Cytokines and the Abdominal Surgeon

MEDICAL INTELLIGENCE UNIT

3

Moshe Schein • Leslie Wise

Cytokines and the Abdominal Surgeon

R.G. LANDES C OM PA N Y

MEDICAL INTELLIGENCE UNIT 3

Cytokines and the Abdominal Surgeon Moshe Schein, M.D., F.C.S. (SA) Associate Professor of Surgery, Cornell University College of Medicine Brooklyn, NY

Leslie Wise, M.D., F.R.C.S. (Eng) Professor of Surgery, Cornell University College of Medicine Brooklyn, NY

R.G. LANDES COMPANY AUSTIN, TEXAS U.S.A.

MEDICAL INTELLIGENCE UNIT Cytokines and the Abdominal Surgeon R.G. LANDES COMPANY Austin, Texas, U.S.A. Copyright © 1998 R.G. Landes Company All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Printed in the U.S.A. Please address all inquiries to the Publishers: R.G. Landes Company, 810 South Church Street, Georgetown, Texas, U.S.A. 78626 Phone: 512/ 863 7762; FAX: 512/ 863 0081

ISBN: 1-57059-536-4

While the authors, editors and publisher believe that drug selection and dosage and the specifications and usage of equipment and devices, as set forth in this book, are in accord with current recommendations and practice at the time of publication, they make no warranty, expressed or implied, with respect to material described in this book. In view of the ongoing research, equipment development, changes in governmental regulations and the rapid accumulation of information relating to the biomedical sciences, the reader is urged to carefully review and evaluate the information provided herein.

Library of Congress Cataloging-in-Publication Data

CIP information applied for but not received at time of publication.

MEDICAL INTELLIGENCE UNIT 3 PUBLISHER’S NOTE

Cytokines and the Abdominal Surgeon

Landes Bioscience produces books in six Intelligence Unit series: Medical, Molecular Biology, Neuroscience, Tissue Engineering, Biotechnology and Environmental. The authors of our books are acknowledged leaders in their fields. Topics are unique; almost without exception, no similar books exist on these topics. Our goal is to publish books in important and rapidly changing areas of bioscience for sophisticated researchers and clinicians. To achieve this goal, we have accelerated our publishing program to conform to the fast pace at which information grows in bioscience. Most of our books are published within to 120 days of receipt of Associate Professor of90 Surgery, the manuscript. WeUniversity would likeCollege to thank our readers for their Cornell of Medicine continuing interest and welcome any comments or suggestions they Brooklyn, NY may have for future books.

Moshe Schein, M.D., F.C.S. (SA)

Judith Kemper Leslie Wise, M.D., F.R.C.S. (Eng)

Professor of Surgery, Production Manager Landes Company Cornell University College of R.G. Medicine Brooklyn, NY

R.G. LANDES COMPANY AUSTIN, TEXAS U.S.A.

CONTENTS 1. Introduction—Cytokines and the Abdominal Surgeon ......................... 1 Moshe Schein and Leslie Wise 2. Relevance of Molecular Biology for the Abdominal Surgeon ................ 3 Philip S. Barie and Steven M. Cohen Introduction ............................................................................................. 3 Structure and Function of DNA and Chromosomes ............................ 4 Human Genome Project ......................................................................... 4 Molecular Diagnostic Tests ..................................................................... 5 Protein Hybridization ............................................................................. 6 Programmed Cell Death ......................................................................... 8 Genetic Predisposition to Disease: Preparing for the Genetic Revolution .............................................. 10 Gene Therapy in Surgery ...................................................................... 11 Antisense Oligonucleotides as Potential Gene Therapy ...................... 13 3. Cytokine Response in Abdominal Surgery ............................................ 17 Edward Lin, Steve E. Calvano, and Stephen F. Lowry Introduction ........................................................................................... 17 Injury and Principal Cytokine Responses ............................................ 19 Clinical Considerations ......................................................................... 25 Conclusions ........................................................................................... 30 4. Cytokines and the Abdominal Operative Wound ................................. 35 Frank J. Thornton and Adrian Barbul Introduction ........................................................................................... 35 Overview of the Repair Process ............................................................ 35 Platelet Degranulation ........................................................................... 36 Immune Cell Influx ............................................................................... 37 Transforming Growth Factor ! .......................................................... 37 Interleukins ............................................................................................ 38 Interferon-∀ ................................................................................................... 38 Tumor Necrosis Factor # ................................................................... 39 Nitric Oxide ........................................................................................... 39 Extracellular Matrix Production ........................................................... 39 Angiogenesis .......................................................................................... 40 Wound Contraction .............................................................................. 40 Epithelialization ..................................................................................... 42 Cytokines in Surgical Site Infection ...................................................... 43 Conclusion ............................................................................................. 44

5. Cytokines and Postoperative Abdominal Adhesions ............................ 53 Shaun G. Appleton and Jeremy N.Thompson Introduction ........................................................................................... 53 Adhesion Formation ............................................................................. 54 Interleukin-1 (IL-1) ............................................................................... 55 Tumor Necrosis Factor-alpha (TNF-#) ............................................... 56 Interleukin-6 (IL-6) ............................................................................... 57 Transforming Growth Factor-beta (TGF-!) ........................................ 57 Other Cytokines ..................................................................................... 58 Conclusions ........................................................................................... 58 6. Cytokines in Blunt Abdominal Trauma ................................................ 63 Matthias W. Wichmann and Eugen Faist Introduction ........................................................................................... 63 Introduction to the Cytokines Studied in Trauma Research .............. 63 Pro- and Anti-Inflammatory Mediators .............................................. 66 Experimental Trauma Research ............................................................ 66 Experimental Immunomodulation Following Trauma-Hemorrhage ...................................................... 67 Clinical Trauma Research ..................................................................... 67 Conclusions ........................................................................................... 68 7. Cytokines in Acute Pancreatitis .............................................................. 73 James Norman and Larry C. Carey Clinical Pancreatitis and the Immune Response ................................. 73 Amplification of the Pancreatitis-Associated Inflammatory Cascade ...................................................................... 75 Local Cytokine Production During Pancreatitis.................................. 76 The Effects of Local Cytokine Production on Pancreatitis Severity ... 76 Cytokine-Induced Acinar Cell Death During Pancreatitis ................. 77 Acute Pancreatitis Induces Systemic Hyperinflammation .................. 77 Antagonism of Cytokines During Experimental Acute Pancreatitis .............................................................................. 79 Is Anti-Cytokine Therapy for Acute Pancreatitis a Clinical Possibility? ......................................................................... 79 8. Cytokines and Perioperative Nutrition ................................................. 83 Hideaki Saito and Lin Ming-Tsan Introduction ........................................................................................... 83 Nutritional Route and Cytokine Biology ............................................. 83 Nonstressed Stable Conditions ............................................................. 84 Tissue Cytokine Biology ........................................................................ 84 Stressed Conditions ............................................................................... 86 Cytokine Biology at Local Inflammatory Sites..................................... 87 Specific Nutrients and Cytokine Biology ............................................. 88 Glutamine .............................................................................................. 89

Arginine .................................................................................................. 90 Nucleotides ............................................................................................ 91 Immune-Enhancing Diets ..................................................................... 91 Anabolic Hormones And Cytokine Biology ........................................ 91 Summary ................................................................................................ 92 9. Cytokine Response to Laparoscopic Surgery ......................................... 97 Harry van Goor and R. Jan A. Goris Introduction ........................................................................................... 97 Rationale for Less Tissue Trauma and Stress Response in Laparoscopy ................................................ 97 Cytokines and Tissue Trauma .............................................................. 98 Cytokines in Laparoscopic Cholecystectomy ....................................... 99 Cytokines in Other Laparoscopic Procedures ...................................... 99 Do Circulating Cytokines Accurately Reflect Tissue Trauma? ......... 100 Cytokines and Outcome After Laparoscopic Surgery ....................... 101 Conclusions ......................................................................................... 103 10. Cytokines in Experimental Peritonitis ................................................. 107 Martin K. Angele, Alfred Ayala, and Irshad H. Chaudry Introduction ......................................................................................... 107 Choice of Models ................................................................................. 108 The Response of Cytokines in Experimental Models of Sepsis and Clinical Sepsis ............................................................ 109 Summary .............................................................................................. 113 11. Hematopoietic Cytokines, G-CSF and Abdominal Surgery ............... 117 Artur Bauhofer, Wilfried Lorenz, Ilhan Celik, Benno Stinner, José Solovera, and Ronald Lorijn Introduction ......................................................................................... 117 Preclinical Biology: Hematopoietic Growth Factors (HGFS) in Infectious Disease ........................................................................ 119 Clinical Reality: The Overall Complexity of Abdominal Infections ................................................................. 123 New Concepts: Demonstrating Effectiveness of Drugs in Clinical Trials and Daily Practice for Abdominal Infections .... 127 A Convincing Example for Increasing Clinical Complexity in Preclinical Trials: Experience with rhg-CSF (Filgrastim) ......... 130 First Explanations for rhg-CSF Positive and Negative Trials: Relevance for Designing and Conducting Definitive Clinical Trials ................................................................................... 130 Perspectives: Filgrastim in Abdominal Surgery ................................. 133 Conclusion ........................................................................................... 133 List of Abbreviations ........................................................................... 134

12. Cytokines and Spontaneous Bacterial Peritonitis ............................... 143 Theresa Propst and A. Propst Introduction ......................................................................................... 143 Pathogenesis ......................................................................................... 143 Diagnosis .............................................................................................. 145 Symptoms ............................................................................................ 145 Differential Diagnosis .......................................................................... 146 Cytokines in Chronic Liver Disease .................................................... 146 Cytokines in SBP ................................................................................. 147 Conclusion ........................................................................................... 149 13. Secondary Peritonitis and Cytokines ................................................... 151 Réne G. Holzheimer Introduction ......................................................................................... 151 Peritonitis and Cytokines in Experimental Models ........................... 151 Dosage of Cytokines ............................................................................ 152 Function of Peritoneal Cells ................................................................ 152 Local Response ..................................................................................... 152 Effect of Cytokines and Growth Factors in Peritonitis ...................... 153 Anti-Endotoxin-Antibodies ................................................................ 153 Pathogens ............................................................................................. 154 Anti-Inflammatory Cytokines ............................................................ 154 Second Hit ............................................................................................ 154 Therapy ................................................................................................ 155 Secondary Peritonitis and Cytokines in Clinical Studies .................. 155 Conclusions ......................................................................................... 156 14. Cytokines and Combination Treatment of Intra-Abdominal Infections ............................................................. 163 William G. Cheadle and Hiram C. Polk, Jr. Introduction ......................................................................................... 163 Overview of Our Current Research .................................................... 163 The Macrophage .................................................................................. 164 The Role of Endotoxin ........................................................................ 165 The Neutrophil .................................................................................... 165 Lymphocytes ........................................................................................ 166 Therapy of Intra-Abdominal Infection .............................................. 166 15. Cytokines, Abdominal Trauma and the “Second Hit” Phenomenon ..................................................... 171 Roger Saadia and Jeffrey Lippman Mutiple Organ Failure After Trauma ................................................. 171 Cytokines and the Postinjury Inflammatory State ............................ 171 “One-Hit” and “Two-Hit” Models ..................................................... 172 Priming and Activation ....................................................................... 173 Clinical Observations .......................................................................... 173 “First Hit” and Cytokine Patterns ...................................................... 174

“Second Hit” and the Inflammatory Response .................................. 175 Conclusion ........................................................................................... 176 16. Cytokines and Mesenteric Ischemia ..................................................... 179 David A. Partrick, Ernest E. Moore and Walter L. Biffl Introduction ......................................................................................... 179 Pathogenesis of Multiple Organ Failure ............................................. 179 Mesenteric Ischemia as a Model of Systemic Inflammation ............. 180 Gut-Derived Mediators of Neutrophil Priming ................................ 181 Interleukin-1b ...................................................................................... 184 Interleukin-6 ........................................................................................ 185 Interleukin-8 ........................................................................................ 188 Clinical Implications ........................................................................... 188 17. The Significance of Intestinal Cytokines ............................................. 197 Per-Olof Hasselgren Introduction ......................................................................................... 197 Cytokines in Inflammatory Bowel Disease ........................................ 197 Studies in Experimental Animals ........................................................ 202 Intestinal Cytokines During Sepsis and Endotoxemia ...................... 202 The Significance of Intestinal Cytokines ............................................ 208 18. Cytokines in Abdominal Vascular Surgery .......................................... 215 C.V. Soong and B.J. Rowlands Introduction ......................................................................................... 215 Evidence of Endotoxemia and Cytokine Production in Abdominal Vascular Surgery ...................................................... 216 The Mechanism of Production of Cytokines in Aortic Surgery ....... 217 The Effects of Elevated Cytokines ....................................................... 221 Potential Therapeutic Manipulation .................................................. 222 Conclusion ........................................................................................... 223 19. Cytokines and Abdominal Organ Transplantation ............................ 233 Stephen W. Chung, Charles H. Scudamore and Reginald Gorczynski Introduction ......................................................................................... 233 The Cytokine Response in Acute Allograft Rejection ........................ 234 Experimental Models of Acute Rejection ........................................... 236 Conclusion ........................................................................................... 240 20. Anti-Cytokine Strategies in Peritonitis: Is Local, Intra-Abdominal, Therapy Possible? .................................................................................. 245 Alfred Ayala and Irshad H. Chaudry Introduction ......................................................................................... 245 Models of Sepsis/Peritonitis ................................................................ 246 Differential Effects of Sepsis/Peritonitis of Immune Cells ................ 247 Potential Intra-Abdominal Pro- or Anti-Inflammatory Cytokine Therapies? ........................................................................ 256 Summary .............................................................................................. 263

21. The Antibiotic-Induced Cytokine Response ....................................... 273 Timothy G. Canty Jr., Edward Boyle Jr., and E. Patchen Dellinger Introduction ......................................................................................... 273 Antibiotics, Bacterial Modulins, and Cytokines ................................ 273 Antibiotic Induced Release of Bacterial Modulins: History .............. 274 Background .......................................................................................... 275 Antibiotic Induced Release of Bacterial Modulins: Endotoxin ......... 275 Antibiotics and Endotoxin (In Vitro) ................................................ 276 Antibiotics and Endotoxin (Animal Studies) .................................... 276 Antibiotics and Endotoxin (Clinical Studies) .................................... 277 Do Antibiotics Differ in Potential for Endotoxin Release? ............... 277 Cytokine Stimulating Microbial Components Other Than LPS ...... 279 Cytokine Synthesis and Antibiotics .................................................... 281 Conclusion ........................................................................................... 281 22. Cytokines and Pentoxifylline in Surgical Sepsis .................................. 285 J. Schröder, K.H. Staubach, F. Stüber and P. Zabel Introduction ......................................................................................... 285 Long-Term Administration of Pentoxifyllin in Surgical Sepsis ........ 286 Effects of Pentoxifylline on Multiple Organ Dysfunction ................. 288 Cytokines and Pentoxifylline .............................................................. 290 Perspectives .......................................................................................... 291 23. Epilogue ................................................................................................. 295 Moshe Schein and Leslie Wise Introduction ......................................................................................... 295 Infection vs Inflammation and Sepsis ................................................ 295 Current Terminology in Clinical Sepsis ............................................. 296 Local Inflammation vs Contamination, Infection or Sepsis ............. 297 Contamination, Infection, Sepsis: A Continuum .............................. 298 Clinical Relevance ................................................................................ 298 Index ....................................................................................................... 303

EDITORS Moshe Schein, MD, FCS (SA) Associate Professor of Surgery, Cornell University College of Medicine New York Methodist Hospital Brooklyn, NY Chapters 1, 23 Leslie Wise, MD, FRCS (Eng) Professor of Surgery, Cornell University College of Medicine and Albert Einstein College of Medicine Chairman, Department of Surgery Brooklyn, NY Chapters 1, 23

CONTRIBUTORS Shaun G. Appleton, FRCS Department of Surgery Chelsea and Westminster Hospital London, UK Chapter 5

Philiph Barrie, MD Professor of Surgery Cornell University College of Medicine New York, NY Chapter 2

Martin K. Angele, MD Research Associate Center for Surgical Research and Department of Surgery Brown University School of Medicine and Rhode Island Hospital Providence, RI Chapter 10

Artur Bauhofer, PhD Institute of Theoretical Surgery, University Marburg Marburg, Germany Chapter 11

Alfred Ayala, PhD Associate Professor Center for Surgical Research and Department of Surgery Brown University School of Medicine and Rhode Island Hospital Providence, RI Chapters 10, 20 Adrian Barbul MD Professor of Surgery, The Johns Hopkins University School of Medicine Baltimore, MD Chapter 4

Walter L. Biffl, MD Department of Surgery University of Colorado Health Sciences Center Denver, CO Chapter 16 Edward Boyle, Jr. MD Department of Surgery, University of Washington Seattle, WA Chapter 21 Steve E. Calvano, PhD UMDNJ-Robert Wood Johnson Medical School Department of Surgery New Brunswick, NJ Chapter 3

Timothy G. Canty, Jr. MD Department of Surgery University of Washington Seattle, WA Chapter 21

Steven M. Cohen, DO Research Fellow Department of Surgery, Cornell University Medical College Chapter 2

Larry C. Carey, MD Professor and Chairman Department of Surgery University of South Florida Tampa, Florida Chapter 7

E. Patchen Dellinger, MD Professor of Surgery Department of Surgery University of Washington Seattle, WA Chapter 21

Ilhan Celik, MD Institute of Theoretical Surgery University of Marburg Marburg, Germany Chapter 11

Eugen Faist, MD Professor or Surgery Ludwig-Maximillians-Universitat, Klinikum Grosshardern Munich, Germany Chapter 6

Irshad H. Chaudry, PhD Professor of Surgery, Molecular Pharmacology, Physiology and Biotechnology Director, Center of Surgical Research Brown University School of Medicine and Rhode Island Hospital Providence, RI Chapters 10, 20 William G. Cheadle, MD Associate Professor of Surgery Veterans Affairs Medical Center, Price Institute for Surgical Research, and the Department of Surgery, University of Louisville School of Medicine Louisville, KY Chapter 14 Stephen W. Chung, MD, PhD Department of Surgery Vancouver Hospital and Health Sciences Centre Vancouver, BC, Canada Chapter 19

R. Jan A. Goris, MD, PhD Professor and Chairman Department of Surgery, University Hospital Nijmegen, Nijmegen, The Netherlands Chapter 9 Reginald Gorczynski, PhD, MD Departments of Immunology University of Toronto Toronto, Canada Chapter 19 Per-Olof Hasselgren, MD Professor of Surgery University of Cincinnati Medical Center Cincinnati, OH Chapter 17 Réne G. Holzheimer, MD, PhD Department of Surgery Martin-Luther-Universität HalleWittenberg Halle, Germany Chapter 13

Edward Lin, DO Department of Surgery The New York Hospital Queens Flushing, New York Chapter 3 Jeffrey Lippman, MB BCh, FFA (SA) Associate Professor-Division of Anesthesiology and Intensive Care The Royal Brisbane Hospital and University of Queensland Brisbane, Australia Chapter 15 Wilfried Lorenz, MD Professor and Director Institute of Theoretical Surgery University Marburg Marburg, Germany Chapter 11 Ronald Lorijn MD, PhD AMGEN-Europe Lucerne, Switzerland Chapter 11 Stephen F. Lowry, MD Professor and Chairman, Department of Surgery UMDNJ-Robert Wood Johnson Medical School New Brunswick, NJ Chapter 3 Lin Ming-Tsan, MD Department of Surgery The University of Tokyo Tokyo, Japan Chapter 8 Ernest E. Moore, MD Professor of Surgery Department of Surgery University of Colorado Health Sciences Center Denver, CO Chapter 16

James Norman, MD Associate Professor of Surgery and Medicine Department of Surgery University of South Florida Tampa, Florida Chapter 7 David A. Partrick, MD Department of Surgery University of Colorado Health Sciences Center Denver, CO Chapter 16 Hiram C. Polk, Jr, MD Professor and Chairman Veterans Affairs Medical Center, Price Institute for Surgical Research, and the Department of Surgery, University of Louisville School of Medicine Louisville, KY Chapter 14 Theresa Propst, MD Department of Internal Medicine University of Innsbruck, Austria Chapter 13 Albert Propst, MD Department of Internal Medicine University of Innsbruck, Austria Chapter 13 Brian J. Rowlands, MD, FRCS Professor of Surgery Queen’s Medical Centre Nottingham, UK Chapter 18 Roger Saadia, MD, FRCS Professor of Surgery, University of the Witwatersrand Chief Surgeon-Baragwanath Hospital Johannesburg, South Africa Chapter 15

Charles H. Scudamore, MD, Msc Department of Surgery Vancouver Hospital and Health Sciences Centre Vancouver, BC, Canada Chapter 19 Hideaki Saito, MD Associate Professor, Department of Surgery The University of Tokyo Tokyo, Japan Chapter 8 J. Schröder, MD Department of General and Thoracic Surgery The Christian-Albrechts-University Hospital Kiel, Germany Chapter 22 José Solovera MD, PhD AMGEN-Europe Lucerne, Switzerland Chapter 11 C.V. Soong, MD, FRCS Senior Registrar Department of Surgery Queen’s University of Belfast Belfast, UK Chapter 18 F. Stüber, MD Department of General and Thoracic Surgery The Christian-Albrechts-University Hospital Kiel, Germany Chapter 22

K.H. Staubach, MD Department of General and Thoracic Surgery The Christian-Albrechts-University Hospital Kiel, Germany Chapter 22 Benno Stinner, MD Department of General Surgery University Marburg Marburg, Germany Chapter 11 Jeremy N.Thompson, MChir, FRCS Department of Surgery Chelsea & Westminster Hospital London Chapter 5 Frank J. Thornton, MB, FRCS(I) Department of Surgical Research, Sinai Hospital of Baltimore and the Johns Hopkins Medical Institutions Baltimore, MD Chapter 4 Harry van Goor, MD, PhD Department of Surgery, University Hospital Nijmegen The Netherlands Chapter 9 Matthias W. Wichmann, MD Ludwig-Maximillians-Universitat Klinikum Grosshardern Munich, Germany Chapter 6 P. Zabel, MD Department of General and Thoracic Surgery The Christian-Albrechts-University Hospital Kiel, Germany Chapter 22

1

CHAPTER 1

Introduction—Cytokines and the Abdominal Surgeon Moshe Schein and Leslie Wise

R

ecent advances in molecular biology permitted the identification of various cytokines as the key “proximal” links in the infectious-inflammatory cascades, mediating local and systemic inflammation, sepsis, and tissue injury. A huge body of data concerning cytokines is available-a MEDLINE search today (December 12, 1997) generated 31897 citations mentioning the word cytokines. We, clinicians find ourselves overwhelmed by such abundant information, understanding less about more. This book was conceived to epitomize and simplify the most complex and confused known facts which imply that cytokines have an important role in many conditions which are commonly encountered by the abdominal surgeon. The authors for the various chapters were chosen based on their previous contributions to the specific topic to be discussed. Both basic researchers and clinicians were selected in order to provide a balanced picture, “tolerated” by scientists and attractive to clinical surgeons. Some repetition and overlapping in the chapters were unavoidable as this is a multi-author book. In chapter 2, Drs. Barie and Cohen from New York, discuss the relevance of molecular biology to the surgeon. Drs. Lin, Calvano and Lowry from New Jersey offer, in chapter 3, a detailed overview on the cytokine-response following abdominal operations. They emphasize that the extent and complexity of surgery, together with the adequacy of the peri-operative care, are important determinants of the ensuing inflammatory response. The role of cytokines in the healing of surgical and nonsurgical wounds is discussed, in chapter 4, by Drs. Thornton and Barboul from Baltimore, who also highlight possible future clinical implications in complicated or compromised wounds. In chapter 5, Messers Appelton and Thompson, from London, deal with the role played by cytokines in the formation of postoperative peritoneal adhesions; cytokine-manipulation to control adhesions appears an attractive future option. The next chapter—no. 6—by Dr. Wichmann and Professor Faist from Munich review the data concerning the effects of trauma on the cytokine network. In chapter 7, Drs. Norman and Carey from Tampa tell us all what is known about cytokines in acute pancreatitis, predicting exciting future therapeutic implications. Drs. Saito and Ming-Tsan from Tokyo (chapter 8) discuss how the different peri-operative nutrition modalities modulate the cytokine-response. In chapter 9, Dr. Van Goor and Professor Goris from the Netherlands summarize the (reduced) cytokine response associated with laparoscopic-minimally invasive surgery. Drs. Angele, Ayala and Chaudry Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

2

Cytokines and the Abdominal Surgeon

from Rhode Island (chapter 10) synthesize what is known about cytokines in experimental peritonitis-offering clinical relevance. In chapter 11, Drs. Bauhofer, Celik, Stinner, Solovera and Lorijn, and Professor Lorenz from Marburg, discuss the potentials of hematopoietic cytokines and G-CSF in the modulation of “surgical sepsis”, emphasizing the complexity of clinical scenarios as opposed to experimental settings. Drs. Propst and Propst from Innsbruck— our only “nonsurgeon” contributors—discuss in chapter 12 the role of cytokines in spontaneous bacterial peritonitis; practical diagnostic and prognostic issues are included. In chapter 13, Dr. Holzheimer from Halle (Germany) focuses on the role of cytokines in secondary bacterial peritonitis, emphasizing the intraperitoneal compartmentalization of the inflammatory response. In chapter 14, Drs. Cheadle and Polk from Louisville sum up their research on the future “combination treatment” of intra-abdominal infections in which anticytokine regimens would be included. Professors Saadia (Johannesburg) and Lippman (Brisbane, Australia) discuss in chapter 15 the issue of cytokines related to the “second hit” phenomenon, best studied hitherto in the settings of multiple trauma. Drs. Patrick, Moore and Biffl from Denver write in chapter 16 about cytokine-related events occurring during mesenteric ischemia. Dr. Hasselgren from Cincinnati (chapter 17) details the significance of intestinal cytokines-locally and systemically. In chapter 18, Mr. Soong from Belfast and Professor Rowlands from Nottingham encapsulate the complex inflammatory events that follow major abdominal aortic surgery, emphasizing the many remaining uncertainties. Chapter 19, by Drs. Chung, Scudamore (Vancouver), and Gorczynski (Toronto), is dedicated to the role of cytokine—be it surveillance of rejection or its prevention—in the field of transplantation. Drs. Ayala and Chaudry, in chapter 20, return to the topic of peritonitis, asking whether local, intra-abdominal anticytokine therapy is possible. In chapter 21, Drs. Canty, Boyle and Dellinger from Seattle consider the significance of the antibioticinduced cytokine response. The last contribution, by Drs. Schroder, Staubach, Stuber and Zabel from Kiel (Germany) discuss their experiments in modulating cytokineresponse in “surgical sepsis” with pentoxifylline. And finally, in chapter 23 the editors “conclude” the discussion, emphasizing the current terminology of local and systemic inflammation versus infection. It is hoped that this book will provide the surgeon with a compact source to learn how cytokines apply to his or her practice and what may be their current and future clinical relevance.

Relevance of Molecular Biology for the Abdominal Surgeon

3

CHAPTER 2

Relevance of Molecular Biology for the Abdominal Surgeon Philip S. Barie and Steven M. Cohen

Introduction

R

ecent advances in cellular and molecular biology, especially the ability to study cell-cell and ligand-receptor interactions, have increased our understanding of the pathophysiology of disease. Important advances have occurred in sepsis and shock, ischemia/reperfusion injury, transplantation immunology, wound healing, atherogenesis, thrombosis and hemostasis, and carcinogenesis. At the same time, the practice of surgery is changing fundamentally. Minimally-invasive abdominal surgery is now routine for many procedures, and developing rapidly for others. This evolution is part of a long-term trend that has seen many procedures become the province of nonsurgeons (e.g., therapeutic endoscopy by gastroenterologists, percutaneous drainage and endovascular prosthetics by interventional radiologists). By extension, it is plausible that the surgery of the future may be a minimally-invasive procedure for gene transfer. The very nature of such procedures makes it likely that nonsurgeons will be very interested in performing such procedures. Molecular biology and clinical medicine are merging (the discipline is already called molecular medicine by some). To continue to provide patient care in an era of gene therapy and biotechnology, the surgeon must understand the principles of molecular biology in order to retain a fundamental understanding of biological processes. This is not a fundamental change in philosophy, but rather it is true to the heritage of surgical education and that has made “surgical basic science” the cornerstone of training and practice for generations. Surgeons who are knowledgeable and prepared will have ample opportunity to become involved in molecular medicine and gene therapy. The surgeon can make a unique contribution to the field. Direct access to diseased tissue provides understanding of the physiologic and anatomic relationships in disease which are essential prerequisites to structural and molecular analysis.1,2 Some diseases, such as oncology, cardiovascular disease, and digestive diseases are areas where specialty “turf ” is becoming anachronistic and the surgeon has evolved to a vital role in a collaborative practice model. Certain disorders are truly surgical diseases, and probably will remain so. Many aspects of infection and the host response; metabolism, nutrition, and wound healing; trauma and burns; organ transplantation; and repair of congenital anomalies: Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

4

Cytokines and the Abdominal Surgeon

Each are examples where surgical investigation and therapy remains paramount. Because of the morbidity and mortality associated with these disorders, and since no other specialty so often encounters these problems in a clinical setting, it behooves surgeons to pioneer the investigation of these surgical disorders by a detailed molecular analysis which will allow development of more effective therapies including strategies based on gene therapy. Many researchers have argued convincingly that the reason for failure of many of the “promising” therapeutic trials for disease processes such as sepsis, is due to an insufficient understanding of the complexity on a molecular basis.1

Structure and Function of DNA and Chromosomes For the practicing surgeon, knowledge of the molecular mechanisms by which a differentiated cell develops and maintains its specialized function is imperative to a more in-depth understanding of human disease and potential treatment. The surgeon’s understanding of the field of molecular biology must have a fundamental basis. It is imperative to understand how inherited information is encoded within DNA, and how this information is regulated and expressed. A variety of new molecular techniques, such as the polymerase chain reaction (PCR), have simplified the investigative work by identifying genes for sequencing and characterizing individual regions of the chromosome.2 Likewise, recombinant DNA technology has made available a number of naturally occurring substances, such as cytokines, which are being studied extensively as potential pharmacotherapeutic agents due to their ability to modulate a variety of host responses.3 Deoxyribonucleic acid (DNA) is the material that encodes all inherited information in normal eukaryotic cells. In the chromosomes, DNA consists of a double helical structure, which allows for the direct copying or replication of genetic instructions.4 During replication, the intertwined strands of DNA are unwound, separated, and then copied by an enzyme known as DNA polymerase, which uses one strand as a template for the synthesis of its complementary partner. The copying is always in a 5' to 3' direction. The only variable part of the DNA strand is the series of four bases along the deoxyribose-phosphate backbone. The sequence of bases in the gene encodes instructions for synthesis of a specific peptide chain. A certain codon (three nucleotide triplets) denotes which regions of genes are to be translated.5 A gene must contain the instructions about the amino acid chain (protein) to be made (the codon or coding region), and it must also include sequences that regulate transcription of a gene which will control the expression of the protein product and ensure that the protein is produced at the correct time in cellular development and differentiation, and that it is made in the correct tissues and in the right amount. Examples of these gene regulatory sites are promoters, or upstream promoter elements, which are regions of DNA to which RNA polymerases or transcription factors bind and initiate RNA synthesis.6 The central dogma of molecular biology describes the flow of genetic information within the cells from DNA to RNA (transcription), and RNA to protein (translation). The completed protein that is released following translation usually must undergo a considerable amount of posttranslational modification before it becomes functional. These modifications include phosphorylation, glycosylation, and cleavage.7

Human Genome Project

Substantial efforts are underway to characterize the human genome.8 The human genome is believed to harbor up to 100,000 genes, of which about one-half have

Relevance of Molecular Biology for the Abdominal Surgeon

5

been sampled to date in the form of expressed sequence tags.6,8,9 More than 16,000 human genes have been mapped relative to a framework map that contains about 1000 polymorphic genetic markers.10 Only a small fraction of the total DNA in the human genome (probably less than 10%) consists of coding information for proteins.11 Recombinant deoxyribonucleic acid (DNA) technology and molecular cloning allow the recovery of multiple copies of a segment of DNA, producing large amounts of material for analysis.9 The gene map will unify existing genetic and physical maps with the nucleotide and protein sequence databases in a fashion that should speed the discovery of genes underlying inherited human disease.10 The human genome project will complete the Herculean task of sequencing the entire human genome early in the 21st century. The human genome contains more than 3 billion nucleotides; the structural detail of their organization will be known precisely. Once known, it is planned that the human neurobiology project will define the function of the approximately 70,000 genes that are expressed in the human brain. There are currently at least 68,000 genes in the brain about which little or nothing is known.11 That is just the beginning, and just the brain. Considering what little we know about cancer genetics and how it already has profound implications for patient care, the enormity and potential of molecular medicine is awesome.

Molecular Diagnostic Tests Classical diagnostic tests (from the clinical laboratory of virtually any facility) target many different ions and macromolecules (enzymes, substrates, metabolites, etc.) at various levels (from cellular to the intact organism). In contrast, molecular diagnostic tests essentially target only two moieties, epitomes and nucleic acid sequences.12 An epitope is a unique space/change occupying region within a parent molecule, for binding by a monoclonal antibody. The epitope may occur singly or repetitively within the same molecule. Because epitopes may change conformation and hence antibody specificity, antibodies can be used not only to identify molecules, cell types, or secretory products, but may also define functional regions or even activate the cell. Information about nucleic acid sequences relates to the precise order of the ribose or deoxyribose sugar bases. Such information can be obtained through one of three methodologies: Annealing (hybridization between two strands of DNA and/or RNA; interactions between nucleic acids and recognition enzymes (restriction/hydrolysis enzymes); or nucleic acid sequencing methodologies (which are cumbersome, expensive, and seldom used). Restriction endonucleases are enzymes that occur naturally, mainly in bacteria, and that cleave foreign DNA. These restriction enzymes, of which more than 400 (with 100 different specificities) have been isolated, are in regular use for recombinant technology. Additionally, the insertion of human DNA into bacterial plasmids or bacteriophages is a keystone of recombinant DNA technology.13 Molecular cloning is the isolation of a specific segment of DNA (such as a gene or part of a gene) and the generation of many identical copies, or clones, of that segment of DNA. Because a given segment of DNA cannot replicate itself, the DNA segment is joined to a vector (or carrier) DNA molecule which is able to replicate itself and can replicate the DNA segment that has been joined to it. The DNA segment, with its vector, is a recombinant DNA molecule.5 In order to provide the necessary milieu for DNA replication, recombinant DNA molecules are inserted into host cells, usually bacteria such as E. coli, which become factories to reproduce large quantities of the recombinant DNA molecule.9

6

Cytokines and the Abdominal Surgeon

Once a particular DNA sequence is isolated, a set of recombinant DNA molecules is constructed from a source that contains the sequence of interest. This set of clones of complimentary DNA (called a cDNA library) should contain all sequences contained in the original source (genomic DNA or messenger RNA). Once a library is constructed, it is screened to identify the specific clone containing the DNA sequence of interest. The most commonly used screening method of a library is nucleic acid hybridization. Once a gene is cloned and sequenced, dysregulation can be studied by a number of techniques. Genes cloned from infectious pathogens can be used for rapid diagnosis of infection. Tumor-specific genes such as oncogenes can be developed into reagents for identification of tumor cells. Cloned genes can be introduced, deleted, or expressed variably in microorganisms, cells, tissues, or intact animals. Recombinant human proteins are already in use therapeutically (e.g., insulin, erythropoietin), and can be used, via hybridoma techniques, to develop monoclonal antibodies for diagnosis.

Protein Hybridization The first recombinant DNA technology was the Southern blot, which can be used for the detection and characterization of a single gene at a concentration ratio of 1:2x107. High molecular weight genomic DNA is purified and then digested with a specific restriction enzyme.14,15 Each lane of a Southern gel (see below) consists of identical DNA digested with several different enzymes, whereas for comparison purposes several different samples are digested with the same enzyme. The DNA fragments are separated on agarose (a molecular sieving gel), denatured into single strands with NaOH, and transferred (blotted) to an immobilizing membrane (nitrocellulose or nylon). The hybridization probes are either cloned gene fragments, their RNA transcripts, or short sequence-specific synthetic oligonucleotides, labeled radioactively after denaturation. Hybridizations are carried out under exacting conditions of anion concentrations, temperature, and time, to favor reassociation of only complementary DNA. The Southern analysis forms the basis of the restriction fragment length polymorphism (RLFP) analysis. Genetic mutations lead to new, abnormally sized restriction fragments. A point mutation that creates a new restriction site between two others causes the disappearance of the parent restriction fragment and the appearance of two new, smaller fragments. The converse is also true. To perform RLFP analysis, test genomic DNA is digested, Southern blotted, and hybridized. Polymorphisms in restriction fragment length become diagnostic indicators of genetic polymorphism. Although the Southern and RLFP methodologies have been largely supplanted by the polymerase chain reaction, they remain useful for the identification of immunologic diseases, infectious disease diagnosis, tissue typing for transplantation, and innumerable inherited disorders. The Northern blot is a simple test that allows identification of the mRNA transcripts of a single gene.14 An aliquot of RNA is purified from a tissue source, separated and denatured by an agarose gel migration, and blotted on to nitrocellulose. Specific species are hybridized to labeled probes as described above. A control blot is performed to detect a gene that is expressed constitutively, such as !-actin, so as to control for undegraded RNA loaded on to the gel. Abnormalities of mRNA production characteristic of several neurologic and infectious diseases can be identified. Northern blots are limited by the fact that relatively large amounts of RNA are re-

Relevance of Molecular Biology for the Abdominal Surgeon

7

quired for detection of single gene transcripts. Some tissue samples (i.e., needle biopsies) may be inadequate for the purpose. Weakly expressed genes are difficult to detect. Northern analysis has largely been supplanted by nuclease protection analysis or the reverse transcriptase polymerase chain reaction (RT-PCR). The PCR is a rapid procedure for rapid in vitro enzymatic amplification of a specific segment of DNA.6 The technology relies heavily on the unique thermodynamic properties of AT and CG base pairs. Armed with even limited information about the sequence information of a protein or gene, very short DNA oligonucleotides complimentary to sequences on the ends of opposite DNA strands of the gene of interest are synthesized, incubated with DNA, free deoxyribonucleotide triphosphates, and purified heat-stable DNA polymerase (Taq). Heat denaturation anneals the oligonucleotide primers perfectly to the complementary sequences on singlestranded DNA. Then, Taq catalyzes the synthesis and elongation of both strands of the gene of interest in opposing directions. Repetitive heating and cooling allow the reaction to repeat indefinitely with automated equipment. When sufficient replication has occurred, direct visual identification with ultraviolet light is possible. As with molecular cloning, PCR has spawned a multitude of experiments that were previously impossible. The number of applications of PCR continues to grow, including direct cloning from genomic DNA or cDNA, in vitro mutagenesis and engineering of DNA, genetic fingerprinting of forensic samples, assays for the presence of infectious agents, prenatal diagnosis of genetic diseases, analysis of allelic sequence variations, analysis of RNA transcript structure, genomic footprinting, and direct nucleotide sequencing of genomic DNA and cDNA.16,17 The sensitivity and power of PCR has allowed for its use in medical genetics, cancer detection, organ transplantation, virology, and specific diagnosis of infectious diseases.18 In particular, PCR has revolutionized the approach to the study of cancer. Direct examination of genomic changes that occur during tumor initiation, progression, and metastasis is now possible.8 The main drawback of PCR technology is its extraordinary sensitivity; falsepositives are possible due to heterologous annealing of primers, or laboratory contamination of even minuscule amounts of DNA. The sensitivity of PCR can be applied to high-resolution identification and quantitation of gene expression by incorporation of a “reverse transcription” step early in the reaction (RT-PCR). Reverse transcription is so named because it is the transcription of DNA from an RNA template. A natural process in the life cycle of retroviruses, it is catalyzed by a family of retroviral enzymes known as reverse transcriptases. In RT-PCR, total RNA from a cell or tissue sample is extracted by one of several standard approaches, then reverse-transcribed into cDNA using primer oligonucleotides. Under ideal conditions, a single cDNA molecule is transcribed off of every mRNA molecule so that quantitative analysis by subsequent PCR is possible. The sensitivity of RT-PCR for detection of infinitesimally low levels of normal gene expression allows for the ready detection of upregulation of pathologic gene expression in disease. The RT-PCR-based estimation of cytokine gene expression has become an important diagnostic tool for the monitoring of chronic inflammatory conditions and the response to therapy.19,20 The enzyme-linked immunoabsorbent assay (ELISA) is the most-used diagnostic application for monoclonal antibodies, being a simple, rapid, and relatively inexpensive analysis for almost any antigenic epitope. There is a high degree of reproducibility, and the technology lends itself well to automation. The radioimmunoassay (RIA), which is a more sensitive and specific but also more resource-intensive assay,

8

Cytokines and the Abdominal Surgeon

is essentially an ELISA with a radiolabeled reagent. There are two steps in the performance of ELISA, antigen immobilization and subsequent identification and quantitation. Immobilizing monoclonal antibodies are coated on to polystyrene plates, while any remaining protein-binding sites on the plates are blocked with albumin. Serial dilutions of the test sample are then allowed to react with the antibody on the plate, and compared to a standard curve of highly-purified test antigens developed in parallel. After washing to remove any unbound antigen, a second monoclonal antibody that is usually directed against a different epitope on the test antigen is added. The second antibody is linked covalently to an enzyme, such as peroxidase or alkaline phosphatase, which then catalyzes a colorimetric reaction once its own substrate and reagents are added. The color reaction is thus directly proportional to the amount of enzyme product present. Thus, ELISA was the first molecular test to use products of hybridoma (monoclonal antibodies) and recombinant DNA (purified reagents) technology; these tests are now in the mainstream of the clinical laboratory.12 The Western blot is even more informative than the ELISA, because the target molecules are first separated on the basis of size.21,22 Size separation is accomplished by electrophoresis in polyacrylamide gels after denaturation with heat and sodium dodecyl sulfate (SDS), which places a negative charge on the protein.21 Treated proteins then migrate through an electrified gel purely on the basis of molecular weight. Reduction with 2-mercaptoethanol, which hydrolyzes disulfide bonds, unfolds proteins and facilitates migration. Separation of proteins based on their isoelectric point in a pH-gradient gel, wherein migration ceases at the isoelectric point when the negative charge is neutralized, is also possible. Isoelectric focusing in one direction, combined with denaturation electrophoresis in the opposite direction, provides for a two-dimensional separation that is virtually unique for most proteins. Once the protein is separated, immunologic identification is carried out. Protein is transferred either electrophoretically or osmotically from the separation gel to a nitrocellulose or nylon membrane, which adsorbs protein avidly. After nonspecific protein blocking, a monoclonal antibody or antiserum is introduced that is specific for the protein in question. An enzyme-labeled second antibody that recognizes the first reagent is then introduced, and the target protein band can be visualized. Incorporation of molecular weight markers and test substrate in parallel electrophoretic and transfer steps makes for positive identification of the test protein. However, some protein epitopes cannot be identified by Western blotting because integrity of the specific epitope is dependent on the native protein configuration. Protein disruption by SDS may make antibody recognition impossible. The demanding technology means that Western blotting is usually performed as a confirmatory test after an initial screening ELISA is positive. This sequence of tests is used clinically for testing for the human immunodeficiency virus.

Programmed Cell Death With the possible exception of cancer genetics, programmed cell death (apoptosis) has generated more interest in biological investigation than any other phenomenon. Apoptosis has been implicated in a host of physiologic processes and disease states, and its potential for regulation as a therapeutic strategy is enormous. Apoptosis is central to the regulated death of cells during normal embryonic development. In adults, physiologic cell death figures prominently in cyclically-stimulated or hormone-dependent tissues such as the endometrium, prostate, adrenal, and mammary gland, as well as in “steady-state” turnover in many tissues, such as small bowel

Relevance of Molecular Biology for the Abdominal Surgeon

9

enterocytes.23 Regulated cell death is fundamental to the development, regulation, and function of the immune system. Cell death in viral infection, the pro-inflammatory host response to infection or major tissue injury, AIDS, neurodegenerative diseases, myocardial ischemia, and a host of malignant tumors. Apoptosis is an innate process by which cells systematically inactivate disassemble, and degrade their own structural and functional components. It can be activated intracellularly by constituitive processes, or activated by extracellular stimuli such as proteins, cytokines, hormones, radiation, oxidative stress, or hypoxia. Control of cell death programs is physiologically stringent, and dysregulated apoptosis may result in excessive, untimely, or inadequate responses. Upregulated apoptosis of neuronal cells may lead to neurodegenerative disease, or severe immunodeficiency syndromes where T-lymphocytes are the target cell. Conversely, suppressed apoptosis may account for chronic viral infections, autoimmune diseases, lympho-proliferative disorders, and neoplastic transformation. From the perspective of the gastrointestinal tract, apoptosis has been implicated in a long list of pathologic entities.24 Apoptosis has been associated with peptic ulcer disease and atrophic gastritis. Numerous hepatocyte diseases have been related to apoptosis, including cholestasis, toxicity from ethanol and other poisons, and viral infections, whereas cholangiocyte-related pathology includes sclerosing cholangitis and primary biliary cirrhosis. Apoptosis has been implicated in the pathogenesis of both acute and chronic pancreatitis, malabsorption syndromes, inflammatory bowel disease, and gastroesophageal reflux disease. Carcinoma of the esophagus, stomach, pancreas, colon, liver, and bile duct have been implicated. Failed apoptosis of sloughed tumor cells may be central to the pathogenesis of metastasis, and for unresponsiveness to antitumor chemotherapy. Apoptosis occurs in distinct stages.23,25 In commitment, the cell becomes irreversibly committed to death after a lethal stimulus has been received. The execution phase is when recognizable structural changes occur. Nuclear chromatin condenses and aggregates as DNA is fragmented progressively. Mitochondrial transmembrane potential is decreased, the electron transport mechanism is uncoupled, and oxygen radical production is increased. The nuclear structure is altered, and mitochondrial function is impaired. In the cytoplasm, protein crosslinking occurs, cytoskeletal filaments aggregate, and the endoplasmic reticulum fuses with the plasma membrane, compromising cellular integrity. The cell becomes spherical, loses contact with its neighbors, shrinks, and projects protruberances known as apoptotic bodies. Phagocytic cells engulf apoptotic cells avidly in the clearance phase. The entire process takes only a few hours to complete. In many ways, apoptosis is controlled by cytokines.26 Two members of the tumor necrosis factor (TNF) family of cell surface signaling molecules, Fas/APO-1/CD95 and the soluble TNF receptor TNFR-1, are potent inducers of apoptosis upon stimulation of their respective ligands, Fas ligand and TNF.27 Fas signaling, which is of critical importance in apoptotic regulation of the immune system, is in turn mediated by known effectors of the death pathway known as interleukin-1b converting enzyme (ICE)-like cysteine proteases. The ICE family proteases are several, including CPP32/Yama/Apopain, which is a mediator of the heat-stress or heat-shock protein response. Proteolysis of several vital cellular proteins, such as those involved in DNA repair, cytoskeletal integrity, and cell signaling, is mediated by ICE-like proteases. Proteolytic cleavage may activate nascent protein functions necessary to execute the death pathway, inactivate countervailing growth signals, and facilitate cellular destruction.23

10

Cytokines and the Abdominal Surgeon

There appears to be a critical integration between regulation of apoptosis and neoplastic transformation. Perhaps best-studied is the apoptosis-promoting function of p53, which exerts its growth-suppressing activity by function as a transcription factor. The most intriguing p53-responsive apoptosis-promoting gene product identified thus far is the Bax protein. The Bax protein was identified for its ability to antagonize the apoptosis-suppressive effects of the Bcl-2 oncogene, which was first identified at the chromosomal breakpoint of certain follicular B-cell lymphomas. Bcl-2 is only one of many apoptosis-regulating genes, but it is unique in that it exerts its effects to promote cell survival by blocking apoptosis rather than promoting proliferation, probably through an antioxidant mechanism of its protein product. Cell death or survival may depend on the ratio of Bcl-2 to Bax gene product. Loss of apoptotic potential, by mutation of p53 or abnormal expression of genes such as Bcl-2, is becoming recognized as an important factor in the development of tumors, by permitting the survival of DNA-damaged cells. Bcl-2 is expressed at the base of colonic crypts.24 Expression is common in colonic dysplasia, adenomas, and carcinomas, suggesting that Bcl-2 alterations occur early in colonic carcinogenesis. Interestingly, Bcl-2 is not expressed in the small intestine epithelium whereas Bax is, which may account for the paucity of small intestine adenocarcinomas. Moreover, cyclooxygenase-2 (COX-2), an isoform that suppresses apoptosis when overexpressed, can be suppressed by sulindac, a nonsteroidal antiinflammatory drug. Sulindac therapy leads to regression of adenomatous polyps in patients with familial adenomatous polyposis, making it possible that drug therapy can restore or promote apoptosis in a therapeutically beneficial way.

Genetic Predisposition to Disease: Preparing for the Genetic Revolution In cancer genetics, most individual genes, once mutated, predispose patients to neoplastic transformation rather than acting as the direct cause. Several mutations may be necessary for neoplastic transformation; some may be acquired through life as somatic mutations, whereas others may be inherited, as is the case with a tumor suppressor gene. The National Cancer Institute has embarked on the Cancer Genome Anatomy Project to define all genes relevant to all malignant tumors.28 Methods for identification of persons at risk to develop cancer are an expected result.29 Examinations of several genes have already had profound implications.30,31 In the case of BRCA1 gene mutations, which have been associated with increased susceptibility to breast and ovarian carcinoma,30 such susceptibility will no longer be determined on the basis of family history, which cannot distinguish between sporadic clustering or inherited predisposition. Risk can be determined with precision; what could be more reassuring than the knowledge that one does not carry inherited increased susceptibility to breast or ovarian cancer? On the other hand, what could be more disconcerting than the knowledge that one does? The availability of such genetic testing raises issues of counseling and confidentiality. Counseling is critical for issues of screening, prophylaxis of disease, and even family planning. Confidentiality is critical in an era where the electronic dispersion of information has made security of medical records of concern. With the possibility of a person’s genome on a microchip already being discussed,32 confidentiality is a major issue. Imagine a scenario in which an employer, about to make a long-term employment commitment for a position of great responsibility, decides that the prospective employee’s medical risk is too great to justify the hiring.

Relevance of Molecular Biology for the Abdominal Surgeon

11

Screening is an integral part of cancer management for the surgeon, and all surgeons involved in oncologic care must be prepared to discuss screening and risk factors with their patients. Referral to a genetic counselor is not always possible, because the number of trained counselors and medical geneticists is inadequate to meet anticipated needs.33 Evidence suggests that without continuing education, the likelihood of misinterpretation of currently-available genetic tests is high.34 Nearly onethird of physicians who received results of a genetic test that detects mutations that cause familial adenomatous polyposis (a known predisposition to colon carcinoma) misinterpreted the test results.34 Equally of concern, only 19% of the patients being tested had received genetic counseling before the test, and only 17% provided written informed consent, implying that many patients were unaware of the discriminatory potential of testing. The elements of patient preparation for genetic testing have been described.35 The personal and family history remain essential, and a physical examination is performed to identify stigmata of already-present disease. The natural history of the suspected disease is reviewed with the patient. The predictive value of the test must be discussed, as well as any other genetic or nongenetic conditions that might influence the onset or severity of the condition. Given that the testing allows the patient and the physician to glimpse the future in the medical sense, the risks, benefits, and limitations of testing must be disclosed. The medical, behavioral, and social implications of positive, negative, and indeterminate tests must be discussed. Available tests and strategies for ongoing surveillance in test-positive subjects should be outlined. Confidentiality is paramount. Persons need to know beforehand to whom (physicians, relatives, insurers, etc.) the results will be disclosed. The logistics of the testing procedure must be disclosed, including cost and expectations for payment. The mechanism for communication of results should be understood. The alternatives to genetic testing and the option of declining to be tested must be disclosed. Written disclosure should be provided and written consent for testing should be obtained. The test itself must be done by a laboratory that is in compliance with Federal Clinical Laboratory Improvement Amendments (1988). Research laboratories must have their results verified by a compliant laboratory before clinical use. Test results are best disclosed in person by a medical professional (e.g., the surgeon) who is knowledgeable about the implications and limitations of testing. The professional who discloses the results should be involved in the testing process in the context of an established relationship with the test subject. The approach to testing must be centered on the individual to be tested. Information must be presented clearly and in a nonjudgmental fashion. The motives behind the request for testing should be understood. Sensitive and empathic support should be provided throughout. How should the surgeon prepare him- or herself for the genetic revolution? Knowledge of basic biological alterations offers the best hope for prevention or therapy.36 Health professionals must make genetics education a priority. The Internet has numerous resources for continuing medical education in medical genetics, and more are becoming available.36

Gene Therapy in Surgery Gene therapy will revolutionize medicine, and surgeons should be part of the revolution. Not only will historically “surgical” diseases be treated, effectively, with

12

Cytokines and the Abdominal Surgeon

gene therapy, but surgically-created access to discrete locations may be a critical component of therapy. Although genetic manipulation of germ cells has raised a major ethical debate—witness the recent furor over the possible cloning of human beings— the genetic manipulation of somatic cells for therapeutic benefit is being pursued vigorously. It is also possible that gene therapy may portend major advances in the prophylaxis of disease. Several factors play a role in gene therapy. First and foremost, the appropriate gene must be identified, sequenced, and replicated. The selected gene must be delivered to the target cell, incorporated into the cell, expressed by the cell (the protein encoded by the inserted DNA must be produced), and expression must occur without toxicity to the host. Toxicity to the target cell may be the goal, however, as in the case with gene therapy for cancer. Incorporation may be accomplished by any of several vectors, including bacterial plasmids (for ex vivo production of human proteins by bacteria), or any of several viruses for in vivo therapy, including retroviruses, adeno-associated viruses or parvoviruses, herpesviruses, and most commonly adenoviruses. There are at least 49 serotypes of adenoviruses, of which Types 2 and 5 are most often exploited. The therapeutic gene is incorporated into the viral genome and after administration to the test subject or patient, the adenovirus infects the target cell. An analogy may be drawn to a “Trojan horse”; the vector deposits the new gene and then either becomes moot or self-destructs. The fact that viral vectors are used to deliver genetic material raises relevant issues related to host defense.37 Viral vectors stimulate host defenses, and could destroy the vector before incorporation takes place. For some diseases, investigators are trying to make the vectors “stealthy”, so that tolerance or nonrecognition occurs long enough (a few hours is long enough) for incorporation. In the case of gene therapy for cancer this is less of an issue, because cytotoxicity is usually the goal of therapy. The host is usually already immunosuppressed by virtue of disease or conventional adjuvant therapy, and the gene may need to be expressed only briefly for lethal cytotoxicity to occur. Several strategies for suppression of tumor growth are being studied, and initial results have been promising.38,39 A converting enzyme may be administered along with a pro-drug, to convert the nontoxic precursor to a cytotoxin. The gene in such a strategy is referred to as a “suicide gene”, because the cell is induced to self-destruct. In one notable example, incorporation of the cytosine deaminase gene from E. coli has been administered via adenovirus vector to patients with metastatic colon cancer.40 Oral 5-fluorocytosine, a clinically-available, relatively nontoxic antifungal agent serves as substrate for cytosine deaminase, which converts it to 5-fluorouracil (5-FU). 5-fluorouracil has known efficacy against colon carcinoma and is a mainstay of conventional adjuvant therapy, but its toxicity to normal mucosal cells is high and sometimes dose-limiting. Regional delivery of 5-FU can be accomplished by an infusion port placed into the hepatic artery, but morbidity is high from a major operative procedure, and systemic toxicity can occur. By direct injection of the adenoviral vector into the metastatic lesions, high local drug concentrations are achieved where they are needed, but systemic toxicity is minimized. In the pilot study, preoperative CT-guided adenoviral vector injection directly into metastatic lesions, followed by surgery one week later, demonstrated more CD8+ (T-helper) lymphocytes in treated as compared to untreated lesions, and there was more apoptosis after treatment as assayed by DNA fragmentation.40

Relevance of Molecular Biology for the Abdominal Surgeon

13

Gene therapy will likely lend itself to many other “surgical” diseases, including cardiovascular disease41,42 and peripheral vascular disease.42-44 Growth-promoting genes may induce the conversion of fibroblasts to cardiac myocytes to improve pump function.41 Gene therapy may stimulate the formation of collateral vessels in areas of ischemic myocardium by introduction of angiogenic growth factors. Coronary collateral flow has been increased after administration of vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF).42 Paradoxically, the VEGF pathway has become an important target of cancer researchers who hope to inhibit angiogenesis and render tumors ischemic.45 Other potential strategies of interest to the vascular surgeon include gene therapy of hypercholesterolemia and atherosclerosis, or direct manipulation of disease peripheral vasculature. Already, genetic material can be delivered directly to an experimental lesion using endovascular insertion techniques. Transplant surgeons are interested in gene therapy both from the standpoint of tissue growth or replacement, and from the standpoint of the regulation and control of the immune response.46 Genetic antiinflammatory therapies will also be of intense interest to the surgical intensivist.47

Antisense Oligonucleotides as Potential Gene Therapy So-called “antisense oligonucleotides” may be of critical value and importance as therapy in the future.47-50 In double-stranded DNA, the nucleotide sequence that contains the information to be translated is called the sense strand. Its complimentary strand is thus the antisense strand. The antisense approach uses short oligonucleotides designed to hybridize to a target mRNA, forming a heteroduplex that inactivates the mRNA and brings protein synthesis to a halt. An attraction of this approach is its potential applicability to any gene product, but much development needs to be done. Antisense therapy could be delivered either as synthetic oligonucleotides delivered exogenously (liposomes, naked DNA), or via the more complex endogenous production of antisense RNA transcripts generated by expression vectors transfected into the cell. Multiple points in the protein synthetic sequence could theoretically be targeted. Active development issues include stability, the length of the sequence (15-20 base pairs is the minimun to ensure sufficient specificity), cellular uptake, target specificity, appropriate negative controls, the potential for oligonucleotide:protein interactions, and cost. Antisense technology has already been used to block cytokine production in vivo, and against intercellular adhesion molecules. In the latter case, there is therapeutic potential for transplant immunosuppression as well as the host response to bacterial infection. In oncology, a number of clinical trials have been initiated against molecular targets including p53, and Bcl-2.51

References 1. Tzeng E, Shears LL, Lotze MT, Billiar TR. Gene therapy. Curr Prob Surg 1996; 33:1032-34. 2. Smith MP. Exploring molecular biology. Arch Surg 1995; 130:811-816. 3. Hebert JC, O’Reilly M, Bednar MM. Modifying the host response to injury. Surg Clin North Am 1995; 75:335-349. 4. Watson GD, Crick FHC. Molecular structure of nucleic acids: a structure for deoxyribose nucleic acid. Nature 1974; 248:765. 5. Lodish H, Baltimore D, Berk A et al. Molecular Cell Biology. 3rd ed. New York: Scientific American Books, Inc. 1995:10-594.

14

Cytokines and the Abdominal Surgeon

6. Lyerly HK, Sullenger BA. Molecular biology in surgery. In: Sabiston DC, Lyerly HK, eds. The Biological Basis of Modern Surgical Practice. 10th ed. Philadelphia: WB Saunders, 1997:16-35. 7. Watson JD, Hopkins NH, Roberts JW, Steitz JA, Weiner AM. Molecular Biology of the Gene. 4th ed. Menlo Park, CA: Benjamin/Cummings Publishing Co., 1987:282-309. 8. Hall CC, Herring JA, Hall TJ. Molecular oncology and the surgeon. Am Surg 1995; 61:156-160. 9. Shore EM, Kaplan FS. Molecular biology for the clinician. Part II: Tools of molecular biology. Clin Orthop Rel Res 1995; 320:247-278. 10. Schuler GD, Boguski MS, Stewart EA, Stein LD et al. A gene map of the human genome. Science 1996; 274:540-546. 11. Rosenberg RN. Molecular neurogenetics. The genome is settling the issue. JAMA 1997; 278:1282-1283. 12. Hardy KJ, Young HA, Lagoo AS. Molecular diagnostics. Immunol Allergy Clin North Am 1994; 14:199-223. 13. Rosenthal N. Tools of the trade. Recombinant DNA. New Engl J Med 1994; 331:599-602. 14. Kroczek RA. Southern and Northern analysis. J Chromatogr 1993; 618:133-145. 15. Borst M, Miller DM. DNA isolation and Southern analysis: A clinician’s view. Am J Med Sci 1990; 299:356-360. 16. Ausubel FM, Brent R, Kingston RE et al. The polymerase chain reaction. In: Current Protocols in Molecular Biology. Massachusetts: John Wiley & Sons, Inc. 1997; (suppl)37:15.1.1-15.1.5. 17. Abbas AK, Lichtman AH, Pober JS. Cellular and Molecular Immunology. Philadelphia, PA: WB Saunders Company, 1994:56-72,109-111. 18. Mariani BD, Martin DS, Levine MJ et al. Polymerase chain reaction detection of bacterial infection in total knee arthroplasty. 1996; 331:11-22. 19. Lagoo-Deenadalayan S, Lagoo S, Barber WH et al. A standardized approach to PCRbased semiquantitation of multiple cytokine gene transcripts from small cell samples. Lymphokine Cytokine Res 1993; 12:59-66. 20. Dallman MJ, Montgomery RA, Larsen CP et al. Cytokine gene expression: Analysis using northern blotting, polymerase chain reaction, and in situ hybridization. Immunol Rev 1991; 119:163-179. 21. Burnette WN. Western blotting. Electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal Biochem 1981; 112:195-208. 22. Fido RJ, Tatham AS, Shewry PR. Western blotting analysis. Methods Molec Biol 1995; 49:423-437. 23. Webb SJ, Harrison DJ, Wyllie AH. Apoptosis: An overview of the process and its relevance to disease. In: Kaufmann SH, ed. Apoptosis. Pharmacological Implications and Therapeutic Opportunities. Adv Pharmacol 1997; 41:1-34. 24. Que F, Gores GJ. Apoptosis and the gastrointestinal system. In: Kaufmann SH, ed. Apoptosis. Pharmacological Implications and Therapeutic Opportunities. Adv Pharmacol 1997; 41:409-428. 25. Allen RT, Hunter WJ III, Agrawal DK. Morphological and biochemical characterization and analysis of apoptosis. J Pharmacol Toxicol Methods 1997; 37:215-228. 26. May WS Jr. Control of apoptosis by cytokines. In: Kaufmann SH, ed. Apoptosis. Pharmacological Implications and Therapeutic Opportunities. Adv Pharmacol 1997; 41:219-246.

Relevance of Molecular Biology for the Abdominal Surgeon

15

27. Eischen CM, Leibson PJ. The Fas pathway in apoptosis. In: Kaufmann SH, ed. Apoptosis. Pharmacological implications and therapeutic opportunities. Adv Pharmacol 1997; 41:107-132. 28. Strausberg RL, Dahl CA, Kuaisner RD. New opportunities for uncovering the molecular basis of cancer. Nat Genet 1997; 15:415-416. 29. Pyeritz RE. Family history and genetic risk factors. Forward to the future. JAMA 1997; 278:1284-1285. 30. Shatuck-Eidens D, Oliphant A, McClure M et al. BRCA1 sequence analysis in women at high risk for susceptibility mutations: Risk factor analysis and implications for genetic testing. JAMA 1997; 278:1242-1250. 31. Gronberg H, Isaacs SD, Smith JR et al. Characteristics of prostate cancer in families potentially linked to the heredity prostate cancer 1 (HPC1) locus. JAMA 1997; 278:1251-1255. 32. DeRisi J, Penland L, Brown PO et al. Use of a cDNA micro array to analyze gene expression patterns in human cancer. Nat Genet 1996; 14:457-460. 33. Stephenson J. As discoveries unfold, a new urgency to bring genetic literacy to physicians. JAMA 1997;278:1251-1255. 34. Predisposition genetic testing for late-onset disorders in adults. A position paper of the National Society of Genetic Counselors. JAMA 1997; 278:1217-1220. 35. Collins FS. Preparing health professionals for the genetic revolution. JAMA 1997; 278:1285-1286. 36. Sikorski R, Peters R. Genomic medicine. Internet resources for medical genetics. JAMA 1997; 278:1212-1213. 37. Wolff G, Worgall S, vanRooijen N et al. Enhancement of in vivo adenovirus-mediated gene transfer and expression by prior depletion of tissue macrophages in the target organ. J Virol 1997; 71:624-629. 38. Lairmore T, Norton JA. Advances in molecular genetics. Am J Surg 1997; 173:37-41. 39. Evoy D, Hirschowitz E, Naama HA et al. In vivo adenoviral-mediated gene transfer in the treatment of pancreatic cancer. J Surg Res 1997; 69:226-231. 40. Crystal RG, Hirschowitz E, Lieberman M et al. Phase I study of direct administration of a replication-deficient adenovirus vector containing the E. coli cytosine deaminase gene in metastatic colon carcinoma of the liver in association with the oral administration of the pro-drug 5-fluorocytosine. Hum Gene Ther 1997; 8:985-1001. 41. Tam SKC, Gu W, Nadal-Ginard B et al. Molecular cardiomyoplasty: Potential cardiac gene therapy for chronic heart failure. J Thorac Cardiovasc Surg 1995; 109:918-924. 42. Rowland RT, Cleveland JC Jr, Meng X et al. Potential gene therapy strategies in the treatment of cardiovascular disease. Ann Thorac Surg 1995; 60:721-728. 43. Clowes AW. Vascular gene therapy in the 21st century. Thromb Hemostasis 1997; 78:605-610. 44. Schwartz LB, Moawad J. Gene therapy for vascular disease. Ann Vasc Surg 1997; 11:189-199. 45. Harris AL. Antiangiogenesis for cancer therapy. Lancet 1997; 349:13-15. 46. Knechtle SJ. Gene therapy and transplantation-a brief review. Transpl Proc 1996; 28(Suppl 1):19-23. 47. Liu M, Slutsky AS. Anti-inflammatory therapies: Application of molecular biology techniques in intensive care medicine. Intensive Care Med 1997; 23:718-731. 48. Aoki M, Morishita R, Higaki J et al. In vivo transfer of antisense oligonucleotides in to the myocardium using HVJ-liposome method. Biochem Biophys Res Comm 1997; 231:540-545.

16

Cytokines and the Abdominal Surgeon

49. Branch A. A hitchhiker’s guide to antisense and nonantisense biochemical pathways. Hepatology 1996; 24:1517-1529. 50. Crooke ST. Advances in understanding the pharmacological properties of antisense oligonucleotides. Adv Pharmacol 1997; 40:1-49.51. 51. Ho PT, Parkinson DR. Antisense oligonucleotides as therapeutics for malignant diseases. Semin Oncol 1997; 24:187-202.

Cytokine Response in Abdominal Surgery

17

CHAPTER 3

Cytokine Response in Abdominal Surgery Edward Lin, Steve E. Calvano, and Stephen F. Lowry

Introduction

A

side from cutaneous procedures, operations on the abdominal viscera remain one of the most frequently performed elective surgeries in the United States. While remarkable perioperative and technical advances have been made in the management of patients with abdominal pathology, the significance of circulating as well as localized mediator release leading to tissue inflammation in abdominal surgery remains relatively unknown. Classically, the stress response comprises activation of the hypothalamic-pituitary-adrenal axis, the acute phase response and the autonomic nervous system.1 Each such response functions, at least acutely, to restore homeostasis to the injured patient. Contemporary investigations of the injury response, mediated by inflammatory molecules referred to as cytokines, have identified their role as a critical mechanism(s) that induce many of the metabolic and physiologic derangements characteristic of surgical or traumatic injury. Cytokines are produced by diverse cell populations and are capable of exerting multiple (pleiotropic) proinflammatory or anti-inflammatory influences.2 These are essential components of the inflammatory response mounted against local injuries and infections (Table 3.1). These responses may be elicited locally via autocrine or paracrine activities or systemically in an endocrine fashion. Our current understanding related to the pathophysiology of these inflammatory mediators have been largely derived from the study of patients with endotoxemia and/or sepsis. The cytokine response to infections or injury is qualitatively similar, but generally magnified in comparison to controlled, elective surgery.3 The clinical manifestation of presumably excessive proinflammatory cytokine activation is often referred to as the systemic inflammatory response syndrome (SIRS).4 Such a response necessitates an intact immunologically responsive host and is not, by definition, obligatorily the result of an infectious process. In clinical practice, the criteria for SIRS, which reflects abnormalities in leukocyte count, respiratory rate, heart rate and body temperature, might also be anticipated in many patients undergoing abdominal surgery. Dysregulation of this normal inflammatory cytokine response, as might occur with secondary traumatic or infectious insults, may augment the state of inflammation leading to shock, tissue injury, and multiple organ failure (MOF).1 Certainly, Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

18

Cytokines and the Abdominal Surgeon

Table 3.1. Catalog of selected cytokines released during controlled surgical injury and their potential beneficial effects TNF-!

∀ PMN release from bone marrow ∀ PMN activation, migration to injured site, degranulation

and superoxide production PMN cytotoxicity against mycotic infections differentiation (activation) of M# M# antiviral/antiparasite activities hepatocyte acute phase reactant (APR) production through IL-6 induction ∀ wound healing/remodeling ∀ endothelial procoagulant activity and leukocyte adhesion ∀ vascular endothelial permeability ∀ neovascularization in wounds ∀ collagen synthesis/fibroblast proliferation ∀ osteoclast activity in bone healing ∀ T-lymphocyte activation and proliferation ∀ PMN release from bone marrow ∀ PMN migration to injured site ∀ differentiation (activation) of M# ∀ granulocyte/macrophage colony-stimulating factor (GM-CSF) ∀ functional PMN restoration ∃ pain perception ∀ %-endorphin release ∀ brain opiate-like receptors ∀ hepatocyte acute phase reactant (APR) production through IL-6 induction ∀ wound healing/remodeling ∀ osteoclast activity in bone healing ∀ overall immunocompetence ∀ lymphokine-activated killers (LAK) production ∀ cytotoxic T-lymphocyte proliferation ∀ reticuloendothelial system (RES) activity ∀ gut barrier immunity ∀ fibroblast antiviral activity ∀ lymphocyte differentiation ∀ B-lymphocyte immunoglobulin production ∀ acute phase reactant (APR) production (e.g., C-reactive protein, fibrinogen, amyloid A, !-1-antitrypsin, haptoglobin) ∀ prostaglandin production ? wound healing properties ∀ chemotaxis of PMN, lymphocytes, M#s to sites of injury and inflammation ∃ cytokine synthesis by lymphocytes and M#s Modulates inflammatory activities of TNF-!, IL-1, IL-6, IL-8, INF-& ∀ ∀ ∀ ∀

IL-1

IL-2

IL-6

IL-8 IL-10

continued...

Cytokine Response in Abdominal Surgery

19

(continued..)

IL-12

IFN-&

∀ ∀ ∀ ∀ ∀ ∀ ∀ ∀ ∀ ∀

Stimulates CD4+ and CD8+ T cells lymphocyte and NK cell proliferation B-lymphocyte immunoglobulin production hematopoiesis IL-2 and IFN-& production M# and PMN activation against invading organisms (including viral) expression of MHC class I and II surface antigens and tumor-associated antigens M# oxidative and cytotoxic activity overall lymphocyte proliferation B-lymphocyte immunoglobulin production IL-1 and TNF-! activity

GM-CSF ∀ myeloproliferation (M#s, PMNs, eosinophils) Partial stimulation of megakaryocyte progenitors ∀ chemotaxis of PMNs and M#s ∀ cytokine production by M#s

the clinical implications of unregulated proinflammatory mediator activities cannot be ignored, as mortalities from MOF in surgical intensive care units remain high and resistant to improvement based upon currently available therapies.2 A proper perspective of the immunobiologic sequelae of cytokine response arising from injury or inflammation may have important clinical applications, particularly in the approach towards the patient undergoing complex abdominal surgery. The catalogue of known cytokines presently approximate 30, and is steadily expanding.3 This chapter will highlight the cytokines that have been more extensively investigated as well as address their potential clinical relevance to the acute response and outcome in major abdominal surgeries.

Injury and Principal Cytokine Responses Tumor Necrosis Factor-! The inflammatory response to severe cross-sectional tissue injury and/or infectious agents evokes a complex cascade of proinflammatory cytokines. Of these, tumor necrosis factor (TNF-!) and interleukin-1 (IL-1) appear to be the earliest and perhaps most potent mediators of the subsequent host response.3 The release of TNF-! in response to acute injury is both rapid and short-lived. Experiments simulating an acute inflammatory response by means of endotoxin challenge have invariably demonstrated a monophasic TNF appearance curve, peaking at approximately 90 minutes.4 This is followed by undetectable levels of bioactive protein within 4 hours (Fig. 3.1). Despite a circulating half-life of less than 20 minutes, the brief appearance of TNF-! seems sufficient to induce the characteristic metabolic and severe hemodynamic changes.9 In addition, the signaling for both secondary cytokines distal in the cascade as well as of intracellular toxins appears to occur within this time frame. This brief appearance of TNF-! and of absent biologic activity over subsequent periods suggests the presence of effective endogenous modulators, which might avert further deleterious consequences from excessive TNF-!

20

Cytokines and the Abdominal Surgeon

Fig. 3.1. Representative circulating TNF-! release in normal adult volunteers upon brief lipopolysaccharide (LPS, endotoxin) challenge at time = 0. A rapid peak is reached at 90 minutes, followed by undetectable levels at 4 hours after challenge.

500

450

400

350

300

250 TNF (pg/mL)

200

150

100

50

0 0

LPS

0.5

1

1.5

2

2.5

3

3.5

4

5

6

7

8

TIME (hour)

activity. Indeed, this has proven to be the case, as several natural antagonist mechanisms to TNF-! production and/or activity have been identified.5 From experimental models of endotoxemia and in severe sepsis, endogenous inhibitors in the form of cleaved extracellular domains of the transmembrane TNF-! receptors (soluble TNF receptors, sTNFR) are readily detectable in the circulation.6 These receptors may serve a protective role by competitively sequestering excess circulating TNF, but are probably only capable of doing so against low levels of TNF-! activity and for relatively brief periods.7,8 The sources of TNF-! synthesis include monocytes/macrophages (M) and T cells, all of which are abundant within the peritoneum and splanchnic tissues9 (Table 3.2). Moreover, with hepatic Kupffer cells representing the single largest concentrated population of M#s in the body,10 peritoneal injury, liver surgery and procedures upon the splanchnic viscera may have profound influences on the generation of inflammatory mediators and homeostatic responses such as acute-phase protein production (Fig. 3.2).

