischaemia-reperfusion injury on donor organs has important implications for the practice of clinical transplantation. A major potential benefit includes the extension of safe ischaemic preservation times for donor hearts, which could have extensive logistical advantages. For example, donor hearts could be used from a larger geographical area and the extended preservation times would allow extra time for tissue matching of recipients with donors. These proposed treatments may also improve the quality of donor organs utilising “normal” ischaemic times, allow the use of more marginal donors (thereby increasing the number of donor organs for transplantation), and improve both short- and long-term transplantation outcomes.

6.4 STUDY LIMITATIONS

The limitations of the studies described in this thesis have been discussed in detail in each chapter. All three studies described in this thesis had relatively short observation times. Donors in Chapters 3, 4 and 5 were managed for a total of six hours, and transplant recipients in Chapter 5 were observed for three hours after weaning off cardiopulmonary bypass support. Whilst the observation times were shorter than the times seen in clinical practice, the results from these studies do provide information on

- 254 - early function and on the potential outcomes that may be extrapolated to the clinical situation.

The effects of the individual components of the hormone resuscitation protocol were not examined in this thesis, nor were the molecular and cellular effects of hormone resuscitation examined. This was beyond the scope of the aims of this thesis. As the hormone resuscitation protocol tested in this thesis had been previously advocated for clinical transplantation (Rosengard et al. 2002; Zaroff et al. 2002), the aim of this thesis was to test the hormone protocol in its entirety and to observe its effects on the donor. Similarly, assessment of the effects of hormone treatments versus those of noradrenaline and IV fluids on the histology, immunohistochemistry, and the expression of cytokines and other inflammatory and protective molecules were not examined in this thesis. These issues were also beyond the scope of this thesis, but they do provide the basis for future work.

The studies in Chapters 3 and 4 did not examine transplantation outcomes and were primarily donor management studies. Whilst these studies were designed to examine the effects of various treatments on the donor and their organs, it would be important to examine the effects of these treatments on transplant outcomes, both short- and long- term in future studies.

Despite these limitations, the studies in this thesis were undertaken using a clinically relevant model of the organ donor and transplantation that incorporated brain death and an extended period of donor management in an intensive care unit-style setting. This has been done in very few large animal studies of donor management and transplantation. By utilising this model, the data generated provide the basis for future human clinical trials into the role of hormone resuscitation, GTN and sodium-hydrogen exchange inhibitors such as cariporide in transplantation.

6.5 FUTURE DIRECTIONS

The results of the studies described in this thesis raise as many questions as they answer. As discussed in the previous sections, the results presented provide some basis for further experimental and clinical studies. Additional experimental studies must still

- 255 - be undertaken to examine the effects of individual hormones on the donor and their organs, including the structure and function of transplantable solid organs. The expression of pro-inflammatory and protective molecules in donor brain death and the effects of various treatments, including hormone treatments, on their expression are also yet to be investigated. Some of these will be addressed by the continuation of this work in this laboratory, utilising tissue and blood collected and stored during the course of these studies.

Given the encouraging results of hormone resuscitation in the organ donor, further clinical studies of this treatment and its effects on transplantation are warranted. As yet, there are no published reports of prospective randomised controlled trials of hormone resuscitation in humans in the manner undertaken in the studies in this thesis. Most reports available to date are retrospective reviews and hence the results described in this thesis form the basis for future studies into donor management and transplantation outcomes.

The results of donor heart preservation in this thesis demonstrate a role for cariporide in mitigating ischaemia-reperfusion injury. Whilst the EXPEDITION trial has cast doubt over the safety of cariporide, it should be noted that the doses and delivery of cariporide in the study in Chapter 5 are quite different to those of the EXPEDITION trial, both in terms of cariporide doses and delivery of the drug. It is yet to be seen whether the use of cariporide in cardioplegic and preservation solution, which is not given systemically to the transplant recipient, would have the same adverse events as those seen in the EXPEDITION trial. In addition, if cariporide were to have adverse effects on humans when used in cardioplegic and storage solutions, then other inhibitors of the sodium- hydrogen exchanger such as enaporide and zoniporide, may prove suitable alternatives. These issues remain to be examined in future experimental and clinical studies.

Despite the number of publications reporting various treatments of the brain-dead organ donor and also examining various treatments to mitigate ischaemia-reperfusion injury, there is still much to be determined. As yet, there is still no universally accepted and proven donor management strategy. Similarly, there is no universally accepted and proven organ preservation strategy. In the quest to develop optimal protocols, there

- 256 - remains a great deal to be done in terms of experimental and clinical studies. It is hoped that this thesis makes a worthwhile contribution towards this endeavour.

- 257 -

REFERENCES

- 258 - REFERENCES

AIHW (2008). Australia's health 2008. Cat. no. AUS 99. Canberra: Australian Institute of Health and Welfare.

Al-khaldi, A. and Robbins, R. C. (2006). New directions in cardiac transplantation. Annual Review of Medicine 57: 455-71.

Allen, D. G. and Xiao, X. H. (2000). Activity of the Na+/H+ exchanger contributes to cardiac damage following ischaemia and reperfusion. Clinical and Experimental Pharmacology and Physiology 27(9): 727-33.

Amado, J. A., Lopez-Espadas, F., Vazquez-Barquero, A., Salas, E., Riancho, J. A., Lopez-Cordovilla, J. J. and Garcia-Unzueta, M. T. (1995). Blood levels of cytokines in brain-dead patients: relationship with circulating hormones and acute-phase reactants. Metabolism: Clinical & Experimental 44(6): 812-6.

Anonymous (1968). A definition of irreversible coma. Report of the Ad Hoc Committee of the Harvard Medical School to Examine the Definition of Brain Death. JAMA 205(6): 337-40.

ANZICS (2008). The Australian and New Zealand Intensive Care Society Statement on Death and Organ Donation. (3rd Edition). Melbourne: ANZICS.

Arita, K., Uozumi, T., Oki, S., Kurisu, K., Ohtani, M. and Mikami, T. (1993). The function of the hypothalamo-pituitary axis in brain dead patients. Acta Neurochirurgica 123(1-2): 64-75.

Avkiran, M. (1999). Rational basis for use of sodium-hydrogen exchange inhibitors in myocardial ischemia. American Journal of Cardiology 83(10A): 10G-17G; discussion 17G-18G.

Avkiran, M. (2003). Basic biology and pharmacology of the cardiac sarcolemmal sodium/hydrogen exchanger. Journal of Cardiac Surgery 18(Suppl 1): 3-12.

- 259 - Avkiran, M. and Marber, M. S. (2002). Na(+)/H(+) exchange inhibitors for cardioprotective therapy: progress, problems and prospects. Journal of the American College of Cardiology 39(5): 747-53.

Avlonitis, V. S., Fisher, A. J., Kirby, J. A., Dark, J. H., Avlonitis, V. S., Fisher, A. J., Kirby, J. A. and Dark, J. H. (2003). Pulmonary transplantation: the role of brain death in donor lung injury. Transplantation 75(12): 1928-33.

Avlonitis, V. S., Wigfield, C. H., Golledge, H. D., Kirby, J. A., Dark, J. H. and Golledge, H. D. R. (2007). Early hemodynamic injury during donor brain death determines the severity of primary graft dysfunction after lung transplantation. American Journal of Transplantation 7(1): 83-90.

Avlonitis, V. S., Wigfield, C. H., Kirby, J. A., Dark, J. H., Avlonitis, V. S., Wigfield, C. H., Kirby, J. A. and Dark, J. H. (2005). The hemodynamic mechanisms of lung injury and systemic inflammatory response following brain death in the transplant donor. American Journal of Transplantation 5(4 Pt 1): 684-93.

Azzam, Z. S., Adir, Y., Crespo, A., Comellas, A., Lecuona, E., Dada, L. A., Krivoy, N., Rutschman, D. H., Sznajder, J. I. and Ridge, K. M. (2004). Norepinephrine increases alveolar fluid reabsorption and Na,K-ATPase activity. American Journal of Respiratory and Critical Care Medicine 170(7): 730-6.

Baldan, N., Toffano, M., Cadrobbi, R., Codello, L., Calabrese, F., Bacelle, L. and Rigotti, P. (1997). Kidney preservation in pigs using celsior, a new organ preservation solution. Transplantation Proceedings 29(8): 3539-40.

Bando, T., Albes, J. M., Schone, J., Wada, H., Hitomi, S., Wahlers, T. and Schafers, H. J. (2000). Significance of cyclic adenosine monophosphate and nitroglycerin in ET- Kyoto solution for lung preservation. Annals of Thoracic Surgery 69(3): 887-91.

Barnard, C. N. (1967). A human cardiac transplant: an interim report of a successful operation performed at Groote Schuur Hospital, Capetown. South African Medical Journal 41: 1271-1274.

- 260 - Baroldi, G., Di Pasquale, G. and Silver, M. D. (1997). Type and extent of myocardial injury related to brain damage and its significance in heart transplantation: a morphometric study. The Journal of Heart and Lung Transplantation 16: 994-1000.

Baumgartner, W. A. (2002). Operative techniques used in heart transplantation. Heart and Lung Transplantation. W. A. Baumgartner, Reitz, B., Kasper, E. and Theodore, J. Philadelphia, W.B. Saunders Company: 180-199.

Baxter, K., Howden, B., Saunder, A. and Jablonski, P. (1999a). Improved cardiac preservation by the addition of nitroglycerine to colloid-free University of Wisconsin solution (MUW). Journal of Heart and Lung Transplantation 18(8): 769-74.

Baxter, K., Howden, B. O. and Jablonski, P. (1999b). Cardiac preservation with nitroglycerine and cyclic adenosine monophosphate. Transplantation Proceedings 31(1- 2): 1046-7.

Baxter, K., Howden, B. O. and Jablonski, P. (2001). Heart preservation with celsior solution improved by the addition of nitroglycerine. Transplantation 71(10): 1380-4.

Beck, W. S. (1991). Haematology. Boston: MIT Press.

Beckman, J. S., Beckman, T. W., Chen, J., Marshall, P. A. and Freeman, B. A. (1990). Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proceedings of the National Academy of Science of the United States of America 87(4): 1620-4.

Beijer, H. J., Maas, A. H. and Charbon, G. A. (1984). A vasopressin-induced decrease in pancreatic blood flow and in pancreatic exocrine secretion in the anesthetized dog. Pflugers Archiv-European Journal of Physiology. 400(3): 324-8.

Belzer, F. O. and Southard, J. H. (1988). Principles of solid-organ preservation by cold storage. Transplantation 45(4): 673-6.

Berthiaume, Y., Folkesson, H. G. and Matthay, M. A. (2002). Lung edema clearance: 20 years of progress: invited review: alveolar edema fluid clearance in the injured lung. Journal of Applied Physiology 93(6): 2207-13.

- 261 - Bhabra, M. S., Hopkinson, D. N., Shaw, T. E. and Hooper, T. L. (1996). Relative importance of prostaglandin/cyclic adenosine monophosphate and nitric oxide/cyclic guanosine monophosphate pathways in lung preservation. Annals of Thoracic Surgery 62(5): 1494-9.

Biguzas, M., Jablonski, P., Howden, B. O., Thomas, A. C., Walls, K., Scott, D. F. and Marshall, V. C. (1990). Evaluation of UW solution in rat kidney preservation. II. The effect of pharmacological additives. Transplantation 49(6): 1051-5.

Birks, E. J., Burton, P. B., Owen, V. J., Latif, N., Nyawo, B. and Yacoub, M. H. (2001). Molecular and cellular mechanisms of donor heart dysfunction. Transplantation Proceedings 33(5): 2749-51.

Bittner, H. B., Chen, E. P., Craig, D. and Van Trigt, P. (1996a). Preload-recruitable stroke work relationships and diastolic dysfunction in the brain-dead organ donor. Circulation 94(9 Suppl): II320-5.

Bittner, H. B., Kendall, S. W., Campbell, K. A., Montine, T. J. and Van Trigt, P. (1995a). A valid experimental brain death organ donor model. Journal of Heart and Lung Transplantation 14(2): 308-17.

Bittner, H. B., Kendall, S. W., Chen, E. P. and Van Trigt, P. (1995b). Endocrine changes and metabolic responses in a validated canine brain death model. Journal of Critical Care 10(2): 56-63.

Bittner, H. B., Kendall, S. W., Chen, E. P. and Van Trigt, P. (1996b). The combined effects of brain death and cardiac graft preservation on cardiopulmonary hemodynamics and function before and after subsequent heart transplantation. Journal of Heart and Lung Transplantation 15(8): 764-77.

Blanche, C., Blanche, D. A., Kearney, B., Sandhu, M., Czer, L. S., Kamlot, A., Hickey, A. and Trento, A. (2001). Heart transplantation in patients seventy years of age and older: A comparative analysis of outcome. Journal of Thoracic and Cardiovascular Surgery 121(3): 532-41.

Blasi-Ibanez, A., Hirose, R., Feiner, J., Freise, C., Stock, P. G., Roberts, J. P., Niemann, C. U., Blasi-Ibanez, A., Hirose, R., Feiner, J., Freise, C., Stock, P. G., Roberts, J. P. and

- 262 - Niemann, C. U. (2009). Predictors associated with terminal renal function in deceased organ donors in the intensive care unit. Anesthesiology 110(2): 333-41.

Boggi, U., Coletti, L., Vistoli, F., Del Chiaro, M., Signori, S., Croce, C., Bartolo, T. V., Pietrabissa, A., Marchetti, P., Capocasale, E., Dalla Valle, R., Mazzoni, M. P. and Mosca, F. (2004). Pancreas preservation with University of Wisconsin and Celsior solutions. Transplantation Proceedings 36(3): 563-5.

Boggi, U., Signori, S., Vistoli, F., Del Chiaro, M., Pietrabissa, A., Croce, C., Barsotti, M., Bartolo, T. V., Amorese, G., Capocasale, E., Della Valle, R., Mazzoni, M. P., Mosca, F. and Bartolo, T. V. (2005). University of Wisconsin solution versus Celsior solution in clinical pancreas transplantation. Transplantation Proceedings 37(2): 1262- 4.

Bolling, S. F., Smolens, I. A. and Pagani, F. D. (2001). Surgical alternatives for heart failure. Journal of Heart and Lung Transplantation 20(7): 729-33.

Boyce, S. W., Bartels, C., Bolli, R., Chaitman, B., Chen, J. C., Chi, E., Jessel, A., Kereiakes, D., Knight, J., Thulin, L., Theroux, P. and Guard During Ischemia Against Necrosis Study Investigators. (2003). Impact of sodium-hydrogen exchange inhibition by cariporide on death or myocardial infarction in high-risk CABG surgery patients: results of the CABG surgery cohort of the GUARDIAN study. Journal of Thoracic and Cardiovascular Surgery 126(2): 420-7.