Interleukin-1 TNF-! also induces the biosynthesis and release of interleukin-1 (IL-1) within Ms and endothelial cells.13 The potency and effects of IL-1 reflect that of TNF-! and they exert several similar physiologic derangements.11 At high doses, both IL-1 and TNF-! induce a state of hemodynamic decompensation reminiscent of septic shock.12 The circulating half-life of IL-1 is less than 10 minutes, a fact which makes its detectability in acute injury or illness less likely than that of TNF-!.13 Among its multiple effects, IL-1 induces the classic inflammatory febrile response to injury by stimulating local prostaglandin activity in the anterior hypothalamus. It also serves as a mediator for attenuating pain perception from surgery by promoting the release of endorphins and increasing the number of central opiate-like receptors.14

Cytokine Response in Abdominal Surgery

21

Table 3.2. Principal sources of selected cytokines TNF-! IL-1

IL-2 IL-6

IL-8

IL-10 IL-12

INF-&

GM-CSF

M# T cells M# B and T cells NK cells Endothelial cells Epithelial cells Keratinocytes Fibroblasts Dendritic cells TH1 cells T cells M# Fibroblasts Endothelial cells M# Endothelial cells T cells Platelet B cells TH2 cells M# PMN Keratinocytes Dendritic cells TH1 cells NK cells M# T cells Fibroblasts Endothelial cells Stromal cells

Endogenous inhibitors of IL-1, known as IL-1 receptor antagonists (IL-1ra), are also released in response to sepsis or endotoxemia.13,14 However, to date, the administration of exogenous sTNFR or IL-1ra have not resulted in any appreciable outcome improvements in clinical sepsis. Distal cytokine mediators, released as part of the inflammatory cascade initiated by TNF-! and IL-1 include IL-2, IL-4, IL-6, IL-8, granulocyte-macrophage colonystimulating factor (GM-CSF) and interferon (IFN).14

Interleukin-2 Although necessary as an inflammatory mediator in promoting immunoglobulin activity and gut barrier integrity, IL-2 is not detected in acute injury.2 Indeed, IL-2 secretion by lymphocytes is impaired following acute injury and several disease states, notably cancer and acquired immunodeficiency syndrome (AIDS).15,16

22

Cytokines and the Abdominal Surgeon

Fig. 3.2. Hypothetical schema for the hepatic acute phase response following injury. Sources of stress and subsequent antigen/endotoxin release may be derived from the splanchnic circulation or from other injury foci (e.g., traumatic, tumor load). The antigen/endotoxin enters the portal circulation and is cleared by hepatic Kupffer cells (M#). Some antigen/endotoxin, particularly when in excess or in liver failure, may escape clearance and enter the systemic circulation. The activated M# is capable of inducing cytokine release that serves as signals for hepatocyte synthesis of acute phase proteins (e.g., haptoglobin, C-reactive protein, ceruloplasmin, complement factors, fibrinogen, -1-antitrypsin, 2-macroglobulin). However, cytokines may also “spill” into the systemic circulation and contribute to the systemic inflammatory response.

Perioperative hemorrhage and transfusions have also been demonstrated to reduce IL-2 production.17,18 Attenuated IL-2 expression following injury partially defines the transient immunocompromised state of the surgical patient. The clinical significance of this depressed IL-2 activity remains unclear. Nevertheless, a nadir in gut barrier IL-2 activity resulting from surgery is hypothesized to predispose the patient to enteric organism activation of the inflammatory cytokine cascade in the gut, with potential inflammation systemically.14 Recent evidence has demonstrated accelerated lymphocyte programmed cell death (apoptosis), in association with diminished IL-2 activity, mediated by the proapoptotic Fas/CD95 cell receptor in the early postoperative period.19 The combined diminution of lymphocyte survival and IL-2 activity may further contribute to the immunocompromised phenotype of the surgical patient. Other studies have demonstrated that following surgical stress, a lymphocyte population density shift from type-1 T-helper cells (TH1, cell-mediated immune re-

Cytokine Response in Abdominal Surgery

23

Fig. 3.3. Specific immunity related to the magnitude of injury as depicted by cholecystectomy, laparoscopically (lesser magnitude) vs. open cholecystectomy (greater magnitude). A type-1 T-helper cell response (TH1) is favored in lesser injuries, with intact cell-mediated immunity against microbial infections. A TH1 response can be activated by the cytokines IFN-&, TNF and IL-1, but IL-12 appears to be the most potent stimulant. A shift toward type-2 T-helper cell response (TH2) is associated with greater injuries and is predominantly immunosuppressive. By nature of its antibody-mediated immunity, it is ineffective against microbial invasions. TH2 cells activate B-lymphocyte IgG4 and IgE production, which are poor complement activators and among the least efficient opsonizing antibodies. Among the cytokines capable of inducing a TH2 response, IL-4 appears to be the most potent. Although not a cytokine, glucocorticoids are potent upregulators of IL-4 production and are effective stimulants of TH2 immunity, explaining in part the immunosuppressive effects of this agent. Although predominantly immunosuppressive, TH2 response to injury is necessary in regulating the excessive inflammatory response mediated by such cytokines as TNF and IL-1. (Concept derived from references 17, 21, 23, 25).

sponse including IL-2, IL-12 and interferon-! production) to type-2 T-helper cells (TH2, antibody-mediated immune response including IL-4, IL-6, IL-10, IL-13 production) is observed.20 This population shift may partially explain the diminished IL-2 activity seen following surgery. Furthermore, the predominant effects of TH2 response are immunosuppressive, which accentuate the risks for postoperative infections17 (Fig. 3.3).

Interleukin-4 The importance of IL-4 is steadily gaining recognition. This glycoprotein molecule is produced by activated TH2 cells with diverse biologic effects on hematopietic cells.21 Of particular importance in its role in antibody-mediated immunity is the capacity to enhance M MHC class II (HLA-DR and HLA-DP) expression and adhesion molecules, making them efficient antigen-presenting cells.22 As a potent antiinflammatory cytokine, it can downregulate several functions associated with activated human Ms, namely the effects of IL-1, TNF-!, IL-6, IL-8, and superoxide production. However, these antiinflammatory effects of IL-4 are not seen with resting monocytes.27 The importance of this cytokine is therefore in its capacity to downregulate the response of inflammatory M’s that are exposed to stimuli

24

Cytokines and the Abdominal Surgeon

such as bacterial endotoxin or inflammatory cytokines (e.g., IL-1 or TNF-!). Several lines of experiments have demonstrated that IL-4 is a potent inducer of programmed cell death in inflammatory Ms, but this effect can be abrogated by IFN.23 Indeed, IL-4 and IFN have been extensively demonstrated to antagonize the effects of each other on B cells.28 IL-4 also appears to increase M susceptibility to the antiinflammatory effects of glucocorticoids.27 Although studied to a lesser extent, IL-13 has been found to share several important properties with IL-4.26

Interleukin-6 There is evidence to suggest that interleukin-6 (IL-6) exerts both pro-inflammatory and anti-inflammatory influences.24 The high frequency for detection of circulating IL-6 during acute injury or stress and the correlation of such levels to surgery made it an indicator of systemic inflammatory response as well as a predictor of surgical morbidity.25 TNF-! and IL-1 are major inducers of IL-6, which can be expressed from virtually all cell types including the intestines. Upon injury, IL-6 levels in the circulation can peak between 4 to 6 hours and persist for as long as 10 days.29 Clinically and experimentally, IL-6 levels appear to be roughly proportional to the extent of tissue injury related to surgery rather than the duration of the surgical procedure itself.26,27 IL-6 is the primary mediator of the hepatic acute phase protein response during injury.31,28,29 Furthermore, it induces polymorphonuclear cell (PMN) activation during injury and inflammation.30 However, IL-6 also delays the phagocytic disposal of senescent or dysfunctional PMNs during injury, thereby potentially perpetuating the inflammatory response.31 The persistence of activated PMNs following surgical injury can also have distant contributions to cardiovascular and pulmonary compromise.29 IL-6 exhibits antiinflammatory properties via attenuation of TNF and IL-1 activity and promotion of soluble TNFR (sTNFR) and IL-1 receptor antagonist (IL-1ra) release.32 However, prolonged and persistent expression of IL-6 has also been associated with immunosuppression and postoperative infectious morbidity.

Interleukin-8 The appearance of interleukin-8 (IL-8) activity is temporally associated with IL-6 following injury and has been proposed as an additional biomarker for the risk of MOF following injury.33 Like IL-6, IL-8 does not produce the hemodynamic instability characteristic of TNF-! and IL-1, but rather serves as a PMN activator and a potent chemoattractant.34 Hence, IL-8 is being increasingly established as a major contributor to organ injury such as acute lung injury not infrequently seen following major surgeries.

Interleukin-10 Interleukin-10 (IL-10) serves as an important endogenous regulatory protein during the inflammatory response. The activity of this cytokine is identified with modulation of TNF-! activity.35 Its appearance in the circulation during endotoxemia, for example, follows that of TNF-!. Supporting experiments have demonstrated that deletion of IL-10 during endotoxemia increases monocyte TNF-! production, but restitution of IL-10 reduces TNF-!.36 IL-10 may also have additional protective roles following injury-induced inflammation by promoting IL-1ra and sTNFR production. In animal experiments, the sustained systemic production of IL-10 during septic peritonitis appears to modulate the systemic inflammatory response. Murine ex-

Cytokine Response in Abdominal Surgery

25

periments have demonstrated rapid induction of IL-10 messenger RNA (mRNA) activity following cecal ligation and puncture (CLP).37 Blocking this activity with anti-IL-10 enhanced mortality following CLP. However, this immunomodulatory effect may also abrogate the necessary proinflammatory response necessary for local clearance of invading organisms.

Interleukin-12 The role of interleukin-12 (IL-12) in the immune response to injury and inflammatory events has gained significant interest. Its capacity to promote the differentiation of TH1 cells and the production of IFN makes it a pivotal molecule in cell-mediated immunity following injury or infection (bacterial, viral and parasitic).38 IL-12 is also implicated in preventing programmed cell death (apoptosis) in selected T-lymphocyte populations following their activation.39

Interferon (IFN) Much of IL-12 biology is mediated through the production and activities of interferon (IFN).40 Stimulated human T helper (TH) cells, through bacterial antigens, IL-2 and IL-12 induce IFN. Upon release from activated T cells, IFN is detectable in vivo within 6 hours following stimulation and has a half-life of approximately 30 minutes. IFN levels peak at 48 to 72 hours and may persist for 7 to 8 days after surgery.14,17 Surgical wounds also demonstrate the presence of IFN production 5 to 7 days following injury.41 IFN has important roles in activating circulating and tissue M#s. Its activation of peritoneal Ms and Kupffer cells has potential immunologic implications in abdominal surgery.42 Furthermore, alveolar M# activation mediated by IFN may be synergistic with IL-8 in the induction of acute lung injury following major surgery.23,25

Clinical Considerations The abdominal surgical patient often presents with a diverse array of preoperative disease states that may influence the postoperative inflammatory response. Such preoperative conditions may include malnutrition, cancer, sepsis, bowel obstruction, peritonitis, critical illness and an element of acquired or induced immunodeficiency. Surgically induced factors that may influence the postoperative inflammation include the extent of cross-sectional tissue injury/resection, wound size, anesthetic regimen, and hemorrhage. In essence, patients with abdominal pathology may exhibit preexisting activation of cytokine expression. Consequently, an interpretation of the postoperative inflammatory response must account for these factors as well as the magnitude of trauma incurred during surgery.

Exploratory Laparotomy Elective exploratory laparotomy would seemingly represent a model system by which to judge the inflammatory response. Murine experiments have demonstrated an impairment in peritoneal M microbicidal activity and antigen presentation following laparotomy.47 However, recovery of normal function in this M population normally occurs on postoperative day 3. This recovery can be accelerated with IFNadministration although such treatment may also augment local inflammatory responses mediated by TNF and IL-1. The local inflammatory response mediated by these M populations through IL-2 and IL-6 activity is also markedly diminished following laparotomy.43

26

Cytokines and the Abdominal Surgeon

The impaired immunocyte and cytokine response following laparotomy in some patients is purported to increase postoperative infectious complications. Systemically, IL-6 levels are found to be elevated following surgical stress, although a clear association with the magnitude of surgical therapy have yet to be established. It has been suggested that the impaired antigen-presentation ability observed in circulating Ms following major abdominal surgery may be mediated, in part, by IL-10 induced downregulation of major histocompatibility II (MHC-II) expression on these cell populations.44

Length of Surgery and Injury Magnitude The correlation between inflammatory cytokine levels, after major abdominal procedures (e.g., hepatic, gastric or pancreatic resections), with the duration of surgery further demonstrate the complexity of this response. Several experimental and clinical studies imply that the type of surgery or the extent of tissue injury are the primary determinants of cytokine responses and that the crossover of these inflammatory mediators into the systemic circulation portends the adverse outcome in certain patients.45 Utilizing IL-6 levels as a potential marker of cytokine responses to elective abdominal surgery, the peak levels are highest in aortic surgery immediately following aortic cross-clamping.46 Levels of IL-6 are lower in colorectal surgery of similar duration.31 Open cholecystectomy had the lowest levels of cytokine activity among the procedures evaluated. In a rat model comparing sham laparotomy to laparotomy plus splanchnic artery occlusion, significant increases in TNF and IL-6 activity were observed in the latter group.32 These attest to the greater contribution of tissue injury to the inflammatory cytokine response than the actual duration of surgery. Cytokine expression following tissue injury must also account for the acuity of this response. While the degree of tissue injury correlates with elevated cytokine levels, the slope of the appearance of cytokine activity during the early postoperative phase also reflects the degree of surgical injury.32 In surgeries of similar duration, an acute and “excessive” rise in IL-1 and IL-6 levels rather than gradual elevations, have been predictive of postoperative complications (e.g., MOF, ARDS) and poorer clinical outcome.30,47

Intestinal Surgery The intestines serve as major sources of inflammatory cytokine production, particularly in response to injurious stimuli or stresses.48 While this inflammatory cytokine response is probably essential for local homeostasis, excessive production can spill into the systemic circulation as demonstrated by hepatic vein cannulation in healthy human subjects.49 The concept of gut barrier dysfunction with resultant bacterial translocation has been widely promoted as a mechanism for induction of cytokine activity.50,51 Only limited data exists to address this concept in humans. In a classic study,52 mesenteric lymph node cultures of patients with bowel obstruction demonstrated positive bacterial cultures in approximately 60% of patients, while only 4% of nonobstructed patients had such findings. Similarly, patients with inflammatory bowel disease (IBD) had higher incidence of bacterial translocation and local cytokine response than patients without IBD.53,54 Should bacterial translocation or regional antigen stimulation exist in a surgical patient, this might serve as a mechanism for enhancing cytokine activity during the early postoperative phase. Unfortunately, necessary studies which

Cytokine Response in Abdominal Surgery

27

integrate translocation, local and systemic mediator responses, and clinical outcome have not been conducted. Brief periods of splanchnic shock and/or hypoperfusion followed by reperfusion might also disrupt normal intestinal barrier function.56 This has implications for procedures where cross-clamping of arterial vessels and splanchnic/organ hypoperfusion may accentuate the local or systemic inflammatory cytokine response. Simple traction of the bowel during surgery has been demonstrated to increase endotoxin and microbial permeability through the gut, leading to rises in IL-6 activities.55 This strongly suggests that extensive visceral manipulation might also contribute to the exaggerated local and systemic cytokine response in abdominal surgery.

Hepatobiliary Surgery The liver Kupffer cells are essential for the clearance of endotoxins derived from the intestinal tract, as well as for the neutralization and/or clearance of inflammatory cytokine mediators.56 Perioperative hepatic impairment may contribute to the excessive systemic inflammatory response and attenuate the acute phase protein response following injuries such as burns, trauma, circulatory shock, anesthesia and surgery.57,58 Obstructive jaundice or liver resection can further impair Kupffer cell function. Bile serves to transport immunoglobulin A (IgA), which binds gut-derived bacterial products for excretion into the feces.59 Bile acids are also capable of neutralizing endotoxin. Hepatic dysfunction preoperatively or from hepatic resections may impair normal bile secretion, which has been associated with increased intestinal endotoxin translocation.60 This may permit endotoxin and inflammatory mediators to bypass the regulatory mechanisms of the liver into the systemic circulation. Indeed, patients undergoing hepatectomy have elevated systemic endotoxin levels compared to their preoperative states.61 The clinical implications of such findings have been suggested, but not yet demonstrated, to correlate with postoperative morbidity or mortality. Surgical injury to the hepatic parenchyma also induces local release, as well as systemic elevations of TNF-! and IL-6.61,62 The local release of inflammatory mediators such as IL-1 may also inhibit hepatocyte proliferation, an effect that can be partially abrogated with the administration of a specific IL-1 antagonist (IL-1ra).62 Experimental evidence has further demonstrated the role of Fas/CD95 ligand (FasL) in promoting hepatocyte apoptosis during systemic inflammatory response.63 Clinically, this may explain a mechanism for liver failure commonly observed after hepatic resections as well as in sepsis.64 Furthermore, the systemic spillage of inflammatory cytokines may explain the lung injury, cardiac complications and other organ dysfunction often observed following major hepatic resections or hepatic cryoablation.65

Nutritional Considerations In addition to maintaining the metabolic and physiologic requirements of the surgical patient, adequate nutrition is also essential for immune competence. Malnutrition is prevalent among patients with abdominal pathology such as cancer, obstruction, and peritonitis. In these patients, a major cause of immune deficit might be an impairment of the normal intestinal barrier.56 Hypothetically, this increases the susceptibility to bacterial and endotoxin translocation and potential activation of local inflammatory mediator activity. Local or systemic release of proinflammatory cytokines may have both acute and indolent sequelae, such as immune dysfunction,

28

Cytokines and the Abdominal Surgeon

increased metabolic demands, and cachexia.56 Indeed, studies have demonstrated that bowel rest and antecedent total parenteral nutrition can exaggerate the inflammatory cytokine response to any subsequent endotoxin challenge.66 Clinically, these preexisting conditions have the potential to adversely affect the outcome of patients undergoing any major surgery that might serve as an additional trigger for cytokine release. It is hypothesized that efforts to maintain or restore intestinal barrier function might minimize further metabolic derangements, restore local immunity and thereby serve to optimize surgical outcomes.67 It has been demonstrated that some forms of cachexia may result from prolonged activity of TNF-!, IL-1 and IL-6, and these mediators can promote peripheral muscle protein mobilization for splanchnic and hepatic use.68 Although cytokine activity may not be either predominant or uniform in this effect, muscle wasting in malnourished surgical patients predisposes them to morbidities such as wound complications, poor healing, weakness, and delayed weaning from ventilator support. The normal inflammatory response can also be markedly altered or exaggerated as a result of infections, surgical stress or injury. In the worst scenario, it may predispose the patient to systemic organ injury (e.g., liver, lung, kidney) and increase late mortality.73 The potential for nutritional modulation of the inflammatory response is controversial. Several purported immunonutritional agents such as glutamine, arginine, omega-3 fatty acids, and fish oil are under intense investigation.69-71 While the capacity of these agents to minimize postoperative morbidities and modulate the cytokine response are reported, these agents are not without adverse effects. Furthermore, clinical outcomes following the use of these agents remain unclear.

Cancer Patients Patients with solid and hematologic malignancies may exhibit proinflammatory cytokine responses that result in significant physiologic and metabolic derangements.72 These cytokine mediators are likely contribute to cancer cachexia, anorexia, anemia, leukocytosis, acute phase responses and coagulopathy.73 These responses may originate from the tumors or from host immunocytes. Several studies in tumor-bearing hosts have demonstrated an exaggerated inflammatory response to insult such as sepsis.48,74 One would expect that surgical injury to patients with abdominal tumor can augment the existing inflammatory cytokine response. In clinical practice, patients undergoing major abdominal cancer operations are frequently critically ill postoperatively, requiring intensive support. The tumor load, malnutrition, and antitumor therapies may all potentially contribute to an ongoing inflammatory response in patients with malignancy. It is yet to be documented that these patients uniformly have an alteration of the cytokine response.

Peritonitis Peritonitis is a frequent indication for urgent surgical intervention. Some reports have demonstrated improved survival in patients capable of mounting adequate local cytokine responses during episodes of intraabdominal sepsis.75 Patients undergoing surgery for peritonitis exhibit an enhanced systemic inflammatory response, even after the initial pathology is properly managed.76,77 It is hypothesized that the response to surgery is additive to the ongoing cytokine response mounted in peritonitis.78,79 Whether this intact inflammatory response correlates to improved outcomes can only be inferred.

Cytokine Response in Abdominal Surgery

29

Hemorrhage Blood loss of varying degrees often accompanies abdominal surgery. While small volume loss is generally well tolerated, greater blood loss and resuscitation perioperatively appears to augment the inflammatory cytokine response.80 In addition, any element of ischemia or shock likely predisposes to antigen translocation across the gut barrier, and this further serves to activate the cytokine cascade. Moreover, an insult such as gut ischemia and reperfusion during surgery generate cytokines such as IL-6, IL-8 that “prime” circulating Ms and PMNs to enhance immunoactivity.81,82 These primed immunocytes, when subjected to a subsequent insult such as endotoxin, may precipitate remote organ injury (e.g., renal failure, ARDS, cardiotoxicity).88 The implications of these responses are applicable in abdominal vascular surgery, major hepatic and pancreatic surgeries, and preoperative gastrointestinal bleeding.

Anesthesia Important advances have been made relating cytokine response to various anesthetic agents. While it is well-documented that effective local/spinal anesthesia attenuates the metabolic and stress response for some abdominal surgeries, it remains to be determined that this similarly influences cytokine activity.83 Studies comparing epidural-general anesthesia to general anesthesia alone demonstrated no change in cytokine response. However, one study reporting humoral inhibition by adding intravenous prednisolone to epidural-general anesthesia was effective in decreasing the global stress response to surgery, including diminished IL-6 secretion.84 These data may suggest that the cytokine response during abdominal surgery is not the result of nociceptive (afferent neural or pain impulse) stimulus. Ketamine has often been advocated in the septic patient for its role in attenuating endotoxin-induced TNF-! production as well as for its cardiovascular stabilizing effects.85 This also requires confirmation by outcome analysis. Anesthesia using propofol and alfentanil have been shown to diminish the expression of IL-6 postoperatively.86 Whereas surgical injury favors the predominance of immunosuppressive TH2 cell population, the use of propofol may partially counter this effect.87

Minimally-Invasive Surgery The paradigm of minimally-invasive surgery is the laparoscopic cholecystectomy. In general, laparoscopic surgery for cholecystectomy is associated with a lesser inflammatory cytokine response when compared to an open technique.88-91 The primary marker for these observations has been IL-6, implying that wound size and the amount of visceral manipulation may be important determinants of the cytokine response.92-94 The diminished inflammatory cytokine response has been hypothesized to be a function of altered peritoneal milieu, which impairs immunocyte activity.95-97 One study has demonstrated that carbon dioxide (CO2) insufflation in humans impairs lymphocyte function for a period of two days secondary to an acidified peritoneal environment.98 The conversion from laparoscopic to open cholecystectomy has also been linked with a shift from TH1 to TH2 immunity and an associated alteration in the pattern of cytokine release.25 Data on cytokine responses to other laparoscopic procedures is scarce. Therefore, no studies are available as yet to address the cytokine response in technically

30

Cytokines and the Abdominal Surgeon

more complex surgery, such as laparoscopic splenectomy, gastric, vascular or endocrine procedures.

Conclusions It is evident that the inflammatory cytokine response is an integral facet of abdominal surgery. The cytokine response results not only from the surgical injury, but is also influenced by the patient’s preoperative status. Cytokine responses following surgery in some immunosuppressed populations and AIDS patients have yet to be addressed. The cytokine mediator response to minimally invasive surgeries remains to be explored from a mechanistic and clinical outcome perspective. Preoperative preparation is perhaps equally important to surgical technique in the optimization of a patient’s surgical outcome. This may include the use of antibiotics, proper nutritional and hemodynamic support. An enhanced understanding of cytokine response in abdominal surgery with respect to certain clinical conditions may serve to minimize any excessive systemic inflammation and improve the surgical outcome. As yet, there is no evidence to suggest that manipulations of the proinflammatory cytokine cascade will provide any clinical benefit to the surgical patient. Supported in part by National Institutes of Health grant GM 34695. Dr. Lin is also supported by The Surgical Society of The New York Hospital Queens.

References 1. Lin E, Calvano SE, Lowry SF. Disordered apoptosis as a mechanism for adverse outcome. In: Vincent JL, ed. Yearbook of Intensive Care and Emergency Medicine, 1997:91-99. 2. Marshall JC, Cook DJ, Christou NV et al. Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med 1995; 23:1638-1652. 3. Damas P, Canivet JL, Groote DD et al. Sepsis and serum cytokine concentrations. Crit Care Med 1997; 25:405-412. 4. Van der Poll T, Lowry SF. Tumor necrosis factor in sepsis: Mediator of multiple organ failure or essential part of host defense? Shock 1995; 3(1):1-12. 5. Lowry SF. Cytokines and their soluble antagonists in sepsis and critical illness. In: Tellado JM, Forse RA, Solomkin JS, eds. Prog Surg 1995; 20:115-124. 6. Rogy MA, Coyle SM, Oldenburg HSA et al. Persistently elevated soluble tumor necrosis factor receptor and interleukin-1 receptor antagonist levels in critically ill patients. J Am Coll Surg 1994; 178:132-138. 7. Van Zee KJ, Kohno T, Fischer E et al. Tumor necrosis factor soluble receptors circulate during experimental and clinical inflammation and can protect against excessive tumor necrosis factor in vitro and in vivo. Proc Natl Acad Sci USA 1992; 89:4845-4849. 8. Van der Poll T, Lowry SF. Endogenous mechanisms regulating TNF and IL-1 during sepsis. In: Vincent JL, ed. Yearbook of Intensive Care and Emergency Medicine. 1995:385-397. 9. Fong Y, Moldawer LL, Shires GT et al. The biologic characteristics of cytokines and their implications in surgical injury. Surg Gyn Obst 1990; 170:363-378. 10. Keogh C, Fong Y, Marano MA et al. Identification of a novel tumor necrosis factor/Cachectin from the livers of burned and infected rats. Arch Surg 1990; 125:79-85. 11. Fischer E, Marano MA, Barber AE et al. Comparison between effects of interleukin-1 administration and sublethal endotoxemia in primates. Am J Physiol 1991; 261:R442R448.

Cytokine Response in Abdominal Surgery

31

12. Lowry SF, Calvano SE. Soluble cytokine and hormonal mediators of immunity and inflammation. In: Howard RJ, Simmons RL, eds. Surgical Infectious Diseases. 3rd ed. Appleton and Lange, 1995; 313-326. 13. Fischer E, Van Zee KJ, Marano M et al. Interleukin-1 receptor antagonist circulates in experimental inflammation and in human disease. Blood 1992; 79: 2196-2200. 14. Van der Poll T, van Deventer SJH, ten Cate H et al. Tumor necrosis factor is involved in the appearance of interleukin 1 receptor antagonist in endotoxemia. J Infect Dis 1994; 169:665-667. 15. Ertel W, Faist E, Nestle C. Kinetics of interleukin-2 and interleukin-6 synthesis following major mechanical trauma. J Surg Res 1990; 48:622-628. 16. Faist E, Schinkel C, Zimmer S. Update on the mechanisms of immune suppression of injury and immune modulation. World J Surg 1996; 20:454-459. 17. Abraham E, Regan RF. The effects of hemorrhage and trauma on interleukin 2 production. Arch Surg 1983; 120:1341-1346. 18 Mack VE, McCarter MD, Naama HA et al. Dominance of T-helper 2-type cytokines after severe injury. Arch Surg 1996; 131:1303-1309. 19. Oka M, Hirazawa K, Yamamoto K et al. Induction of Fas-mediated apoptosis on circulating lymphocytes by surgical stress. Ann Surg 1996; 223(4):434-440. 20. Decker D, Schondorf M, Bidlingmaier F et al. Surgical stress induces a shift in the type-1/type-2 T-helper cell balance, suggesting down-regulation of cell-mediated and up-regulation of antibody-mediated immunity commensurate to trauma. Surgery 1996; 326-325. 21. Keegan AD, Ryan JJ, Paul WE. IL-4 regulates growth and differentiation by distinct mechanisms. The Immunologist 1996; 4(6):194-198. 22. Mangan DF, Baldwin R, Wahl SM. IL-4 enhances programmed cell death (apoptosis) in stimulated human monocytes. J Immunol 1992; 148(6):1812-1816. 23. Mangan DF, Wahl SM. Differential regulation of human monocyte programmed cell death (apoptosis) by chemotactic factors and proinflammatory cytokines. J Immunol 1991; 147(10):3408-3412. 24. Biffl WL, Moore EE, Moore FA. Interleukin-6 in the injured patient. Ann Surg 1996; 224(5):647-664. 25. Baigrie RJ, Lamont PM, Kwiatkowski D. Systemic cytokine response after major surgery. Br J Surg 1992; 79:757-760. 26. Cruickshank AM, Fraser WD, Burns HJG et al. Response of serum interleukin-6 in patients undergoing elective surgery of varying severity. Clin Sci 1990; 79:161-165. 27. Bitterman H, Kinarty A, Lazarovich H et al. Acute release of cytokines is proportional to tissue injury induced by surgical trauma and shock in rats. J Clin Immunol 1991; 11(4):184-192. 28. Oldenburg HSA, Rogy MA, Lazarus DD et al. Cachexia and the acute-phase protein response in inflammation are regulated by interleukin-6. Eur J Immunol 1993; 23:1889-1894. 29. Berger D, Bolke E, Seidelmann M et al. Time-scale of interleukin-6, myeloid related proteins (MRP), C reactive protein (CRP), and endotoxin plasma levels during the postoperative acute phase reaction. Shock 1997; 7(6):422-426. 30. Lee A, Whyte KB, Haslett C. Inhibition of apoptosis and prolongation of neutrophil functional longevity by inflammatory mediators. J Leukocyte Biol 1993; 54:283-288. 31. Biffl WL, Moore EE, Moore FA et al. Interleukin-6 delays neutrophil apoptosis. Arch Surg 1996; 131:24-30. 32. Tilg H, Trehu E, Atkins MB. Interleukin-6 (IL-6) as an antiinflammatory cytokine: induction of circulating IL-1 receptor antagonist and soluble tumor necrosis factor receptor p55. Blood 1994; 83:113-118.

32

Cytokines and the Abdominal Surgeon

33. Patrick DA, Moore FA, Moore EE et al. The inflammatory profile of interleukin-6, interleukin-8, and soluble intercellular adhesion molecule-1 in postinjury multiple organ failure. Am J Surg 1996; 172:425-431. 34. Van Zee KJ, Fischer E, Hawes AS et al. Effects of intravenous IL-8 administration in nonhuman primates. J Immunol 1992; 148(6):1746-1752. 35. Gerard C, Bruyns C, Marchant A et al. Interleukin 10 reduces the release of tumor necrosis factor and prevents lethality in experimental endotoxemia. J Exp Med 1993; 177:547-550. 36. Van der Poll T, Marchant A, Keogh CV et al. Interleukin-10 impairs host defense in murine pneumococcal pneumonia. J Infect Dis 1996; 174:994-1000. 37. Van der Poll T, Marchant A, Buurman WA et al. Endogenous IL-10 protects mice from death during septic peritonitis. J Immunol 1995; 155:5397-5401. 38. Trinchieri G. Interleukin-12: a cytokine produced by antigen-presenting cells with immunoregulatory functions in the generation of T-helper cells type 1 and cytotoxic lymphocytes. Blood 1994; 84:4008-4027. 39. Jelinek DF, Braaten JK. Role of IL-12 in human B lymphocyte proliferation and differentiation. J Immunol 1995; 154:1606-1613. 40. Heinzel FP, Rerko DM, Ling P et al. Interleukin 12 is produced in vivo during endotoxemia and stimulates synthesis of gamma interferon. Infect Immunol 1994; 62:4244-4249. 41. Barbul A, Regan MB. The regulatory role of T lymphocytes in wound healing. J Trauma 1990; 30:S97-S102. 42. Redmond HP, Hofmann K, Shou J et al. Effects of laparotomy on systemic macrophage function. Surgery 1992; 111:647-655. 43. Zellweger R, Ayala A, Zhu XL et al. Effect of surgical trauma on splenocyte and peritoneal macrophage immune function. J Trauma 1995; 39(4):645-650. 44. Klava A, Windsor ACJ, Farmery SM et al. Interleukin-10: a role in the development of postoperative immunosuppression. Arch Surg 1997; 132:425-429. 45. Roumen RH, Hendriks T, van der Ven-Jongekrijg J et al. Cytokine patterns in patients after major vascular surgery, hemorrhagic shock, and severe blunt trauma. Ann Surg 1993; 218(6):769-776. 46. Barry MC, Kelly C, Burke P et al. Immunological and physiological responses to aortic surgery: effect of reperfusion on neutrophil and monocyte activation and pulmonary function. Br J Surg 1997; 84:513-519. 47. Meduri GU, Headley S, Kohler G et al. Persistent elevation of inflammatory cytokines predicts a poor outcome in ARDS. Chest 1995; 107:1062-1073. 48. Ogle CK, Guo X, Hasselgren PO et al. The gut as a source of inflammatory cytokines after stimulation with endotoxin. Eur J Surg 1997; 163:45-51. 49. Fong Y, Marano MA, Moldawer LL et al. The acute splanchnic and peripheral tissue metabolic response to endotoxin in humans. J Clin Invest 1990; 85:1896-1904. 50. Mowat AM, Viney JL. The anatomical basis of intestinal immunity. Immunol Rev 1997; 156:145-166. 51. Reynolds JV, O’Farrelly C, Feighery C et al. Impaired gut barrier function in malnourished patients. Br J Surg 1996; 83:1288-1291. 52. Deitch EA. Simple intestinal obstruction causes bacterial translocation in man. Arch Surg 1989; 207:549-554. 53. Palmer HR, Duerden BI, Holdsworth CD. Bacteriological and endotoxin studies in cases of ulcerative colitis submitted to surgery. Gut 1980; 21:851-854. 54. Sher ME, D’Angelo AJ, Stein TA et al. Cytokines in Crohn’s colitis. Am J Surg 1995; 169(1):133-136. 55. Brinkmann A, Wolf CF, Berger D et al. Perioperative endotoxemia and bacterial translocation during major abdominal surgery: evidence for the protective effect of endogenous prostacyclin? Crit Care Med 1996; 24(8):1293-1301.

Cytokine Response in Abdominal Surgery

33

56. Enayati P, Brennan MF, Fong Y. Systemic and liver cytokine activation. Arch Surg 1994; 129:1159-1164. 57. Kimura F, Miyazaki M, Suwa T et al. Reduced hepatic acute-phase response after simultaneous resection for gastrointestinal cancer with synchronous liver metastases. Br J Surg 1996; 83:1002-1006. 58. Ferri M, Gabriel S, Gavelli A et al. Bacterial translocation during portal clamping for liver resection. Arch Surg 1997; 132:162-165. 59. Fujita T, Kobayashi S, Tomoyuki S et al. Relationship between circulating secretory immunoglobulin A levels and portal blood cytokine levels during major abdominal surgery. Arch Surg 1997; 132:124-127. 60. Bailey ME. Endotoxin, bile salts and renal function in obstructive jaundice. Br J Surg 1976; 63:774-778. 61. Van Leeuwen PAM, Hong RW, Rounds JD et al. Hepatic failure and coma after liver resection is reversed by manipulation of gut contents: The role of endotoxin. Surgery 1991; 110:169-175. 62. Boermeester MA, Straatsburg IH, Houdijk APJ et al. Endotoxin and interleukin-1 related hepatic inflammatory response promotes liver failure following partial hepatectomy. In: Boermeester MA (ed). Pathogenesis and Modulation of the Systemic Inflammatory Response in Endotoxemia and Sepsis. 1994; 63-84. 63. Kondo T, Suda T, Fukuyama H et al. Essential roles of the Fas ligand in the development of hepatitis. Nature Med 1997; 3(4):409-413. 64. Wang P, Ba ZF, Chaudry IH. Mechanism of hepatocellular dysfunction during early sepsis. Arch Surg 1997; 132:364-370. 65. Yanaga K, Yamamoto TK, Nishizaki T et al. Cardiac complications after hepatic resection. Br J Surg 1996; 83:1448-1451. 66. Braxton CC, Coyle SM, Montegut WJ et al. Parenteral nutrition alters monocyte TNF receptor activity. J Surg Res 1995; 59:23-28. 67. Windsor ACJ, Klava A, Somers SS et al. Manipulation of local and systemic host defence in the prevention of perioperative sepsis. Br J Surg 1995; 82:1460-1467. 68. Fong Y, Marano M, Barber A et al. Total parenteral nutrition and bowel rest modify the metabolic response to endotoxin in humans. Ann Surg 1989; 210(4):449-457. 69. Wigmore SJ, Fearon KCH, Ross JA. Modulation of human hepatocyte acute phase protein production in vitro by n-3 and n-6 polyunsaturated fatty acids. Ann Surg 1997; 225(1):103-111. 70. Li J, Kudsk KA, Janu P et al. Effect of glutamine-enriched total parenteral nutrition on small intestinal gut-associated lymphoid tissue and upper respiratory tract immunity. Surgery 1997; 121:542-549. 71. Parry-Billings M, Baigrie RJ, Lamont PM et al. Effects of major and minor surgery on plasma glutamine and cytokine levels. Arch Surg 1992; 127:1237-1240. 72. Ng EH, Lowry SF. Nutritional support and cancer cachexia. Hem Onc Clin N Am 1991; 5(1):161-184. 73. Mealy K. Proinflammatory cytokines and the host response to cancer. In: Vincent JL ed. Yearbook of Intensive Care and Emergency Medicine 1995; 398-412. 74. Sherry B, Gelin J, Fong Y et al. Anticachectin/tumor necrosis factor-alpha antibodies attenuate the development of cachexia in two murine transplantable tumor models. FASEB J 1989; 3:1556-1562. 75. Riche F, Panis Y, Laisne MJ et al. High tumor necrosis factor serum level is associated with increased survival in patients with abdominal septic shock: A prospective study in 59 patients. Surgery 1996; 120:801-807. 76. Sautner T, Gotzinger P, Redl-Wenzl EM et al. Does reoperation for abdominal sepsis enhance the inflammatory host response? Arch Surg 1997; 132:250-255. 77. Schein M, Wittmann DH, Holzheimer R et al. Hypothesis: Compartmentalization of cytokines in intraabdominal infection. Surgery 1996; 119:694-700.

34

Cytokines and the Abdominal Surgeon

78. Hammond JMJ, Potgieter PD. The influence of surgery on cytokines in patients with intraabdominal sepsis. Anaesth Intens Care 1996; 24:430-434. 79. Tang GJ, Kuo CD, Yen TC et al. Perioperative plasma concentrations of tumor necrosis factor and interleukin-6 in infected patients. Crit Care Med 1996; 24:423-428. 80. Ayala A, Lehman DL, Herdon CD et al. Mechanism of enhanced susceptibility to sepsis following hemorrhage. Arch Surg 1994; 129:1172-1178. 81. Ayala A, Wang P, Ba ZF et al. Differential alterations in plasma IL-6 and TNF levels after trauma and hemorrhage. Am J Physiol 1991; 260:R167-R171. 82. Botha AJ, Moore FA, Moore EE et al. Sequential systemic platelet-activating factor and interleukin-8 primes neutrophils in patients with trauma at risk of multiple organ failure. Br J Surg 1996; 83:1407-1412. 83. Norman JG, Fink GW. The effects of epidural anesthesia on the neuroendocrine response to major surgical stress: A randomized prospective trial. Am Surgeon 1997; 63:75-80. 84. Schulze S, Sommer P, Bigler D et al. Effect of combined prednisolone, epidural analgesia, and indomethacin on the systemic response after colonic surgery. Arch Surg 1992; 127:325-331. 85. Schmidt H, Ebeling D, Bauer H et al. Ketamine attenuates endotoxin-induced leukocyte adherence in rate mesenteric venules. Crit Care Med 1995; 23:2008-2014. 86. Crozier TA, Muller JE, Quittkat D et al. Effect of anaesthesia on the cytokine responses to abdominal surgery. Br J Anaesth 1994; 72:280-285. 87. Salo M, Pirttikangas CO, Pulkki K. Effects of propofol emulsion and thiopentone on T helper cell type-1/type-2 balance in vitro. Anaesth 1997; 52:341-344. 88. Chambrier C, Chassard D, Bienvenu J et al. Cytokine and hormonal changes after cholecystectomy. Ann Surg 1996; 224(2):178-182. 89. Jakeways MSR, Mitchell V, Hashim IA et al. Metabolic and inflammatory responses after open or laparoscopic cholecystectomy. Br J Surg 1994; 81:127-131. 90. Deuss U, Dietrich J, Kaulen D et al. The stress response to laparoscopic cholecystectomy: investigation of endocrine parameters. Endoscopy 1994; 26:235-238. 91. Mealy K, Gallagher H, Barry M et al. Physiological and metabolic responses to open and laparoscopic cholecystectomy. Br J Surg 1992; 79:1061-1064. 92. Glaser F, Sannwald GA, Buhr HJ et al. General stress response to conventional and laparoscopic cholecystectomy. Ann Surg 1995; 221:372-380. 93. Joris J, Cigarini I, Legrand M et al. Metabolic and respiratory changes after cholecystectomy performed via laparotomy or laparoscopy. Br J Anaesth 1992; 69:341-345. 94. Iwanaka T, Arkovitz M, Arya G et al. Evaluation of operative stress and peritoneal macrophage function in minimally invasive operations. J Am Coll Surg 1997; 184:357-363. 95. Targarona E, Pons MJ, Balague C et al. Acute phase is the only significantly reduced component of the injury response after laparoscopic cholecystectomy. World J Surg 1996; 20:528-534. 96. Redmond HP, Watson WG, Houghton T et al. Immune function in patients undergoing open vs laparoscopic cholecystectomy. Arch Surg 1994; 129:1240-1246. 97. Kobayashi E, Yoshida T, Yamauchi H et al. Immune function in patients undergoing open vs laparoscopic cholecystectomy. (letter) Arch Surg 1995; 130:676. 98. Evrard S, Falkenrodt A, Park A et al. Influence of CO2 pneumoperitoneum on systemic and peritoneal cell-mediated immunity. World J Surg 1997; 21:353-357.

Cytokines and the Abdominal Operative Wound

35

CHAPTER 4

Cytokines and the Abdominal Operative Wound Frank J. Thornton and Adrian Barbul

Introduction

P

rofound advances in our understanding of the healing process have occurred in the past decade.1,2 The concept of endogenous cytokine release which synchronizes the complex processes of cellular proliferation and matrix deposition within wounds has brought us closer to a therapeutic approach to wound healing.3-5 Laparotomy wounds include several layers: skin, subcutaneous tissue, fascia, muscle and peritoneum. In the past it was considered essential to carry out closure of the abdominal wall in individual layers, with particular emphasis on closure of the peritoneal layer.6-8 More recent animal and clinical studies have shown that healing of a midline incision occurs through formation of a dense fibrous scar that juxtaposes the two surfaces of the wound en masse.9

Overview of the Repair Process The normal repair process is a complex series of interactive processes including inflammation, fibroplasia with collagen deposition and maturation.2 Sequentially, repair begins with exposure of thrombocytes to exposed collagen, followed rapidly by chemotaxis of inflammatory cells and fibroblasts into the wound site. Next, activation of macrophages and fibroblasts results in matrix production and continued proliferation of fibroblasts until wound healing is complete (Fig. 4.1.). Cytokines have the potential to initiate, sustain and terminate these complex biological events resulting in successful wound repair.5,10-15 This review describes the temporal sequence of cellular events occuring in the abdominal surgical wound and the role cytokines play in orchestrating the process. The cytokines known to contribute to the healing process include platelet-derived growth factor (PDGF),16,17 platelet factor 4,18-21 transforming growth factors-! (TGF- !)22 and ∀ (TGF-∀), 23 fibroblast growth factors (basicFGF/FGF-2 and acidicFGF/FGF-1),24-26 vascular endothelial growth factor (VEGF),27,28 keratinocyte growth factor (KGF),29 IFN-#,30 interleukins 1, 2, 4, 6, 8,10, and 1231-35 and tumor necrosis factor-! (TNF-!).36-38 These peptides mediate their effect via specific receptors located on target cells. They can act locally on adjacent cells (paracrine), travel systemically (endocrine) or exert their effect on the cell of origin (autocrine or intracrine). Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

36

Cytokines and the Abdominal Surgeon

Fig. 4.1. Scheme of cellular involvement in the wound healing process.

Platelet Degranulation Performance of a laparotomy wound is followed immediately by bleeding at the tissue edges. Platelet contact with the exposed collagen of the damaged tissue results in adherence and aggregation of these thrombocytes as well as a release reaction during which !-granules within the platelets release platelet factor-4 (PF-4), ∀ -thromboglobulin, PDGF, and TGF- ∀ , as well as other prostaglandins and leukotrienes. PDGF is derived predominantly from platelets and activated macrophages but also from keratinocytes, smooth muscle cells, activated fibroblasts and endothelial cells.17 PDGF is chemotactic to monocytes, neutrophils, fibroblasts and smooth muscle cells.39 It activates inflammatory cells further, promoting the inflammatory response of healing and stimulates collagenase production by fibroblasts which help in debriding the wound edges of nonviable tissue. Collagenase production may in turn initiate the early decrease in wound breaking strength experienced in uncomplicated repair.40 PDGF is also strongly mitogenic to mesenchymal cells, such as fibroblasts and smooth muscle cells, inducing resident fibroblasts around the wound to proliferate and modulate their phenotype as repair proceeds.41 PF-4 is also strongly chemotactic to inflammatory cells18 and fibroblasts as is ∀-thromboglobulin and TGF-∀.42 PDGF has also been implicated in inducing transcription of genes encoding cytokines necessary for directing cell activation, migration and communication. In rodents, the JE gene has been shown to play a critical role in the wound healing cascades and has a human homolog which encodes the monocyte chemotactic factor-1 (MCF-1).16 PDGF induces transcription of this gene, as well as others including several platelet !-granule proteins, together named the superfamily of small in-

Cytokines and the Abdominal Operative Wound

37

ducible genes (SIG). IL-8, an activator and chemotactic factor for neutrophils and T lymphocytes43,44 is another member of the SIG family and protects endothelial cells from neutrophil contact thus protecting the endothelial surface against neutrophilmediated damage.45 Macrophage inflammatory proteins (MIP-1 and 2) are chemotactic to, and induce degranulation of polymorphonuclear cells further stimulating the inflammatory reaction46 as well as recruiting monocytes to the wound site.47 Platelet basic protein, ∀-thromboglobulin, connective tissue-activating peptide III (CTAP-III) and neutrophil-activating peptide-2 (NAP-2) are variant forms derived from a common gene product. CTAP-III is mitogenic for human dermis48 while monocyte derived NAP-2 stimulates the release of elastase from human neutrophils.49

Immune Cell Influx After haemostasis is achieved the inflammatory process continues up until the third postoperative day. Cytokine induced chemotactic forces attract neutrophils into the wound with subsequent phagocytosis of bacteria. However, in the absence of significant surgical site infection, these cells are not essential for wound healing to occur.50 Monocytes appear in the abdominal wound within 48-96 hours of surgery to become the wound macrophages51 and are essential to the healing process.52-54 Their influx into the wound and eventual role in the repair process depends largely on the generation of a diverse range of mononuclear phagocyte chemoattractants in the wound site.55 PDGF and TGF-∀ are nonspecific chemoattractants produced during the platelet release reaction39 recruiting monocytes, neutrophils and many other cells to the wound site. Macrophage Chemotactic Protein (MCP-1) is produced by macrophages themselves as well as fibroblasts and endothelial cells in response to stimulation with IFN-#, TNF-!, IL-1∀ and LPS.56,57 Therefore, not only do monocytes respond to specific chemoattractants but they themselves, once established in the wound, secrete a repertoire of cytokines to further amplify the inflammatory process and selectively recruit cells to the wound. As previously stated PDGF and TGF-∀ while primarily derived from platelet a-granules are also secreted later in the repair process by macrophages to function as coordinators of matrix remodeling and angiogenesis as well as chemotactic factors. IL-8, produced by macrophages, also participates in angiogenesis.58 Many of the cytokines and growth factors secreted by fibroblasts, macrophages and the other wound cells display pleitrophic effects being influential in cell migration and mitosis, matrix deposition and remodelling as well as angiogenesis. PDGF and TGF-∀ are among these growth factors as well as insulinlike growth factor (IGF-1), FGF and the interleukin family of cytokines. The central role of lymphocyte in the wound is now well recognised.59-61 Lymphocytes like macrophages appear to regulate wound healing through cytokine release.62

Transforming Growth Factor (TG-∀) TGF-∀ is released from platelet a-granules in a latent form which is transformed to its active form by plasmin, a product of the coagulation cascade.63 It exists in five isoforms with the relative abundance of these subtypes determining the degree of scarring seen in the mature wound. TGF-∀1 and ∀2 are implicated in scarring, while TGF-∀3 is a potential antiscarring agent.64 TGF-∀1 is the predominant isoform in wound fluid65 and affects almost all pathways in the wound repair process. In the active form TGF-∀ is strongly chemotactic for inflammatory cells and fibroblasts, with this effect being lost above or below an “optimum” concentration.66-68 The effect of TGF-∀ on fibroblast proliferation is still not completely understood. In vitro

38

Cytokines and the Abdominal Surgeon

studies have demonstrated inhibition of proliferation.69 TGF-∀ also induces PDGF production by fibroblasts and induces fibronectin, procollagen type I and glycosaminoglycan synthesis in these cells. It exerts chemoattractant forces on cells required to potentiate repair in the wound and subsequently stimulates granulation tissue formation by these cells. Probably its greatest influence is on matrix formation through its action on the fibroblast cell. TGF-∀ regulates genetic transcription of many matrix proteins including collagen, fibronectin, glycosaminoglycans.70 Also, some of the negative effects of antineoplastic agents on the healing process are mediated through changes in TGF-∀ expression either at a transcriptional or a post-transcriptional level e.g., mRNA expression for TGF-∀ is reduced in adriamicin-impaired skin wounds71 and exogenous TGF-∀ reverses this effect in experimental wound chambers.72 The collagen deposition promoted by TGF-∀ has been shown in experimental models to be paralleled by increased breaking strength of the wound.73 TGF-∀ is also inhibitory to endothelial cell proliferation despite its ability to stimulate angiogenesis when injected in vivo.74

Interleukins In the wound site IL-1 is predominantly a product of the monocyte population. It regulates connective tissue regeneration by stimulating fibroblast proliferation and fibroblast synthesis of collagen (in absence of serum),75,76 fibronectin,77 hyaluronic acid78 and elastin.79 It acts as a costimulus with TNF-! to produce plasminogen activator80 and collagenase33,81 both of which help in the digestion of pericellular ground substance and thus promote increased cell motility which is especially important in epithelialization. IL-2 also participates in wound healing by increasing collagen synthesis.32 The effect of blood transfusion on the repair process has been ascribed to diminution of wound IL-2 levels.82,83 IL-6 has been implicated in suppressing fibroblast proliferation.35 Recent evidence also suggests that reoperative trauma for cases of abdominal sepsis induces an early postoperative increase in IL-6 levels.84 This surge in IL-6 production has been associated with postoperative hypotension often seen in these septic patients and may contribute to the resultant hypoxia in the wound and to compromised healing. IL-4 and IL-10 are inhibitors of macrophage metalloproteinases, they also suppress iNOS production while simultaneously inducing arginase.85 TGF-∀ similarly downregulates iNOS expression while incresing arginase activity. IFN-# has the opposite effect.86 Shearer et al have interpreted this as a regulatory phenomenon mediated by inflammatory cytokines, creating an NO regulated cytotoxic environment early in the wound healing process. This is followed by an increase in arginase activity later in the repair process which produces an environment more favorable to fibroblast proliferation and collagen synthesis. IL-12 is a potent inhibitor of angiogenesis87 possibly as a result of its induction of IFN-#.88

Interferon-# IFN-# is another major participant in the repair process. It selectively decreases collagen production in fibroblasts and type I and III procollagen mRNA levels while inducing fibronectin at a transcriptional level.30,89,90 Similar down-regulation of collagen synthesis has been shown in vivo.91 Along with other lymphokines, IL-4 and IL-10, it inhibits matrix metalloproteinase production by macrophages but has little effect on enzyme activity in other wound cells.34,92,93 It also participates in the induc-

Cytokines and the Abdominal Operative Wound

39

tion of the inducible isoform of nitric oxide synthase (iNOS) and may exert its effects in the wound through the free radical NO, the by-product of iNOS activity. This latter pathway may also be responsible for the activation and increased cytotoxicity seen in macrophages stimulated by IFN-#.94

Tumor Necrosis Factor-! TNF-! in the wound serves to regulate macrophage proliferation while differentiation occurs,95 inhibits collagen synthesis38,96 and at higher concentrations increases collagenase production97-99 (Fig. 4.2). Other activities of TNF-! include chemotaxis and invasive migration of fibroblasts into the wound,36,100 induction of other proinflammatory cytokines, including IL-1 and specialized cytokines IL-6 and IL-8 thus promoting the early inflammatory response essential to normal repair.13,101

Nitric Oxide The precise role of the free radical, nitric oxide (NO) in wound healing still defies full explanation. NO synthesis is inducible by cytokines and lipopolysaccharide, in particular IL-1, TNF-! and IFN-#.102 It is synthesised in wounds by the inducible isoform of the enzyme nitric oxide synthase (iNOS). Hypoxia, which is inevitably present in laparotomy wounds is a costimulus with inflammatory cytokines for the production of NO in macrophages.103 Macrophage activation in an anoxic environment has also been shown to increase TNF-! and IL-6 production.104 Our laboratory has demonstrated decreased wound healing in murine cutaneous wounds after systemic inhibition of inducible nitric oxide synthase (iNOS).105 Collagen synthesis by wound fibroblasts also varies with NO concentration, collagen synthesis increasing at low concentrations of NO and decreasing at higher concentrations.106 This finding may reflect the in vivo scenario seen in systemic sepsis where we have shown decreased collagen synthesis in colonic anastomoses paralleled by a large increase in iNOS activity.107 Overall, these results suggest that there exists a level of NO at which wound healing is maximized but above or below which the process is suboptimal. More recently Schäffer et al have demonstrated that wound fibroblasts are phenotypically altered during the healing process to synthesize NO which in turn, regulates their collagen synthetic and contractile activities.108 In human keratinocytes iNOS activity is induced by IFN-# while IFN-! and TGF-∀ antagonize this effect.109 NO inhibits endothelial cell proliferation and bFGF-induced phenotypic modulations.110 Also, TGF-∀ down-regulates iNOS expression in vivo.111 These findings further support the view that an intricate web of interactions exists between cytokines and NO in the healing process. Further work is necessary to clarify the biological and clinical relevance of these interactions.

Extracellular Matrix Production Fibroblasts, the primary source of extracellular matrix, take on a range of functions as they evolve with the healing wound. During the first 3 days after wounding, fibroblasts refrain from migrating into the wound site. Instead they proliferate at the wound edge under the potent mitogenic influence of PDGF.17 The failure of these cells to migrate into the wound site during this time is not fully understood. However, the presence of fibronectin surface receptors and a noncompliant cytoskeletal framework have been proposed as causative factors.112 A phenotypic change occurs after day 3 resulting in fibroblast invasion into the wound site. The primary difference in these cells from the proliferating phenotype is the expression of vitronectin receptors, which has been shown by in vitro studies to

40

Cytokines and the Abdominal Surgeon

be induced by FGF.113,114 The time for adequate FGF accumulation and subsequent induction of this receptor to occur may explain the 3-4 day delay in fibroblast migration. The wound fibroblasts begin immediately to secrete matrix proteins. Wound cytokines influence the fibroblast rate of synthesis of these proteins which constitute the ECM (Table 4.1). Unfortunately most information concerning the regulation of fibroblast metabolism and collagen secretion is derived from in vitro work, with findings dependent on the presence or absence of serum. Much less evidence exists for their role in vivo. Interferon-# decreases collagen type I and III synthesis in fibroblasts at a transcriptional level30 with little effect on other proteins. TNF-! also reduces collagen synthesis while simultaneously inducing collagenase production.38 Its synergistic effect on collagen production with #-interferon has been suggested as a possible down-regulatory mechanism for matrix production towards the end of the repair process37,115 (Fig. 4.2). TGF-∀ and IL-1 have been found to enhance collagen synthesis in fibroblasts70,116,117 and furthermore have increased wound healing when applied topically. 118-120 In addition TGF- ∀ induces tissue inhibitor of metalloproteinase and blocks collagenase induction by other cytokines.121 The balance of cytokines in the wound during the first three days of healing may determine the collagenase activity and therefore the suture holding capacity of the wound. This is especially relevant to laparotomy wounds where the breaking strength in rats has been shown to decrease by 47% during the first 48 hours.40

Angiogenesis Due to vascular damage at the time of surgery the centre of a wound is hypoxic. New blood vessel formation or angiogenesis is a prerequisite to successful completion of wound repair.4,122 The soluble factors responsible for neo-vascularization remain incompletely defined. FGF-2 was the first angiogenesis factor to be isolated and acts as an endothelial cell mitogen, stimulating tube formation and proteinase secretion by endothelial cells as a prelimenary step in neo-capillary formation.26,123 The proteinases cause breakdown of the vascular basement membrane permitting vascular endothelial cell invasion during new vessel formation.124 Fibronectin is also essential to this process as a substrate which modulates endothelial cell shape fascilitating its growth.125 FGF is synthesised by many cells, including macrophages, keratinocytes and neutrophils. It has a high affinity for heparin as well as the syndecan family of cell surface receptors, both of which may influence the mobilization of bFGF. These syndecan receptors may also determine upon which cells in the wound site it exerts its regulatory influence.126-128 The mechanism by which FGF is secreted by the cell is incompletely elucidated.129 One theory suggests that heparin sulphate plays a role, and provides a mechanism of storage for FGF in the basement membrane.130 FGF possesses mitogenic properties when applied to keratinocytes with a potential role in re-epithelialization of the wound.131 In vivo studies have demonstrated more rapid re-epithelialization with topical application of FGF-2 while in combination with PDGF it facilitated earlier capillary ingrowth in wounds of diabetics.132,133 Other potential cytokine mediators of angiogenesis are IL-1, TGF-!, EGF, G-CSF, GM-CSF and TGF-∀74,134-136 while platelet factor 4 and IL-12 inhibit angiogenesis.87,137

Wound Contraction By day 7 fibroblasts have become the main cellular component of granulation tissue. At about this time fibroblasts take on characteristics normally seen in smooth

Cytokines and the Abdominal Operative Wound

41

Table 4.1. Cytokines-sources and functions in the wound Cytokine

Cellular Source in Wound Function in Repair Process

PDGF

Platelet, activated macrophage, activated fibroblast, keratinocyte, smooth muscle cell and endothelial cell

Chemotactic to monocytes, neutrophils, smooth muscle cells and fibroblasts. Activates inflammatory cells. Mitogenic to mesenchymal cells. Weakly angiogenic. Phenotypically alters fibroblast to myofibroblast. Stimulates collagenase synthesis by fibroblasts. Induction of the SIG family of genes.

TGF-∀

Platelet, lymphocyte, macrophage, keratinocyte, fibroblast, endothelial cell, smooth muscle cell

Strongly chemotactic to macrophages, neutrophils, lymphocytes and fibroblasts. Induces fibroblast proliferation and phenotypic alteration to myofibroblast form. Stimulates angiogenesis. Induces PDGF, procollagen, glycosaminoglycan and fibronectin synthesis by the fibroblast. Upregulates TIMPs production while blocking collagenase induction.

FGF -1 and -2 Basement membrane storage. Macrophage, neutrophil, keratinocyte, lymphocyte.

Strongly angiogenic: endothelial cell mitosis, tube formation and proteinase secretion. Mitogenic to keratinocytes.

IFN-#

Lymphocyte, macrophage

Enhances neutrophil and macrophage function. Antagonizes phenotypic change in fibroblast to its contractile counterpart. Decreases proliferation of fibroblasts and collagen synthesis. Induces fibronection transcription. Induces other cytokines and NO.

IL-1

Macrophage, keratinocyte, Lymphocyte differentiation and fibroblast, endothelial cell, activation. Increases collagenase and smooth muscle cell. plasminogen activator synthesis. Induces collagen synthesis in fibroblasts and their proliferation. Decreases synthesis of TIMPs. Induces keratinocyte migration.

IL-6

Keratinocyte, fibroblast, macrophage,endothelial cell, lymphocyte

continued...

Lymphocyte proliferation. Suppresses fibroblast proliferation. Has been implicated in causing wound hypoxia in sepsis (see text).

42

Cytokines and the Abdominal Surgeon

Table 4.1. (continued) Cytokine

Cellular Source in Wound Function in Repair Process

TNF-!

Macrophage, keratinocyte

Decreases collagen synthesis and increases collagenase activity. Stimulates cytokine production and NO.

EGF

Macrophage, platelet, keratinocyte

Stimulates proliferation and migration of epithelial cells.

TGF-!

Macrophage, keratinocyte Angiogenesis. (40% homology with EGF) Growth of epithelial cells, endothelial cells and fibroblasts.

GM-CSF

Keratinocyte,fibroblast, lymphocyte, endothelial cell

Indirectly induces phenotypic change of fibroblast to myofibroblast.

Nitric Oxide

Most mammalian cells when stimulated with LPS or inflammatory cytokines

Still poorly understood. Fibroblast collagen synthesis appears to vary with NO concentration as does fibroblast proliferation. Inhibits endothelial proliferation.

muscle cells, including !-smooth muscle actin expression, and have been named myofibroblasts.138-140 PDGF and TGF-∀ have been implicated in phenotypically changing the fibroblast into its myofibroblastic counterpart.41,141,142 GM-CSF also promote cell metamorphosis by accumulating macrophages in the wound site which in turn release cytokines that induce !-smooth muscle actin expression.143,144 Through their regulation of cytokine release ECM components may also control fibroblast phenotypic change.145,146 IFN-# was found to antagonize this myofibroblastic change by decreasing !-smooth muscle actin mRNA and protein147 which correlates well with this cytokines antiproliferative activity148 and decrease of collagen synthesis149,150 all of which appear to act as “quenching” factors in the repair process. Interferon-! and -∀ appear to have similar inhibitory effects on wound contraction.62

Epithelialization Epithelialization plays a minor role in the primary healing of uncomplicated laparotomy wounds because of the small surface area to be traversed by migrating epithelial cells. Nevertheless its successful completion establishes a barrier to contamination of the deeper wound layers.151 Reepithelialization commences within several hours of the injury and involves early migration of keratinocytes from the wound edge.152,153 These cells undergo phenotypic changes as they migrate over the underlying wound matrix, becoming flattened and elongated,154,155 expressing actin and myosin156 and loosing their junctional attachment to adjacent cells and basement membrane.153 Cytokines control many of these events and have been shown in clini-

Cytokines and the Abdominal Operative Wound

43

Fig 4.2. Inflammatory cytokines influence the net collagen production in the wound site through their effect on fibroblast proliferation, collagen synthesis and collagen degradation.

cal studies to augment reepithelialization. Epidermal growth factor (EGF), one of the first growth factors to be discovered, stimulates the proliferation and migration of epithelial cells157 as does TGF-!, a structurally similar cytokine.158 These cytokines exert their effect by increasing expresion of the collagen integrin !2∀1.159 IL-1 also induces keratinocyte migration but through a different pathway.119,160 TNF-! and IL-1 also upregulate keratinocyte secretion of a urokinase-type plasminogen activator (uPA) and expression of its cell-surface associated receptor uPA-R. Plasminogen that is also bound to a membrane binding site is converted by uPA to plasmin which is then available for proteolysis of pericellular glycoproteins.80 IL-1 also increases collagenase activity which also contributes to increased cell motility.81

Cytokines in Surgical Site Infection Surgical site infection increases the morbidity and mortality for the surgical patient especially when it develops into abdominal wound dehiscence. Knowledge of cytokine behaviour in the infected wound is incomplete. It is known that the bacterial presence in a wound induces thrombocytopenia.161,162 Since the primary source of wound cytokines in early repair is the platelet we can postulate that a temporal imbalance occurs in wound cytokine concentrations as a result of infection. Increased numbers of neutrophils and macrophages are also seen in infected wounds which further alter the balance of cytokines in the wound site. Hypoxia of the wound has been demonstrated to promote surgical site infection.163 Macrophage activation under hypoxic conditions induces production of increased quantities of the cytokines TNF-! and IL-1b and for prolonged periods.104 These cytokines act synergistically to increase the production of matrix metalloproteinases, while simultaneously decreasing the endogenous production of tissue inhibitor of metalloproteinases (TIMPs).164 The alterations in cytokine behaviour in surgical site sepsis needs to be properly elucidated before the application of exogenous recombinant cytokines as a therapeutic answer to the problem.

44

Cytokines and the Abdominal Surgeon

Conclusion In the context of wound healing endogenous cytokines act as multifunctional signaling peptides regulating a wide spectrum of activities. Their role as mediator between such divergent cell populations as the immune cells, fibroblasts, endothelial cells and keratinocytes establishes a harmonious environment which eliminates contaminant organisms and nonviable tissue, promotes matrix deposition, angiogenesis and re-epithelialization, with the ultimate goal of ensuring a rapid and secure wound closure especially in the case of laparotomy wounds where dehiscence is associated with high morbidity and mortality. Our ever increasing knowledge of cytokines and our ability to manufacture recombinant forms is bringing us to a new era in the management of the complicated or compromised wound.3,5 Factors to be remembered, however, before widespread application of topical cytokine therapy are firstly the appropriate timing and quantity applied since some cytokines work in a narrow dosage and time range and secondly, that administration of combinations of growth factors may be more therapeutic than single cytokine application. Also difficulties encountered in translating in vitro data to in vivo situations need to be overcome.165 Nevertheless, the future of advances in wound healing appear to lie in the domain of cytokines. As a contemporary, Ambroise Paré might rephrase—“I dressed the wound; God healed it—with cytokines”.166

References 1. Barbul A. Immune aspects of wound repair. [Review]. Clin Plast Surg 1990; 17:433-42. 2. Hammar H. Wound healing. [Review]. Int J Dermatol 1993; 32:6-15. 3. Pierce GF, Mustoe TA. Pharmacologic enhancement of wound healing. [Review]. Annu Rev Med 1995; 46:467-81. 4. Hockel M, Schlenger K, Doctrow S et al. Therapeutic angiogenesis. [Review]. Arch Surg 1993; 128:423-9. 5. Peschel C, Huber C, Aulitzky WE. Clinical applications of cytokines. [Review]. Presse Med 1994; 23:1083-91. 6. diZerega G.S. The peritoneum and its response to surgical injury. Prog Clin Biol Res 1990; 358:1-11. 7. Ellis H. Wound repair. Reaction of the peritoneum to injury. Ann R Coll Surg Engl 1978; 60:219-21. 8. Irvin TT, Koffman CG, Duthie HL. Layer closure of laparotomy wounds with absorbable and nonabsorbable suture materials. Br J Surg 1976; 63:793-6. 9. Bucknall TE, Ellis H. Wound Healing for Surgeons. 1st ed., Eastbourne: Bailliere Tindall, 1984. 10. Blitstein-Willinger E. The role of growth factors in wound healing. [Review]. Skin Pharmacol 1991; 4:175-82. 11. Hatamochi A, Mori K, Ueki H. Role of cytokines in controlling connective tissue gene expression. [Review]. Arch Dermatol Res 1994; 287:115-21. 12. Kulozik M, HC Mauch C, Scharffetter K, Krieg TH. Cytokine Regulation of Collagen Metabolism During Wound Healing in vitro and in vivo. In: Janssen H, Robertson JIS, eds. Wound Healing. Wrightson Biomedical Publishing Ltd., 1991. 13. Lowry, SF. Cytokine mediators of immunity and inflammation. [Review]. Arch Surg 1993; 128:1235-41. 14. McKay IA, Leigh IM. Epidermal cytokines and their roles in cutaneous wound healing. [Review]. Br J Dermatol 1991; 124:513-8. 15. Slavin J. The role of cytokines in wound healing. [Review]. J Pathol 1996; 178:5-10.

Cytokines and the Abdominal Operative Wound

45

16. Deuel TF, Kawahara RS, Mustoe TA, Pierce AF. Growth factors and wound healing: platelet-derived growth factor as a model cytokine. [Review]. Annu Rev Med 1991; 42:567-84. 17. Heldin C. Role of platelet-derived growth factor in vivo. In: Clark RAF, ed. The Molecular and Cellular Biology of Wound Repair. New York: Plenum Press, 1996; 2:249-273. 18. Deuel TF, Senior, RM, Chang D et al. Platelet factor 4 is chemotactic for neutrophils and monocytes. Proc Natl Acad Sci USA 1981; 78:4584-7. 19. Zhang X, Chen L, Bancroft DP et al. Crystal structure of recombinant human platelet factor 4. Biochemistry 1994; 33:8361-6. 20. Gengrinovitch S, Greenberg SM, Cohen T et al. Platelet factor-4 inhibits the mitogenic activity of VEGF121 and VEGF165 using several concurrent mechanisms. J Biol Chem 1995; 270:15059-65. 21. Hansell P, Maione TE, Borgstrom P. Selective binding of platelet factor 4 to regions of active angiogenesis in vivo. Am J Physiol 1995. 22. Derynck R. Transforming growth factor alpha. [Review]. Cell 1988; 54:593-5. 23. Roberts AB. Transforming growth factor-∀: Activity and efficacy in animal models of wound healing. Wound Repair and Regeneration 1995; 3:408-418. 24. Davidson JM, Broadley KN. Manipulation of the wound-healing process with basic fibroblast growth factor. [Review]. Ann N Y Acad Sci 1991; 638:306-15. 25. Slavin J. Fibroblast growth factors: At the heart of angiogenesis. [Review]. Cell Biol Int 1995; 19:431-44. 26. Gospodarowicz D. Fibroblast growth factor. Chemical structure and biologic function. [Review] Clin Orthop 1990. 27. Brown LF, Yeo KT, Berse B et al. Expression of vascular permeability factor (vascular endothelial growth factor) by epidermal keratinocytes during wound healing. J Exp Med 1992; 176:1375-9. 28. Keck PJ, Hauser SD, Krivi G et al. Vascular permeability factor, an endothelial cell mitogen related to PDGF. Science 1989; 246:1309-12. 29. Rubin JS, Bottaro DP, Chedid M et al. Keratinocyte growth factor as a cytokine that mediates mesenchymal-epithelial interaction. [Review]. Exs 1995; 74:191-214. 30. Czaja MJ, Weiner FR, Eghbali M et al. Differential effects of gamma-interferon on collagen and fibronectin gene expression. J Biol Chem 1987; 262:13348-51. 31. Baggiolini M, Moser B, Clark-Lewis I. Interleukin-8 and related chemotactic cytokines. The Giles Filley Lecture. [Review]. Chest 1994; 105:95S-98S. 32. Barbul A, Knud-Hansen J, Wasserkrug HL, Efron G. Interleukin 2 enhances wound healing in rats. J Surg Res 1986; 40:315-9. 33. Graham MF, Willey A, Adams J, Yager D, Diegelmann RF. Interleukin 1 beta downregulates collagen and augments collagenase expression in human intestinal smooth muscle cells. Gastroenterology 1996; 110:344-50. 34. Lacraz S, Nicod LP, Chicheportiche R et al. IL-10 inhibits metalloproteinase and stimulates TIMP-1 production in human mononuclear phagocytes. J Clin Invest 1995; 96:2304-10. 35. Mateo RB, Reichner JS, Albina JE. Interleukin-6 activity in wounds. Am J Physiol 1994; 266:R1840-4. 36. Schirren CG, Scharffetter K, Hein R et al. Tumor necrosis factor alpha induces invasiveness of human skin fibroblasts in vitro. J Invest Dermatol 1990; 94:706-10. 37. Scharffetter K, Heckmann M, Hatamochi A et al. Synergistic effect of tumor necrosis factor-alpha and interferon-gamma on collagen synthesis of human skin fibroblasts in vitro. Exp Cell Res 1989; 181:409-19. 38. Regan MC, Kirk SJ, Hurson M et al. Tumor necrosis factor-alpha inhibits in vivo collagen synthesis. Surgery 1993; 113:173-7.

46

Cytokines and the Abdominal Surgeon

39. Pierce GF, Mustoe TA, Lingelbach J et al. Platelet-derived growth factor and transforming growth factor-beta enhance tissue repair activities by unique mechanisms. J Cell Biol 1989; 109:429-40. 40. Hogstrom H, Haglund U. Postoperative decrease in suture holding capacity in laparotomy wounds and anastomoses. Acta Chir Scand 1985; 151:533-5. 41. Antoniades H N, Galanopoulos T, Neville-Golden J et al. Injury induces in vivo expression of platelet-derived growth factor (PDGF) and PDGF receptor mRNAs in skin epithelial cells and PDGF mRNA in connective tissue fibroblasts. Proc Natl Acad Sci USA 1991; 88:565-9. 42. Senior RM, Griffin GL, Huang JS et al. Chemotactic activity of platelet alpha granule proteins for fibroblasts. J Cell Biol 1983; 96:382-5. 43. Yoshimura T, Matsushima K, Tanaka S et al. Purification of a human monocytederived neutrophil chemotactic factor that has peptide sequence similarity to other host defense cytokines. Proc Natl Acad Sci USA 1987; 84:9233-7. 44. Larsen CG, Anderson AO, Appella E et al. The neutrophil-activating protein (NAP-1) is also chemotactic for T lymphocytes. Science 1989; 243:1464-6. 45. Gimbrone MJ, Obin MS, Brock AF et al. Endothelial interleukin-8: A novel inhibitor of leukocyte-endothelial interactions. Science 1989; 246:1601-3. 46. Wolpe SD, Cerami A. Macrophage inflammatory proteins 1 and 2: Members of a novel superfamily of cytokines. [Review] Faseb J 1989; 3:2565-73. 47. DiPietro LA, Polverini PJ, Rahbe SM, Kovacs EJ. Modulation of JE/MCP-1 expression in dermal wound repair. Am J Pathol 1995; 146:868-75. 48. Castor CW, Miller JW, Walz DA. Structural and biological characteristics of connective tissue activating peptide (CTAP-III) a major human platelet-derived growth factor. Proc Natl Acad Sci USA 1983; 80:765-9. 49. Walz A, Zwahlen R, Baggiolini M. Formation and biological properties of neutrophil activating peptide 2 (NAP-2). [Review] Adv Exp Med Biol 1991; 305:39-46. 50. Simpson DM, Ross R. The neutrophilic leukocyte in wound repair a study with antineutrophil serum. J Clin Invest 1972; 51:2009-23. 51. Stewart RJ, Duley JA, Dewdney J et al. The wound fibroblast and macrophage. II: Their origin studied in a human after bone marrow transplantation. Br J Surg 1981; 68:129-31. 52. van Furth R, Nibbering PH, van Dissel JT, Diesselhoff-den Dulk MM. The characterization, origin, and kinetics of skin macrophages during inflammation. J Invest Dermatol 1985; 85:398-402. 53. Leibovich SJ, Ross R. A macrophage-dependent factor that stimulates the proliferation of fibroblasts in vitro. Am J Pathol 1976; 84:501-14. 54. Leibovich SJ, Ross R. The role of the macrophage in wound repair. A study with hydrocortisone and antimacrophage serum. Am J Pathol 1975; 78:71-100. 55. Riches DWH. Macrophage involvement in wound repair, remodeling and fibrosis. In: RAF Clark, ed. The Molecular and Cellular Biology of Wound Repair, 2nd ed. New York: Plenum Press 1996:95-143. 56. Kunkel SL, Standiford T, Kasahara K, Strieter RM. Stimulus specific induction of monocyte chemotactic protein-1 (MCP-1) gene expression. [Review]. Adv Exp Med Biol 1991; 305:65-71. 57. Baggiolini M, Dewald B, Moser B. Interleukin-8 and related chemotactic cytokines— CXC and CC chemokines. [Review]. Adv Immunol 1994; 55:97-179. 58. Strieter RM, Kunkel SL, Elner VM et al. Interleukin-8. A corneal factor that induces neovascularization. Am J Pathol 1992; 141:1279-84. 59. Barbul A, Breslin RJ, Woodyard JP et al. The effect of in vivo T helper and T suppressor lymphocyte depletion on wound healing. Ann Surg 1989; 209:479-83.

Cytokines and the Abdominal Operative Wound

47

60. Barbul A, Shawe T, Rotter SM et al. Wound healing in nude mice: A study on the regulatory role of lymphocytes in fibroplasia. Surgery 1989; 105:764-9. 61. Peterson JM, Barbul A, Breslin RJ et al. Significance of T-lymphocytes in wound healing. Surgery 1987; 102:300-5. 62. Barbul A, Regan MC. Immune involvement in wound healing. [Review]. Otolaryngol Clin North Am 1995; 28:955-68. 63. Miyazono K, Ichijo H, Heldin CH. Transforming growth factor-beta: latent forms, binding proteins and receptors. [Review]. Growth Factors 1993; 8:11-22. 64. Shah M, Foreman DM, Ferguson MW. Control of scarring in adult wounds by neutralising antibody to transforming growth factor beta. Lancet 1992; 339:213-4. 65. Longaker MT, Bouhana KS, Harrison MR et al. Wound healing in the fetus: Possible role for inflammatory macrophages and transforming growth factor-beta isoforms. Wound Repair and Regulation 1994; 2:104-112. 66. Wahl SM, Hunt DA, Wakefield LM et al. Transforming growth factor type beta induces monocyte chemotaxis and growth factor production. Proc Natl Acad Sci USA 1987; 84:5788-92. 67. Roberts AB, Sporn MB. The transforming growth factor-∀. In: Sporn Roberts AB, ed. Handbook of experimental pharmacology. Peptide growth factors and their receptors. New York: Springer-Verlag, 1990:419-72. 68. Brandes ME, Mai UE, Ohura K, Wahl SM. Type I transforming growth factor-beta receptors on neutrophils mediate chemotaxis to transforming growth factor-beta. J Immunol 1991; 147:1600-6. 69. Thornton SC, Por SB, Walsh BJ et al. Interaction of immune and connective tissue cells: I. The effect of lymphokines and monokines on fibroblast growth. J Leukoc Biol 1990; 47:312-20. 70. Ignotz RA, Massague J. Transforming growth factor-beta stimulates the expression of fibronectin and collagen and their incorporation into the extracellular matrix. J Biol Chem 1986; 261:4337-45. 71. Salomon GD, Kasid A, Bernstein E et al. Gene expression in normal and doxorubicin-impaired wounds: importance of transforming growth factor-beta. Surgery 1990; 108:318-22. 72. Lawrence WT, Norton JA, Sporn MB et al. The reversal of an Adriamycin induced healing impairment with chemoattractants and growth factors. Ann Surg 1986; 203:142-7. 73. Mustoe TA, Pierce GF, Thomason A et al. Accelerated healing of incisional wounds in rats induced by transforming growth factor-beta. Science 1987; 237:1333-6. 74. Roberts AB, Sporn MB, Assoian RK et al. Transforming growth factor type beta: rapid induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in vitro. Proc Natl Acad Sci USA 1986; 83:4167-71. 75. Kahari VM, Heino J, Vuorio E. Interleukin-1 increases collagen production and mRNA levels in cultured skin fibroblasts. Biochim Biophys Acta 1987; 929:142-7. 76. Duncan MR, Berman B. Differential regulation of collagen, glycosaminoglycan, fibronectin, and collagenase activity production in cultured human adult dermal fibroblasts by interleukin 1-alpha and beta and tumor necrosis factor-alpha and beta. J Invest Dermatol 1989; 92:699-706. 77. Krane SM, Dayer JM, Simon LS, Byrne MS. Mononuclear cell-conditioned medium containing mononuclear cell factor (MCF), homologous with interleukin 1, stimulates collagen and fibronectin synthesis by adherent rheumatoid synovial cells: Effects of prostaglandin E2 and indomethacin. Coll Relat Res 1985; 5:99-117. 78. Hamerman D, Wood DD. Interleukin 1 enhances synovial cell hyaluronate synthesis. Proc Soc Exp Biol Med 1984; 177:205-10.

48

Cytokines and the Abdominal Surgeon

79. Mauviel A, Chen YQ, Kahari VM et al. Human recombinant interleukin-1 beta upregulates elastin gene expression in dermal fibroblasts. Evidence for transcriptional regulation in vitro and in vivo. J Biol Chem 1993; 268:6520-4. 80. Bechtel MJ, Reinartz J, Rox JM et al. Upregulation of cell-surface-associated plasminogen activation in cultured keratinocytes by interleukin-1 beta and tumor necrosis factor-alpha. Exp Cell Res 1996; 223:395-404. 81. Postlethwaite AE, Lachman LB, Mainardi CL, Kang AH. Interleukin 1 stimulation of collagenase production by cultured fibroblasts. J Exp Med 1983; 157:801-6. 82. Tadros T, Wobbes T, Hendriks T. Opposite effects of interleukin-2 on normal and transfusion-suppressed healing of experimental intestinal anastomoses. Ann Surg 1993; 218:800-8. 83. Tadros T, Wobbes T, Hendriks T. Blood transfusion impairs the healing of experimental intestinal anastomoses. Ann Surg 1992; 215:276-81. 84. Sautner T, Gotzinger P, Redl-Wenzl E et al. Does reoperation for abdominal sepsis enhance the inflammatory host response? Arch Surg 1997; 132:250-255. 85. Corraliza IM, Soler G, Eichmann K et al. Arginase induction by suppressors of nitric oxide synthesis (IL-4, IL-10 and PGE2) in murine bone-marrow-derived macrophages. Biochem Biophys Res Commun 1995; 206:667-73. 86. Shearer JD, Richards JR, Mills CD, Caldwell MD. Differential regulation of macrophage arginine metabolism: A proposed role in wound healing. Amer J Physiol 1997; 272 (Endocrinol Metab 35):E181-E190. 87. Voest EE, Kenyon BM, O’Reilly MS et al. Inhibition of angiogenesis in vivo by interleukin 12 [see comments]. J Natl Cancer Inst 1995; 87:581-6. 88. Arbiser JL. Angiogenesis and the skin: A primer. [Review]. J Am Acad Dermatol 34:486-97. 89. Stephenson ML, Krane SM, Amento EP et al. Immune interferon inhibits collagen synthesis by rheumatoid synovial cells associated with decreased levels of the procollagen mRNAs. Febs Lett 1985; 180:43-50. 90. Rosenbloom J, Feldman G, Freundlich B, Jimenez SA. Inhibition of excessive scleroderma fibroblast collagen production by recombinant gamma-interferon. Association with a coordinate decrease in types I and III procollagen messenger RNA levels. Arthritis Rheum 1986; 29:851-6. 91. Granstein RD, Murphy GF, Margolis RJ et al. Gamma-interferon inhibits collagen synthesis in vivo in the mouse. J Clin Invest 1987; 79:1254-8. 92. Shapiro SD, Campbell EJ, Kobayashi DK, Welgus HG. Immune modulation of metalloproteinase production in human macrophages. Selective pretranslational suppression of interstitial collagenase and stromelysin biosynthesis by interferongamma. J Clin Invest 1990; 86:1204-10. 93. Lacraz S, Nicod L, Galve de RB et al. Suppression of metalloproteinase biosynthesis in human alveolar macrophages by interleukin-4. J Clin Invest 1992; 90:382-8. 94. Le J, Prensky W, Yip YK et al. Activation of human monocyte cytotoxicity by natural and recombinant immune interferon. J Immunol 1983; 131:2821-6. 95. Witsell AL, Schook LB. Tumor necrosis factor alpha is an autocrine growth regulator during macrophage differentiation [published erratum appears in Proc Natl Acad Sci USA 1993 May 15;90(10):4763]. Proc Natl Acad Sci USA 1992; 89:4754-8. 96. Solis HJ, Brenner DA, Chojkier M. Tumor necrosis factor alpha inhibits collagen gene transcription and collagen synthesis in cultured human fibroblasts. J Biol Chem 1988; 263:5841-5. 97. Dayer JM, Beutler B, Cerami A. Cachectin/tumor necrosis factor stimulates collagenase and prostaglandin E2 production by human synovial cells and dermal fibroblasts. J Exp Med 1985; 162:2163-8.

Cytokines and the Abdominal Operative Wound

49

98. Callaghan MM, Lovis RM, Rammohan C, Lu Y, Pope RM. Autocrine regulation of collagenase gene expression by TNF-alpha in U937 cells. J Leukoc Biol 1996; 59:125-32. 99. Westermarck J, Hakkinen L, Fiers W, Kahari VM. TNF-R55-specific form of human tumor necrosis factor-alpha induces collagenase gene expression by human skin fibroblasts. J Invest Dermatol 1995; 105:197-202. 100. Postlethwaite AE, Seyer JM. Stimulation of fibroblast chemotaxis by human recombinant tumor necrosis factor alpha (TNF-alpha) and a synthetic TNF-alpha 31-68 peptide. J Exp Med 1990; 172:1749-56. 101. Salomon GD, Kasid A, Cromack DT et al. The local effects of cachectin/tumor necrosis factor on wound healing. Ann Surg 1991; 214:175-80. 102. Davies MG, Fulton GJ, Hagen PO. Clinical biology of nitric oxide. [Review]. Br J Surg 1995; 82:1598-610. 103. Melillo G, Taylor LS, Brooks A et al. Regulation of Inducible Nitric Oxide Synthase Expression in IFN-∃-Treated Murine Macrophages Cultured Under Hypoxic Conditions. Journal of Immunology 1996; 157:2638-2644. 104. Albina JE, Henry W Jr, Mastrofrancesco B, Martin BA, Reichner JS. Macrophage activation by culture in an anoxic environment. J Immunol 1995; 155:4391-6. 105. Schaffer MR, Tantry U, Gross SS et al. Nitric oxide regulates wound healing. J Surg Res 1996; 63:237-40. 106. Witte MB, Schaeffer MR, Barbul A. Phenotypic induction of Nitric Oxide is critical for synthetic function in wound fibroblasts. Surgical Forum.1996; 703-704. 107. Thornton FJ, Ahrendt G, Schäffer MR et al. Sepsis impairs anastomotic collagen gene expression and synthesis—A possible role for nitric oxide. J Surg Res (In Press) 1997. 108. Schäffer M, Efron PA, Thornton FJ et al. Nitric oxide: An autocrine regulator of wound fibroblast synthetic function. J Immunol 1997; 158:2375-2381. 109. Arany I, Brysk MM, Brysk H, Tyring SK. Regulation of inducible nitric oxide synthase mRNA levels by differentiation and cytokines in human keratinocytes. Biochem Biophys Res Commun 1996; 220:618-22. 110. RayChaudhury A, Frischer H, Malik AB. Inhibition of endothelial cell proliferation and bFGF-induced phenotypic modulation by nitric oxide. J Cell Biochem 1996; 63:125-134. 111. Vodovotz Y, Geiser AG, Chesler L et al. Spontaneously increased production of nitric oxide and aberrant expression of the inducible nitric oxide synthase in vivo in the transforming growth factor beta 1 null mouse. J Exp Med 1996; 183:2337-42. 112. Akiyama SK, Yamada SS, Chen WT, Yamada KM. Analysis of fibronectin receptor function with monoclonal antibodies: roles in cell adhesion, migration, matrix assembly, and cytoskeletal organization. J Cell Biol 1989; 109:863-75. 113. Sepp NT, Li LJ, Lee KH, Brown EJ et al. Basic fibroblast growth factor increases expression of the alpha v beta 3 integrin complex on human microvascular endothelial cells. J Invest Dermatol 1994; 103:295-9. 114. Clark RA. Regulation of fibroplasia in cutaneous wound repair. [Review]. Am J Med Sci 1993; 306:42-8. 115. Kulozik M, Heckmann M, Mauch C et al. Cytokine regulation of collagen metabolism during wound healing in vitro and in vivo. In: H Janssen, Rooman R, Robertson JIS, ed. Wound Healing. 33-39. Petersfield: Wrightson Medical Publishing Ltd, 1991. 116. Martens MF, Huyben CM, Hendriks T. Collagen synthesis in fibroblasts from human colon: regulatory aspects and differences with skin fibroblasts. Gut 1992; 33:1664-70.

50

Cytokines and the Abdominal Surgeon

117. Postlethwaite AE, Raghow R, Stricklin GP et al. Modulation of fibroblast functions by interleukin 1: Increased steady-state accumulation of type I procollagen messenger RNAs and stimulation of other functions but not chemotaxis by human recombinant interleukin 1 alpha and beta. J Cell Biol 1988; 106:311-8. 118. Beck LS, Chen TL, Mikalauski P, Ammann AJ. Recombinant human transforming growth factor-beta 1 (rhTGF-beta 1) enhances healing and strength of granulation skin wounds. Growth Factors 1990; 3:267-75. 119. Mertz PM, Sauder DL, Davis SC et al. IL-1 as a potent inducer of wound re-epithelization. Prog Clin Biol Res 1991; 365:473-80. 120. Sauder DN, Kilian PL, McLane JA et al. Interleukin-1 enhances epidermal wound healing. Lymphokine Res 1990; 9:465-73. 121. Edwards DR, Murphy G, Reynolds JJ et al. Transforming growth factor beta modulates the expression of collagenase and metalloproteinase inhibitor. Embo J 1987; 6:1899-904. 122. Arnold F, West DC. Angiogenesis in wound healing. [Review]. Pharmacol Ther 1991; 52:407-22. 123. Mignatti P, Mazzieri R, Rifkin DB. Expression of the urokinase receptor in vascular endothelial cells is stimulated by basic fibroblast growth factor. J Cell Biol 1991; 113:1193-201. 124. Mignatti P, Tsuboi R, Robbins E et al. In vitro angiogenesis on the human amniotic membrane: requirement for basic fibroblast growth factor-induced proteinases. J Cell Biol 1989; 108:671-82. 125. Ingber DE. Fibronectin controls capillary endothelial cell growth by modulating cell shape. Proc Natl Acad Sci USA 1990; 87:3579-83. 126. Klagsbrun M, D’Amore PA. Regulators of angiogenesis. [Review]. Annu Rev Physiol 1991; 53:217-39. 127. Bernfield M, Kokenyesi R, Kato M et al. Biology of the syndecans: a family of transmembrane heparan sulfate proteoglycans. [Review]. Annu Rev Cell Biol 1992; 8:365-93. 128. Tekotte H, Engel M, Margolis RU, Margolis RK. Disaccharide composition of heparan sulfates: brain, nervous tissue storage organelles, kidney, and lung. J Neurochem 1994; 62:1126-30. 129. Mignatti P, Rifkin DB. Release of basic fibroblast growth factor, an angiogenic factor devoid of secretory signal sequence: A trivial phenomenon or a novel secretion mechanism? [Review] J Cell Biochem 1991; 47:201-7. 130. D’Amore PA. Modes of FGF release in vivo and in vitro. [Review]. Cancer Metastasis Rev 1990; 9:227-38. 131. O’Keefe EJ, Chiu ML, Payne RJ. Stimulation of growth of keratinocytes by basic fibroblast growth factor. J Invest Dermatol 1990; 90:767-9. 132. Hebda PA, Klingbeil CK, Abraham JA, Fiddes JC. Basic fibroblast growth factor stimulation of epidermal wound healing in pigs. J Invest Dermatol 1990; 95:626-31. 133. Greenhalgh DG, Sprugel KH, Murray MJ, Ross R. PDGF and FGF stimulate wound healing in the genetically diabetic mouse. Am J Pathol 1990; 136:1235-46. 134. Mahadevan V, Hart IR, Lewis GP. Factors influencing blood supply in wound granuloma quantitated by a new in vivo technique. Cancer Res 1989; 49:415-9. 135. Schreiber AB, Winkler ME, Derynck R. Transforming growth factor-alpha: A more potent angiogenic mediator than epidermal growth factor. Science 1986; 232:1250-3. 136. Bussolino F, Mantovani A. Effect of granulocyte-macrophage colony-stimulating factor on endothelial cells. Blood 1991; 78:2475-6. 137. Maione TE, Gray GS, Petro J et al. Inhibition of angiogenesis by recombinant human platelet factor-4 and related peptides. Science 1990; 247:77-9.

Cytokines and the Abdominal Operative Wound

51

138. Sappino AP, Schurch W, Gabbiani G. Differentiation repertoire of fibroblastic cells: expression of cytoskeletal proteins as marker of phenotypic modulations. [Review]. Lab Invest 1990; 63:144-61. 139. Finesmith TH, Broadley KN, Davidson JM. Fibroblasts from wounds of different stages of repair vary in their ability to contract a collagen gel in response to growth factors. J Cell Physiol 1990; 144:99-107. 140. Schmitt-Graff A, Desmouliere A, Gabbiani G. Heterogeneity of myofibroblast phenotypic features: an example of fibroblastic cell plasticity. [Review]. Virchows Arch 1994; 425:3-24. 141. Desmouliere A, Geinoz A, Gabbiani F, Gabbiani G. Transforming growth factorbeta 1 induces alpha-smooth muscle actin expression in granulation tissue myofibroblasts and in quiescent and growing cultured fibroblasts. J Cell Biol 1993; 122:103-11. 142. Montesano R, Orci L. Transforming growth factor beta stimulates collagen-matrix contraction by fibroblasts: implications for wound healing. Proc Natl Acad Sci U S A 1988; 85:4894-7. 143. Vyalov S, Desmouliere A, Gabbiani G. GM-CSF-induced granulation tissue formation: relationships between macrophage and myofibroblast accumulation. Virchows Arch B Cell Pathol Incl Mol Pathol 63:231-9. 144. Rubbia-Brandt L, Sappino AP, Gabbiani G. Locally applied GM-CSF induces the accumulation of alpha-smooth muscle actin containing myofibroblasts. Virchows Arch B Cell Pathol Incl Mol Pathol 1991; 60:73-82. 145. Desmouliere A, Gabbiani G. Modulation of fibroblastic cytoskeletal features during pathological situations: the role of extracellular matrix and cytokines. [Review]. Cell Motil Cytoskeleton 1994; 29:195-203. 146. Streuli CH, Schmidhauser C, Kobrin M et al. Extracellular matrix regulates expression of the TGF-beta 1 gene. J Cell Biol 1993; 120:253-60. 147. Desmouliere A, Rubbia-Brandt L, Abdiu A et al. Alpha-smooth muscle actin is expressed in a subpopulation of cultured and cloned fibroblasts and is modulated by gamma-interferon. Exp Cell Res 1992; 201:64-73. 148. Duncan MR, Berman B. Gamma interferon is the lymphokine and beta interferon the monokine responsible for inhibition of fibroblast collagen production and late but not early fibroblast proliferation. J Exp Med 1985; 162:516-27. 149. Granstein RD, Deak MR, Jacques SL et al. The systemic administration of gamma interferon inhibits collagen synthesis and acute inflammation in a murine skin wounding model. J Invest Dermatol 1989; 93:18-27. 150. Granstein RD, Flotte TJ, Amento EP. Interferons and collagen production. [Review]. J Invest Dermatol 95:75S-80S. 151. Woodley DT, Chen JD, Kim JP et al. Re-epithelialization. Human keratinocyte locomotion. [Review]. Dermatol Clin 1993; 11:641-6. 152. Winter GD. Formation of the scab and the rate of epithelisation of superficial wounds in the skin of the young domestic pig. 1962. J Wound Care 1995; 4:366-7. 153. Stenn K Sa. Re-epithelialization. In: Clark PM, ed. The Molecular and Cellular Biology of Wound Repair. New York: Plenum Press 1988; 321-335. 154. Odland G, Ross R. Human wound repair. I. Epidermal regeneration. J Cell Biol 1968; 39:135-51. 155. Kim JP, Zhang K, Chen JD et al. Mechanism of human keratinocyte migration on fibronectin: unique roles of RGD site and integrins. J Cell Physiol 1992; 151:443-50. 156. Bereiter-Hahn J, Strohmeier R, Kunzenbacher I et al. Locomotion of Xenopus epidermis cells in primary culture. J Cell Sci 1981; 52:289-311. 157. Cohen S. The epidermal growth factor (EGF). [Review]. Cancer 1987; 51:1787-91.

52

Cytokines and the Abdominal Surgeon

158. Derynck R. Transforming growth factor-alpha: Structure and biological activities. [Review]. J Cell Biochem 1986; 32:293-304. 159. Chen JD, Kim JP, Zhang K et al. Epidermal growth factor (EGF) promotes human keratinocyte locomotion on collagen by increasing the alpha 2 integrin subunit. Exp Cell Res 1993; 209:216-23. 160. Chen JD, Lapiere JC, Sauder DN et al. Interleukin-1 alpha stimulates keratinocyte migration through an epidermal growth factor/transforming growth factor-alphaindependent pathway. J Invest Dermatol 1995; 104:729-33. 161. Dudgeon LS, Goadby HK. The examination of the tissues and some observations on the blood platelets of rabbits at intervals of five minutes, and later, after intravenous inoculation of Staphylococcus aureus and Indian ink. J Hyg 1931; 31:247. 162. Weil MH, Spink WW. A comparison of shock due to endotoxin with anaphylactic shock. J Lab Clin Med 1957; 50:502. 163. Hunt TK, Niinikoski J, Zederfeldt B. Role of oxygen in repair processes. Acta Chir Scand 1972; 138:109-10. 164. Mauviel A. Cytokine regulation of metalloproteinase gene expression. [Review] J Cell Biochem 1993; 53:288-95. 165. Postlethwaite AE, Kang AH. Advantages and limitations of in vitro models of wound healing and tissue repair. Prog Clin Biol Res 1988; 266:237-42. 166. Pare A. The classic. Compound fracture of leg, Pare’s personal care (MII, 328). Clin Orthop 1983; 178:3.

Cytokines and Postoperative Abdominal Adhesions

53

CHAPTER 5

Cytokines and Postoperative Abdominal Adhesions Shaun G. Appleton and Jeremy N.Thompson

Introduction

A

dhesions are the fibrinous deposits that form between mesothelial surfaces following injury or inflammation. Following abdominal surgery, over 90% of people will develop peritoneal adhesions.1 As the number of laparotomies performed has increased so has the incidence of adhesions and their complications, notably bowel obstruction, infertility and pelvic pain. Acquired adhesions result from peritoneal injury which may be caused by infection, ischemia, a reaction to foreign materials such as sutures, gauze particles and glove powder, or from the coaption of two denuded peritoneal surfaces.2,3 In developed countries adhesions are the leading cause of bowel obstruction accounting for more than 40% of all cases of intestinal obstruction4 and 60-70% of those involving the small bowel.5 In developing countries the commonest cause of small bowel obstruction is still strangulated hernia because elective abdominal surgery and early hernia repair are relatively rare. About 1% of patients will develop intestinal obstruction due to postoperative adhesions within one year of surgery and 0.5% within four weeks of their operation.1 A survey of 1200 hospitals in Germany estimated that 2.6% laparotomies were performed for adhesional bowel obstruction,6 while a review of British surgeons revealed that on average, adhesive obstruction accounted for three to four laparotomies and another seven to eight admissions for conservative management per general surgeon per year.7 In women the most common causes of adhesions are prior surgery, infection and endometriosis.8 The relationship between pelvic adhesions and infertility is well established9-11 and adhesiolysis can increase pregnancy rates among previously infertile women. The relationship between adhesions and pelvic pain is less clear-cut though nearly 80% of patients report a disappearance or substantial reduction in pelvic pain after adhesiolysis.12 As well as the surgical workload, adhesions impose a huge financial burden; in 1988 in the United States there were over 282,000 hospitalizations for adhesiolysis costing an estimated 1.18 billion US dollars.13

Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

54

Cytokines and the Abdominal Surgeon

Adhesion Formation In order to appreciate the complexities of adhesion formation, an understanding of the anatomy and pathophysiology of the peritoneum must be considered. The peritoneum is a unique organ covering an area of 10,000 cm2 in an adult, almost as much as the skin.14 It consists of a single layer of mesothelial cells resting on a highly vascularised, but loosely packed, connective tissue framework. A small volume of fluid within the peritoneal cavity provides lubrication between the parietal and visceral surfaces. This peritoneal fluid contains a variable number of resident peritoneal macrophages that exhibit minimal biological activity at rest. However, when stimulated by peritoneal injury or infection they release large quantities of interleukin-1 (IL-1! and IL-1∀), interleukin-6 (IL-6), interleukin-8 (IL-8) and tumour necrosis factor-alpha (TNF-!).15 These proinflammatory and chemotactic mediators effect the production of an inflammatory exudate and a change in the peritoneal white cell population. Large numbers of polymorphonuclear neutrophils (PMNs) appear and, in the absence of infection disappear within three to four days. Macrophages follow and reach a peak about the fifth day. It is these postsurgical macrophages that play a vital role in the modulation of inflammation by secreting a range of substances, including cyclooxygenase and lipoxygenase metabolites, plasminogen activators, plasminogen activator inhibitors, collagenase, elastase, IL-1 and IL-6, TNF and other cytokines.16 Peritoneal macrophages also recruit new mesothelial cells onto the surface of the injured peritoneum. These cells are thought to be derived from the primitive mesenchymal cells supporting the mesothelial layer and differentiate in response to cytokines and other macrophage-secreted mediators.5 Islands of new mesothelial cells appear throughout the injured area and rapidly proliferate accomplishing re-epithelialization within eight days.17 It is a unique feature of the mesothelium that, unlike dermal defects, both large and small peritoneal defects heal at the same speed.18 Mesothelial cells themselves are an important source of inflammatory and fibrogenic mediators secreting IL-1, IL-6 and IL-8 when appropriately stimulated.19-21 This stimulus is mediated by IL-1∀ and TNF-! released from resident and infiltrated macrophages.20 Several groups have investigated the cytokine levels in peritoneal fluid and plasma in response to surgery.40-47 They have shown an initial rise in TNF-! within 3 hours of surgery followed by rises in IL-1∀ and IL-6 within 6 hours of surgery. IL-6 reached the highest and most sustained concentrations locally and could also be detected in the plasma23 (Fig. 5.1). The inflammatory peritoneal exudate formed in response to injury is rich in fibrin and lysis of fibrin deposits is a physiological property of normal mesothelial surfaces.24 Mesothelial cells possess plasminogen-activating activity;25 tissue plasminogen activator (tPA) is the major plasminogen activator in human peritoneum.26 This property is important in preventing fibrinous adhesions developing between tissue surfaces within body cavities. Peritoneal injury results in rapid synthesis and release of plasminogen activator inhibitors-1 and -2 by mesothelial, endothelial and inflammatory cells, with loss of plasminogen-activating activity.27 Studies of peritoneal fluid after elective operation have demonstrated that this loss occurs over six to twelve hours and is associated with marked increases in the concentrations of plasminogen activator inhibitors-1 and -2. The production of plasminogen activator inhibitors appears to be mediated directly by the inflammatory cytokines in the peritoneal fluid following injury. TNF-!, IL-1 and IL-6 have all been shown both individually and synergistically to stimulate plasminogen activator-1 production by human mesothelial cells in culture.28

Cytokines and Postoperative Abdominal Adhesions

55 Fig. 5.1. Concentrations of cytokines in the peritoneal fluid following surgery (mean ± SEM).

The duration of the reduction in plasminogen-activating activity on the peritoneal surface probably determines the fate of the initial fibrinous adhesions. Prolonged depression of fibrinolytic activity allows the organisation and permanent formation of fibrous adhesions. Recovery of plasminogen-activating activity within three to four days leads to lysis of fibrin and adhesion-free tissue healing (Fig. 5.2). Much of the experimental work on adhesions and cytokines has investigated the role of the early inflammatory mediators; IL-1, IL-6 and TNF-! and the fibrogenic cytokine TGF-∀.

Interleukin-1 (IL-1) IL-1 consists of two subtypes ! and ∀ with a similar spectrum of biological activity. IL-1 is produced by all forms of antigen-presenting cells and is an important and ubiquitous early mediator of inflammation. In vivo effects of IL-1 include expression of adhesion molecules on vascular endothelium, regulation of fibrin deposition

56

Cytokines and the Abdominal Surgeon Release of cytokines, arachidonic acid metabolites, oxygen free radicals and growth factors

TNF-! IL-1 IL-6

TNF-∀

Plasminogen Activator Inhibitors

INJURY

INFLAMMATION

FIBRIN DEPOSITION

Plasminogen activators

FIBRIN REMNANTS

LYSIS

INGROWTH OF FIBROBLASTS AND CAPILLARIES

RESOLUTION

PERMANENT ADHESION

Fig. 5.2. Pathogenesis of adhesions.

and lysis, collagen deposition by fibroblasts, stimulation of fibroblast proliferation, increase in collagenase production by fibroblasts and modulation of granulocyte production.29 IL-1 and TNF-! overlap in many of their biological activities. In a clinical setting levels of IL-1 in peritoneal fluid rise within six hours of operation and reach a peak at 12 hours.40 Levels of IL-1 (! and ∀) in the plasma are low or undetectable following surgical operations,40,44,47 but one study has shown a significant and persistent rise in levels of IL-1 receptor antagonist in the plasma following minor surgical trauma.47 In a rat adhesion model, recombinant IL-1! given intraperitoneally at operation and for five days afterwards resulted in increased adhesion formation compared to controls.30 However, this effect was not seen in rats treated with IL-1! at two weeks after surgery30 highlighting the importance of the cytokine in the initial inflammatory response. Intravenous administration of antibodies to IL-1 immediately after surgery reduced adhesion formation in a rat model and this effect was potentiated in rats given both anti-IL-1 and anti-TNF-! antibodies.31 Histologically these adhesions also demonstrated lower levels of collagen deposition and fibroblast invasion.

Tumor Necrosis Factor-alpha (TNF-!) Much interest has been focused on the effects of the pro-inflammatory cytokine TNF-! on wound healing and fibroblast function.32 The effects of TNF-! are similar to IL-1 in stimulating the production of prostaglandins by different types of fibroblasts and TNF-! induces the synthesis and release of IL-1 from monocytes, fibroblasts and endothelial cells. TNF-! is the earliest cytokine to be produced in the peritoneal cavity after injury reaching peak concentrations between three and eight hours after surgery and remaining elevated for 24 hours.40 The rise in concentration of TNF-! in the perito-

Cytokines and Postoperative Abdominal Adhesions

57

neal fluid is proportional to both the length of surgery and degree of peritoneal bacterial contamination.42 Increased levels of TNF-! have also been found in the peritoneal fluid of women with endometriosis and infertility.38 Women with endometriosis are said to be more prone to postoperative adhesions,33 presumably due to the localized sterile inflammatory process and the large numbers of activated peritoneal macrophages. One group of investigators has shown TNF-! to be a good biological marker for intra-abdominal adhesion formation.34 They subjected three groups of rats to laparotomy: the first group acted as a control and had only saline irrigation of the peritoneal cavity, the second group underwent caecal abrasion while the third group had a limited small bowel resection and anastomosis. Levels of TNF-! were measured in the blood and peritoneal fluid at 30, 90 and 180 minutes after the operation. At 3 weeks the rats were killed and the adhesions graded. There were more adhesions in the groups which had caecal abrasion and small bowel resection compared to the control but there was also a significant correlation between the grade of adhesions and the concentration of TNF-! in the peritoneal fluid and plasma. Rats given intravenous antibodies to TNF-! preoperatively showed no significant reduction in adhesion formation unless also given antibodies to IL-1 when there was a synergistic reduction in adhesion formation.31 Peritoneal fibrinolytic activity plays a pivotal role in the pathophysiology of adhesions35 and TNF has been shown to mediate the release of plasminogen activator inhibitor-1 by human peritoneal mesothelial cells.36

Interleukin-6 (IL-6) This multifunctional cytokine is secreted by T cells, fibroblasts and macrophages in response to injury. IL-6 is a normal constituent of peritoneal fluid and several investigators have found elevated levels of IL-6 in the peritoneal fluid of women with adhesions, endometriosis and pelvic inflammatory disease.37-39 Following surgical injury there is a marked rise in IL-6 levels in both peritoneal fluid40-43 and plasma.40,44-47 This rise starts within six hours of the beginning of the operation, reaches a peak at 12 hours40 and is proportional to the duration of the operation.48 Peak concentrations of IL-6 are 300-fold greater in the peritoneal fluid than in the blood. This difference may be explained by a number of factors, including incomplete peritoneal absorption, dilution and first-pass hepatic metabolism. However it seems likely that the systemic IL-6 response is a secondary reflection of events within the peritoneal cavity. The rise in IL-6 levels in peritoneal fluid is more sustained than those of IL-1∀ and TNF-! reaching a plateau between 24 and 72 hours.46 The induction of IL-6 synthesis is probably stimulated by TNF-! and IL-1∀ released by resident mononuclear phagocytes following surgical injury.48,49 Intravenous injection of IL-6 prior to adhesion formation in a rat model resulted in a slight but significant increase in the grade of intraperitoneal adhesions, while treatment with anti-IL-6 antibody preoperatively showed a decrease in adhesion grade.50

Transforming Growth Factor-beta (TGF-∀) The central role of TGF-∀ in wound healing and tissue fibrosis is well recognized.51,52 Released initially by platelets and subsequently by macrophages, activated T cells and neutrophils, TGF-∀ is a potent regulator of inflammation and fibrosis. It

58

Cytokines and the Abdominal Surgeon

has a chemotactic action on fibroblasts and inflammatory cells and promotes angiogenesis, cell proliferation, differentiation and deposition of extracellular matrix.53,54 The evidence for the role of TGF-∀ in the formation and regulation of adhesions is based on (1) the role of TGF-∀ in fibrosis in other tissues; (2) the presence and (3) activity of TGF-∀ and/or its receptor in the peritoneal cavity and in adhesions; (4) the effect of exogenous TGF-∀ or its neutralizing antibody on the formation of adhesions in experimental models.52 Excessive or sustained production of TGF-∀1 is a key molecular mediator of tissue fibrosis in many organs. Topical application of TGF-∀ enhances experimental wound healing55 and intravenous administration leads to liver and kidney fibrosis in rats. 56 All three isoforms of TGF- ∀ ( TGF- ∀1 , TGF- ∀ 2, TGF- ∀3) are present immunohistochemically in surgically induced adhesions in a rat model. Fibroblasts, inflammatory cells and the endothelial and smooth-muscle cells of arterioles in the fibrous tissue all stain for TGF-∀.57 The intact peritoneum demonstrates minimal staining for TGF-∀3, but once injured by surgery or inflammation, levels of all three isoforms rise substantially.58 TGF-∀1 is also found in the peritoneal fluid of patients who have developed adhesions following intraperitoneal immunotherapy with interleukin-2 (IL-2) and IL-2 activated killer cells.58 Rats given TGF-∀ following surgery develop more severe adhesions than controls and histologically their adhesions have a greater density of inflammatory cells and fibroblasts.59 However administration of neutralizing antibody to TGF-∀ (all isoforms) is not effective in reducing the level of fibrous adhesions though it does reduce the cellularity of the fibrous tissue.59,60 More specific blockade by giving antibodies to the different TGF-∀ isoforms has shown a significant reduction in adhesion formation in a rat model when using anti-TGF-∀1 but not anti-TGF-∀2.60

Other Cytokines Epidermal growth factor (EGF) is known to stimulate fibroblast proliferation and collagen deposition.61 EGF, transforming growth factor-! (TGF-!) and the EGF receptor have been identified immunohistochemically in fibrous adhesions from both rats and patients.62 The highest immunostaining for EGF was associated with the inflammatory cells infiltrated into the fibrous adhesion. Interleukin-2 (IL-2) is produced by activated T lymphocytes and is known to stimulate the production of fibrogenic mediators from peripheral blood leucocytes.58 In rat peritoneal macrophage culture IL-2 induces the expression of platelet-derived growth factor (PDGF-! and ∀ chains) within two hours of treatment.63 As PDGF is a fibrogenic cytokine this suggests that it may play a role in the well-recognized development of adhesions in patients receiving intraperitoneal immunotherapy with IL-2-activated killer cells and IL-2.

Conclusions The healing of injured mesothelial surfaces is a complex process integrating both the coagulation cascade and the immune system in an effort to arrest bleeding, prevent infection and restore tissue integrity and function. Any imbalance in the control mechanisms of this process will lead to an impairment of wound healing or excess tissue formation—fibrosis; or adhesion formation. Cytokines have been shown to play a key role in adhesion formation following operation or other injury to the peritoneum. Adhesions form soon after surgery and

Cytokines and Postoperative Abdominal Adhesions

59

much attention has been focused on the cytokines that mediate the early inflammatory reaction. By experimentally augmenting or blocking the action of these cytokines the degree of adhesion formation can be altered. Cytokine manipulation to control adhesion formation is an attractive option. However further research is necessary into the complex interactions that occur between mediators during fibrin deposition and fibrinolysis. In addition, the localization of cytokine modification to the areas of adhesion formation, minimizing the systemic effects of therapy and preventing adhesions without obviating the beneficial aspects of the inflammatory process are all potential problems that remain to be addressed.

References 1. Menzies D, Ellis H. Intestinal obstruction from adhesions-how big is the problem? Ann R Coll Surg Eng 1990; 72:60-63. 2. Ellis H. The causes and prevention of intestinal adhesions. Br J Surg 1982; 69:241-243. 3. Ellis H. Prevention and treatment of adhesions. Infect Surg 1983:803-807. 4. Menzies D. Postoperative adhesions: Their treatment and relevance in clinical practice. Ann R Coll Surg Engl 1993; 75:147-153. 5. Menzies D. Peritoneal adhesions: incidence, cause and prevention. Ann Surg 1992; 24(Pt I):29-45. 6. Treutner K-H, Bertram P, Loser S et al. Prevention and treatment of peritoneal adhesions - a questionnaire survey on 1200 hospitals in Germany. Der Chirurg 1995; 66:398-403. 7. Scott-Coombes DM, Vipond M, Thompson JN. General surgeons attitudes to the treatment and prevention of abdominal adhesions. Ann R Coll Surg Eng 1993; 75:123-128. 8. Pijlman BM, Dorr PJ, Brommer EJP et al. Prevention of adhesions: Review. Eur J Obstet Gynecol Reprod Biol 1994; 53:155-163. 9. Trimbos-Kemper TCM, Trimbos JB, van Hall EV. Adhesion formation after tubal surgery: results of the eighth-day laparoscopy in 188 patients. Fertil Steril 1985; 43:395-400. 10. Diamond E. Lysis of postoperative pelvic adhesions in infertility. Fertil Steril 1979; 31:287-295. 11. Frantzen C, Schlösser HW. Microsurgery and postinfectious tubal infertility. Fertil Steril 1982; 38:397420. 12. diZerega GS. Biochemical events in peritoneal tissue repair. Eur J Surg 1997; Suppl 577:10-16. 13. Fox RN, Larsen JW, Stillman RJ et al. Economic impact of hospitalizations for lower abdominal adhesiolysis in the United States in 1988. Surg Gynecol Obstet 1993; 176:271-276. 14. Esperanza MJ, Collins DL. Peritoneal dialysis efficiency in relation to body weight. J Paediatr Surg 1966; 1:162-169. 15. Topley N, Mackenzie R, Jörres A et al. Cytokine networks in CAPD: interactions of resident cells during inflammation in the peritoneal cavity. Perit Dial Int 1993; 13:282-285. 16. Rodgers KE, diZerega GS. Function of peritoneal exudate cells after abdominal surgery. J Invest Surg 1993; 6:9-23. 17. Raftery AT. Regeneration of parietal and visceral peritoneum: An electron microscopical study. J Anat 1973; 115:321-328.

60

Cytokines and the Abdominal Surgeon

18. Ellis H, Harrison W, Hugh TB. The healing of the peritoneum under normal and pathological conditions. Br J Surg 1965; 52:471-476. 19. Topley N, Brown Z, Jörres A et al. Human peritoneal mesothelial cells synthesize IL-8: synergistic induction by interleukin-1∀ and tumour necrosis factor-!. Am J Pathol 1993; 142:1876-1886. 20. Betjes MGH, Tuk CW, Struijk DG et al. Interleukin-8 production by human peritoneal mesothelial cells in response to tumour necrosis factor a, interleukin-1 and medium conditioned by macrophages co-cultured with Staphylococcus epidermidis. J Infect Dis 1993; 168:1202-1210. 21. Douvdevani A, Rapaport A, Konforty A et al. Human peritoneal mesothelial cells synthesize IL-1a and b. Kidney Int 1994; 46:993-1001. 22. Topley N, Williams JD. The role of the peritoneal membrane in the control of inflammation in the peritoneal cavity. Kidney Int 1994; 46:S71-78. 23. Scott-Coombes DM, Whawell SA, Thompson JN. Peritoneal cytokine response in surgery. Br J Surg 1994; 81:756. 24. Porter JM, McGregor FH Jr, Mullen DC et al. Fibrinolytic activity of mesothelial surfaces. Surg Forum 1969; 20:80-82. 25. Raftery AT. Regeneration of peritoneum: A fibrinolytic study. J Anat 1979; 129:659-664. 26. Vipond MN, Whawell SA, Thompson JN et al. Peritoneal fibrinolytic activity and intra-abdominal adhesions. Lancet 1990; 335:1120-1122. 27. Whawell SA, Wang Y, Fleming KA et al. Localisation of plasminogen activator inhibitor-1 production in inflamed appendix by in situ mRNA hybridisation. J Pathol 1993; 169:67-71. 28. Whawell SA, Thompson JN. Cytokine induced release of plasminogen activator inhibitor-1 by human mesothelial cells. Eur J Surg 1995; 161:315-317. 29. Dinarello CA, Mier JW. Lymphokines. N Engl J Med 1987;317:940-5. 30. Hershlag A, Otterness IG, Bliven ML et al. The effect of interleukin-1 on adhesion formation in the rat. Am J Obstet Gynecol 1991; 165:771-774. 31. Kaidi A, Nazzal M, Gurchumelidze T et al. Preoperative administration of antibodies against tumour necrosis factor-Alpha (TNF-!) and Interleukin-1 (IL-1) and their impact on peritoneal adhesion formation. Am Surg 1995; 58:302-306. 32. Rapala K. The effect of tumour necrosis factor-! on wound healing. Ann Chirurg Gynae 1996; 85 (Suppl.211):1-53. 33. Haney AF. Endometriosis, macrophages, and adhesions. Prog Clin Biol Res 1993; 381:19-44. 34. Kaidi AA, Gurchmelidze T, Nazzal M et al. Tumour necrosis factor-alpha: a marker for peritoneal adhesion formation. J Surg Res 1995; 58:516-518. 35. Vipond MN, Whawell SA, Thompson JN et al. Peritoneal fibrinolytic activity and intra-abdominal adhesions. Lancet 1990; 335:1120-1122. 36. Whawell SA, Scott-Coombes DM, Vipond MN et al. Tumour necrosis factor-mediated release of plasminogen activator inhibitor 1 by human peritoneal mesothelial cells. Br J Surg 1994; 81:214-216. 37. Buyalos RP, Watson JM, Funari VA et al. Elevated interleukin-6 levels in peritoneal fluid of patients with pelvic pathology. Fertil Steril 1992; 58:302-306. 38. Overton C, Fernandez-Shaw S, Hicks B et al. Peritoneal fluid cytokines and the relationship with endometriosis and pain. Hum Reprod 1996; 11:380-386. 39. Rier SE, Parsons AK, Becker JL. Altered interleukin-6 production by peritoneal leukocytes from patients with endometriosis. Fertil Steril 1994; 61:294-299. 40. Badia JM, Whawell SA, Scott-Coombes DM et al. Peritoneal and systemic cytokine response to laparotomy. Br J Surg 1996; 83(3):347-348.

Cytokines and Postoperative Abdominal Adhesions

61

41. Tokunaga A, Onda M, Fujita I et al. Sequential changes in the cell mediators of peritoneal and wound fluids after surgery. Jpn J Surg 1993; 23:841-844. 42. Tsukada K, Katoh H, Shiojima M et al. Concentrations of cytokines in peritoneal fluid after abdominal surgery. Eur J Surg 1993; 159:475-479. 43. Fujita I, Onda M, Tokunaga A. Expression of interleukin-6 and tumour necrosis factor in peritoneal mononuclear cells after gastrectomy. Wound Rep Regen 1995; 3:168-175. 44. Baigre RJ, Lamont PM, Kwiatkowski D et al. Systemic cytokine response after major surgery. Br J Surg 1992; 79:757-760. 45. Ohzato H, Yoshizaki K, Nishimoto N et al. Interleukin-6 as a new indicator of inflammatory status: detection of serum levels of interleukin-6 and C-reactive protein after surgery. Surgery 1992; 111:201-209. 46. Sakamoto K, Arakawa H, Mita S et al. Elevation of circulating interleukin-6 after surgery: factors influencing the serum level. Cytokine 1994; 6:181-186. 47. Grzelak I, Olszewski WL, Zaleska M et al. Blood cytokine levels rise even after minor surgical trauma. J Clin Immunol 1996; 16(3):159-164. 48. Van Deuren M, Dofferhoff ASM, van der Meer JWM. Cytokines and the response to infection. J Pathol 1992; 168:349-356. 49. Topley N, Jörres A, Luttmann W et al. Human peritoneal mesothelial cells synthesize IL-6: Induction by IL-1∀ and TNF-!. Kidney Int 1993; 43:226-233. 50. Saba A, Kaidi A, Godziachvili, V et al. Effects of Interleukin-6 and its neutralising antibodies on peritoneal adhesion formation and wound healing. Am Surg 1996; 62:569-572. 51. Border WA, Noble NA. Transforming growth factor ∀ in tissue fibrosis. N Engl J Med 1994; 331(19):1286-1292. 52. Chegini N. The role of growth factors in peritoneal healing: transforming growth factor-beta. Eur J Surg (Suppl.) 1997; (577):17-23. 53. Wahl SM, Hunt DA, Wakefield LM et al. Transforming growth factor type ∀ induces monocyte chemotaxis and growth factor production. Proc Natl Acad Sci USA 1987; 84:5788-5792. 54. Postlethwaite AE, Keski-Oja J, Moses HL et al. Stimulation of the chemotactic migration of human fibroblasts by transforming growth factor ∀. J Exp Med 1987; 165:251-256. 55. Roberts AB, Sporn MB. Physiological actions and clinical applications of transforming growth factor ∀ (TGF-∀). Growth Factors 1993; 8:1-9. 56. Terrell TG, Working PK, Chow CP et al. Pathology of recombinant human transforming growth factor-b1 in rats and rabbits. Int Rev Exp Pathol 1993; 34:43-67. 57. Chegini N, Gold LI, Williams RS et al. Localization of transforming growth factor beta isoforms (TGF-beta1, TGF-beta2, TGF-beta3) in surgically induced pelvic adhesions in the rat. Obstet Gynecol 1994; 83:449-454. 58. Kovacs EJ, Brock B, Silber IE et al. Production of fibrogenic cytokines by interleukin2-treated peripheral blood leukocytes: Expression of transforming growth factorbeta and platelet-derived growth factor B chain genes. Obstet Gynecol 1993; 82:29-36. 59. Williams RS, Rossi AM, Chegini N et al. Effect of transforming growth factor b on postoperative adhesion formation and intact peritoneum. J Surg Research 1991; 52:65-70. 60. Lucas PA, Warejcka DJ, Young HE et al. Formation of abdominal adhesions is inhibited by antibodies to transforming growth factor-beta1. J Surg Res 1996; 65:135-138.

62

Cytokines and the Abdominal Surgeon

61. Laato M, Niinikoski J, Lundberg C et al. Effect of epidermal growth factor on experimental granulation tissue. J Surg Res 1986; 41:252-255. 62. Chegini N, Simms J, Williams RS et al. Identification of epidermal growth factor, transforming growth factor-alpha and epidermal growth factor receptor in surgically induced pelvic adhesions in the rat and intraperitoneal adhesions in the human. Am J Obs Gyn 1994; 171:321-327. 63. Kovacs EJ, Van Stedum S, Neuman JE et al. Selective induction of PDGF gene expression in peritoneal macrophages by interleukin-2. Immunobiology 1994; 190(3):263-274.

Cytokines in Blunt Abdominal Trauma

63

CHAPTER 6

Cytokines in Blunt Abdominal Trauma Matthias W. Wichmann and Eugen Faist

Introduction

B

lunt abdominal trauma may become a life-threatening situation for the trauma victim. A recent study by Brenneman et al1 showed that 73% of nonsurvivors following blunt trauma suffered from abdominal injuries. Furthermore, 57% of the surviving trauma victims presented with abdominal injuries.1 In total 65% (50/76) of all patients with severe blunt trauma (ISS 50) suffered from abdominal injuries.1 A study of 433 patients with multiple injuries by Faist et al2 showed abdominal involvement in 29% of all cases. Severe blunt abdominal trauma is not only associated with soft-tissue injury, but also occurs frequently in conjunction with some form of blood loss (hemorrhage). This can be concluded from the observation that most patients with blunt abdominal trauma present with lacerations of the liver and/or spleen as well as retroperitoneal hematoma.1 In this context it is well recognized that liver and/or spleen are the most frequently injured organs following blunt abdominal trauma.3 Laceration of the liver or the spleen is associated with blood loss into the organ itself or into the abdominal cavity once the organ-capsule is ruptured. Studying the effects of blunt abdominal trauma on the cytokine system should, therefore, focus on the effects of soft-tissue trauma and/or hemorrhage on this system of cellular mediators. With regard to this, the term cytokines includes lymphokines, interferons, colony stimulating factors and tumor necrosis factors.4 A number of experimental animal studies as well as clinical investigations have been carried out focusing on the effects of soft-tissue trauma and/or hemorrhage on the cytokine system. These studies mainly involved the following parameters: Interleukin- (IL-) 1, IL-2, IL-3, IL-4, IL-6, IL-8, IL-10, IL-12, IL-13, Tumor-Necrosis-Factor (TNF), Transforming-Growth-Factor (TGF), and Interferon (IFN). This chapter will provide an account of the results obtained in experimental animals, as well as in clinical studies on trauma patients.

Introduction to the Cytokines Studied in Trauma Research Cytokines are proteins of relatively low molecular mass which regulate all important biological processes, i.e., cell growth, cell activation, inflammation, immunity, Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

64

Cytokines and the Abdominal Surgeon

tissue repair, fibrosis and morphogenesis.5 These proteins are effective at very low concentrations (pg/ml), which is due to their mode of action involving high-affinity receptors on the cell surface.5

Interleukin-1 IL-1 is a cytokine produced by activated mononuclear phagocytes and is a mediator of inflammation. It shares many properties in common with tumor necrosis factor. IL-1 is comprised of two principal polypeptides of 17 kDa which have the same biological activities and bind to the same receptor on cell surfaces. While the one IL-1 acts as a membrane-associated substance the other IL-1 is found in the circulation. The effects of IL-1 differ according to whether it is produced in lower or in higher concentrations. At low concentrations, the effects are mainly immunoregulatory, at higher levels IL-1 produces endocrine effects (fever, release of acutephase proteins).6 Virtually all cells of the body have receptors for IL-1 and can respond to it. In the brain IL-1 induces fever and it causes increased release of corticosteroids. Furthermore, the production of prostaglandins and degradative enzymes is induced in B and T lymphocytes.5

Interleukin-2

IL-2 is a 15.5 kDa glycoprotein which is synthesized by CD4+ T helper lymphocytes; a small amount is also produced by CD8+ T cells. The amount of IL-2 produced by CD4+ T helper cells is a principal marker of the strength of an immune response. Inadequate IL-2 formation can, therefore, lead to antigen-specific T lymphocyte anergy. High levels of IL-2 are required for stimulation of natural killer cells.6 IL-2 is used in experimental cancer therapy, especially for renal cancer; its benefit may be related to the activation of cells that can produce cytotoxic anticancer effects.5

Interleukin-3

IL-3 is a 20 kDa lymphokine which is synthesized by activated CD4+ T helper lymphocytes and acts as a colony-stimulating factor. It promotes proliferation and differentiation of other lymphocytes as well as mast cell proliferation.6 IL-3 stimulates the growth of precursors of all the hemopoietic lineages.5

Interleukin-4

IL-4 is a 20 kDa cytokine which is synthesized mainly by CD4+ T helper cells and by activated mast cells. IL-4 induces switching of B lymphocytes to produce IgE and it may have some role in cell-mediated immunity. In mice IL-4, furthermore, serves as a growth and differentiation factor for B cells.6 In macrophages IL-4 induces MHC class II expression and inhibits cytokine release (TNF, IL-1, IL-6).7

Interleukin-6 IL-6 is a 26 kDa cytokine produced by endothelial cells, mononuclear phagocytes, fibroblasts, activated T lymphocytes, and a number of neoplasms. IL-6 is secreted in response to IL-1, IFN-! or TNF and mainly acts on hepatocytes and B cells. Hepatocytes form acute-phase proteins and B lymphocytes differentiate into antibody forming cells in response to IL-6.6 IL-6 is a central regulatory protein involved in the control of acute phase protein synthesis in hepatocytes.7

Cytokines in Blunt Abdominal Trauma

65

Interleukin-8 IL-8 is an 8 kDa protein produced by macrophages and endothelial cells. It has a strong chemotactic effect on T lymphocytes and neutrophils and it has antiviral, immunomodulatory, as well as antiproliferative properties. Furthermore, IL-8 blocks the neutrophil-mediated injury; on the other hand it facilitates neutrophil adherence to endothelial cells. Moreover, it is involved in inflammation and cell-migration.6

Interleukin-10 IL-10 is an 18 kDa polypeptide which acts as a cytokine synthesis inhibitory factor. It is synthesized by CD4+ and CD8+ T lymphocytes, monocytes, macrophages, activated B cells, and keratinocytes. IL-10 inhibits cytokine synthesis (TNF, IL-1, IL-6) by TH1 cells, blocks antigen presentation, and the formation of interferon. IL-10 also inhibits antigen presentation by macrophages, thereby blocking the formation of IL-1, IL-6 and TNF.6

Interleukin-12 IL-12 is a heterodimeric molecule comprised of 35 and 40 kDa chains. It is a growth factor for activated CD4+ and CD8+ T cells and for natural killer cells. IL-12 may have future importance for the therapeutic use in the treatment of tumors or infections.6

Interleukin-13 IL-13 has a molecular weight of 10 kDa and is synthesized by activated T cells and inhibits inflammatory cytokine production in human peripheral blood lymphocytes.8 It can be considered a modulator of B cell responses and, furthermore, suppresses the cytotoxic functions of monocytes/macrophages and the generation of proinflammatory cytokines.6

Tumor Necrosis Factor-∀ TNF-∀ is a 50 kDa cytotoxic monokine which participates in inflammation, wound healing, and remodeling of tissue. It can, furthermore, induce septic shock and cachexia. TNF-∀ is synthesized by a number of cells including monocytes, macrophages, T lymphocytes, B lymphocytes, and natural killer cells.6 TNF stimulates the release of IL-1, IL-6, IL-8, platelet activating factor, leukotrienes, thromboxane A2, and prostaglandins.7

Transforming Growth Factor TGF is a 25 kDa multi functional cytokine which plays a role in immunosuppression following trauma and supports wound healing as well as collagen synthesis. TGF, furthermore, counteracts the activation of T and B lymphocytes.9 It has been described as an anti-inflammatory cytokine which inhibits the synthesis of monocyte/macrophage-derived proinflammatory cytokines.10 TGF decreases TNF-∀ and IL-6 synthesis on a transcriptional level, while reducing IL-1 release on a posttranscriptional level.7

Interferon-! IFN-! is a glycoprotein that is a 21 to 24 kDa homodimer synthesized by activated T lymphocytes and natural killer cells. It has antiproliferative and antiviral properties and is a strong activator of mononuclear phagocytes as well as natural killer cells. IFN-! induces expression of class II MHC molecules on a number of

66

Cytokines and the Abdominal Surgeon

different cells and contributes to the differentiation of B and T lymphocytes.6 A decrease of immune activation leads to increased production of IFN-! and an increase in antigen-presenting-cell function. Excessive production of IFN-! may play a part in the induction of autoimmune diseases.5

Pro- and Anti-Inflammatory Mediators Despite the magnitude of effects and target tissues/cells of the above mentioned mediators, some of these cytokines are also referred to as pro- or anti-inflammatory mediators. So-called proinflammatory mediators are: Tumor Necrosis Factor, IL-1, IL-6 and IL-8. Anti-inflammatory mediators are: IL-4, IL-10, IL-13, Transforming Growth Factor.7

Experimental Trauma Research A number of different animals, such as rats, mice, pigs, rabbits and baboons, have been used to study the effects of blood loss and/or soft-tissue trauma on immune responses. Nonetheless, most of the work performed in this area of research was carried out using rodents. This is due to the fact that these animals are cheaper than larger animals, easier to handle and, furthermore, standardized immunological assays exist for these animals, especially for mice.11 Blunt abdominal trauma is associated with some form of tissue injury. In this regard it is well known that tissue trauma per se induces significant depression of cell-mediated and humoral immunity.12-14 Moreover, it was observed that hemorrhagic shock without soft-tissue trauma also results in significant depression of humoral and cell-mediated immunity.15-17 In addition it was observed that simple hemorrhage results in increased susceptibility to sepsis.18 Nonetheless, a recent study demonstrated that tissue trauma or hemorrhagic shock alone, does not produce a long-lasting depression of cell-mediated and humoral immunity.19 In their study Zellweger et al19 compared the effects of laparotomy (soft-tissue trauma) alone, hemorrhagic shock alone and of the combination of soft-tissue trauma with hemorrhage on macrophage and splenocyte function at 5 days after the experiment. In this study a significant depression of splenocyte and macrophage immune function at 120 hrs after trauma was only observed in animals subjected to soft-tissue trauma in conjunction with hemorrhagic shock.19 These findings indicate that only severe blunt abdominal trauma which results in soft-tissue trauma and (major) blood loss should be considered for studies of the cytokine system. With regard to this, it has been observed that soft-tissue trauma in conjunction with hemorrhagic shock resulted in a long-lasting depression of cellmediated immunity as indicated by depressed splenocyte proliferative capacity, depressed IL-2 and IL-3 release as well as depressed macrophage IL-6 release at 5 days after onset of the experiment.19 This observation of markedly depressed splenocyte and macrophage function was confirmed in a number of subsequent studies.20-22 It is of interest, that in one of these studies it was shown that closed bone fracture alone did not produce any deleterious effects on macrophage and splenocyte immune function, as opposed to the effects of soft-tissue trauma alone.22 In additional physiological studies using rats instead of mice it was observed that trauma-hemorrhage results in endothelial cell dysfunction, impaired hepatocellular function, depressed cardiac output and impaired microcirculation.23-25

Cytokines in Blunt Abdominal Trauma

67

Experimental Immunomodulation Following Trauma-Hemorrhage Several studies were carried out to determine the potentially beneficial effects of different drugs and hormones on the depressed cell-mediated and humoral immunity following trauma-hemorrhage. In this respect beneficial effects of pharmacological interventions are defined as the return of cell-mediated and humoral immune functions to levels comparable to those observed in sham-operated animals. With regard to this, beneficial effects were observed following the administration of interferon-!,26,27 ATP-MgCl2,28-30 n-3 polyunsaturated fatty acids,31 ibuprofen,32 chloroquine,33-35 anti-TNF antibodies,36 diltiazem,37 chemically modified/nonanticoagulant heparin,25,38 pentoxifylline,39 PAF-antagonists,40 prolactin41,42 and melatonin.21,43 While a number of these interventions were not as effective in the clinical setting as in the laboratory, some of these drugs still have to be clinically tested, e.g., ATPMgCl2, chemically modified heparin, n-3 polyunsaturated fatty acids, prolactin and melatonin.

Clinical Trauma Research The response to trauma begins in the immune system at the moment of injury and all patterns of trauma (sepsis, hemorrhage, ischemia, ischemia-reperfusion, softtissue trauma, burns) share the ability to activate macrophages and induce the release of proinflammatory cytokines.44,45 The production and release of all cytokines so far examined is altered by physiologic stress and it is well recognized that changes in the release of various cytokines have an important role in mediating functional changes of the immune as well as the circulatory system.46 During recent years a large number of clinical investigators have studied the effects of trauma on immunological parameters as indicated by circulating cytokine levels and/or the in vitro release of cytokines by peripheral blood mononuclear cells. Circulating levels of IL-6 and TNF were studied following elective surgery/trauma and while no changes of circulating TNF levels following surgery/trauma were observed, the authors reported a rise in acute phase protein levels (C-reactive protein, CRP) following the peak circulating levels of IL-6.47 Nonetheless, the observed changes in plasma IL-6 levels showed only poor correlation with blood loss, fever, white cell count or duration of surgery.47 Another study confirmed the observed correlation between circulating CRP and IL-6 levels.48 These investigators, however, reported a close relationship between the duration of surgery and the circulating levels of IL-6, an observation which is supported by another, more recent, study.48,49 Moreover, this correlation between trauma, rise in circulating IL-6 and CRP levels was subsequently confirmed in a study involving patients undergoing pancreaticoduodenectomy.50 In a different study involving patients with major surgery, TNF∀ and IFN-! were not detected in the circulation, while an early and short-lived IL-1 response to trauma could be observed by the authors.51 Again a rise in circulating IL-6 was observed following surgical trauma in this study.51 A study on trauma patients by Fabian et al52 did not show any overproduction of TNF in whole blood following trauma, which makes the therapeutic targeting of TNF in exaggerated inflammatory response difficult to justify. On the other hand, an in vitro study by Kim et al8 showed a strong immunomodulatory potential of IL-13 on the increased production of TNF-∀, IL-1, IL-6 and IL-8 by peripheral monocytes obtained from trauma patients.

68

Cytokines and the Abdominal Surgeon

An analysis of serial postresuscitation patient samples by Hoch et al53 demonstrated rapid increases in IL-6 and IL-8 within 12 hours after injury. Increased levels of IL-6 and IL-8 were observed for more than 5 days in patients with an ISS 25, while no significant elevation of TNF-∀ could be observed at any time-point.53 Surprisingly, in a recent study by Hauser et al54 no increased gene expression of the proinflammatory cytokines TNF-∀, IL-1 and IL-6 was observed in peripheral blood mononuclear cells from trauma patients. Furthermore, these authors reported a significant increase in IL-10 expression, while no change in IL-4 could be detected.54 This observation, which is inconsistent with the occurrence of systemic endotoxemia and subsequent global immunocyte activation after trauma may be due to the fact that the patterns of gene expression do not always correlate well with the detected levels of cytokine in the circulation and/or the culture supernatant of immunocytes. In a recent study by Schinkel et al45 it was shown that massive trauma causes the immediate activation of chemotactic cytokines (IL-8, ENA-78) and soluble adhesion molecules (sE-selectin, sP-selectin) within minutes after trauma which subsequently persist for a number of weeks (depending on the trauma mechanism). Nonetheless, no correlation between serum concentrations of these mediators and the prediction of infectious complications or outcome could be observed in this study.45 Despite the above mentioned studies and the so far inconsistent information which was collected in clinical studies it appears that increased generation of IL-1 and IL-6 (and maybe TNF) contribute to the acute phase response and hypermetabolic phase which accompanies trauma.46 Furthermore, severe immunosuppression which involves T- and B-cell function may result from the various effects of stress-related alterations in cytokine release.46 In this respect it has been proposed that traumatic stress causes a severe disintegration of the intact monocyte-T cell interaction, which is associated with profound changes in macrophage forward-regulatory capacities and substantial depression of T cell function.55 Prostaglandin E2 (PGE2) is considered as a major mediator of this stress/trauma induced immune suppression, since increased PGE2 synthesis and release results in decreased T cell mitogenesis, decreased IL-2 release and IL-2 receptor expression.55 Furthermore, PGE2 causes a shift of T-helper activities towards the TH2 direction which results in increased release of the immunosuppressive cytokines IL-4 and IL-10.55 Whether or not this proposed central role for PGE2 in the severe immune dysfunction following trauma holds true in the clinical situation certainly needs further thorough investigation. As far as cytokine alterations following blunt abdominal trauma in patients are concerned, it needs to be emphasized that blunt abdominal trauma alone, without blood loss and hypotension and/or without major tissue destruction probably does not induce significant changes of circulating cytokines. We, therefore, focused in this chapter on the known effects of hemorrhagic shock and/or soft-tissue injury as a consequence of blunt abdominal trauma on the cytokine network.

Conclusions To date a large body of information and data exists concerning the effects of trauma on the cytokine network. This data, however, was collected mainly in experimental animal research. Clinical studies did not always support the experimental laboratory results since patients and clinical trauma settings usually do not compare very well with the laboratory. Nevertheless, a clear cause and effect relationships between trauma, the acute phase response and the cytokine network (represented by IL-6) has been established. This indicates that the detection methods for the cytokines

Cytokines in Blunt Abdominal Trauma

69

possibly involved in the host reaction to trauma may not always be as accurate as necessary and, furthermore, it is very well possible that researchers are not always looking for the correct cytokine in the correct body fluid at the correct point of time following trauma. Finally, clinical researchers are limited in their access to certain body fluids/tissues (e.g., cerebrospinal fluid, liver tissue, splenic tissue). All of these limitations make clinical research of the cytokine system much more difficult than laboratory research and a lot of clinical research still remains to be carried out before routine determination of cytokines following trauma will be established as a means of clinical work-up of trauma-patients. Moreover, a long way is ahead of clinical researchers before the observed immunological changes following traumatic insult can be beneficially influenced by immunomodulatory therapeutic interventions. These immune system derived interventions should protect all cellular host defense compartments from hyper-activation as well as exhaustion and we are convinced that only a combination of drugs and therapeutical efforts can effectively control this posttraumatic dyshomeostasis of the various cell systems.55

References 1. Brenneman FD, Boulanger BR, McLellan BA et al. Acute and Long-Term Outcomes of Extremely Injured Blunt Trauma Victims. J Trauma 1995; 39:320-324. 2. Faist E, Baue AE, Dittmer H. Multiple organ failure in polytrauma patients. J Trauma 1983; 23:775-787. 3. Davis KA, Brody JM, Cioffi WG. Computed tomography in blunt hepatic trauma. Arch Surg 1996; 131:255-260. 4. Rook G. Cel-Mediated Immune Reactions. In: Roitt I, Brostoff J, Male D, eds. Immunology. 3rd ed. St. Louis; Baltimore; Boston: Mosby, 1993:8.1-8.16. 5. Feldman M. Cell Cooperation in the Antibody Response. In: Roitt I, Brostoff J, Male D, eds. Immunology. 3rd ed. St. Louis; Baltimore; Boston: Mosby, 1993: 7.1-7.16. 6. Cruse JM, Lewis RE. Illustrated Dictionary of Immunology. Boca Raton; New York; London: CRC Press, 1995. 7. Ertel W, Scholl FA, Trentz O. The role of anti-inflammatory mediators for the control of systemic inflammation following severe injury. In: Faist E, Baue AE, Schildberg FW, eds. The Immune Consequences of Trauma, Shock and Sepsis. Mechanisms and Therapeutic Approaches. Lengerich; Berlin; Duesseldorf: Pabst Science Publishers, 1996:453-470. 8. Kim C, Schinkel C, Fuchs D et al. Interleukin-13 Effectively Down-regulates the Monocyte Inflammatory Potential During Traumatic Stress. Arch Surg 1995; 130:1330-1336. 9. Bank U, Reinhold D, Kunz D et al. Regulation of neutrophil degranulation by interleukin-6 and transforming growth factor. In: Faist E, Baue AE, Schildberg FW, eds. The Immune Consequences of Trauma, Shock and Sepsis. Mechanisms and Therapeutic Approaches. Lengerich; Berlin; Duesseldorf: Pabst Science Publishers, 1996:335-345. 10. Cavaillon J, Marie C, Pitton C et al. Regulation of neutrophil derived IL-8 production by anti-inflammatory cytokines. In: Faist E, Baue AE, Schildberg FW, eds. The Immune Consequences of Trauma, Shock and Sepsis. Mechanisms and Therapeutic Approaches. Lengerich; Berlin; Duesseldorf: Pabst Science Publishers, 1996:327-334. 11. Chaudry IH, Ayala A. Models of hemorrhage and hemorrhagic shock. In: Chaudry IH, Ayala A, eds. Immunological Aspects of Hemorrhage. Austin; Georgetown: R.G. Landes Company, 1992:35-40.

70

Cytokines and the Abdominal Surgeon

12. McRitchie DI, Girotti MJ, Rotstein OD et al. Impaired antibody production in blunt trauma. Arch Surg 1990; 125:91-96. 13. Faist E, Ertel W, Mewes A et al. Trauma-induced alterations of the lymphokine cascade. In: Faist E, Ninnemann J, Green D, eds. Immune Consequences of Trauma, Shock, and Sepsis: Mechanisms and Therapeutic Approaches. Berlin: Springer Verlag, 1989:79-94. 14. Stephan RN, Mitsuyoski S, Conrad PJ et al. Depressed antigen presentation function and membrane interleukin 1 activity of peritoneal macrophages after laparotomy. Surgery 1987; 102:147-154. 15. Stephan RN, Kupper TS, Geha AS et al. Hemorrhage without tissue trauma produces immunosuppression and enhances susceptibility to sepsis. Arch Surg 1987; 122:62-68. 16. Chaudry IH, Ayala A, Ertel W et al. Editorial Review. Hemorrhage and resuscitation: Immunological Aspects. Am J Physiol 1990; 259:R663-R678. 17. Schmand JF, Ayala A, Chaudry IH. Effects of trauma, duration of hypotension, and resuscitation regimen on cellular immunity following hemorrhagic shock. Crit Care Med 1994; 22:1076-1083. 18. Ayala A, Perrin MM, Wagner MA et al. Enhanced susceptibility to sepsis following simple hemorrhage: Depression of Fc and C3b receptor mediated phagocytosis. Arch Surg 1990; 125:70-75. 19. Zellweger R, Ayala A, Zhu X et al. Effect of surgical trauma on splenocyte and peritoneal macrophage immune function. J Trauma 1995; 39:645-650. 20. Wichmann MW, Zellweger R, DeMaso CM et al. Mechanisms of immunosuppression in males following trauma hemorrhage: Critical role of testosterone. Arch Surg 1996; 131:1186-1192. 21. Wichmann MW, Zellweger R, DeMaso CM et al. Melatonin administration attenuates depressed immune functions after trauma-hemorrhage. J Surg Res 1996; 63:256-262. 22. Wichmann MW, Zellweger R, Williams C et al. Immune function is more compromised following closed bone fracture and hemorrhagic shock than hemorrhage alone. Arch Surg 1996; 995-1000. 23. Wang P, Ba ZF, Chaudry IH. Endothelial cell dysfunction occurs very early following trauma-hemorrhage and persists despite fluid resuscitation. Am J Physiol 1993; 265:H973-H979. 24. Wang P, Ba ZF, Lu MC et al. Measurement of circulating blood volume in vivo after trauma-hemorrhage and hemodilution. Am J Physiol 1994; 266:R368-R374. 25. Wang P, Ba ZF, Chaudry IH. Chemically modified heparin improves hepatocellular function, cardiac ouput, and microcirculation after trauma-hemorrhage and resuscitation. Surgery 1994; 116:169-176. 26. Ayala A, Wang P, Chaudry IH. Insights into the mechanism by which interferongamma improves macrophage function following hemorrhage and resuscitation. J Surg Res 1993; 54:322-327. 27. Ertel W, Morrison MH, Ayala A et al. Interferon-gamma attenuates hemorrhageinduced suppression of macrophage and splenocyte functions and decreases susceptibility to sepsis. Surgery 1992; 111:177-187. 28. Chaudry IH, Ohkawa M, Clemens MG. Improved mitochondrial function following ischemia and reflow by ATP-MgCl2. Am J Physiol 1984; 246:R799-R804. 29. Chaudry IH. Use of ATP following shock and ischemia. Ann N Y Acad Sci 1990; 603:130-141. 30. Wang P, Ba ZF, Morrison MH et al. Mechanism of the beneficial effects of ATPMgCl2 following trauma-hemorrhage and resuscitation: Downregulation of inflammatory cytokine (TNF, IL-6) release. J Surg Res 1992; 52:364-371.