Breda, M. A., Drinkwater, D. C., Laks, H., Bhuta, S., Corno, A. F., Davtyan, H. G. and Chang, P. (1989). Prevention of reperfusion injury in the neonatal heart with leukocyte- depleted blood. Journal of Thoracic and Cardiovascular Surgery 97(5): 654-65.

Brunicardi, F. C., Dyen, Y., Brostrom, L., Kleinman, R., Colonna, J., Gelabert, H. and Gingerich, R. (2000). The circulating hormonal milieu of the endocrine pancreas in healthy individuals, organ donors, and the isolated perfused human pancreas. Pancreas 21(2): 203-11.

Budrikis, A., Bolys, R., Liao, Q., Ingemansson, R., Sjoberg, T. and Steen, S. (1998). Function of adult pig hearts after 2 and 12 hours of cold cardioplegic preservation. Annals of Thoracic Surgery 66(1): 73-8.

- 263 - Buerke, M., Rupprecht, H. J., vom Dahl, J., Terres, W., Seyfarth, M., Schultheiss, H. P., Richardt, G., Sheehan, F. H. and Drexler, H. (1999). Sodium-hydrogen exchange inhibition: novel strategy to prevent myocardial injury following ischemia and reperfusion. American Journal of Cardiology 83(10A): 19G-22G.

Busuttil, R. W., George, W. J. and Hewitt, R. L. (1975). Protective effect of methylprednisolone on the heart during ischemic arrest. Journal of Thoracic and Cardiovascular Surgery 70(6): 955-65.

Bynum, T. E. and Fara, J. W. (1980). Hepatic artery response to vasopressin. American Journal of Physiology 239(5): G378-81.

Carrel, A. and Guthrie, C. C. (1905). The transplantation of veins and organs. American Journal of Medicine 10: 1101-1102.

Carrier, O., Jr., Wedell, E. K. and Barron, K. W. (1978). Specific alpha-adrenergic receptor desensitization in vascular smooth muscle. Blood Vessels 15(4): 247-58.

Cass, M. H. and Brock, R. (1959). Heart excision and replacement. Guy's Hospital Report 108: 285-290.

Castella, M., Buckberg, G. D., Tan, Z. and Ignarro, L. J. (2002). Myocyte and endothelial effects of preconditioning the jeopardized heart by inhibiting Na/H exchange. Journal of Thoracic and Cardiovascular Surgery 124(6): 1113-21.

Cavallari, A., Cillo, U., Nardo, B., Filipponi, F., Gringeri, E., Montalti, R., Vistoli, F., D'Amico, F., Faenza, A., Mosca, F., Vitale, A. and D'Amico, D. (2003). A multicenter pilot prospective study comparing Celsior and University of Wisconsin preserving solutions for use in liver transplantation. Liver Transplantation 9(8): 814-21.

Caves, P. K., Billingham, M. E., Stinson, E. B. and Shumway, N. E. (1974). Serial transvenous biopsy of the transplanted human heart. Improved management of acute rejection episodes. Lancet 1: 821-826.

Chambers, D. J. and Hearse, D. J. (1999). Developments in cardioprotection: "polarized" arrest as an alternative to "depolarized" arrest. Annals of Thoracic Surgery 68(5): 1960-6.

- 264 - Chang, V. P. (1984). Reintroduction of cardiac transplantation in Australia. The Medical Journal of Australia 140(12): 692-3.

Changani, K. K., Fuller, B. J., Bell, J. D., Taylor-Robinson, S. D., Moore, D. P. and Davidson, B. R. (1999). Improved preservation solutions for organ storage: a dynamic study of hepatic metabolism. Transplantation 68(3): 345-55.

Chen, E. P., Bittner, H. B., Kendall, S. W. and Van Trigt, P. (1996). Hormonal and hemodynamic changes in a validated animal model of brain death. Critical Care Medicine 24(8): 1352-9.

Chen, J. M., Cullinane, S., Spanier, T. B., Artrip, J. H., John, R., Edwards, N. M., Oz, M. C. and Landry, D. W. (1999). Vasopressin deficiency and pressor hypersensitivity in hemodynamically unstable organ donors. Circulation 100(19 Suppl): II244-6.

Chen, R. H. (1996). The scientific basis for hypocalcemic cardioplegia and reperfusion in cardiac surgery. Annals of Thoracic Surgery 62(3): 910-4.

Chiari, P., Hadour, G., Michel, P., Piriou, V., Rodriguez, C., Budat, C., Ovize, M., Jegaden, O., Lehot, J. J. and Ferrera, R. (2000). Biphasic response after brain death induction: prominent part of catecholamines release in this phenomenon. Journal of Heart and Lung Transplantation. 19(7): 675-82.

Christie, J. D., Bavaria, J. E., Palevsky, H. I., Litzky, L., Blumenthal, N. P., Kaiser, L. R. and Kotloff, R. M. (1998). Primary graft failure following lung transplantation. Chest 114(1): 51-60.

Christie, J. D., Edwards, L. B., Aurora, P., Dobbels, F., Kirk, R., Rahmel, A. O., Taylor, D. O., Kucheryavaya, A. Y. and Hertz, M. I. (2008). Registry of the International Society for Heart and Lung Transplantation: twenty-fifth official adult lung and heart/lung transplantation report--2008. Journal of Heart and Lung Transplantation 27(9): 957-69.

Clayton, H. A., Swift, S. M., Turner, J. M., James, R. F. and Bell, P. R. (2000). Non- heart-beating organ donors: a potential source of islets for transplantation? Transplantation 69(10): 2094-8.

- 265 - Conte, J. V., Fleischer, K. J. and Baumgartner, W. A. (2002). Evaluation and management of heart donors. Heart and Lung Transplantation. W. A. Baumgartner, Kasper, E., Reitz, B. and Theodore, J. Philadelphia, W.B. Saunders Company: 155-160.

Cooper, D. K. and Novitzky, D. (1988a). Thyroid function in human cardiac allograft donors. Transplantation 45(3): 671-2.

Cooper, D. K., Novitzky, D. and Wicomb, W. N. (1988b). Hormonal therapy in the brain-dead experimental animal. Transplantation Proceedings 20(5 Suppl 7): 51-4.

Cooper, D. K., Novitzky, D. and Wicomb, W. N. (1989). The pathophysiological effects of brain death on potential donor organs, with particular reference to the heart. Annals of the Royal College of Surgeons of England 71(4): 261-6.

Copeland, J. G. (2001). Advanced medical therapy does not render heart transplantation obsolete for ambulatory end-stage heart failure patients: a debate. Journal of Heart and Lung Transplantation 20(7): 725-8.

Corris, P. A. (2002). Non-heart beating lung donation: aspects for the future. Thorax 57(2).

Cropper, J. R., Hicks, M., Ryan, J. B. and Macdonald, P. S. (2003a). Cardioprotection by cariporide after prolonged hypothermic storage of the isolated working rat heart. Journal of Heart and Lung Transplantation 22(8): 929-36.

Cropper, J. R., Hicks, M., Ryan, J. B. and Macdonald, P. S. (2003b). Enhanced cardioprotection of the rat heart during hypothermic storage with combined Na+ - H+ exchange inhibition and ATP-dependent potassium channel activation. Journal of Heart and Lung Transplantation 22(11): 1245-53.

Cruickshank, J. M., Neil-Dwyer, G., Degaute, J. P., Hayes, Y., Kuurne, T., Kytta, J., Vincent, J. L., Carruthers, M. E. and Patel, S. (1987). Reduction of stress/catecholamine-induced cardiac necrosis by beta 1-selective blockade. Lancet 2(8559): 585-9.

Csont, T. and Ferdinandy, P. (2005). Cardioprotective effects of glyceryl trinitrate: beyond vascular nitrate tolerance. Pharmacology & Therapeutics 105(1): 57-68.

- 266 - Csont, T., Szilvassy, Z., Fulop, F., Nedeianu, S., Pali, T., Tosaki, A., Dux, L. and Ferdinandy, P. (1999). Direct myocardial anti-ischaemic effect of GTN in both nitrate- tolerant and nontolerant rats: a cyclic GMP-independent activation of KATP. British Journal of Pharmacology 128(7): 1427-34.

Cushing, H. (1902). Some experimental and clinical observation concerning states of increased intracranial hypertension. American Journal of the Medical Sciences 124: 375-400.

D'Amico, T. A., Meyers, C. H., Koutlas, T. C., Peterseim, D. S., Sabiston, D. C., Jr., Van Trigt, P. and Schwinn, D. A. (1995). Desensitization of myocardial beta-adrenergic receptors and deterioration of left ventricular function after brain death. Journal of Thoracic and Cardiovascular Surgery 110(3): 746-51.

D'Armini A, M., Grande, A. M., Rinaldi, M., Goggi, C. and Vigano, M. (2001). Prospective randomized clinical study of Celsior vs University of Wisconsin in double lung transplant. Journal of Heart and Lung Transplantation 20(2): 183.

Del Rizzo, D. F., Menkis, A. H., Pflugfelder, P. W., Novick, R. J., McKenzie, F. N., Boyd, W. G. and Kostuk, W. J. (1999). The role of donor age and ischemic time on survival following orthotopic heart transplantation. Journal of Heart and Lung Transplantation 18(4): 310-9.

Demikhov, V. P. (1962). Experimental Transplantation of Vital Organs. New York: Consultants' Bureau.

Demmy, T. L., Biddle, J. S., Bennett, L. E., Walls, J. T., Schmaltz, R. A. and Curtis, J. J. (1997). Organ preservation solutions in heart transplantation--patterns of usage and related survival. Transplantation 63(2): 262-9.

Depret, J., Teboul, J. L., Benoit, G., Mercat, A. and Richard, C. (1995). Global energetic failure in brain-dead patients. Transplantation 60(9): 966-71.

Di Giantomasso, D., May, C. N., Bellomo, R., Di Giantomasso, D., May, C. N. and Bellomo, R. (2002). Norepinephrine and vital organ blood flow. Intensive Care Medicine 28(12): 1804-9.

- 267 - DiBardino, D. J. (1999). The history and development of cardiac transplantation. Texas Heart Institute Journal 26(3): 198-205.

Dimopoulou, I., Tsagarakis, S., Anthi, A., Milou, E., Ilias, I., Stavrakaki, K., Charalambidis, C., Tzanela, M., Orfanos, S., Mandragos, K., Thalassinos, N. and Roussos, C. (2003). High prevalence of decreased cortisol reserve in brain-dead potential organ donors. Critical Care Medicine 31(4): 1113-7.

Dobson, G. P. and Jones, M. W. (2004). Adenosine and lidocaine: a new concept in nondepolarizing surgical myocardial arrest, protection, and preservation. Journal of Thoracic and Cardiovascular Surgery. 127(3): 794-805.

Donnelly, A. J. and Djuric, M. (1991). Cardioplegia solutions. American Journal of Hospital Pharmacy 48(11): 2444-60.

Drinkwater, D. C., Rudis, E., Laks, H., Ziv, E., Marino, J., Stein, D., Ardehali, A., Aharon, A., Moriguchi, J. and Kobashigawa, J. (1995). University of Wisconsin solution versus Stanford cardioplegic solution and the development of cardiac allograft vasculopathy. Journal of Heart and Lung Transplantation. 14(5): 891-6.

Drummond, C. S., Wise, R., McLaughlin, J. S. and Foster, A. H. (1997). Critical components responsible for superior rat heart preservation with Celsior solution and comparison with Belzer-University of Wisconsin solution (abstract). Journal of Heart and Lung Transplantation 16: 79.

Du, Z. Y., Hicks, M., Jansz, P., Rainer, S., Spratt, P. and Macdonald, P. (1998). The nitric oxide donor, diethylamine NONOate, enhances preservation of the donor rat heart. Journal of Heart and Lung Transplantation 17(11): 1113-20.

Duranski, M. R., Greer, J. J., Dejam, A., Jaganmohan, S., Hogg, N., Langston, W., Patel, R. P., Yet, S. F., Wang, X., Kevil, C. G., Gladwin, M. T. and Lefer, D. J. (2005). Cytoprotective effects of nitrite during in vivo ischemia-reperfusion of the heart and liver. Journal of Clinical Investigation 115(5): 1232-40.

Edmonds, H. L., Jr., Cannon, H. C., Jr., Garretson, H. D. and Dahlquist, G. (1986). Effects of aerosolized methylprednisolone on experimental neurogenic pulmonary injury. Neurosurgery 19(1): 36-40.

- 268 - Ellett, J. D., Evans, Z. P., Fiorini, J. H., Fiorini, R. N., Haines, J. K., Schmidt, M. G. and Chavin, K. D. (2008). The use of the Papworth cocktail is detrimental to steatotic livers after ischemia-reperfusion injury. Transplantation. 86(2): 286-92.

Excell, L., Hee, K. and Russ, G., Eds. (2008). ANZOD Registry Report 2008. Adelaide, Australia and New Zealand Organ Donation Registry.

Faenza, A., Catena, F., Nardo, B., Montalti, R., Capocasale, E., Busi, N., Boggi, U., Vistoli, F., Di Naro, A., Albertazzi, A., Mosca, F. and Cavallari, A. (2001). Kidney preservation with university of Wisconsin and Celsior solution: a prospective multicenter randomized study. Transplantation 72(7): 1274-7.

Fedalen, P. A., Piacentino, V., 3rd, Jeevanandam, V., Fisher, C., Greene, J., Margulies, K. B., Houser, S. R., Furukawa, S., Singhal, A. K. and Goldman, B. I. (2003). Pharmacologic pre-conditioning and controlled reperfusion prevent ischemia- reperfusion injury after 30 minutes of hypoxia/ischemia in porcine hearts. Journal of Heart and Lung Transplantation. 22(11): 1234-44.

Feneley, M. P., Skelton, T. N., Kisslo, K. B., Davis, J. W., Bashore, T. M. and Rankin, J. S. (1992). Comparison of preload recruitable stroke work, end-systolic pressure- volume and dP/dtmax-end-diastolic volume relations as indexes of left ventricular contractile performance in patients undergoing routine cardiac catheterization. Journal of the American College of Cardiology 19(7): 1522-30.

Ferdinandy, P., Szilvassy, Z., Csont, T., Csonka, C., Nagy, E., Koltai, M. and Dux, L. (1995). Nitroglycerin-induced direct protection of the ischaemic myocardium in isolated working hearts of rats with vascular tolerance to nitroglycerin. British Journal of Pharmacology 115(7): 1129-31.