Cytokines in Blunt Abdominal Trauma

71

31. Ertel W, Morrison MH, Ayala A et al. Modulation of macrophage membrane phospholipids by n-3 polyunsaturated fatty acids increases interleukin-1 release and prevents suppression of cellular immunity following hemorrhagic shock. Arch Surg 1993; 128:15-21. 32. Ertel W, Morrison MH, Meldrum DR et al. Ibuprofen restores cellular immunity and decreases susceptibility to sepsis following hemorrhage. J Surg Res 1992; 53:55-61. 33. Ertel W, Morrison MH, Ayala A et al. Chloroquine attenuates hemorrhagic shockinduced immunosuppression and decreases susceptibility to sepsis. Arch Surg 1992; 127:70-76. 34. Ertel W, Morrison MH, Ayala A et al. Chloroquine attenuates hemorrhagic shock induced suppression of Kupffer cell antigen presentation and MHC class II antigen expression through blockade of tumor necrosis factor and prostaglandin release. Blood 1991; 78:1781-1788. 35. Zhu X, Ertel W, Ayala A et al. Chloroquine inhibits macrophage tumour necrosis factor- mRNA transcription. Immunology 1993; 80:122-126. 36. Ertel W, Morrison MH, Ayala A et al. Anti-TNF monoclonal antibodies prevent haemorrhage induced suppression of Kupffer cell antigen presentation and MHC class II antigen expression. Immunology 1991; 74:290-297. 37. Meldrum DR, Ayala A, Chaudry IH. Mechanism of diltiazem’s immunomodulatory effects after hemorrhage and resuscitation. Am J Physiol 1993; 265:C412-C421. 38. Zellweger R, Ayala A, Zhu X et al. A novel nonanticoagulant heparin improves splenocyte and peritoneal macrophage immune function after trauma-hemorrhage and resuscitation. J Surg Res 1995; 59:211-218. 39. Wang P, Ba ZF, Zhou M et al. Pentoxifylline restores cardiac output and tissue perfusion following trauma-hemorrhage and decreases susceptibility to sepsis. Surgery 1993; 114:352-359. 40. Zellweger R, Ayala A, Schmand JF et al. PAF-antagonist administration after hemorrhage-resuscitation prevents splenocyte immunodepression. J Surg Res 1995; 59:366-370. 41. Zellweger R, Wichmann MW, Ayala A et al. Prolactin: a novel and safe immunomodulating hormone for the treatment of immunodepression following severe hemorrhage. J Surg Res 1996; 63:53-58. 42. Zhu X, Zellweger R, Ayala A et al. Prolactin inhibits the increased cytokine gene expression in Kupffer cells following haemorrhage. Cytokine 1996; 8:134-140. 43. Wichmann MW, Haisken JM, Ayala A et al. Melatonin administration following hemorrhagic shock decreases mortality from subsequent septic challenge. J Surg Res 1996; 65:109-114. 44. Harris BH, Gelfand JA. The immune response to trauma. Semin Pediatr Surg 1995; 4:77-82. 45. Schinkel C, Faist E, Zimmer S et al. Kinetics of circulating adhesion molecules and chemokines after mechanical trauma and burns. Eur J Surg 1996; 162:763-768. 46. Abraham E. Effects of stress on cytokine production. Methods Achiev Exp Pathol 1991; 14:45-62. 47. Pullicino EA, Carli F, Poole S et al. The relationship between the circulating concentrations of interleukin 6 (IL-6), tumor necrosis factor (TNF) and the acute phase response to elective surgery and accidental injury. Lymphokine Res 1990; 9:231-238. 48. Yoshizaki K. Clinical significance of cytokines-interleukin 6 in disease. Rinsho Byori 1990; 38:375-379. 49. Sakamoto K, Arakawa H, Mita S et al. Elevation of circulating interleukin 6 after surgery: factors influencing the serum level. Cytokine 1994; 6:181-186.

72

Cytokines and the Abdominal Surgeon

50. Wortel CH, van Deventer SJ, Aarden LA et al. Interleukin-6 mediates host defense responses induced by abdominal surgery. Surgery 1993; 114:564-570. 51. Baigrie RJ, Lamont PM, Dallman M et al. The release of interleukin-1 beta (IL-1) precedes that of interleukin 6 (IL-6) in patients undergoing major surgery. Lymphokine Cytokine Res 1991; 10:253-256. 52. Fabian TC, Croce MA, Fabian MJ et al. Reduced tumor necrosis factor production in endotoxin-spiked whole blood after trauma: Experimental results and clinical correlation. Surgery 1995; 118:63-72. 53. Hoch RC, Rodriguez R, Manning T et al. Effects of accidental trauma on cytokine and endotoxin production [see comments]. Crit Care Med 1993; 21:839-845. 54. Hauser CJ, Lagoo S, Lagoo A et al. Human peripheral mononuclear cells do not show proinflammatory patterns of cytokine transcription in early trauma: a preliminary report. Shock 1995; 4:247-250. 55. Faist E, Schinkel C, Zimmer S. Update on the Mechanisms of Immune Suppression of Injury and Immune Modulation. World J Surg 1996; 20:454-459.

Cytokines in Acute Pancreatitis

73

CHAPTER 7

Cytokines in Acute Pancreatitis James Norman and Larry C. Carey

A

cute pancreatitis is a serious clinical entity for which no directed therapy exists beyond simple supportive care. More than 250,000 people are hospitalized with a bout of acute pancreatitis every year in the United States, a rate which is similar in most European countries. Although the causative etiology differs from report to report depending on the type of hospital and the country from which the study originated, most experts now believe the cause is fairly uniform world wide. Alcohol and gallstones combine as etiologic factors for approximately 90% of all cases, with the remainder being secondary to bile duct instrumentation, hyperlipidemia, hypercalcemia, trauma, and a few other rare causes. The actual inciting event leading to the development of pancreatic inflammation is still poorly understood. Most investigators believe that the normal process of proenzyme production, packaging, and secretion is impeded leading to one enzyme activating another, a process termed colocalization. The prematurely activated enzymes are then preferentially secreted at the basolateral rather than apical aspect of the acinar cell. The end result is the presence of activated digestive enzymes within the pancreatic interstitium. Many treatments have been proposed and tested which aim to attenuate this autodigestive process, yet a number of clinical trial have failed to show a benefit.1

Clinical Pancreatitis and the Immune Response Regardless of whether the initiating event is alcohol, gallstones or some other less common cause, acute pancreatitis progresses in a predictable manner through several pathologic stages (Fig. 7.1). The first is the intracellular disturbances described above which lead to activation of digestive enzymes. Very quickly, a local inflammatory process is initiated which results in the local production of inflammatory mediators. Virtually all patients with acute pancreatitis will experience these two phases, with some resolving completely at this point. Most patients, however, will go on to enter the third phase which is characterized by a systemic hyperinflammatory state expressed as the development of fever, tachycardia, tachypnea, and mild acid-base disturbances.2 Although this systemic hyperinflammatory state is usually mild, occasionally it may be very severe resulting in overt distant organ failure. The forth and final stage of pancreatitis is relatively uncommon and occurs a number of days or even weeks following the initial inflammatory stages and is characterized by the development of local complications such as infected necrosis, pancreatic abscess, or pseudocyst formation. Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

74

Cytokines and the Abdominal Surgeon

Fig. 7.1. The four pathologic stages of acute pancreatitis. Pancreatitis progresses through four rather predictable states beginning with the activation of intrapancreatic enzymes leading to the rapid development of local inflammation. Activated enzymes as well as products of local inflammation can have dramatic effects on acinar cell viability. The third phase of acute pancreatitis begins shortly after the second and is characterized by the signs and symptoms associated with the systemic production of inflammatory mediators. The trigger for this systemic hyperinflammatory response is still unclear but the mediators produced are ultimately responsible for the distant organ dysfunction, hypoperfusion, tachycardia, and fluid shifts known collectively as the systemic inflammatory response syndrome (SIRS). The fourth phase occurs in a minority of acute pancreatitis patients a week or more into the disease and is characterized by the development of local complications such as infected necrosis and pseudocyst formation.

The key to understanding the complex pathophysiology of acute pancreatitis lies in determining why some patients fail to develop systemic hyper-inflammation and the systemic inflammatory response syndrome (SIRS) associated with it. These patients have a quick, uncomplicated recovery. This response is contrasted by others who rapidly progress into wide-spread inflammatory mediator over-production leading to distant organ dysfunction. Significant strides were taken in understanding this process in the early 1990s with the discovery that acute pancreatitis was associated with the appearance of dramatic levels of circulating inflammatory cytokines. A number of investigators followed the appearance of cytokines in patients hospitalized with pancreatitis hoping to find a marker predictive of disease severity. Each has shown that circulating levels of IL-6 and IL-8 (and often IL-1 and TNF) were higher in patients with complicated pancreatitis than they were in those with simple uncomplicated disease. When compared to Ranson’s and APACHE scoring, serum cytokine concentrations were usually found to be more specific in predicting the severity of acute pancreatitis, end-organ failure, overall mortality and even the duration of hospital stay. These close correlations provided the first direct evidence that cytokines play a major role in determining just how far a patient progresses down the path from local to systemic illness.3 IL-6 is a member of the inflammatory cytokine family which is well characterized as the primary inducer of the acute phase protein response during all types of

Cytokines in Acute Pancreatitis

75

insults in direct correlation to their severity.4 It is not surprising, therefore, that the severity of pancreatitis would be manifest by the degree and duration of IL-6 elevation. Patients with complicated and lethal pancreatitis have significantly higher serum levels of IL-6 than those with simple uncomplicated disease. This correlation is so high that a number of authors advocate measurement of IL-6 during the first 24 hours of acute pancreatitis onset to help decide the need for more aggressive resuscitation and monitoring techniques including whether to put a patient in the ICU and initiate invasive hemodynamic monitoring. IL-8 is also recognized as an inflammatory cytokine whose primary target cell is the neutrophil, where it causes degranulation and the release of enzymes (e.g., elastase) capable of tissue destruction when overproduced5 (Fig. 7.2). The relationship between pancreatitis severity and circulating levels of IL-8 parallels IL-6 production closely. Results from recent clinical trials of a platelet activating factor antagonist have confirmed the ability of these two cytokines to predict outcome and the occurrence of organ dysfunction,6,7 yet the lack of rapid, reproducible means to determine IL-6 or IL-8 levels by most hospitals has thus far kept these mediators from becoming more commonly used in the clinical setting. As such tests become available, it is likely that measurement of IL-6 and/or IL-8 early in the course of pancreatitis will become routine. Although IL-6 and IL-8 are reliable measures of systemic stress, their production is not felt to be the driving force behind pancreatitis-associated organ dysfunction and therefore they are not potential targets for antagonism at this time. In contrast, IL-1 and TNF are the primary members of the inflammatory cytokine family, are primary inducers of IL-6 and IL-8 production, and are believed to initiate and/or propagate nearly all of the detrimental consequences of severe pancreatitis, such as fever, hypoperfusion, circulatory collapse, shock, metabolic acidosis, cardiac dysfunction, and the production of ARDS. The presence of both in the serum and ascites of patients with severe pancreatitis has been documented by a number of investigators, but this is an inconsistent finding. Production of both of these cytokines is closely regulated and occurs only briefly in most circumstances, and as during sepsis, their presence in the systemic circulation during pancreatitis portends a very poor prognosis. As will be discussed subsequently, nearly all experimental models of pancreatitis have implicated IL-1 and TNF as the major pathologic cytokines associated with local and systemic tissue destruction. It must be remembered that the presence of inflammatory cytokines in the serum is not a prerequisite for them to be responsible for the development of organ dysfunction and SIRS.

Amplification of the Pancreatitis-Associated Inflammatory Cascade Although there has been a recent emphasis on the importance of cytokine activity during pancreatitis, a number of other inflammatory mediators are also believed to play a major role in the progression of pancreatitis into a systemic process.8 Those which contribute significantly include platelet activating factor (PAF), nitric oxide (NO), and reactive oxygen intermediates, with bradykinin and compliment likely playing a less important role. Figure 7.2 demonstrates the close relationships and means for amplification which exist between each of the aforementioned compounds produced during severe acute pancreatitis. It is the cumulative effect of these mediators which eventually leads to vascular leakage, hypovolemia, ARDS, shock, and MOSF. As illustrated, essentially all are produced under the influence of inflammatory cytokines or play an active role in initiating or amplifying the cytokine cascade. For

76

Cytokines and the Abdominal Surgeon

example, PAF is a very potent vasodilator and leukocyte activator which has been implicated in the development of SIRS during acute pancreatitis. The production of PAF is so intimately tied to that of IL-1 and TNF that inhibition of PAF attenuates IL-1 and TNF production just as inhibition of either of these cytokines attenuates PAF production. Although our emphasis has been placed on the role of cytokines, their production influences the appearance of other mediators which also play a large role in determining pancreatitis outcome.

Local Cytokine Production During Pancreatitis Concomitant with the clinical observations of cytokine and PAF activity during acute pancreatitis was the finding that all animal models of pancreatitis demonstrate the same phenomenon. Inflammatory cytokines are produced in the pancreas itself within 30 minutes of experimental pancreatitis induction, which is often prior to appreciable changes in pancreatic histology.9,10 Although it has been suggested that the acinar cell may produce IL-1 and TNF, the majority of evidence to date suggests that leukocytes which invade the pancreas in great number during pancreatitis are the predominant source. One of the most important observations made during these experiments was that the concentration of TNF and IL-1 within pancreatic tissues was several orders of magnitude higher than they were in the serum, levels which are known to be toxic to many cell types. These findings were substantiated by showing high levels of TNF in the portal vein of animals with pancreatitis which was significantly higher than corresponding systemic levels. It is believed that the liver serves to clear TNF produced in the pancreas thereby preventing it from reaching the general circulation. These findings also help explain the difficulty in isolating TNF routinely from the serum of patients with pancreatitis.

The Effects of Local Cytokine Production on Pancreatitis Severity Since inflammatory cytokines are present in high concentrations within the pancreatic parenchyma during acute pancreatitis, a number of experiments have been conducted to determine if they were capable of initiating pancreatitis. Studies on isolated human pancreata perfused with large amounts of IL-1 or TNF found little or no evidence that these mediators alone could propagate pancreatitis. Further, in vitro studies showed that IL-1 and TNF could not induce co-localization or the activation or release of enzymes from acinar cells, but exposure to TNF in concentrations found within the pancreas during pancreatitis was capable of killing the cells. Good evidence has also been found that IL-6 plays a regulatory role within the acinar cell through induction of regulatory genes including CCK receptor expression.11 These findings may help explain why cytokine blockade is capable of decreasing pancreatitis severity while all cytokines tested to date lack the ability to initiate enzyme activation or the development of pancreatitis. The use of knockout animals has also helped investigate whether IL-1 and/or TNF are necessary for the initiation and development of pancreatitis by showing that pancreatitis does develop in the absence of receptors for either of these two cytokines, yet its severity/lethality never reaches that of wild-type mice12,13 (Fig. 7.3). Combined, these studies have helped establish that pancreatitis is not triggered or induced by IL-1 and/or TNF, yet both play an important and very detrimental role to its progression.

Cytokines in Acute Pancreatitis

77

Fig. 7.2. Amplification of the Pancreatitis-Associated Inflammatory Cascade. Acute pancreatitis incites an inflammatory cascade within the pancreas as well as in multiple tissues of the body. The systemic activation of leukocytes results in the production of a number of powerful mediators such as interleukin (IL)-1, tumor necrosis factor (TNF), platelet activating factor (PAF), IL-6, IL-8, and nitric oxide (NO). Each of these mediators possess the ability to feed back and amplify the production of the others. Pancreatitis has the propensity to incite this process beyond the point of normal control mechanisms. The end result is a cascade of events leading to the release of multiple mediators in tissues such as the lungs, liver, pancreas, and spleen.

Cytokine-Induced Acinar Cell Death During Pancreatitis Another mechanism by which TNF (and possibly IL-1) may contribute to the severity of pancreatitis is through the induction of apoptosis of acinar cells. Apoptosis is a well orchestrated process by which nucleated cells are induced to kill themselves.14 This process is distinctly different from simple necrosis which is not energy dependant and does not follow a preset cascade of events. Several investigators have recently demonstrated that acute pancreatitis is associated with the induction of acinar cell apoptosis, the degree of which mimics the severity of pancreatitis. These studies suggested that a portion of pancreatic cell death which was previously believed to be due to simple necrosis mediated by autodigestion, may in fact be apoptotic cell death mediated by TNF produced within the inflamed gland.15,16 Although it is still unclear whether apoptosis is harmful or even protective during acute pancreatitis, it seems apparent that intra-pancreatic cytokines play a significant regulatory role in acinar cell viability.

Acute Pancreatitis Induces Systemic Hyperinflammation Further insight into the role of IL-1 and TNF in the propagation of distant organ failure was given by the observation that both cytokines are produced systemically

78

Cytokines and the Abdominal Surgeon

Fig. 7.3. The importance of IL-1 and TNF activity on pancreatitis severity. The importance of IL-1 and TNF to the progression of pancreatitis is demonstrated by the use of transgenic knockout animals which lack active receptors for either IL-1 or TNF. Although the loss of cytokine activity has no effect on the initiation of pancreatitis, the end result is a much less severe hyperinflammatory response and a significant reduction in mortality.

during pancreatitis and not just within the pancreas. Regardless of the animal model used, IL-1 and TNF are produced in the spleen, lung, and liver with pancreatic production always preceding that in distant sites, by hours or even days, depending on the rapidity with which pancreatitis develops.17 Patients with severe pancreatitis possess circulating leukocytes which produce significantly more TNF, IL-6, IL-8 (and to a lesser degree, IL-1) while possessing circulating monocytes, lymphocytes, and PMNs which are hyperstimulated when compared to those with mild uncomplicated pancreatitis. Just as was seen in various animal models, these mediators are overproduced without regard to the etiology of clinical pancreatitis. Inflammatory cytokine production (along with PAF activity) within the pulmonary parenchyma is now believed to be the direct cause of ARDS whether during sepsis or acute pancreatitis. What remains a mystery, however, is the signal responsible for causing distant tissues to respond to pancreatic inflammation in this manner. As shown in Figure 7.1, local pancreatic cytokine production is likely to contribute to systemic cytokine gene induction, but this is probably not the primary mechanism. High levels of activated trypsin and other pancreatic enzymes can induce cytokine production from macrophages in vitro, but it is still unclear if activated enzymes in concentrations typically seen during clinical pancreatitis can induce this reaction, or if some other as of yet unknown mediator is responsible. Together, the activation of leukocytes systemically and their production of inflammatory cytokines in multiple tissues during acute pancreatitis resembles that seen in

Cytokines in Acute Pancreatitis

79

severe sepsis and undoubtedly plays a central role in the development of the clinical syndrome we know as acute pancreatitis.

Antagonism of Cytokines During Experimental Acute Pancreatitis As it became clear in the early 1990s that pancreatitis was associated with the appearance of inflammatory cytokines, a number of investigators postulated that their antagonism would have beneficial effects. The first studies utilized anti-TNF antibodies, soluble TNF receptors, or IL-1 receptor antagonists. Regardless of the approach, these studies showed that proximal blockade of the cytokine cascade would significantly decrease pancreatic edema, necrosis, and inflammation while decreasing mortality by more than half. Delayed (therapeutic) cytokine antagonism was shown to be effective as well, becoming the first pancreatitis experiment to demonstrate this critical correlation to clinical pancreatitis. It is now fairly well established that interruption of the cytokine cascade at any point between gene induction and receptor binding will attenuate experimental pancreatitis as long as the blockade takes place prior to the onset of severe organ dysfunction. It is less clear how effective delaying therapy beyond this point will be, but the efficacy of these agents is likely to fall off just as was seen in a number of sepsis trials.

Is Anticytokine Therapy for Acute Pancreatitis a Clinical Possibility? A number of treatments are now available to treat specific aspects of the clinical syndrome of acute pancreatitis. Ventilator management has progressed to the point that ARDS is a less common cause of pancreatitis associated mortality, yet it continues to be the major cause of morbidity leading to the need for intensive care monitoring.2 Similarly, the judicious use of broad spectrum antibiotics such as imipenem and cefuroxime has been shown to reduce the number of septic complications and improve survival in a subgroup of patients with acute necrotizing pancreatitis.3 The need for a specific therapy for all patients with this disease aimed at preventing distant organ dysfunction and the sequelae of pancreatitis associated systemic hyperinflammation, however, still remains. A great deal has been learned over the past few years about the production of inflammatory cytokines during acute pancreatitis. At the same time, the dominant role these few mediators play in the pathogenesis of SIRS and organ dysfunction is now becoming much better understood. There is little doubt that preventing the effect of IL-1, TNF, or PAF dramatically alters the expected course of experimental pancreatitis. The question remaining is whether similar antagonism during clinical pancreatitis would benefit patients with this disease. One of the major considerations to be examined is the suitability of acute pancreatitis to cytokine antagonism in the clinical setting and whether a therapeutic window for such antagonism exists.3 Figure 7.4 details the typical presentation and subsequent development of organ dysfunction in patients with pancreatitis. Although some patients present a day or two after the onset of pain, the vast majority seek medical attention within 8 to 12 hours. Similarly, few patients presenting shortly after the onset of pain will exhibit organ dysfunction at that time. By the second and third day, however, the incidence of organ dysfunction rises rapidly to distinguish between patients likely to have a protracted and complicated course from those with simple uncomplicated pancreatitis. It is now known that cytokine production during clinical pancreatitis begins shortly after pain onset but does not peak until 36 to

80

Cytokines and the Abdominal Surgeon

Fig. 7.4. Time course of acute pancreatitis progression demonstrating a therapeutic window for inflammatory mediator antagonism. The majority of patients with acute pancreatitis will present within 18 hours after the onset of pain. This is followed closely by inflammatory cytokine production usually lasting several days. Although distant organ dysfunction is occasionally manifest at the time of presentation, the vast majority of patients develop severe systemic manifestations of pancreatitis 2 to 4 days later. This type of presentation allows for an interventional window during which time specific inflammatory mediator antagonists could be administered to attenuate or block the development of distant organ dysfunction/failure.

48 hours later. If the goal is to prevent distant organ dysfunction (the major cause of pancreatitis morbidity and mortality) this clinical scenario provides a theoretical therapeutic window beginning at hospital presentation and lasting for two to three days (Fig. 7.4). This has been supported by the results of the European phase III trial which showed a more beneficial effect when a PAF antagonist was given within 48 hours of pain onset when compared to those receiving the antagonist between 48 and 72 hours after pain onset or even later. Directed therapy for acute pancreatitis has eluded physicians for many years. This clinical goal has taken many different forms, some of which worked poorly even in experimental animals. Cytokine antagonism has been shown to be effective in virtually all experimental animal models of acute pancreatitis even when implemented in a delayed fashion. Anticytokine therapy in patients with acute pancreatitis has distinct advantages over similar therapies during sepsis in that pancreatitis patients will declare the onset of their disease and therefore the initiation of the cytokine cascade. In contrast, septic patients typically declare themselves only with the onset of systemic organ dysfunction, usually well after the cytokine cascade has been initiated or even peaked. The therapeutic window provided by pain onset during acute pancre-

Cytokines in Acute Pancreatitis

81

atitis may allow antagonism of specific mediators prior to the initiation of dramatic cascades which eventually lead to inflammatory mediator overproduction. Since the majority of the morbidity and mortality of this disease is due to this cascade of events, acute pancreatitis may be ideally suited for this type of antagonism. As we learn more about these powerful mediators, a scenario is likely to develop which will allow specific therapies during different phases of pancreatitis to prevent progression of gland destruction and the development of systemic complications.

References 1. Leach SD, Gorelick FS, Modlin IM. New perspectives on acute pancreatitis. Scand J Gastroenterol 1992; 27:Suppl 192:29-38. 2. Steinberg W, Tenner S. Acute pancreatitis. New England J Med 1994;330:1198-1220. 3. Norman J. The role of cytokines in the pathogenisis of acute pancreatitis. Amer J Sur 1997 in press. 4. Kishimoto T, Akira S, Narazaki M et al. Interleukin-6 family of cytokines and gp130. Blood 1995; 86:1243-1254. 5. Baggiolini M, Loetscher P, Moser B. Interleukin-8 and the chemokine family. Int J Immunopharmac 1995;17:103-108. 6. Kingsnorth AN, Galloway SW, Formela LJ. Randomized, double-blind phase II trial of Lexipafant, a platelet-activating factor antagonist, in human acute pancreatitis. British J Surg (1995); 82:1414-1420. 7. ingsnorth AN. Early treatment with lexipafant, a platelet activating factor antagonist reduces mortality in acute pancreatitis: A double blind, randomized, placebo controlled study. Gastroenterology; 112:A453. 8. Kinsgnorth AN. The role of cytokines and their inhibitors in acute pancreatitis. Gut 1997; 40:1-4. 9. Norman J, Fink G, Franz M. Acute pancreatitis induces intrapancreatic tumor necrosis factor gene expression. Arch of Surg 1995; 130:966-970. 10. Fink G, Norman J. Specific changes in the pancreatic expression of the interleukin1 family of genes during experimental acute pancreatitis. Cytokine. In Press. 11. Dusetti NJ, Ortiz EM, Mallo GV et al. Pancreatitis-associated protein I (PAP I), an acute phase protein induced by cytokines. Identification of two functional interleukin-6 response elements in the rat PAP I promoter region. J Biol Chem 1995; 270:22417-22421, 12. Norman JG, Fink G, Franz M et al. Active interleukin-1 receptor required for maximal progression of acute pancreatitis. Ann of Surg 1996; 223:163-169. 13. Denham W, Fink G, Norman J. Transgenic animals demonstrate modest additive detrimental effects of IL-1 and TNF during acute pancreatitis. Gastroenterology 1997. In Press. 14. Nagata, S. Apoptosis by death factor. Cell 1997; 69:355-365. 15. Gukoyskaya AS, Sandoval D, Zaninovich V et al. Tumor necrosis factor regulates cell death in caerulein-induced pancreatitis. Gastroenterology 1996; 111:A2791. 16. Kaiser AM, Saluja AK, Sengupta A et al. Relationship between severity, necrosis, and apoptosis in five models of experimental acute pancreatitis. Am J Physiol 1995; 269:C1295-C1304. 17. Norman J, Fink G, Denham W et al. Tissue specific cytokine production during experimental acute pancreatitis: A problem mechanism for distant organ dysfunction. Dig Dis and Sciences 1997; 42:1783-1788.

Cytokines and Peri-Operative Nutrition

83

CHAPTER 8

Cytokines and Peri-Operative Nutrition Hideaki Saito and Lin Ming-Tsan

Introduction urgical stress triggers the production of a variety of cytokines by immune cells.1,2 These cytokines initiate immunological, hematological and metabolic alterations in association with surgical stress. Appropriate synthesis and release of cytokines, such as TNF, IL-1, IL-6 and IFN-! are essential for normal host defense. In contrast, exaggerated systemic cytokine responses may be harmful to the host. Recent studies have demonstrated that the production of cytokines following surgical insult is influenced by nutritional support. This production may be influenced by nutritional status, nutritional routes, specific nutrients such as glutamine and fatty acids, and anabolic hormones (Table 8.1). These nutritional modalities may exert beneficial biological effects in response to surgical insult through the modulation of cytokine production. Understanding the role of nutrients in cytokine production in normal states, as well as under surgical stress, facilitates choosing the most appropriate nutritional support for preventing complications and improving outcomes of postoperative patients. This article focuses on the role of nutrition in the modulation of cytokine responses in surgical stress, especially in the perioperative period.

S

Nutritional Route and Cytokine Biology Previous investigations have suggested the advantages of enteral nutrition over parenteral nutrition following surgical insult. These advantages include maintenance of intestinal barriers,3 improvement of systemic and local protein metabolism,3,4 and better stress hormone responses.3,5 Antecedent or early enteral feeding has been shown to improve outcomes in various animal models and in trauma patients, as compared to parenteral feeding.6-9 However, the precise mechanisms by which enteral feeding maintains immunity and improves outcomes, under conditions of surgical stress, are unclear. Recent investigations have demonstrated that part of the effectiveness of the enteral route may lie in appropriate cytokine production by the immune cells in response to inflammatory stimuli.

Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

84

Cytokines and the Abdominal Surgeon

Table 8.1. Nutritional modalities that modulate cytokine biology 1. Nutritional routes a. Parenteral b. Enteral 2. Specific nutrients a. Polyunsaturated fatty acids b. Glutamine c. Arginine d. Nucleotides e. Vitamins E and D f. Combinations of specific diets 3. Anabolic hormones a. Growth hormone b. Insulin-like growth factor 1

Nonstressed Stable Conditions Circulating Cytokines

Lowry et al10 measured circulating TNF, IL-1 and IL-6 levels in normal volunteers and patients before and after one week of parenteral nutrition. Systemic TNF, IL-1 and IL-6 levels were undetectable in healthy volunteers before and after 1 week of TPN. Both TNF and IL-6 were detected in the systemic circulation in some patients both before and after TPN, but post-TPN levels were not significantly increased. Fong et al5 also reported that circulating TNF levels in nonstressed subjects receiving TPN were indistinguishable from those in subjects on enteral feeding. These results suggest that parenteral nutrition has little effect on circulating cytokines without an associated overt stress. In contrast, Santos et al11 demonstrated higher plasma levels of TNF and IL-6 in volunteers receiving enteral feeding than in those receiving TPN. We investigated the effect of preoperative nutritional route on the cytokine response in patients with colorectal cancer (Fig. 8.1).12 The levels of IL-6 and IL-8 were higher in the TPN patients than in the orally fed patients, indicating that effects stimulating the production of systemic cytokines may have been present in the TPN patients before surgery. Thus, human data concerning the effect of nutritional route on circulating cytokines in an unstressed condition are contradictory. Animal studies have demonstrated that the route of nutritional supply does not influence serum cytokine levels under nonstressed stable conditions.7,13

Tissue Cytokine Biology Evidence is accumulating that the gut is a cytokine-producing organ. Gut associated lymphoid tissues include lymphocytes, mast cells, macrophages, Paneth cells and M cells. These cells are a rich source of cytokines such as IL-1,2, 4, 5, 6, TNF, and IFN-!.14 In addition, enterocytes15 produce cytokines. Ogle et al16 have demonstrated messenger RNA expression for TNF-∀, IL-1 and IL-6 to be increased in the jejunum of parenterally fed rats, as compared with chow-fed controls, suggesting local stimulation of intestinal cytokine-producing cells by bacteria and endotoxin following parenteral nutrition.

Cytokines and Peri-Operative Nutrition

85

Fig. 8.1. Systemic IL-6 responses before and after colon surgery with different nutritional routes (left). The IL-6 levels were higher in the TPN than in the oral group before the operation (*p=0.09 vs the oral group). Plasma IL-6 levels were significantly higher in the TPN group than in the oral group on postoperative day 1(**p<0.05 vs the oral group). Plasma IL-6 levels on postoperative day 1 vs amounts of intraoperative blood loss (right). Plasma IL6 levels showed a positive correlation with the amount of blood loss in the TPN group (P<0.05, r=0.881), but not in the oral group. The difference in the regression relation between the TPN and oral groups was significant (*p<0.01, test by z score). From Lin M, Saito H, Fukushuma R et al. Preoperative total parenteral nutrition influences postoperative systemic cytokine responses after colorectal surgery. Nutrition 1997; 13:8-12.

We determined the effects of nutritional route on the TNF levels in peritoneal lavage fluid using rats (Fig. 8.2).7 The TNF levels were significantly higher in parenterally fed rats than in those on enteral feeding. One possible source of the increase in peritoneal TNF is immune cells in the gut associated lymphoid tissue. Another possibility is that bacteria or bacterial products, such as endotoxin translocated from the gut to the peritoneal cavity, stimulated resident peritoneal cells even under stable conditions. Luminal flora have been proposed to influence the production of cytokines by mouse peritoneal macrophages.17 It is noteworthy that LPS stimulated production of TNF by resident peritoneal cells was significantly diminished in the parenterally as compared to the enterally fed rats in our study (Fig. 8.2).7 Development of endotoxin tolerance may secondarily suppress TNF production by resident peritoneal cells. The liver, spleen, and lung may be more important sites of cytokine production than the gut, because of their numerous tissue macrophages. Rats fed TPN instead of chow have higher levels of cell-associated TNF-∀ in the liver.13 Billiar et al18 demonstrated that Kupffer cells from rats with prolonged intestinal bacterial overgrowth produced more IL-1 in response to LPS than cells from control rats. They hypothesized that gut-derived bacteria/endotoxin primes Kupffer cells directly or indirectly by increasing portal levels of lymphokines and eicosanoids via the stimulation of immune cells in the gut-associated lymphoid tissue. TPN promote bacterial overgrowth and translocation. Thus, TPN, but not enteral nutrition, may activate Kupffer cells to produce excessive cytokines in response to subsequent stimuli. The advantages of enteral over parenteral nutrition may be due, in part, to well-regulated cytokine production by the liver.18

86

Cytokines and the Abdominal Surgeon

Fig. 8.2. TNF levels in peritoneal lavaged fluid (left) and TNF production by LPS stimulated resident peritoneal cells (right) in a rat peritonitis model. Enteral and parenteral feeding groups received an identical standard hyperalimentation solution for seven days. Modified from Lin MT, Saito H, Fukushima R et al. Route of nutritional supply influences local, systemic, and remote organ responses to intraperitoneal bacterial challenge. Ann Surg 1996; 223:84-93.

Expressions of splenic and intestinal cytokines are different after parenteral nutrition.16 Splenic mRNA expressions for TNF-∀ and IL-6 were decreased, whereas the intestinal expressions of cytokine mRNA were increased in rats receiving oneweek TPN. In addition, productions of TNF-∀ and IL-1 by splenic macrophages were decreased following TPN in both LPS-untreated and LPS-treated cultures. Moreover, TNF production in response to LPS by pulmonary macrophages was significantly decreased in TPN-fed rats as compared to chow-fed rats.19 A possible explanation for the impaired production of cytokines by splenic and pulmonary macrophages after TPN is the induction of endotoxin tolerance. Depressed in vitro function of splenic and pulmonary macrophages following TPN may lead to increased infectious complications after surgical insults.

Stressed Conditions Circulating Cytokine Levels

Fong et al5 compared the effects of nutritional route on the immunologic and metabolic responses to intravenous endotoxin administration in normal human volunteers. Antecedent TPN significantly worsened systemic manifestations of endotoxemia as compared to enteral nutrition (EN). Peak levels of arterial epinephrine and glucagon were significantly higher in the TPN than in the EN subjects. Circulating TNF levels after endotoxin administration were also significantly higher in the TPN group than in the EN group. It is noteworthy that hepatic venous levels of TNF were consistently higher than arterial levels, suggesting that the splanchnic bed is an important source of TNF during sepsis and endotoxemia. It is hypothesized that food deprivation in the gut may increase bacterial/endotoxin translocation from the gut lumen into the portal circulation.10 Exposure of Kupffer cells to translocated bacteria/endotoxin would prime the immune system, thereby enhancing TNF production in response to subsequent endotoxin challenge.

Cytokines and Peri-Operative Nutrition

87

In contrast, Santos et al11 showed that TNF responses to intravenous endotoxin in humans receiving TPN were similar to those of subjects receiving enteral nutrition. However, a diminished plasma IL-6 response was observed in the TPN groups. The disparity in results between these two studies may be due to differences in study design and methodology: diet quantity, duration of nutritional support, TPN regimens and stimulation of cytokine production. We investigated whether preoperative nutritional routes influence the systemic cytokine responses in patients following surgery for colorectal cancer.12 Plasma IL-6 and IL-8 levels were significantly higher on postoperative day 1 in the TPN group than in the oral group. The IL-6 levels on postoperative day 1 showed a significant positive regression relation with the amounts of blood loss only in the TPN group (Fig. 8.1). The slope of the regression line was steeper in the TPN than in the orally fed group. Thus, routes of nutritional supply may have an impact on the production of systemic cytokines postoperatively. Our study suggests that preoperative nutrition via the enteral route may provide better regulation of systemic cytokine responses following surgery than parenteral nutrition. An animal experiment conducted in our laboratory also demonstrated that the antecedent nutritional route influences systemic cytokine responses after intraperitoneal bacterial challenge.7 TPN-rats had higher serum TNF levels after challenge than the enterally fed rats. One reason for the higher serum TNF in parenteral nutrition may be altered gut-liver interactions.5,10 Another possibility is that the depressed local bacterial clearance in the setting of parenteral nutrition leads to systemic spread of bacteria, which may ultimately initiate greater systemic production of TNF. Exaggerated production of systemic TNF may induce multiple organ failure and thereby be fatal. In contrast, serum IFN-! levels were higher in the enteral group than in the TPN group after bacterial challenge (Fig. 8.3). Increased systemic IFN-! levels may protect TEN (total enteral nutrition) rats from infection and thereby result in better survival.20,21

Cytokine Biology at Local Inflammatory Sites Various cytokines, such as TNF, IL-1, and IFN-!, are important humoral factors in host defense at the local inflammatory site. The antecedent nutritional route influences cytokine responses at local inflammatory sites and in remote organs as well as systemic responses to surgical insult.7 We compared the effect of feeding route on the cytokine response in a rat peritonitis model. The enteral and parenteral feeding groups received an identical standard hyperalimentation solution for seven days. Then, E. coli were injected intraperitoneally. The survival rate was better in the enteral group than in the parenteral group. The TPN group had higher bacterial colony counts in peritoneal lavaged fluid than the enteral group after the E. coli challenge. In addition, numbers of peritoneal exudative cells were significantly lower in the TPN than in the enteral group. Moreover, after the challenge, the enteral group had higher levels of TNF and IL-1 in peritoneal lavaged fluid. Significant positive correlations between numbers of peritoneal exudate cells and the amounts of peritoneal TNF and IL-1 were seen only in the TEN group. Furthermore, local IFN-! levels rose significantly after challenge only in the TEN group (Fig. 8.3). In our model, postchallenge IFN-! levels in BALF were also significantly higher in the TEN than in the TPN group. Thus, enteral nutrition increases inflammatory cell numbers and induces more favorable regulation of cytokine production at the local inflammatory site. These responses may result in more effective bacterial killing, and reduced systemic spread

88

Cytokines and the Abdominal Surgeon

Fig. 8.3. The effects of antecedent nutritional route on circulating blood and peritoneal lavaged fluid levels of IFN-!. Enteral and parenteral feeding groups received an identical standard hyperalimentation solution for seven days. Then, 3x108 E. coli were injected intraperitoneally. Modified from Lin MT, Saito H, Fukushima R et al. Route of nutritional supply influences local, systemic, and remote organ responses to intraperitoneal bacterial challenge. Ann Surg 1996; 223:84-93.

of bacteria. In addition, immune cells in remote organs can be mobilized to the inflammatory site more effectively. Immediate enteral nutrition decreases infectious complications, especially pneumonia and intra-abdominal abscess, in trauma patients.8,9,22 The results obtained with our peritonitis model may elucidate the mechanisms underlying the advantages of initiating enteral feeding immediately after surgical stress.

Specific Nutrients and Cytokine Biology Polyunsaturated Fatty Acids (PUFA) Linoleic acid (C18:2 n6) and alpha-linolenic acid (C18:3 n3) are essential fatty acids. These PUFAs have been shown to influence eicosanoid metabolism, cell membrane fluidity, and receptor orientation.23 The administration of PUFAs alters the immune and inflammatory responses to surgical insults.24 PUFAs exert their effects through diverse mechanisms. PUFAs, specifically n-3 PUFAs, have been shown to modulate the ability of immune cells to produce cytokines.25 One of the mechanisms whereby PUFAs may influence cytokine production is related to an alteration of the eicosanoid synthesis pathway from the prostaglandin (PG) 2 series to 3 series, and from the leukotriene (LT) 4 series to 5 series. In general, PGE2 generated from n6 fatty acids suppresses the production of TNF and IL-1.23 In contrast, LTB4 enhances the production of these cytokines. The balance between these eicosanoids may regulate cytokine production. Administration of dietary n-3 fatty acid downregulates mononuclear cell cytokine production in response to endotoxin.26-28 Peripheral mononuclear cells isolated from healthy volunteers receiving fish oil supplements released less TNF, IL-1 and/or IL-6 when stimulated with LPS in vitro than those isolated before n-3 supplementation.26,27 Kupffer cells isolated from rats fed n-3 rich fish oil produced less TNF when triggered by LPS than did those from rats fed corn oil, which is rich in n-6 fatty acids.29 In contrast, n-3 fatty acid-supplemented diets upregulate LPS-stimulated TNF production by peritoneal adherent macrophages.30 Thus, the effect of dietary fatty acid

Cytokines and Peri-Operative Nutrition

89

supplementation on the production of cytokines by immune cells is complex.31 The modulatory effects exerted by various fats depend upon the duration of feeding and the dietary fatty acid composition. Commercial fat emulsions used in TPN contain an abundance of n-6 PUFAs. The fat emulsions may change immune cell functions after surgical insults through altered cytokine biology. Long-term use of an LCT-based emulsion increases plasma TNF-∀ levels in humans.32 In contrast, TPN regimens containing no lipid, LCT, or MCT/LCT do not alter TNF, IL-1 or IL-6 levels in patients who have undergone gastrointestinal surgery.33 Future studies must be designed to elucidate the role of fatty acid supplementation in postoperative cytokine production.

Glutamine Glutamine has been regarded as a conditionally essential amino acid, during catabolic illness. Glutamine is important in several key metabolic processes of lymphocytes and macrophages.34 Glutamine breakdown may provide intermediates for the biosynthesis of ATP, nucleotides which are required for the synthesis of DNA and mRNA, and nitrogen for the formation of glucosamine, GTP and NAD+.34 Thus, glutamine plays an important role in maintaining the immune function of lymphocytes and macrophages. Exogenous glutamine augments functions of immune cells, such as lymphocytes, macrophages35,36 and neutrophils.37,38 In vitro, glutamine reportedly enhances LPS stimulated IL-1 production by mouse peritoneal macrophages,36 suggesting that glutamine availability is essential for the synthesis of IL-1. Glutamine also increases conconavalin A-stimulated production of IL-2 by rat lymphocytes.39 The ability of macrophages and lymphocytes to produce cytokines depends upon the glutamine concentration. In addition, glutamine increases anti-CD3 stimulated production of TNF and IFN by human peripheral blood mononuclear cells.40 However, in the “in vitro” study, the TNF production was observed in the absence of glutamine and reached a plateau at a glutamine concentration 0.1 mM. In contrast, IFN production requires exogenous glutamine and at least 0.5 mM glutamine is required for optimal IFN production. The different patterns of TNF and IFN-! production may be due to differences in the main cellular source of these cytokines: monocytes for TNF versus T lymphocytes for IFN.40 Neutrophils produce TNF-∀, IL-1 and IL-8.41 Neutrophil activation by these cytokines increases the production of reactive oxygen metabolites. We investigated the effect of glutamine on the in vitro E. coli-killing activity of PMNs and its relation with secreted cytokines in postoperative patients.38 Following culture with patient neutrophils the number of viable E. coli decreased significantly as the in vitro glutamine concentration was increased from 500 to 1000 nmol/ml. However, glutamine supplementation produced no appreciable changes in TNF-∀, IL-1b or IL-8 levels in cell culture supernatants after a 2 hour incubation. These observations suggest that the enhancement of neutrophil bactericidal function by glutamine may not be directly mediated by these cytokines. In an in vivo study, Barber et al42 demonstrated that a glutamine supplemented enteral diet had no effect on the systemic cytokine response (TNF, IL-6) following endotoxin challenge in rats. We also conducted an animal study which revealed that glutamine supplementation does not influence the levels of TNF-∀, IL-8 and IFN-! in either peritoneal lavage fluid or plasma in a rat protracted peritonitis model, while glutamine supplementation enhanced both peritoneal and hepatic bacterial clearance.43 Furthermore, our subsequent clinical study demonstrated that enteral

90

Cytokines and the Abdominal Surgeon

glutamine supplementation has no influence on levels of IL-8 and transforming growth factor- # at operative wound site in patients undergoing subtotal esophagectomy.44 Thus, there is no direct evidence that enteral glutamine supplementation enhances cytokine production in response to surgical stress. Glutamine-supplemented TPN, reportedly, had no effects on LPS stimulated TNF or IL-1 production by peripheral mononuclear cells from volunteers.45 O’Riordain et al46 showed T-cell DNA synthesis to be enhanced after 5 days of glutamine-supplemented TPN in patients undergoing colorectal surgery, while glutamine-supplementation had no effects on IL-2, TNF or IL-6 production by LPS-stimulated peripheral mononuclear cells. These studies indicate that glutamine-enriched TPN does not influence cytokine production by peripheral mononuclear cells under either stable conditions or surgical stress. In contrast, another animal study done in our laboratory revealed the beneficial effects of antecedent glutamine-enriched TPN on local, systemic, and splenic cytokine responses to intraperitoneal bacterial challenge.47 Rats received isocaloric and isonitrogenous alanyl-glutamine(Ala-Gln) enriched TPN or standard TPN for seven days, and were then challenged intraperitoneally with E. coli. The TNF levels in peritoneal lavaged fluid were significantly higher in the Ala-Gln-TPN than in the standard-TPN group after bacterial challenge. Hepatic TNF concentrations were significantly increased only in the Ala-Gln-TPN group. Moreover, glutamine supplementation greatly enhanced IFN-! responses. The splenic IFN-! level after challenge was significantly higher in the Ala-Gln-TPN than in the standard-TPN group. Antecedent Ala-Gln-TPN also enhanced spontaneous in vitro IFN-! production by splenocytes, consisting primarily of lymphocytes. IFN-! is a potent immunoenhancing cytokine.20,48 Thus, the enhanced splenic IFN-! response may exert protective effects against bacterial challenge in animals receiving Ala-Gln-TPN. Further clinical investigations are needed to clarify the effects of glutamine-TPN on IFN-! responses to surgery, trauma and sepsis.

Arginine L-arginine supplementation has beneficial effects on immune defense after surgical trauma.49,50 Peripheral blood lymphocyte responses and mitogenesis improved with L-arginine supplementation in surgical patients.51 In addition, dietary arginine supplementation increased the number of tumor infiltrating lymphocytes expressing CD 16 and CD 56 in patients with colorectal cancer.52 L-arginine’s mechanisms of action remain unclear. It has been suggested that the key role of arginine in host defense is the production and release of NO.53 L-arginine itself, or a product of arginine metabolism such as NO, may induce cytokine production.54 It is possible that the beneficial effects of L-arginine supplementation in surgical stress are associated with altered cytokine biology. L-arginine increased IL-2 receptor expression and IL-2 production by lymphocytes in both traumatized55 and tumor-bearing56 animals. In contrast, Brittenden et al57 showed that L-arginine supplementation stimulated lymphocyte mitogenesis and natural killer cell activity in patients with breast cancer, while no increases in serum levels of IL-1, IL-2, IFN-! and TNF were seen. The authors speculated that increases in the production of IL-2, as well as other cytokines, may occur in the early period after oral arginine intake, and that this would enhance natural cytotoxicity. Another mechanism by which L-arginine affects cytokine production may be via stimulating the release of GH and IGF-1, as described below.

Cytokines and Peri-Operative Nutrition

91

Nucleotides Dietary nucleotides may be important for maintaining immune function during protein starvation.58 Exogenous nucleotides have been shown to improve nonspecific as well as specific host defenses.58 Animals fed a nucleotide-deficient diet showed impaired production of IL-2, IL-1 and IFN-!, indicating that nucleotides are essential for cytokine production.59,60 A mononucleotide/nucleotide mixture given intraperitoneally to mice enhanced IL-4 and IL-5 production by splenic T helper cells, in response to mitogens, as compared with a nucleotide-free diet.61 Both cytokines are produced by T helper cells and are involved in B cell differentiation and antibody formation. However, IL-4 and IL-5 production by purified T helper cells was not altered by the addition of RNA to the culture medium.

Immune-Enhancing Diets Clinical studies have shown a beneficial effect on the incidence of infection and the duration of hospitalization in surgical patients receiving diets enriched with specific nutrients.62-67 Since each of these nutrients (n-3 fatty acids, arginine, glutamine and nucleotides) has immune enhancing effects mediated partly through cytokines, it is anticipated that the combinations of these dietary additives will also modulate cytokine biology and enhance immune responses in surgical stress. Moore et al65 compared an immune-enhancing diet enriched with glutamine, arginine, n-3 fatty acids and nucleotides with a standard stress enteral diet in the early postinjury period. The enriched diet significantly increased total lymphocyte, T-lymphocyte and T-helper cell numbers. In addition, the diet reduced the incidence of abdominal abscess and multiple organ failure. However, neither circulating cytokine levels, nor stimulation of IL-1, IL-2, IL-6, TNF and IFN-! production by peripheral mononuclear cells differed between the enriched and standard diets. In contrast, Senkal et al68 investigated the effects of an enteral diet enriched with arginine, RNA and n-3 fatty acids on cytokine production in whole blood cultures after abdominal operations. The enriched diet reduced the spontaneous expressions of TNF-∀ and IL-6, as compared with an isonitrogenous isocaloric placebo diet. Postoperative patients receiving the supplemented diet also experienced more rapid recovery in the expression of IL-1# and IL-2 receptors after phytohemagglutinin stimulation. The authors68 suggested that the altered cytokine responses may explain, at least in part, the improved clinical outcomes of surgical patients given the supplemented enteral nutrition.

Anabolic Hormones and Cytokine Biology Several recent studies have focused on the roles of growth hormone(GH) and insulin-like growth factor 1(IGF-1), in regulating the immune system along with the metabolic system.69-72 These anabolic hormones augment the morphology and function of lymphoid cells. It is noteworthy that GH increases the production of IL-1 and IFN-! by murine splenocytes stimulated with LPS.73 In contrast, a high concentration of GH decreased the productions of IL-1 and TNF in response to LPS by human blood mononuclear cells.74 Both GH and IGF-1 are powerful modulators of the functions of myeloid cells, such as polymorphonuclear cells and macrophages.69,71,75 These anabolic hormones prime macrophages for the production of cytokines including TNF, IL-1 and IL-6.76 We have studied the effects of GH and IGF-1 on the host cytokine response against bacterial infection.75 Mice were given normal saline, GH, or IGF-1 subcutaneously

92

Cytokines and the Abdominal Surgeon

Fig. 8.4. The effect of anabolic hormones on IL-1 and IL-6 production by peritoneal exudative cells in a mouse peritonitis model. Peritoneal exudative cells (PEC) harvested at 4 hrs after intraperitoneal challenge with 1 x 108 colony-forming units of E. coli, were incubated in vitro with 10 mg of lipopolysaccharide for 24 hours. Interleukin 1 (IL-1) and interleukin 6 (IL-6) levels in the PEC culture supernatants were measured. GH indicates growth hormone; IGF-I, insulin-like growth factor I. *p<0.05 vs control; #p<0.05 vs control and IGF-I. Mean ± SEM. (Modified from ref. 75).

for 6 days and then challenged with an intraperitoneal E. coli injection. The survival times of the GH and IGF-I groups were significantly longer than those of the control group. The numbers of bacteria in the peritoneal lavaged fluids, of both the GH and the IGF-I group, were lower than that in the control group after bacterial challenge. The LPS-stimulated production of IL-1 by peritoneal exudative cells was greater in GH and IGF-1 treated animals than in the control group (Fig. 8.4). IL-6 production by the exudative cells was also enhanced by GH. The results suggest that the enhancement of in vitro cytokine production by peritoneal exudative cells, as seen in both the GH and the IGF-I group, may improve host defense via these mechanisms. In contrast to enhanced in vitro cytokine production by peritoneal exudative cells, plasma TNF, IL-1 and IL-6 levels after challenge were significantly lower in the hormone-treated than in the nontreated control group. Elsasser et al77 also reported that recombinant bovine somatotropin blunted the plasma TNF response to endotoxin in vivo. Thus, both GH and IGF-1 appear to modulate cytokine responses by preventing the undesired toxic effects of excessive cytokines in the systemic circulation and by promoting cytokine activities at inflammatory sites.

Summary Nutritional routes, specific nutrients and anabolic hormones are all capable of modulating cytokine production, leading to a variety of cytokine responses to surgical insult. The modulation of cytokine production by these nutritional modalities may be beneficial for the prevention of complications and thereby improve the outcome of patients undergoing surgery. However, few clinical trials have focused on the effects of perioperative nutritional support on cytokine biology and its relation

Cytokines and Peri-Operative Nutrition

93

to clinical outcome. Moreover, the mechanisms by which nutritional therapy modulates cytokine production are still unclear. The regulation of cytokine synthesis may occur at the transcriptional or post-transcriptional level. A modulatory effect, exerted either directly or indirectly, by nutritional support can occur at either level. Further studies are required to determine how these nutritional modalities influence the molecular processes of cytokine biosynthesis.

References 1. Tracey KJ, Lowry SF. The role of cytokine mediators in septic shock. Advances in Surgery 1990; 23:21-56. 2. Souba WW. Cytokine control of nutrition and metabolism in critical illness. Current Problems in Surgery 1994; 31:577-643. 3. Saito H, Trocki O, Alexander JW et al. The effect of route of nutrient administration on the nutritional state, catabolic hormone secretion and gut mucosal integrity after burn injury. J Pareter Enteral Nutr 1987; 11:1-7. 4. Fletcher JP, Little JM. A comparison of parenteral nutrition and early postoperative enteral feeding on the nitrogen balance after major abdominal surgery. Surgery 1986; 100:21-24. 5. Fong Y, Marano MA, Barber A et al. Total parenteral nutrition and bowel rest modify the metabolic response to endotoxin in humans. Ann Surg 1989; 210:449-457. 6. Kudsk KA, Stone JM, Carpenter G et al. Enteral and parenteral nutrition influences mortality after hemoglobin-E. coli peritonitis in normal rats. J Trauma 1983; 23:605-609. 7. Lin MT, Saito H, Fukushima R et al. Route of nutritional supply influences local, systemic, and remote organ responses to intraperitoneal bacterial challenge. Ann Surg 1996; 223:84-93. 8. Kudsk KA, Croce MA, Fabian TC et al. Enteral versus parenteral Feeding. Effects on septic morbidity after blunt and penetrating abdominal trauma. Ann Surg 1992; 215:503-513. 9. Moore FA, Feliciano DV, Andrassy RJ et al. Early enteral feeding, compared with parenteral, reduces postoperative septic complications. The results of a meta-analysis. Ann Surg 1992; 216:172-183. 10. Lowry SF, Thompson WAI. Nutrient modification of inflammatory mediator production. New Horizons 1994; 2:164-74. 11. Santos AA, Rodrick ML, Jacobs DO et al. Does the route of feeding modify the inflammatory response? Ann Surg 1994; 220:155-63. 12. Lin M, Saito H, Fukushuma R et al. Preoperative total parenteral nutrition influences postoperative systemic cytokine responses after colorectal surgery. Nutrition 1997; 13:8-12. 13. Rock CS, Barber AE, Ng E-H et al. TPN vs. oral feeding: Bacterial translocation, cytokine response and mortality after Escherichia coli LPS administration. Surg Forum 1990; 41:14-16. 14. Marshall JC. The gut-liver axis in multiple organ failure. In: Lamy M, Thijs LG, eds. Mediators of sepsis. Berlin: Springer-Verlag, 1992:221-233. 15. Michalsky MP, Deitch EA, Ding J et al. Interleukin-6 and tumor necrosis factor production in an enterocyte cell model (Caco-2) during exposure to Escherichia coli. Shock 1997; 7:139-146. 16. Ogle CK, Zuo L, Mao JX et al. Differential expression of intestinal and splenic cytokines after parenteral nutrition. Arch Surg 1995; 130:1301-1307. 17. Nicaise P, Gleizes A, Forestier F et al. Influence of intestinal bacterial flora on cytokine (IL-1, IL-6 and TNF-∀) production by mouse peritoneal macrophages. European Cytokine Network 1993; 4:133-138.

94

Cytokines and the Abdominal Surgeon

18. Billiar TR, Maddaus MA, West MA et al. Intestinal gram-negative bacterial overgrowth in vivo augments the in vitro response of Kupffer cells to endotoxin. Ann Surg 1988; 208:532-540. 19. Shou J, Lappin J, Daly JM. Impairment of pulmonary macrophage function with total parenteral nutrition. Ann Surg 1994; 219:291-297. 20. Gennari R, Alexander JW, Eaves PT. IFN-! decreases translocation and improves survival following transfusion and thermal injury. J Surg Res 1994; 56:530-536. 21. Redmond HP, Leon P, Lieberman MD et al. Impaired macrophage function in severe protein-energy malnutrition. Arch Surg 1991; 126:192-196. 22. The Veterans Affairs total parenteral nutrition cooperative study group. Perioperative total parenteral nutrition in surgical patients. New Engl J Med 1991; 325:525-532. 23. Chavali SR, Forse RA. The role of w-3 polyunsaturated fatty acids on immune responses during infection and inflammation. In: Forse RA, ed. Diet, Nutrition, and Immunity. Boca Raton: CRC Press, 1994:179-186. 24. Alexander JW, Saito H, Ogle CA et al. The importance of lipid type in the diet after burn injury. Ann Surg 1986; 204:1-8. 25. Meydani SN, Dinarello CA. Influence of dietary fatty acids on cytokine production and its clinical implications. Nutrition in Clinical Practice 1993; 8:65-72. 26. Endres S, Ghorbani R, Kelley VE et al. The effect of dietary supplementation with n-3 polyunsaturated fatty acids on the synthesis of interleukin-1 and tumor necrosis factor by mononuclear cells. New Engl J Med 1989; 320:265-271. 27. Meydani SN, Endres S, WoodsM et al. Oral (n-3) fatty acid supplementation suppresses cytokine production and lymphocyte proliferation: Comparison between young and older women. J Nutr 1991; 121:547-555. 28. Yaqoob P, Calder P. Effects of dietary lipid manipulation upon inflammatory mediator production by murine macrophages. Cellular Immunology 1995; 163:120-128. 29. Billar TR, Bankey PE, Svingen BA al. Fatty acid intake and Kupffer cell function: Fish oil alters eicosanoid and monokaine production to endotoxin stimulation. Surgery 1988; 104:343-349. 30. Chang HR, Arsenijevic D, Pechere JC et al. Dietary supplementation with fish oil enhances in vivo synthesis of tumor necrosis factor. Immunol Lett 1992; 34:13-17. 31. Tappia PS, Grimble RF. Complex modulation of cytokine induction by endotoxin and tumour necrosis factor from peritoneal macrophages of rats by diets containing fats of different saturated, monounsaturated and polyunsaturated fatty acid composition. Clinical Science 1994; 87:173-178. 32. Gogos C, Zoumbos NC, Makri M. Tumor necrosis factor production by human mononuclear cells during total parenteral nutrition containing long-chain triglycerides. Nutrition 1992; 8:26-29 33. Arias-Diaz J, Rodriguez JM, Vara E et al. NO2/NO3 and cytokine plasma profiles under different postoperative parenteral nutrition regimens. Nutrition 1996; 12:89-92. 34. Calder PC. Glutamine and the immune system. Clinical Nutrition 1994; 13:2-8. 35. Parry-Billings M, Evans J, Calder PC et al. Does glutamine contribute to immunosuppression after major burns? Lancet 1990; 336(8714):523-5. 36. Wallace C, Keast D. Glutamine and macrophage function. Metabolism: Clinical & Experimental 1992; 41:1016-1020. 37. Ogle CK, Ogle JD, Mao JX et al. Effect of glutamine on phagocytosis and bacterial killing by normal and pediatric burn patient neutrophils. J Parent Ent Nutr 1994; 18:128-133. 38. Furukawa S, Saito H, Inaba T et al. Glutamine enhanced bacterial killing by neutrophils from postoperative patients. Nutrition 1997; in press. 39. Calder PC. Fuel utilization by cells of the immune system. Proc Nutr Society 1995; 54:65-82.

Cytokines and Peri-Operative Nutrition

95

40. Horig H, Spagnoli GC, Filgueira L et al. Exogenous glutamine requirement is confined to late events of T cell activation. J Cell Biochem 1993; 53:343-351. 41. Cassatella MA. The production of cytokines by polymorphonuclear neutrophils. Immunology Today 1995; 16:21-26. 42. Barber AE, Jones WGd, Minei JP, Harry M et al. Glutamine or fiber supplementation of a defined formula diet: impact on bacterial translocation, tissue composition, and response to endotoxin. J Parenter Enteral Nutr 1990; 14:335-343. 43. Furukawa S, Saito H, Inaba T et al. Glutamine-enriched enteral diet enhances bacterial clearance in protracted bacterial peirtonitis, regardless of glutamine form. J Parenter Enteral Nutr 1997; 21:208-214. 44. Fukatsu K, Hiramatsu T, Saito H et al. The effect of glutamine supplementation on interleukin-8 and transforming growth factor-# levels at operative wound sites. J Parenter Enteral Nutr 1997; 21:S13 (abstract). 45. Braxton CC, Coyle SM, van der Poll T et al. Influence of glutamine-supplemented TPN on in vitro and in vivo responses to endotoxin. Surg Forum 1995; 46:21-23. 46. O’Riordain MG, Fearon KC, Ross JA et al. Glutamine-supplemented total parenteral nutrition enhances T-lymphocyte response in surgical patients undergoing colorectal resection. Ann Surg 1994; 220:212-221. 47. Lin MT, Saito H, Hashiguchi Y et al. Alanyl-glutamine enriched total parenteral nutrition (TPN) improves splenic IFN-! response agaist intraperitoneal bacterial challenge. J Parenter Enteral Nutr 1997; 21:S5(abstract). 48. Ertel W, Morrison MH, Ayala A et al. Interferon-! attenuates hemorrhage-induced suppression of macrophage and splenocyte functions and decreases susceptibility to sepsis. Surgery 1992; 111:177-187. 49. Barbul A. Arginine: biochemistry, physiology, and therapeutic implications. J Parenter Enteral Nutr 1986; 10:227-238. 50. Saito H, Trocki O, Wang S et al. Metabolic and immune effects of dietary arginine supplementation after burn. Arch Surg 1987; 122:784-789. 51. Daly JM, Reynolds J, Thom A et al. Immune and metabolic effects of arginine in the surgical patient. Ann Surg 1988; 208:512-523. 52. Heys SD, Segar A, Payne S et al. Dietary supplementation with L-arginine: modulation of tumour-infiltrating lymphocytes in patients with colorectal cancer. Brit J Surg 1997; 84:238-241. 53. Gianotti L, Alexander JW, Pyles T et al. Arginine-supplemented diets improve survival in gut-derived sepsis and peritonitis by modulating bacterial clearance. The role of nitric oxide. Ann Surg 1993; 217:643-654. 54. Ketteler M, Border WA, Noble NA. Cytokines and L-arginine in renal injury and repair. Amer J Physiol 1994; 267(2 Pt 2):F197-207. 55. Cerra FB. Immune system modulation: Nutritional and pharmacological approaches. Crit Care Med 1990; 18:S85 56. Reynolds JV, Daly JM, Shou J et al. Immunologic effects of arginine supplementation in tumor bearing and non tumor bearing hosts. Ann Surg 1990; 211:202-209. 57. Brittenden J, Park KG, Heys SD et al. L-arginine stimulates host defenses in patients with breast cancer. Surgery 1994; 115:205-212. 58. Van Buren CT, Kulkarni AD, Rudolph FB. The role of nucleotides in adult nutrition. Journal of Nutrition 1994; 124(1 Suppl):160S-164S. 59. Chandra RK, Kumari S. Nutrition and immunity: An overview. Journal of Nutrition 1994; 124(8 Suppl):1433S-1435S. 60. Kulkarni AD, Rudolph FB, VanBuren CT. Nucleotide nutrition dependent immunosurveillance: Natural killer cell cytotoxity, !-interferon production, and polymorphonuclear cell function. In: Forse, RA, ed. Diet, nutrition, and immunity. Boca Raton: CRC Press, 1994:229-235.

96

Cytokines and the Abdominal Surgeon

61. Jyonouchi H, Sun S, Zhang-Shanbhag L et al. Polynucleotides compensate for impaired T-dependent antibody production induced in C57B1/6 mice by a nucleotidefree diet both in vivo and in vitro, but a mononucleotide-nucleoside mixture is effective only in vivo. Journal of Nutrition 1995; 125:1578-1586. 62. Gottschlich MM, Jenkins M, Warden GD et al. Differntial effects of three enteral dietary regimens on selected outcome variables in burn patients. J Parenter Enteral Nutr 1990; 14:225-236. 63. Daly JM, Lieberman MD, Goldfine J et al. Enteral nutrition with supplemental arginine, RNA, and omega-3 fatty acids in patients after operation: Immunologic, metabolic, and clinical outcome. Surgery 1992; 112:56-67. 64. Kudsk K, Minard G, Croce M et al. A rondomized trial of isonitrogenous enteral diets after severe trauma. An immune-enhancing diet reduces septic complications. Ann Surg 1996; 224:531-543. 65. Moore FA, Moore EE, Kudsk KA et al. Clinical benefits of an immune-enhancing diets for early postinjury enteral feeding. J Trauma 1994; 37:607-615. 66. Bower RH, Cerra FB, Bershadsky B et al. Early enteral nutrition of a formula (Impact) supplemented with arginine, nucleotides, and fish oil in intesnsive care unit patients: Results of a multicenter, prospective randomized clinical trial. Crit Care Med 1995; 23:436-449. 67. Daly JM, Weintraub FN, Shou J et al. Enteral nutrition during multimodality therapy in upper gastrointestinal cancer patients. Ann Surg 1995; 221:327-338. 68. Senkal M, Kemen M, Homann HH et al. Modulation of postoperative immune response by enteral nutrition with a diet enriched with arginine, RNA, and omega3 fatty acids in patients with upper gastrointestinal cancer. Eur J Sur 1995; 161:115-122. 69. Kelley KW. The role of growth hormone in modulation of the immune response. Ann N Y Acad Sci 1990; 594:95-103. 70. Gala RR. Prolactin and growth hormone in the regulation of the immune system. Proc Soc Exp Biol Med 1991; 198:513-527. 71. Wiedermann CJ, Reinisch N, Kahler C, Braunsteiner H. Regulation of myeloid phagocyte development and function by growth hormone: A review. J Pediatr Endocrinol 1993; 6:85-91. 72. Saito H, Inoue T, Fukatsu K et al. Growth hormone and the immune response to bacterial infection. Hormone Research 1996; 45:50-54. 73. Sommese L, Donnarumma G, de l’Ero C et al. Growth hormone modulates IL-∀ and IFN-! release by murine splenocytes activated by LPS or porins of Salmonella typhimurium. J Med Microbiol 1996; 45:40-47. 74. Kappel M, Hansen MB, Diamant M et al. In vitro effects of human growth hormone on the proliferative responses and cytokine production of blood mononuclear cells. Horm Metab Res 1994; 26:612-614. 75. Inoue T, Saito H, Fukushima R et al. Growth hormone and insulin-like growth factor I enhance host defense in a murine sepsis model. Arch Surg 1995; 130:1115-1122. 76. Edwards CK III, Lorence RM, Dunham DM et al. Hypophysectomy inhibits the synthesis of tumor necrosis factor ∀ by rat macrophages: Partial restoration by exogenous growth hormone or interferon g. Endocrinology 1991; 128:989-996. 77. Elsasser TH, Fayer R, Rumsey TS et al. Recombinant bovine somatotropin blunts plasma tumor necrosis factor-∀, cortisol, and thromboxane-B2 responses to endotoxin in vivo. Endocrinology 1994; 134:1082-1088.

Cytokine Response to Laparoscopic Surgery

97

CHAPTER 9

Cytokine Response to Laparoscopic Surgery Harry van Goor and R. Jan A. Goris

Introduction

S

ince the introduction of laparoscopic cholecystectomy in France in 1986 a new era of ‘minimal’ access general surgery has begun.1 Virtually every abdominal operation has thus far been done through a laparoscope.2-4 Laparoscopic cholecystectomy has become standard; the value of other procedures, such as laparoscopic Nissen fundoplication, inguinal hernia repair and colectomy is currently investigated.5,6 Laparoscopic surgery has many (potential) advantages above open surgery: less postoperative pain, less postoperative complications, shorter hospital stay, better cosmetic result, quicker convalescence and earlier return to work and social activities, and cost savings.7-9 The majority of these benefits have been related to reduced tissue trauma and a subsequently reduced postoperative stress response. 10-12 Proinflammatory cytokines play a pivotal role in this acute-phase response to trauma.13-15 In this review we examined the role of cytokines in laparoscopic surgery.

Rationale for Less Tissue Trauma and Stress Response in Laparoscopy A simple answer to the question ‘why is minimal invasive surgery less traumatic’ is that the wound is smaller. The wound is the primary source of pain, infection and psychological stress and it is probably true for many abdominal operations that the trauma of the surgical wound is greater than the trauma to the operative field. However, wounds are often not smaller in laparoscopic surgery, particularly not when large pieces of tissue have to be removed or extracorporeal anastomoses are created. More likely, the surgical trauma inflicted to gain and to hold access to the operative field is much greater in laparotomy compared with laparoscopy. Mechanical tissue retractors may cause greater pain than that of the incision alone and prolonged wound retraction enhances the risk of wound-infection and dehiscence, which are the source of considerable morbidity.16 It has been observed during laparoscopic cholecystectomy that stretching the fascia to remove a thick gallbldder results in more postoperative discomfort than a short extension of the incision.17 An additional traumatizing effect of manual dissection and laparotomy pads (gauzes) in the abdominal cavity during laparotomy is conceivable but is poorly Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

98

Cytokines and the Abdominal Surgeon

investigated.18 Glovepowder and woven gauzes cause intra-abdominal inflammation with release of cytokines from peritoneal macrophages and T-lymphocytes.18-20 Because open surgery is performed with large instruments, the amount of tissue incised with each move of the scissors or knife is large, causing a certain amount of bloodloss. Moves are smaller in laparoscopic surgery and the bloodloss must be less to maintain adequate visualisation of the operative field. The benefits of reduced bloodloss to the postoperative stress response are evident. Exposure of the abdominal cavity to the atmosphere in open surgery results in dehydration and cooling of exposed organs with subsequent inflammation, including edema, clotting disturbances, and organ dysfunction.16 Moreover, it has been suggested that exogenous endotoxin derived from air promotes bacterial translocation from the gut into the peritoneal cavity with subsequent release of inflammatory mediators.21 The increased pressure in the abdominal cavity, the often prolonged operative times and the continuous CO2 insufflation and flow might be potential disadvantages of laparoscopy in terms of stress response to tissue damage.22-25 Although in some reports operating time differences were held responsible for differences found in the acute-phase response, sound data are lacking.26-29 CO2 in the abdominal cavity may have beneficial instead of adverse effects on the inflammatory response. In vitro exposure of peritoneal macrophages to CO2 resulted in a profound but reversible inhibition of lipopolysaccharide-stimulated cytokine release.30

Cytokines and Tissue Trauma Within hours of surgical trauma a cascade of events is stimulated that mediate the inflammatory response. Activation of the complement system and of neutrophils are early responses to operation and may lead to the release of biologically potent mediators of inflammation, including complement split products C3a and C5a and elastase.11,29,31 Cytokines are synthesized and released by a variety of cells, including monocytes, macrophages, lymphocytes, epithelial and endothelial cells, fibroblasts and parenchymal cells of the gastrointestinal viscera, and orchestrate the body’s response to tissue trauma. These cytokines generate a wide range of local and systemic effects incorporating immune, metabolic, hemodynamic, central nervous and endocrine responses, in order to benefit the host.32,33 The main actions of cytokines after surgical trauma occur locally and influence a number of local processes like debridement of dead tissue, control of infection, wound healing and tissue remodelling.33-35 The local production of cytokines contributes to the systemic disturbances observed following surgery. These disturbances include fever, anorexia, increased O2 consumption, weight loss, fat and protein catabolism and hypoaminoacidemia.32,33,36,37 The increased synthesis and release of acute phase proteins by hepatocytes, including C-reactive protein (CRP), and coagulation (tissue factor, fibrinogen) and fibrinolytic (plasminogen activators and activator inhibitors) factors are other important manifestations of cytokine activation.32,33,38-40 It seems that release of cytokines and cytokine-induced responses are proportional to the severity of surgical trauma. Interleukin-6 (IL-6), an inducer of the acute phase response, whose secretion is stimulated by other cytokines such as IL-1 and tumor necrosis factor (TNF ), is considered the main early sensitive marker of tissue trauma.40-43 Also CRP correlates with the magnitude of surgical trauma.26,44 Therefore, especially IL-6 and CRP are evaluated in studies on the extent of surgical trauma, for example studies comparing laparoscopy with laparotomy.