Finfer, S., Bohn, D., Colpitts, D., Cox, P., Fleming, F. and Barker, G. (1996). Intensive care management of paediatric organ donors and its effect on post-transplant organ function. Intensive Care Medicine 22(12): 1424-32.

Finkelstein, I., Toledo-Pereyra, L. H. and Castellanos, J. (1987). Physiologic and hormonal changes in experimentally induced brain dead dogs. Transplantation Proceedings 19(5): 4156-8.

- 269 - Fiser, D. H., Jimenez, J. F., Wrape, V. and Woody, R. (1987). Diabetes insipidus in children with brain death. Critical Care Medicine 15(6): 551-3.

Fisher, A. J., Dark, J. H. and Corris, P. A. (1998). Improving donor lung evaluation: a new approach to increase organ supply for lung transplantation. Thorax. 53(10): 818-20.

Fisher, A. J., Donnelly, S. C., Hirani, N., Haslett, C., Strieter, R. M., Dark, J. H. and Corris, P. A. (2001). Elevated levels of interleukin-8 in donor lungs is associated with early graft failure after lung transplantation. American Journal of Respiratory and Critical Care Medicine 163(1): 259-65.

Fleischer, K. J. and Baumgartner, W. A. (2002). Heart preservation. Heart and Lung Transplantation. W. A. Baumgartner, Reitz, B., Kasper, E. and Theodore, J. Philadelphia, W.B. Saunders Company: 135-141.

Folkesson, H. G., Norlin, A., Wang, Y., Abedinpou, r. P. and Matthay, M. A. (2000). Dexamethasone and thyroid hormone pretreatment upregulate alveolar epithelial fluid clearance in adult rats. Journal of Applied Physiology 88(2): 416-24.

Follette, D. M., Rudich, S. M. and Babcock, W. D. (1998). Improved oxygenation and increased lung donor recovery with high-dose steroid administration after brain death. Journal of Heart and Lung Transplantation. 17(4): 423-9.

Fukuhiro, Y., Wowk, M., Ou, R., Rosenfeldt, F. and Pepe, S. (2000). Cardioplegic strategies for calcium control: low Ca(2+), high Mg(2+), citrate, or Na(+)/H(+) exchange inhibitor HOE-642. Circulation 102(19 Suppl 3): III319-25.

Fung, J. J. (2000). Use of non-heart-beating donors. Transplantation Proceedings 32: 1510-1.

Gamez, P., Cordoba, M., Ussetti, P., Carreno, M. C., Alfageme, F., Madrigal, L., Nunez, J. R., Calatayud, J., Ramos, M., Salas, C., Varela, A. and Lung Transplant Group of the Puerta de Hierro, H. (2005). Lung transplantation from out-of-hospital non-heart-beating lung donors. one-year experience and results. Journal of Heart and Lung Transplantation 24(8): 1098-102.

- 270 - Gao, L., Hicks, M. and MacDonald, P. S. (2005). Improved preservation of the rat heart with celsior solution supplemented with cariporide plus glyceryl trinitrate. American Journal of Transplantation 5(8): 1820-6.

Garcia-Fages, L. C., Antolin, M., Cabrer, C., Talbot, R., Alcaraz, A., Lozano, F. and Manyalich, M. (1991). Effects of substitutive triiodothyronine therapy on intracellular nucleotide levels in donor organs. Transplantation Proceedings 23(5): 2495-6.

Garlicki, M. (2003). May preservation solution affect the incidence of graft vasculopathy in transplanted heart? Annals of Transplantation 8(1): 19-24.

Garlicki, M., Kolcz, J., Rudzinski, P., Kapelak, B., Sadowski, J., Wojcik, S., Pietrzyk, E., Frasik, W., Drukala, J. and Dziatkowiak, A. (1999). Myocardial protection for transplantation. Transplantation Proceedings 31(5): 2079-83.

Gasser, M., Waaga, A. M., Laskowski, I. A. and Tilney, N. L. (2000). The influence of donor brain death on short and long-term outcome of solid organ allografts. Annals of Transplantation. 5(4): 61-7.

Gasser, M., Waaga, A. M., Laskowski, I. A. and Tilney, N. L. (2001). Organ transplantation form brain-dead donors:its impact on short- and long-term outcome revisited. Transplantation Reviews 15(1): 1-10.

Gerstenkorn, C., Oliveira, D., MacPhee, I. and Chang, R. (2000). Non-heart-beating donors for renal transplantation. Lancet 356: 1854.

Gibson, H., Seldon, W. A. and Windsor, H. M. (1969). A heart transplantation: 1. Clinical features. The Medical Journal of Australia 1(13): 663-665.

Gifford, R. R., Weaver, A. S., Burg, J. E., Romano, P. J., Demers, L. M. and Pennock, J. L. (1986). Thyroid hormone levels in heart and kidney cadaver donors. Journal of Heart Transplantation 5(3): 249-53.

Glantz, S. A. and Slinker, B. K. (2001). Primer of applied regression and analysis of variance. New York: McGraw-Hill.

- 271 - Glower, D. D., Spratt, J. A., Kabas, J. S., Davis, J. W. and Rankin, J. S. (1988). Quantification of regional myocardial dysfunction after acute ischemic injury. American Journal of Physiology. 255(1 Pt 2): H85-93.

Glower, D. D., Spratt, J. A., Snow, N. D., Kabas, J. S., Davis, J. W., Olsen, C. O., Tyson, G. S., Sabiston, D. C., Jr. and Rankin, J. S. (1985). Linearity of the Frank- Starling relationship in the intact heart: the concept of preload recruitable stroke work. Circulation. 71(5): 994-1009.

Goarin, J. P., Cohen, S., Riou, B., Jacquens, Y., Guesde, R., Le Bret, F., Aurengo, A. and Coriat, P. (1996). The effects of triiodothyronine on hemodynamic status and cardiac function in potential heart donors. Anesthesia & Analgesia 83(1): 41-7.

Goldsmith, S. R. (1987). Vasopressin as vasopressor. American Journal of Medicine 82(6): 1213-9.

Golling, M., Mehrabi, A., Blum, K., Jahnke, C., Kellner, H., Bud, O., Hashemi, B., Breitkreutz, R., Becker-Brandenburg, K., Schemmer, P., Gebhard, M. M., Herfarth, C. and Kraus, T. (2003). Effects of hemodynamic instability on brain death-induced prepreservation liver damage. Transplantation 75(8): 1154-9.

Gores, P. F., Gillingham, K. J., Dunn, D. L., Moudry-Munns, K. C., Najarian, J. S. and Sutherland, D. E. (1992). Donor hyperglycemia as a minor risk factor and immunologic variables as major risk factors for pancreas allograft loss in a multivariate analysis of a single institution's experience. Annals of Surgery 215(3): 217-30.

Gores, P. F., Viste, A., Hesse, U. J., Moudry-Munns, K. C., Dunn, D. L. and Sutherland, D. E. (1990). The influence of donor hyperglycemia and other factors on long-term pancreatic allograft survival. Transplantation Proceedings 22(2): 437-8.

Gramm, H. J., Meinhold, H., Bickel, U., Zimmermann, J., von Hammerstein, B., Keller, F., Dennhardt, R. and Voigt, K. (1992a). Acute endocrine failure after brain death? Transplantation 54(5): 851-7.

Gramm, H. J., Zimmermann, J., Meinhold, H., Dennhardt, R. and Voigt, K. (1992b). Hemodynamic responses to noxious stimuli in brain-dead organ donors. Intensive Care Medicine 18(8): 493-5.

- 272 - Griepp, R. B. (1979). A decade of human heart transplantation. Transplantation Proceedings 11: 285-292.

Griepp, R. B. and Ergin, M. A. (1984). The history of experimental heart transplantation. Heart Transplantation 3(2): 145-51.

Guesde, R., Barrou, B., Leblanc, I., Ourahma, S., Goarin, J. P., Coriat, P. and Riou, B. (1998). Administration of desmopressin in brain-dead donors and renal function in kidney recipients. Lancet 352(9135): 1178-81.

Hannon, J. P., Bossone, C. A. and Wade, C. E. (1990). Normal physiological values for conscious pigs used in biomedical research. Laboratory Animal Science 40(3): 293-298.

Hardy, J. D., Chavez, C. M., Kurrus, F. D., Neely, W. A., Eraslan, S., Turner, M. D., Fabian, L. W. and Labecki, T. D. (1964). Heart transplantation in man. Developmental studies and report of a case. JAMA 188: 1132-1140.

Harrison, D. G. and Bates, J. N. (1993). The nitrovasodilators. New ideas about old drugs. Circulation 87(5): 1461-7.

Hassanein, W. H., Zellos, L., Tyrrell, T. A., Healey, N. A., Crittenden, M. D., Birjiniuk, V. and Khuri, S. F. (1998). Continuous perfusion of donor hearts in the beating state extends preservation time and improves recovery of function. Journal of Thoracic and Cardiovascular Surgery 116(5): 821-30.

Hausenloy, D. J. and Yellon, D. M. (2004). New directions for protecting the heart against ischaemia-reperfusion injury: targeting the Reperfusion Injury Salvage Kinase (RISK)-pathway. Cardiovascular Research 61(3): 448-60.

Hesse, U. J., Gores, P. F. and Sutherland, D. E. (1989). Serum amylase and plasma glucose levels in pancreas cadaver donors: correlation with functional status of the pancreatic graft. Transplantation Proceedings 21(1 Pt 3): 2765-6.

Hicks, M., Hing, A., Gao, L., Ryan, J. and Macdonald, P. S. (2006). Organ Preservation. Transplantation Immunology: Methods and Protocols. P. Hornick and Rose, M. Totowa, Humana Press Inc. 333: 331-373.

- 273 - Hing, A. J., Hicks, M., Garlick, S. R., Gao, L., Kesteven, S. H., Faddy, S. C., Wilson, M. K., Feneley, M. P. and Macdonald, P. S. (2007). The effects of hormone resuscitation on cardiac function and hemodynamics in a porcine brain-dead organ donor model. American Journal of Transplantation 7(4): 809-17.

Hing, A. J., Watson, A., Hicks, M., Gao, L., Faddy, S. C., McMahon, A. C., Kesteven, S. H., Wilson, M. K., Jansz, P., Feneley, M. P. and Macdonald, P. S. (2009). Combining cariporide with glyceryl trinitrate optimises cardiac preservation during porcine heart transplantation. American Journal of Transplantation 9: 2048-56.

Hirt, S. W., Wahlers, T., Jurmann, M., Fieguth, H. G., Dammenhayn, L. and Haverich, A. (1992). Improvement of currently used methods for lung preservation with prostacyclin and University of Wisconsin solution. Journal of Heart and Lung Transplantation 11(4 Pt 1): 656-64.

Holmes, C. L., Landry, D. W. and Granton, J. T. (2003). Science review: Vasopressin and the cardiovascular system part 1--receptor physiology. Critical Care 7(6): 427-34.

Holt, E., Sjaastad, I., Lunde, P. K., Christensen, G. and Sejersted, O. M. (1999). Thyroid hormone control of contraction and the Ca(2+)-ATPase/phospholamban complex in adult rat ventricular myocytes. Journal of Molecular and Cellular Cardiology 31(3): 645-56.

Howlett, T. A., Keogh, A. M., Perry, L., Touzel, R. and Rees, L. H. (1989). Anterior and posterior pituitary function in brain-stem-dead donors. A possible role for hormonal replacement therapy. Transplantation 47(5): 828-34.

Humbert, M., Sitbon, O. and Simonneau, G. (2004). Treatment of pulmonary arterial hypertension. The New England Journal of Medicine 351(14): 1425-36.

Hurtado, C. and Pierce, G. N. (2001). Sodium-hydrogen exchange inhibition: pre-versus post-ischemic treatment. Basic Research in Cardiology 96(4): 312-7.

Ignarro, L. J., Napoli, C. and Loscalzo, J. (2002). Nitric oxide donors and cardiovascular agents modulating the bioactivity of nitric oxide: an overview. Circulation Research 90(1): 21-8.

- 274 - Jahania, M. S., Sanchez, J. A., Narayan, P., Lasley, R. D. and Mentzer, R. M., Jr. (1999). Heart preservation for transplantation: principles and strategies. Annals of Thoracic Surgery 68(5): 1983-7.

Jamall, I. N., Pagel, P. S., Hettrick, D. A., Lowe, D., Kersten, J. R., Tessmer, J. P. and Warltier, D. C. (1997). Positive inotropic and lusitropic effects of triiodothyronine in conscious dogs with pacing-induced cardiomyopathy. Anesthesiology 87(1): 102-9.

Jassem, W., Koo, D. D., Cerundolo, L., Rela, M., Heaton, N. D. and Fuggle, S. V. (2003). Cadaveric versus living-donor livers: differences in inflammatory markers after transplantation. Transplantation 76(11): 1599-603.

Jeevanandam, V. (1997). Triiodothyronine: spectrum of use in heart transplantation. Thyroid 7(1): 139-45.

Jeevanandam, V., Todd, B., Hellman, S., Eldridge, C., McClurken, J. and Addonizio, V. P. (1993). Use of triiodothyronine replacement therapy to reverse donor myocardial dysfunction: creating a larger donor pool. Transplantation Proceedings 25(6): 3305-6.

Jeevanandam, V., Todd, B., Regillo, T., Hellman, S., Eldridge, C. and McClurken, J. (1994). Reversal of donor myocardial dysfunction by triiodothyronine replacement therapy. Journal of Heart and Lung Transplantation 13(4): 681-7; discussion 685-7.

Jordan, J. E., Zhao, Z. Q. and Vinten-Johansen, J. (1999). The role of neutrophils in myocardial ischemia-reperfusion injury. Cardiovascular Research 43(4): 860-78.

Julious, S. A. (2004). Sample sizes for clinical trials with normal data. Statistics In Medicine 23(12): 1921-86.

Kantrowicz, A., Huller, J. D., Joos, H., Cerruti, M. M. and Carstensen, H. E. (1968). Transplantation of the heart in an infant and an adult. American Journal of Cardiology 22: 782-790.

Karam, G. (2003). [Safety of the use of Celsior in kidney-pancreas transplantation]. Prog Urol 13(1): 46-9.

- 275 - Karmazyn, M., Gan, X. T., Humphreys, R. A., Yoshida, H. and Kusumoto, K. (1999). The myocardial Na(+)-H(+) exchange: structure, regulation, and its role in heart disease. Circulation Research 85(9): 777-86.

Karunanithi, M. K. and Feneley, M. P. (2000). Single-beat determination of preload recruitable stroke work relationship: derivation and evaluation in conscious dogs. Journal of the American College of Cardiology. 35(2): 502-13.

Karunanithi, M. K., Michniewicz, J., Copeland, S. E. and Feneley, M. P. (1992). Right ventricular preload recruitable stroke work, end-systolic pressure-volume, and dP/dtmax-end-diastolic volume relations compared as indexes of right ventricular contractile performance in conscious dogs. Circulation Research. 70(6): 1169-79.

Kasper, E. K. and Achuff, S. C. (2002). Clinical evaluation of potential heart transplant recipients. Heart and Lung Transplantation. W. A. Baumgartner, Reitz, B., Kasper, E. and Theodore, J. Philadelphia, W.B. Saunders Company: 57-63.

Kawashima, M., Bando, T., Nakamura, T., Isowa, N., Liu, M., Toyokuni, S., Hitomi, S. and Wada, H. (2000). Cytoprotective effects of nitroglycerin in ischemia-reperfusion- induced lung injury. American Journal of Respiratory and Critical Care Medicine 161(3 Pt 1): 935-43.

Kaye, M. P. (1993). The registry of the International Society for Heart Transplantation: tenth official report. Journal of Heart and Lung Transplantation 12: 541-548.

Keck, T., Banafsche, R., Werner, J., Gebhard, M. M., Herfarth, C. and Klar, E. (2001). Desmopressin impairs microcirculation in donor pancreas and early graft function after experimental pancreas transplantation. Transplantation. 72(2): 202-9.

Keogh, A. and Pettersson, R., Eds. (2008). Australia and New Zealand Cardiothoracic Organ Transplant Registry 2007 Report. Sydney, St. Vincent's Hospital.

Keogh, A. M., Howlett, T. A., Perry, L. and Rees, L. H. (1988). Pituitary function in brain-stem dead organ donors: a prospective survey. Transplantation Proceedings 20(5): 729-30.

- 276 - Kevelaitis, E., Oubenaissa, A., Mouas, C., Peynet, J. and Menasche, P. (2001). Ischemic preconditioning with opening of mitochondrial adenosine triphosphate-sensitive potassium channels or Na/H exchange inhibition: which is the best protective strategy for heart transplants? Journal of Thoracic and Cardiovascular Surgery 121(1): 155-62.

Kim, Y. I., Herijgers, P., Laycock, S. K., Van Lommel, A., Verbeken, E. and Flameng, W. J. (1998a). Na+/H+ exchange inhibition improves long-term myocardial preservation. Annals of Thoracic Surgery 66(2): 436-42.

Kim, Y. I., Herijgers, P., Van Lommel, A., Verbeken, E. and Flameng, W. (1998b). Na+/H+ exchange inhibition improves post-transplant myocardial compliance in 4-hour stored donor hearts. Cardiovascular Surgery 6(1): 67-75.

Kinoshita, Y., Yahata, K., Yoshioka, T., Onishi, S. and Sugimoto, T. (1990). Long-term renal preservation after brain death maintained with vasopressin and epinephrine. Transplant International 3(1): 15-8.

Kirk, R., Edwards, L. B., Aurora, P., Taylor, D. O., Christie, J., Dobbels, F., Kucheryavaya, A. Y., Rahmel, A. O. and Hertz, M. I. (2008). Registry of the International Society for Heart and Lung Transplantation: eleventh official pediatric heart transplantation report--2008. Journal of Heart and Lung Transplantation 27(9): 970-7.

Kirklin, J. K., Naftel, D. C., Bourge, R. C., McGiffin, D. C., Hill, J. A., Rodeheffer, R. J., Jaski, B. E., Hauptman, P. J., Weston, M. and White-Williams, C. (2003). Evolving trends in risk profiles and causes of death after heart transplantation: a ten-year multi- institutional study. Journal of Thoracic & Cardiovascular Surgery 125(4): 881-90.

Kirklin, J. K., Pambukian, S. V., McGiffin, D. C., Benza, R. L., Kirklin, J. K., Pambukian, S. V., McGiffin, D. C. and Benza, R. L. (2004). Current outcomes following heart transplantation. Seminars in Thoracic & Cardiovascular Surgery 16(4): 395-403.

Kirklin, J. K., Young, J. B. and McGiffin, D. C. (2002). Heart Transplantation. New York: Churchill Livingstone.

- 277 - Kishida, A., Kurumi, Y. and Kodama, M. (1997). Efficacy of prostaglandin I2 analog on liver grafts subjected to 30 minutes of warm ischemia. Surgery Today 27(11): 1056- 60.

Kitai, T., Tanaka, A., Terasaki, M., Okamoto, R., Ozawa, K., Morikawa, S. and Inubushi, T. (1993). Energy metabolism of the heart and the liver in brain-dead dogs as assessed by 31P NMR spectroscopy. Journal of Surgical Research 55(6): 599-606.

Klein, I. and Ojamaa, K. (2001). Thyroid hormone and the cardiovascular system. New England Journal of Medicine 344(7): 501-9.

Koo, D. D., Welsh, K. I., McLaren, A. J., Roake, J. A., Morris, P. J. and Fuggle, S. V. (1999). Cadaver versus living donor kidneys: impact of donor factors on antigen induction before transplantation. Kidney International 56(4): 1551-9.

Kristo, G., Yoshimura, Y., Ferraris, S. P., Jahania, S. A., Mentzer, R. M., Jr. and Lasley, R. D. (2004). The preischemic combination of the sodium-hydrogen exchanger inhibitor cariporide and the adenosine agonist AMP579 acts additively to reduce porcine myocardial infarct size. Journal of the American College of Surgeons 199(4): 586-94.

Kutsogiannis, D. J., Pagliarello, G., Doig, C., Ross, H., Shemie, S. D., Kutsogiannis, D. J., Pagliarello, G., Doig, C., Ross, H. and Shemie, S. D. (2006). Medical management to optimize donor organ potential: review of the literature. Canadian Journal of Anaesthesia 53(8): 820-30.

Kwan, J. C., Gao, L., Macdonald, P. S. and Hicks, M. (2007). Decrease in apoptosis associated with cardioprotective strategies in a model of donor heart preservation (abstract). Journal of Molecular and Cellular Cardiology 42(6): S81.

Kwan, J. C., Gao, L., Macdonald, P. S. and Hicks, M. (2008). Protective effect of glyceryl trinitrate and cariporide in a model of donor heart preservation: Activation of Erk 1/2 pathway (abstract). Transplantation 86(2S): 1150.

Lagiewska, B., Pacholczyk, M., Szostek, M., Walaszewski, J. and Rowinski, W. (1996). Hemodynamic and metabolic disturbances observed in brain-dead organ donors. Transplantation Proceedings 28(1): 165-6.

- 278 - Lansman, S. L., Ergin, M. A. and Griepp, R. B. (1995). The history of heart and heart- lung transplantation. Thoracic Transplantation. S. J. Shumway and Shumway, N. E. Cambridge, Blackwell Science Inc.: 3-14.

Liard, J. F., Deriaz, O., Schelling, P. and Thibonnier, M. (1982). Cardiac output distribution during vasopressin infusion or dehydration in conscious dogs. American Journal of Physiology 243(5).

Light, J. A., Sasaki, T. M., Ayuino, A. Q., Barhyte, D. Y. and Gage, F. (2000). Excellent long-term graft survival with kidneys from uncontrolled non-heart beating donor. Transplantation Proceedings 32: 186-7.

Lima, B., Rajagopal, K., Petersen, R. P., Shah, A. S., Soule, B., Felker, G. M., Rogers, J. G., Lodge, A. J. and Milano, C. A. (2006). Marginal cardiac allografts do not have increased primary graft dysfunction in alternate list transplantation. Circulation 114(1 Suppl): I27-32.

Lin, H., Okamoto, R., Yamamoto, Y., Maki, A., Ueda, J., Tokunaga, Y., Yamamoto, S., Mori, K., Tanaka, K. and Yamaoka, Y. (1989a). Hepatic tolerance to hypotension as assessed by the changes in arterial ketone body ratio in the state of brain death. Transplantation 47(3): 444-8.

Lin, H., Yamamoto, Y., Okamoto, R., Ueda, J., Yamamoto, S., Mori, K., Tanaka, K., Yamaoka, Y. and Ozawa, K. (1989b). Hepatic functional difference between brain death hypotension and hypovolemic hypotension in liver donation. Transplantation Proceedings 21(1 Pt 2): 2389-91.

Loehe, F., Preissler, G., Annecke, T., Bittmann, I., Jauch, K. W. and Messmer, K. (2004). Continuous infusion of nitroglycerin improves pulmonary graft function of non- heart-beating donor lungs. Transplantation 77(12): 1803-8.

Lower, R. R. and Shumway, N. E. (1960). Studies on orthotopic homotransplantation of the canine heart. Surgical Forum 11: 18-19.

Lyons, J. M., Pearl, J. M., McLean, K. M., Akhter, S. A., Wagner, C. J., Pandalai, P. K. and Duffy, J. Y. (2005). Glucocorticoid administration reduces cardiac dysfunction after brain death in pigs. Journal of Heart and Lung Transplantation 24(12): 2249-54.

- 279 - Macdonald, P. (2008). Heart transplantation: who should be considered and when? Internal Medicine Journal 38(12): 911-7.

Mackersie, R. C., Bronsther, O. L. and Shackford, S. R. (1991). Organ procurement in patients with fatal head injuries. The fate of the potential donor. Annals of Surgery 213(2): 143-50.

Maczewski, M. and Beresewicz, A. (2003). Role of nitric oxide and free radicals in cardioprotection by blocking Na+/H+ and Na+/Ca2+ exchange in rat heart. European Journal of Pharmacology 461(2-3): 139-47.

Magliato, K. E. and Trento, A. (2001). Heart transplantation - surgical results. Heart Failure Reviews 6(3): 213-9.

Manaka, D., Okamoto, R., Yokoyama, T., Yamamoto, Y., Washida, M., Ino, K., Yamaoka, Y., Kumada, K. and Ozawa, K. (1992). Maintenance of liver graft viability in the state of brain death. Synergistic effects of vasopressin and epinephrine on hepatic energy metabolism in brain-dead dogs. Transplantation 53(3): 545-50.

Manaka, D., Okamoto, R., Yokoyama, T., Yamamoto, Y., Yamaoka, Y. and Ozawa, K. (1990). Hemodynamic stability and metabolic preservation of the liver under simultaneous administration of vasopressin and epinephrine in brain-dead dogs. Transplantation 49(6): 1188-90.

Mandal, A. K., King, K. E., Humphreys, S. L., Maley, W. R., Burdick, J. F. and Klein, A. S. (2000). Plasmapheresis: an effective therapy for primary allograft nonfunction after liver transplantation. Transplantation 70(1): 216-20.

Mann, F. C., Priestly, J. T., Markowitz, J. and Yater, W. M. (1933). Transplantation of the intact mammalian heart. Archives of Surgery 26: 219-224.

Manzarbeitia, C. Y., Ortiz, J. A., Jeon, H., Rothstein, K. D., Martinez, O., Araya, V. R., Munoz, S. J. and Reich, D. J. (2004). Long-term outcome of controlled, non-heart- beating donor liver transplantation. Transplantation 78(2): 211-5.

Marcus, E., Wong, S. N. and Luisada, A. A. (1953). Homologous heart grafts: transplantation of the heart in dogs. Surgical Forum 2: 212-217.

- 280 - Mariot, J., Sadoune, L. O., Jacob, F., Dousset, B., Perrier, J. F., Jacob, C., Strub, P. and Voltz, C. (1995). Hormone levels, hemodynamics, and metabolism in brain dead organ donors. Transplantation Proceedings 27(1): 793-4.

Maron, M. B., Holcomb, P. H., Dawson, C. A., Rickaby, D. A., Clough, A. V. and Linehan, J. H. (1994). Edema development and recovery in neurogenic pulmonary edema. Journal of Applied Physiology 77(3): 1155-63.

Marques, R. G., Rogers, J., Chavin, K. D., Baliga, P. K., Lin, A., Emovon, O., Afzal, F., Baillie, G. M., Taber, D. J., Ashcraft, E. E. and Rajagopalan, P. R. (2004). Does treatment of cadaveric organ donors with desmopressin increase the likelihood of pancreas graft thrombosis? Results of a preliminary study. Transplantation Proceedings 36(4): 1048-9.

Martin, J., Lutter, G., Ihling, C., Siepe, M., Wagner, S., Hilberath, J., Kemper, M., Sarai, K. and Beyersdorf, F. (2003a). Myocardial viability twenty-four hours after orthotopic heart transplantation from non-heart-beating donors. Journal of Thoracic and Cardiovascular Surgery 125(6): 1217-28.

Martin, J., Sarai, K., Yoshitake, M., Haberstroh, J., Takahashi, N., Lutter, G. and Beyersdorf, F. (1998). Orthotopic transplantation of pig hearts harvested after 30 min of normothermic ischemia: controlled reperfusion with blood cardioplegia containing the Na+-H+-exchange inhibitor HOE 642. European Journal of Cardio Thoracic Surgery 14(6): 607-14.

Martin, J., Sarai, K., Yoshitake, M., Haberstroh, J., Takahashi, N., Lutter, G. and Beyersdorf, F. (1999). Successful orthotopic pig heart transplantation from non-heart- beating donors. Journal of Heart and Lung Transplantation 18(6): 597-606.

Martin, J., Yerebakan, C., Goebel, H., Benk, C., Krause, M., Derjung, G., Lutter, G., Siegenthaler, M. and Beyersdorf, F. (2003b). Viability of the myocardium after twenty- four-hour heart conservation--a preliminary study. Thoracic and Cardiovascular Surgeon 51(4): 196-203.

- 281 - Masson, F., Thicoipe, M., Gin, H., de Mascarel, A., Angibeau, R. M., Favarel- Garrigues, J. F. and Erny, P. (1993). The endocrine pancreas in brain-dead donors. A prospective study in 25 patients. Transplantation 56(2): 363-7.

Matthay, M. A., Folkesson, H. G. and Clerici, C. (2002). Lung epithelial fluid transport and the resolution of pulmonary edema. Physiological Reviews 82(3): 569-600.

McCrystal, G. D., Pepe, S., Esmore, D. S. and Rosenfeldt, F. L. (2004). The challenge of improving donor heart preservation. Heart, Lung and Circulation 13(1): 74-83.

McCully, J. D. and Levitsky, S. (2003). The mitochondrial K(ATP) channel and cardioprotection. Annals of Thoracic Surgery 75(2): S667-73.

McElhinney, D. B., Khan, J. H., Babcock, W. D. and Hall, T. S. (2001). Thoracic organ donor characteristics associated with successful lung procurement. Clinical Transplantation 15(1): 68-71.