Cytokine Response to Laparoscopic Surgery

99

Cytokines in Laparoscopic Cholecystectomy Comparison of cytokine releases between open and laparoscopic cholecystectomy has frequently been done.10,12,21,27,45-49 In general, open cholecystectomy induces a rise in circulating IL-6 within the first 3 hours (peak values 100 to 200 pg/ml), which persists for 12 to 24 after surgery, whereas in laparoscopic cholecystectomy circulating IL-6 is either undetectable or present in small amounts with significantly lower peak values (<50 pg/ml).12,49,51 Serum CRP levels show a similar pattern with peak values 12-24 hours later than those of IL-6.46,48,49 O’Dwyer did not find a difference in IL-6 and CRP levels between laparoscopicand mini-cholecystectomy.50 One would expect that the similar cytokine response was attributed to reduction of surgical trauma in the mini-cholecystectomy group. However, the authors explained their results by the longer operating time required for laparoscopic cholecystectomy and the constant peritoneal distension, probably because IL-6 levels in the laparoscopic group were relatively high (200 pg/ml). In concordance, several authors have reported a correlation between the magnitude of the metabolic response and the duration of surgery but it has never been identified as an independent risk factor.26,27,29 In a prospective study from our group, serum IL-6 did not increase both after laparoscopic and open cholecystectomy in patients under age 60.46 Only patients above 60 years of age, who underwent open cholecystectomy, had elevated IL-6 levels. CRP was significantly lower at 12 and 24 hours after laparoscopic cholecystectomy. Berggren et al could not demonstrate significant differences in circulating cytokines in patients below the age of 65, who underwent laparoscopic or open cholecystectomy.47 Increased operative bloodloss, which has been reported as an important factor enhancing cytokine response in major surgery,26 did not account for the elevated IL-6 serum lvels, since bloodloss was similar in elderly and younger patients, who underwent open cholecystectomy. It was hypothesized that older patients display a more intense inflammatory reaction, which is particularly apparent in relatively minor surgery such as cholecystectomy.46 On the other hand, younger patients might have a higher turnover of cytokines like IL-6. In most studies additional stress factors, including preceding endoscopic retrograde cholangio-panreatography (ERCP) and cholecystitis were not taken into account analysing circulating cytokines. The combination of ERCP followed by laparoscopic cholecystectomy dramatically increases the inflammatory response over laparoscopic cholecystectomy only.27 Similar results were obtained for complicated compared to uncomplicated open cholecystectomy.45 It is likely that the inflammatory response is already switched on and macrophages are in a primed state due to injury or infection, whereby the cholecystectomy acts as a “second hit.”51 Other cytokine measurements, such as IL-1, IL-2 and TNF, do not give unequivocal results in laparoscopic cholecystectomy. In some studies, significantly lower values of intraoperative and postoperative IL-1 beta and TNF have been observed.12,45,52 In the majority of studies serum levels of these cytokines remained under the detection limit after surgery and appeared to be unreliable indicators for postsurgical stress. The early and brief appearance in the circulation of these cytokines after injury and the random blood sampling in most studies may explain the absence of correlation with surgical stress.33

Cytokines in Other Laparoscopic Procedures There are scarce data on cytokine release in relation to other laparoscopic procedures than cholecystectomy. Harmon et al assessed intra- and postoperative changes

100

Cytokines and the Abdominal Surgeon

in serum IL-1 and IL-6 in the first 72 hours after elective laparoscopic or open colectomy.28 The IL-1 levels remained undetectable in virtually all patients irrespective of operative technique or time point of measurement. The IL-6 levels of the laparoscopic group were significantly lower than those of the open group between 3 and 24 hours after induction of anesthesia. There was no correlation between peak IL-6 concentrations and operative times or bloodloss. It is surprising that in such a major laparoscopic procedure reduction of tissue trauma is demonstrated in comparison with conventional colectomy. This reduction might not be attributed to the minimal access, as suggested for laparoscopic cholecystectomy, because in the majority of laparoscopic colectomies an incision is made to retrieve the specimen and to do an extracorporal anastomosis. IL-6 levels were increased intra- and postoperatively in a study comparing laparoscopic to abdominal hysterectomy.29 The only difference observed between the two patient groups was a significant increase in IL-6 at the time of the removal of the uterus in patients undergoing laparoscopic hysterectomy. This difference was explained by a longer operating time and by the extensive use of diathermia in the laparoscopic group. Schrenk et al studied metabolic responses after laparoscopic preperitoneal and open hernia (Shouldice) repair.53 No differences between these procedures were found in postoperative levels of IL-1, IL-6, CRP and TNF. There was only an increase of elastase in the laparoscopic group, which was attributed to the pneumoperitoneum. The absence of significant differences in serum cytokine levels might be due to the relatively low concentrations measured (<10 pg/ml IL-6) in both groups or the small number of samples (immediately after surgery and at the first and second postoperative day). From these low concentrations it appears that the metabolic response to open hernia repair is negligible. Therefore superiority of laparoscopic techniques may hardly be demonstrated in hernia repair.

Do Circulating Cytokines Accurately Reflect Tissue Trauma? It is questionable if measurements of circulating cytokines accurately reflect the inflammatory response to surgical trauma. In many experimental and clinical conditions, including wound healing, meningitis, rheumatoid arthritis and sarcoidosis, local tissue levels of cytokines are dramatically increased, whereas they are low or remain undetectable in the circulation.35,54-57 These cytokines may exert important local paracrine influences.33 Furthermore, there are cell-associated forms of cytokines, which function by cell-contact in the absence of any circulating form.58 Scott-Coombes et al pointed out that the cytokine response to abdominal surgery is mainly restricted to the abdominal cavity, and that increased plasma evels of cytokines may be considered as “spillover.”59 Recently, this compartmentalized peritoneal cytokine response has also been described for bacterial peritonitis, in which local response better reflects the severity of the disease and its prognosis than the systemic response.60 Thus, even though the majority of clinical studies involving cytokines in laparoscopy have sought to detect cytokines in the circulation, tissue concentrations of these proteins are more likely to reflect tissue trauma and to be biologically relevant than those in the circulation. Questioning the value of circulating cytokines recent studies focused on changes in cell-mediated and antibody-mediated immunity after laparoscopic surgery.61-64 Redmond and associates showed a significant increase in superoxide and TNF release by monocytes at day 1 and 3 after open cholecystectomy, demonstrating a more pronounced immunological depression than in laparoscopic cholecystectomy.61 Remarkably, activation of monocytes was present at time points, whereas serum TNF

Cytokine Response to Laparoscopic Surgery

101

levels were generally under the detection limit. A difference in monocyte activation, as determined by human leukocyte antigen (HLA)-DR expression, was not found between laparoscopic and open surgery; expression was suppressed in an equal manner.65 Surgical stress induces a shift in the type2/type1 T-helper cell balance (IL-4/interferon(IFN)-gamma and CD23/HLA-DR ratios) towards a Th2-type immune response.64,65 This shift seemed to be minor in laparoscopic procedures.64 It is, however, questionable if determination of this shift is more helpful in quantifying surgical stress than measuring circulating cortisol, androgens and growth hormone, which regulate the Th1 and Th2 activities. The lesser degree of down regulation of Th1 immune response (e.g., cell-mediated immunity) concords with the finding of a better preserved delayed hypersensitivity after laparoscopic colon resection.62,63

Cytokines and Outcome After Laparoscopic Surgery Reduced tissue trauma and subsequent diminished inflammatory response after laparoscopic surgery have been associated with less perioperative pain, a reduction in postoperative complications, a shorter hospitalization and a faster recovery. However, a relationship between cytokine levels and the outcome after laparoscopic surgery is not evident (Table 9.1). Reduction of perioperative pain and improved pulmonary function are common findings, whereas a decrease of circulating cytokines is only demonstrated in about half of the studies. In an attempt to determine the relationship between perioperative pain and/or analgesic requirements and cytokines several pitfalls may be encountered. First, epidural anesthesia reduces the stress response to operation, resulting in lower IL-6 levels.66 Second, the degree of insufflation pressure affects postoperative pain and demands for analgesics.22 Third, a lower placement of the ports, such as in the French technique of laparoscopic cholecystectomy, leads to less postoperative pain, upper abdominal in particular, in comparison with higher placement of ports such as in the American technique.67 Of great interest is the relation between cytokines and the development of postoperative complications, especially infections. An increased release of inflammatory mediators is associated with immunological alterations, which may render patients more susceptible for infection, and infection is an important cause of postoperative pain and prolonged hospital stay. There are no data on circulating cytokines and postoperative infections in laparoscopic compared to open surgery. Enhanced monocyte release of O2– and TNF in the open surgery group correlated with significantly higher postoperative septic complications in a randomized comparative study of patients undergoing laparoscopic or open cholecystectomy.61 IFN-gamma and IL-2 production by Th1-helper cells seemed to be less down-regulated in laparoscopic compared to open surgery.64 A better preserved IFN-gamma (a potential biological prophylactic antisepsis agent) and IL-2 production is associated with a reduction in wound infections.68 This beneficial effect may be explained by less down-regulation of HLA-DR molecules on monocytes after laparoscopic surgery. However, the level of monocyte HLA-DR expression did not differ between laparoscopic and open procedures and monocytes were similarly refractory to further stimulation by interferon (IFN)-gamma in vitro.65 Thus, other mechanisms are probably (also) involved. A reduction of the inflammatory response by laparoscopic surgery is of particular importance in the immuno-compromised, for example HIVpositive patients, who have a poor outcome in terms of mortality and postoperative infectious complications after surgery.69

*IL-6! CRP! IL-6= CRP=

IL-6= CRP= IL-6! CRP!

IL-6= IL-1= CRP= TNF∃= IL-6= CRP=

1992

1992

1992 1994

1995

Mealy11

Roumen46 Berggren47

O’Dwyer50 1995 Karayiannatis49 1997

1996

Joris10

Schrenk53**

Ellström29*** 168h

120h

168h 24h

48h <24h

48h

48h

48h

Duration of measurement

reduced only at day 0 –

– reduced first 12 hrs. reduced reduced

reduced at 48 hrs. reduced

reduced

Postoperative pain

AOS=

–

PEFR# AOS# –

VC# FEV1# PaO2# VC# FEV1# PaO2# – –

–

Pulmonary function

–

0

1 2

5 1

3

2

–

Decrease in hospital stay (days)

–

–

yes –

– yes

–

–

–

Earlier return to previous activities

*In patients over age 60. **Hernia repair. ***Hysterectomy. VC=vital capacity, FEV1=forced expiratory volume in 1 sec., PaO2=arterial partial pressure of oxygen, PaCO2=arterial partial pressure of carbon dioxide, PEFR=peak expiratory flow rate, AOS=average oxygen saturation

CRP!

IL-6! Il-1∀! IL-8 = IL-6! CRP!

1992

Glaser12

Circulating cytokines and CRP

Year

First author

Table 9.1. Circulating cytokines and postoperative pain, pulmonary function, hospital stay and recovery in laparoscopic compared to open cholecystectomy, hernia repair and hysterectomy.

102 Cytokines and the Abdominal Surgeon

Cytokine Response to Laparoscopic Surgery

103

Conclusions The literature reviewed supports the notion that laparoscopic surgery is associated with reduced tissue trauma and inflammatory response. This minor stress response is mainly reflected by lower plasmal evels of IL-6; studies of other circulating cytokines reveal equivocal results. Thus, the advantages of laparoscopic surgery in terms of reduced pain and faster recovery can not be explained by decreased circulating cytokines. The intraperitoneal cytokine release and local activity might better reflect tissue trauma and correlate with outcome after laparoscopic surgery. More experimental and clinical studies are required to assess the effect of laparoscopic surgery on intraperitoneal cytokines, and cell-mediated and antibody-mediated immunity, particularly with regard to the development of postoperative infections.

References 1. Muehe E. The first cholecystectomy through the laparoscope. Langenbecks Arch Chir 1986; 369:804. 2. Wexner SD, Cohen SM, Johansen OB et al. Laparoscopic colorectal surgery: a prospective assessment and current perspective. Br J Surg 1993; 80:1602-1605. 3. Watson DI, Devitt PG, Game PA. Laparoscopic Billroth II gastrectomy for early gastric cancer. Br J Surg 1995; 82:661-662. 4. Cuschieri A. Laparoscopic surgery of the pancreas. JR Coll Surg Edinb 1994; 39:178-84. 5. Pursnani KG, Sataloff DM, Zayas F et al. Evaluation of the antireflux mechanism following laparoscopic fundoplication. Br J Surg 1997; 84:1157-1161. 6. Liem MSL, Graaf Y van der, Zwart RC et al for the COALA trial group. A randomized comparison of physical performance following laparoscopic and open inguinal hernia repair. Br J Surg 1997; 84:64-67. 7. Reddick EJ, Olsen DO. Laparoscopic cholecystectomy: Passing fancy or legitimate treatment option? Gastroenterol 1990; 99:1527-1529. 8. Gadacz TR, Talamini MA. Traditional versus laparoscopic cholecystectomy. Am J Surg 1991; 161:336-338. 9. Vitale GC, Collet D, Larson GM et al. Interruption of professional and home activity after laparoscopic cholecystectomy among French and American patients. Am J Surg 1991; 161:396-398. 10. Joris J, Cigarini I, Legrand M et al. Metabolic and respiratory changes after cholecystectomy performed via laparotomy or laparoscopy. Br J of Anaesthesia 1992; 69:341-345. 11. Mealy K, Gallagher H, Barry M et al. Physiological and metabolic responses to open and laparoscopic cholecystectomy. Br J Surg 1992; 79:1061-1064. 12. Glaser F, Sannwald GA, Buhr HJ et al. General stress response to conventional and laparoscopic cholecystectomy. Ann Surg; 221:372-380. 13 Fong Y, Moldawer LL, Shires T et al. The biological characteristics of c ytokines and their implication in surgical injury. Surg Gynecol Obstet 1990; 170:363-378. 14. Pullicino EA, Carli F, Poole S et al. The relationship between the circulating concentrations of interleukin-6 (il-6), tumor necrosis factor (TNF) and the acute phase response to elective surgery and accidental injury. Lymphokine Res 1990; 9:231-238. 15. Sehgal PB. Interleukin-6: molecular pathophysiology. J Invest Dermatol 1990; 94:2S-6S. 16. Hunter JG, Sackier JM. Minimally invasive high tech surgery: Into the 21st century. In: Hunter JG, Sackier JM, eds. Minimally Invasive Surgery, McGraw Hill Inc., USA 1993:3-6.

104

Cytokines and the Abdominal Surgeon

17. Bordelon BM, Hobday KA, Hunter JG. Incision extension is the optimal method of difficult gallbladder extraction at laparoscopic cholecystectomy. Surg Endosc 1992; 6:225-227. 18. Duron JJ, Elian N, Olivier O. Postoperative peritoneal adhesions and foreign bodies. Eur J Surg 1997; 579(S):15-16. 19. Renz H, Gemsa D. Effects of powder on infection risks and associated mechanisms. Eur J Surg 1997; 579(S):35-38. 20. Ignatius JA, Hartmann WH. The glove starch peritonitis syndrome. Ann Surg 1972; 175:388-397. 21. Kobayashi E, Yoshida T, Yamauchi H et al. Immune function in patients undergoing open vs. laparoscopic cholecystectomy (Letter). Arch Surg 1995; 130:676. 22. Wallace DH, Serpell MG, Baxter JN et al. Randomized trial of different insufflation pressures for laparoscopic cholecystectomy. Br J Surg 1997; 84:455-458. 23. Demers Ph, Ratelle R, Boudreault D et al. Comparison of haemodynamic and respiratory effect during laparoscopic cholecystectomy using carbon dioxide peritoneum or laprolift. Ann Chir 1996; 50:593-600. 24. Cooper GM, Scottins AM, Ward ID et al. Laparoscopy—a stressful procedure. Anaesthesia 1982; 37:266-269. 25. Shenkin A, Fraser WD, Series J et al. The serum interleukin 6 response to elective surgery. Lymphokine Res 1989; 8:123-7. 26. Murata A, Ogawa M, Yasada T et al. Serum Interleukin-6, C-reactive protein and pancreatic secretory trypsin inhibitor (PSTI) as acute phase reactants after major thoraco-abdominal surgery. Immunol Invest 1990; 19:271-278. 27. Cho JM, LaPorta AJ, Clark JR et al. Response of serum cytokines in patients undergoing laparoscopic cholecystectomy. Surg Endosc 1994; 8:1380-1384. 28. Harmon GD, Senagore AJ, Kilbride MJ et al. Interleukin-6 response to laparoscopic and open colectomy. Dis Colon Rectum 1994; 37(8):754-759. 29. Ellström M, Bengtsson A, Tylman M et al. Evaluation of tissue trauma after laparoscopic and abdominal hysterectomy: Measurements of neutrophil activation and release of Interleukin-6, Cortisol and C-reactive protein. J Am Coll Surg 1996; 182:423-430. 30. West MA, Baker J, Bellingham BA. Kinetics of decreased LPS-stimulated cytokine release by macrophages exposed to CO2. J Surg Res 1996; 63:269-275. 31. Wanscher M, Knudsen F, Antonsen S et al. Release of granulocyte elastase and complement changes during hysterectomy. Scand J Clin Lab Invest 1989; 49:255-258. 32. Souba WW. Cytokine control of nutrition and metabolism in critical illness. Curr Probl Surg 1994; 31:577-652. 33. Fong Y, Lowry SF. Cytokines and the cellular response to injury and infection. In: Meakins JL, ed. Surgical Infections: Diagnosis and treat ment. New York: Scientific American Inc. 1994:65-86. 34. Frater-Schroder M, Risau W, Hallman R et al. Tumour necrosis factor type alpha, a potent inhibitor of endothelial cell growth in vitro, is angiogenic in vivo. Proc Natl Acad Sci 1987; 84:5277. 35. Ford HR, Hoffmann RA, Wing EJ et al. Characterization of wound cytokines in the sponge matrix model. Arch Surg 1989; 124:1422. 36. Mealy K, Robinson BG, Majzoub JA et al. Hypothalamic-pituitary-adrenal axis regulation by tumor necrosis factor. Wiley-Liss, New York, 1990:225-30. 37. Rothwell N. CRH is involved in the pyrogenic and thermogenic effects of interleukin1 beta in the rat. J Physiol 1989; 256:111-115. 38. Deventer SLH van, Bueller HR, Cate JW ten et al. Experimental endotoxemia in humans: analysis of cytokine release and coagulation, fibrinolysis and complement pathways. Blood 1990; 76:2520-2527.

Cytokine Response to Laparoscopic Surgery

105

39. Poll T van der, Bueller HR, Cate JW ten et al. Activation of coagulation after administration of tumor necrosis factor to normal subjects. N Engl J Med 1990; 322:1622-1627. 40. Levi M, Cate H ten, Bauer KA et al. Inhibition of endotoxin-induced activation of coagulation and fibrinolysis by pentoxiphylline or by a monoclonal antitissue factor antibody in chimpanzees. J Clin Invest 1994; 93:114-120. 41. Libert C, Brouckaert P, Shaw A et al. Induction of IL-6 by human and murine recombinant IL-1 in mice. Eur J Immunol 1990; 20:691-694 42. Heinrich PC, Castell JV, Andus T. Review article: Interleukin-6 and the acute phase response. Biochem J 1990; 265:621-636 43. Cruickshank AM, Fraser WD, Burns HJG et al. Response of serum interleukin-6 in patients undergoing elective surgery of varying severity. Clin Science 1990; 79:161-165. 44. Dominioni L, Dionigi R, Cividini F. Determinations of C-reactive protein and alpha-I-antitrypsin for quantitative assessment of surgical trauma. Eur J Surg Res 1980; Suppl 1:33. 45. Svoboda P, Kantorova I, Ochmann J et al. Serum levels of cytokines as a measure of response to stress after various types of elective gallbladder surgery. (Abs) Rozhl Chir 1994; 73(8):400-402 46. Roumen RMH, Meurs Van PA, Kuijpers JHC et al. Serum interleukin-6 and Creactive protein responses in patients after laparoscopic or convential cholecystectomy. Eur J Surg 1992; 158:541-544. 47. Berggren U, Gordh T, Grama D et al. Laparoscopic versus open cholecystectomy: hospitalization, sick leave, analgesia and trauma responses. Br J Surg 1994; 81:1362-1365. 48. McMahon AJ, O’Dwyer PJ, Cruikshank A et al. A comparison of the metabolic response to minilaparotomy and laparoscopic cholecystectomy. Br J Surg 1993; 80:1255-1258. 49. Karayiannakis AJ, Makri GG, Mantzioka A et al. Systemic stress response after laparoscopic or open cholecystectomy: A randomized trial. Br J Surg 1997; 84:467-471. 50. O’Dwyer PJ, McMahon AJ. Minicholecystectomy. In: Darzi A, Grace PA, Pitt HA, Bouchier-Hayes D, eds. Techniques in the management of gallstone disease. Oxford: Blackwell Science Ltd, 1995:84-89. 51. Waydhas C, Nast-Kolb D, Kick M et al. Posttraumatic inflammatory response, secondary operations, and late multiple organ failure [abstract]. J Trauma 1994; 74:165. 52. Yoshida T, Kobayashi E, Suminaga Y et al. Laparoscopic cholecystectomy minimally impairs postoperative cardiorespiratory and muscle performance [letter}. Br J Surg 1995; 82:996-997. 53. Schrenk P, Bettelheim P, Woisetschlaeger R et al. Metabolic responses after laparoscopic and open hernia repair. Surg Endosc 1996; 10:628-632. 54. Nelson S, Bagby GJ, Bainton BG et al. Compartmentalization of intraalveolar and systemic lipopolysaccharide-induced tumor necrosis factor and the pulmonary inflammatory response. J Infect Dis 1989; 159:189. 55. Helfgott DC, Tatter SB, Santhanam U et al. Multiple forms of IFN-beta 2/IL-6 in serum and body fluids during acute bacterial infections. J Immunol 1989; 142:948. 56. Hancock WW, Kobzik L, Colby AJ et al. Detection of lymphokines and lymphokine receptors in pulmonary sarcoidosis; immunologic evidence that inflammatory macrophages express IL-2 receptors. Am J Pathol 1986; 123:1. 57. Husby G, Williams RC Jr. Immunohistochemical studies of interleukin-2 and gamma-interferon in rheumatoid arthritis. Arthritis Rheum 1985; 28:174.

106

Cytokines and the Abdominal Surgeon

58. Okubo A, Sone S, Tanaka M et al. Membrane-associated interleukin-1 alpha as a mediator of tumor cell killing by human blood monocytes fixed with paraformaldehyde. Cancer Res 1989; 49:265. 59. Scott-Coombes DM, Whawell SA, Thompson JN. Peritoneal cytokine response in surgery. Br J Surg 1994; 81:756. 60. Schein M, Wittmann DH, Holzheimer R et al. Hypothesis: Compartmentalization of cytokines in intra-abdominal infection. Surgery 1996; 119:694-700. 61. Redmond HP, Watson RW, Houghton T et al. Immune function in patients undergoing open vs. laparoscopic cholecystectomy. Arch Surg 1994; 129:1240-1246. 62. Allendorf JD, Bessler M, Whelan RL et al. Better preservation of immune function after laparoscopic asssisted vs. open bowel resection in a murine model. Dis Colon Rectum 1996; 39:S67-72. 63. Bessler M, Whelan RL, Halverson A et al. Is immune function better preserved after laparoscopic versus open bowel resection? Surg Endosc 1994; 8:881-883. 64. Decker D, Schoendorf M, Bidlingmaier F et al. Surgical stress induces a shift in the type-1/type-2 T-helper cell balance, suggesting down-regulation of cell-mediated and up-regulation of antibody-mediated immunity commensurate to the trauma. Surgery 1996; 119:316-325. 65. Klava A, Windsor A, Boylston AW et al. Monocyte activation after open and laparoscopic surgery. 1997; 84:1152-1156. 66. Kehlet H. The stress response to surgery: Release mechanisms and the modifying effect of pain relief. Acta Chirurgica Scandinavica 1988; 550:S22-28. 67. Kum CK, Eypasch E, Aljarizi et al. Randomized comparison of pulmonary function after the ‘French’ and ‘American’ techniques of laparoscopic cholecystectomy. Br J Surg 1996; 83:938-941. 68. Wilmore DW. Homeostasis: bodily changes in trauma and surgery. In: Sabiston DC, ed. Textbook of surgery:the biological basis of modern surgical practice. 14th ed. Philadelphia: Saunders, 1991:19-33. 69. Tanner AG, Hartley JE, Darzi A et al. Laparoscopic surgery in patients with human immunodeficiency virus. Br J Surg 1994; 81:1647-1648.

Cytokines in Experimental Peritonitis

107

CHAPTER 10

Cytokines in Experimental Peritonitis Martin K. Angele, Alfred Ayala, and Irshad H. Chaudry

Introduction

D

espite the use of specific antibiotics, aggressive operative intervention, intravenous hyperalimentation, and improved monitoring, sepsis continues to be the major cause of death in most surgical intensive care units.1 Sepsis has been found to frequently cause pulmonary, renal, and subsequent multiple organ failure,2-4 thus making it necessary to investigate the relationship between sepsis and remote organ dysfunctions. However, the precise etiology contributing to the development of organ failure remains unclear. A number of studies have suggested that the link between cell and organ dysfunction, associated with multiple organ failure, lies in the initial increase of the proinflammatory cytokines such as interleukin-1 (IL-1), interleukin-6 (IL-6), and tumor necrosis factor (TNF) produced in response to septic insult.5-8 Support for this hypothesis comes from several studies which indicate that elevated levels of TNF are associated with increased mortality and infectious complications in septic patients.6,9,10 Furthermore, studies of patients with meningococcal septic shock demonstrated that serum TNF levels on admission were found to be a better prognostic parameter for outcome than leukocyte count, platelet count, or low blood pressure.11 In this regard, a fatal outcome in patients with meningococcal sepsis was associated with an increase in plasma IL-1, IL-6 and TNF levels.12 In addition, monocyte hyporesponsiveness to stimulation in vitro from critically ill patients was associated with increased mortality.13 Clinically, sepsis commonly presents itself as a two-phase process in which the patient initially demonstrates a hyperdynamic-hypermetabolic phase typified by increased cardiac output, metabolic rate and fever.14,15 Subsequently, the patient progresses into the hypodynamic-hypometabolic state characterized by a compromised circulatory system, depressed organ functions, and decreased tissue metabolism, which eventually culminates in death.14,15 Although patient studies continue to provide useful information, it is difficult to perform controlled studies in septic patients since there is a wide diversity of underlying diseases, infective organisms, injury severity, nutritional state, age, etc. In view of this, it is advantageous to utilize animal models which simulate the clinical condition in order to allow us to better define the pathophysiology of the septic process and the components/mechanisms responsible for the associated morbidity and mortality seen in septic patients. Currently, investigations into the process of sepsis-induced organ failure are being greatly Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

108

Cytokines and the Abdominal Surgeon

advanced with the development of transgenic mice. The use of transgenic mice, or mice in which a known genetic variation exists, allows for control of certain variables and may provide new mechanistic answers concerning sepsis and septic shock studies. A wide variety of models to simulate different aspects of sepsis and septic shock have been introduced and utilized. It is imperative, however, that the investigator choose the right model in order to obtain specific information concerning the hyperdynamic and/or hypodynamic circulatory state of sepsis.

Choice of Models Investigation into the pathophysiology of a process, such as sepsis and septic shock, requires the use of a well-characterized and reproducible model. Commonly used models for sepsis and septic shock produce either the hyperdynamic or hypodynamic phase of sepsis. However, few models are available which permit us to study the progressive changes from hyperdynamic to hypodynamic circulatory state of sepsis. The focus of our laboratory has been to develop and use a more clinically relevant model of sepsis, which permits us to examine the progressive changes during sepsis. Several investigators use small animal models, e.g., rodents, for experimental purposes because of the following advantages: rodents are inexpensive and easy to care for, widely available, can be obtained genetically identical, the same age and sex, on the same diet, and specific pathogen free. In this manner, biological variables are minimize. Nonetheless, there are certain limitations in using small animal models. For instance, some of the findings observed might be strain related. An example of this is the use of C3H/HeN versus C3H/HeJ endotoxin tolerant mice.16 In addition, most animal studies utilize healthy animals subjected to sepsis, whereas patients at risk for sepsis may have been traumatized or severely sick for some time before sepsis supervenes. The development of animal models for the study of sepsis began after Borden et al17 proposed that gram-negative shock was induced by endotoxins. Many models were thus developed that utilized a bolus injection of endotoxin (ETX) into laboratory animals, as well as in human volunteers. However, it was found that the clinical scenario that resulted from endotoxin injection differed greatly from that documented in gram-negative sepsis.18 Gram-negative sepsis in normovolemic patients is characterized by decreased peripheral resistance, increased cardiac output, hypotension, and oliguria.19 In contrast, hypotension following endotoxin injection is associated with increased peripheral resistance.20 Following the injection of a large dose of endotoxin (in mg/kg body weight range) in animals, blood pressure and cardiac output decreases rapidly, leading to death within a few hours.20 In light of these findings, investigators attempted to simulate the hypermetabolic state of clinical sepsis by introducing chronic models of endotoxemia.21 The model of chronic low-dose endotoxemia utilizes a prewarmed Alzet micro-osmotic pump (Model 1003D; Alza Corp., Palo Alto, CA) loaded with endotoxin (0.075 mg of lipopolysaccharide from Escherichia coli 055:B5 in 100 ml saline) which is implanted into the abdominal cavity. Although the dose of endotoxin utilized in such a model produces sustained systemic ETX levels, this chronic release model does not produce tachycardia, hyperinsulinemia or hyperglycemia variations commonly seen in patients during early sepsis. The intravenous administration of concentrated live E. coli organisms as a model of sepsis has been used by a number of investigators.22-26 Such a model, however, also does not simulate the situation of sepsis in patients, since a bolus injection of a massive bacterial load of Escherichia coli, pseudomonas aeriginosa etc. overwhelms the host.27 This was evidenced by the fact that infusion of live E. coli into guinea pigs was

Cytokines in Experimental Peritonitis

109

found to elicit either a hyperdynamic or hypodynamic circulatory state, depending on the route and dose of bacteria administration.28 Other sepsis models include implantation of gelatin capsules containing barium sulfate and rat feces into the peritoneal cavity of rats.29 This model is lethal within 3 days in most animals, and the autopsy of those animals revealed a generalized peritonitis. However, the severity of this model varies with changes in the bacterial content due to dietary manipulation prior to the capsule implantation. For instance, if the rats are fed a meat diet instead of the usual grain diet, the mortality in this model is significantly decreased. Ryan et al introduced a small animal model of cecal ligation to produce peritonitis.30 However, studies in our laboratory have shown that cecal ligation alone led to localized intraabdomineal abscess which the rats tolerated quite well, and none of the animals died when followed for periods up to one month.22 Furthermore, blood cultures taken from such animals were found to be negative.22 In view of the fact that cecal ligation alone was unable to produce septic shock, we have utilized a simple and reproducible model of sepsis in rats and mice which requires cecal ligation followed by needle puncture of the cecum (CLP).5,31 In brief, the procedure in mice is as follows: isolation and ligation of the cecum with a 5-0 suture, followed by puncturing the ligated cecum twice with a 22 gauge needle. A small amount of cecal contents is then extruded through the puncture holes, and the cecum is returned to the abdominal cavity. The midline incision is closed in layers with nonabsorbable suture, and the animals are resuscitated with 1 ml of Ringer’s lactate subcutaneously. In rats, an 18 gauge needle is used and the animals are resuscitated with 3 ml/100g BW of Ringer’s lactate subcutaneously. Blood cultures taken from experimental animals following cecal ligation and puncture were found to be positive for gram-positive (e.g., Streptococcus bovis) and gram-negative bacteria (e.g., Bacteroides fragilis, E. coli, Klebsiella, Proteus mirabilis) as soon as 1 hour following CLP.22,31 Furthermore, CLP produces an early hyperdynamic, hypermetabolic state, followed by a hypodynamic, hypometabolic state and thus mimics the characteristics of sepsis seen in humans.32 Moreover, the severity of the model can be altered by increasing or decreasing the size of the needle puncture by using 18 gauge versus 22 gauge. Using an 18 gauge instead of a 22 gauge needle for puncturing the ligated cecum resulted in an increased mortality rate from the septic challenge.31 An additional advantage of this model is that it allows one to subsequently excise the necrotic cecum and irrigate the peritoneal cavity. Thus, such a model closely resembles perforated appendicitis. Recent studies of Wilson et al demonstrated that higher volume of resuscitation following CLP were associated with a decreased mortality.33 Thus, the issue of fluid resuscitation should be taken into consideration in the use of experimental sepsis models. Since many of the pathophysiological responses observed following sepsis in patients, as well as in the animal models mentioned above, appear to be mediated by cytokines, the response of cytokines in experimental models of endotoxemia, bacteremia, and sepsis and in septic patients are described below.

The Response of Cytokines in Experimental Models of Sepsis and Clinical Sepsis Proinflammatory Cytokines Proinflammatory cytokines are proteinaceous mediators produced by inflammatory cells. The function of proinflammatory cytokines is to communicate and interact between somatic tissues in the presence of an inflammatory stimulus. In

110

Cytokines and the Abdominal Surgeon

contrast to mediators derived from the classical endocrine system, cytokines are paracrine agents which act locally in the variety of tissues where they are produced. Two proinflammatory cytokines, TNF-! and IL-1, appear to play predominant roles in the normal inflammatory response. These cytokines, besides being synonymous with a worsening clinical scenario, also exert beneficial effects in the normal physiologic inflammatory response. They increase neutrophil margination and activate the antimicrobial activity of monocytes, macrophages, neutrophils, and eosinophils.34-37 On the other hand, increased TNF-! and IL-1 production are responsible for the acute phase response characterized by fever and anorexia. Moreover, there is evidence that when excessive acute concentrations or chronic long-term increased plasma levels of proinflammatory cytokines occur, the adverse physiological effects of the cytokines are predominant. Several studies indicate that there is an association between increased infectious complications, mortality rates and elevated plasma levels of proinflammatory cytokines, in particular TNF-!.5,6,9-12,38 Calandra et al reported a progressive decline in TNF levels in survivors of septic shock, whereas TNF levels remained persistently elevated after initial diagnosis and attempted treatment in nonsurvivors from septic shock.39 Further support for a potentially toxic effect of TNF comes from a study of Tracey et al who demonstrated lethality of animals following an infusion of 100 mg/kg TNF.40 Similarly, IL-1 injection resulted in a shocklike state in baboons which was characterized by tachycardia, a dramatic decrease in mean arterial pressure, lactic acidemia, and initiation of a proinflammatory cascade.41

Plasma Cytokine Levels Following Endotoxemia and Bacteremia Serum profiles of TNF, IL-1 and IL-6 following endotoxin administration have been well described in both animal experiments,42,43 and patient studies.44,45 The results demonstrated an early, but transient, elevation in TNF followed by increased IL-1∀ and IL-6 levels. Additional studies have examined the effect of TNF-! on mortality by inhibiting TNF’s activity with a specific antibody. Studies from Remick et al indicate that following the onset of sepsis by CLP, administration of TNF antibodies failed to prevent lethality.46 In contrast, in a model of lethal endotoxemia, TNF-! appears to play a key role,47,48 since anti-TNF antibodies attenuated the lethal effect of endotoxemia and prevented the development of lethal shock (in rodent models).49-51 In a model of Escherichia coli injection in baboons, however, lethality was only decreased when the administration of TNF antibodies was initiated 2 hr before the injection of endotoxin.52 Simultaneous injection of TNF antibodies and Escherichia coli had no beneficial effect on the outcome of the animals.52 Evidence that the release of TNF might initiate the release of the other proinflammatory cytokines comes from a study which showed that administration of anti-TNF antibodies down-regulated the production of IL-1 and IL-6 release.43 Shalaby et al42 further suggested that besides endotoxin and TNF-!, IL-1 is able to up-regulate IL-6. TNF-! and IL-1 are known to also stimulate the production of other cytokines that perpetuate the sepsis cascade. Particularly important is the local release of IL-8, which recruits and activates neutrophils, resulting in tissue damage and organ dysfunction.53 In addition to the release of inflammatory cytokines, the sepsis cascade involves the production and release of specific cytokines with predominantly anti-inflammatory activities, e.g., IL-4 and IL-10. Those cytokines appear to decrease the synthesis of IL-1 and TNF in septic patients.53 These above mentioned studies, however, were performed using a bolus injection of endotoxin or bacteria. These models of sepsis, however, simulate only the hyperdynamic or hypodynamic state of sepsis. Thus, these models allow studies of the cellular and subcellu-

Cytokines in Experimental Peritonitis

111

lar alterations during either the hyperdynamic or hypodynamic state of sepsis and not the progressive changes from hyperdynamic to hypodynamic state. In view of this, our laboratory has been interested in investigating the progressive changes from hyperdynamic to hypodynamic circulation using the CLP model.

Plasma Cytokine Levels Following Polymicrobial Sepsis The early response to sepsis is characterized by increased systemic release of macrophage/monocyte derived mediators, including proinflammatory cytokines, prostanoids, platelet-activating factor, complement, etc. This response to the bacterial stimulants has been suggested to eventually lead to a cascade of events which result in cellular as well as organ dysfunction.54 In this regard, we and others have detected the early circulating levels of proinflammatory cytokines, such as TNF, IL-1 and IL-6 in septic models.5,55-57 Polymicrobial sepsis as produced by cecal ligation and puncture induced a persistent elevation of plasma TNF-!, which lasted for approximately 10 hours, followed by steady increase in plasma IL-1 levels, and enhanced plasma IL-6 levels at all time points compared to sham animals.57 Hadjiminas et al have documented that enhanced plasma TNF levels were reflected by an increase in transcriptional activity of this gene early after CLP.58 Within 3 hours following CLP, a rapid increase in mRNA of both TNF-! and IL-1∀ is induced in the liver and in the intestine.33 However, the intensity of the proinflammatory response may have been partially suppressed by administration of fluid for resuscitation after surgery in the model of CLP. In this respect, Wilson et al found that 1.0 ml as compared to 0 ml of fluid resuscitation following CLP decreased the IL-1∀ mRNA expression in the liver and small intestine33 indicating that fluid resuscitation attenuates the early cytokine mRNA expression after peritonitis in this model of polymicrobial sepsis.

Release of Proinflammatory Cytokines by Different Macrophages Macrophage cytokine release is differentially affected during the early hyperdynamic/hypermetabolic state of sepsis dependent on the tissue macrophage population examined. In peritoneal macrophages harvested 1 hr after CLP, an increase in the release of IL-1, IL-6, and TNF-! in the absence of added LPS occurred.5,56 However, only the release of IL-1 from peritoneal macrophages remains elevated beyond 1 hr post-CLP.5 Nonetheless, the capacity of peritoneal macrophages to release IL-1, IL-6, and TNF-! in response to an in vitro stimulus such as LPS is significantly decreased after the induction of sepsis as compared to macrophages from sham animals.59 At one hour after CLP, a transient increase in the release of IL-6 by unstimulated Kupffer cells has been found, as compared to sham operated animals.16 In contrast to peritoneal macrophages, IL-1 and TNF-! release from unstimulated Kupffer cells remained unchanged in post CLP animals.16 Similar to the peritoneal macrophages, however, the capacity of Kupffer cells to release proinflammatory cytokines in response to LPS in vitro was significantly decreased following CLP.16 In contrast to peritoneal macrophages, splenic macrophages do not show enhanced IL-1 and IL-6 release following the induction of sepsis. However, in response to LPS, a significantly decreased capacity of splenic macrophages to release cytokines in response to LPS is evident. Alveolar macrophages do not release enhanced amounts of cytokines following CLP in the absence of added LPS.56 However, the addition of LPS significantly enhanced the release of proinflammatory cytokines as compared to unstimulated alveolar macrophages. While there is a trend towards suppressed production in the cells taken from CLP animals, this was not significantly different from sham controls.56

112

Cytokines and the Abdominal Surgeon

In summary, the results indicate that late sepsis is characterized by a dysfunction of all macrophage populations, as evidenced by decreased capacity to release proinflammatory cytokines in response to LPS.5,16,56,59 This immune dysfunction might contribute to the observed mortality following septic challenge. Moreover, there are interactions between macrophages and T cells. Our laboratory has demonstrated that increased levels of IL-6 enhance the release of transforming growth factor-∀ (TGF-∀) 24 hr following the onset of sepsis, which is well beyond the early elevation of systemic proinflammatory cytokines.60 Since in vitro TGF-∀ decreases IL-2 release, elevated TGF-∀ plasma levels might contribute to the depression of splenocyte IL-2 release observed after the induction of sepsis as described in detail below.

Splenocyte Function Following the Induction of Sepsis A progressive depression in splenocyte immune functions occurs following the induction of sepsis by CLP.54 Splenocytes harvested 24 hr post-CLP show a significantly depressed IL-2, IL-3 and interferon-# (IFN-#) release as well as splenocyte proliferation in response to mitogenic stimulation.61 IL-2, IL-3 and IFN-# are considered to be the key mediators in up-regulating cell mediated immunity. The upregulation of anti-inflammatory cytokines could contribute to the decreased splenic IL-2 and IL-3 release under these conditions. In this regard, significantly higher levels of IL-4 and IL-10, important anti-inflammatory cytokines, are released by splenocytes following the induction of sepsis.61 Furthermore, administration of recombinant human IL-10 prior to endotoxin injection in baboons significantly decreased the LPS induced cytokine response, as evidenced by decreased TNF-!, IL-6, IL-8, and IL-12 levels. 62 It should be noted, however, that inhibition of proinflammatory cytokine production during bacterial sepsis may be associated with an impaired clearance of bacteria leading to decreased survival rates, as it has been demonstrated in mouse models of pneumonia.63-65 The use of transgenic mice (C3H/HEJ) tolerant to endotoxin has demonstrated that endotoxin may not be required for initiating the ensuing cytokine chain reaction. Peritoneal macrophages obtained after cecal ligation and puncture from endotoxin tolerant mice exhibit spontaneous release of TNF-!, IL-1, and IL-6,5 suggesting that macrophages can be primed to release cytokines by other nonendotoxin mechanisms.5 Furthermore, studies from our laboratory indicate that chronic exposure to low plasma endotoxin levels comparable to those in CLP mice is not an adequate stimulus to account for the suppression of lymphokine secretion following sepsis.61 We therefore speculate that some other agent/factor or component(s) of bacteria, acting alone or synergistically with endotoxin, may contribute to the culmination of events leading to the cellular immune dysfunction seen following septic challenge. Almost all of the animal studies mentioned above were performed in male animals, although it is known that there are differences in both humoral and cell mediated immune responses between males and females. Studies that examined immune functions in female rodents following the induction of sepsis by CLP were conducted by Zellweger et al.66 The results indicate that females in the proestrus state of the estrus cycle maintain splenic immune functions 24 hr following CLP (as evidenced by the absence of a marked decrease in splenocyte proliferation, IL-2, and IL-3 release in female mice) as compared to depressed responses in males.66 Furthermore, this immunologically better positioned immune response in females was associated with a significantly higher survival rate of female proestrus mice as compared to male mice when subjected to sepsis.66 The lower plasma testosterone and/or the higher

Cytokines in Experimental Peritonitis

113

plasma estrogen and prolactin levels in proestrus female animals appear to be responsible for the better maintained immune functions in females as compared to males. Thus, attempts to mimic female plasma hormone levels in males by administration of estrogen or prolactin and/or blocking testosterone receptors might prevent the dysfunction of the immune system in male septic patients.

Summary Although various experimental models of endotoxemia, bacteremia, septicemia and peritonitis have been described and each has their advantages and limitations, the choice of model should depend on the hypothesis to be tested. Furthermore, depending on the sepsis-peritonitis model and the sex of the animals, different cytokine profiles following the onset of sepsis are observed. In view of this, results obtained in animal models of peritonitis have to be evaluated carefully for extrapolation to the clinical situation. A model which progresses from a hypermetabolic/ hyperdynamic circulatory state to a hypodynamic/hypometabolic state, such as our CLP model, would appear to be appropriate for further defining the detailed pathophysiological mechanism involved in clinical sepsis. Investigations concerning the function of the various cytokines and the control of their release has lead to a better understanding of the intracellular response to infection. Such information appears to be important for the clinician who encounters a septic patient in the intensive care unit. Knowledge of the patient’s cytokine levels may give the clinician some indicator of the intracellular milieu and possibly insight into cellular changes taking place, allowing him/her a better understanding of how to treat such a critically ill patient. More refinements towards the rapid and perhaps also the online measurements of cytokines are, however, needed before the full benefits of such measurements are effectively translated to better management of septic patients. Although various cytokine therapies in septic patients so far have not yielded satisfactory results, the lack of beneficial effects might be related to the timing and dose of anticytokine administration. It is our hypothesis that total blockade/neutralization of cytokines will not be helpful to the host. Instead, modulation of cytokine production/release leading to the restoration of cellular homeostasis might be a better approach under those conditions. Since sepsis continues to be the major cause of morbidity and mortality in intensive care units, it is important to further investigate the pathophysiology of sepsisperitonitis so that the underlying mechanism responsible for organ dysfunction, multiple organ failure, and late deaths can be further defined. Controlled studies, using well-characterized animal models which simulate a clinical scenario, continue to be important for further defining the pathophysiological changes of sepsis. A more detailed understanding of the pharmacology of the proinflammatory cytokines cascade and the physiologic effects, in particular of TNF-! and IL-1, will allow a more comprehensive approach to control their detrimental as well as salutary effects. Although these treatments hold much promise for the future management of severely traumatized and septic patients, careful evaluation of both the benefits and complications of therapy is needed before widespread clinical use can be envisioned.

References 1. Marshall J, Sweeney D. Microbial infection and the septic response in critical surgical illness. Arch Surg 1990; 125:17-23. 2. Baue AE. Multiple organ failure. In: Baue AE, ed. Multiple Organ Failure: Patient Care and Prevention, St. Louis, MO: Mosby Year Book, 1990:421-470.

114

Cytokines and the Abdominal Surgeon

3. Deitch EA. Multiple organ failure: Summary and overview. In: Deitch EA, ed. Multiple Organ Failure, New York: Thieme Medical Publishers, 1990:285-299. 4. Faist E, Baue AE, Dittmer H. Multiple organ failure in polytrauma patients. J Trauma 1983; 23:775-787. 5. Ayala A, Kisala JM, Felt JA, Perrin MM, Chaudry IH. Does endotoxin tolerance prevent the release of inflammatory monokines (IL-1, IL-6, or TNF) during sepsis? Arch Surg 1992; 127:191-197. 6. Marks JD, Marks CB, Luce JM et al. Plasma tumor necrosis factor in patients with septic shock. Am Rev Respir Dis 1990; 141:94-97. 7. Starnes HF, Pearce MK, Tewari A et al. Anti-IL-6 monoclonal antibodies protect against lethal Escherichia coli infection and lethal tumor necrosis factor-a challenge in mice. J Immunol 1990; 145:4185-4191. 8. Hack CE, DeGroot ER, Felt-Bersma RJF et al. Increased plasma levels of interleukin6 in sepsis. Blood 1989; 74:1704-1714. 9. Damas P, Reuter A, Gysen P et al. Tumor necrosis factor and interleukin-1 serum levels during severe sepsis in humans. Crit Care Med 1989; 17:975-978. 10. Marano MA, Fong Y, Moldawer LL et al. Serum cachectin/tumor necrosis factor in critically ill patients with burns correlates with infection and mortality. Surg Gynecol Obstet 1990; 170:32-38. 11. Waage A, Halstensen A, Espevik T. Association between tumor necrosis factor in serum and fatal outcome in patients with meningococcal disease. Lancet 1987; i:355-357. 12. Waage A, Brandtzaeg P, Halstensen A, Kierulf P, Espevik T. The complex pattern of cytokines in serum from patients with meningococcal septic shock. Association between interleukin 6, interleukin 1, and fatal outcome. J Exp Med 1989; 169:333-338. 13. Rogy MA, Oldenburg SA, Coyle R et al. Correlation between acute physiology and chronic health evaluation (APACHE) III score and immunological parameters in critacally ill patients with sepsis. British J Surgery 1996; 83:396-400. 14. Cerra FB. Multiple organ failure syndrome. In: Bihari DJ, Cerra FB, eds. Multiple Organ Failure, Fullerton, CA: Society of Critical Care Med. 1989:1-24. 15. Groeneveld AB, Bronsveld W, Thijs LG. Hemodynamic determinants of mortality in human septic shock. Surgery 1986; 99:140-153. 16. Ayala A, Perrin MM, Wang P, Chaudry IH. Sepsis induces an early increased spontaneous release of the hepatocellular stimulatory factor (IL-6) by Kupffer cells in both endotoxin tolerant or intolerant mice. J Surg Res 1992; 52:635-641. 17. Borden CW, Hall WH. Fatal transfusion reactions from massive bacterial contamination of blood. N Engl J Med 1951; 245:760-765. 18. Waisbren BA. Gram-negative shock and endotoxin shock. Am J Med 1964; 36:819-824. 19. MacLean LD, Mulligan WG, McLean APH, Duff JH. Patterns of septic shock in man—A detailed study of 56 patients. Ann Surg 1967; 166:543-562. 20. Swan KG, Jacobson ED. Hemodynamics of endotoxin shock in the conscious animal. Surg Gynecol Obstet 1967; 125:1041-1046. 21. Fish RE, Spitzer JA. Continuous infusion of endotoxin from an osmotic pump in the conscious, unrestrained rat: A unique model of chronic endotoxemia. Circ Shock 1984; 12:135-149. 22. Wichterman KA, Baue AE, Chaudry IH. Sepsis and septic shock—A review of laboratory models and a proposal. J Surg Res 1980; 29:189-201. 23. Fink MP, Heard SO. Laboratory models of sepsis and septic shock. J Surg Res 1990; 49:186-196. 24. Jones SB, Westfall MV, Sayeed MM. Plasma catecholamines during E. coli bacteremia in conscious rats. Am J Physiol 1988; 254:R470-R477.

Cytokines in Experimental Peritonitis

115

25. Archer LT. Hypoglycemia in conscious dogs in liver E. coli septicemia: A chronic study. Circ Shock 1976; 3:93-97. 26. Cryer PE, Herman CM, Sode J. Carbohydrate metabolism in the baboon subjected to gram-negative (E. coli) septicemia. Ann Surg 1971; 174:91-100. 27. Chaudry IH, Ayala A, Singh G, Wang P, Hauptman JG. Rodent models of endotoxemia and sepsis. In: Schlag G, Redl H, eds. Pathophysiology of Shock, Sepsis and Organ Failure. Berlin: Springer-Verlag. 1993:1048-1059. 28. Durkot MJ, Wolfe RR. Hyper and hypodynamic models of sepsis in guinea pigs. J Surg Res 1989; 46:118-122. 29. Onderdonk AB, Weinstein WM, Sullivan NM, Gorbach SL. Experimental intraabdominal abscesses in rats: Quantitative bacteriology of infected animals. Infect Immun 1974; 10:1256-1259. 30. Ryan NT, Blackburn GL, Clowes GHA. Differential tissue sensitivity to elevated endogenous insulin levels during experimental peritonitis in rats. Metabolism 1974; 23:1081-1089. 31. Baker CC, Chaudry IH, Gaines HO, Baue AE. Evaluation of factors affecting mortality rate after sepsis in murine cecal ligation and puncture model. Surgery 1983; 94:331-335. 32. Chaudry IH, Wichterman KA, Baue AE. Effect of sepsis on tissue adenine nucleotide levels. Surgery 1979; 85:205-211. 33. Wilson MA, Chou MC, Spain DA et al. Fluid resuscitation attenuates early cytokine mRNA expression after peritonitis. J Trauma 1996; 41:622-627. 34. Keogh C, Fong Y, Marano MA et al. Identification of a novel tumor necrosis factor a/cachectin from the livers of burned and infected rats. Arch Surg 1990; 125:79-85. 35. Chen W, Havell EA, Moldawer LL et al. Interleukin 1: An important mediator of host resistance against Pneumocystis carinii. J Exp Med 1992; 176:713-718. 36. Kohno H, Thompson RC, Tucker K, Russell D. Tumor necrosis factor binding protein inhibits rodent streptococcal cell wall induced arthritis. Lymphokine Cytokine Res 1993; 12:376. 37. Dinarello CA. Interleukin-1 and interleukin-1 antagonism. Blood 1991; 77:1627-1652. 38. Feldbush TL, Hobbs MV, Severson CD, Ballas ZF, Weiler JM. Role of complement in the immune response. Fed Proc 1984; 43:2548-2552. 39. Molloy RG, Mannick JA, Rodrick ML. Cytokines, sepsis and immunomodulation. Br J Surg 1993; 80:289-297. 40. Tracey KJ, Lowry SF, Fahey TJI et al. Cachectin/tumor necrosis factor induces lethal shock and stress hormone responses in the dog. Surg Gynecol Obstet 1987; 164:415-422. 41. Lejeune P, Lagadec P, Onier N et al. Nitric oxide involvement in tumor-induced immunosuppression. J Immunol 1994; 152:5077-5083. 42. Shalaby MR, Waage A, Espevik T. Endotoxin, tumor necrosis factor-alpha and interleukin 1 induce interleukin 6 production in vivo. Clin Immunol Immunopath 1989; 53:488-498. 43. Fong Y, Tracey KJ, Moldawer LL et al. Antibodies to cachectin/tumor necrosis factor reduce interleukin 1 beta and interleukin 6 appearance during lethal bacteremia. J Exp Med 1989; 170:1627-1633. 44. Hesse DG, Tracey KJ, Fong Y et al. Cytokine appearance in human endotoxemia and primate bacteremia. Surg Gynecol Obstet 1988; 166:147-153. 45. Michie HR, Manogue KR, Spriggs DR et al. Detection of circulating tumor necrosis factor after endotoxin administration. N Engl J Med 1988; 318:1481-1486. 46. Eskandari MK, Bolgos G, Miller C et al. Antitumor necrosis factor antibody therapy fails to prevent lethality after cecal ligation and puncture or endotoxemia. J Immunol 1992; 148:2724-2730.

116

Cytokines and the Abdominal Surgeon

47. Tracey KJ, Vlassara H, Cerami A. Cachectin/tumour necrosis factor. Lancet 1989; i:1122-1125. 48. Michie HR, Spriggs DR, Manogue KR et al. Tumor necrosis factor and endotoxin induce similar metabolic responses in human beings. Surgery 1988; 104:280-286. 49. Beutler B, Milsark IW, Cerami A. Passive immunization against cachectin/tumor necrosis factor protects mice from lethal effect of endotoxin. Science 1985; 229:869-871. 50. Mathison JC, Wolfson E, Ulevitch RJ. Participation of tumor necrosis factor in the mediation of gram negative bacterial lipopolysaccharide-induced injury in rabbits. J Clin Invest 1988; 88:1925-1937. 51. Hinshaw LB, Tecamp-Olson P, Chang ACK et al. Survival of primates in LD100 septic shock following therapy with antibody to tumor necrosis factor (TNF-alpha). Circ Shock 1990; 30:279-292. 52. Tracey KJ, Fong Y, Hesse DG et al. Anticachectin/TNF monoclonal antibodies prevent septic shock during lethal bacteremia. Nature 1987; 330:662-664. 53. Fisher CJ Jr, Opal SM, Dhainaut JF et al. Influence of antitumor necrosis factor monoclonal antibody on cytokine levels in patients with sepsis. Crit Care Med 1993; 21:S436-S440. 54. Schlag G, Redl H. Pathophysiology of Shock, Sepsis, and Organ Failure. In: Schlag, G, Redl H, eds. Berlin Heidelberg: Springer-Verlag, 1993:1-1165. 55. Border J. Hypothesis: Sepsis, multiple systems organ failure, and the macrophage. Arch Surg 1988; 123:285-286. 56. Ayala A, Perrin MM, Kisala JM, Ertel W, Chaudry IH. Polymicrobial sepsis selectively activates peritoneal but not alveolar macrophage to release inflammatory mediators (IL-1, IL-6 and TNF). Circ Shock 1992; 36:191-199. 57. Ertel W, Morrison MH, Wang P et al. The complex pattern of cytokines in sepsis—Association between prostaglandins, cachectin and interleukins. Ann Surg 214:141-148. 58. Hadjiminas DJ, McMasters KM, Peyton JC, Cheadle WG. Tissue tumor necrosis factor mRNA expression following cecal ligation and puncture or intraperitoneal injection of endotoxin. J Surg Res 1994; 56:549-555, 1994. 59. Ayala A, Deol ZK, Lehman DL et al. Does endotoxin play a major role in inducing the depression of macrophage function during polymicrobial sepsis? Arch Surg 1995; 130:1178-1185. 60. Ayala A, Knotts JB, Ertel W, Chaudry IH et al. Role of interleukin 6 and transforming growth factor-beta in the induction of depressed splenocyte responses following sepsis. Arch Surg 1993; 128:89-95. 61. Ayala A, Deol ZK, Lehman DL et al. Polymicrobial sepsis but not low dose endotoxin infusion causes decreased splenocyte IL-2/IFN-gamma release while increasing IL-4/IL-10 production. J Surg Res 1994; 56:579-585. 62. Poll V, Jansen PM, Montegut WJ et al. Effects of IL-10 on systemic inflammatory responses during sublethal primate endotoxemia. J Immunol 1997; 158:1971-1975. 63. Joyce DA, Gibbons DP, Green D et al. Two inhibitors of pro-inflammatory cytokine release, interleukin-10 and interleukin-4, have contrasting effects on the release of soluble p75 tumor necrosis factor receptor by cultured monocytes. Eur J Immunol 1994; 24:2699-2705. 64. Gosselin D, DeSanctis J, Boule M et al. Role of tumor necrosis factor alpha in innate resistance to mouse pulmonary infection with Pseudomonas aeruginosa. Infect Immunol 1995; 63:3272-3278. 65. Greenberger MJ, Strieter RM, Kunkel SL et al. Neutralization of IL-10 increases survival in a murine model of Klebsiella pneumonia. J Immunol 1995; 155:722-729. 66. Zellweger R, Ayala A, Stein S et al. Females in proestrus state tolerate sepsis better than males. Crit Care Med 1997; 25:106-110.

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

117

CHAPTER 11

Hematopoietic Cytokines, G-CSF and Abdominal Surgery Artur Bauhofer, Wilfried Lorenz, Ilhan Celik, Benno Stinner, José Solovera, and Ronald Lorijn

Introduction Some Old and Some New Reasons for Conflicting Results: Success in Preclinical Biology and Failure in Clinical, Evidence-Based Medicine

P

ostoperative infection, sepsis and death in the course of sepsis are bad outcomes of surgical interventions. With increasing resistances of microbes to antibiotics and with an almost century-constant mortality rate of 30-50% in septic shock, new ways of combating infectious diseases are required. As one of the new ways, the modulation of the host response in infections attains new dimensions with the availability of biotechnically engineered products. On one hand, it has been shown in many immune compromised patients that the infection cannot be cleared until the host defense has been restored. On the other hand, an overwhelming SIRS (systemic inflammatory response syndrome) in the course of infection is detrimental, too. An adequate response is needed for the eradication of invading microbes with no clinically relevant or persistent harm for the body (Fig. 11.1). This seems achievable with the newly developed drugs in the field of cytokines. Indeed, in the first two decades after their discovery the two main clinical strategies with cytokines were the reduction of SIRS and the enhancement of the antimicrobial response of the host. However, almost all of the definite clinical studies to show effectiveness (multicenter randomized trials) which were performed till today, intended to reduce the SIRS (Table 11.1). Antibodies neutralizing endotoxin1,2 or TNF-!,3 soluble receptors binding and inactivating TNF-! in the blood stream or receptor antagonists (e.g., against IL-1)4 were used among others, but these expensive undertakings with almost 1000 patients in each trial showed no significant beneficial effect on mortality and in part they were even detrimental! Several reasons for this disaster were given: patients’ heterogeneity, redundancy of the cytokine pathways, different effects of cytokines in different phases of the pathophysiological state of sepsis. Trial errors were the most frequent reasons cited.5,6

Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

118

Cytokines and the Abdominal Surgeon

Fig. 11.1. SIRS elicited in the perioperative situation: operation, trauma and/or infection. An adequate local and systemic response is essential for eliminating micro-organisms, tissue debris for successful wound healing and for keeping homeostasis. An excessive or inadequate SIRS causes progressive tissue damage, shock and MOF. Modified from: Natanson C, WD Hoffman, AF Suffredini et al. Selected treatment strategies for septic shock based on proposed mechanisms of pathogenesis. Ann Intern Med 1994; 120:771-783.

However, several important aspects were missed in the present debate: 1) The complexity of the clinical management was not adequately simulated7 in animal and early human studies before the definitive randomized controlled clinical trials were started. A new experimental concept: clinical modelling randomized trial (CMRT) is needed to simulate systematically in animals the complex clinical situation.8 2) New endpoints9 including both health status measurements such as the McPeek index used for assessing postoperative recovery10,11 and quality-of-life indices are needed in addition to mortality rate as the true endpoint. 3) The management of patients in sepsis is characterized by numerous if-then interventions which compensate trial-specific omissions in the control group of sepsis studies. In addition, the clinical algorithms for interventions are extremely heterogeneous12 so that clinical practice guidelines are mandatory to reduce this heterogeneity.

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

119

Table 11.1. Summary of placebo controlled randomized trials with new agents in sepsis

agent methylprednisolone E5 monoclonal anti-endotoxin Ab E5 monoclonal anti-endotoxin Ab HA-1A human monoclonal anti-endotoxin Ab HA-1A human monoclonal anti-endotoxin Ab monoclonal anti-TNF-! Ab monoclonal anti-TNF-! Ab monoclonal anti-TNF-! Ab dimeric TNF-! receptor IL-1 receptor antagonist (rh-IL-1ra) platelet-activating factor receptor antagonist (PAF): BN 52021 - trial 1 (PAF): BN 52021 - trial 2

source

mortality rate [%] treatment placebo p-value

Bone et al Bone et al

34 30

25 26

0.06* 0.21

Greenman et al

38

41

0.72

McCloskey et al

33

32

0.86

Ziegler et al

24

34

0.12

Abraham et al Carlet et al Reinhard et al Sadoff et al Fisher et al

30-31 32-43 38-56 30-53 29-31

33 40 41 30 34

0.33 - 0.61 0.13 - 0.45 ...§ 0.02* 0.22

Dhainaut et al

42

51

0.17

Dhainaut et al

44

50

...§

* results suggest a harmful effect from treatment, § p-values were not reported. Adopted from: Bone RC. Why sepsis trials fail. JAMA 1996; 276:565-566.

Since—by their marked pleiotropic action—a clinical use of most cytokines (i.e., IL-1,13 IFN-!, IFN-∀14) will possibly be restricted to very specified clinical indications despite their proven activity in viral, bacterial and fungal infection, an increased interest was developed for another group of mediators in cell-to-cell communication: the hematopoietic growth factors (HGFs). They are more target-specific in their action than the classical cytokines, especially those factors (G-CSF, GM-CSF and M-CSF) which increase particularly the number and functions of specific phagocytes during infection.

Preclinical Biology: Hematopoietic Growth Factors (HGFs) in Infectious Disease Synopsis The development of cells in the hematopoietic system from stem cells (CD 34 positive cells) into different blood cell lineages and the proliferation of the progenitor cells themselves are regulated by stimulation and complex interaction of HGFs (Fig. 11.2). HGFs are glycoproteins with a molecular mass between 13 and 21 kDa15 These factors are expressed by a variety of cells including fibroblasts, blood cells and epithelial cells. The circulating number and the life-span of differentiated hematopoietic cells is determined by HGFs (Fig. 11.2). Blood cells involved in host defense

120

Cytokines and the Abdominal Surgeon

are stimulated to be more sensitive to chemoattractants like chemokines and bacterial products. The phagocytic and bacteriolytic activity of monocytes and granulocytes is also regulated by these factors. By altering the cellular host defense against microbes in the leucopenic and normopenic host, HGFs are candidates for a new type of drugs in infectious diseases, specially in patients with impaired neutrophil and monocyte function. The most important of them—depending on research and drug development—are listed in the following sections. All of them stimulate individually specific cell functions—but even more important—they alter and synergize their activities in combination.16 Once again, we are dealing therefore with complexity in the real clinic setting.

Stem Cell Factor (SCF) SCF is the ligand of the protooncogene receptor c-kit (CD 117). SCF stimulates very primitive pluripotent CD 34 positive cells (Fig. 11.2). In combination with other HGFs (G-CSF, GM-CSF, IL-3, IL-7 and rh EPO) colony formation is induced17-20 and sensitivity of progenitor cells to other growth factors is increased. For example the low reactivity to G-CSF in patients with severe congenital neutropenia is enhanced with SCF and the rate of perioperative infectious complications is reduced.21,22

Interleukin-3 (IL-3) At present there are two high affinity receptors for IL-3 postulated. One high affinity IL-3 receptor is described and formed by aggregation of the IL-3 specific 150 kDa membrane glycoprotein and GM-R∀ which also constitutes the large subunit of the GM-CSF receptor. IL-3 acts on more primitive hematopoietic progenitor cells which are less sensitive to GM-CSF.23 The effects on more lineage-specific cells are secondary in comparison to G-CSF and GM-CSF. Macrophages stimulated by bacterial endotoxin (LPS) are enhanced by IL-3 to secrete proinflammatory cytokines including IL-1, IL-6 and TNF-!. IL-3 has limited effects on neutrophils but enhances the activity of eosinophils and basophils.24-26

Megakaryocyte Growth and Development Factor (MGDF) Both synonyms MGDF and TPO (thrombopoetin) are in use for the new HGF stimulating platelet formation. MGDF exerts its effect by binding to the receptor, encoded by the protooncogene c-mlp, which is expressed on a large number of cells of hematopoietic origin. MGDF regulates early and late stages of platelet formation by increasing size and number of megakaryocytes (Fig. 11.2). It is also involved in platelet activation.27,28 Beside these effects on platelets, however, MGDF primes neutrophil responses to the bacterial wall-derived peptide: n-formyl-met-leu-phe (FMLP). FMLP induces an early oxidative burst and stimulates the production and release of the chemokine IL-8 from neutrophils.29

Macrophage Colony Stimulating Factor (M-CSF) M-CSF exerts its effect by binding to the M-CSF receptor homodimer which is encoded by the protooncogene c-fms (CD115). This protooncogene is predominantly expressed on cells of the monocyte lineage.20 M-CSF stimulates the development of monocytes (Fig. 11.2), increases their circulating number and regulates their functional activities. In macrophages M-CSF increases antibody dependent cellular cytotoxicity, but stimulates also antibody-independent intracellular killing of various types of micro-organisms (bacteria, fungi, vira). M-CSF induces in monocytes and macrophages the synthesis of proinflammatory factors: cytokines such as IL-1, IFN-#

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

121

Fig. 11.2. Interaction of HGFs and cytokines in hematopoesis. Starting from CD 34 positive totipotent/pluripotent stem cells later stages (multipotent, bipotent or unipotent progenitors) are derived that can be identified by means of sets of cell surface markers. The synergistic interaction of the single factors in cell differentiation is: BFU-E = burst forming unit erythroid, CFU-Eo = colony forming unit eosinophil; CFU-GEMM = colony forming unit granulocyte, erythrocyte, macrophage, monocyte; CFU-MEG = colony forming unit megacaryocyte; CFU-GM = colony forming unit granulocyte, macrophage; CFU-M colony forming unit monocyte; CFU-G = colony forming unit granulocyte; CD14+ = CD14 positive cell; MEG-CSA = megakaryocyte-colony stimulating activity; NP = neutrophil precursor. Modified from: Ibelgaufts H. Dictionary of cytokines. Editiones Roche, Basel, Switzerland. 1995.

122

Cytokines and the Abdominal Surgeon

and TNF-!, but also G-CSF, and plasminogen activator, prostaglandines, thromboxanes and reactive oxygen species.30

Granulocyte-Macrophage Colony Stimulating Factor (GM-CSF) The GM-CSF receptor is mainly expressed on immature myeloid cells (Fig. 11.2). Its density decreases with increasing maturation of these cells. Lymphoid cells do not express the receptor.31 The receptor is also expressed on endothelial cells and on small cell lung carcinoma cells. GM-CSF stimulates the proliferation of neutrophil and monocyte progenitors as well as that of eosinophils and basophils. GM-CSF has a less proliferative effect on mature neutrophils and cell release activity on neutrophils from the bone marrow.32 However, GM-CSF prolongs the life-span of neutrophils, enhances cell functions and chemotaxis.33 The half life of phagocytes is prolonged by this factor by suppressing apoptotic removal.34 Deficient neutrophil phagocytosis and oxidative burst are stimulated.35 GM-CSF potentiates the production of TNF-!,36 which in adequate levels (Fig. 11.1) is beneficial for the phagocytic response and intracellular micro-organism killing.37 However, in septic shock (excessive SIRS, Fig. 11.1) the excessive production and release of TNF-! is often detrimental to the patient.38 Cytokine secretion and migration of monocytes and neutrophils may be enhanced by the increase of the hyaluron binding receptor CD 44 on these cells.39 Brief exposure of neutrophils to GM-CSF promotes their movement to the site of infection probably by increasing the expression of surface receptors for chemotactic stimuli and adhesion. Longer exposure appears to inhibit the movement. This suggests that GM-CSF may firstly attract neutrophils to the side of infection and then prevents their movement away.33,40 GM-CSF also increases the MHC class-II receptor expression on monocytes which is important for an active monocytic response to microbes.41

Granulocyte Colony Stimulating Factor (G-CSF) Among the hematopoietic cytokines, G-CSF has obtained the greatest clinical importance. So far more than 1,200,000 patients have received rhG-CSF (filgrastim).42 Recombinant human G-CSF for clinical use is, however, available both in a nonglycosylated (filgrastim) and a glycosylated (lenograstim) form. Resulting from alternative splicing, at least four forms of G-CSF high affinity human receptors are described. The receptors are exclusively localised on cells of the neutrophil lineage, on early stages of monocytes and on placenta cells, endothelial cells and some carcinoma cells.43,44 G-CSF has several functions which are all beneficial in the control of infection and limiting excessive SIRS (Fig. 11.1). 1) G-CSF increases dramatically the number of granulocytes in the peripheral blood and tissues by decreasing their maturation time in the bone marrow and releasing more neutrophils from storage pools in the bone marrow into the blood stream. In addition, it increases the survival time of peripheral blood neutrophils in humans—a very important function since the half life time for them is only about 8 hours. G-CSF achieves this goal at lower concentrations than those required for stimulating colony formation.45,46 The mechanism is reduction of apoptosis, especially effective in combination with IFN-# while chemotaxis and fungal killing (C. albicans) is still preserved.34,47 2) G-CSF stimulates and enhances the activity of neutrophil granulocytes which is the general base for its great effect in infectious diseases. This was shown first in a series of animal models,36,48-51 summarized in several reviews

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

123

(Nelson,52 Dale et al53). However, as will be shown later, it was also successful now in some clinical trials on community acquired pneumonia54 and AIDS. The mechanisms involved in the increased activity of neutrophils include potentiation of binding of several bacterial agonists for superoxide ion production,55 augmenting by this way neutrophil cytotoxicity to microbes.56 The microbial agents are bacterial products which are derived from the cell wall from gram-positive (i.e., peptidoglycans, lipoteichonic acid) and gram-negative bacteria (i.e., endotoxin, porins57). Downward in the mediator cascade the same is true for cytokines and chemokines which are induced by the bacterial products or by trauma. G-CSF increases also the density of cellular surface receptors for laminin and fibronectin on neutrophils, especially by upregulation to normality in septic patients.58 The killing of E. coli by purified neutrophils is synergistically induced by the antibiotics ofloxacin or ceftazidime in combination with filgrastim.59,60 3) G-CSF reduces effectively the systemic48,49 and local release by liver and mesenterium50 of the proinflammatory mediator TNF-! in sepsis. The overwhelming SIRS reaction is reduced by attenuation of TNF-!.61 In humans, G-CSF attenuates the release of TNF and other proinflammatory cytokines in response to endotoxin and bacterial products.62,63 These last mentioned articles are outstanding studies in human subjects and have considerable relevance for our clinical understanding of the effects of G-CSF. During infection, G-CSF acts on two levels: systemically the formation and release of additional phagocytes is increased and locally neutrophils are recruited and activated by the induction of chemokines. This process continues until the infection is eradicated and the production of IL-1 and TNF-! is decreased. The latter induces in a positive feedback loop the formation of G-CSF by a variety of blood cells.16,64

Clinical Reality: The Overall Complexity of Abdominal Infections

In an international clearing conference on ‘Surgical Research around the World’7 the participants unanimously recognized four quite different forces which influenced clinical research enormously in only the short period of 10-15 years: globalized information, rise of molecular biology, outcome movement and the concept of complexity. The latter is inherited in our common language, but has risen now to a subject of scientific and methodological interest. Indeed, abdominal infections were chosen as a clinical example in which the concept, the terminology and methods of assessing complexity could be exemplified (Fig. 11.3). What is complex? The simplest way to describe a complex system is a great many independent variables (agents, effectors, mediators) interacting with each other in a great many ways.7,65 This definition does not predict the model of handling the many variables, which might be stochastic, deterministic, chaotic or fractal or something else.7 The system in the real world (the patient in his clinical situation) is modelled (by the doctor, the nurses, the health managers) using particular items and the relations between the items. If the items and their relationship are affected by or changes over time, the term dynamic instead of static systems is used. This is usually the case in clinical situations. The more items and relations a system demonstrates the greater is its complexity. Intricacy is an expression of the inhomogeneity of the items. In clinical practice, the complexity can be measured by clinical algorithms structure analysis (CASA).66,67 Using the same clinical situation (e.g., anastomotic leakage after

124

Cytokines and the Abdominal Surgeon

Colonic resection and postoperative leakage of anastomosis - Removal of foci

Laparotomy

Peritonitis no

• - Intraoperative lavage Raising of anasto• -mosis colostomy Performance of new • -anastomosis • -i.v. Antibiotics

Planned relaparotomy no

yes

- Transfer to ICU

yes

Improvement of health no

Observation

Improvement of health yes

no

Normal ward

Fig. 11.3. Measurement of clinical guidelines (algorithms) complexity. n CASA (scores) = 2 Dx + D0 + ∃ Li , when Dx = number of decision box (➱), D0 i =1

number of all action and state boxes (❑), Li = weighted sum of all boxes between loop origin and re-entry and n = number of loops. The seed algorithm was proposed from all individual centers for a consensus process in Marburg, January 7-9, 1997. ❍ = all steps in the algorithm in which the different centers and countries disagreed. Reproduced with permission from: Lorenz et al. Surgical research around the world. In: Surgical research: basic principles and clinical practice. Troidl H, ed. Heidelberg, New York: Springer. 1997, in press.

resection for colorectal cancer) the CASA values for managing the patient (caserelated clinical practice guidelines) may vary from 19-168 units (Fig. 11.3). Complexity and intricacy are important analytic concepts for our routine clinical world. Hence we specify them for abdominal infections or one of the targets for filgrastim and other hematopoietic growth factors in three domains: heterogeneity of the patients’ risks for abdominal infection, heterogeneity of treatment concepts in different countries or centers and heterogeneity of the individual treatment of the single case and its influence on the cytokine and inflammatory mediators’ network.

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

125

Table 11.2. Proven risk factors for severe abdominal infection and sepsis Main or Concomitant Disease

Reference

Tumors with: overwhelming malignancy chemo- or radiotherapy liver cirrhosis malnutrition diabetes mellitus

Cheadle WG et al (111) Link H et al (112) Cheadle WG et al (111) Windsor JA, Negro GC et al (113,114) Delamaire MD et al (115)

Immune Suppression by Therapy: liver transplantation liver irradiation blood transfusion high age as ‘concomitant disease’

Yokoyama et al (116) Yokoyama et al (116) Morgan AS et al (117) McLauchlan GJ et al (118)

Heterogeneity of the Patients’ Risk For Abdominal Infection Peritonitis or sepsis can be elicited by different events (Fig. 11.1). Infection follows contamination.68 In most cases infection starts locally (i.e., from perforated appendicitis or pancreatitis) and involves also a variety of microbes (gram-negative, gram-positive, aerobic and anaerobic bacteria or fungi). By systemic reaction of the whole subject to the different types of infection SIRS is induced. However, it is mostly the extent of SIRS which is crucial for the outcome of the patient. An adequate SIRS (Fig. 11.1) will help the body to eliminate the microbes while an excessive or an inadequate reaction will lead to a further destruction of tissues and the development of the complex situation of multiple organ failure (MOF). The patients in trials on abdominal infections show a variety of main diseases. In addition, they have very often several concomitant diseases which are associated with high risk for severe infection and sepsis. Proven risk factors are shown in Table 11.2. These risk factors will define the immunological status of the patient which determines the outcome after abdominal operation and infection fundamentally beside the surgical management. This is a common experience in mortality and morbidity conferences.