McPherson, J., Ed. (1997). Manual Of Use And Interpretation Of Pathology Tests. Sydney, The Royal College of Pathologists of Australasia.

Mellemgaard, K. (1966). The alveolar-arterial oxygen difference: its size and components in normal man. Acta Physiol Scand 67(1): 10-20.

Menasche, P., Pradier, F., Peynet, J., Grousset, C., Mouas, C., Bloch, G. and Piwnica, A. (1991). Limitation of free radical injury by reduced glutathione: an effective means of improving the recovery of heart transplants. Transplantation Proceedings 23(5): 2440-2.

Menasche, P., Termignon, J. L., Pradier, F., Grousset, C., Mouas, C., Alberici, G., Weiss, M., Piwnica, A. and Bloch, G. (1994). Experimental evaluation of Celsior, a new heart preservation solution. European Journal of Cardio-Thoracic Surgery 8(4): 207- 13.

Mentzer Jr, R. M., Bartels, C., Bolli, R., Boyce, S., Buckberg, G. D., Chaitman, B., Haverich, A., Knight, J., Menasche, P., Myers, M. L., Nicolau, J., Simoons, M., Thulin, L. and Weisel, R. D. (2008). Sodium-hydrogen exchange inhibition by cariporide to reduce the risk of ischemic cardiac events in patients undergoing coronary artery bypass

- 282 - grafting: results of the EXPEDITION study. Annals of Thoracic Surgery 85(4): 1261- 70.

Mertes, P. M., Carteaux, J. P., Jaboin, Y., Pinelli, G., el Abassi, K., Dopff, C., Atkinson, J., Villemot, J. P., Burlet, C. and Boulange, M. (1994a). Estimation of myocardial interstitial norepinephrine release after brain death using cardiac microdialysis. Transplantation 57(3): 371-7.

Mertes, P. M., el Abassi, K., Jaboin, Y., Burtin, P., Pinelli, G., Carteaux, J. P., Burlet, C., Boulange, M. and Villemot, J. P. (1994b). Changes in hemodynamic and metabolic parameters following induced brain death in the pig. Transplantation 58(4): 414-8.

Meyers, C. H., D'Amico, T. A., Peterseim, D. S., Jayawant, A. M., Steenbergen, C., Sabiston, D. C., Jr. and Van Trigt, P. (1993). Effects of triiodothyronine and vasopressin on cardiac function and myocardial blood flow after brain death. Journal of Heart and Lung Transplantation 12(1 Pt 1): 68-79.

Michel, P., Hadour, G., Rodriguez, C., Chiari, P. and Ferrera, R. (2000). Evaluation of a new preservative solution for cardiac graft during hypothermia. Journal of Heart and Lung Transplantation 19(11): 1089-97.

Michel, P., Vial, R., Rodriguez, C. and Ferrera, R. (2002). A comparative study of the most widely used solutions for cardiac graft preservation during hypothermia. Journal of Heart and Lung Transplantation 21(9): 1030-9.

Minnear, F. L. and Connell, R. S. (1982). Prevention of aconitine-induced neurogenic pulmonary edema (NPE) with hypovolemia or methylprednisolone. Journal of Trauma 22(2): 121-8.

Mohara, J., Morishita, Y., Takahashi, T., Oshima, K., Yamagishi, T., Takeyoshi, I. and Matsumoto, K. (1999). A comparative study of Celsior and University of Wisconsin solutions based on 12-hr preservation followed by transplantation in canine models. Journal of Heart and Lung Transplantation 18(12): 1202-10.

Morgan, J. A., John, R., Weinberg, A. D., Remoli, R., Kherani, A. R., Vigilance, D. W., Schanzer, B. M., Bisleri, G., Mancini, D. M., Oz, M. C. and Edwards, N. M. (2003).

- 283 - Long-term results of cardiac transplantation in patients 65 years of age and older: a comparative analysis. Annals of Thoracic Surgery 76(6): 1982-7.

Movahed, A., Reeves, W. C., Mehta, P. M., Gilliland, M. G., Mozingo, S. L. and Jolly, S. R. (1994). Norepinephrine-induced left ventricular dysfunction in anesthetized and conscious, sedated dogs. International Journal of Cardiology. 45(1): 23-33.

Mukadam, M. E., Harrington, D. K., Wilson, I. C., Mascaro, J. G., Rooney, S. J., Thompson, R. D., Nightingale, P. and Bonser, R. S. (2005). Does donor catecholamine administration affect early lung function after transplantation? Journal of Thoracic and Cardiovascular Surgery 130(3): 926-7.

Mullen, J. C., Bentley, M. J., Modry, D. L. and Koshal, A. (2001). Extended donor ischemic times and recipient outcome after orthotopic cardiac transplantation. Canadian Journal of Cardiology 17(4): 421-6.

Myers, M. L. and Karmazyn, M. (1996). Improved cardiac function after prolonged hypothermic ischemia with the Na+/H+ exchange inhibitor HOE 694. Annals of Thoracic Surgery 61(5): 1400-6.

Nagareda, T., Kinoshita, Y., Tanaka, A., Hasuike, Y., Terada, N., Nishizawa, Y., Fujita, M. Q., Kuroda, H., Yawata, K. and Aozasa, K. (1989). Clinicopathological study of livers from brain-dead patients treated with a combination of vasopressin and epinephrine. Transplantation 47(5): 792-7.

Nagareda, T., Kinoshita, Y., Tanaka, A., Takeda, M., Sakano, T., Yawata, K., Sugimoto, T., Nishizawa, Y. and Terada, N. (1993). Clinicopathology of kidneys from brain-dead patients treated with vasopressin and epinephrine. Kidney International 43(6): 1363-70.

Najarian, J. S. and Simmons, R. L., Eds. (1972). Transplantation. Philadelphia, Lea & Febiger.

Naka, Y., Chowdhury, N. C., Oz, M. C., Smith, C. R., Yano, O. J., Michler, R. E., Stern, D. M. and Pinsky, D. J. (1995). Nitroglycerin maintains graft vascular homeostasis and enhances preservation in an orthotopic rat lung transplant model. Journal of Thoracic and Cardiovascular Surgery 109(2): 206-10.

- 284 - Nakatani, T., Ishikawa, Y., Kobayashi, K. and Ozawa, K. (1991). Hepatic mitochondrial redox state in hypotensive brain-dead patients and an effect of dopamine administration. Intensive Care Medicine 17(2): 103-7.

Nardo, B., Catena, F., Cavallari, G., Montalti, R., Di Naro, A., Faenza, A. and Cavallari, A. (2001). Randomized clinical study comparing UW and Celsior solution in liver preservation for transplantation: preliminary results. Transplantation Prococeedings 33(1-2): 870-2.

Nawata, S., Sugi, K., Ueda, K., Nawata, K., Kaneda, Y. and Esato, K. (1996). Prostacyclin analog OP2507 prevents pulmonary arterial and airway constriction during lung preservation and reperfusion. Journal of Heart and Lung Transplantation 15(5): 470-4.

Neptune, W. B., Cookson, B. A., Bailey, C. P., Appler, R. and Rajkowski, F. (1953). Complete homologous heart transplantation. Archives of Surgery 66: 174-178.

NHMRC (2004). Australian Code of Practice for the Care and Use of Animals for Scientific Purpose. Canberra: Australian Government.

Nickless, D. K., Rabinov, M., Richards, S. M., Conyers, R. A. and Rosenfeldt, F. L. (1998). Continuous perfusion improves preservation of donor rat hearts: importance of the implantation phase. Annals of Thoracic Surgery 65(5): 1265-72.

Novitzky, D. (1991). Triiodothyronine replacement, the euthyroid sick syndrome, and organ transplantation. Transplantation Proceedings 23(5): 2460-2.

Novitzky, D. (1996a). Novel actions of thyroid hormone: the role of triiodothyronine in cardiac transplantation. Thyroid 6(5): 531-6.

Novitzky, D. (1997a). Detrimental effects of brain death on the potential organ donor. Transplantation Proceedings 29(8): 3770-2.

Novitzky, D. and Cooper, D. K. (1988a). Results of hormonal therapy in human brain- dead potential organ donors. Transplantation Proceedings 20(5 Suppl 7): 59-62.

- 285 - Novitzky, D., Cooper, D. K., Chaffin, J. S., Greer, A. E., DeBault, L. E. and Zuhdi, N. (1990). Improved cardiac allograft function following triiodothyronine therapy to both donor and recipient. Transplantation 49(2): 311-6.

Novitzky, D., Cooper, D. K., Human, P. A., Reichart, B. and Zuhdi, N. (1988b). Triiodothyronine therapy for heart donor and recipient. Journal of Heart Transplantation 7(5): 370-6.

Novitzky, D., Cooper, D. K., Morrell, D. and Isaacs, S. (1987a). Brain death, triiodothyronine depletion, and inhibition of oxidative phosphorylation: relevance to management of organ donors. Transplantation Proceedings 19(5): 4110-1.

Novitzky, D., Cooper, D. K., Morrell, D. and Isaacs, S. (1988c). Change from aerobic to anaerobic metabolism after brain death, and reversal following triiodothyronine therapy. Transplantation 45(1): 32-6.

Novitzky, D., Cooper, D. K., Muchmore, J. S. and Zuhdi, N. (1989a). Pituitary function in brain-dead patients. Transplantation 48(6): 1078-9.

Novitzky, D., Cooper, D. K. and Reichart, B. (1986). Value of triiodothyronine (T3) therapy to brain-dead potential organ donors. Journal of Heart Transplantation 5(6): 486-7.

Novitzky, D., Cooper, D. K. and Reichart, B. (1987b). Hemodynamic and metabolic responses to hormonal therapy in brain-dead potential organ donors. Transplantation 43(6): 852-4.

Novitzky, D., Cooper, D. K. and Reichart, B. (1987c). The value of hormonal therapy in improving organ viability in the transplant donor. Transplantation Proceedings 19(1 Pt 3): 2037-8.

Novitzky, D., Cooper, D. K., Rosendale, J. D. and Kauffman, H. M. (2006). Hormonal therapy of the brain-dead organ donor: experimental and clinical studies. Transplantation 82(11): 1396-401.

Novitzky, D., Cooper, D. K. and Swanepoel, A. (1989b). Inotropic effect of triiodothyronine (T3) in low cardiac output following cardioplegic arrest and

- 286 - cardiopulmonary bypass: an initial experience in patients undergoing open heart surgery. European Journal of Cardio Thoracic Surgery 3(2): 140-5.

Novitzky, D., Cooper, D. K., Wicomb, W., Novitzky, D., Cooper, D. K. C. and Wicomb, W. (2008). Hormonal therapy to the brain-dead potential organ donor: the misnomer of the "Papworth cocktail". Transplantation 86(10): 1479-80.

Novitzky, D., Cooper, D. K. and Zuhdi, N. (1988d). Triiodothyronine therapy in the cardiac transplant recipient. Transplantation Proceedings 20(5 Suppl 7): 65-8.

Novitzky, D., Human, P. A. and Cooper, D. K. (1988e). Inotropic effect of triiodothyronine following myocardial ischemia and cardiopulmonary bypass: an experimental study in pigs. Annals of Thoracic Surgery 45(1): 50-5.

Novitzky, D., Rose, A. G. and Cooper, D. K. C. (1988f). Injury of myocardial conduction tissue and coronary artery smooth muscle following brain death in the baboon. Transplantation 45: 964-966.

Novitzky, D., Wicomb, W. N., Cooper, D. K., Rose, A. G., Fraser, R. C. and Barnard, C. N. (1984). Electrocardiographic, hemodynamic & endocrine changes occurring during experimental brain death in the chacma baboon. The Journal of Heart Transplantation IV(1): 63-69.

Novitzky, D., Wicomb, W. N., Cooper, D. K. and Tjaalgard, M. A. (1987d). Improved cardiac function following hormonal therapy in brain dead pigs: relevance to organ donation. Cryobiology 24(1): 1-10.

Nutt, M. P., Fields, B. L., Belzer, F. O. and Southard, J. H. (1991). Comparison of continuous perfusion and simple cold storage for rabbit heart preservation. Transplantation Proceedings 23(5): 2445-6.

Nygaard, C. E., Townsend, R. N. and Diamond, D. L. (1990). Organ donor management and organ outcome: a 6-year review from a level 1 trauma center. Journal of Trauma 30: 728-32.

O'Connell, P. J., Hawthorne, W. J., Simond, D., Chapman, J. R., Chen, Y., Patel, A. T., Walters, S. N., Burgess, J., Weston, L., Stokes, R. A., Moran, C. and Allen, R. (2005).

- 287 - Genetic and functional evaluation of the level of inbreeding of the Westran pig: a herd with potential for use in . Xenotransplantation 12(4): 308-15.

Obermaier, R., von Dobschuetz, E., Keck, T., Hopp, H. H., Drognitz, O., Schareck, W., Hopt, U. T. and Benz, S. (2004). Brain death impairs pancreatic microcirculation. American Journal of Transplantation 4(2): 210-5.

Okereke, O. U., Frazier, O. H., Reece, I. J., Painvin, G. A., Radovancevic, B. and Cooley, D. A. (1987). Cause and importance of donor neurogenic myocardial injury in cardiac transplantation. Transplantation Proceedings 19: 1034-5.

Ootaki, Y., Kamohara, K., Akiyama, M., Zahr, F., Kopcak, M. W., Jr., Dessoffy, R. and Fukamachi, K. (2007). Ventral cardiac denervation increased right coronary arterial blood flow. Int J Cardiol 114(3): 309-14.

Ou, R., Gavin, J. B., Esmore, D. S. and Rosenfeldt, F. L. (1999). Increased temperature reduces the protective effect of University of Wisconsin solution in the heart. Annals of Thoracic Surgery 68(5): 1628-34; discussion 1634-5.

Oz, M. C., Pinsky, D. J., Koga, S., Liao, H., Marboe, C. C., Han, D., Kline, R., Jeevanandam, V., Williams, M. and Morales, A. (1993). Novel preservation solution permits 24-hour preservation in rat and baboon cardiac transplant models. Circulation 88(5 Pt 2): II291-7.

Papp, M., Varga, B., Varga, G. and Folly, G. (1983). Pancreatic secretion: I. Effects of vasopressin and glucagon on pancreatic blood flow and secretion. Mount Sinai Journal of Medicine. 50(5): 439-40.

Pearl, J. M., Drinkwater, D. C., Jr., Laks, H., Stein, D. G., Capouya, E. R. and Bhuta, S. (1992a). Leukocyte-depleted reperfusion of transplanted human hearts prevents ultrastructural evidence of reperfusion injury. Journal of Surgical Research 52(4): 298- 308.

Pearl, J. M., Drinkwater, D. C., Laks, H., Capouya, E. R. and Gates, R. N. (1992b). Leukocyte-depleted reperfusion of transplanted human hearts: a randomized, double- blind clinical trial. Journal of Heart and Lung Transplantation 11(6): 1082-92.

- 288 - Pedotti, P., Cardillo, M., Rigotti, P., Gerunda, G., Merenda, R., Cillo, U., Zanus, G., Baccarani, U., Berardinelli, M. L., Boschiero, L., Caccamo, L., Calconi, G., Chiaramonte, S., Dal Canton, A., De Carlis, L., Di Carlo, V., Donati, D., Montanaro, D., Pulvirenti, A., Remuzzi, G., Sandrini, S., Valente, U. and Scalamogna, M. (2004). A comparative prospective study of two available solutions for kidney and liver preservation. Transplantation 77(10): 1540-5.

Pennefather, S. H., Bullock, R. E., Mantle, D. and Dark, J. H. (1995). Use of low dose arginine vasopressin to support brain-dead organ donors. Transplantation 59(1): 58-62.

Perez-Blanco, A., Caturla-Such, J., Canovas-Robles, J. and Sanchez-Paya, J. (2005). Efficiency of triiodothyronine treatment on organ donor hemodynamic management and adenine nucleotide concentration. Intensive Care Medicine 31: 943-8.

Pienaar, H., Schwartz, I., Roncone, A., Lotz, Z. and Hickman, R. (1990). Function of kidney grafts from brain-dead donor pigs. The influence of dopamine and triiodothyronine. Transplantation 50(4): 580-2.

Pinsky, D. J., Naka, Y., Chowdhury, N. C., Liao, H., Oz, M. C., Michler, R. E., Kubaszewski, E., Malinski, T. and Stern, D. M. (1994a). The nitric oxide/cyclic GMP pathway in organ transplantation: critical role in successful lung preservation. Proceedings of the National Academy of Sciences of the United States of America 91(25): 12086-90.

Pinsky, D. J., Oz, M. C., Koga, S., Taha, Z., Broekman, M. J., Marcus, A. J., Liao, H., Naka, Y., Brett, J. and Cannon, P. J. (1994b). Cardiac preservation is enhanced in a heterotopic rat transplant model by supplementing the nitric oxide pathway. Journal of Clinical Investigation 93(5): 2291-7.

Pisarenko, O. I., Solomatina, E. S., Ivanov, V. E., Studneva, I. M., Kapelko, V. I. and Smirnov, V. N. (1985). On the mechanism of enhanced ATP formation in hypoxic myocardium caused by glutamic acid. Basic Research In Cardiology 80(2): 126-34.

Plenz, G., Eschert, H., Erren, M., Wichter, T., Bohm, M., Flesch, M., Scheld, H. H. and Deng, M. C. (2002). The interleukin-6/interleukin-6-receptor system is activated in donor hearts. Journal of the American College of Cardiology 39(9): 1508-12.

- 289 - Pocock, S. J. (1983). Clinical Trials: A Practical Approach: John Wiley & Sons.

Potapov, E. V., Ivanitskaia, E. A., Loebe, M., Mockel, M., Muller, C., Sodian, R., Meyer, R. and Hetzer, R. (2001). Value of cardiac troponin I and T for selection of heart donors and as predictors of early graft failure. Transplantation. 71(10): 1394-400.

Potapov, E. V., Wagner, F. D., Loebe, M., Ivanitskaia, E. A., Muller, C., Sodian, R., Jonitz, B. and Hetzer, R. (2003). Elevated donor cardiac troponin T and procalcitonin indicate two independent mechanisms of early graft failure after heart transplantation. International Journal of Cardiology. 92(2-3): 163-7.

Powner, D. J. and Darby, J. M. (2000a). Management of variations in blood pressure during care of organ donors. Progress in Transplantation 10(1): 25-30; quiz 31-2.

Powner, D. J., Darby, J. M. and Kellum, J. A. (2004). Proposed treatment guidelines for donor care. Progress in Transplantation 14(1): 16-26.

Powner, D. J., Darby, J. M. and Stuart, S. A. (2000b). Recommendations for mechanical ventilation during donor care. Progress in Transplantation 10(1): 33-8.

Powner, D. J., Hendrich, A., Lagler, R. G., Ng, R. H. and Madden, R. L. (1990). Hormonal changes in brain dead patients. Critical Care Medicine 18(7): 702-8.

Powner, D. J., Hendrich, A., Nyhuis, A. and Strate, R. (1992). Changes in serum catecholamine levels in patients who are brain dead. Journal of Heart and Lung Transplantation 11(6): 1046-53.

Powner, D. J. and Kellum, J. A. (2000c). Maintaining acid-base balance in organ donors. Progress in Transplantation 10(2): 98-103.

Powner, D. J., Kellum, J. A. and Darby, J. M. (2000d). Abnormalities in fluids, electrolytes, and metabolism of organ donors. Progress in Transplantation 10(2): 88-94.

Pratschke, J., Kofla, G., Wilhelm, M., Vergopoulos, A., Laskowski, I., Shaw, G. D., Tullius, S. G., Volk, H. D., Neuhous, P. and Tilney, N. L. (2002). Improvements in early behaviour of kidney allografts after donor treatment. Transplantation Proceedings 34(6): 2213-4.

- 290 - Pratschke, J., Kofla, G., Wilhelm, M. J., Vergopoulos, A., Laskowski, I., Shaw, G. D., Tullius, S. G., Volk, H. D., Neuhaus, P. and Tilney, N. L. (2001a). Improvements in early behavior of rat kidney allografts after treatment of the brain-dead donor. Annals of Surgery 234(6): 732-40.

Pratschke, J., Wilhelm, M. J., Kusaka, M., Basker, M., Cooper, D. K., Hancock, W. W. and Tilney, N. L. (1999). Brain death and its influence on donor organ quality and outcome after transplantation. Transplantation 67(3): 343-8.

Pratschke, J., Wilhelm, M. J., Kusaka, M., Beato, F., Milford, E. L., Hancock, W. W. and Tilney, N. L. (2000). Accelerated rejection of renal allografts from brain-dead donors. Annals of Surgery 232(2): 263-71.

Pratschke, J., Wilhelm, M. J., Laskowski, I., Kusaka, M., Paz, D., Tullius, S. G., Neuhaus, P., Hancock, W. W. and Tilney, N. L. (2001b). The influence of donor brain death on long-term function of renal allotransplants in rats. Transplantation Proceedings 33(1-2): 693-4.

Qayumi, A. K., Jamieson, W. R., Rosado, L. J., Tomlinson, C. W., Schulzer, M., McConville, B., Gillespie, K. and Wong, A. (1991). Preservation techniques for heart transplantation: comparison of hypothermic storage and hypothermic perfusion. Journal of Heart and Lung Transplantation 10(4): 518-26.

Radomski, M. W., Palmer, R. M. and Moncada, S. (1987a). The anti-aggregating properties of vascular endothelium: interactions between prostacyclin and nitric oxide. British Journal of Pharmacology 92(3): 639-46.

Radomski, M. W., Palmer, R. M. and Moncada, S. (1987b). Endogenous nitric oxide inhibits human platelet adhesion to vascular endothelium. Lancet 2(8567): 1057-8.

Randell, T. T. (2004). Medical and legal considerations of brain death. Acta Anaesthesiologica Scandinavica 48(2): 139-44.

Randell, T. T. and Hockerstedt, K. A. (1992). Triiodothyronine treatment in brain-dead multiorgan donors-a controlled study. Transplantation 54: 736-8.

- 291 - Randell, T. T. and Hockerstedt, K. A. V. (1993). Triiodothyronine treatment is not indicated in brain-dead multi-organ donors: a controlled study. Transplantation Proceedings 25: 1552-3.

Reich, D. J., Munoz, S. J., Rothstein, K. D., Nathan, H. M., Edwards, J. M., Hasz, R. D. and Manzarbeitia, C. Y. (2000). Controlled non-heart-beating donor liver transplantation: a successful single center experience, with topic update. Transplantation 70(8): 1159-66.

Reitz, B. A. (2002). History of heart and heart-lung transplantation. Heart and Lung Transplantation. W. A. Baumgartner, Reitz, B., Kasper, E. and Theodore, J. Philadelphia, W.B. Saunders Company: 3-14.

Remadi, J. P., Baron, O., Roussel, J. C., Al Habash, O., Treilhaud, M., Despins, P., Duveau, D. and Michaud, J. L. (2002). Myocardial preservation using Celsior solution in cardiac transplantation: early results and 5-year follow-up of a multicenter prospective study of 70 cardiac transplantations. Annals of Thoracic Surgery 73(5): 1495-9.

Rezkalla, S. H. and Kloner, R. A. (2002). No-reflow phenomenon. Circulation 105(5): 656-62.

Richens, D., Junius, F., Hill, A., Keogh, A., Macdonald, P., McGoldrick, J. and Spratt, P. (2001). Clinical study of crystalloid cardioplegia vs aspartate-enriched cardioplegia plus warm reperfusion for donor heart preservation. Transplantation Proceedings 25(1 Pt 2): 1608-1610.

Ririe, D. G., Butterworth, J. F., Royster, R. L., MacGregor, D. A. and Zaloga, G. P. (1995). Triiodothyronine increases contractility independent of beta-adrenergic receptors or stimulation of cyclic-3',5'-adenosine monophosphate. Anesthesiology 82(4): 1004-12.

Robertson, K. M., Hramiak, I. M. and Gelb, A. W. (1989). Endocrine changes and haemodynamic stability after brain death. Transplantation Proceedings 21(1 Pt 2): 1197-8.

- 292 - Rosendale, J. D., Chabalewski, F. L., McBride, M. A., Garrity, E. R., Rosengard, B. R., Delmonico, F. L. and Kauffman, H. M. (2002). Increased transplanted organs from the use of a standardized donor management protocol. American Journal of Transplantation 2(8): 761-8.

Rosendale, J. D., Kauffman, H. M., McBride, M. A., Chabalewski, F. L., Zaroff, J. G., Garrity, E. R., Delmonico, F. L. and Rosengard, B. R. (2003a). Hormonal resuscitation yields more transplanted hearts, with improved early function. Transplantation 75(8): 1336-41.

Rosendale, J. D., Kauffman, H. M., McBride, M. A., Chabalewski, F. L., Zaroff, J. G., Garrity, E. R., Delmonico, F. L. and Rosengard, B. R. (2003b). Aggressive pharmacologic donor management results in more transplanted organs. Transplantation 75(4): 482-7.

Rosendale, J. D., Kauffman, H. M., McBride, M. A., Skjei, T. C., Zaroff, J. G., Garrity, E. R., Delmonico, F. L. and Rosengard, B. R. (2004). Hormonal resuscitation associated with more transplanted organs with no sacrifice in survival (abstract). Transplantation 78(2, Supplement 1): 17.

Rosengard, B. R., Feng, S., Alfrey, E. J., Zaroff, J. G., Emond, J. C., Henry, M. L., Garrity, E. R., Roberts, J. P., Wynn, J. J., Metzger, R. A., Freeman, R. B., Port, F. K., Merion, R. M., Love, R. B., Busuttil, R. W. and Delmonico, F. L. (2002). Report of the Crystal City meeting to maximize the use of organs recovered from the cadaver donor. American Journal of Transplantation 2(8): 701-11.

Rostron, A. J., Avlonitis, V. S., Cork, D. M., Grenade, D. S., Kirby, J. A. and Dark, J. H. (2008). Hemodynamic resuscitation with arginine vasopressin reduces lung injury after brain death in the transplant donor. Transplantation 85(4): 597-606.

Rudis, M. I., Basha, M. A. and Zarowitz, B. J. (1996). Is it time to reposition vasopressors and inotropes in sepsis? Critical Care Medicine 24(3): 525-37.

Ruiz-Meana, M., Garcia-Dorado, D., Pina, P., Inserte, J., Agullo, L. and Soler-Soler, J. (2003). Cariporide preserves mitochondrial proton gradient and delays ATP depletion in

- 293 - cardiomyocytes during ischemic conditions. American Journal of Physiology - Heart and Circulatory Physiology 285(3).

Rupprecht, H. J., vom Dahl, J., Terres, W., Seyfarth, K. M., Richardt, G., Schultheibeta, H. P., Buerke, M., Sheehan, F. H. and Drexler, H. (2000). Cardioprotective effects of the Na(+)/H(+) exchange inhibitor cariporide in patients with acute anterior myocardial infarction undergoing direct PTCA. Circulation 101(25): 2902-8.

Russo, M. J., Chen, J. M., Sorabella, R. A., Martens, T. P., Garrido, M., Davies, R. R., George, I., Cheema, F. H., Mosca, R. S., Mital, S., Ascheim, D. D., Argenziano, M., Stewart, A. S., Oz, M. C. and Naka, Y. (2007). The effect of ischemic time on survival after heart transplantation varies by donor age: an analysis of the United Network for Organ Sharing database. Journal of Thoracic & Cardiovascular Surgery 133(2): 554-9.

Ryan, J. B., Hicks, M., Cropper, J. R., Garlick, S. R., Kesteven, S. H., Wilson, M. K., Feneley, M. P. and Macdonald, P. S. (2003a). Cariporide (HOE-642) improves cardiac allograft preservation in a porcine model of orthotopic heart transplantation. Transplantation 75(5): 625-31.

Ryan, J. B., Hicks, M., Cropper, J. R., Garlick, S. R., Kesteven, S. H., Wilson, M. K., Feneley, M. P. and Macdonald, P. S. (2003b). Sodium-hydrogen exchanger inhibition, pharmacologic ischemic preconditioning, or both for extended cardiac allograft preservation. Transplantation 76(5): 766-71.

Ryan, J. B., Hicks, M., Cropper, J. R., Garlick, S. R., Kesteven, S. H., Wilson, M. K., Feneley, M. P. and Macdonald, P. S. (2003c). Functional evidence of reversible ischaemic injury immediately after sympathetic storm associated with experimental brain death. The Journal of Heart and Lung Transplantation 22(8): 922-928.

Ryan, J. B., Hicks, M., Cropper, J. R., Garlick, S. R., Kesteven, S. H., Wilson, M. K., Feneley, M. P. and Macdonald, P. S. (2003d). The initial rate of troponin I release post- reperfusion reflects the effectiveness of myocardial protection during cardiac allograft preservation. European Journal of Cardio-Thoracic Surgery 23(6): 898-906.