Heterogeneity of Treatment Concepts in Various Countries and Medical Centers Pursuing old traditions and individual expertise (intuition), but in the absence of evidence based on randomized controlled clinical trials which have demonstrated superior treatment regimens, the therapy of abdominal infections is very heterogeneous.12,69,70 In France, for instance, on one hand an antimycotic drug is applied prophylactically and therapeutically with antibiotics71,72 which is quite debatable in the US, UK and Germany. On the other hand, intraoperative and/or postoperative lavage of the abdomen is recommended very often in Germany—with physiological salt solution or with taurolidine73—whereas such procedures are totally rejected in surgical units of other countries.74-76

126

Cytokines and the Abdominal Surgeon

Leucocytes TNA-RII TNA-RI IL-6 Elastase CRP

OP1

OP2 G-CSF

Days OP3 AB 2

Fig. 11.4. Inflammatory parameters before and after G-CSF application. OP1 = elective colectomy. OP2 = relaparotomy and intraoperative lavage. OP3 = planned relaparotomy and intraoperative lavage. AB1 = co-amoxiclav. AB2 = cefuroxime, metronidazole, amikacin. CRP = C-reactive protein. Reproduced with permission from: Reimund et al. Management of leucopenic sepsis. Lancet 1995; 346:382-383.

The whole management of an ICU is a very important factor in demonstrating effectiveness of sepsis treatment. The availability of advanced technological support has been associated significantly with lower risk—adjusted mortality.77 Highly specialized units, for example stroke units, significantly reduced mortality in comparison to general wards which was demonstrated in two meta-analyses.78,79 The incidence of surgical wound infections was influenced by the operative techniques,80 but again also by the perioperative management. Normothermia, for instance, reduced significantly the incidence of surgical wound infections.81 However, in diffuse peritonitis planned relaparotomy (see Fig. 11.3) did not show any advantage over relaparotomy on demand in several outcome variables.82

Heterogeneity of the Individual Patient’s Management: Influence on Inflammatory Parameters Very often, the single patient with abdominal infections is not managed according to an accepted clinical practice guideline, but from day to day on individual demands—or what the doctors believe what they are. The complexity of such a management and the influences of that on inflammatory parameters and cytokine levels is illustrated by a case of a 36 year old man with familial adenomatous polyposis (FAP) who developed 3 days after proctomucosectomy with ileoanal J-pouch and loop ileostomy a leucopenic sepsis.83 Based on preclinical and some early human data, it was decided to administer 300 ∝g Neupogen® (filgrastim) subcutaneoulsy as part of an experimental emergency treatment after the emergency relaparotomy. Two days later, a planned second relaparotomy was performed (Fig. 11.4). Most inflammatory parameters (C-reactive protein, elastase, IL-6, soluble tumor necrosis factor receptor I and II (sTNF-RI and II) were considerably increased before filgrastim application (Fig. 11.4). After administration of filgrastim the leukocyte count increased threefold within 2 h and the IL-6 concentration fell by 80% within 10 h. Probably these effects were caused by filgrastim, as reported by others,63 but interpretation is confounded by other treatments including antibiotics and especially the planned

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

127

relaparotomy. In the complex clinical situation the outcome can therefore not be attributed to a single intervention. There is a clear need to monitor the different inflammatory parameters (including cytokines), in order to understand more about the impact of the surgical and medical strategies on the immunological status of the patient.

New Concepts: Demonstrating Effectiveness of Drugs in Clinical Trials and Daily Practice for Abdominal Infections Beside new endpoints in randomized trials which were the subject of an extended analysis in another paper,9 the failures of trials in abdominal infections and sepsis can be attacked by two other concepts: clinical algorithms and clinic modelling randomized trials in animals. First, clinical complexity can be reduced by the development of clinical algorithms. Using this important tool and, second, designing animal studies which simulate the complexity of the clinical situation (CMRT: clinic modelling randomized trials) we tried to come to grips with the problems of complexity and heterogeneity.

Clinical Algorithms—Reducing Complexity Heterogeneity of clinical management was shown and complexity was reduced by setting up an international study on clinical algorithms. Clinical cases were collected from 22 centers in Europe and the US related to the treatment of postoperative anastomotic leakage after segmental resection of the colon for cancer. Each center summarized the case in form of an extended discharge summary and transformed it into a time-action table—day by day. After collection of all cases from the centers, case-specific algorithms with a standardized graphic format67 were developed. By abstraction from the case specific algorithm center specific algorithms were designed. One of these algorithms which was used as a seed algorithm for the consensus process (Nominal Group Process) is illustrated in Figure 11.3. To reduce the heterogeneity of the clinical management between the various centers, a Nominal Group Process was conducted in Marburg in 1997.12 Delegates of the 22 centers participated at this meeting and searched for consensus about the seed algorithm. The resulting algorithm of the group process was only developed in part since the heterogeneity between the centers was too large (Fig. 11.3). This was the main result of the meeting. However, the consensus process was not blocked and is still in progress. There is still hope and reasonable enthusiasm that the guideline will help to reduce heterogeneity in the management of sepsis both during daily clinical practice and during the conduction of a multicenter study. Guidelines and protocols, today, are an essential part of good clinical quality practice (GCQP).

Clinic Modelling Randomized Trials (CMRT): Increasing Complexity In order to investigate the effects of cytokines and other new immunoregulatory drugs in the complex clinical world, more and more reductionistic experiments in biomedical sciences are not sufficient for predicting clinical effectiveness and for a contribution to evidence-based medicine. A new study type modelling the clinical scenario and real clinical trial conditions was therefore conceptualized for laboratory animals (Table 11.3). • It combines both items which model the clinical scenario in which the promising new drug is intended to be administered, and items which model a

128

Cytokines and the Abdominal Surgeon

Table 11.3. Summary of the clinical and trial conditions applied in clinical modelling randomized trials (CMRT) in the rat with abdominal contamination and infection (ACI). Items Modelling Clinical Scenario

Items Modelling Randomized Trial

• sepsis mortality rate of 50%

• study population: large sample possible! • resistance to antibiotics included • randomization: block design • abscess formation (pus) to be shown • blinding procedures • course of the experiment similar to clinic • dose-response curves time and dose of drug administration (responsiveness of the model) perioperative volume substitution • reproducible with various stool probes anaesthesia and time intervals (reliability of the model) antibiotic prophylaxis and therapy • endpoint mortality rate within 5 days, surgery (laparotomy) life tables contamination with human stool • evaluation of morbidity postoperative analgesia • statistical analysis as in randomized trials sample size calculation • analysis of cellular functions from blood mortality rate as true endpoint and organ specimens (e.g. log rank test) • risk factors included: age, diabetes, • survival analysis cardio-pulmonar, alcohol (e.g. Kaplan-Meier curves) • guide-lines (algorithms) • analysis of cellular parameters in relation to outcome Reproduced with permission of Bauhofer A et al. Clinic modelling randomized trials in animals as a new concept for designing clinical sepsis trials: antibiotics modulate G-CSF effectiveness in severe abdominal infection in rats. J Clin Invest. 1997, in preparation.

randomized controlled clinical trial with a considerable number of animals as study patients. • The study produces in the control group a 5 day mortality rate of about 50% and includes clinical details such as an extended laparotomy and analgesia with opioids what occurs regularly in patients. • The items modelling a randomized trial include first a defined randomization and blinding procedure, a prestudy calculation of sample size and analysis of cellular parameters (migration, formation of oxygen radicals etc.) in relation to outcome. CMRTs have to be performed in right time, e.g., in parallel to phase I and phase II clinical trials (Fig. 11.5) to investigate the interaction of the new drug with various operative and perioperative interventions such as anaesthesia, antibiotics, reoperation and lavage before the definite trials can be conducted. Furthermore, clinically relevant end points must be selected, especially a survival rate which is not only defined by the toxicity of a SIRS inducer. To demonstrate effectiveness of a new drug, the difference in mortality between the control treat-

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

129

Fig. 11.5. Increase of complexity in the experimental setting. Beginning with the most reductionistic cell and molecular experiments, pharmacological experiments in the whole organ and animal reaching the new intermediate between perchance and clinic CMRT. In the clinic complexity increases from case reports to controlled phase II trials culminating in multicenter phase III trials. Reproduced with permission from: Bauhofer A et al. Clinic modelling randomised trials in animals as a new concept for designing clinical sepsis trials: antibiotics modulate G-CSF effectiveness in severe abdominal infection in rats. J Clin Invest. 1997, in preparation.

ment and an additive treatment should realistically be situated between 20-30%. The clinical effects of 100% as in reductionistic models should be avoided as clinically unrealistic. • For these reasons, finally statistically significant results can only be obtained using a higher number of animals per group as usually applied in biomedical studies, e.g., 15-20 animals/group. As a consequence of that rodents have to be used as a realistic compromise between ethical considerations and the similarity to the clinical situation. • In general, the trial design should involve the activity not only of basic scientists, but also that of clinicians who really know about the items of the clinical situation. • A realistic sample size calculation i.e., by the formula of Friedman84 with relevant parameters (2a = 0.05, b = 0.1, d = 0.5) is needed. • And last but not least, not only positive results but also results without effect or negative results have to be designed, studied and reported to avoid conditions in the definitive trial which are prone to failure. Also dropouts have to be listed. This is completely different from reporting animal data in biomedical sciences.

130

Cytokines and the Abdominal Surgeon

A Convincing Example for Increasing Clinical Complexity in Preclinical Trials: Experience with rhG-CSF (Filgrastim) CMRTs with rats suffering from severe Abdominal Contamination and Infection (ACI) were performed to investigate the effectiveness of rhG-CSF in reducing the mortality rate. The following variables were introduced: anaesthesia, volume loading, antibiotic prophylaxis, operation, infection with human standardized stool bacteria and analgesia.8,48 The items modelling clinical scenario and those modelling randomized trials with ACI corresponding to those in Table 11.3. CMRT with ACI was validated by testing reproducibility with various stool samples (male, female, mixed samples) and demonstration of dose response curves (LD5O = 1.0-1.25 ml/kg, LD100 > 1,5 ml/kg stool). All antibiotics tested induced a shift of the dose response curves (LD5O = 1.25-1.5 ml/kg stool) using clinically relevant dosages. Microbiological analysis of the stool samples showed similar patterns of aerobic and anaerobic bacteria in each probe. Antibiotics were selected by their common clinical usage in several Western countries. With all antibiotics tested a survival rate in the range of 50% was obtained in CMRT-ACI. Metronidazole was additively introduced in the trials to close the gap of the antibiotics in their effectiveness against anaerobes. Especially the second and third generation cephalosporins were less active against anaerobes. The introduction of rhG-CSF as adjuvant prophylaxis showed enormous differences. A repetitive dosing of rhG-CSF on three consecutive days (s.c. 20 ∝g/kg at 12 h before, 12 h after and 36 h after ACI) as mentioned by Cohen et al85 increased the survival rate compared to a single application (50 ∝g/kg at 12 hbefore ACI).

Positive CMRTs with rhG-CSF An additive increase of the survival rate was obtained by the combination of rhG-CSF with co-amoxiclav (10 mg/kg), cefuroxime/metronidazole (10/3.5 mg/kg), amikacin/metronidazole (15/3 mg/kg) and ofloxacin/metronidazole (5/2 mg/kg). Prophylaxis of rhG-CSF in combination with the third generation cephalosporin cefotaxime/ metronidazole (p = 0.09, one sided Fisher test) showed only a modest increase in survival rate (Table 11.4a).

Negative CMRTs with rhG-CSF No beneficial effect was seen by the therapeutical application of rhG-CSF at +1 h, +2 h or 6 hours after operation or at the time of operation (data not shown). Ofloxacin and the cephalosporins were less active without metronidazole. In combination with the third generation cephalosporin ceftriaxone, no increase of the survival rate was seen alone nor in combination with metronidazole (Table 11.4b).

First Explanations for rhG-CSF Positive and Negative Trials: Relevance for Designing and Conducting Definitive Clinical Trials Why are the two third generation cephalosporins ceftriaxone and cefotaxime less potent in combination with rhG-CSF than the other antibiotics tested? There are several explanations. Cellular functions of phagocytic cells (granulocytes and macrophages) are altered by antibiotics. Migration to the side of infection, phagocytosis and lysis of the microbes by radicals and proteases are the most important cellular functions for the elimination of microbes.86-88 For this reason the WBC count, the PMN ratio, the capacity of granulocytes from rats with ACI to produce O2– radicals and to migrate in direction to a chemoattractive stimulus were analyzed (Table 11.5). 24 h after operation without antibiotics and rhG-CSF the WBC, the migratory activ-

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

131

Table 11.4. The 120 hour survival rate of rats in CMRT with ACI in combination with rhG-CSF a) Positive CMRT Results Antibiotics

Dosage (mg/kg)

co-amoxiclav 10 cefuroxime/ 10 / 3.5 metronidazole amikacin/ 15 / 3 metronidazole imipenem 3 ofloxacin/ 5/2 metronidazole cefotaxime/ 6/2 metronidazole

N

Survival Rates of the Rats AB alone AB + rhG AB + rhG-CSF 1x CSF 3x

P-Values

12 18 12

41% 44% 50%

83% – 75%

– 82% 92%

< 0.05 < 0.05 < 0.05

18

44%

–

82%

< 0.05

12 18 18

75% 28% 50%

67% –

100% 56% 83%

0.06 0.08 < 0.05

18

35%

–

65%

0.09

b) Negative CMRT Results Antibiotics

Dosage (mg/kg)

N

Survival Rates of the Rats AB Alone AB + rhG AB + rhG -CSF1x -CSF 3x

ofloxacin cefotaxime ceftriaxone ceftriaxone/ metronidazole

6 8 10 5/2

12 12 12 11

33% 50% 42% 55%

17% 75% 33% 46%

50% 75% 1.0 46%

P-Values

0.14 0.24 1.0

The antibiotics (AB) were given alone, in combination with a single shot of rhG-CSF at -12h with 50 ∝g/kg or with a repeated s.c. application of rhG-CSF at -12h, +12h and at +36h (20 ∝g/kg). In separate studies in the control groups without antibiotics no rat survived. N is the number of animals per group. P-Values were determined in the one-sided Fisher test. Reproduced with permission from Bauhofer A et al. Clinic modelling randomized trials in animals as a new concept for designing clinical sepsis trials: antibiotics modulate G-CSF effectiveness in severe abdominal infection in rats. J Clin Invest. 1997, in preparation.

ity and the production of O 2 – were down-regulated. Using co-amoxiclav prophylacticaly the migratory activity was elevated and in combination with rhGCSF restored to the control level of healthy rats (Table 11.5). The O2– production was not increased by co-amoxiclav, but in combination with rhG-CSF this function of the granulocytes was restored, too. In contrast, with ceftriaxone and rhG-CSF no restoration of the cellular functions (migration and O2– production) were seen. Other antibiotics like clindamycin and netilmicin suppress antimicrobial cellular functions of granulocytes, too,89 but rhG-CSF synergises with ofloxacin in stimulating neutrophil bactericidal functions in vitro.59 In 65 of 240 patients suffering from recurrent infection, a lack of oxygen radical production or reduced chemotaxis was

132

Cytokines and the Abdominal Surgeon

Table 11.5. Blood cell parameters and cellular functions Experiment A WBC 106 cells/ml

groups (n = 8)

control rhG-CSF only ACI ACI, AB1 ACI, rhG-CSF, AB1

11.7 ± 3.3 22.7 ± 5.8 8.0 ± 2.4 8.3 ± 2.2 7.6 ± 1.8

PMN %

16 31 34 30 26

±4 ±8 ± 14 ± 10 ± 13

PMN migration in PMA induced O2% nmol/min/ x106 + activated serum cells 284 251 141 251 312

± 75 ± 110 ± 29 * ± 131 ± 172 *

0.32 ± 0.25 ± 0.19 ± 0.20 ± 0.32 ±

0.13 0.08 0.07 § 0.04 0.12 §

Experiment B groups (n = 8)

WBC 106 cells/ml

control 12.2 ± 3.7 ACI, AB2 8.9 ± 2. ACI, rhG-CSF, AB2 10.7 ± 2.1

PMN %

13 ± 3.9 37 ± 10 35 ± 5

PMN migration in PMA induced O2– % nmol/min/ x106 + activated serum cells 250 ± 62# 199 ± 71# 215 ± 83

0.39 ± 0.12$ 0.30 ± 0.06$ 0.29 ± 0.09

Blood cell parameters and cellular functions of two sets (A + B) of experiments from healthy control rats, healthy rats treated with rhG-CSF only (20 ∝g/kg, 12h before and 12h after ACI), from ACI rats, ACI rats with antibiotic prophylaxis (AB1 = co-amoxiclav, AB2 = ceftriaxone/ metronidazole) and ACI rats with AB and rhG-CSF prophylaxis (12h before and after operation, 20 ∝g/kg). Blood was sampled 24h after ACI. WBC = White blood cell count. Data are the mean ± SD. Two group significance analysis was performed with the Wilcoxon-test. * p < 0.01, § p = 0.06, # p < 0.05, § p = 0.09. Reproduced with permission from Bauhofer A et al. Clinic modelling randomised trials in animals as a new concept for designing clinical sepsis trials: antibiotics modulate G-CSF effectiveness in severe abdominal infection in rats. J Clin Invest. 1997, in preparation.

found.90 Finally, as demonstrated in patients, animal experiments and isolated organs,91 antibiotics have also cardiovascular adverse effects in complex surgical cases. Rapid lysis of gram-negative bacteria is associated with a considerable release of free endotoxin. Endotoxin is a potent stimulator of the release of proinflammatory cytokines like TNF-!. Whole blood incubated with Haemophilus influenzae type b and ceftriaxone resulted in a greater release of TNF-! than incubated with imipenem.92 Third generation cephalosporins like ceftriaxone have a high affinity to penicillin binding protein 1 and 3 (PBP 1+3). These PBP’s induce the formation of bacteria filaments and release high amounts of endotoxin.93 Imipenem, an other betalactam antibiotic has a higher affinity to PBP-2, form’s spheroids and releases less endotoxin. The different mode of bacterial elimination could be also an explanation for the different survival curves obtained by the antibiotics in CMRT with ACI. In-vitro cell culture experiments have shown that penicillin G, imipenem, netilmicin and vancomycin did not influence CSF (colony stimulating factor) secretion by monocytes or T-lymphocytes. In contrast the cephalosporins (ceftazidine and cefuroxime) suppress GM-CSF and G-CSF secretion of endothelial cells and T-lymphocytes.94 In sepsis patients treated with cephalosporins the synergistic acti-

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

133

vation of phagocytic cells by different CSFs may be disturbed. The antibiotic uptake by granulocytes is increased in the presence of rhG-CSF.95 This increased uptake will be only a benefit for some antibiotics.

Perspectives: Filgrastim in Abdominal Surgery Filgrastim (rhG-CSF) is a good candidate for clinical use in abdominal surgery. So far more than 1,200,000 patients worldwide received rhG-CSF42 without marked side effects. Several case reports83,96-98 and a few phase I trials with surgical patients and infectious disease have shown a benefit by filgrastim prophylaxis and treatment.54,99-101 Filgrastim seems to be safe not only in neutropenic but also in normopenic patients.102 Before new large sepsis trials with filgrastim will be initiated the trial conditions must be redefined. CMRTs in animals are needed to define the trial conditions by the analysis of cytokine interaction with the complex clinical scenario. We have demonstrated that the antibiotic regime has to be well selected for both study purpose as well as daily use when given in combination with cytokines. Sepsis patients must be handled in a more standardized way. Clinical guidelines will help to reduce treatment heterogeneity in centers and between centers. For this reason we have developed clinical algorithms for the treatment of postoperative leakage of anastomosis after colon resection. In a defined group of patients with high risk (ASA III) a sepsis prophylaxis with filgrastim will be of benefit. A randomized multicenter placebo controlled trial with filgrastim as prophylaxis before colon resection is under development. Not only in sepsis but also in other diseases filgrastim is a promising candidate for new treatment strategies. In a placebo controlled trial with filgrastim in patients with diabetes mellitus developing diabetic foot, filgrastim showed benefits in an earlier eradication of pathogens from the infected ulcer, a shorter hospital stay and a shorter duration of the antibiotic treatment.103 G-CSF shows also beneficial effects in colitis by the attenuation of the inflammatory response and in glycogen storage disease Ib-related colitis.104 Crohns disease is an other disease with an immunological disorder and a new possible indication of filgrastim.63 Also in AIDS patients filgrastim and GM-CSF will have a great future in treatment of specific HIV-related opportunistic infections.105 In HIV patients the number of functional granulocytes is decreased by apoptosis. In vitro, the incubation of granulocytes from AIDS patients with rhG-CSF prolongs the survival time of the cells by decreasing apoptosis.106 Stimulating the host defence against microbes and reducing the inflammatory reaction are the promising good effects of filgrastim and other related factors like GM-CSF. Only by designing and executing appropriate placebo controlled clinical studies we will be able to understand how to use these natural hematopoietic growth factors adequately in nonneutropenic infectious diseases.

Conclusion Several new promising drugs for the modulation of the immunological response to systemic infection were engineered by gene technology. The most important factors altering blood cell differentiation and cellular host defence against microbes are summarized in the first part of the article. All cytokines used in clinical sepsis trials were introduced after experiments in cell cultures and pharmacological experiments. In these reductionistic approaches anti-inflammatory activities and effectiveness was demonstrated in the reduction of the mortality rate. This was in opposite to clinical trials were no benefit could be observed. Several reasons (i.e., inadequate trial design, wrong dosage, patient and center heterogeneity, insufficient number of patients)

134

Cytokines and the Abdominal Surgeon

were discussed. One critical topic in most of the trials is the selection of the real endpoint.9 We introduced two new concepts to improve the quality of sepsis trials. First, the clinical management heterogeneity between and in centers should be reduced. The way to achieve this goal are case-based clinical algorithms. Second, the complexity in animal trials should be increased by the new concept of clinic modelling randomized trials in animals. An example was demonstrated with rhG-CSF prophylaxis in rats with abdominal contamination and infection (ACI). The results show the importance of CMRT trials for trial conditions. Definite proof by a randomized trial in humans is lacking for filgrastim, but was given in another situation: we have shown the positive transfer of animal data107 into a clinical trial in the case of life-threatening anaphylactoid reactions occurring in the perioperative period in surgery. The histamine H1 + H2-prophylaxis and volume loading after induction of anaesthesia108 was effective in the clinical setting. The positive effect of filgrastim to increase the survival rate and to increase the antimicrobial function of granulocytes is dependent on antibiotics. This is a classical example for complexity in the clinical setting. Hence more complex animal experiments modelling the clinical situation are fundamental for the development of successful clinical trials in the field of pleiotropic cytokines.

List of Abbreviations AB ACI AIDS ASA

Antibiotic Abdominal Contamination and Infection Acquired Immunodeficiency Syndrome American Society of Anesthesia Score for Preoperative Physical Status BFU-E Burst Forming Unit-Erythroid c-fms CD 115: M-CSF Receptor c-kit CD 117: SCF-Receptor c-mlp MGDF-Receptor CD Cluster of Determination CD14+ CD14 Positive Cell CFS Colony Stimulating Factor CFU-EO Colony Forming Unit Eosinophil CFU-G Colony Forming Unit Granulocyte CFU-GEMM Colony Forming Unit Granulocyte, Erythrocyte, Macrophage, Monocyte CFU-GM Colony Forming Unit Granulocyte, Macrophage CFU-M Colony Forming Unit Monocyte CFU-MEG Colony Forming Unit Megacaryocyte CMRT Clinic Modelling Randomized Trials CRP C-Reactive Protein EPO Erythropoietin FAP Familial Adenomatous Polyposis FMLP n-formyl-met-leu-phe #-IFN Gamma Interferon GCQP Good Clinical Quality Practice G-CSF Granulocyte Colony Stimulating Factor GM-CSF Granulocyte-Macrophage Colony Stimulating Factor

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

GM-R∀ HGF IFN-!/-∀ IL-1ra IL LPS M-CSF MEG-CSA MGDF MHC MOF NP PBP rh SCF SIRS sTNF-RI/II TNF-! TPO WBC

135

Granulocyte-Macrophage-Receptor-Beta Subunit Hematopoietic Growth Factor Interferon-Alpha/Beta Interleukin Receptor Antagonist 1 Interleukin Lipopopolysaccharides Macrophage Colony Stimulating Factor Megakaryocyte Colony Stimulating Activity Megakoryocyte Growth and Development Factor (Synonymous Thrombopoetin (TPO)) Major Histocompatibility Complex Multiple Organ Failure Neutrophil Precursor Penicillin Binding Protein Recombinant Human Stem Cell Factor Systemic Inflammatory Response Syndrome Soluble TNF-Receptor I/II Tumor Necrosis Factor Alpha Thrombopoetin White Blood Cell Count

Acknowledgments The authors thank the Lucerne Group for Consensus-Assisted Development of the Study Protocol on Prevention of Abdominal Sepsis: Example G-CSF for necessary and valuable suggestions in the Marburg Conference, January 1997. The support of H. Sitter in constructing clinical algorithms was especially appreciated. The study was supported by grant of Deutsche Forschungsgemeinschaft, Germany (Ba 1560/2-2), Amgen Inc. Members of the Lucerne sepsis study group: Bauer (Altötting), Bauhofer (Marburg), Black (Bristol), Celik (Marburg), Dietz (Delmenhorst), Duda (Mainz), Encke, Hanisch (Frankfurt), Fagniez (Paris), Farndon (Bristol), Feifel, Menger (Homburg Saar), Fingerhut (Paris), Geks (Marburg), Goris, Reemst (Nijmegen), Greger (Lichtenfels), Hay (Paris), Hesterberg (Kassel), Horeyseck (Siegburg), Izbicki, Schneider, Broelsch (Hamburg), Koller, Knobloch (Marburg), Lacaine (Paris), Lefering (Cologne), Lorenz (Marburg), Lorijn (Lucerne), Margolis, Pliskin (Beer Sheva), Neugebauer (Cologne), Nyström (Linköping), Platzer (Basle), Rothmund (Marburg), Schein (New York), Sitter (Marburg), Solomkin (Cincinatti), Stinner (Marburg), Troidl (Cologne), Van der Auwera (Basle), van Deventer (Amsterdam), Voigt (Marburg), Wendel (Konstanz), Willatts (Bristol), Wittmann (Milwaukee), Wyatt (London).

References 1. Ziegler EJ, Fisher CJJ, Sprung CL et al. Treatment of Gram-negative bacteriemia and septic shock with HA-1A human monoclonal antibody against endotoxin. N Engl J Med 1991; 324:429-436. 2. Greenman RL, Schein RMH, Martin MA et al. A controlled clinical trial of E5 murine monoclonal IgM antibody to endotoxin in the treatment of Gram-negative sepsis. JAMA 1991; 266:1097-1102.

136

Cytokines and the Abdominal Surgeon

3. Fisher CJ, Opal SM, Dhainaut J-F et al. Influence of an antitumor necrosis factor monoclonal antibody on cytokine levels in patients with sepsis. Crit Care Med 1993; 21:318-327. 4. Fisher CJ, Slotman GJ, Opal SM. Initial evaluation of human recombinant interleukin-1 receptor antagonist in the treatment of sepsis syndrome: A randomized, open-label, placebo controlled multicenter trial. Crit Care Med 1994; 22:12-21. 5. Lorenz W, Neugebauer E, Pilz G, Werdan K et al. Methodology of clinical trials in sepsis: Discussion about the trial protocol of supplemental immunoglobulin therapy in sepsis and septic shock. Theor Surg 1994; 9:10-67. 6. Bone RC. Why sepsis trials fail. JAMA 1996; 276:565-566. 7. Lorenz W, McKneally H, Troidl H, Banerjee JK et al. Surgical research around the world. In: Surgical Research: Basic Principles and Clinical Practice. Troidl H, ed. Springer, Heidelberg, New York. 1997, in press. 8. Bauhofer A, Celik I, Lorenz W et al. Clinic modelling randomised trials in animals as a new concept for designing clinical sepsis trials: Antibiotics modulate G-CSF effectiveness in severe abdominal infection in rats. J Clin Invest 1997, in preparation. 9. Lorenz W, Troidl H, Solomkin JS et al. Endoscopic surgery: Innovation versus evaluation. Theory—the second step: Testing-outcome measurements. World J Surg. 1997, submitted. 10. McPeek B, Gasko M, Mosteller F. Measuring outcome from anesthesia and operation. Theor Surg 1986; 1:2-9. 11. Lorenz W, Dick W, Junginger T et al. Induction of anaesthesia and perioperative risk: influence of antihistamine H1 + H2-prophylaxis and volume substitution with Haemaccel-35 on cardiovascular and respiratory disturbances and histamine release. Protocol of a controlled clinical trial. Theor Surg 1988; 3:55-77. 12. Encke A, Hanisch E et al. Evaluationsmodelle zur Therapieplanung / Standardisierung am Beispiel Sepsis. Langenbecks Arch Chir Sppl 1997; 2:323-329. 13. Kullberg BJ. Trends in immunotherapy of fungal infections. Eur J Clin Microbiol Infect Dis 1997; 16:51-55. 14. Baron S, Tyring SK, Fleischmann WR Jr et al. The Interferons: Mechanisms of action and clinical applications. JAMA 1991; 266(10):1375-1382. 15. Nicola NA. Haematopoetic cell growth factors and their receptors. Annu Rev Biochem 1989; 58:45-77. 16. Kullberg BJ, van der Meer JWM. Infectious diseases: The role of haematopoetic growth factors. 1996:1. 17. Billips LG, Petitte D, Dorshkind K et al. Differential roles of stromal cells, interleukin-7, and kit-ligand in the regulation of B lymphopoiesis. Blood 1992; 79:1185-1192. 18. Heyworth CM, Whetton AD, Nicholls S, Zsebo K, Dexter TM. Stem cell factor directly stimulates the development of enriched granulocyte-macrophage colony-forming cells and promotes the effects of other colony-stimulating factors. Blood 1992; 80:2230-2236. 19. Kirshenbaum AS, Goff JP, Kessler SW et al. Effect of IL-3 and stem cell factor on the appearance of human basophils and mast cells from CD34+ pluripotent progenitor cells. J Immunol 1992; 148:772-777. 20. Metcalf D. The colony-stimulating factors: Discovery to clinical use. Phil Trans R Soc Lond. 1992; 333:147-173. 21. Hestdal K, Welte K, Lie SO et al. Severe congenital neutropenia: Abnormal growth and differentiation of myeloid progenitors to granulocyte colony-stimulating factor (G-CSF) but normal response to G-CSF plus stem cell factor. Blood 1993; 82:2991-2997.

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

137

22. de Revel T, Appelbaum FR, Storb R et al. Effects of granulocyte colony-stimulating factor and stem cell factor, alone and in combination, on the mobilization of peripheral blood cells that engraft lethally irradiated dogs. Blood 1994; 83:3795-3799. 23. Emerson SG, Yang T-C, Clark SC, Long MW. Human recombinant granulocytemacrophage colony stimulating factor and interleukin 3 have overlapping but distinct hematopoietic activities. J Clin Invest 1988; 82:1282-1287. 24. Davies MG, Hagen P-O. Systemic inflammatory response syndrome. Br J Surg 1997; 84:920-935. 25. Lopez AF, To LB, Yu-Chung Yang et al. Stimulation of proliferation, differentiation and function of human cells by primate interleukin 3. Proc Natl Acad Sci USA 1987; 84:2761-2765. 26. Warringa RA, Koenderman L, Kok PT et al. Modulation and induction of eosinophil chemotaxis by granulocyte-macrophage colony-stimulating factor and interleukin-3. Blood 1991; 77:2694-2700. 27. Wendling F, Maraskovsky E, Debili N et al. cMpl ligand is a humoral regulator of megakaryocytopoiesis. Nature 1994; 369:571-574. 28. Montrucchio G, Brizzi MF, Calosso G et al. Effects of recombinant human megakaryocyte growth and development factor on platelet activation. Blood 1996; 87:2762-2768. 29. Brizzi FM, Battaglia E, Rosso A et al. Regulation of polymorphonuclear cell activation by thrombopoitin. J Clin Invest 1997; 99:1576-1584. 30. Khwaja A, Johnson B IE, Addison K et al. In vivo effects of macrophage colonystimulating factor on human monocyte function. Br J Haematol 1991; 77:25-31. 31. Cannistra SA, Koenigsmann M, DiCarlo J et al. Differentiation-associated expression of two functionally distinct classes of granulocyte-macrophage colony-stimulating factor receptors by human myeloid cells. J Biol Cem 1990; 265(21): 12656-12663. 32. Lord BI, Gurney H, Chang J et al. Haematopoetic cell kinetics in humans treated with rGM-CSF. Int J Cancer 1992; 50:26-31. 33. Weisbart RH, Kwan L, Golde DW, Gasson JC. Human GM-CSF primes neutrophils for enhanced oxidative metabolism in response to the major physiological chemoattractants. Blood 1987; 69:18-21. 34. Williams GT, Smith CA, Spooncer E et al. Haematopoetic colony stimulating factors promote cell survival by suppressing apoptosis. Nature 1990; 343:76-79. 35. Weisbart RH, Kacena A, Schuh A, Golde DW. GM-CSF induces human neutrophil IgA-mediated phagocytosis by an IgA Fc receptor activation mechanism. Nature 1988; 332:647-648. 36. Barsig JD, Bundschuh T, Hartung A et al. Control of faecal peritoneal infection in mice by colony-stimulating factors. J Infect Dis 1996; 174:790-799. 37. Echtenacher BF, Falk W, Männel DN, Kramer PH. Requirement of endogenous tumor necrosis factor/cachectin for recovery from experimental peritonitis. J Immunol 1990; 145:3762-3766. 38. Danis VA, Franic GM, Rathjen DA, Brooks PM. Effects of granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-2, interferon-gamma (IFN-gamma), tumour necrosis factor-alpha (TNF-alpha) and IL-6 on the production of immunoreactive IL-1 and TNF-alpha by human monocytes. Clin Exp Immunol 1991; 85:143-150. 39. Aman MJ, Stockdreher K, Thews A et al. Regulation of immunomodulatory functions by granulocyte-macrophage colony-stimulating factor and granulocyte colonystimulating factor in vivo. Ann Hematol 1996; 73:231-238. 40. Arnout MA, Wang EA, Clark SC, Sief CA. Human recombinant granulocyte-macrophage colony-stimulating factor increases cell-to-cell adhesion and surface expression of adhesion-promoting surface glycoproteins on mature granulocytes. J Clin Invest 1986; 78:597-601.

138

Cytokines and the Abdominal Surgeon

41. Döcke WD, Syrbe U, Meinecke A et al. Improvement of monocyte function—a new therapeutic approach? In: Reinhart K, Eyrich K, Sprung C, eds. Sepsis-Current Perspectives in Pathophysiology and Therapy. Heidelberg:Springer-Verlag. 1994:473-500. 42. Welte K, Grabrilove J, Bronchud MH et al. Filgrastim (r-metHuG-CSF): The first 10 years. Blood 1996; 88:1907-1929. 43. Fukunaga R, Ishizaka-Ikeda E, Pan C-X et al. Functional domains of the granulocyte colony-stimulating factor receptor. EMBO J 1991; 10:2855-2865. 44. Avalos BR, Gasson JC, Hedvat C et al. Human granulocyte colony-stimulating factor: biological activities and receptor characterization on haematopoetic cells and small lung cancer cell lines. Blood 1990; 75:851-857. 45. Begley CG, Lopez AF, Nicola NA et al. Purified colony-stimulating factors enhance the survival of human neutrophils and eosinophils in vitro: Rapid and sensitive microassay for colony-stimulating factors. Blood 1996; 68:162-166. 46. Lord BI, Bronchud MH, Owens S et al. The kinetics of human granulopoiesis following treatment with granulocyte colony-stimulating factor in vivo. Proc Natl Acad Sci USA 1989; 86:9499-9503. 47. Rex JH, Bhalla SC, Cohen DM et al. Protection of human polymorphonuclear leukocyte function from the deleterious effects of isolation, irradiation, and storage by interferon-gamma and granulocyte-colony-stimulating factor. Transfusion 1995; 35:605-611. 48. Lorenz W, Reimund K-P, Weitzel F et al. Granulocyte colony-stimulating factor prophylaxis before operation protects against lethal consequences of postoperative peritonitis. Surgery 1994; 116:925-934. 49. Goergen I, Hartung T, Leist M et al. Granulocyte colony-stimulating factor treatment protects rodents against lipopolysaccharide-induced toxicity via suppression of systemic tumor necrosis factor alpha. J Immunol 1994; 149:918-924. 50. Bauhofer A, Celik I, Lorenz W et al. The antibiotic regime has substantial influence on the positive effect of rhG-CSF in sepsis. Shock Supplement 1997; 7:78 (Abstract). 51. Hebert JC, O’Reilly M, Gamelli RL. Protective effect of recombinant human granulocyte colony-stimulating factor against pneumococcal infections in splenecto-mised mice. Arch Surg 1990; 125:1075-1078. 52. Nelson S. Role of granulocyte colony-stimulating factor in the immune response to acute bacterial infection in the nonneutropenic host: an overview. Clin Infect Dis 1994; 18:197-204. 53. Dale DC, Liles WC, Summer WR, Nelson S. Granulocyte colony-stimulating factor: role and relationships in infectious diseases. J Infect Dis 1994; 172:1061-1075. 54. Wunderink RG, Leeper KV, Schein RMH et al. Clinical response to filgrastim (r-metHuG-CSF) in pneumonia with severe sepsis. Am J Respir Crit Care Med 1996; 153:A123 (Abstract). 55. Lopez AF, Nicola NA, Burgess AW et al. Activation of granulocyte cytotoxic function by purified mouse colony-stimulating factors. J Immunol 1983; 131:2983-2988. 56. Silver GM, Gamelli RL, O’Reilly M. The beneficial effect of granulocyte colony stimulating factor (G-CSF) in combination with gentamicin on survival after pseudomonas burn wound infection. Surgery 1989; 106:452-456. 57. Henderson B, Wilson M. Modulins: A new class of cytokine-inducing, pro-inflammatory bacterial virulence factors. Inflamm Res 1995; 44:187-197. 58. Simms HH, D’Amico R. Granulocyte colony-stimulating factor reverses septic shockinduced polymorphonuclear leukocyte dysfunction. Surgery 1994; 115:85-93. 59. Kropec A, Lemmen SW, Grundmann HJ et al. Synergy of simultaneous administration of ofloxacin and granulocyte colony-stimulating factor in killing of Escherichia coli by human neutrophils. Infection 1995; 23:48/298-50/300.

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

139

60. Daschner FD, Grundmann H, Anding K, Lemmen S. Combined effect of human neutrophils, ceftazidime and granulocyte colony-stimulating factor on killing of Escherichia coli. Eur J Clin Microbiol Infect Dis 1995; 14:536-539. 61. Hartung T, Volk HD, Wendel A. G-CSF-an anti-infammatory cytokine. J Clin Invest 1995; 2:195-201. 62. Hartung T, Döcke W-D, Gantner F et al. Effect of granulocyte colony-stimulating factor treatment on ex vivo blood cytokine response in human volunteers. Blood 1995; 85:2492-2489. 63. Pajkrt D, Manten A, van der Poll T et al. Modulation of cytokine release and neutrophil function by granulocyte colony-stimulating factor during endotoxemia in humans. Blood 1997; 90:1415-1424. 64. Tazi A, Nioche S, Chastre J et al. Spontaneous release of granulocyte colony-stimulating factor (G-CSF) by alveolar macrophages in the course of bacterial pneumonia and sarcoidosis: Endotoxin-dependent and endotoxin-independent G-CSF release by cells recovered by bronchoalveolar lavage. Am J Respir Cell Mol Biol 1991; 4:140-147. 65. Waldrop MM. Complexity. The emerging science at the edge of order and chaos. New York: Simon & Schuster. 1992:1. 66. Pearson SD, Margolis CZ, Davis S et al. The clinical algorithm nosology: A method for comparing algorithmic guidlines. Med Dec Making 1992; 12:123-131. 67. Sitter H, Prünte H, Lorenz W. A new version of the programm ALGO for clinical algorithms. In: Brender J et al, eds. Medical information Europe ‘96. IOS Press, 1996:654-657. 68. Nyström, P.-O. Transition from contamination to infection: implications in colonic surgery. Eur J Surg 1997; 162 Suppl.576:42-46. 69. Antonelli M, Moro ML, Capelli O et al. Risk factors for early onset pneumonia in trauma patients. CHEST 1994; 105:224-228. 70. Schein M, Wittmann DH, Lorenz W. Duration of antibiotic treatment in surgical infections of the abdomen. Eur J Surg Suppl 1996; 576:1-75. 71. Fingerhut A, Terville J-P, Hay J-M, Parmentier G. Choice of antibiotics in two French hospitals. Eur J Surg 1997; 162:576:63-65. 72. Frileux P, Parc Y. Pancreatic infection. Eur J Surg 1996; 162(576):53-55. 73. Baker DM, Jones JA, Nguyen-van-Tam JS et al. Taurolidine peritoneal lavage as prophylaxis against infection after elective colorectal surgery. Br J Surg 1994; 81:1054-1056. 74. Pollock AV. Reviews on wound and peritoneal lavage: Concepts and clinical trials for decision making. II. Peritonal lavage. Theor Surg 1993; 8:103-110. 75. Schein M, Gecelter G, Freinkel W et al. Peritoneal lavage in abdominal sepsis A controlled clinical study. Arch Surg 1990; 125:1132-1135. 76. Edmiston CE, Goheen MP, Kornhall S et al. Fecal peritonitis: Microbial adherence to serosal mesothelium and resistance to peritoneal lavage. World J Surg 1990; 14:176-183. 77. Shortell SM, Zimmerman JE, Rousseau DM et al. The performance of intensive care units: does good management make a difference? Med Care 1994; 32:508-525. 78. Langhorne P, Williams BO, Gilchrist W, Howie K. Do stroke units save lives? Lancet 1993; 342:395-398. 79. Dennis M, Langhorne P. So stroke units save lives: Where do we go from here? BMJ 1994; 309:1273-1277. 80. Pollock A, Evans M. Surgical audit. Anonymous Butterworth, Heinemann, Oxford, London, Boston. 1993:1-278. 81. Kurz A, Sessler DI, Lenhardt R. The study of wound infection and temperature group. Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization. N Eng J Med 1996; 334:1209-1215.

140

Cytokines and the Abdominal Surgeon

82. Hau T, Ohmann C, Wolmershäuser A et al. Planned relaparotomy vs relaparotomy on demand in the treatment of intra-abdominal infections. Arch Surg 1995; 130:1193-1197. 83. Reimund K-P, Lorenz W, Celik I et al. Management of leucopenic sepsis. Lancet 1995; 346:382-383. 84. Friedman LM. Fundamentals of clinical trials. PSG Publishing Company Inc: Littelton, Massachusetts. 1985. 85. Cohen AM, Zsebo KM, Inoue H et al. In vivo stimulation of granulopoesis by recombinant human granulocyte colony-stimulating factor. Proc Natl Acad Sci USA 1987; 84:2484-2488. 86. Doherty GF, Ingbar SH. Stimulation by phagocytosis of the deiodination of L-thyroxine in human leukocytes. Science 1972; 176:1039-1041. 87. Solomkin JS, Cotta LA, Brodt JK, Hurst JM. Regulation of neutrophil superoxide production in sepsis. Arch Surg 1985; 120:93-98. 88. Snyderman R, Goetzl EJ. Molecular mechanisms of leukocyte chemotaxis. Science 1997; 213:830-837. 89. Sheng FC, Freischlag J, Backstrom B et al. The effect of in vivo antibiotics on neutrophil (PMN) activity in rabbits with peritonitis. J Surg Res 1987; 43:239-245. 90. Brenneis H, A Schmidt, P Blaas-Mautner et al. Chemotaxis of polymorphonuclear neutrophils (PMN) in patients suffering from recurrent infection. Eur J Clin Invest 1997; 23:693-698. 91. Künneke MB, Stinner I et al. Cardiovascular adverse effects of antibiotics in complex surgical cases. Eur J Surg suppl 1996; 576:24-28. 92. Arditi M, Kabat W, Yogev R. Antibiotic-induced bacterial killing stimulates tumor necrosis factor alpha release in whole blood. J Infect Dis 1993; 167:240-244. 93. Jackson JJ, Kropp H. Variations in beta-lactam antibiotic-induced release of endotoxin: in vivo relevance. In: Morrison DC, Ryan JL, eds. Novel therapeutic strategies in the treatment of sepsis. New York, Basel, Hong Kong: Marcel Dekker, Inc. 1996:325-347. 94. Lenhoff S, Olofsson T. Effects of immunosuppressive drugs and antibiotics on GMCSF and G-CSF secretion in vitro by monocytes, T lymphocytes and endothelial cells. Br J Haematol 1996; 95:33-38. 95. McKenna PJ, Nelson S, Andresen J. Filgratim (rhuG-CSF) enhances ciprofloxacin uptake and bactericidal activity of human neutrophils in vitro. Am J Respir Crit Care Med. 1996; 153:A535(Abstract). 96. Vogel C, Spadafora P, Horowitz B et al. Myonecrosis due to Clostridium septicum in a patient with unexplained neutropenia: successful treatment with granulocyte colony-stimulating factor. South Med J 1995; 88:765-768. 97. Mohri N, Akamo Y, Takeyama H et al. Perforated acute appendicitis in a patient with AIDS/HIV infection: report of a case. Surg Today 1995; 25:62-64. 98. Kamp Hv, Berg Ev, Timens W et al. Sweet’s syndrome in myeloid malignancy: a report of two cases. Br J Surg 1994; 86:415-417. 99. Weiss M, Gross-Weege W, Schneider M et al. Enhancement of neutrophil function by in vivo filgrastim treatment for prophylaxis of sepsis in surgical intensive care patients. J Crit Care 1994; 10:21-26. 100. Weiss M, Gross Weege W, Harms B et al. Filgrastim (RHG-CSF) related modulation of the inflammatory response in patients at risk of sepsis or with sepsis. Cytokine. 1996; 8:260-265. 101. Liang DC, Chen SH, Lean SF. Role of granulocyte colony-stimulating factor as adjunct therapy for septicemia in children with acute leukemia. Am J Hematol 1995; 48:76-81.

Hematopoietic Cytokines, G-CSF and Abdominal Surgery

141

102. Pajkrt D, Van Deventer SJ. Is G-CSF safe and useful in the treatment of infectious diseases in the non-neutropenic host? Intensive Care Med 1995; 23:1-2. 103. Gough AM, Clapperton N, Rolando AVM et al. Randomised placebo controled trial of granulocyte-colony stimulating factor in diabetic foot infection. Lancet 1997; 350:855-859. 104. Hommes DW, Meenan J, Dijkhuizen S et al. Efficacy of recombinant granulocyte colony-stimulating factor (rhG-CSF) in experimental colitis. Clin Exp Immunol 1996; 106:529-533. 105. Frumkin LR. Role of granulocyte colony stimulating factor and granulocyte-macrophage colony-stimulating factor in the treatment of patients with HIV infection. Curr Opin Hematol 1997; 4:200-206. 106. Pitrak DL, Tsai HC, Mullane KM et al. Accelerating neutrophil apoptosis in the aquired immunodeficency syndrome. J Clin Invest 1996; 98:2714-2719. 107. Lorenz W, Doenicke A. H1 and H2 blockade: A prophylactic principle in anaesthesia and surgery against histamine-release responses of any degree of severity: part 2. New Engl Reg Allergy Proc 1997; 6:37-57. 108. Lorenz W, Duda D, Dick W et al, and the trial group Mainz/Marburg. Incidence and clinical importance of perioperative histamine release: Randomised study of volume loading and antihistamines after induction of anaesthesia. Lancet 1994; 343:933-940. 109. Natanson C, Hoffman WD, Suffredini AF et al. Selected treatment strategies for septic shock based on proposed mechanisms of pathogenesis. Ann Intern Med 1994; 120:771-783. 110. Ibelgaufts H. Dictionary of cytokines. Editiones Roche, Basel, Switzerland. 1995. 111. Cheadle WG, Mercer Jones M, Heinzelmann M, Polk Jr HC. Sepsis and septic complications in the surgical patient: Who is at risk? Shock 1996; 6(Suppl)1:S6-9. 112. Link H, Boogaerts MA, Carella AM et al. A controlled trial of recombinant human granulocyte-macrophage colony-stimulating factor after total body irradiation, highdose chemotherapy, and autologous bone marrow transplantation for acute lymphoblastic leukemia or malignant lymphoma. Blood 1992; 80:2188-2195. 113. Windsor JA. Underweight patients and the risks of major surgery. World J Surg 1993; 17:165-172. 114. Negro G, Iannuzzi C, Terracciano CA et al. Evaluation of the surgical risk in general surgery: usefulness of a predictive system based on statistical analysis. Ann Ital Chir 1992; 63:123-126. 115. Delamaire M, Maugendre D, Moreno M et al. Impaired leucocyte functions in diabetic patients. Diabetic Med 1997; 14:29-34. 116. Yokoyama T, Kodama T, Takesue Y et al. Perioperative managements for postoperative severe infections in compromised host. Nippon Geka Gakkai Zasshi 1996; 97:1060-1065. 117. Morgan AS, Connecticut H. Risk factors for infection in the trauma patient. J Natl Med Assoc 1992; 84:1019-1023. 118. McLauchlan GJ, Anderson ID, Grant IS, Fearon KCH. Outcome of patients with abdominal sepsis treated in an intensive care unit. Br J Surg 82:524-529.

Cytokines and Spontaneous Bacterial Peritonitis

143

CHAPTER 12

Cytokines and Spontaneous Bacterial Peritonitis Theresa Propst and Albert Propst

Introduction

C

ytokines are polypeptides that possess a wide spectrum of inflammatory, metabolic, and immunologic regulatory properties.1,2 The liver represents an important site of synthesis and at the same time the major clearance organ for several cytokines.3 Consequently, several pathological phenomena, which are mediated by inflammatory cytokines such as fever, malaise, cachexia, and cholestasis are observed in chronic liver disease.4 By their capacity to control fibrogenesis, cytokines may also play an important role in the development of cirrhotic transformation of the liver.5 The present article is focused on cytokine levels in chronic liver disease, especially of cytokine levels in the ascitic fluid of patients with cirrhosis together with evidence of spontaneous bacterial peritonitis, which is a serious complication of advanced chronic liver disease with portal hypertension.

Pathogenesis Spontaneous bacterial peritonitis (SBP) is a major problem encountered in patients with liver cirrhosis, with an incidence of 5-30% of all infections and with an overall incidence of 60-75% of all severe infections in patients with chronic liver disease.6-9 Hospital mortality constitutes about 50% of all SBP cases. Pathogenesis of SBP appears multifactorial and can be explained as resulting from derangements of one or more components of the host defense system. Over the last few years the pathogenesis of SBP has gradually become clearer. To understand the immunological reactions and cytokine production in ascitic fluid (AF) of patients with cirrhosis, some comments on the pathogenesis of SBP need to be made. Patients with advanced chronic liver disease are incompetent in protein synthesis and an AF protein concentration < 1g/dl (< 10 g/l) is well known to be a risk factor for SBP development. The reason for this is, at least in part, the close correlation of the opsonic activity (endogenous antimicrobial activity) of AF with its total protein concentration. Patients with detectable opsonic activity are protected from SBP, whereas in those patients with undetectable opsonic activity, SBP develops within a few days. There is a positive correlation between opsonic activity and AF complement C3-concentration. The more dilute the ascitic fluid, the more dilute the opsonic proteins. An AF protein content <1.0 g/dl (< 10 g/l) or an AF complement C3-concentration < 20 mg/dl Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

144

Cytokines and the Abdominal Surgeon

Fig. 12.1. Pathogenetic model of spontaneous bacterial peritonitis development in patients with chronic liver disease and portal hypertension. sbp= spontaneous bacterial peritonitis, res= reticuloendothelial system.

(< 0.2 g/l) has no bactericidal activity.10 In summary, AF opsonic activity correlates with AF total protein concentration and with AF complement C3-concentration, and patients whose AF is dilute and deficient in these parameters are predisposed to SBP. However, additional risk factors for SBP development have also been defined (Fig. 12.1). The enteric nature of most organisms which cause SBP implicate the gut as the source of infection. Transmural migration of bacteria in patients with cirrhosis-associated portal hypertension occurs from the gut lumen across the mucosa and submucosal lymphatics into AF. Circumstances which promote translocation include bacterial overgrowth in the gut, portal hypertension and gastrointestinal bleeding in cirrhosis. It is also known that shunting of portal blood away from reticuloendothelial cells and the inability of the cirrhotic patients reticuloendothelial system (RES) to clear particulate matter from the blood is correlated with an SBP infection-related death. The ability of some organisms but not others to pass from the gut to the AF implies that a different kind of filter mechanism may exist between the gut and the AF in cirrhosis. The filter appears to be at the levels of the mucosa and the mesenteric lymph nodes.11 Furthermore, patients with cirrhosis have altered gut flora which also could promote translocation. Increased mucosal permeability in patients with cirrhosis facilitates translocation of bacteria from the gut to mesenteric lymph nodes and AF. In addition, bacteremia is common in patients with SBP (>50%). Over the last few years there have been discussions as to whether AF peritoneal monocytes, macrophages, peritoneal polymorphonuclear cells (PMN) or inhibitory immunoregulatory substances can be defined as local causes of compromised immune reaction in patients with SBP. Serum neutrophil dysfunction is quite common in patients with cirrhosis and such a defect in host defense against infection can be expected to lead to frequent and prolonged episodes of bacteremia. Furthermore, the entry of PMNs into the AF is an indicator of failure of local opsonins in combination with peritoneal macrophages to control the infection. Patients with advanced chronic liver disease can therefore be defined as presenting a more advanced stage of immunocompromise.

Cytokines and Spontaneous Bacterial Peritonitis

145

Table 12.1. Pathogenic strains and frequency of aerobic and anaerobic bacteria in spontaneous bacterial peritonitis (SBP) Aerobic Bacteria Negative Gram Stain E. coli Citrobacter Klebsiella Proteus Enterobacter Positive Gram Stain Streptococci Group D Viridans Pneumonia Staphylococcus aureus Anaerobic Bacteria Bacteroides, Clostridium, Lactobacillus

55% 5% 5% 3% 3%

13% 10% 3% 1% 2%

Diagnosis Awareness of the full spectrum of SBP is essential in avoiding its underdiagnosis. The diagnosis of SBP is based on the analysis of AF. Abdominal paracentesis and proper AF analysis are required before a diagnosis of SBP can be made. In patients with refractory ascites the presence of ascitic infection must be excluded. Prior to the 1980s abdominal paracentesis was not performed regularly because of fear of complications of the procedure. Now that the complication rate is known to be <1%, in many Liver Units routine paracentesis is performed at the time of admission of patients with decompensated liver cirrhosis.12 Diagnosis of SBP is defined as PMN >250 cells/mm3 with a positive AF culture.13 Conventional cultures have been found to detect bacterial growth in 42-43% of samples of neutrocytic ascites, whereas bedside inoculation of blood culture bottles with AF detects bacterial growth in 93%. SBP is essentially always a monomicrobial infection with a low colony count very similar to bacteremia. Culturing AF as if it were blood would be predicted to be superior to culturing the fluid as if it were urine or stool. In fact, prospective studies demonstrated the superiority of the blood culture bottle technique to the conventional method.14 Most episodes of SBP are caused by E. coli, Streptococci and Klebsiella (Table 12.1). Although AF Gram stains are frequently ordered, they are seldom helpful. A Gram stain of AF for detection of SBP is analogous to a Gram stain of blood in bacteremia; it is only positive in the presence of an overwhelming infection. In the two largest and most recent studies, determination of pH and the lactate content of AF were not found to be helpful as they are too insensitive, nonspecific and in addition expensive.15

Symptoms The symptoms and signs of infection may be very subtle and easily misinterpreted. Minor changes in mental status, that would only be detected by family members or a physician who is very close to the patient, may be the only clinical evidence

146

Cytokines and the Abdominal Surgeon

of infection. SBP is a syndrome ranging in severity from being totally asymptomatic to being fulminant, characterized by fever, hypotension, abdominal pain and the development or worsening of encephalopathy (Table 12.2). Symptoms alone may not be helpful in differentiating SBP from gastrointestinal tract perforation since they may be subtle or absent in both conditions. Patients with previous hepatic encephalopathy have an increased incidence of SBP.16 Once SBP is considered, paracentesis is indicated to obtain a definitive diagnosis. Since it is a relatively safe procedure and there is no evidence that it predisposes patients to SBP, diagnostic paracentesis should be performed at the onset of ascites or in patients with known ascites who develop new clinical symptoms such as fever or general deterioration.

Differential Diagnosis Differentiating SBP from secondary peritonitis is essential because the appropriate treatment of the latter is surgical; when laparotomy is carried out in a patient with undiagnosed SBP, operative mortality is as high as 80%.17 Signs and symptoms are not helpful in differentiating secondary peritonitis from SBP. Nevertheless, patients with perforation peritonitis fulfil at least two of the following chemical criteria: 1) AF total protein level of >1 g/dl; 2) a glucose level of <50 mg/dl; and 3) a lactate dehydrogenase level of more than 225 mU/ml.18 Polymicrobial infection is unusual in SBP but is universal in peritonitis caused by perforation. For differentiation of SBP from secondary peritonitis (not associated with perforation) the response of the AF PMN count to antibiotic therapy may be helpful as 48 h after appropriate treatment, the concentration of AF neutrophils is below the pretreatment value in all episodes of SBP, but in only two-thirds of patients with secondary peritonitis.18

Cytokines in Chronic Liver Disease Most published data on cytokine levels in liver disease focus on alcoholic liver disease. Khoruts et al reported elevated plasma levels of tumor necrosis factor-alpha (TNF-!), interleukin-1-alpha (IL-1a), and interleukin-6 (IL-6) in 50% of patients with stable alcoholic liver cirrhosis.19 In another more recent study, serum levels of IL-1b, IL-6, TNF-!, interferon gamma (IFN-∀), and C-reactive protein (CRP) were investigated in 264 patients with chronic liver diseases of different etiologies.20 This study showed that serum levels of IL-1b, IL-6, TNF-!, IFN-∀ were significantly elevated in patients with chronic liver disease, that endogenous cytokine patterns were stage-dependent and that these changes were independent of the etiology of the underlying liver disease.20 Levels of IL-1b, IL-6 and TNF-! in patients with cirrhosis were significantly higher than in noncirrhotic patients; however, no correlation between the inflammatory disease nature of chronic liver disease and endogenous cytokine levels were found. The significantly higher serum levels of proinflammatory cytokines in patients with cirrhosis might reflect increased lipopolysaccharide-induced macrophage stimulation or impaired clearance. Therefore, the phenomenon of enhanced endogenous cytokines in chronic liver disease may be a consequence of chronic liver failure leading to enhanced endogenous lipopolysaccharide levels and decreased cytokine clearance. In patients with decompensated cirrhosis, IL-6 plasma levels were found to be significantly different between infected and noninfected patients.21 Levels above 200 pg/ml were always found in infected patients giving a sensitivity of 100% and a specificity of 74%. TNF-! plasma levels were less sensitive (95%) and specific (68%) for the diagnosis of bacterial infection at a threshold of 50 pg/ml, but were more closely related to a poor patient outcome.21

Cytokines and Spontaneous Bacterial Peritonitis

147

Table 12.2. Frequency of symptoms in spontaneous bacterial peritonitis (SBP). Symptoms

Frequency (%)

Fever Chills Abdominal pain Decreased bowel sounds Hypotension Encephalopathy

54% 14% 51% 21% 5% 51%

Cytokines in SBP Cytokines are well known to be crucially involved in the regulation of immune and phagocytic reactions and also act as the central mediators of the inflammation response. Elevated IL-6 levels have been detected so far in the serum of patients with bacterial infections and septic shock caused mainly by Gram-negative bacteria.22 The literature reviewed strongly supports the notion that bacterial peritonitis is associated with a significant and mainly compartmentalized peritoneal cytokine response that reflects the severity of the disease and its prognosis. The dynamics of cytokines in the plasma and peritoneal fluid were determined simultaneously in patients with cirrhosis and SBP and primary peritonitis complicating chronic ambulatory peritoneal dialysis (CAPD).23-26 Peritoneal mesothelial cells and peritoneal macrophages, collected from patients undergoing CAPD during an episode of peritonitis, secreted increased amounts of IL-1, IL-6, TNF-! and IL-8.23,24 In a recent study significantly enhanced levels of IL-6 and its secondary mediators alpha-1-antitrypsin (AAT) and C-reactive protein (CRP) in the AF of patients with SBP were found.25 Significantly higher IL-6, TNF-! and neopterin levels were also found in the AF of patients with cirrhosis who later died of septic complications, which indicates the prognostic value of these cytokines.25 No differences could be detected in the AF levels of interleukin-2, interleukin-2 receptor and INF-∀ of cirrhotic patients with and without SBP.25 Indeed, measurement of IL-6 does not appear to be practical as a screening test for SBP, since the measurement of PMN count is simpler and more widely available. However, determination of IL-6, TNF-! and neopterin levels may be good prognostic parameters in patients with severe and life threatening infection.25 Furthermore, peritoneal levels of IL-6 and TNF-! decreased within the first 48 hours of effective antibiotic therapy.26 In all performed studies, peritoneal levels of cytokines in peritonitis were found to be much higher than systemic levels, indicating a compartmentalized inflammatory process and suggesting that cytokines do not equilibrate readily between the peritoneal cavity and plasma.25-27 The largest fraction of peritoneal cytokines probably derives from macrophages after they contact bacteria and their byproducts. Furthermore, damage to the intestinal barrier may allow the translocation of luminal endotoxin, which is frequently found in ascites, or even of certain cytokines into the portal and lymphatic circulation.28 Although most clinical studies hitherto measured circulatory cytokine levels, it appears that local peritoneal cytokine concentrations,

148

Cytokines and the Abdominal Surgeon

exerting their effects in a paracrine fashion, are more likely to be of primary biologic and clinical importance. Elevated cytokine and acute phase protein levels as well as enhanced neopterin and GM-CSF levels found in the AF of SBP patients in one of the above mentioned studies further suggests that liver disease patients do mount an immunological response to infection in the AF compartment.25 The ratio of IL-6 (ascites) to IL-6 (serum) is not determined by a decreased hepatic cytokine clearance but indicates rather that IL-6 is produced locally in the peritoneal cavity. The continous synthesis of IL-6 in ascites and the elevated levels of several other cytokines in case of infection suggest that cells of the immune system, as e.g., periteoneal macrophages, might be the primary source of cytokine release. The fact that there was no difference of IL-1 concentrations in serum and ascites, with and without SBP, can be explained as being due to the inhibition of IL-1 production by the high concentrations of IL-6.29 As cytokine levels of AF, with the exception of IL-1 and INF-∀, were found to be significantly higher than serum concentrations, impairment of peritoneal macrophages in patients with chronic liver disease may not be causally related to the development of SBP. However, further investigations on peritoneal macrophage function in patients with chronic liver disease and SBP are required to prove this hypothesis. In patients with advanced chronic liver disease, a reduced chemotactic effect of serum was found.30 Furthermore, the chemotactic effect of AF is also considered to be lower than that of serum and is proportional to the local complement C3-concentrations. Chemotaxis-inhibiting factors were also identified in AF of patients with chronic liver disease, their pathogenetic relevance being limited.30 In a recent study, the function of normal PMN cells in the AF of patients with cirrhosis was assessed by a chemilluminescence method using preopsonized zymosan as stimulus and evidence of the presence of suppressive factors was provided by AF dilution.31 In addition, there was a deficiency in supportive factors other than complement C3.31 The impaired production of oxidative metabolites by PMN explains partly the high susceptibility of cirrhotic patients to SBP independently of complement C3 levels.31 In another study, plasma complement activation via the classic pathway, indicated by C4d/C4, was shown to be significantly increased in patients with SBP compared with uninfected patients.32 In addition, there was no difference between these two groups in ascitic complement activation of either the classical or alternative pathways.32 In SBP, decreased plasma C3 and C4 are primarily caused by increased activation of the classical pathway and not by an impaired hepatic synthesis. Activation and consumption of C3 is one factor causing the low ascitic C3 concentrations observed in SBP.32 In conclusion, the susceptibility of cirrhotic patients to SBP is due to impaired PMN oxidative metabolites and inhibitory intracellular and extracellular factors in addition to low complement C3 levels modulating PMN functions.31 The in vitro effect of GM-CSF on phagocytosis, phagocytic index and chemotaxis of AF PMN cells from cirrhotic patients support this view of immunodeficiency, as after GM-CSF stimulation, the percentage of phagocytic PMNs and chemotaxis in infected patients was significantly higher.33 The high IL-6 (>10 000 pg/ml) and TNF-! levels (>100pg/ml) found in SBP are due to local cytokine production from peritoneal macrophages. The increased levels of the secondary mediators such as AAT, CRP and neopterin as well as complement activation via the classical pathway strongly suggest that patients with decompensated liver disease do mount an immunological response in the peritoneal cavity.

Cytokines and Spontaneous Bacterial Peritonitis

149

Conclusion The compartmentalization of the cytokine cascades in SBP fits into the concept that the circulating systemic concentration of cytokines may be misleading and fail to reflect their tissue concentration or local biologic activity. The use of cytokine serum concentrations in the form of a cytokine scoring in chronic liver disease has been frustrated by the fact that circulatory concentrations of the free bioactive cytokines may be negligible, yet significant amounts of cytokine are present in the tissue level such as the peritoneal cavity. Therefore in outcome prediction local estimation of cytokines may better reflect the severity of an initially local process such as SBP. For this reason, measurements of IL-6 and TNF-! in AF may represent useful markers both for the diagnosis and prognosis of SBP and also for monitoring the treatment of SBP in cirrhotic patients.

References 1. Arai K, Lee F, Miyajima A et al. Cytokines: coordinators of immune and inflammatory responses. Ann Rev Biochem 1990; 59:783-836. 2. Fujita T, Reis LFL, Watanabe N et al. Induction of the transcription factor IRF-1 and interferon-# mRNAs by cytokines and activators of second-messenger pathways. Proc Natl Acad Sci 1989; 86:9936-9940. 3. Andus T, Bauer J, Gerok W. Effects of cytokines on the liver. Hepatology 1991; 13:364-375. 4. Thiele DL. Tumor necrosis factor: The acute phase response and the pathogenesis of alcoholic liver disease. Hepatology 1989; 9:497-499. 5. Kovacs EJ. Fibrogenic cytokines: The role of immune mediators in the development of scar tissue. Immunol Today 1991; 12:17-23. 6. Hoefs JC, Runyon BA. Spontaneous bacterial peritonitis. Dis Mon 1985; 31:1-48. 7. Rimola A, Soto R, Bory F et al. Reticuloendothelial system phagocytic activity in cirrhosis and its relationship to bacterial infections and prognosis. Hepatology 1984; 4:53-58. 8. Wyke RJ. Problems of bacterial infection in patients with liver disease. Gut 1987; 28:623-641. 9. Felisart J, Rimola A, Arroyo V et al. Cefotaxime is more effective than is ampicillin-tobramycin in cirrhotics with severe infections. Hepatology 1985; 5:457-462. 10. Runyon BA. Low protein-concentration ascitic fluid is predisposed to spontaneous bacterial peritonitis. Gastroenterology 1986; 91:1343-1346. 11. Berg RD, Garlington AW. Translocation of certain indigenous bacteria from the gastrointestinal tract to the mesenteric lymph nodes and other organs in a gnotobiotic mouse model. Infect Immun 1979; 23:403-411. 12. Runyon BA. Paracentesis of ascitic fluid: A safe procedure. Arch Intern Med 1986; 146:2259-2261. 13. Runyon BA, Hoefs JC. Ascitic fluid analysis before, during and after spontaneous bacterial peritonitis. Hepatology 1985; 5:257-259. 14. Siersema PD, de Marie S, van Zeijl JH et al. Blood culture bottles are superior to lysis-centrifugation tubes of bacteriological diagnosis of spontaneous bacterial peritonitis. J Clin Microbiol 1992; 30:667-669. 15. Runyon BA, Antillon MR. Ascitic fluid pH and lactate: Insensitive and nonspecific tests in detecting ascitic fluid infection. Hepatology 1991;13:929-935. 16. Almidal TP, Skinhoej P. Spontaneous bacterial peritonitis: Incidence, diagnosis and prognosis. Scand J Gastroenterol 1987; 22:295-300. 17. Garrison RN, Cryer HM, Howard DA et al. Clarification of risk factors for abdominal operations in patients with hepatic cirrhosis. Ann Surg 1984; 199:648-655.

150

Cytokines and the Abdominal Surgeon

18. Akriviadis EA, Runyon BA. The value of algorithm in differentiating spontaneous from secondary bacterial peritonitis. Gastroenterology 1990; 98:127-133. 19. Khoruts A, Stahnke L, McClain CJ et al. Circulating tumor necrosis factor, interleukin-1 and interleukin-6 concentrations in chronic alcoholic patients. Hepatology 1991; 13:267-276. 20. Tilg H, Wilmer A, Vogel W et al. Serum levels of cytokines in chronic liver diseases. Gastroenterology 1992; 103:264-274. 21. Le-Moine O, Deviere J, Devaster JM et al. Interleukin-6: An early marker of bacterial infection in decompensated cirrhosis. J Hepatol 1994; 20:819-824. 22. Calandra T, Gerain J, Heumann D et al. High circulating levels of interleukin-6 in patients with septic shock: Evolution during sepsis, prognostic value and interplay with other cytokines. Am J Med 1991; 91:23-29. 23. Betjes MG, Tuk CW, Struijk DG et al. Interleukin-8 production by human peritoneal mesothelial cells in response to tumor necrosis factor-alpha, and interleukin conditioned by macrophages cocultured with staphylococcus epidermidis. J Infect Dis 1993; 168:1202-1210. 24. Topley N, Jorres A, Luttmann E. Human peritoneal mesothelial cells synthesize interleukine-6: Induction by IL-1 beta and TNF alpha. Kidney Int 1993; 43:226-233. 25. Propst T, Propst A, Herold M et al. Spontaneous bacterial peritonitis is associated with high levels of interleukin-6 and its secondary mediators in ascitic fluid. Eur J Clin Invest 1993; 23:832-836. 26. Zeni F, Tardy B, Vindimian M et al. High levels of tumor necrosis factor-alpha and interleukin-6 in the ascitic fluid of cirrhotic patients with spontaneous bacterial peritonitis. Clin Infect Dis 1993; 17:218-223. 27. Pruimboom WM, Bac DJ, van-Dijk AP et al. Levels of soluble intercellular adhesion molecule 1, eicosanoids and cytokines in ascites of patients with liver cirrhosis, peritoneal cancer and spontaneous bacterial peritonitis. Int J Immuno-pharmacol 1995; 17:375-384. 28. Deitch EA. Cytokines yes, cytokines no, cytokines maybe? Crit Care Med 1993; 21:817-819. 29. Schindler R, Mancilla J, Endres S et al. Correlation and interactions in the production of interleukin-6 (IL-6), interleukin-1 (IL-1) and tumor necrosis factor (TNF) in human blood mononuclear cells: Il-6 supresses IL-1 and TNF. Blood 1990; 75:40-47. 30. Horing E, Otto D, Von-Gaisberg U. Influence of ascites on the chemotaxis of granulocytes in patients with cirrhosis. J Gastroenterol-Hepatol 1995; 10:186-191. 31. Lebrun L, Pelletier G, Briantais MJ et al. Impaired functions of normal peripheral polymorphonuclear leukocytes in cirrhotic ascitic fluid. J Hepatol 1992; 16:98-101. 32. Bird G, Senaldi G, Panos M et al. Activation of the classical complement pathway in spontaneous bacterial peritonitis. Gut 1992; 33:307-311. 33. Garcia-Gonzalez M, Boixeda D, Herrero D et al. Effect of granulocyte-macrophage colony-stimulating factor on leukocyte function in cirrhosis. Gastroenterology 1993; 105:527-531.

Secondary Peritonitis and Cytokines

151

CHAPTER 13

Secondary Peritonitis and Cytokines Réne G. Holzheimer

Introduction

P

eritonitis is still a life threatening complication for every surgical patient, which can occur after trauma, intra-abdominal operations or spontaneously.1 Depending on localization or cause the mortality rate of peritonitis is between 40-70% according to literature. The principles of peritonitis treatment established by Kirschner, e.g., source control, debridement and lavage, are still valid.2-4 Even the development of new, potent antibiotics within the last twenty years did not dramatically reduce the mortality rate of peritonitis. Intensive care therapy has influenced the survival of peritonitis patients. However, the treatment was focused on symptomatic treatment of organ dysfunction.5 With the identification and recombinant development of cytokines the possibility to influence the course of peritonitis before organ dysfunction seem to available soon. Our knowledge of the pathophysiological immune response during peritonitis has been enlarged during the last 10 years. However, a uniform, efficient immunological concept of peritonitis treatment is not yet available. The purpose of this chapter is to summarize the main developments in cytokines in secondary peritonitis with regard to experimental and clinical studies.