Ryan, J. B., Hicks, M., Cropper, J. R., Garlick, S. R., Kesteven, S. H., Wilson, M. K., Macdonald, P. S. and Feneley, M. P. (2002). The preload recruitable stroke work

- 294 - relationship as a measure of left ventricular contractile dysfunction in porcine cardiac allografts. European Journal of Cardio-Thoracic Surgery 22(5): 738-45.

Ryan, J. B., Hicks, M., Cropper, J. R., Nicholson, A., Kesteven, S. H., Wilson, M. K., Feneley, M. P. and Macdonald, P. S. (2003e). Lazaroid (U74389G)-supplemented cardioplegia: results of a double-blind, randomized, controlled trial in a porcine model of orthotopic heart transplantation. Journal of Heart & Lung Transplantation 22(3): 347-56.

Ryan, J. B., Wilson, M. K., Hicks, M., Nicholson, A., Kesteven, S. H., Junius, F., Feneley, M. P. and Macdonald, P. S. (2000). A brain dead donor model for porcine orthotopic cardiac transplantation for assessment of cardiac allograft preservation. Heart, Lung and Circulation 9: 78-81.

Salim, A., Martin, M., Brown, C., Inaba, K., Roth, B., Hadjizacharia, P., Mascarenhas, A., Rhee, P. and Demetriades, D. (2007). Using thyroid hormone in brain-dead donors to maximize the number of organs available for transplantation. Clinical Transplantation 21(3): 405-9.

Salim, A., Vassiliu, P., Velmahos, G. C., Sava, J., Murray, J. A., Belzberg, H., Asensio, J. A. and Demetriades, D. (2001). The role of thyroid hormone administration in potential organ donors. Archives of Surgery. 136(12): 1377-80.

Salim, A., Velmahos, G. C., Brown, C., Belzberg, H. and Demetriades, D. (2005). Aggressive organ donor management significantly increases the number of organs available for transplantation. Journal of Trauma Injury Infection & Critical Care 58(5): 991-4.

Sanchez-Urdazpal, L., Gores, G. J., Ferguson, D. M. and Krom, R. A. (1991). Improved liver preservation with addition of iloprost to Eurocollins and University of Wisconsin storage solutions. Transplantation 52(6): 1105-7.

Sarko, J., Pollack, C. V., Jr., Sarko, J. and Pollack, C. V., Jr. (2002). Cardiac troponins. Journal of Emergency Medicine 23(1): 57-65.

- 295 - Sayen, M. R., Rohrer, D. K. and Dillmann, W. H. (1992). Thyroid hormone response of slow and fast sarcoplasmic reticulum Ca2+ ATPase mRNA in striated muscle. Molecular and Cellular Endocrinology 87(1-3): 87-93.

Sazontseva, I. E., Kozlov, I. A., Moisuc, Y. G., Ermolenko, A. E., Afonin, V. V. and Ilnitskiy, V. V. (1991). Hormonal response to brain death. Transplantation Proceedings 23(5).

Scheule, A. M., Jost, D., Beierlein, W., Zurakowski, D., Haas, J., Vogel, U., Miller, S., Wendel, H. P. and Ziemer, G. (2003). Sodium-hydrogen inhibitor cariporide (HOE 642) improves in situ protection of hearts from non-heart-beating donors. Journal of Heart and Lung Transplantation 22(12): 1335-42.

Schneider, A., Toledo-Pereyra, L. H., Zeichner, W. D., Allaben, R. and Whitten, J. (1983). Effect of dopamine and pitressin on kidneys procured and harvested for transplantation. Transplantation 36(1): 110-1.

Schnuelle, P., Berger, S., de Boer, J., Persijn, G. and van der Woude, F. J. (2001). Effects of catecholamine application to brain-dead donors on graft survival in solid organ transplantation. Transplantation 72(3): 455-63.

Schnuelle, P., Lorenz, D., Mueller, A., Trede, M. and van der Woude, F. J. (1999). Donor catecholamine use reduces acute allograft rejection and improves graft survival after cadaveric renal transplantation. Kidney International 56: 738-46.

Scudamore, C. H. C. R. and Rolles, K. (1984). The multiple organ donor: The Cambridge experience. Transplantation Proceedings 16: 225-.

Segel, L. D., Follette, D. M., Castellanos, L. M., Hayes, R., Baker, J. M. and Smolens, I. V. (1997). Steroid pretreatment improves graft recovery in a sheep orthotopic heart transplantation model. Journal of Heart and Lung Transplantation 16(4): 371-80.

Seguin, C., Devaux, Y., Grosjean, S., Siaghy, E. M., Mairose, P., Zannad, F., de Talance, N., Ungureanu-Longrois, D. and Mertes, P. M. (2001). Evidence of functional myocardial ischemia associated with myocardial dysfunction in brain-dead pigs. Circulation 104(12 Suppl 1): I197-201.

- 296 - Shivalkar, B., Van Loon, J., Wieland, W., Tjandra-Maga, T. B., Borgers, M., Plets, C. and Flameng, W. (1993). Variable effects of explosive or gradual increase of intracranial pressure on myocardial structure and function. Circulation 87(1): 230-9.

Singhal, A. K., Abrams, J. D., Mohara, J., Hasz, R. D., Nathan, H. M., Fisher, C. A., Furukawa, S. and Goldman, B. I. (2005). Potential suitability for transplantation of hearts from human non-heart-beating donors: data review from the Gift of Life Donor Program. Journal of Heart and Lung Transplantation 24(10): 1657-64.

Skerrett, D., Mor, E., Curtiss, S. and Mohandas, K. (1996). Plasmapheresis in primary dysfunction of hepatic transplants. Journal of Clinical Apheresis 11(1): 10-3.

Slinker, B. K. and Glantz, S. A. (1988). Multiple linear regression is a useful alternative to traditional analyses of variance. American Journal of Physiology 255(3 Pt 2): R353- 67.

Smith, M. (2004). Physiologic changes during brain stem death-lessons for management of the organ donor. Journal of Heart and Lung Transplantation 23(9 Suppl): s217-22.

Snow, T. R., Deal, M. T., Connelly, T. S., Yokoyama, Y. and Novitzky, D. (1992). Acute inotropic response of rabbit papillary muscle to triiodothyronine. Cardiology 80(2): 112-7.

Sorajja, P., Cable, D. G. and Schaff, H. V. (1997). Hypothermic storage with University of Wisconsin solution preserves endothelial and vascular smooth-muscle function. Circulation 96(9 Suppl): II-297-302; discussion II-302-3.

Southard, J. H., van Gulik, T. M., Ametani, M. S., Vreugdenhil, P. K., Lindell, S. L., Pienaar, B. L. and Belzer, F. O. (1990). Important components of the UW solution. Transplantation 49(2): 251-7.

Steen, S., Sjoberg, T., Pierre, L., Liao, Q., Eriksson, L. and Algotsson, L. (2001). Transplantation of lungs from a non-heart-beating donor. Lancet 357(9259): 825-9.

Steenbergen, C., Murphy, E., Watts, J. A. and London, R. E. (1990). Correlation between cytosolic free calcium, contracture, ATP, and irreversible ischemic injury in perfused rat heart. Circulation Research 66(1): 135-46.

- 297 - Steendijk, P., Baan, J., Jr., Van der Velde, E. T. and Baan, J. (1998). Effects of critical coronary stenosis on global systolic left ventricular function quantified by pressure- volume relations during dobutamine stress in the canine heart. Journal of the American College of Cardiology 32(3): 816-26.

Steiner, E., Koller, J., Geleff, S., Dietze, O., Konigsrainer, A. and Margreiter, R. (1988). Who can serve as a pancreas donor? Transplantation Proceedings 20(5): 1005-6.

Stevens, R. M., Salik Jahania, M., Mentzer, R. M., Jr. and Lasley, R. D. (2004). Sodium-hydrogen exchange inhibition attenuates in vivo porcine myocardial stunning. Annals of Thoracic Surgery 77(2): 651-7.

Stinson, E. B., Dong, E., Jr., Schroeder, J. S., Harrison, D. C. and Shumway, N. E. (1968). Initial clinical experience with heart transplantation. American Journal of Cardiology 22: 791-803.

Sugimoto, T., Sakano, T., Kinoshita, Y., Masui, M. and Yoshioka, T. (1992). Morphological and functional alterations of the hypothalamic-pituitary system in brain death with long-term bodily living. Acta Neurochirurgica 115(1-2): 31-6.

Suleiman, M. S., Halestrap, A. P. and Griffiths, E. J. (2001). Mitochondria: a target for myocardial protection. Pharmacology & Therapeutics 89(1): 29-46.

Swindle, M. M. (1984). Swine as replacements for dogs in the surgical teaching and research laboratory. Laboratory Animal Science 34(4): 383-5.

Swindle, M. M., Horneffer, P. J., Gardner, T. J., Gott, V. L., Hall, T. S., Stuart, R. S., Baumgartner, W. A., Borkon, A. M., Galloway, E. and Reitz, B. A. (1986). Anatomic and anesthetic considerations in experimental cardiopulmonary surgery in swine. Laboratory Animal Science 36(4): 357-61.

Swindle, M. M. and Smith, A. C. (1998). Comparative anatomy and physiology of the pig. Scandanavian Journal of Laboratory Animal Science 25(Suppl 1): 11-21.

Swindle, M. M., Smith, A. C. and Hepburn, B. J. (1988). Swine as models in experimental surgery. Journal of Investigative Surgery 1(1): 65-79.

- 298 - Szostek, M., Gaciong, Z., Danielelewicz, R., Lagiewska, B., Pacholczyk, M., Chmura, A., Laskowski, I., Walaszewski, J. and Rowinski, W. (1997). Influence of thyroid function in brain stem death donors on kidney allograft function. Transplantation Proceedings 29(8): 3354-6.

Takada, M., Nadeau, K. C., Hancock, W. W., Mackenzie, H. S., Shaw, G. D., Waaga, A. M., Chandraker, A., Sayegh, M. H. and Tilney, N. L. (1998). Effects of explosive brain death on cytokine activation of peripheral organs in the rat. Transplantation 65(12): 1533-42.

Takeuchi, M., Odake, M., Takaoka, H., Hayashi, Y. and Yokoyama, M. (1992). Comparison between preload recruitable stroke work and the end-systolic pressure- volume relationship in man. European Heart Journal 13 Suppl E: 80-4.

Taniguchi, S., Kitamura, S., Kawachi, K., Doi, Y. and Aoyama, N. (1992). Effects of hormonal supplements on the maintenance of cardiac function in potential donor patients after cerebral death. European Journal of Cardio Thoracic Surgery 6(2): 96- 101.

Tanoue, Y., Morita, S., Ochiai, Y., Haraguchi, N., Tominaga, R., Kawachi, Y. and Yasui, H. (1999). Nitroglycerin as a nitric oxide donor accelerates lipid peroxidation but preserves ventricular function in a canine model of orthotopic heart transplantation. Journal of Thoracic and Cardiovascular Surgery 118(3): 547-56.

Taylor, D. O., Edwards, L. B., Aurora, P., Christie, J. D., Dobbels, F., Kirk, R., Rahmel, A. O., Kucheryavaya, A. Y. and Hertz, M. I. (2008). Registry of the International Society for Heart and Lung Transplantation: twenty-fifth official adult heart transplant report--2008. Journal of Heart and Lung Transplantation 27(9): 943-56.

Taylor, D. O., Edwards, L. B., Boucek, M. M., Trulock, E. P., Aurora, P., Christie, J., Dobbels, F., Rahmel, A. O., Keck, B. M. and Hertz, M. I. (2007). Registry of the International Society for Heart and Lung Transplantation: twenty-fourth official adult heart transplant report--2007. Journal of Heart and Lung Transplantation 26(8): 769- 81.

- 299 - Taylor, D. O., Edwards, L. B., Boucek, M. M., Trulock, E. P., Deng, M. C., Keck, B. M. and Hertz, M. I. (2005). Registry of the International Society for Heart and Lung Transplantation: twenty-second official adult heart transplant report--2005. Journal of Heart and Lung Transplantation 24(8): 945-55.

Taylor, D. O., Edwards, L. B., Boucek, M. M., Trulock, E. P., Keck, B. M. and Hertz, M. I. (2004). The Registry of the International Society for Heart and Lung Transplantation: twenty-first official adult heart transplant report--2004. Journal of Heart and Lung Transplantation 23(7): 796-803.

Taylor, D. O., Edwards, L. B., Boucek, M. M., Trulock, E. P., Waltz, D. A., Keck, B. M. and Hertz, M. I. (2006). Registry of the International Society for Heart and Lung Transplantation: twenty-third official adult heart transplant report--2006. Journal of Heart and Lung Transplantation 25(8): 869-79.

Taylor, D. O., Stehlik, J., Edwards, L. B., Aurora, P., Christie, J. D., Dobbels, F., Kirk, R., Kucheryavaya, A. Y., Rahmel, A. O. and Hertz, M. I. (2009). Registry of the International Society for Heart and Lung Transplantation: Twenty-sixth official adult heart transplant report - 2009. Journal of Heart and Lung Transplantation 28(10): 1007- 22.

Terasaki, P. I., Cecka, J. M., Gjerston, D. W. and Takemoto, S. (1995). High survival rates of kidney transplants from spousal and living unrelated donors. New England Journal of Medicine 333(6): 333-336.

Teshima, Y., Akao, M., Jones, S. P. and Marban, E. (2003). Cariporide (HOE642), a selective Na+-H+ exchange inhibitor, inhibits the mitochondrial death pathway. Circulation 108(18): 2275-81.

Thabut, G., Vinatier, I., Brugiere, O., Leseche, G., Loirat, P., Bisson, A., Marty, J., Fournier, M. and Mal, H. (2001). Influence of preservation solution on early graft failure in clinical lung transplantation. American Journal of Respiratory and Critical Care Medicine 164(7): 1204-8.

- 300 - Thabut, G., Vinatier, I., Stern, J. B., Leseche, G., Loirat, P., Fournier, M. and Mal, H. (2002). Primary graft failure following lung transplantation: predictive factors of mortality. Chest 121(6): 1876-82.

Theroux, P. (1999). Protection of the myocardial cell during ischemia. American Journal of Cardiology 83(10A): 3G-9G.

Theroux, P., Chaitman, B. R., Danchin, N., Erhardt, L., Meinertz, T., Schroeder, J. S., Tognoni, G., White, H. D., Willerson, J. T. and Jessel, A. (2000). Inhibition of the sodium-hydrogen exchanger with cariporide to prevent myocardial infarction in high- risk ischemic situations. Main results of the GUARDIAN trial. Guard during ischemia against necrosis (GUARDIAN) Investigators. Circulation 102(25): 3032-8.

Tilney, N. L., Paz, D., Ames, J., Gasser, M., Laskowski, I. and Hancock, W. W. (2001). Ischemia-reperfusion injury. Transplantation Proceedings 33(1-2): 843-4.

Todd, G. L., Baroldi, G., Pieper, G. M., Clayton, F. C. and Eliot, R. S. (1985a). Experimental catecholamine-induced myocardial necrosis. I. Morphology, quantification and regional distribution of acute contraction band lesions. Journal of Molecular and Cellular Cardiology 17(4): 317-38.

Todd, G. L., Baroldi, G., Pieper, G. M., Clayton, F. C. and Eliot, R. S. (1985b). Experimental catecholamine-induced myocardial necrosis. II. Temporal development of isoproterenol-induced contraction band lesions correlated with ECG, hemodynamic and biochemical changes. Journal of Molecular and Cellular Cardiology 17(7): 647-56.

Trivieri, M. G., Oudit, G. Y., Sah, R., Kerfant, B. G., Sun, H., Gramolini, A. O., Pan, Y., Wickenden, A. D., Croteau, W., Morreale de Escobar, G., Pekhletski, R., St Germain, D., Maclennan, D. H. and Backx, P. H. (2006). Cardiac-specific elevations in thyroid hormone enhance contractility and prevent pressure overload-induced cardiac dysfunction. Proceedings of the National Academy of Sciences of the United States of America 103(15): 6043-8.

Tune, J. D., Richmond, K. N., Gorman, M. W., Feigl, E. O., Tune, J. D., Richmond, K. N., Gorman, M. W. and Feigl, E. O. (2002). Control of coronary blood flow during exercise. Experimental Biology and Medicine 227(4): 238-50.

- 301 - Tuttle-Newhall, J. E., Collins, B. H., Kuo, P. C. and Schroeder, R. A. (2003). Organ donation and treatment of the multi-organ donor. Current Problems in Surgery 40(5): 253-310.

Valero, R. (2002). Donor management: one step forward. American Journal of Transplantation 2(8): 693-4. van der Hoeven, J. A., Lindell, S., van Schilfgaarde, R., Molema, G., Ter Horst, G. J., Southard, J. H. and Ploeg, R. J. (2001). Donor brain death reduces survival after transplantation in rat livers preserved for 20 hr. Transplantation. 72(10): 1632-6. van der Hoeven, J. A., Moshage, H., Schuurs, T., Nijboer, M., Van Schilfgaarde, R. and Ploeg, R. J. (2003). Brain death induces apoptosis in donor liver of the rat. Transplantation 76(8): 1150-4. van der Hoeven, J. A., Ploeg, R. J., Postema, F., Molema, G., de Vos, P., Girbes, A. R., van Suylichem, P. T. and Ter Horst, G. J. (1999a). Induction of organ dysfunction and activation of inflammatory markers in donor liver and kidney during hypotensive brain death. Transplantation Proceedings 31(1-2): 1006-7. van der Hoeven, J. A., Ploeg, R. J., Postema, F., Molema, I., de Vos, P., Girbes, A. R., van Suylichem, P. T., van Schilfgaarde, R. and Ter Horst, G. J. (1999b). Induction of organ dysfunction and up-regulation of inflammatory markers in the liver and kidneys of hypotensive brain dead rats: a model to study marginal organ donors. Transplantation 68(12): 1884-90.

Vega, J. D., Ochsner, J. L., Jeevanandam, V., McGiffin, D. C., McCurry, K. R., Mentzer, R. M., Jr., Stringham, J. C., Pierson, R. N., 3rd, Frazier, O. H., Menkis, A. H., Staples, E. D., Modry, D. L., Emery, R. W., Piccione, W., Jr., Carrier, M., Hendry, P. J., Aziz, S., Furukawa, S. and Pham, S. M. (2001). A multicenter, randomized, controlled trial of Celsior for flush and hypothermic storage of cardiac allografts. Annals of Thoracic Surgery. 71(5): 1442-7.

Venkateswaran, R. V., Patchell, V. B., Wilson, I. C., Mascaro, J. G., Thompson, R. D., Quinn, D. W., Stockley, R. A., Coote, J. H. and Bonser, R. S. (2008). Early donor

- 302 - management increases the retrieval rate of lungs for transplantation. Annals of Thoracic Surgery 85(1): 278-86.

Verma, S., Fedak, P. W., Weisel, R. D., Butany, J., Rao, V., Maitland, A., Li, R. K., Dhillon, B. and Yau, T. M. (2002). Fundamentals of reperfusion injury for the clinical cardiologist. Circulation 105(20): 2332-6.

Vigneswaran, W. T., Edwards, B. S. and McGregor, C. G. (1993). Extended donor use for heart transplantation. Annals of Thoracic Surgery 56(3): 552-3. von Oppell, U. O., Pfeiffer, S., Preiss, P., Dunne, T., Zilla, P. and Reichart, B. (1990). Endothelial cell toxicity of solid-organ preservation solutions. Annals of Thoracic Surgery 50(6): 902-10.

Wada, H., Liu, C. J., Hirata, T., Bando, T. and Kosaka, S. (1099). Effective 30-hour preservation of canine lungs with modified ET-Kyoto solution. Annals of Thoracic Surgery 61(4): 1099-105.

Wallwork, J., Jones, K., Cavarocchi, N., Hakim, M. and Higenbottam, T. (1987). Distant procurement of organs for clinical heart-lung transplantation using a single flush technique. Transplantation 44(5): 654-8.

Wan, S., DeSmet, J. M., Antoine, M., Goldman, M., Vincent, J. L. and LeClerc, J. L. (1996). Steroid administration in heart and heart-lung transplantation: is the timing adequate? Annals of Thoracic Surgery 61(2): 674-8.

Washida, M., Okamoto, R., Manaka, D., Yokoyama, T., Yamamoto, Y., Ino, K., Yamamoto, N., Osaki, N., Shimahara, Y. and Yamaoka, Y. (1992). Beneficial effect of combined 3,5,3'-triiodothyronine and vasopressin administration of hepatic energy status and systemic hemodynamics after brain death. Transplantation 54(1): 44-9.

Webb, W. R. and Howard, H. S. (1957). Cardio-pulmonary transplantation. Surgical Forum 8: 313-317.

Weiss, S., Kotsch, K., Francuski, M., Reutzel-Selke, A., Mantouvalou, L., Klemz, R., Kuecuek, O., Jonas, S., Wesslau, C., Ulrich, F., Pascher, A., Volk, H. D., Tullius, S. G., Neuhaus, P. and Pratschke, J. (2007). Brain death activates donor organs and is

- 303 - associated with a worse I/R injury after liver transplantation. American Journal of Transplantation 7(6): 1584-93.

Wheeldon, D., Sharples, L., Wallwork, J. and English, T. (1992). Donor heart preservation survey. Journal of Heart and Lung Transplantation 11(5): 986-93.

Wheeldon, D. R., Potter, C. D., Jonas, M., Wallwork, J. and Large, S. R. (1994). Using "unsuitable" hearts for transplantation. European Journal of Cardio Thoracic Surgery 8(1): 7-9.

Wheeldon, D. R., Potter, C. D., Oduro, A., Wallwork, J. and Large, S. R. (1995). Transforming the "unacceptable" donor: outcomes from the adoption of a standardized donor management technique. Journal of Heart and Lung Transplantation 14(4): 734- 42.

White, F. C. and Blolor, C. M. (1986). The pig as model for myocardial ischemia. Swine in Biomedical Research. M. E. Tumbleson. New York, Plenum Press. 1: 481-490.

White, M., Wiechmann, R. J., Roden, R. L., Hagan, M. B., Wollmering, M. M., Port, J. D., Hammond, E., Abraham, W. T., Wolfel, E. E. and Lindenfeld, J. (1995). Cardiac beta-adrenergic neuroeffector systems in acute myocardial dysfunction related to brain injury. Evidence for catecholamine-mediated myocardial damage. Circulation 92(8): 2183-9.

Wicomb, W., Cooper, D. K., Hassoulas, J., Rose, A. G. and Barnard, C. N. (1982a). Orthotopic transplantation of the baboon heart after 20 to 24 hours' preservation by continuous hypothermic perfusion with an oxygenated hyperosmolar solution. Journal of Thoracic and Cardiovascular Surgery 83(1): 133-40.

Wicomb, W. N. and Collins, G. M. (1989a). 24-hour rabbit heart storage with UW solution. Effects of low-flow perfusion, colloid, and shelf storage. Transplantation 48(1): 6-9.

Wicomb, W. N., Cooper, D. K. and Barnard, C. N. (1982b). Twenty-four-hour preservation of the pig heart by a portable hypothermic perfusion system. Transplantation 34(5): 246-50.

- 304 - Wicomb, W. N., Cooper, D. K., Lanza, R. P., Novitzky, D. and Isaacs, S. (1986a). The effects of brain death and twenty-four hours storage by hypothermic perfusion on donor heart function in the pig. Journal of Thoracic and Cardiovascular Surgery 91(6): 896- 909.

Wicomb, W. N., Cooper, D. K. and Novitzky, D. (1986b). Impairment of renal slice function following brain death, with reversibility of injury by hormonal therapy. Transplantation 41(1): 29-33.

Wicomb, W. N., Cooper, D. K. C., Novitzky, D. and Barnard, C. N. (1984). Cardiac transplantation following storage of the donor heart by a portable hypothermic perfusion system. Annals of Thoracic Surgery 37: 243-248.

Wicomb, W. N., Novitzky, D., Cooper, D. K., Wells, M. and Hill, J. D. (1989b). Early extranuclear effect of triiodothyronine (T3) on tissue slices: relevance to organ donor viability. Transplantation Proceedings 21(1 Pt 2): 1263-4.

Wieselthaler, G. M., Chevtchik, O., Konetschny, R., Moidl, R., Mallinger, R., Mares, P., Griessmacher, A., Grimm, M., Wolner, E. and Laufer, G. (1999). Improved graft function using a new myocardial preservation solution: Celsior. Preliminary data from a randomized prospective study. Transplantation Proceedings 31(5): 2067-8.

Wigfield, C. H., Golledge, H. D., Shenton, B. K., Kirby, J. A. and Dark, J. H. (2002). Ameliorated reperfusion injury in lung transplantation after reduction of brain death induced inflammatory graft damage in the donor. Journal of Heart and Lung Transplantation 21(1): 57.

Wijdicks, E. F. (2001). The diagnosis of brain death. New England Journal of Medicine 344(16): 1215-21.

Wijdicks, E. F. (2002). Brain death worldwide: accepted fact but no global consensus in diagnostic criteria. Neurology 58(1): 20-5.

Wildhirt, S. M., Weis, M., Schulze, C., Conrad, N., Rieder, G., Enders, G., Ihnken, K., von Scheidt, W. and Reichart, B. (2000). Effects of Celsior and University of Wisconsin preservation solutions on hemodynamics and endothelial function after cardiac

- 305 - transplantation in humans: a single-center, prospective, randomized trial. Transplant International 13(Suppl 1): S203-11.

Wilhelm, M. J., Pratschke, J., Beato, F., Taal, M., Kusaka, M., Hancock, W. W. and Tilney, N. L. (2000a). Activation of the heart by donor brain death accelerates acute rejection after transplantation. Circulation 102(19): 2426-33.

Wilhelm, M. J., Pratschke, J., Laskowski, I. A., Paz, D. M. and Tilney, N. L. (2000b). Brain death and its impact on the donor heart-lessons from animal models. Journal of Heart and Lung Transplantation 19(5): 414-8.

Windsor, H. M. (1969). A heart transplantation: 3. Surgical and postoperative surgical aspects. The Medical Journal of Australia 1(13): 666-670.

Wood, K. E., Becker, B. N., McCartney, J. G., D'Alessandro, A. M. and Coursin, D. B. (2004). Care of the potential organ donor. New England Journal of Medicine 351(26): 2730-9.

Yamaoka, Y., Taki, Y., Gubernatis, G., Nakatani, T., Okamoto, R., Yamamoto, Y., Ishikawa, Y., Ringe, B., Bunzendahl, H. and Oellerich, M. (1990). Evaluation of the liver graft before procurement. Significance of arterial ketone body ratio in brain-dead patients. Transplant International 3(2): 78-81.

Yang, Q., Zhang, R. Z., Yim, A. P. and He, G. W. (2004). Histidine-tryptophan- ketoglutarate solution maximally preserves endothelium-derived hyperpolarizing factor- mediated function during heart preservation: comparison with University of Wisconsin solution. Journal of Heart and Lung Transplantation 23(3): 352-9.

Yeh, T., Jr., Wechsler, A. S., Graham, L. J., Loesser, K. E., Sica, D. A., Wolfe, L. and Jakoi, E. R. (1999). Acute brain death alters left ventricular myocardial gene expression. Journal of Thoracic and Cardiovascular Surgery 117(2): 365-74.

Yoshioka, T., Sugimoto, H., Uenishi, M., Sakamoto, T., Sadamitsu, D., Sakano, T. and Sugimoto, T. (1986). Prolonged hemodynamic maintenance by the combined administration of vasopressin and epinephrine in brain death: a clinical study. Neurosurgery 18(5): 565-7.

- 306 - Zaroff, J. G., Rosengard, B. R., Armstrong, W. F., Babcock, W. D., D'Alessandro, A., Dec, G. W., Edwards, N. M., Higgins, R. S., Jeevanandum, V., Kauffman, M., Kirklin, J. K., Large, S. R., Marelli, D., Peterson, T. S., Ring, W. S., Robbins, R. C., Russell, S. D., Taylor, D. O., Van Bakel, A., Wallwork, J. and Young, J. B. (2002). Consensus conference report: maximizing use of organs recovered from the cadaver donor: cardiac recommendations, March 28-29, 2001, Crystal City, Va. Circulation 106(7): 836-41.

Zembala, M., Religa, Z., Majewski, J., Knapik, P., Bohatyrewicz, R., Banas, S., Pakosiewicz, W., Juszczyk, G., Czapla, J., Nadziakiewicz, P., Glanc, W., Puzio, J. and Latasiewicz, H. (1996). Extended donor criteria--a possible way to enhance the number of potential heart donors. Annals of Transplantation 1(1): 44-5.

Zuppa, A. F., Nadkarni, V., Davis, L., Adamson, P. C., Helfaer, M. A., Elliott, M. R., Abrams, J. and Durbin, D. (2004). The effect of a thyroid hormone infusion on vasopressor support in critically ill children with cessation of neurologic function. Critical Care Medicine 32(11): 2318-22.

- 307 -