Peritonitis and Cytokines in Experimental Models Endotoxin which is released from cell walls of disintegrating gram-negative pathogens is a major trigger for cytokine release in peritonitis, although it may not be the only cause for macrophage cytokine production.6 Peritonitis results in protease activation and protease inhibitor consumption, especially in the peritoneal fluid.7 This may lead to a breakdown of C3 complement and IgG in peritonitis exudate.8 Bacterial components other than endotoxin may induce dysfunction in the peritoneal macrophages capacity to produce proinflammatory cytokines during sepsis and peritonitis.9,10 However, endotoxin can be frequently detected in the peritoneal exudate and plasma of patients with peritonitis.11,12 The activation of macrophages in peritonitis leads to a complex release of cytokines.13 This activation occurs via specific LPS receptors, e.g., CD14, CD11/18 family, LPS receptor, scavenger receptor. The identification of specific cell membrane targets for LPS has important implications for immunotherapy.14,15 Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

152

Cytokines and the Abdominal Surgeon

It has been demonstrated that the infusion of either LPS or TNF can mimic the effects of sepsis.16 Proinflammatory cytokine levels correlated with the prognosis of peritonitis and treatment failure. Reduced TNF plasma levels correlated with increased survival;17 constantly increased peritoneal levels correlated with poor outcome.18 The demonstration of detrimental TNF effects lead to the concept to block endotoxin or TNF. However, there was early evidence that I.V. anti-TNF antibodies decreased mortality only in I.V. sepsis and not in peritonitis.19 Anti-TNF-antibodies reduced IL-1 and IL-6 plasma levels in I.V. sepsis and not in peritonitis. Anti-LPS-antibodies, however, were able to induce protection in peritonitis and to reduce pathogens, plasma TNF, IL-1 and IL-6 levels. The peritoneal cytokine levels remained unchanged.20 Conversely, peritoneal cytokine production correlates with outcome in peritonitis and pretreatment with monophoryl lipid A as well as induction of endotoxin tolerance may reduce cytokine production.21

Dosage of Cytokines The effect of cytokines on the immune response—beneficial or detrimental— may depend on the dose and the concentration of cytokines.22 Anti-TNF-antibodies administration after CLP induction increased the mortality, while the addition of TNF increased the survival rate.23 Interferon-! increased TNF and IL-6 plasma levels and subsequent mortality. Blocking interferon-! resulted in increased survival without any effect on the cytokine levels.24 It seems to be likely that Interferon-! increases sensitivity of cells to endotoxin.25

Function of Peritoneal Cells Peritoneal macrophages may change their function during the course of peritonitis and therefore may influence the immune modulation. While antigen recognition is the main function during the early phase, cytokine production prevails during later stages of peritonitis. TNF plasma levels may be lower during peritonitis compared to I.V. induced sepsis.26 Anti-TNF-antibodies reduced migration of granulocytes and monocytes into the peritoneum with increased number of pathogens.27 Inflammatory cytokines may induce adhesion and fibrinogen production by stimulation of plasminogen activation inhibitor in mesothelial cells.28,29 The mortality decreasing effect of Pentoxifyllin which is known to decrease TNF production in peritonitis may be due to reduced adhesion and fibrinogen production.30 Peritoneal macrophages and mesothelial cells are responsible for initiation, amplification and termination of the inflammatory response. The surface of mesothelial cells, where bacteria colonize, may be an important place of interaction between macrophages and mesothelial cells. 31 Microbial colonization of the peritoneal mesothelial surface is a rapid and stable phenomena following penetration injury to the distal bowel. Mesothelial populations are resistant to intraperitoneal lavage.32 S. aureus adherence to mesothelial cells is increased following preincubation of mesothelial cells with IL-1. Treating mesothelial cells with IFN-! reduces adherence of S. aureus.33 Other cell population in the peritoneum have not been studied until recently. Mast cells in the vicinity of blood vessels may be important for the synthesis of leukotrienes which are responsible for PMN recruitment.34,35

Local Response Local TNF production is an important factor for tissue damage and organ dysfunction. Peritoneal macrophage TNF mRNA increased after CLP and i.p. LPS injection; however, TNF mRNA decreased faster after i.p. LPS injection. Both forms of

Secondary Peritonitis and Cytokines

153

infection are able to induce TNF production in lungs and liver.36 Bacterial infection of the peritoneal cavity may induce a slow release of cytokines which are important for the local immune response. IL-1 may induce the lethal effects of sepsis and TNF may be more important to launch the local immune response.37 Increased local, intestinal IL-6 production has been demonstrated after CLP38 and TNF levels were higher in portal vein than in hepatic veins.39 Within encapsulated abscess LPS binding protein (LBP) and bactercidal/permeability increasing protein (BPI) have been measured. LBP binds to the lipid A component of the bacterial endotoxin and facilitates its delivery to the CD14 antigen on the macrophage where inflammatory cytokines are released. The neutrophil granular protein bactericidal/permeability increasing protein (BPI) competes with LBP for endotoxin binding and functions as a molecular antagonist of LBP-endotoxin interactions. Within abscess cavities BPI is available in sufficient quantities for effective competition with LBP for endotoxin. BPI may attenuate the local inflammatory response and the systemic toxicity of endotoxin release during gram-negative infections.40 Endotoxin derived from enteric bacteria might play an important role in the pathogenesis of lung injury and anti-endotoxin agents, such VVN1 222-5 appear to protect against endogenous bacterial endotoxin related disorders in severe hemorrhagic shock.41,42

Effect of Cytokines and Growth Factors in Peritonitis Different results were reported when cytokines or growth factors were added in peritonitis. G-CSF increased survival rate in peritonitis, probably by reducing TNF levels.43-45 High dose G-CSF decreased endotoxin and TNF, increased peripheral neutrophils and improved cardiopulmonary function.46 Conversely, macrophages, which produce G-CSF, were not pivotal after bacterial translocation or septic shock in the knock-out mouse model.47 GM-CSF, which was administered after the onset of peritonitis, was not beneficial and inhibited the neutrophil migration into the peritoneal cavity.48 IL-1, an inflammatory cytokine, downregulated TNF and IL-6 plasma levels, decreased organ dysfunction and mortality when added before the induction of peritonitis.49 This is supported by the finding that blockade of Kupffer cells decreased IL-1 and survival.50 IL-2, which is known to be essential for immune response after thermal injury and CLP51 induced influx of neutrophils in peritonitis and increased survival.52 IL-2 administration and induction of peritonitis should be performed simultaneously to achieve a protection by IL-2.53

Anti-Endotoxin-Antibodies The notion that anti-endotoxin-antibodies may be beneficial in peritonitis by reduction of cytokines54 is further supported by other studies. Pretreatment with anti-endotoxin-antibodies were protective in peritonitis and reduced plasma TNF levels and splenocyte TNF production.55 E5 monoclonal antibodies reduced mortality, endotoxin and TNF in peritonitis, but not endothelin.56 Endotoxin neutralizing protein (ENP) decreased endotoxin and TNF; however, mortality was only reduced when ENP was added together with gentamicin.57 Further studies with anti-endotoxin antibodies were published with controversial results. Some of them, however, revealed important pathophysiological mechanisms of the function of anti-endotoxin-antibodies. Type specific anti-endotoxinantibodies were protective in peritonitis by Fc-mediated clearance of both bacteria and endotoxin.58

154

Cytokines and the Abdominal Surgeon

IgG and IgM anti-LPS mAbs exert protective capacity by extracellular neutralization of LPS, while Fc-receptor mediated cellular uptake also may serve to bypass macrophage activation and TNF secretion by promoting internalization and intracellular neutralization.59 While some recent reports demonstrated significant activity of antibodies, not all anti-endotoxin antibodies have been demonstrated to bind to endotoxin or to be beneficial in peritonitis. Selected models may have a clear influence on the results.60-63

Pathogens Treatment failures of intra-abdominal infection may be due in part to the presence of resistant pathogens at the site of infection.64 Enterococcus plays an important role in the mechanisms of bacterial synergism in experimental peritonitis.65,66 The LPS induced cytokine immune response and the bacterial surface characteristics may be more important for the killing of invading pathogens than previously thought.67 In vitro studies have revealed that antibiotics may release different amounts of endotoxin depending on the type of Penicillin binding protein.68 There may be important functional relationships between the immune response and resistant pathogens not yet clarified.69,70

Anti-Inflammatory Cytokines There is a growing body of information available on the onset of inflammatory response in secondary peritonitis. However, there is less information available on the termination of inflammation and the role of anti-inflammatory cytokines. It was generally believed that the anti-inflammatory response is launched after the inflammatory response. However, we and others have demonstrated that both inflammatory and anti-inflammatory cytokines are released simultaneously during the inflammatory response.71 Anti-inflammatory cytokines seem to suppress inflammatory cytokines. IL-10 administration prolonged survival in septic mice72 and reduced mortality in severe peritonitis;73 anti-IL-10-antibodies given before CLP increased mortality rate.74 Pretreatment with anti-IL-10 increased plasma TNF levels, whereas IL-1 and IFN-! could not be detected. IL-10 mRNA was observed in liver, spleen and lungs after CLP. The increased mortality rate with anti-IL-10 pretreatment could not be influenced by anti-TNF-antibodies.75 TNF exerts its effects by two cell surface receptors, TNF R I and II, also referred to as p55 and p75 receptors, respectively. TNF-R are transmembrane proteins which on cleavage of their extracellular domains result in the release of soluble fragments sTNF-R. sTNF-R increases markedly during infection and may serve to modulate TNF bioactivity. In endotoxin sensitive and resistant mice it was demonstrated that upon infection with LPS or live gram-negative bacteria there may be 2 separately regulated pathways that control sTNF-R shedding. Peritoneal macrophages of endotoxin sensitive mice responded to LPS stimulation; in contrast macrophages of resistant mice showed only a modest response.76 Endotoxin induces downmodulation of monocyte and granulocyte TNF surface receptors in humans in vivo which may represent a mechanism to reduce excessive activity of TNF during systemic infection.77

Second Hit Trauma may prime macrophages in such a way that a second hit to the immune system by infection or sepsis may lead to inadequate immune response. Trauma induces changes in endotoxin kinetics and PMN function in a model of trauma and posttraumatic peritonitis.78 Synergism between trauma and infection was observed

Secondary Peritonitis and Cytokines

155

for IL-1, but not for TNF.79 After thermal injury and CLP macrophage production of inflammatory cytokines and arachidonic acid production has been downregulated and was associated with increased mortality.80

Therapy Therapeutic intervention may influence the immune response. Resuscitation with fluids influenced TNF mRNA and IL-1 mRNA production in liver and intestines in intraabdominal sepsis.81 Topical applied antiseptics or antibiotics may influence cytokine release in the peritoneal cavity or compounds used for other indications than sepsis may affect cytokine production, coagulation disturbances and mortality.82

Secondary Peritonitis and Cytokines in Clinical Studies Increased TNF and IL-6 plasma levels correlated with outcome and APACHE II scores in several studies.84-86 However, the determination of plasma cytokines is hampered by the interference with plasma proteins and receptors. Cell-associated cytokines may give a more realistic picture of the inflammatory response.87 The information on the kinetics of cytokines during peritonitis is incomplete. TNF and IL-6 plasma levels were decreased before death in peritonitis and may indicate an anergic response induced by T-cell suppression.88 However increased elastase production (a marker for PMN activation89) and increased neopterin production (a marker for macrophage activation90) in peritonitis do not support this notion. Plasma IL-1 and IL-6 were increased even after large abdominal operations reflecting more the operative trauma than the infection.91-93 The significance of local cytokine production has been supported by several studies. After colectomy IL-6 levels were higher in portal vein than in systemic circulation94 supporting the hypothesis of bacterial translocation in portal and lymphatic circulation.95 Bacterial peritonitis induces local release of proinflammatory cytokines and secondary mediators with subsequent interaction of endothelial cells and neutrophils, microcirculatory dysfunction and tissue damage.96 The peritoneal cavity may be already cleared from pathogens by lavage, while the local release of inflammatory cytokines may continue.97 Peritoneal levels of endotoxin, TNF, IL-1, IL-6 and elastase may be several fold higher than systemic levels. Peritoneal TNF and elastase levels decreased in survivors and remain elevated in nonsurvivors.80 This lead to the conclusion that the mechanisms in sepsis and peritonitis may be similar. However, the immune response occurs in two functional different compartments and the intensity in both compartments may influence outcome.98 This is also supported by studies in other compartments of the body.99 Treatment interventions may influence and modulate cytokine production. IL-6 levels correlated with mean arterial pressure in severe peritonitis. Reoperation caused hypotension which may have induced an early increase in IL-6 plasma levels.100 Antibiotics can release, according to the type of penicillin binding protein (PBP), different amounts of endotoxin.101,102 In surgical intensive care patients, PBP 3 specific antibiotics induced more often endotoxin release than PBP 2-specific antibiotics.103 Several clinical studies with anti-endotoxin-antibodies and anti-TNF-antibodies have been performed in septic patients including patients with peritonitis. The clinical significance of endotoxin is still disputed. Endotoxin may not be the trigger for proinflammatory cytokine release,104 however, it may help in the early detection of anastomotic leaks by endotoxin determination.105 HA-1A reduces mortality in septic patients with endotoxemia and lowers serum TNF levels.106 Other studies were

156

Cytokines and the Abdominal Surgeon

not successful in reducing the mortality of septic patients. This may be due to conceptual and organizational weakness of some studies. The fact that results from animal experiments and clinical studies are not well reflected in the design of these studies is further supported by many investigators. High circulating levels of IL-1ra and sTNF-R and the relatively small proportion of patients developing Endotoxin Core Antibody depletion may contribute to the limitations of therapies to augment natural defenses against endotoxin or proinflammatory cytokines.107

Conclusions Endotoxin and cytokines play an important role in secondary peritonitis and contribute to the outcome of this life-threatening complication. The effect cytokines may have on the immune response depends on concentration, location and other co-factors, e.g., BPI and LBP. The addition of growth factors and cytokines to the treatment arsenal in peritonitis may not be advocated at this time with regard to the controversial results in animal and clinical studies. Cells and pathogens in the peritoneal cavity interact with each other and cytokines play an important part in this communication. However, our information on this network is rather limited. Therapeutic interventions, e.g., resuscitation, antibiotics, surgery, modulate the cytokine levels and thereby the immune response. The blockade of cytokines and endotoxin was not successful in most clinical studies. Newly developed anti-endotoxin-antibodies and compounds which block the endotoxin-induced activation of cells seem to be more promising. The concept of anti-inflammatory response in humans is not yet clear and needs further investigations and well planned studies to reveal the pathophysiological consequences of cytokines in secondary peritonitis.

References 1. Malangoni MA. Pathogenesis and treatment of intra-abdominal infection. Surg Gynecol Obstet 1990; 171(Suppl):31-34. 2. Schein M, Gecelter G, Freinkel W, Gerding H, Becker PJ. Peritoneal lavage in abdominal sepsis. A controlled clinical study. Arch Surg 1990; 125(9):1132-1135. 3. Wittmann DH, Aprahamian C, Bergstein JM. Etappenlavage: Advanced diffuse peritonitis managed by planned multiple laparotomies utilizing zippers, slide fastener, and Velcro analogue for temporary abdominal closure. World J Surg 1990; 14(2):218-226. 4. Nathens AB, Rotstein OD. Therapeutic options in peritonitis. Surg Clin North Am 1994; 74(3):677-692. 5. Offenbartl K, Bengmark S. Intraabdominal infections and gut origin sepsis. World J Surg 1990; 14(2):191-195. 6. Ayala A, Kisala JM, Felt JA et al. Does endotoxin tolerance prevent the release of inflammatory monokines (interleukin 1, interleukin 6, or tumor necrosis factor) during sepsis? Arch Surg 1992; 127(2):191-196. 7. Delshammar M, Lasson A, Ohlsson K. Proteases and protease inhibitor balance in peritonitis with different causes. Surgery 1989; 106(3):555-562. 8. Billing A, Kortmann H, Frohlich D, Jochum M. Breakdown of C3 complement and IgG in peritonitis exudate—pathophysiological aspects and therapeutic approach. Prog Clin Biol Res 1989; 308:527-533. 9. Ayala A, Deol ZK, Lehman DL et al. Does endotoxin play a major role in inducing the depression of macrophage function during polymicrobial sepsis? Arch Surg 1995; 130(11):1178-1184.

Secondary Peritonitis and Cytokines

157

10. Ayala A, Deol ZK, Lehman DL et al. Polymicrobial sepsis but not low-dose endotoxin infusion causes decreased splenocyte IL-2/IFN-gamma release while increasing IL-4/IL-10 production. J Surg Res 1994; 56(6):579-585. 11. Schoeffel U, Jacobs E, Ruf G et al. Intraperitoneal micro-organisms and the severity of peritonitis. Eur J Surg 1995; 161(7):501-508. 12. Berger D, Beger HG. The pathophysiological bases of peritonitis therapy. Chirurg 1992; 63(3):147-152. 13. Ertel W, Morrison MH, Wang P et al. The complex pattern of cytokines in sepsis. Ann Surg 1991; 214:141-148. 14. Morrison DC, Lei MG, Kirikae T, Chen TY. Endotoxin receptors on mammalian cells. Immunobiology 1993; 187(3-5):212-226. 15. Ulevitch RJ, Tobias PS. Receptor-dependent mechanisms of cell stimulation by bacterial endotoxin. Annu Rev Immunol 1995:13(Imm.). 16. van der Poll T, van Deventer SJ, Buller HR, Sturk A, ten Cate JW. Comparison of the early dynamics of coagulation activation after injection of endotoxin and tumor necrosis factor in healthy humans. Prog Clin Biol Res 1991; 367:55-60. 17. Mayoral JL, Schweich CJ, Dunn DL. Decreased tumor necrosis factor production during the initial stages of infection correlates with survival during murine gramnegative sepsis. Arch Surg 1990; 125:24-28. 18. Rasmussen LT, Fandrem J, Seljelid R. Dynamics of blood components and peritoneal fluid during treatment of murine E. coli sepsis with beta-1,3-D-polyglucose derivatives. II. Interleukin 1, tumour necrosis factor, prostaglandin E2, and leukotriene B4. Scand J Immunol 1990; 32(4):333-340. 19. Bagby GJ, Plessala KJ, Wilson LA et al. Divergent efficacy of antibody to tumor necrosis factor-alpha in intravascular and peritonitis models of sepsis. J Infect Dis 1991; 163(1):83-88. 20. Zanetti G, Heumann D, Gerain J et al. Cytokine production after intravenous or peritoneal gram-negative bacterial challenge in mice. Comparative protective efficacy of antibodies to tumor necrosis factor-alpha and to lipopolysaccharide. J Immunol 1992; 148(6):1890-1897. 21. Astiz ME, Saha DC, Carpati CM, Rackow EC. Induction of endotoxin tolerance with monophosphoryl lipid A in peritonitis: importance of localized therapy. J Lab Clin Med 1994; 123(1):89-93. 22. Natanson C, Hoffman WD, Suffredini AF et al. Selected treatment strategies for septic shock based on proposed mechanisms of pathogenesis. Ann Intern Med 1994; 120(9):771-783. 23. Echtenacher B, Falk W, Mannel DN, Krammer PH. Requirement of endogenous tumor necrosis factor/cachectin for recovery from experimental peritonitis. J Immunol 1990; 145(11):3762-3766. 24. Kohler J, Heumann D, Garotta G et al. IFN-gamma involvement in the severity of gram-negative infections in mice. J Immunol 1993; 151(2):916-921. 25. Jurkovich GJ, Mileski WJ, Maier RV et al. Interferon gamma increases sensitivity to endotoxin. J Surg Res 1991; 51(3):197-203. 26. McMasters KM, Cheadle WG. Regulation of macrophage TNF alpha, IL-1 beta, and Ia (I-A alpha) mRNA expression during peritonitis is site dependent. J Surg Res 1993; 54(5):426-430. 27. van Furth R, van Zwet TL, Buisman AM, van Dissel JT. Antitumor necrosis factor antibodies inhibit the influx of granulocytes and monocytes into an inflammatory exudate and enhance the growth of Listeria monocytogenes in various organs. J Infect Dis 1994; 170(1):234-237. 28. Whawell SA, Thompson JN. Cytokine-induced release of plasminogen activator inhibitor-1 by human mesothelial cells. Eur J Surg 1995; 161(5):315-318.

158

Cytokines and the Abdominal Surgeon

29. Thompson JS. The intestinal response to critical illness. Am J Gastroenterol 1995; 90(2):190-200. 30. Chalkiadakis GE, Kostakis A, Karayannacos PE. Pentoxifylline in the treatment of experimental peritonitis in rats. Arch Surg 1985; 120: 1141-1144. 31. Topley N, Mackenzie RK, Williams JD. Macrophages and mesothelial cells in bacterial peritonitis. Immunobiology 1996; 195(4-5):563-573. 32. Edmiston CE Jr, Goheen MP, Kornhall S et al. Fecal peritonitis: Microbial adherence to serosal mesothelium and resistance to peritoneal lavage. World J Surg 1990; 14(2):176-183. 33. Glancey G, Cameron JS, Ogg C, Poston S. Adherence of Staphylococcus aureus to cultures of human peritoneal mesothelial cells. Nephrol Dial Transplant 1993; 8(2):157-162. 34. Ramos BF, Qureshi R, Olsen KM, Jakschik BA. The importance of mast cells for the neutrophil influx in immune complex-induced peritonitis in mice. J Immunol 1990; 145(6):1868-1873. 35. Ramos BF, Zhang Y, Qureshi R, Jakschik BA. Mast cells are critical for the production of leukotrienes responsible for neutrophil recruitment in immune complexinduced peritonitis in mice. J Immunol 1991; 147(5):1636-1641. 36. Hadjiminas DJ, McMasters KM, Peyton JC, Cheadle WG. Tissue tumor necrosis factor mRNA expression following cecal ligation and puncture or intraperitoneal injection of endotoxin. J Surg Res 1994; 56(6):549-555. 37. Acton RD, Dahlberg PS, Uknis ME et al. Differential sensitivity to Escherichia coli infection in mice lacking tumor necrosis factor p55 or interleukin-1 p80 receptors. Arch Surg 1996; 131(11):1216-1221. 38. Meyer TA, Noguchi Y, Ogle CK et al. Endotoxin stimulates interleukin-6 production in intestinal epithelial cells. A synergistic effect with prostaglandin E2. Arch Surg 1994; 129(12):1290-1294. 39. Grewal HP, Kotb M, el Din AM et al. Induction of tumor necrosis factor in severe acute pancreatitis and its subsequent reduction after hepatic passage. Surgery 1994; 115(2):213-221. 40. Opal SM, Palardy JE, Marra MN et al. Relative concentrations of endotoxin-binding proteins in body fluids during infection. Lancet 1994; 344(8920):429-431. 41. Bahrami S, Yao YM, Leichtfried G et al. Monoclonal antibody to endotoxin attenuates hemorrhage-induced lung injury and mortality in rats. Crit Care Med 1997; 25(6):1030-1036. 42. Xu D, Qi L, Guillory D et al. Mechanisms of endotoxin-induced intestinal injury in a hyperdynamic model of sepsis. J Trauma 1993; 34(5):676-682. 43. O’Reilly M, Silver GM, Greenhalgh DG et al. Treatment of intra-abdominal infection with granulocyte colony-stimulating factor. J Trauma 1992; 33(5):679-682. 44. Molloy RG, Holzheimer R, Nestor M et al. Granulocyte-macrophage colony-stimulating factor modulates immune function and improves survival after experimental thermal injury. Br J Surg 1995; 82(6):770-776. 45. Lorenz W, Reimund KP, Weitzel F et al. Granulocyte colony-stimulating factor prophylaxis before operation protects against lethal consequences of postoperative peritonitis. Surgery 1994; 116(5):925-934. 46. Eichacker PQ, Waisman Y, Natanson C et al. Cardiopulmonary effects of granulocyte colony-stimulating factor in a canine model of bacterial sepsis. J Appl Physiol 1994; 77(5):2366-2373. 47. Feltis BA, Jechorek RP, Erlandsen SL, Wells CL. Bacterial translocation and lipopolysaccharide-induced mortality in genetically macrophage-deficient op/op mice. Shock 1994; 2(1):29-33.

Secondary Peritonitis and Cytokines

159

48. Toda H, Murata A, Oka Y et al. Effect of granulocyte-macrophage colony-stimulating factor on sepsis-induced organ injury in rats. Blood 1994; 83(10):2893-2898. 49. Lange JR, Alexander HR, Merino MJ et al. Interleukin-1 alpha prevention of the lethality of Escherichia coli peritonitis. J Surg Res 1992; 52(6):555-559 1992. 50. Callery MP, Mangino MJ, Kamei T, Flye MW. Interleukin-6 production by endotoxin-stimulated Kupffer cells is regulated by prostaglandin E2. J Surg Res 1990; 48(6):523-527. 51. Gough DB, Jordan A, Mannick JA, Rodrick MI. Impaired cell-mediated immunity in experimental abdominal sepsis and the effect of interleukin 2. Arch Surg 1992; 127(7):859-863. 52. Maddaus MA, Simmons RL. Intraperitoneal administration of recoombinant interleukin-2 protects against lethal IP Gram-negative sepsis by induction of a neutrophil influx. Surgical Forum, Chicago, American College of Surgeons 1988:105-106. 53. Chong KT. Prophylactic administration of interleukin-2 protects mice from lethal challenge with Gram-negative bacteria. Infect Immun 1987; 55:668-673. 54. Zanetti G, Glauser MP, Baumgartner JD. Anti-endotoxin antibodies and other inhibitors of endotoxin. New Horiz 1993; 1(1):110-119. 55. Battafarano RJ, Burd RS, Kurrelmeyer KM et al. Inhibition of splenic macrophage tumor necrosis factor alpha secretion in vivo by antilipopolysaccharide monoclonal antibodies. ArchSurg 1994; 129(2):179-180. 56. Lundblad R, Giercksky KE. Synergistic effect of E5 imipenem, and volume support during fulminant intraabdominal sepsis in rats [published erratum appears in J Infect Dis 1995 Aug;172(2):612]. J Infect Dis 1995; 172(1):152-160. 57. Saladino R, Garcia C, Thompson C et al. Efficacy of a recombinant endotoxin neutralizing protein in rabbits with Escherichia coli sepsis. Circ Shoc 1994; 42(2):104-110. 58. Burd RS, Cody CS, Raymond CS, Dunn DL. Anti-endotoxin monoclonal antibodies protect by enhancing bacterial and endotoxin clearance. Arch Surg 1993; 128(2):145-150. 59. Burd RS, Battafarano RJ, Cody CS et al. Anti-endotoxin monoclonal antibodies inhibit secretion of tumor necrosis factor-alpha by two distinct mechanisms. Ann Surg 1993; 218(3):250-259. 60. Christ WJ, Asano O, Robidoux AL et al. E5531 a pure endotoxin antagonist of high potency. Science 1995; 268(5207):80-83. 61. Marra MN, Thornton MB, Snable JL et al. Endotoxin-binding and -neutralizing properties of recombinant bactericidal/permeability-increasing protein and monoclonal antibodies HA-1A and E5 [see comments]. Crit Care Med 1994; 22(4): 559-565. 62. Stack AM, Saladino RA, Siber GR et al. A comparison of bactericidal/permeabilityversus increasing protein variant recombinant endotoxin-neutralizing protein for the treatment of Escherichia coli sepsis in rats. Crit Care Med 1997; 25(1):101-105. 63. Bailat S, Heumann D, Le Roy D et al. Similarities and disparities between corespecific and O-side-chain-specific antilipopolysaccharide monoclonal antibodies in models of endotoxemia and bacteremia in mice. Infect Immun 1997; 65(2): 811-814. 64. Christou NV, Turgeon P, Wassef R et al. Management of intra-abdominal infections. The case for intraoperative cultures and comprehensive broad-spectrum antibiotic coverage. The Canadian Intra-abdominal Infection Study Group Arch Surg 1996; 131(11):1193-1201. 65. Montravers P, Andremont A, Massias L, Carbon C. Investigation of the potential role of Enterococcus faecalis in the pathophysiology of experimental peritonitis. J Infect Dis 1994; 169(4):821-830.

160

Cytokines and the Abdominal Surgeon

66. Montravers P, Mohler J, Saint Julien L, Carbon C. Evidence of the proinflammatory role of Enterococcus faecalis in polymicrobial peritonitis in rats. Infect Immun 1997; 65(1):144-149. 67. Cross A, Asher L, Seguin M et al. The importance of a lipopolysaccharide-initiated, cytokine-mediated host defense mechanism in mice against extraintestinally invasive Escherichia coli. J Clin Invest 1995; 96(2):676-686. 68. Jackson JJ, Kropp. H beta-Lactam antibiotic-induced release of free endotoxin: in vitro comparison of penicillin-binding protein (PBP) 2-specific imipenem and PBP 3-specific ceftazidime [see comments]. J Infect Dis 1992; 165(6):1033-1041. 69. Muller Alouf, Alouf H, Gerlach JE et al. Human pro- and anti-inflammatory cytokine patterns induced by Streptococcus pyogenes erythrogenic (pyrogenic) exotoxin A and C superantigens. Infect Immun 1996; 64(4):1450-1453. 70. Tanaka Y, Jotwani R, Watanabe K et al. Effect of Escherichia coli lipopolysaccharide on Bacteroides fragilis abscess formation and mortality in mice. Microbiol Immunol 1994; 38(2):97-102. 71. Holzheimer RG, Groß J Maseizik T, Steinmetz WG et al. Ischemia and endotoxin mediated inflammatory response (TNF, IL-6, IL-10, TNF-R I + II) in aortic aneurysm repair Shock 1995; 3(Suppl):15-16. 72. Napolitano LM, Campbell C. Interleukin-10 (IL-10) decreases mortality in a lethal murine trauma/sepsis model. Abstract. Proceedings of the Fifteenth Annual Meeting of the Surgical Infection Society. Louisville April 20-22 1992. 73. Kato T, Murata A, Ishida H et al. Interleukin 10 reduces mortality from severe peritonitis in mice. Antimicrob Agents Chemother 1995; 39(6):1336-1340. 74. van der Poll T, Jansen J, Levi M et al. Regulation of interleukin 10 release by tumor necrosis factor in humans and chimpanzees. J Exp Med 1994; 180(5): 1985-1988. 75. van der Poll T, Marchant A, Buurman WA et al. Endogenous IL-10 protects mice from death during septic peritonitis. J Immunol 1995; 155(11):5397-5401. 76. Carpenter A, Evans TJ, Buurman WA et al. Differences in the shedding of soluble TNF receptors between endotoxin-sensitive and endotoxin-resistant mice in response to lipopolysaccharide or live bacterial challenge. J Immunol 1995; 155(4):2005-2012. 77. van der Poll T, Calvano SE, Kumar A et al. Endotoxin induces downregulation of tumor necrosis factor receptors on circulating monocytes and granulocytes in humans. Blood 1995; 86(7):2754-2759. 78. Rokke O, Revhaug A, Giercksky KE. PMN activity and endotoxin kinetics in peritonitis induced after moderate trauma. Acta Chir Scand 1989; 155(10):497-502. 79. Rokke, O, Revhaug A, Seljelid R, Rekvig O. The synergistic effect of trauma and infection on interleukin-1 but not tumor necrosis factor liberation during posttraumatic gram-negative septicemia. Eur Surg Res 1993; 25(1):1-10. 80. Holzheimer RG, Schein M, Wittmann DH. Inflammatory response in peritoneal exudate and plasma of patients undergoing planned relaparotomy for severe secondary peritonitis. Arch Surg 1995; 130(12):1314-1319. 81. Wilson MA, Chou MC, Spain DA et al. Fluid resuscitation attenuates early cytokine mRNA expression after peritonitis. J Trauma 1996; 41(4):622-627. 82. Rosman C, Westerveld GJ, van Oeveren W, Kooi K, Bleichrodt RP. Effect of intraperitoneal antimicrobials on the concentration of bacteria, endotoxin and tumor necrosis factor in abdominal fluid and plasma in rats. Eur Surg Res 1996; 28(5):351-360. 83. Bahrami S, Yao YM, Shiga H et al. Comparison of the efficacy of pentoxifylline and albifyllin (HWA 138) on endotoxin-induced cytokine production, coagulation disturbances, and mortality. Shock 1996; 5(6):424-428.

Secondary Peritonitis and Cytokines

161

84. Damas P, Ledoux D, Nys M et al. Cytokine serum level during severe sepsis in human IL-6 as a marker of severity. Ann Surg 1992; 215(4):356-362. 85. Fugger R, Zadrobilek E, Gotzinger P et al. Perioperative TNF alpha and IL-6 concentrations correlate with septic state, organ function and APACHE II scores in intra-abdominal infection. Eur J Surg 1993; 159(10):525-529. 86. Patel RT, Deen KI, Youngs D, Warwick J, Keighley MR. Interleukin 6 is a prognostic indicator of outcome in severe intra-abdominal sepsis Br J Surg 1994; 81(9): 1306-1308. 87. Munoz C, Carlet J, Fitting C et al. Dysregulation of in vitro cytokine production by monocytes during sepsis. J Clin Invest 1991; 88(5):1747-1754. 88. Hamilton G, Hofbauer S, Hamilton B. Endotoxin TNF-alpha, interleukin-6 and parameters of the cellular immune system in patients with intraabdominal sepsis. Scand J Infect Dis 1992; 24(3):361-368. 89. Duswald KH, Jochum M, Schramm E, Fritz H. Released granulocyte elastase: An indicator of pathobiochemical alterations in septicemia after abdominal surgery. Surgery 1985; 98:892-898. 90. Strohmaier W, Redl H, Schlag G, Inthorn D. D-erythro-neopterin plasma levels in intennsive care patients with and without septic complications. Crit Care Med 1987; 15:757-760. 91. Baigrie RJ, Lamont PM, Kwiatkowski D, Dallman MJ, Morris PJ. Systemic cytokine response after major surgery. Br J Surg 1992; 79(8):757-760. 92. Ueo H, Inoue H, Honda M. Production of interleukin-6 at operative wound sites in surgical patients. J Am Coll Surg 1994; 179:326-332. 93. Glaser F, Sannnwald GA, Buhr H. General stress response to conventional and laparoscopic cholecystectomy. Ann Surg 1995; 221:372-380. 94. Riche F, Dosquet C, Panis Y et al. Levels of portal and systemic blood cytokines after colectomy in patients with carcinoma or Crohn’s disease. J Am Coll Surg 1995; 180(6):718-724. 95 Deitch EA, Xu D, Franko L, Ayala A, Chaudry IH. Evidence favoring the role of the gut as a cytokine-generating organ in rats subjected to hemorrhagic shock. Shock 1994; 1(2):141-145. 96. Livingston DH, Mosenthal AC, Deitch EA. Sepsis and multiple organ dysfunction syndrome: a clinical-mechanistic overview. New Horizons 1995; 3:257-266. 97. Aprahamian C, Schein M, Wittmann D. Cefotaxime and metronidazole in severe intra-abdominal infection. Diagn Microbiol Infect Dis 1995; 22:183-188. 98. Schein, M, Wittmann, DH, Holzheimer R, Condon RE. Hypothesis: Compartmentalization of cytokines in intraabdominal infection. Surgery 1996; 119(6):694-700. 99. Boujoukos AJ, Martich GD, Supinski E, Suffredini AF. Compartmentalization of the acute cytokine response in humans after intravenous endotoxin administration. J Appl Physiol 1993; 74(6):3027-3033. 100. Sautner T, Gotzinger P, Redl Wenzl EM et al. Does reoperation for abdominal sepsis enhance the inflammatory host response? Arch Surg 1997; 132(3):250-255. 101 Prins JM, van Deventer SJ, Kuijper EJ, Speelman P. Clinical relevance of antibioticinduced endotoxin release. Antimicrob Agents Chemother 38(6):1211-1218. 102. Sawyer RG, Adams RB, May AK, Rosenlof LK, Pruett TL. Anti-tumor necrosis factor antibody reduces mortality in the presence of antibiotic-induced tumor necrosis factor release. Arch Surg 1996; 128(1):73-77. 103. Holzheimer RG, Hirte JF, Reith B et al. Different endotoxin release and IL-6 plasma levels after antibiotic administration in surgical intensive care patients. J Endotoxin Res 1996; 3:261–267.

162

Cytokines and the Abdominal Surgeon

104. Kelly JL, O’Sullivan C, O’Riordain M et al. Is circulating endotoxin the trigger for the systemic inflammatory response syndrome seen after injury? Ann Surg 1997; 225(5):530-541. 105. Junger W, Junger WG, Miller K et al. Early detection of anastomotic leaks after colorectal surgery by measuring endotoxin in the drainage fluid. Hepatogastroenterology. 1996; 43(12):1523-1529. 106. Wortel CH, von der Mohlen MA, van Deventer SJ et al. Effectiveness of a human monoclonal anti-endotoxin antibody (HA-1A) in gram-negative sepsis: relationship to endotoxin and cytokine levels [see comments]. J Infect Dis 1992; 166(6): 1367-1374. 107. Goldie AS, Fearon KC, Ross JA et al. Natural cytokine antagonists and endogenous antiendotoxin core antibodies in sepsis syndrome. The Sepsis Intervention Group. JAMA 1995; 274(2):172-177.

Cytokines and Combination Treatment of Intra-Abdominal Infections

163

CHAPTER 14

Cytokines and Combination Treatment of Intra-Abdominal Infections William G. Cheadle and Hiram C. Polk, Jr.

Introduction

A

bdominal infection and subsequent sepsis remains a leading cause of multiple organ failure (MOF) despite improved antibiotics, early surgical intervention, and effective organ support. Highlighting this problem is the fact that the current mortality rate from MOF in patients with abdominal sepsis has not improved from 47% over the past 10 years as reported in a 1995 British study.1 In a more recent study, organ failure was not correlated with recurrent infection.2 Nearly 73% of the patients from that study who stayed in the hospital longer than 10 days developed organ failure. Virtually all of the deaths were in this group, but only one patient developed recurrent infection. This implies that sepsis and organ failure occur more commonly in a setting of an effective local host defense response, yet the prognosis remains poor in such patients. There is thus a real need for new anti-infective therapies (or modification of existing ones) and for combination therapy with anticytokine strategies. Organ failure subsequent to sepsis has been attributed to the systemic inflammatory response syndrome due to gram-negative aerobes, endotoxin, and the circulatory release of proinflammatory cytokines such as tumor necrosis factor-alpha (TNF-∀) and interleukin 1 (IL-1).3 In contrast, after experimental fecal peritonitis, there are very low serum levels of TNF-∀ or IL-1 but a local sustained increase in organ (lung and liver) expression of TNF-∀ and IL-l mRNA.4 Local production of these cytokines may upregulate neutrophil and endothelial adhesion molecules, causing neutrophil sequestration.5 The correlation between serum endotoxin and outcome in human sepsis studies recently has been questioned.6 Even though circulating cytokine concentrations (TNF-∀, IL-1, IL-6) are reported to be of predictive value in human sepsis studies, only the measurement of serum IL-6 has been of consistent value.7

Overview of Our Current Research We have postulated that remote organ injury is in part neutrophil-mediated and have used cecal ligation and puncture (CLP) in the mouse as our model. Neutrophil Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

164

Cytokines and the Abdominal Surgeon

sequestration into the liver and lung occurs rapidly after CLP, and it is accompanied by a reduction in the peripheral blood neutrophil count. Use of an antineutrophil antibody to deplete neutrophils has abrogated both the liver and lung injury, which is typically observed despite an increase in serum endotoxin levels. In fact, serum TNF, IL-1, and endotoxin levels are low after CLP, and mortality is similar in endotoxin-resistant (TNF-deficient) animals. Levels of bacteremia, rather than endotoxemia, have correlated with tissue proinflammatory cytokine mRNA, protein, and neutrophil sequestration. Pentoxifylline, which increases peripheral blood cell elasticity and reduces TNF production, has reduced mortality, lung neutrophil sequestration, and IL-1 mRNA, but increased peritoneal leukocyte accumulation after CLP. A blockade of integrin adhesion molecules (CD11b/CD18) has reduced peritoneal neutrophil migration but increased lung and liver neutrophil sequestration, liver injury, and bacteremia. The C-X-C chemokines, macrophage inflammatory-2 (MIP-2) and KC, are chemoattractants that bind to endothelial cells and potentially activate neutrophils rolling along the endothelium. Liver, lung, and peritoneal leukocytes after CLP have shown an early increase in MIP-2 and KC mRNA and protein expression by both alveolar macrophages and lung neutrophils. Even though antibodies to IL-1 have reduced lung MIP-2 mRNA expression after CLP, neutrophil sequestration is unchanged. Intraperitoneal injection of MIP-2 has caused peritoneal neutrophil migration, and antibody to MIP-2 has reduced peritoneal neutrophil migration after CLP. In mast cell-deficient mice injection of MIP-2 has resulted in less peritoneal neutrophil migration when compared with normal controls. Thus far, we have established a role for chemokines and adhesion molecules in the regulation of peritoneal neutrophil migration after peritonitis but have failed to alter organ neutrophil sequestration with anticytokine or adhesion molecule therapies. Inhibition of peritoneal neutrophil migration after CLP paradoxically increases organ neutrophil sequestration and injury. Organ neutrophil sequestration after peritonitis may be due to mechanical factors such as a reduction in neutrophil deformability rather than local proinflammatory cytokine production or adhesion molecule upregulation. It will be important in future trials of combination therapy to consider both augmentation of local host defense responses, as well as mechanisms by which organ failure occurs in the setting of peritonitis.

The Macrophage The macrophage plays a pivotal role in the immune response to bacterial peritonitis. In addition to antigen presentation in association with class II MHC expression, macrophages display a variety of important functions that include phagocytosis, chemotaxis, and secretion of cytokines and other immunologically active substances, including IL-1, IL-6, TNF, and transforming growth factor-beta. These cells also synthesize lysozyme, proteases, and C proteins, as well as surface receptors for immunoglobulins, cytokines, and many other substances involved in regulation of their own function. While macrophages express some of these products constitutively, many macrophage functions are subject to regulation by autocrine, paracrine, and endocrine signals. The process of macrophage “activation” involves expression of many of these substances that correlate with the acquisition of functional characteristics such as bactericidal and tumoricidal activity. Several agents have been shown to induce or enhance macrophage activation, including interferon-gamma, endotoxin, and muramyl dipeptide.8-10 Macrophages are the major sources of IL-1 and TNF-∀. While not produced in appreciable levels in unstimulated macrophages, these

Cytokines and Combination Treatment of Intra-Abdominal Infections

165

cytokines are induced by treatment with a variety of agents, including interferon and endotoxin-lipopolysaccaride (LPS).11,12 Schein et al13 have shown that acute abdominal infection in patients with peritonitis is likely to be accompanied by high levels of proinflammatory cytokines and endotoxin in the peritoneal cavity, despite low levels in the serum.

The Role of Endotoxin The systemic response to gram-negative sepsis has been attributed in part to the release of proinflammatory cytokines such as TNF-∀ and IL-1, in response to LPS. Proinflammatory cytokines have been shown to upregulate the expression of adhesion molecules on neutrophils and endothelium and also to stimulate tissue C-X-C chemokine release, which may result in tissue neutrophil sequestration. This resultant sequestration leads to tissue injury via neutrophil activation and degranulation. Many of these findings have been demonstrated in animal models challenged with large doses of gram-negative bacteria or LPS.14 In human sepsis studies, the causative role of LPS in the pathogenesis of gram-negative shock and organ failure has not been definitely established.15 Following relative failures of recent anti-LPS treatment trials, the relationship between circulating LPS and the severity of septic shock has been questioned.16 In addition, the role of TNF-∀ as a common mediator of organ damage in sepsis is not as clear as originally thought17 and a relationship of antecedent TNF-∀ activity to the evolution of clinical organ dysfunction has yet to be defined. An interesting method of counteracting the effects of endotoxin has been studied in depth by Kodama and co-workers18 from the Shiga University in Japan. They have bound polymyxin-B to polystyrene and used extracorporeal filtration as a venovenous means to filter endotoxin. Preliminary data from a series of 136 patients demonstrated enhanced survival to concomitant, but nonrandomized controls.18 The treatment of sepsis in dogs that had been injected with both endotoxin and E. coli followed by arteriovenous filtration using a column that was filled with these polymyxin-B immobilized fibers (PMX-F) significantly increased the survival rate.19 PMX-F endotoxin filtration has been used with antibiotics and nonspecific immune stimulation in a rodent sepsis model and demonstrated an additive survival benefit compared to any of the individual agents used alone.20

The Neutrophil

Anderson et al21 have suggested that the neutrophil is central to remote organ failure, and that an increase in the numbers of circulating neutrophils is common in these clinical settings.22 Organ injury, however, has been postulated to occur after the integrin adhesion molecule complex on neutrophils has bound to the ICAM-1 receptor on endothelium, allowing neutrophils to migrate across the vascular barrier into tissue parenchyma.23-24 Botha et al23 demonstrated an upregulation of CD11b on circulating neutrophils from patients who progress to MSOF when compared to patients with similar injury who do not. The role of the neutrophil was also examined in an animal model of CLP. The use of an antineutrophil antibody reduced both lung and liver injury, even though serum endotoxin levels were actually elevated in the antipolymorphonuclear leukocytes (PMN)-treated animals.2 Neutrophils, which clearly have been shown to mediate tissue injury in other models of inflammation by inappropriate degranulation and oxygen radical production, mediate such organ injury during intra-abdominal infection as well.

166

Cytokines and the Abdominal Surgeon

Lymphocytes Lymphocytes have been classically divided into B, T, and natural killer cells. More recently the T helper (CD4) cells have been divided into subsets of Th1 and Th2, depending on the cytokine profiles. Th1 cells secrete IL-2 and interferon-gamma, and Th2 cells secrete IL-4, IL-10, and IL-13. The former appear to augment the host’s inflammatory response, while the latter inhibit it. Kelly et al25 have shown that mortality from CLP was increased up to 10 days after burn injury, and that this enhanced susceptibility may be mediated by a switch from a predominant T helper-1 lymphocyte response to a T helper-2-type response. Treatment with anti-IL-10 abrogated this, indicating that IL-10 may be a trigger of immune suppression after the burn injury. Decker et al26 demonstrated a shift toward a predominant T helper-2 cytokine pattern in patients, which was more pronounced after open cholecystectomy than when done laparoscopically. IL-4, a T helper-2 cytokine, was markedly increased in the open cholecystectomy patients compared to those who underwent laparoscopic cholecystectomy. The type 2 T helper shift may be an indicator of general immune suppression after stress, and restoration of this shift may reduce the risk of postoperative infection. IL-10, however, has been used successfully in several animal models of infection and pancreatitis to enhance survival after the insult. Preliminary studies show that it is safe in humans, but dosage and timing are critical.27

Therapy of Intra-Abdominal Infection Combination therapy for peritonitis has been limited clinically to conventional sepsis trials in which large groups of heterogenous patients receive a particular biologic therapy in conjunction with conventional antibiotics. Unfortunately, no therapy has been shown to reduce the 28-day all-cause mortality from sepsis, which the Food and Drug Administration requires before licensing for general use. Strategies that have been subjected to such clinical trials include anti-endotoxin antibodies, antiTNF antibodies, IL-1 receptor antagonists, soluble TNF receptors, and antagonists of bradykinin, platelet-activating factor, and tissue factor. Delineation of the specific pathways of cytokine production in the response to peritonitis will be critical for design of future therapeutic strategies. Most of the above interventions have been based on models of endotoxemia and bacteremia that are not clinically relevant to peritonitis. Experimentally, IL-1 receptor antagonist has been shown to increase survival from CLP, a clinically relevant model of peritonitis.28 However, administration of antiTNF antibodies to mice after CLP increases mortality,29 suggesting that in this model of sepsis, IL-1 is inappropriately expressed and not TNF. Pentoxifylline, a phosphodiesterase inhibitor, decreases endotoxin-induced production of TNF, mainly by reducing TNF mRNA levels.30 When stimuli other than endotoxin have been used, pentoxifylline was not a potent inhibitor of TNF synthesis and had a more pronounced effect on IL-1.31 This suggests that different stimuli induce the synthesis of proinflammatory mediators through separate pathways, and that the steps that are blocked by pentoxifylline may affect TNF or IL-1, depending on the stimulus. We found that pentoxifylline improved survival but did not reduce the CLP-induced TNF mRNA content of lung tissue or peritoneal macrophages, whereas IL-1 mRNA levels were significantly reduced in both tissues.32 The effectiveness of continuous versus intermittent antibiotic administration in the treatment of many infections is supported by the consistent results of several investigators.33-35 The efficacy of cephalosporins is predominantly independent of

Cytokines and Combination Treatment of Intra-Abdominal Infections

167

concentration and is more dependent on the duration of time that the tissue-serum antibiotic concentration levels exceed the minimum inhibitory concentration (MIC).36 The area below the concentration curve is similar to intermittent versus continuous infusion; however, the shape of the curve is different, with continuous infusion achieving a more sustained level above the MIC. In general, the short halflife of cephalosporins and #-lactams favors a rapid clearance, and intermittent dosing allows for regrowth of gram-negative bacilli whenever the concentrations are below MIC or minimum bactericidal concentrations. These factors, in addition to the lack of a postantibiotic effect, are further arguments for continuous administration of cephalosporins. Therapy with multiple interventions, including cytokines, growth factors, and antibiotics has received little attention in the literature, despite the fact that there are multiple redundant mechanisms at play in the patient with peritonitis. Using a model of soft tissue infection, produced by a contaminated foreign body (suture) placed into the thigh, combination therapy improved survival.37 Mice were given granulocyte-macrophage colony stimulating factor, TNF, and the immune adjuvant MDP, in combination with either ampicillin/sulbactam or cefoxitin. Mice given either of these combinations had better survival rates than controls receiving no therapy or those receiving only part of the combination. The mechanisms at play are far from being defined, but the idea of appropriately timed and sequenced therapy deserves further investigation.

References 1. McLauchlan GJ, Anderson ID, Grant IS et al. Outcome of patients with abdominal sepsis treated in an intensive care unit. Br J Surg 1995; 82:524-529. 2. Cheadle WG, Mercer-Jones M, Wickel DJ et al. Poor clinical outcome from peritonitis is primarily due to pulmonary and other organ failure, not recurrent peritoneal infection. Ann Surg 1997; 225:744-753. 3. Casey LC, Balk RA, Bone RC. Plasma cytokine and endotoxin levels correlate with survival in patients with the sepsis syndrome. Ann Intern Med 1993; 119:771-778. 4. Hadjiminas DJ, McMasters KM, Peyton JC et al. Tissue tumor necrosis factor mRNA expression following cecal ligation and puncture or intraperitoneal injection of endotoxin. J Surg Res 1994; 56:549-555. 5. Windsor AC, Mullen PG, Fowler AA et al. Role of the neutrophil in adult respiratory distress syndrome. Br J Surg 1993; 80:10-17. 6. Hurley JC. Reappraisal of the role of endotoxin in the sepsis syndrome. Lancet 1993; 341:1133-1135. 7. Pruitt JH, Copeland EM, Moldawer LL. Interleukin-1 and interleukin-1 antagonism in sepsis, systemic inflammatory response syndrome, and septic shock. Shock 1995; 3:235-251. 8. Uhing RJ, Adams DO. Molecular events in the activation of murine macrophages. Agents Actions 1989; 26:9-14. 9. Adams DO. Molecular interactions in macrophage activation. Immunology Today 1989; 10:33-35. 10. Lambert LE, Paulnock DM. Differential induction of activation markers in macrophage cell lines by interferon-gamma. Cell Immunol 1989; 120:401-418. 11. Kovacs EJ, Radzioch D, Young HA et al. Differential inhibition of IL-1 and TNF-alpha mRNA expression by agents which block second messenger pathways in murine macrophages. J Immunol 1988; 141:3101-3105. 12. Mannel DN, Moore RN, Mergenhagen SE. Macrophages as a source of tumoricidal activity (tumor-necrotizing factor). Infect Immun 1980; 30:523-530.

168

Cytokines and the Abdominal Surgeon

13. Schein M, Wittmann DH, Holzheimer R, Condon RE. Hypothesis: Compartmentalization of cytokines in intra-abdominal infection. Surgery 1996; 119:694-700. 14. Creasey AA, Stevens P, Kenney JC et al. Endotoxin and cytokine profile in plasma of baboons challenged with lethal and sublethal Escherichia coli. Circulatory Shock 1991; 33:84-91. 15. Van Dervort AL, Danner RL. Antiendotoxin approaches to septic shock therapy. Crit Care Med 1994; 22:539-541. 16. Hurley JC. Reappraisal of the role of endotoxin in the sepsis syndrome. Lancet 1993; 341:1133-1135. 17. Van der Poll T, Lowry SF. Tumor necrosis factor in sepsis: Mediator of multiple organ failure or essential part of host defense? Shock 1995; 3:1-12. 18. Kodama M, Tani T, Hanasawa K. Extracorporeal removal in the septic patient by toraymyxin clinical results in a phase II or III in Japan [abstract]. Shock 1997; 7(suppl.):6. 19. Hanasawa K, Tani T, Kodama M. New approach to endotoxin and septic shock by means of polymyxin B immobilized fiber. Surg Gynecol Obstet 1989; 168:323-331. 20. Cheadle WG, Hanasawa K, Gallinaro RN et al. Endotoxin filtration and immune stimulation improve survival from gram-negative sepsis. Surgery 1991; 110:785-792. 21. Anderson BO, Brown JM, Harken AH. Mechanism of neutrophil mediated tissue injury. J Surg Res 1991; 52:170-179. 22. Peterson VM, Ambruso DR, eds. Phagocytes Production and Function Following Burn Injury. Austin: RG Landes Co, 1994. 23. Botha AJ, Moore FA, Moore EE et al. Early neutrophil sequestration after injury: a pathogenic mechanism for multiple organ failure. J Trauma 1995; 39:411-417. 24. Weiss SJ. Tissue destruction by neutrophils. N Engl J Med 1989; 320:365-376. 25. Kelly J, Soberg C, Lyons A et al. Anti-IL-10 antibody restores burn-induced defects in t-cell function. Surgery (in press). 26. Decker D, Schondorf M, Bidlingmaier F et al. Surgical stress induces a shift in the type-1/type-2 T-helper cell balance, suggesting down-regulation of cell-mediated and up-regulation of antibody-mediate and up-regulation of antibody-mediated immunity commensurate to the trauma. Surgery 1996; 119:316-325. 27. Moldawer LL. The clinical potential of rHIL-10 in ischemia-reperfusion injury [abstract]. Shock 1997; 7(suppl.):77. 28. Alexander HR, Doherty GM, Venzon DJ et al. Recombinant interleukin-1 receptor antagonist (IL-1ra): effective therapy against gram-negative sepsis in rats. Surgery 1992; 112:188-194. 29. Echtenacher B, Falk W, Mannel DN et al. Requirement of endogenous tumor necrosis factor/cachectin for recovery from experimental peritonitis. J Immunol 1990; 145:3762-3766. 30. Strieter RM, Remick DG, Ward PA et al. Cellular and molecular regulation of tumor necrosis factor-alpha production by pentoxifylline. Biochem Biophys Res Commun 1988; 155:1230-1236. 31. Prabhakar U, Lipshutz D, Truneh A. Inhibition of CD44, CD45 and LFA-3 mediated cytokine release from human monocytes by SK&F 86002 and pentoxifylline. Int J Immunopharmacol 1993; 15:205-209. 32. Hadjiminas DJ, McMasters KM, Robertson SE et al. Enhanced survival from cecal ligation and puncture with pentoxifylline is associated with altered neutrophil trafficking and reduced interleukin-expression but not inhibition of tumor necrosis factor synthesis. Surgery 1994; 116:348-355. 33. Livingston DH, Wang MT. Continuous infusion of cefazolin is superior to intermittent dosing in decreasing infection after hemorrhagic shock. Am J Surg 1993; 165:203-207.

Cytokines and Combination Treatment of Intra-Abdominal Infections

169

34. Roosendaal R, Bakker-Woudenberg IA, van den Berghe-van Raffe M et al. Impact of the dosage schedule on the efficacy of ceftazidime, gentamicin and ciprofloxacin in Klebsiella pneumoniae pneumonia and septicemia in leukopenic rats. Eur J Clin Microbiol Infect Dis 1989; 8:878-887. 35. Naziri W, Cheadle WG, Trachtenberg LS et al. Increased antibiotic effectiveness in a model of surgical infection through continuous infusion. Am Surg 1995; 61:11-5. 36. Nishida M, Murakawa T, Kamimura T et al. Bactericidal activity of cephalosporins in an in vivo model simulating serum levels. Antimicrob Agents Chemother 1978; 14:6-12. 37. Gaar EE, Naziri W, Cheadle WG et al. Improved survival in simulated surgical infection with combined cytokine, antibiotic and immunostimulant therapy. Br J Surg 1994; 81:1309-1311.

Cytokines, Abdominal Trauma and the “Second Hit” Phenomenon

171

CHAPTER 15

Cytokines, Abdominal Trauma and the “Second Hit” Phenomenon Roger Saadia and Jeffrey Lippman

Mutiple Organ Failure After Trauma

I

n the 1970s and well into the 1980s, infection, often in the form of an occult intra-abdominal focus, was thought to be the most common cause of multiple organ failure (MOF) after trauma. It has more recently become apparent that a systemic inflammatory response syndrome (SIRS) is the prelude to multiple organ dysfunction; importantly, it is well documented now that this exaggerated inflammatory response need not be triggered by infection.1 More complex “inflammatory” models have been devised to explain the development of postinjury MOF in the absence of infection. These models have to account for the causation by the initial traumatic insult of an inflammatory state, sometimes reversible at first, that could progress to an uncontrolled response and organ failure. Furthermore, they have to be consistent with clinical observations and, at the same time, be susceptible of corroboration by advancing knowledge in the biology of inflammation and sepsis.

Cytokines and the Postinjury Inflammatory State

Cytokines2 are protein mediators produced by leukocytes, macrophages and endothelial cells, amongst others, that are essential to a variety of homeostatic functions. Excessive production and release, however, initiate widespread tissue injury and result in organ dysfunction. There has been recently rapid accumulation of data implicating cytokines in the pathobiology of SIRS. Four cytokines in particular have been incriminated: tumour necrosis factor ∀ (TNF∀), interleukin 1# (IL-1#), interleukin 6 (IL-6) and interleukin 8 (IL-8). The cytokine cascade is initiated when a stimulus such as Gram-negative bacterial endotoxin induces the production and release of “proximal” cytokines which include TNF∀ and IL-1#; these are responsible for most of the pathophysiological disturbances characteristic of sepsis. Endotoxin and the “proximal” cytokines stimulate the production of “distal” cytokines, such as IL-6 and IL-8. The role of these “distal“cytokines in sepsis is still incompletely defined, but they seem to intensify and perpetuate the inflammatory response. Excessive and prolonged elevations of circulating IL-6 levels in patients after trauma have been associated with complications and mortality.3 IL-8 is a potent activator and chemoattractant for polymorphonuclear leukocytes and is thought to mediate neutrophilic tissue inflammation resulting in organ dysfunction,4 particularly in the lung. Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

172

Cytokines and the Abdominal Surgeon

Interleukin-10 (IL-10) can inhibit the production of other cytokines (TNF∀, IL-1#, IL-6) and interferon !; it can also stimulate IL-1 receptor antagonist production in monocytes; in clinical studies it has been particularly studied in patients with sepsis and is considered to be a downregulator of the inflammatory response. For organ injury to occur in the inflammatory milieu, activated neutrophils need to marginate in the postcapillary venules, adhere to the inflamed endothelium and migrate into the interstitium.5 The latter two steps are mediated by adhesion molecules of the #2 integrin family (the heterodimers CD18/CD11a and CD18/CD11b). Their ligands on the endothelium are the intercellular adhesion molecules-1 and -2 (ICAM-1 and ICAM-2) that belong to the IgG family. Integrin-mediated adhesion is the result of endothelial and neutrophil activation by proinflammatory mediators and the slowing down of neutrophil flow. This slowing, or rolling, effect is mediated by another group of adhesion molecules known as selectins. These L-, E-, P-selectins and others are structurally related glycoproteins. L-selectin is present on neutrophils under basal conditions, and E- and P-selectins are expressed on activated endothelium. P-selectin expression follows stimulation by histamine, complement or thrombin; E-selectin expression is caused by endothelial activation by endotoxin, TNF∀ and IL-1#. Neutrophil migration is the result of attraction by chemotactic cytokines (chemokines), in particular IL-8, neutrophil activating protein-2 (NAP-2) and epithelial cell-derived neutrophil activating protein (ENA-78).

“One-Hit” and “Two-Hit” Models

It has been suggested6 that, as a result of successful resuscitation after significant trauma, an early inflammatory state sets in; from a teleological perspective, this is presumably beneficial for the host. The systemic inflammatory response resolves as the patient recovers. However, in this conceptual framework, the response is assumed to be excessive at times. There are thus said to be two patterns of SIRS-induced MOF. In the “one-hit” model, the initial traumatic insult is so massive that it causes a severe SIRS and an early MOF syndrome. In the “two-hit” model, a less severe injury is responsible for a moderate state of SIRS, sometimes associated with subclinical multiple organ dysfunction; subsequent inflammatory insults (“second” and “sequential” hits), infectious or noninfectious, may amplify the pre-existing inflammatory state into an exaggerated SIRS sufficient to induce delayed MOF. In the “two-hit” model, the initial traumatic event is said to prime the host, in a way such that an innocuous second inflammatory insult (the activating event) triggers an uncontrolled inflammatory response and MOF.6,7 The “two-hit” model purports to account for the development of MOF after trauma in the clinical setting. In order to validate the model in this context, it would be necessary to provide evidence for the following phenomena: 1) the initial trauma (“first hit”) is responsible for the production of proinflammatory mediators (inter alia cytokines); 2) a subsequent insult (“second hit”) occurring in the aftermath of the initial trauma is responsible for the activation of the same or other mediators; 3) this secondary mediator outburst is temporally related to the development of MOF. The formulation of the theory still falls short in defining precisely what these sequential sublethal activating events might be in the clinical setting; a long list of possible insults come to mind in the context of the severe polytrauma patient treated in an intensive care unit: a relook laparotomy, internal fixation of long bone fractures, the performance of a tracheostomy, an intercurrent infection (pneumonia), a complication of invasive monitoring or the transfer to the radiology suite of a poten-

Cytokines, Abdominal Trauma and the “Second Hit” Phenomenon

173

tially unstable patient. It is important to note the emphasis put by the theory on the moderate magnitude of the activating stimulus: the “second hit” would be relatively inconsequential should it occur as an isolated event. This renders the concepts of initial priming and subsequent activation crucial to the “second hit” hypothesis.

Priming and Activation There is a large body of experimental evidence supporting the concept that a host might be rendered vulnerable to a mildly injurious activating stimulus. Low doses of endotoxin, chemotactic agents, various inflammatory mediators and cytokines prime inflammatory cells like polymorphonuclear neutrophils and macrophages such that a subsequent otherwise relatively innocuous stimulus promotes significant tissue injury.8-12 In a rabbit model,13 hemorrhage caused an increased susceptibility of the cardiovascular and pulmonary systems to the adverse effect of endotoxin administered in an otherwise harmless dose; the authors suggested that part of the interaction of hemorrhagic shock and endotoxin administration may be the result of macrophage stimulation leading to production of TNF∀ and IL-1#; in addition, the observed profound leukopenia in the animals may be the consequence of cytokine-induced endothelial proadhesiveness. In an attempt to validate their “second hit” hypothesis, the Denver group have developed a rodent model14 to study neutrophil priming and reactivation: in a series of experiments, they were able to demonstrate that superior mesenteric artery clamping (analogous to shock-induced splanchnic hypoperfusion) induces an ischemia-reperfusion phenomenon capable of priming neutrophils in the mesenteric bed; these neutrophils enter subsequently the systemic circulation and gain access to the pulmonary vascular system, where they are susceptible of reactivation; at this stage, the administration of a small dose of endotoxin unleashes their cytotoxic effects. Macrophage cytokine production is a complex phenomenon, and there is evidence that a wide range of different inflammatory stimuli can have both positive and negative regulatory influences; the final integrated host response to repetitive insults depends on the nature, duration and order of exposure to these stimuli. This is illustrated in a series of experiments on murine macrophages demonstrating qualitative variations in TNF∀ and IL-1# production that are dependent on the nature of both pretreatment stimulus and activating stimulus.15

Clinical Observations A number of clinical studies lend support to the characterization of postinjury MOF as occurring in two different temporal patterns: early and late. In addition, they suggest that the mechanisms of SIRS-induced MOF may be different in at least one respect: whereas infection is rarely associated with the development of early MOF, it is often a trigger of late MOF. Faist et al16 noted that 44% of their postinjury MOF patients developed organ failure within 12 to 36 hours, while in the remaining 56% MOF set in late (7 to 8 days); the latter group was uniformly associated with sepsis. In another German study,17 two similar peaks of organ failure were observed; in 50% of the late MOF patients, infections were chronologically related to the onset of organ failure. In their study of postinjury ventilator-associated pneumonias,18 the Denver group observed that in the 14 patients developing MOF within 3 days, pneumonia was a trigger of MOF in only four of eleven cases of pneumonia; in contrast, in the remaining 14 patients with late onset MOF, eight out of the nine pneumonias were considered to be causally related to the development of MOF. In another study19

174

Cytokines and the Abdominal Surgeon

of 70 patients with MOF after trauma by the same group, an identical temporal bimodal distribution of MOF onset was noted; however, early and late MOF had a similar incidence of major infections, although sepsis seemed to be more important in causing late organ failure.

“First Hit” and Cytokine Patterns There is compelling evidence that priming and activation are phenomena that occur at a cellular level. While this finding is necessary to set the second-hit hypothesis on firm ground, it is by no means sufficient. It is of vital importance for the theory to examine the available clinical data on the production of cytokines in the period between initial trauma and onset of MOF and, if possible, to observe the possible influence of intercurrent clinical events on cytokine release. A large number of clinical studies have documented the early elevation of cytokine levels, especially IL-6, after thermal injury and mechanical trauma. The IL-6 response after burn injury has been well summarized:3 the predominant source of IL-6 seems to be the injured skin, and highest concentrations are found around the site of burn and in dysfunctional end-organs; postburn IL-6 release is significant, proportionate to the severity of burn and persistent; high peaks are predictors of septic complications and mortality. In a study of 20 injured patients requiring emergency laparotomy, Moore et al20 measured portal and systemic levels of endotoxin, TNF∀ and IL-6; of these, only IL-6 was found to be elevated immediately after the injury; this was followed by a steady decline after 6 hours; levels did not correlate with the development of MOF. In a subsequent study21 performed a few years later in the same institution, serial plasma levels of IL-6, IL-8 and soluble ICAM-1 were measured in 27 high-risk trauma patients; the findings were somewhat different: high levels IL-6 and IL-8 were found early after injury and persisted at least 3 to 4 days and predicted MOF; there was also a late elevation of soluble ICAM-1 the significance of which is not entirely clear. Roumen et al22 investigated the course of cytokine levels in a mixed group of 66 patients with multiple trauma, with a ruptured abdominal aortic aneurysm, and after elective repair of abdominal aortic aneurysms; TNF∀ and IL-1# were significantly higher immediately after admission in patients who developed organ dysfunction and in nonsurvivors; a sustained but mild elevation of IL-6 was noted in the patients with subsequent MOF. Hoch et al23 reported that severe trauma produces rapid large increases in circulating concentrations of IL-6 and IL-8, but TNF∀ remained low whilst the presence of endotoxin could not be detected. There is very little published data on circulating IL-10 levels in trauma patients. The presence of this cytokine was correlated with various clinical parameters in a retrospective study of 66 multiple trauma patients;24 detectable levels of IL-10 within 72 hours were associated with early hypotension and subsequent infectious complications, but did not predict pulmonary, hepatic or renal dysfunction; the role of IL-10 in trauma remains unclear. Other clinical studies have investigated primarily the postinjury patterns of adhesion proteins. In a series of 50 trauma patients,25 it was noted that moderate injury resulted in a minimal increase in circulating E- and P-selectins; in more severely injured patients who subsequently developed septic complications, organ dysfunction or eventually died, there were elevated levels of E-selectin; the increase in P-selectin level was not statistically significant. In a retrospective study26 of trauma patients that included burns, high concentrations of IL-8, ENA-78 and soluble P-selectin were measured very early after the injury and persisted several weeks thereafter; E-selectin was increased only during the first week after major trauma.

Cytokines, Abdominal Trauma and the “Second Hit” Phenomenon

175

“Second Hit” and the Inflammatory Response The data quoted above indicate that an early cytokine response is triggered by trauma. In order to validate clinically the “second hit” hypothesis, evidence is needed to show that a clinical event of some sort is responsible for the re-activation of the inflammatory machinery that had been primed in the immediate aftermath of the initial injury. Comparative dynamic studies of inflammatory mediators levels of the whole period between initial trauma and recovery or development of MOF are rare. The required information has to be extrapolated sometimes from studies of nontrauma patients or from studies that have investigated mediators other than cytokines. In their mixed group of patients, Roumen et al22 noted a rather inexplicable cytokine response. After an initial high level of IL-1# and TNF∀ in the patients who developed organ dysfunction, the levels in these patients and in those without MOF equalized between day 1 and 4 as a result of a rise in the levels of the non-MOF patients; subsequently, the levels in these patients decreased, while they remained elevated in the MOF patients; the highest levels of IL-6 were found in the first 6 hours after injury. At no stage was it possible, in this study, to correlate cytokine profiles with the presence of endotoxemia. In another study,27 sharp increases of TNF∀ and IL-6 were observed after day 3 in the nonsurvivors of severe trauma while the levels remained low in the survivors, but no explanation for this sudden rise was offered. Commenting explicitly on their observation of increased selectin levels in their high risk trauma patients and a possible “second hit” event, Simons et al25 could find no evidence in support of such a link. There are to date no other clinical studies of trauma patients documenting a clearly triggered cytokine response, within days of the initial injury, that can be incriminated in the occurrence of MOF. There are, however, two recent studies investigating specifically the role of subsequent surgical procedures as secondary inflammatory insults in the development of late MOF. In a series of 106 trauma patients requiring secondary operations (facial reconstruction, pelvic or long bone osteosynthesis, tracheostomy and others),28 it was found that preoperative levels of neutrophil elastase, C-reactive protein and platelet count were good predictors of subsequent MOF, whilst other preoperative parameters (blood pressure, heart rate, bilirubin, creatinine, urine output, lactate, pH and coagulation status) were unable to differentiate between MOF and non-MOF patients. The authors concluded, inter alia, that “secondary operations …may act as a second-hit in severely injured patients and trigger postoperative organ failure”. Indeed this may well be the case, but this is not a conclusion supported by the results of the study; the findings had merely identified better markers of MOF than more commonly used parameters. Support for the “second hit” phenomenon would have required the demonstration of an amplified inflammatory response following the secondary intervention and leading to MOF. A more pertinent study29 has tackled the question of the direct effect of a re-look laparotomy on the cytokine profile; in a series of 15 patients who had presented initially with severe peritonitis (of nontraumatic origin) and who required at least one relaparotomy, it was found that postoperative IL-6 levels were sharply elevated; this increase occurred before the development of hypotension (and the consequent requirement for greater vasopressor support), leading the authors to speculate on a possible causal link between the two events; neither endotoxin nor TNF∀ levels showed significant changes in the postoperative course. The study did not provide information on possible differential IL-6 levels in survivors and nonsurvivors.

176

Cytokines and the Abdominal Surgeon

Conclusion The priming of inflammatory cells by an initial event into a hyper-reactive state, such that a secondary stimulus elicits an exaggerated response is a phenomenon that has been documented in a variety of experimental settings. It is furthermore an essential underpinning to the “second hit” hypothesis put forward to explain the development of postinjury MOF. However, to date, there are insufficient data to validate this hypothesis in the clinical context. Priming and activation were shown to occur at a cellular level in the laboratory. It is, at this stage, unwarranted to infer that a patient can been primed by a traumatic event and subsequently go on to develop MOF as a result of an amplification of his inflammatory response brought about by an intercurrent pneumonia or a relook laparotomy; there is no denying that an infection or a secondary operation may be responsible for the deterioration of a trauma patient and the aggravation of his organ dysfunction; however, in order to validate the “two-hit” theory, it would be necessary to prove that this deterioration is mediated by a reactivated inflammatory outburst caused directly by an infectious or noninfectious insult. Available clinical studies have documented an initial peak of circulating cytokines occurring early after trauma; even this phenomenon needs to be consolidated by larger and more rigorous studies, as current findings are often inconsistent and difficult to reproduce, not least because endotoxin and cytokine assays may be unreliable. The evidence that late MOF is heralded by a subsequent peak, triggered by a secondary event, is lacking. In addition, postinjury cytokine and endotoxin patterns are not consistent with the clinical observation that late MOF is often precipitated by an infectious insult. While caution is required before the “onehit” and “two-hit” conceptual framework is adopted, larger studies in critically injured patients are necessary for the elucidation of postinjury MOF.

References 1. Saadia R, Lipman J. Multiple organ failure after trauma. No unifying hypothesis yet found. Br Med J 1996; 313:573-574. 2. Blackwell TS, Christman JW. Sepsis and cytokines: Current status. Br J Anaesth 1996; 77:110-117. 3. Biffl WL, Moore EE, Moore FA, Peterson VM. Interleukin-6 in the injured patient. Marker of injury or mediator of inflammation? Ann Surg 1996; 224:647-664. 4. Marty C, Misset B, Tamion F, Fitting C, Carlet J, Cavaillon JM. Circulating interleukin-8 concentrations in patients with multiple organ failure of septic and nonseptic origin. Crit Care Med 1994; 22:673-679. 5. Benton LD, Khan M, Greco RS. Integrins, adhesion molecules and surgical research. Surg Gynecol Obstet 1993; 177:311-327. 6. Moore FA, Moore EE. Evolving concepts in the pathogenesis of postinjury multiple organ failure. Surg Clin North Am 1996; 75:257-277. 7. Sauaia A, Moore FA, Moore EE, Lezotte DC. Early risk factors for postinjury multiple organ failure. World J Surg 1996; 20:392-400. 8. Anderson BO, Harken AH. Multiple organ failure: Inflammatory priming and activation sequences promote autologous tissue injury. J Trauma 1990; 30 (Supplement):S44-S49. 9. Forehand JR, Pabst MJ, Phillips WA, Johnston RB. Lipopolysaccharide priming of human neutrophils for an enhanced respiratory burst: Role of intracellular free calcium. J Clin Invest 1989; 83:74-83.

Cytokines, Abdominal Trauma and the “Second Hit” Phenomenon

177

10. Brom J, Konig W. Cytokine-induced (interleukins-3, -6 and -8, and tumour necrosis factor-beta) activation and deactivation of human neutrophils. Immunology 1992; 75:281-285. 11. Vercellotti GM, Yin HQ, Gustafson KS, Nelson RD, Jacob HS. Platelet-activating factor primes neutrophil responses to agonists: role in promoting neutrophil-mediated endothelial damage. Blood 1988; 71:1100-1107. 12. Biffl WL, Moore EE, Moore FA, Carl VS, Kim FJ, Franciose RJ. Interleukin-6 potentiates neutrophil priming with platelet-activating factor. Arch Surg 1994; 129:1131-1136. 13. Mileski WJ, Winn RK, Harlan JM, Rice CL. Sensitivity to endotoxin in rabbits is increased after hemorrhagic shock. J Appl Physiol 1992; 73:1146-1149. 14. Moore EE, Moore FA, Franciose RJ, Kim FJ, Biffl WL, Banerjee A. The postischemic gut serves as a priming bed for circulating neutrophils that provoke multiple organ failure. J Trauma 1994; 37:881-887. 15. West MA, Bennet T, Clair L. Reprogrammed macrophage tumor necrosis factor and interleukin-1 release with inflammatory pretreatment: Differential regulation by endotoxin and zymosan. J Trauma 1995; 39:404-410. 16. Faist E, Baue AE, Dittmer H, Heberer G. Multiple organ failure in polytrauma patients. J Trauma 1983; 23:775-787. 17. Waydhas C, Nast-Kolb D, Jochum M, Trupka A, Lenk S, Fritz H. Inflammatory mediators, infection, sepsis and multiple organ failure after severe trauma. Arch Surg 1992; 127:460-467. 18. Sauaia A, Moore FA, Moore EE, Haenel JB, Read RA. Pneumonia: Cause or symptom of postinjury multiple organ failure? Am J Surg 1993; 166:606-611. 19. Moore FA, Sauaia A, Moore EE, Haenel JB, Burch JM, Lezotte DC. Postinjury multiple organ failure: A bimodal phenomenon. J Trauma 1996; 40:501-512. 20. Moore FA, Moore EE, Poggetti R, McAnena OJ, Peterson VM, Abernathy CM, Parsons PE. Gut bacterial translocation via the portal vein: A clinical perspective with major torso trauma. J Trauma 1991; 31:629-638. 21. Partrick DA, Moore FA, Moore EE, Biffl WL, Sauaia A, Barnett CC. The inflammatory profile of interleukin-6, interleukin-8 and soluble intercellular adhesion molecule-1 in postinjury multiple organ failure. Am J Surg 1996; 172:427-431. 22. Roumen RMH, Hendriks T, van der Ven-Jongekrijg, Nieuwenhuijzen GAP, Sauerwein RW, van der Meer JWM, Goris RJA. Cytokine patterns in patients after major vascular surgery, hemorrhagic shock and severe blunt trauma. Relation with subsequent adult respiratory distress syndrome and multiple organ failure. Ann Surg 1993; 218:769-776. 23. Hoch RC, Rodriguez R, Manning T, Bishop M, Mead P, Shoemaker WC, Abraham E. Effects of accidental trauma on cytokine and endotoxin production. Crit Care Med 1993; 21:839-845. 24. Sherry RM, Cue JI, Goddard JK, Parramore JB, DiPiro JT. Interleukin-10 is associated with the development of sepsis in trauma patients. J Trauma 1996; 40:613-617. 25. Simons RK, Hoyt DB, Winchell RJ, Rose RM, Holbrook T. Elevated selectin levels after severe trauma: a marker for sepsis and organ failure and a potential target for immunomodulatory therapy. J Trauma 1996; 41:653-662. 26. Schinkel C, Faist E, Zimmer S, Piltz S, Walz A, Rose R, Höcherl E, Herndon D, Schildberg FW. Kinetics of circulating adhesion molecules and chemokines after mechanical trauma and burns. Eur J Surg 1996; 162:763-768. 27. Svoboda P, Kantorová I, Ochmann J. Dynamics of interleukin 1, 2 and 6 and tumor necrosis factor alpha in multiple trauma patients. J Trauma 1994; 36:336-340.

178

Cytokines and the Abdominal Surgeon

28. Waydhas C, Nast-Kolb D, Trupka A, Zettl R, Kick M, Wiesholler J, Schweiberer L, Jochum M. Posttraumatic inflammatory response, secondary operations and late multiple organ failure. J Trauma 1996; 40:624-631. 29. Sautner T, Götzinger P, Redl-Wenzl EM, Dittrich K, Felfernig M, Sporn P, Roth E, Függer R. Does reoperation for abdominal sepsis enhance the inflammatory host response? Arch Surg 1997; 132:250-255.

Cytokines and Mesenteric Ischemia

179

CHAPTER 16

Cytokines and Mesenteric Ischemia David A. Partrick, Ernest E. Moore and Walter L. Biffl Supported in part by National Institutes of Health Grants P50GM49222 and T32GM08315

Introduction

C

ytokines are a family of endogenous proteins or glycoproteins that exist in both nascent and rapidly transcribable form within monocytes or macrophages. These cytokines, in a complex interaction with other inflammatory mediators as well as with classic neurohumoral hormones, are capable of influencing a wide range of biologic effects, many of which are essential to normal or beneficially adaptive host responses. There are likely overlapping stimulatory and inhibitory functions between these mediators that regulate the host response to injury. However, it is becoming increasingly clear that excessive or prolonged influence of these mediators can adversely influence critical organ function and survival. Multiple organ failure (MOF) remains the most common cause of delayed mortality in surgical intensive care units.1 Once established, MOF defies our standard critical care supportive measures: mortality ranges from 40-100%, and is related directly to the number and duration of organ failures.2 However, despite intensive investigation, the pathophysiology of this syndrome remains unclear. Our trauma research center has actively sought to define and characterize postinjury MOF, and to elucidate the mechanisms responsible for its progression. Clinical studies have consistently identified early shock to be an important risk factor for postinjury MOF,3,4 and it has additionally long been recognized that the gut is particularly susceptible to shock due to disproportionate splanchnic vasoconstriction. Moreover, with the advent of gastric tonometry, persistent gastric mucosal acidemia has been shown to be a predictor of early MOF.5 Clinically, mesenteric artery revascularization of ischemic bowel has been shown to lead to significant multiple organ dysfunction, potentially as a result of intestinal ischemia and reperfusion injury.6 This chapter reviews the production of specific cytokines in response to mesenteric ischemia and reperfusion as a surrogate of shock and their role as markers of systemic inflammation as well as possible mediators of subsequent tissue injury and MOF.

Pathogenesis of Multiple Organ Failure Until the mid-1980s, it was felt that major infection or subclinical translocation of bacteria and endotoxin were the common precipitating events for postinjury MOF; however, a large number of trauma patients were subsequently found to develop Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

180

Cytokines and the Abdominal Surgeon

MOF without an identifiable source of infection.7 We performed a prospective trial to verify bacterial translocation in patients sustaining major trauma but were unable to detect elevations of endotoxin or bacteria in either the portal or systemic circulation.8 Based on this unexpected finding, we developed a global hypothesis that splanchnic hypoperfusion elaborated mediators provoking a dysfunctional inflammatory response, ultimately resulting in postinjury MOF.7,9,10 Following major tissue injury, hemorrhagic shock, and resuscitation, patients develop an early physiologic state of systemic hyperinflammation. This was traditionally referred to as the “injury stress response” and was believed to be primarily driven by counter-regulatory stress hormones.11,12 Subsequently, over the past decade a variety of inflammatory cascades and effector cells have also been implicated in this stress response,13,14 and collectively this condition has been redefined as the systemic inflammatory response syndrome (SIRS).10,15 Certain patients appear to be vulnerable (i.e., primed) such that early secondary inflammatory insults amplify SIRS resulting in MOF (“two-event” model). An attractive unifying concept in this postinjury MOF paradigm is diffuse neutrophil (PMN) mediated tissue injury.16 Previous work showed that PMNs may be either hyper- or hypoactive following major trauma, and both functional states have been implicated in the pathogenesis of MOF.17-19 Our research efforts have focused on early PMN hyperactivity. PMNs are essential in host defense and tissue repair after injury. However, excessive PMN function has been implicated in the pathogenesis of capillary leak syndromes resulting in decreased perfusion and end organ damage.16,17,20 These syndromes include the adult respiratory distress syndrome (ARDS) and MOF. PMNs are uniquely equipped to promote inflammation as well as cause direct tissue injury via oxygen-dependent (NADPH oxidase) and independent (proteolytic enzymes) mechanisms of cytotoxicity in the microenvironment of tight adhesion to endothelial cells (EC).21-24 PMN-EC adhesion and migration are necessary steps in acute inflammation; priming and activation of PMNs appear to be pivotal in determining cytotoxicity. Conceptually, in the “two-event model” of MOF, severely injured patients are resuscitated into similar states of SIRS, during which they are vulnerable to a secondary inflammatory insult that may precipitate MOF (Fig. 16.1). The correlate of this model at the cellular level is the priming and activation of PMNs.

Mesenteric Ischemia as a Model of Systemic Inflammation Shock, hypotension, and inadequate tissue perfusion have been commonly implicated as etiologic agents of MOF. Presumably, the hypoperfusion state results in cellular injury, which in turn leads to clinically definable impairment of organ function. The gut is particularly vulnerable to hypoperfusion following multisystem trauma because of disproportionate splanchnic vasoconstriction, mediated predominantly via the renin-angiotensin axis.25-27 Unrecognized flow-dependent oxygen consumption may produce ongoing mesenteric ischemia in a patient judged adequately resuscitated by standard measures.3,28 Moreover, mesenteric hypoperfusion may be compounded by the ensuing reperfusion injury with secondary elaboration of cytokines and other inflammatory agents. We developed a rodent model of mesenteric ischemia/reperfusion (gut I/R) to simulate the postinjury state, with the hypothesis that gut I/R creates a local environment that primes circulating PMNs which may become activated at distant sites, causing multiple organ injury.29 Our gut I/R model consists of 45 minutes of superior mesenteric artery (SMA) clamping followed by 6 hours of reperfusion. In preliminary studies, this was a nonlethal injury,

Cytokines and Mesenteric Ischemia

181

Fig. 16.1. The “two-event” model of postinjury multiple organ failure (MOF). The initial insult (first event) causes mesenteric hypoperfusion, resulting in systemic inflammatory response syndrome. Some patients are primed such that a second event results in progression to MOF.

but primed the animal such that an activating stimulus (low dose endotoxin) resulted in lung injury and mortality.30 Notably, pulmonary PMN sequestration temporally preceded, but was insufficient to cause, lung injury in the absence of an activating stimulus (Fig. 16.2). Our hypothesis then became that splanchnic I/R provokes distal organ injury via a mechanism involving priming of circulating PMNs in the reperfused splanchnic bed. In our gut I/R model, we had noted that circulating PMNs become primed at 2 hours of reperfusion.31 To determine whether priming occurred in the splanchnic circulation, we measured priming of PMNs drawn from the splanchnic inflow (aorta) as well as the splanchnic outflow (portal vein) after 90 min of reperfusion. We found primed PMNs (as measured by superoxide anion production) in the portal vein but not the aorta (Fig. 16.3).32 Around this same time, Johnston and colleagues33 showed that mediators of gut origin are responsible for I/R-induced injury by devising an animal model demonstrating that gut I/R could produce injury independent of the liver. Our next series of studies focused on the mechanism of PMN priming in the splanchnic circulation. We had not found evidence of bacterial translocation in our severely injured patients who developed MOF,8 and therefore we were interested in whether LPS was involved in PMN priming in our I/R model. We measured LPS levels in the plasma of laparotomized rodents vs. those subjected to splanchnic I/R, and found no difference. In addition, elimination of LPS by E5 monoclonal antibody had no effect on PMN priming after I/R.34 As we had previously suspected, something other than LPS is responsible for PMN priming and activation after mesenteric I/R. We began our search for inflammatory mediators capable of PMN priming, which were generated in the gut.

Gut-Derived Mediators of Neutrophil Priming Although numerous inflammatory mediators including various lipid and protein compounds have been identified as contributors to SIRS and MOF, this chapter will focus on the contribution of cytokines as relevant gut-derived mediators of PMN priming. These cytokines include tumor necrosis factor-∀ (TNF-∀), IL-1b, IL-6, and IL-8.

Lung MPO (units/gm of tissue)

Fig. 16.2. The sequential insult rodent model of MOF. A) Lung myeloperoxidase (MPO) reflecting pulmonary PMN sequestration. LPS alone increased pulmonary PMN sequestration to the same extent as sequential insults, gut I/R plus endotoxin. B) 125I albumin lung/blood ratio reflecting lung injury (leak). Only the sequential insults of gut I/R plus endotoxin caused demonstrable lung injury. C) Animal mortality at 18 hours of reperfusion was 40% in those subjected to gut I/R followed by endotoxin. LAP = sham laparotomy; I/R = 45 minutes of intestinal ischemia; LAP + LPS = laparotomy plus endotoxin (LPS) 6 hours later; I/R + LPS = I/R plus LPS 6 hours later. * denotes difference from LAP and I/R; ** denotes difference from LAP, I/R, and LAP + LPS (p < 0.05).

Cytokines and the Abdominal Surgeon

25

A *

15 10 5 0 I/R

LAP+LPS I/R+LPS

B 0.2

**

0.15

0.1 0.05

0 LAP

I/R

C

LAP+LPS I/R+LPS

**

40

Mortality (%)

*

20

LAP

I-125 Albumin Lung/Blood Ratio

182

30

20 10

0 LAP

I/R

LAP+LPS I/R+LPS

Tumor Necrosis Factor-a TNF is a peptide of approximately 17 kd derived principally from stimulated macrophages. Tracey et al35,36 first reported MOF in rats after the intravenous administration of recombinant TNF. Others subsequently confirmed a pathogenic role for both TNF and PMNs in this type of inflammatory organ injury. The result of PMN exposure to TNF includes increased superoxide anion (O2–) generation,37 increased PMN aggregation and adherence to endothelial cells,38 release of leukotriene B4,39 and enhanced phagocytosis.40 TNF-∀ is now well-established as a pivotal cytokine responsible for the clinical manifestations of shock induced by endotoxin or bacteremia.41-43 However, its role in the setting of mesenteric ischemia/reperfusion is only beginning to be elucidated. Caty et al44 found that intestinal ischemia in rats was

Cytokines and Mesenteric Ischemia

183

40

nmol O-2/106cells/minute

- fMLP + fMLP

30

*

20

10

0

Aorta

Portal Vein

Fig. 16.3. Ischemic reperfused gut serves as a priming bed for PMNs. Bar graph shows primed state of PMNs isolated from the aorta and portal vein following 45 minutes of intestinal ischemia and 90 minutes of reperfusion. Priming measured as superoxide anion (O2–) generation following formylmethionyl-leucyl-phenylalanine (fMLP) stimulation. There was significant priming among the PMNs exiting the gut (portal vein) contrasted to no evidence of priming at entry (aorta). * denotes difference from aorta samples (p < 0.05).

associated with TNF elevation and that reperfusion for periods of 15 and 30 minutes generated 5- to 10-fold increases in circulating TNF levels. They also found that antiTNF antibodies attenuated the increase in pulmonary microvascular permeability but did not prevent pulmonary PMN sequestration, thus implying an in vivo role for TNF generated from intestinal I/R in the process of PMN activation. Caty et al’s pioneering work was confirmed by Squadrito et al45 who also demonstrated that antibodies against TNF protected animals from end-organ injury secondary to splanchnic artery occlusion. Altavilla et al46 elaborated on this work and found that TNF specifically induced E-selectin production following intestinal I/R. Using another animal model, Welborn et al47 showed that brief visceral ischemia (30 min) produces lung injury that is dependent on TNF-∀ and IL-1. Human intestinal segments undergoing 30 min of ischemia followed by reperfusion produce TNF-∀ in concentrations capable of increasing PMN adhesion in the microcirculation via upregulation of E-selectin and ICAM-1 adhesion molecules.48 Indeed, even isolated enterocytes are capable of producing TNF-∀.49 Furthermore, Marcus et al50 demonstrated that TNF-∀ and IL-1 are able to increase cell adhesion molecule expression and endothelial permeability independent of PMN adhesion-activation. Such findings support a role for TNF-∀ and IL-1 in local as well as distant organ injury secondary to mesenteric I/R. The mechanism of induction of TNF bioactivity after intestinal I/R is not completely clear, but many studies point to the translocation of endotoxin or bacteria

184

Cytokines and the Abdominal Surgeon

across the gut mucosa as the stimulus for TNF production.51,52 However, Chaudry53 and Ayala et al54 have shown that hemorrhagic shock causes an enhanced systemic release of TNF-∀ that is not the result of endotoxin release. Expanding on this concept of endotoxin-independent release of TNF-∀, Tamion et al55 demonstrated specifically that intestinal I/R caused by hemorrhage and resuscitation leads to increased plasma levels of TNF-∀ and may be a major trigger for macrophage TNF-∀ mRNA expression. Endotoxin was not detectable in the hemorrhagic group. TNF-∀ has also been shown to be produced from the gut following hemorrhage in the absence of detectable bacteria in the portal circulation.56,57 In the previously mentioned study by Squadrito et al,45 endotoxin was undetectable in the systemic plasma of animals subjected to splanchnic artery occlusion shock. In vivo, TNF may also be produced from macrophage populations in other organs such as the liver58 or spleen59 after intestinal I/R, yielding higher levels of the potentially damaging compound. In a rodent model of intestinal I/R injury, intestinal ischemia has been shown to decrease portal blood flow 66% and reperfusion to reduce hepatic artery flow by 80% compared to sham-operated animals.60 Hepatic ischemia may thus contribute significantly to the production of proinflammatory cytokines following mesenteric I/R. Although different mechanisms may be involved in the subsequent development of MOF resulting from mesenteric ischemia, the generation of excessive amounts of TNF seems to play a pivotal role.

Interleukin-1b IL-1∀ and IL-1b are proinflammatory cytokines which share many of the proinflammatory properties of TNF-∀. The various physiologic effects of IL-1 have recently been reviewed by Dinarello.61,62 The activated mononuclear phagocyte is the major source of IL-1, although endothelial cells, keratinocytes, PMNs, and lymphocytes are also capable of producing this cytokine.63 IL-1 may be produced in large amounts from the small intestine and liver due to large resident macrophage populations in these organs. Whereas IL-1∀ and IL-1b are the product of two separate genes and display only 30% amino acid identity, they bind with equal affinity to the same receptor.64 TNF and IL-1 display many similar actions and not only stimulate their own release, but also the release of each other, thus amplifying the cascade of inflammatory mediators after injury.65,66 Production of IL-1 is significantly increased following surgical trauma, thermal injury and hemorrhagic shock.67-69 In animal models, IL-1 appears to mediate visceral ischemia-reperfusion injury in conjunction with TNF-∀. Several studies have demonstrated that mesenteric I/R promotes release of TNF-∀ and IL-1 which leads to PMN sequestration and increased capillary permeability in the lung. Welborn et al47 documented that lung injury following 30 minutes of supraceliac aortic occlusion could be blocked by pretreatment with anti-IL-1 antibody. Following hemorrhage and resuscitation, Shenkar et al70 found that mRNA levels for IL-1b increased among intraparenchymal pulmonary mononuclear cells along with IL-6. Similar to TNF-∀, IL-1b is released from reperfused human intestine and the concentration in this venous effluent is capable of increasing PMN adhesion in the microcirculation.48 This process involves the upregulation of both E-selectin and ICAM-1. The combined effects of IL-1 and TNF in this study were not additive but they were greater than either cytokine alone. Therefore, in order to reduce PMN adhesion and ameliorate I/R-induced lung injury, in vivo therapies may require antagonism of both TNF-∀ and IL-1b.

Cytokines and Mesenteric Ischemia

185

Interleukin-6 Interleukin-6 (IL-6) is considered an integral mediator of the physiologic acute phase response to injury;71 however, excessive and prolonged elevations of circulating IL-6 levels in patients following trauma, burns, and elective surgery are associated with morbidity and mortality.72-76 The precise role of IL-6 in mediating adverse postinjury events remains ill-defined; moreover, the specific source and stimuli of IL-6 are not entirely clear. IL-6 is inducible in nearly every human tissue and cell type; numerous cytokines, growth factors, hormones, neuropeptides, leukotrienes, and microbial products stimulate its production. Blood-borne IL-6 may be liberated by circulating myeloid or lymphoid cells, or may represent “spillover” from the gut, liver, wound, or other local tissues. The short half-life of IL-6 in serum (less than 1 hour) combined with the fact that IL-6 is constantly detected in blood, suggests continuous production of IL-6. An intestinal source of IL-6 following surgery and trauma has been suggested. Baigrie et al77 found significantly higher elevations of IL-6 in inferior mesenteric veins than systemic veins during and after aortic crossclamping in aortic surgery. Similarly, Wortel et al78 detected IL-6 in higher concentrations in portal vs. peripheral venous blood in patients after pancreatic and hepatic resections. Deitch et al56 found higher IL-6 concentrations in portal vs. cardiac blood after hemorrhagic shock in rats. Thus, IL-6 appears to be generated in the gut after systemic insults. In fact, Meyer et al79 have demonstrated IL-6 production in intestinal mucosa in vivo. Tamion et al55 found that intestinal I/R upon hemorrhage and resuscitation is a major trigger for IL-6, as well as TNF-∀ mRNA expression. The complex interaction between IL-6 and TNF has been further elucidated by Yao et al.80 Using a monoclonal antibody to cause TNF blockade, they demonstrated that the release of IL-6 in the setting of acute mesenteric I/R may be mediated in part via TNF-dependent mechanisms. IL-6 is capable of promoting hyperinflammation and organ injury. IL-6 has been shown to increase endothelial permeability in the absence of PMNs.81 In addition, accumulating evidence suggests that IL-6 mediates inflammation by modulating the functional repertoire of the mature PMN. Circumstantial evidence comes from a recent study showing coordinate increases in circulating concentrations of IL-6 and PMN elastase after aortic cross-clamping,82 raising the possibility that IL-6 may be involved in reperfusion injury. CABG has similarly been shown to cause an increase in IL-6, and we have recently documented enhanced PMN priming with respect to superoxide anion production and elastase release in patients undergoing elective CABG (Partrick et al, unpublished data). Oka et al83 demonstrated that patients with postoperative complications had higher concentrations of IL-6 than those without complications; furthermore, PMN priming for release of elastase was also higher in the cohort with complications. Redmond et al84 demonstrated that open cholecystectomy patients, shown to have higher IL-6 concentrations, had increased PMN chemotaxis and priming for superoxide anion release compared with laparoscopic cholecystectomy patients. In addition, the open cholecystectomy group had more septic complications. Our Trauma Center Research Program has focused on the priming and activation sequences of human PMNs, and has examined IL-6 as a potential mediator in promoting postinjury MOF. We have studied the effects of IL-6 on PMN cytotoxic function in vitro. IL-6 by itself does not prime PMNs for superoxide release; however, when quiescent PMNs are incubated with IL-6 and then exposed to PAF, the PMNs are primed for superoxide anion production (Fig. 16.4).85 The concentration of PAF required for this effect is 100-fold lower than that required to

186

Cytokines and the Abdominal Surgeon

Superoxide, nmol/min per 106 PMNs

200

150

* 100

50

0

Control

IL-6

PAF

IL-6 and PAF

Neutralized IL-6 and PAF

Fig. 16.4. Priming for superoxide anion production by PMNs following incubation with IL-6 (10 ng/mL for 60 minutes), platelet-activating factor (PAF) (0.1 ng/mL for 5 minutes), IL-6 followed by PAF, and neutralized IL-6 followed by PAF (for 60 minutes). * denotes difference from the control PMNs, IL-6 and PAF-incubated PMNs, and PMNs incubated with neutralized IL-6 and PAF (p < 0.01).

prime PMNs without prior IL-6 stimulation. Mullen et al86 have similarly demonstrated synergy between IL-6 and TNF-∀ in PMN priming. Thus, these data suggest that IL-6 can sensitize PMNs to the effects of a subsequent inflammatory mediator. This may be particularly important in a local site of inflammation where IL-6 is concentrated; PMNs can be made more sensitive to inflammatory mediators and their cytotoxic potential can be enhanced, further exacerbating the inflammatory process. IL-6 may also affect PMN-mediated inflammation by modulating apoptosis, or programmed cell death. PMN flux into the interstitial space is unidirectional, and PMNs must meet their fate in situ. The physiological mechanism of obligatory PMN clearance is believed to be apoptosis. Senescent, apoptotic PMNs initiate a signal prompting macrophage phagocytosis. This cytophagocytosis resolves inflammation without activation of the macrophage or exocytosis of cytotoxic PMN contents. Recently, this process has been identified as an important mechanism in the normal resolution of inflammatory processes and the limitation of inflammatory tissue injury. We have found that IL-6 in vitro, in concentrations approximating those at inflammatory sites, delays PMN apoptosis, resulting in a larger population of nonapoptotic (surviving) PMNs with a greater collective capacity for superoxide production than untreated PMNs (Fig. 16.5).87 This IL-6 mediated delay of PMN apoptosis could postpone the clearance of PMNs from a site of inflammation, pro-

Cytokines and Mesenteric Ischemia

187

A 50

% Survival

40

*

*

30

20

0 0

1

10

100

IL-6, ng/mL

Fig. 16.5. A) Dose-response curve for the effects of IL-6 on PMN apoptosis. PMN survival (viable PMNs without morphologic evidence of apoptosis divided by the total number of PMNs counted in the sample) was assessed after 24 hours in culture. IL-6 at doses of 10 and 100 ng/mL resulted in greater PMN survival than that in untreated controls. * denotes p < 0.05. B) Effects of IL-6 and heatdenatured IL-6 (X-IL-6) on PMN apoptosis compared with untreated (buffer) PMNs. IL-6 delayed PMN apoptosis; this effect was abrogated by denaturation of the IL-6 protein before incubation (X-IL-6). * denotes difference from buffer and X-IL-6 (p < 0.05).

B 60

*

% Survival

50 40 30 20 10 0 Buffer

IL-6

X-IL-6

long release of reactive oxygen metabolites and proteolytic enzymes, and aggravate PMN-mediated tissue injury and organ failure. Furthermore, an important proximal step in the process of PMN-mediated cytotoxicity involves the adherence of PMNs to the endothelium. Endothelial ICAM-1 is a critical mediator of PMN adherence and subsequent migration through the endothelium. We have identified ICAM-1 as a pivotal regulator of PMN-mediated cytotoxicity.88 Recent evidence has implicated IL-6 in upregulating ICAM-1 following ischemia/reperfusion injury in the heart.89 Thus, IL-6 can promote PMN adhesion to the endothelium, sensitize PMNs to priming by endogenous inflammatory mediators, and delay PMN apoptosis. These collective effects could facilitate PMN-mediated tissue injury and the exacerbation of the hyperinflammatory response after injury. Although a direct causal link has not been made between IL-6 and MOF, there is an increasing body of evidence suggesting that IL-6 is a pivotal component of the hyperinflammatory cytokine cascade driving SIRS. Simms and D’Amico90 studied patients with SIRS, and found that PMNs had upregulated functions during SIRS;

188

Cytokines and the Abdominal Surgeon

this upregulation persisted in patients who developed MOF, but not in those who did not develop MOF. An anti-IL-6 antibody was effective in reducing PMN oxidative responses, implicating IL-6 as the primary PMN stimulant in the circulation. Gennari et al91 demonstrated that IL-6 neutralization decreased IL-6 concentrations and bacterial translocation, and improved bacterial killing; in addition, a correlation between IL-6 concentrations and survival time was observed. While TNF and IL-1 may also be generated in the gut and have been strongly implicated in tissue injury in animal models, IL-6 has been the cytokine most consistently associated with adverse clinical outcomes. For example, we measured IL-6 and TNF concentrations in patients sustaining major trauma.8 IL-6 concentrations were elevated immediately after injury, but TNF concentrations were not higher than uninjured controls at any time. Furthermore, in a separate population of severely injured patients, we found that IL-6 and IL-8 concentrations correlated with the development of postinjury MOF (Fig. 16.6).76 Hoch et al92 found elevations of IL-6 and IL-8 in proportion to the severity of injury in trauma patients, but detected no endotoxin or IL-1, and no elevations of TNF-∀. Meade et al93 similarly found elevated concentrations of IL-6 and IL-8 after injury, but could not detect IL-1b, TNF-∀ or endotoxin at any time. In this study, both IL-6 and IL-8 concentrations remained elevated for more prolonged periods in patients who developed ARDS. In patients undergoing elective surgery, DiPadova et al94 found early increases in IL-6 after surgery, but did not see increases in IL-1 or TNF. Marty et al95 showed that IL-6 concentrations were higher in the nonsurvivors than in the surviving MOF patients, and that IL-6 elevations were similar in MOF of septic and nonseptic origin. In liver transplant patients with MOF and suspected bacterial infections, Rosenbloom et al96 found IL-6, but not IL-1b or TNF, to correlate with organ failure.

Interleukin-8 Interleukin-8 (IL-8) has been identified as a unique neutrophil-activating cytokine97 which increases endothelial permeability,98 regulates the transvascular migration of PMNs during acute inflammatory responses,99 and primes neutrophils for enhanced superoxide production.100,101 In fact, IL-8 has been referred to as a chemokine due to its strong chemotactic properties. IL-8 is produced by monocytes,102 endothelial cells,103 fibroblasts,104 and neutrophils.105 Similar to IL-6, excessive and prolonged elevations of circulating IL-8 levels in patients following trauma, burns, and elective surgery are associated with ARDS, MOF and increased mortality.76,92,95,106,107 Investigators have shown that IL-8 gene expression is induced in human endothelial cells and monocytes by hypoxia or anoxia-hyperoxia.108,109 Compared to IL-6 or TNF, however, there is far less data concerning mesenteric ischemia and IL-8. Using a murine I/R model of supra-mesenteric artery occlusion, Tsuruma et al110 documented that plasma IL-8 concentrations increased 6 to 12 hours following reperfusion. This elevation was correlated to the ischemia time and degree of gut mucosal damage. In a further group of experiments, this group of investigators found a mechanism of heat-shock protein-induced protection against intestinal I/R injury.111 This involved inhibition of IL-8 production and, theoretically, the subsequent prevention of PMN activation and chemotaxis.

Clinical Implications Current evidence strongly implicates mesenteric I/R and the production of cytokines in the genesis of malignant hyperinflammation resulting in MOF. The pre-

Cytokines and Mesenteric Ischemia

189

A 2000

*

IL-6, pg/ml

1500

* 1000

*

500

*

0 12

36

60

84

132

Time Postinjury (hours) B

Fig. 16.6. Serial plasma levels of IL6 and IL-8 in 27 high-risk trauma patients (n = 9 MOF, n = 18 nonMOF) (A) Plasma concentrations of IL-6 in MOF patients ( ) vs. ) over the non-MOF patients ( 5 day postinjury time period. IL-6 levels were elevated in MOF patients compared with non-MOF patients at 12, 36, 84 and 132 hours postinjury. (B) Plasma concentrations of IL-8 in MOF patients ( ) vs. non-MOF patients ( ) over the 5 day postinjury time period. IL-8 levels were elevated in MOF patients compared with non-MOF patients at 12, 36 and 84 hours postinjury. Results are shown as mean ± SEM. * denotes difference from non-MOF patients (p < 0.05).

1250

IL-8, pg/ml

1000

*

750

*

500

*

250

0 12

36

60

84

132

Time Postinjury (hours)

cise mechanistic events and relevant temporal associations remain to be elucidated. Until we develop agents able to interfere with the inflammatory cascade, we must focus on prevention and supportive care. The principles of aggressive resuscitation,28 early fracture fixation and tissue debridement,112 and early enteral nutritional support113 have all been shown to be beneficial in attenuating the inflammatory response and MOF. Resuscitation should be targeted towards augmenting the splanchnic circulation. Recently, Ivatury and colleagues114 showed that resuscitation of patients to supranormal levels of oxygen delivery and consumption did not consistently normalize gastric mucosal pH; however, when normalization of gastric mucosal pH was used as an endpoint of resuscitation, there was improved survival. This suggests that direct monitoring of the splanchnic circulation is preferable to systemic goal-oriented resuscitation. Evolving immune-enhancing enteral diets appear promising by enabling the gut to withstand stress.113

190

Cytokines and the Abdominal Surgeon

Various other compounds have been investigated to block the proinflammatory effects of these cytokines following mesenteric I/R. Soluble receptors for TNF-∀115 and a receptor antagonist for IL-1116 have been cloned and administered parenterally in an attempt to neutralize these cytokines and prevent PMN activation.117 Lung reperfusion injury in rabbits has been prevented by the use of a monoclonal antibody against IL-8,118 but we are not aware of any clinical trials using this antibody. However, numerous other inflammatory mediators are produced from experimental mesenteric I/R (PAF, LTB4, IL-6, IL-8, IF-!, and C5a), as well as hemorrhage/resuscitation in humans clinically. It is therefore unlikely that the use of a single drug would be effective in preventing endogenous PMN priming/activation sequences unless administered very early postinjury. Another group of cytokines typically generated from the TH2-type lymphocyte cells are notable for their immunosuppressive properties including IL-4, IL-10, and IL-13.119 For example, IL-10 has previously been shown to modulate TNF, IL-1b, and IL-8 cytokine gene expression and production in PMNs,120 as well as selectively upregulate IL-1 receptor antagonist.121 IL-10 has recently been shown to reduce the severity of local and systemic inflammation in a murine model of intestinal I/R when given before or after reperfusion injury.122 It specifically reduced the circulating levels of TNF-∀ and IL-6. IL-10 may, therefore, be an important physiologic regulator of cytokine production from w and its potential role in inflammatory responses needs to be further investigated. Benzydamine has been shown to cause a marked reduction of serum levels of TNF-∀ and IL-1b in a murine model of septic shock, whereas IL-6 was unaffected.123 In a rat model of hemorrhage and resuscitation, pentoxifylline down-regulated TNF and IL-6 levels.124 Following splanchnic artery occlusion and reperfusion, lipocortin 1 has more recently been shown to inhibit PMN migration and accumulation into reperfused tissues, and thus may show promise in ameliorating the outcome of I/R-induced shock.125 In summary, there are many exciting interventions currently being investigated that may protect patients from mesenteric I/R-mediated complications.

References 1. Sauaia A, Moore FA, Moore EE et al. Epidemiology of trauma deaths: a reassessment. J Trauma 1995; 38:185-193. 2. Knaus WA, Wagner DP. Multiple systems organ failure: Epidemiology and prognosis. Crit Care Clin 1989; 5:221-232. 3. Shoemaker WC, Appel PL, Kram HB. Tissue oxygen debt as a determinant of lethal and nonlethal postoperative organ failure. Crit Care Med 1988; 16:1117-1120. 4. Moore FA, Moore EE, Poggetti RS et al. Postinjury shock and early bacteremia: A lethal combination. Arch Surg 1992; 127:893-897. 5. Chang MC, Cheatham ML, Nelson LD et al. Gastric tonometry supplements information provided by systemic indicators of oxygen transport. J Trauma 1994; 37(3):488-494. 6. Harward TRS, Brooks DL, Flynn TC et al. Multiple organ dysfunction after mesenteric artery revascularization. J Vasc Surg 1993; 18:459-469. 7. Goris JA, Boekhoerst TPA, Nuytinck JKS et al. Multiple-organ failure: Generalized autodestructive inflammation? Arch Surg 1985; 120:1109-1115. 8. Moore FA, Moore EE, Poggetti R et al. Gut bacterial translocation via the portal vein: A clinical perspective with major torso trauma. J Trauma 1991; 31:629-638. 9. Nuytinck HKS, Offermans XJMW, Kubat K et al. Whole-body inflammation in trauma patients: An autopsy study. Arch Surg 1988; 123:1519-1524.

Cytokines and Mesenteric Ischemia

191

10. Moore FA, Moore EE. Evolving concepts in the pathogenesis of postinjury multiple organ failure. Surg Clin North Am 1995; 75:257-277. 11. Reichlin S. Neuroendocrine-immune interactions. N Engl J Med 1993; 329: 1246-1253. 12. Lilly MP, Gann DS. The hypothalamic-pituitary-adrenal-immune axis. A critical assessment. Arch Surg 1992; 127:1463-1475. 13. Baue AE. The horror autotoxicus and multiple-organ failure. Arch Surg 1992; 127:1451-1462. 14. Waydhas C, Nast-Kolb D, Jochum M et al. Inflammatory mediators, infection, sepsis, and multiple organ failure after severe trauma. Arch Surg 1992; 127:460-467. 15. Demling R, LaLonde C, Saldinger P. Multiple-organ dysfunction in the surgical patient: Pathophysiology, prevention and treatment. Curr Probl Surg 1993; 30:347-414. 16. Vedder NB, Futy BW, Winn RK et al. Role of neutrophils in generalized reperfusion injury associated with resuscitation from shock. Surgery 1989; 106:509-516. 17. Windsor ACJ, Mullen PG, Fowler AA et al. Role of the neutrophil in adult respiratory distress syndrome. Br J Surg 1993; 80:10-17. 18. Moore FA, Moore EE, Read RA. Postinjury multiple organ failure: Role of extrathoracic injury and sepsis in adult respiratory distress syndrome. New Horiz 1993; 1:538-549. 19. Botha AJ, Moore FA, Moore EE et al. Postinjury neutrophil priming and activation: An early vulnerable window. Surgery 1995; 118:358-365. 20. Welbourn CRB, Goldman G, Paterson IS et al. Pathophysiology of ischemia reperfusion injury: central role of the neutrophil. Br J Surg 1991; 78:651-655. 21. Weiss SJ. Tissue destruction by neutrophils. N Engl J Med 1989; 320:365-376. 22. Ward PA, Varani J. Mechanisms of neutrophil-mediated killing of endothelial cells. J Leukoc Biol 1990; 48:97-102. 23. Wahl SM, Feldman GM, McCarthy JB. Regulation of leukocyte adhesion and signaling in inflammation and disease. J Leukoc Biol 1996; 59:789-796. 24. Partrick DA, Moore EE, Moore FA et al. Lipid mediators upregulate CD11b and prime for concordant superoxide and elastase release in human neutrophils. J Trauma 1997; 43:297-303. 25. McNeill JR, Stark RD, Greenway CV. Intestinal vasoconstriction after hemorrhage: roles of vasopressin and angiotensin. Am J Physiol 1970; 219:1342-1347. 26. Adar R, Franklin A, Spark RF et al. Effect of dehydration and cardiac tamponade on superior mesenteric artery flow: Role of vasoactive substances. Surgery 1976; 79(5):534-543. 27. Bailey RW, Bulkley GB, Hamilton SR et al. Protection of the small intestine from nonocclusive mesenteric ischemic injury due to cardiogenic shock. Am J Surg 1987; 153:108-116. 28. Moore FA, Haenel JB, Moore EE et al. Incommensurate oxygen consumption in response to maximal oxygen availability predicts postinjury multiple organ failure. J Trauma 1992; 33:58-67. 29. Poggetti RS, Moore FA, Moore EE et al. Liver injury is a reversible neutrophilmediated event following gut ischemia. Arch Surg 1992; 127:175-179. 30. Koike K, Moore FA, Moore EE et al. Endotoxin after gut ischemia/reperfusion causes irreversible lung injury. J Surg Res 1992; 52(6):656-662. 31. Koike K, Moore EE, Moore FA et al. Phospholipase A2 inhibition decouples lung injury from gut ischemia-reperfusion. Surgery 1992; 112:173-180. 32. Franciose RJ, Moore EE, Moore FA et al. Postischemic gut is the priming bed for PMNs. Surgical Forum 1993; XLIV:142-144.

192

Cytokines and the Abdominal Surgeon

33. Johnston TD, Fischer R, Chen Y et al. Lung injury from gut ischemia: Insensitivity to portal blood flow diversion. J Trauma 1993; 35:508-511. 34. Koike K, Moore EE, Moore FA et al. Gut ischemia/reperfusion produces lung injury independent of endotoxin. Crit Care Med 1994; 22:1438-1444. 35. Tracey KJ, Beutler B, Lowry SF et al. Shock and tissue injury induced by recombinant human cachectin. Science 1986; 234:470-474. 36. Tracey KJ, Fong Y, Hesse DG et al. Anti-cachectin/TNF monoclonal antibodies prevent septic shock during lethal bacteraemia. Nature 1987; 330:662-666. 37. Tsujimoto M, Yokota S, Vilcek J et al. Tumor necrosis factor provokes superoxide anion generation from neutrophils. Biochem Biophys Res Commun 1986; 137:1094-1098. 38. Pober JS, Cotran RS. Cytokines and endothelial cell biology. Physiol Rev 1990; 70:427-451. 39. Meyer JD, Yurt RW, Duhaney R et al. Tumor necrosis factor-enhanced leukotriene B4 generation and chemotaxis in human neutrophils. Arch Surg 1988; 123:1454-1458. 40. Shalby MR, Aggarwal BB, Rinderknecht E et al. Activation of human polymorphonuclear neutrophil function by interferon-gamma and tumor necrosis factor. J Immunol 1985; 135:2069. 41. Remick DG, Kunkel RG, Larrick JW et al. Acute in vivo effects of human recombinant tumor necrosis factor. Lab Invest 1987; 56:583-590. 42. Starnes HF, Warren RS, Jeevanandam M et al. Tumor necrosis factor and the acute metabolic response to tissue injury in man. J Clin Invest 1988; 82:1321-1325. 43. Strieter RM, Kunkel SL, Bone RC. Role of tumor necrosis factor-a in disease states and inflammation. Crit Care Med 1993; 21:S447-S463. 44. Caty MG, Guice KS, Oldham KT et al. Evidence for tumor necrosis factor-induced pulmonary microvascular injury after intestinal ischemia-reperfusion injury. Ann Surg 1990; 212(6):694-700. 45. Squadrito F, Altavilla D, Ioculano M et al. Passive immunization with antibodies against tumor necrosis factor (TNFa) protects from the lethality of splanchnic artery occlusion shock. Circ Shock 1992; 37:236-244. 46. Altavilla D, Squadrito F, Canale P et al. Tumor necrosis factor induces E-selectin production in splanchnic artery occlusion shock. Am J Physiol 1995; 268: H1412-H1417. 47. Welborn MB, Douglas WG, Abouhamze Z et al. Visceral ischemia-reperfusion injury promotes tumor necrosis factor (TNF) and interleukin-1 (IL-1) dependent organ injury in the mouse. Shock 1996; 6(3):171-176. 48. Wyble CW, Desai TR, Clark ET et al. Physiologic concentrations of TNFa and IL-1b released from reperfused human intestine upregulate E-Selectin and ICAM-1. J Surg Res 1996; 63:333-338. 49. Ogle CK, Guo X, Hasselgren PO et al. The gut as a source of inflammatory cytokines after stimulation with endotoxin. Eur J Surg 1997; 163:45-51. 50. Marcus BC, Wyble CW, Hynes KL et al. Cytokine-induced increases in endothelial permeability occur after adhesion molecule expression. Surgery 1996; 120:411-417. 51. Jiang J, Bahrami S, Leichtfried G et al. Kinetics of endotoxin and tumor necrosis factor appearance in portal and systemic circulation after hemorrhagic shock in rats. Ann Surg 1995; 221(1):100-106. 52. Yao YM, Sheng ZY, Yu Y et al. The potential etiologic role of tumor necrosis factor in mediating multiple organ dysfunction in rats following intestinal ischemiareperfusion injury. Resuscitation 1995; 29:157-168. 53. Chaudry IH. Cellular mechanisms in shock and ischemia and their correction. Am J Physiol 1983; 245:R117-R134.

Cytokines and Mesenteric Ischemia

193

54. Ayala A, Perrin MM, Meldrum DR et al. Hemorrhage induces an increase in serum TNF which is not associated with elevated levels of endotoxin. Cytokine 1990; 2:170-174. 55. Tamion F, Richard V, Lyoumi S et al. Gut ischemia and mesenteric synthesis of inflammatory cytokines after hemorrhagic or endotoxic shock. Am J Physiol 1997; 273:G314-G321. 56. Deitch EA, Xu D, Franko L et al. Evidence favoring the role of the gut as a cytokinegenerating organ in rats subjected to hemorrhagic shock. Shock 1994; 1(2):141-145. 57. Mainous MR, Ertel W, Chaudry IH et al. The gut: A cytokine generating organ in systemic inflammation? Shock 1995; 4(3):193-199. 58. Colletti LM, Remick DG, Burtch GD et al. Role of tumor necrosis factor-a in the pathophysiologic alterations after hepatic ischemia/reperfusion injury in the rat. J Clin Invest 1990; 85:1936-1943. 59. Cohen S, Lahat N, Kinarty A et al. Ex vivo secretion of tumor necrosis factor and interleukin-2 by rat splenocytes after intestinal ischemia and shock. Lymphokine and Cytokine Research 1992; 11(5):215-220. 60. Turnage RH, Kadesky KM, Myers SI et al. Hepatic hypoperfusion after intestinal reperfusion. Surgery 1996; 119(2):151-160. 61. Dinarello CA, Wolff SM. The role of interleukin-1 in disease. N Engl J Med 1993; 328:106-113. 62. Dinarello CA. Biologic basis for interleukin-1 in disease. Blood 1996; 87(6): 2095-2147. 63. Dinarello CA. Biology of interleukin 1. FASEB J 1988; 2:108-115. 64. March CJ, Mosley B, Larsen A et al. Cloning, sequence and expression of two distinct human interleukin-1 complementary DNAs. Nature 1985; 315:641-647. 65. Dinarello CA, Cannon JG, Wolff SM et al. Tumor necrosis factor (cachectin) is an endogenous pyrogen and induces production of interleukin-1. J Exp Med 1986; 163:1433-1450. 66. Le J, Vilcek J. Tumor necrosis factor and interleukin-1: Cytokines with multiple overlapping biological activities. Lab Invest 1987; 56:234-248. 67. Abraham E, Richmond NJ, Chang YH. Effects of hemorrhage on interleukin-1 production. Circ Shock 1988; 25:33-40. 68. Grzelak I, Olszewski WL, Rowinski W. Blood mononuclear cell production of IL-1 and IL-2 following moderate surgical trauma. Eur Surg Res 1989; 21:114-122. 69. O’Riordain MG, Mannick JA, Rodrick ML et al. Modulation of macrophage hyperactivity improves survival in a burn-sepsis model. Arch Surg 1992; 127:152-158. 70. Shenkar R, Coulson WF, Abraham E. Hemorrhage and resuscitation induce alterations in cytokine expression and the development of acute lung injury. Am J Respir Cell Mol Biol 1994; 10(3):290-297. 71. Castell JV, Gomez-Lechon MJ, David M et al. Interleukin-6 is the major regulator of acute phase protein synthesis in adult human hepatocytes. FEBS Lett 1989; 242:237-239. 72. Damas P, Ledoux D, Nys M et al. Cytokine serum level during sevre sepsis in human IL-6 as a marker of severity. Ann Surg 1992; 215:356-362. 73. Drost AC, Burleson DG, Cioffi WG et al. Plasma cytokines after thermal injury and their relationship to infection. Ann Surg 1993; 218:74-78. 74. Roumen RMH, Hendriks T, Ven-Jongekrijg Jvd et al. Cytokine patterns in patients after major vascular surgery, hemorrhagic shock, and severe blunt trauma. Relation with subsequent adult respiratory distress syndrome and multiple organ failure. Ann Surg 1993; 218(6):769-776. 75. Meduri GU, Headley S, Kohler G et al. Persistent elevation of inflammatory cytokines predicts a poor outcome in ARDS. Chest 1995; 107:1062-1073.

194

Cytokines and the Abdominal Surgeon

76. Partrick DA, Moore FA, Moore EE et al. The inflammatory profile of interleukin-6, interleukin-8, and soluble intercellular adhesion molecule-1 in postinjury multiple organ failure. Am J Surg 1996; 172(5):425-431. 77. Baigrie RJ, Lamont PM, Whiting S et al. Portal endotoxin and cytokine responses during abdominal aortic surgery. Am J Surg 1993; 166:248-251. 78. Wortel CH, deventer SJHv, Aarden LA et al. Interleukin-6 mediates host defence responses induced by abdominal surgery. Surgery 1993; 114:564-570. 79. Meyer TA, Wang J, Tiao GM et al. Sepsis and endotoxemia stimulate intestinal interleukin-6 production. Surgery 1995; 118:336-342. 80. Yao YM, Bahrami S, Redl H et al. IL-6 release after intestinal ischemia/reperfusion in rats is under partial control of TNF. J Surg Res 1997; 70:21-26. 81. Maruo N, Morita I, Shirao M et al. IL-6 increases endothelial permeability in vitro. Endocrinology 1992; 131:710-714. 82. Kawamura T, Wakusawa R, Okada K et al. Elevation of cytokines during open heart surgery with cardiopulmonary bypass: Participation of interleukin 8 and 6 in reperfusion injury. Can J Anesth 1993; 40:1016-1021. 83. Oka Y, Murata A, Nishijima J et al. Enhanced attachment and elastase-releasing capacity of neutrophils after surgery. Am J Surg 1994; 167:405-411. 84. Redmond HP, Watson RWG, Houghton T et al. Immune function in patients undergoing open vs. laparoscopic cholecystectomy. Arch Surg 1994; 129:1240-1246. 85. Biffl WL, Moore EE, Moore FA et al. Interleukin-6 potentiates neutrophil priming with platelet-activating factor. Arch Surg 1994; 129:1131-1136. 86. Mullen PG, Windsor ACJ, Walsh CJ et al. Tumor necrosis factor-a and interleukin6 selectively regulate neutrophil function in vitro. J Surg Res 1995; 58:124-130. 87. Biffl WL, Moore EE, Moore FA et al. Interleukin-6 delays neutrophil apoptosis. Arch Surg 1996; 113:24-30. 88. Barnett CC, Moore EE, Moore FA et al. Intercellular adhesion molecule-1 promotes neutrophil-mediated cytotoxicity. Surgery 1995; 118:171-176. 89. Kukielka GL, Youker KA, Hawkins HK et al. Regulation of ICAM-1 and IL-6 in myocardial ischemia; effect of reperfusion. Ann NY Acad Sci 1994; 574:258-270. 90. Simms HH, D’Amico R. Polymorphonuclear leukocyte dysregulation during the systemic inflammatory response syndrome. Blood 1994; 83:1398-1407. 91. Gennari R, Alexander JW, Pyles T et al. Effects of animurine interleukin-6 on bacterial translocation during gut-derived sepsis. Arch Surg 1994; 129:1191-1197. 92. Hoch RC, Rodriguez R, Manning T et al. Effects of accidental trauma on cytokine and endotoxin production. Crit Care Med 1993; 21(6):839-845. 93. Meade P, Shoemaker WC, Donnelly TJ et al. Temporal patterns of hemodynamics, oxygen transport, cytokine activity, and complement activity in the developent of adult respiratory distress syndrome after severe injury. J Trauma 1994; 36(5): 651-657. 94. DiPadova F, Pozzi C, Tondre MJ et al. Selective and early increase of IL-1 inhibitors, IL-6 and cortisol after elective surgery. Clin Exp Immunol 1991; 85:137-142. 95. Marty C, Misset B, Tamion F et al. Circulating interleukin-8 concentrations in patients with multiple organ failure of septic and nonseptic origin. Crit Care Med 1994; 22(4):673-679. 96. Rosenbloom AJ, Pinsky MR, Bryant JL et al. Leukocyte activation in the peripheral blood of patients with cirrhosis of the liver and SIRS: Correlation with serum interleukin-6 levels and organ dysfunction. JAMA 1995; 274:58-65. 97. Baggiolini M, Walz A, Kunkel SL. Neutrophil-activating peptide-1/interleukin 8, a novel cytokine that activates neutrophils. J Clin Invest 1989; 84:1045-1049. 98. Biffl WL, Moore EE, Moore FA et al. Interleukin-8 increases endothelial permeability independent of neutrophils. J Trauma 1995; 39(1):98-103.

Cytokines and Mesenteric Ischemia

195

99. Huber AR, Kunkel SL, Todd RF et al. Regulation of transendothelial neutrophil migration by endogenous interleukin-8. Science 1991; 254:99-102. 100. Daniels RH, Finnen MJ, Hill ME et al. Recombinant human nomocyte IL-8 primes NADPH-oxidase and phospholipase A2 activation in human neutrophils. Immunol 1992; 75:157-163. 101. Wozniak A, Betts WH, Murphy GA et al. Interleukin-8 primes human neutrophils for enhanced superoxide anion production. Immunology 1993; 79:608-615. 102. Schroder JM, Mrowietz U, Morita E et al. Purification and partial biochemical characterization of a human monocyte-derived neutrophil-activating peptide that lacks interleukin 1 activity. J Immunol 1987; 139:3474-3483. 103. Gimbrone MA, Obin MS, Brock AF et al. Endothelial interleukin-8: A novel inhibitor of leukocyte-endothelial interactions. Science 1989; 246:1601-1603. 104. Larsen CG, Anderson AO, Oppenheim JJ et al. Production of interleukin-8 by human dermal fibroblasts and keratinocytes in response to interleukin-1 or tumor necrosis factor. Immunology 1989; 68:31-36. 105. Bazzoni F, Cassatella MA, Rossi F et al. Phagocytosing neutrophils produce and release high amounts of the neutrophil-activating peptide 1/interleukin 8. J Exp Med 1991; 173:771-774. 106. Donnelly SC, Strieter RM, Kunkel SL et al. Interleukin-8 and development of adult respiratory distress syndrome in at-risk patient groups. Lancet 1993; 341:643-647. 107. Vindenes H, Ulvestad E, Bjerknes R. Increased levels of circulating interleukin-8 in patients with large burns: relation to burn size and sepsis. J Trauma 1995; 39(4):635-640. 108. Metinko AP, Kunkel SL, Standiford TJ et al. Anoxia-hyperoxia induce monocytederived interleukin-8. J Clin Invest 1992; 90:791-798. 109. Karakurum M, Shreeniwas R, Chen J et al. Hypoxic induction of interleukin-8 gene expression in human endothelial cells. J Clin Invest 1994; 93:1564-1570. 110. Tsuruma T, Yagihashi A, Hirata K et al. Evaluation of plasma IL-8 (CINC) concentration during ischemia and after reperfusion in the small intestine. Transplant Proc 1996; 28(3):1917-1918. 111. Tsuruma T, Yagihashi A, Matsuno T et al. The heat-shock protein 70 family reduces ischemia/reperfusion injury in small intestine. Transplant Proc 1996; 28(5):2629-2630. 112. Border JR, Hassett J, LaDuca J et al. The gut origin septic states in blunt multiple trauma (ISS = 40) in the ICU. Ann Surg 1987; 206:427-448. 113. Moore FA, Moore EE, Kudsk KA et al. Clinical benefits of an immune-enhancing diet for early postinjury enteral feeding. J Trauma 1994; 37:607-615. 114. Ivatury RR, Simon RJ, Havriliak D et al. Gastric mucosal pH and oxygen delivery and oxygen consumption indices in the assessment of adequacy of resuscitation after trauma: A prospecitve, randomized study. J Trauma 1995; 39(1):128-136. 115. Schall RJ, Lewis M, Koller KJ et al. Molecular cloning and expression of a receptor for human tumor necrosis factor. Cell 1990; 61:361-370. 116. Carter DB, Deibel MR, Dunn CJ et al. Purification, cloning, expression and biological characterization of an interleukin-1 receptor antagonist protein. Nature 1990; 344:633-638. 117. Dinarello CA, Gelfand JA, Wolff SM. Anticytokine strategies in the treatment of the systemic inflammatory response syndrome. JAMA 1993; 269:1829-1835. 118. Sekido N, Mukaida N, Harada A et al. Prevention of lung reperfusion injury in rabbits by a monoclonal antibody against interleukin-8. Nature 1993; 365:654-657. 119. Faist E, Schinkel C, Zimmer S. Update on the mechanisms of immune suppression of injury and immune modulation. World J Surg 1996; 20:454-459.

196

Cytokines and the Abdominal Surgeon

120. Cassatella MA, Meda L, Bonora S et al. Interleukin 10 (IL-10) inhibits the release of proinflammatory cytokines from human polymorphonuclear leukocytes. Evidence for an autocrine role of tumor necrosis factor and IL-1b in mediating the production of IL-8 triggered by lipopolysaccharide. J Exp Med 1993; 178:2207-2211. 121. Cassatella MA, Meda L, Gasperini S et al. Interleukin 10 (IL-10) upregulates IL-1 receptor antagonist production from lipopolysaccharide-stimulated human polymorhonuclear leukocytes by delaying mRNA degradation. J Exp Med 1994; 179:1695-1699. 122. Lane JS, Todd KE, Lewis MPN et al. Interleukin-10 reduces the systemic inflammatory response in a murine model of intestinal ischemia/reperfusion. Surgery 1997; 122:288-294. 123. Sironi M, Pozzi P, Polentarutti N et al. Inhibition of inflammatory cytokine production and protection against endotoxin toxicity by benzydamine. Cytokine 1996; 8(9):710-716. 124. Wang P, Ba ZF, Morrison MH et al. Mechanism of the beneficial effects of pentoxifylline on hepatocellular function after trauma hemorrhage and resuscitation. Surgery 1992; 112:451-458. 125. Cuzzocrea S, Tailor A, Zingarelli B et al. Lipocortin 1 protects against splanchnic artery occlusion and reperfusion injury by affecting neutrophil migration. J Immunol 1997; 159(10):5089-5097.

The Significance of Intestinal Cytokines

197

CHAPTER 17

The Significance of Intestinal Cytokines Per-Olof Hasselgren Supported in part by Grant #8510 from the Shriners of North America

Introduction

T

he important role of the intestinal mucosa in the metabolic and immunologic responses to trauma, sepsis and endotoxemia has become increasingly recognized during recent years. The intestine has many important functions in addition to its traditional role in absorption of nutrients. Among those functions, mucosal production of proteins, including gut hormones1-4 and acute-phase proteins,5 is important for the metabolic response to both local and systemic inflammation. An intact mucosal membrane is thought to prevent bacterial translocation and may reduce the risk for multiple organ failure in critically ill patients.6 An additional important function of the intestine is cytokine production. Cytokines are involved in the intestinal response to injury, sepsis and local inflammation both because they are produced in the mucosa in these conditions and because they may mediate several of the metabolic and functional changes seen in intestinal mucosa during inflammation. The significance of intestinal cytokines is summarized in Figure 17.1. Cytokines are produced in the intestinal mucosa in response to local inflammation, such as inflammatory bowel disease (IBD), and in response to systemic inflammation, such as seen during sepsis and endotoxemia and following trauma. Intestinal cytokines may be released into the circulation and reach the liver through the portal vein. Intestinal cytokines may therefore influence metabolic events in the liver and may also have systemic effects. Perhaps more importantly, cytokines produced in the intestinal mucosa may have local effects exerted through autocrine or paracrine mechanisms. In this chapter, the production of intestinal cytokines in IBD and during systemic inflammation is reviewed together with cellular and molecular mechanisms of mucosal cytokine production. In addition, the significance of mucosal cytokines is discussed with regard to their local effects in the intestine and to their potential systemic effects.

Cytokines in Inflammatory Bowel Disease A large number of studies in patients with IBD have provided evidence that both ulcerative colitis (UC) and Crohn’s disease (CD) are associated with upregulated Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

198

Cytokines and the Abdominal Surgeon

Fig. 17.1. Both local and systemic inflammation are associated with increased cytokine production in the intestine. Intestinal cytokines can be released into the circulation and may influence the liver or have systemic effects. In addition, intestinal cytokines can have local effects exerted through autocrine or paracrine mechanisms.

mucosal production of pro-inflammatory and regulatory cytokines. Experimental animal models of colitis have been used to examine the regulation of mucosal cytokine production and the role of the cytokines in the pathogenesis of IBD. The role of intestinal cytokines in IBD was reviewed extensively recently7 and only a relatively limited overview is given here.

Studies in Patients In early reports, evidence was found that IL-1 production by peripheral blood monocytes was increased in patients with IBD.8-10 In subsequent studies, IL-1 production by lamina propria mononuclear cells isolated from surgical specimens was increased in patients with IBD, supporting the concept that the local mucosal production of IL-1 is increased in these patients.11 In several studies, the mucosal IL-1 levels correlated with the degree of mucosal inflammation.12,13 There is evidence that mononuclear cells of the lamina propria are the major source of IL-1 in patients with IBD with no or only minor contribution from enterocytes.14 The synthesis of both IL-1∀ and IL-1# is increased in lamina propria cells from patients with IBD (Fig 17.2). Interestingly, the amount of IL-1 produced by isolated lamina propria cells and measured in fresh mucosal biopsies was higher in patients with CD than in patients with UC.8 Similarly, IL-1 mRNA levels were increased in mucosal biopsies from pa-

The Significance of Intestinal Cytokines

199

CD UC

Fig. 17.2. IL-1∀ and IL-1# concentrations in intestinal lamina propria mononuclear cells from 6 control patients, 4 patients with Crohn’s disease (CD) and 4 patients with ulcerative colitis (UC). The IL-1∀ concentrations were significantly (p < 0.01) greater in cells from CD and UC patients than in cells from control patients. Reproduced with permission from: Youngman KR et al. Localization of intestinal interleukin 1 activity and protein and gene expression to lamina propria cells. Gastroenterology 1993; 104:749-758.

tients with inactive CD but not UC.15 Although somewhat controversial, these observations suggest that the pattern of cytokine expression may be different in CD and UC and could possibly be used to differentiate between the two diseases. Closely related to the mucosal production of IL-1 is the production of IL-1 receptor antagonist (IL-1ra). This is an anti-inflammatory cytokine produced by the same cells that produce IL-1 and downregulates the activity of IL-1. Recent studies suggest that the balance between IL-1 and IL-1ra production may be altered with a decreased ratio of IL-1ra/IL-1 in mucosa of patients with IBD.16 The mucosal production of both IL-1 and IL-1ra was increased in patients with IBD but the relative increase in IL-1 production was greater than that of IL-1ra, thus explaining why the IL-1ra/IL-1 ratio was decreased in inflammatory bowel disease (Fig 17.3). In addition to IL-1, there is evidence that IL-6 production is increased in patients with IBD. Elevated serum levels of IL-6 and increased IL-6 production by peripheral blood mononuclear cells were reported in IBD patients17 and some studies suggest that this response may be more common in patients with CD than in patients with UC.18 In more recent studies, production of IL-6 by isolated lamina propria cells was stimulated in patients with active IBD and increased expression of IL-6 mRNA further

200

Cytokines and the Abdominal Surgeon

Fig. 17.3. Total IL-1 (+) and IL-1ra levels in freshly isolated intestinal mucosal cells from 16 control, 12 CD, and 12 UC patients. The IL-1ra/IL-1 ratio was lower in patients with IBD than in control patients, consistent with a mucosal imbalance between IL-1 and IL-1ra production. * p < 0.001 vs control; ** p < 0.01 vs controls. Reprodueced with permission from: CasiniRaggi V et al. Mucosal imbalance of IL-1 and IL-1 receptor antagonist in inflammatory bowel disease. A novel mechanism of chronic intestinal inflammation. J Immu 1995; 154:2434-2440.

supported the concept of mucosal production of IL-6 in these patients.19 Similar to IL-1, studies showed a good correlation between mucosal IL-6 levels and the degree of inflammation.20 There is evidence that both IL-6 and the IL-6 receptor (IL-6R) are expressed in cells of the intestinal mucosa. In a recent study, the cellular expression of IL-6R was

The Significance of Intestinal Cytokines

201

increased both at the mRNA and protein levels in actively involved mucosa of patients with IBD21 further illustrating the important role of IL-6 in IBD. Mucosal IL-6 is probably produced by multiple cell types. In a study by Kusugami et al,19 lamina propria macrophages produced significant amounts of IL-6 in unstimulated cultures and these cells accounted for the majority of increased IL-6 production by isolated lamina propria mononuclear cells from patients with active IBD. In addition to macrophages, intestinal T cells and B cells may also be a source of IL-6.19 Recent studies from our22 and other laboratories23 provided evidence that the enterocyte as well expresses IL-6. Immunohistochemistry and Northern blot analysis showed that IL-6 protein and transcripts were present in colonic epithelial cells of patients with active IBD.19 Among the pro-inflammatory cytokines, the evidence for increased production in patients with IBD is strongest for IL-1 and IL-6, whereas the involvement of TNF is more controversial, both with regards to circulating24 and tissue levels.25 When semi-quantitative PCR was applied, no differences in the gene expression of TNF were found in mucosal samples from patients with active or inactive IBD or control patients.20 In another study, TNF transcripts could not be detected by PCR in the majority of mucosal samples from patients with active IBD.15 High TNF concentrations in stools from patients with CD and UC may originate from neutrophils transmigrating into the intestinal lumen.26 In addition to pro-inflammatory cytokines, there is evidence that several regulatory cytokines, in particular IL-2 and the interferons, are also involved in IBD. Studies in isolated lamina propria mononuclear cells suggest that CD is an IL-2 hyperresponsive condition, whereas UC is an IL-2 hypo-responsive condition.27 Two interesting clinical observations are supportive of this concept. In patients with active CD and who developed AIDS, the bowel disease went into remission when the number of circulating IL-2 producing cells fell.28 In contrast, patients with CD in remission had flare-ups when they were treated with IL-2 for malignancies.29 The major functions of interferon-∀ IFN-∀ and IFN-# are their antiviral properties whereas IFN-! is a key cytokine for mediating immune functions.30 Furthermore, IFN-! enhances both the expression and effects of certain pro-inflammatory cytokines. In recent studies from our laboratory, IFN-! interacted synergistically with IL-1# in stimulating IL-6 production in cultured human intestinal epithelial cells.22 Several studies in patients with CD have shown that lamina propria mononuclear cells release IFN-! at an increased rate and that IFN-! mRNA is expressed in these cells.31 In contrast to CD, IFN-! production or mRNA levels do not seem to be upregulated in UC.31 Thus, interferons may be more important for the mucosal immune process in CD than in UC. The mechanisms by which IFN-! influences the inflammatory response in CD are complex and likely include recruitment of peripheral macrophages to the mucosa, increased epithelial expression of MHC class II antigens, enhanced intercellular adhesion molecule I expression by both endothelium and mononuclear cells, the presence of IL-2 receptor-positive macrophages and the presence of multinucleated giant cells.32 The multiple biological effects of IFN in the intestinal mucosa are summarized in Figure 17.4. In addition to the cytokines described above, there is evidence that other cytokines as well are involved in IBD. Among them, the chemotactic cytokine chemokines, in particular IL-8 and monocyte chemoattractant peptide-1 (MCP-1), and the colonystimulating factors, are especially important. The role of these cytokines in IBD was reviewed recently.7

202

Cytokines and the Abdominal Surgeon HLA-DR Intestinal Epithelial Barrier

tight-junctions IFN-!!

IFN-!!

IF

N∀

IFN-!!

∃ PERIPHERAL M∃ RECRUITMENT

MGC GENERATION

ICAM-1 POLARIZATION

Fig. 17.4 Proposed mechanisms by which interferons modulate the inflammatory response in IBD. EC: endothelial cells; ICAM-I: intercellular adhesion molecule-I; MGC: multi-nucleated giant cell. Reproduced with permission from: Pallone F et al. The interferon system in inflammatory bowel disease. In: Fiocchi C, ed. Cytokines in Inflammatory Bowel Disease. Austin: RG Landes Co, 1996:57-67.

Studies in Experimental Animals A number of experimental models of intestinal inflammation have been described, including inflammation induced by the administration of dextran sulfate sodium, trinitrobenzenesulfonic acid/ethanol, acetic acid, indomethacin, and formalin-immune complexes.33 By using such models and treating animals with cytokine antibodies or cytokine receptor antagonists, the important role of IL-1 in intestinal inflammation has been confirmed. For example, early treatment with IL-1ra of rabbits with experimental colitis induced by formalin-immune complexes reduced the inflammatory index, edema and necrosis of the intestinal wall.34 Similar beneficial effects of IL-1ra were noticed in other models of intestinal inflammation as well, supporting the important role of IL-1 in mucosal inflammation35 (Fig. 17.5). In contrast to IL-1, the role of other cytokines in experimental intestinal inflammation is not as well documented. Interestingly, a protective role of the anti-inflammatory cytokine IL-10 was suggested from studies in which IL-10 knock out mice developed chronic enterocolitis, predominantly in the duodenum and proximal colon.36 The proposed mechanism of colitis in this model was overproduction of proinflammatory cytokines, such as IL-1, and of IFN-!, due to a lack of suppression by IL-10. In other experiments, administration of IL-10 attenuated the inflammatory response in animals with experimental colitis.

Intestinal Cytokines During Sepsis and Endotoxemia In addition to local inflammation, as seen in IBD, systemic inflammation as well is associated with increased mucosal cytokine production, further supporting the

The Significance of Intestinal Cytokines

203

Fig. 17.5. Experimental colitis was induced in rats by subserosal injection of purified group A streptococcal peptidoglycan-polysaccharide. Groups of rats were treated with human recombinant IL1ra (open bars) or solvent hatched bars). Treatment with IL-1ra attenuated both the acute and chronic inflammatory response. * p < 0.05. Reproduced with permission from: McCall RD et al. Tissue interleukin 1 and interleukin-1 receptor antagonist expression in enterocolitis in resistant and susceptible rats. Gastroenterology 1994; 106:960-972.

role of the gut in the metabolic and immunologic response to a number of different conditions, including sepsis, endotoxemia,37 hemorrhagic shock,38 and burn injury.39 In this section of the chapter, evidence for upregulated cytokine production in these conditions is reviewed. In addition, the cellular sources and the mechanisms of mucosal cytokine production and the significance of intestinal cytokines are discussed.

Mucosal Cytokines During Sepsis and Endotoxemia

In studies by Mester et al,40 tissue levels of IL-1 were increased in the mucosa of small intestine during endotoxemia in mice and increased IL-1 mRNA levels41 suggested that mucosal IL-1 production during endotoxemia was regulated at the transcriptional level. Similarly, Nathens et al42 found that TNF-∀ and TNF mRNA levels were increased in the small bowel after endotoxin injection suggesting that the expression of this cytokine as well is upregulated during systemic inflammation. More recent studies in our laboratory have focused on mucosal IL-6 production during sepsis and endotoxemia. Among the pro-inflammatory cytokines, IL-6 may be particularly important considering its role in both localintestinal and systemic immune responses and its metabolic effects during severe infection. Mucosal IL-6 has been implicated in the induction of immunoglobulin A secretion by Peyer’s patch B cells, in local macrophage differentiation, and in T-cell proliferation.43 In addition, the acute phase response in both liver44 and enterocytes5 is regulated by IL-6.

204

Cytokines and the Abdominal Surgeon

Fig. 17.6. IL-6 levels in jejunal mucosa of mice at different time points after subcutaneous injection of 10 mg/kg of endotoxin (solid line) or corresponding volumes of sterile saline (broken line). * p < 0.05 vs control. Reproduced with permission from: McCall RD et al. Tissue interleukin 1 and interleukin-1 receptor antagonist expression in enterocolitis in resistant and susceptible rats. Gastroenterology 1994; 106:960-972.

When mice were injected with endotoxin, IL-6 levels in jejunal mucosa increased rapidly and were elevated 4-5 times above control levels after 1 h37 (Fig. 17.6). This response was associated with increased expression of IL-6 mRNA, suggesting that the mucosal production of IL-6 during endotoxemia was regulated at the transcriptional level. When sepsis was induced in mice by cecal ligation and puncture, mucosal IL-6 protein and mRNA levels were again increased, further supporting the role of this cytokine in the response to systemic inflammation. In the same study,37 the increase in mucosal IL-6 levels was blunted following treatment of endotoxemic mice with the nitric oxide synthase inhibitor N-nitro-L-arginine or indomethacin, suggesting that nitric oxide and prostaglandins may be involved in the regulation of mucosal IL-6 production during endotoxemia. The results described above, together with experimental studies in hemorrhagic shock38 and burn injury39 support the role of the intestine as a cytokine producing organ in systemic inflammation. Thus, in this respect, the gut is an active participant, rather than a passive bystander, in the immunologic and metabolic response to critical illness.

The Enterocyte Produces Cytokines Although cells in the lamina propria and submucosa probably participate in the response to systemic inflammation, there is evidence that the enterocyte as well may be an important source of cytokines in these conditions. For example, Ogle et al39 reported that enterocytes isolated from normal or burned guinea pigs produced IL-1,

The Significance of Intestinal Cytokines

205

Fig. 17.7. Release of IL-6 by cultured IEC-6 cells (a rat intestinal epithelial cell line) following treatment of the cells with endotoxin1 g/ml) for 1-4 days. Control cells were cultured without addition of endotoxin to the culture medium. * p < 0.05 vs control. Reproduced with permission from: Meyer TA et al. Sepsis and endotoxemia stimulate intestinal interleukin-6 production. Surgery 1995; 118:336-342.

IL-6 and TNF after stimulation with LPS and that enterocytes from burned guinea pigs produced more IL-6 than enterocytes from unburned controls. Other reports support the concept that the enterocyte may be a significant source of cytokines. In studies using immunohistochemistry, evidence was found that human intestinal epithelial cells express both IL-6 and IL-6R.45 When different cultured intestinal epithelial cell lines were examined, mRNA for various cytokines were detected, including IL-1∀, IL-1#, IL-8 and IL-10.46 Thus, enterocytes may be a source of both pro-inflammatory, anti-inflammatory and regulatory cytokines. The ability of enterocytes to produce cytokines is particularly important considering the strategic position of enterocytes between the intestinal bacterial flora and the internal milieu.

Regulation of Cytokine Production In the Enterocyte In previous experiments, we examined the regulation of IL-6 production in cultured IEC-6 cells, a rat intestinal epithelial cell line.47 In those studies, IL-6 production was stimulated by LPS in a dose- and time-dependent manner (Figs. 17.7 and 17.8) and prostaglandin E2 interacted with LPS in a synergistic fashion. Addition of indomethacin to the cultured enterocytes blunted the effect of LPS, supporting the role of prostaglandin in IL-6 production. PCR showed that IL-6 mRNA was not constitutively expressed in the IEC-6 cells but was induced by LPS, suggesting that LPS upregulated the IL-6 production at the transcriptional level (Fig. 17.9). Although other studies support the concept that endotoxin can induce cytokine production through a direct effect on enterocytes,23 contradictory results have been

206

Cytokines and the Abdominal Surgeon

Fig. 17.8. Release of IL-6 by IEC-6 cells cultured for 3 days in the presence of different concentrations of endotoxin. Endotoxin stimulated IL-6 production in a dose-dependent manner. * p < 0.05 vs 0 ∝g/ml endotoxin. From Meyer et al47 with permission.

Fig. 17.9. PCR products from IEC-6 cells cultured in medium with endotoxin (LPS, 1 ∝g/ ml) for 1, 2 or 3 days or in medium without endotoxincontrol). #actin was used as standard. Results suggest that IL-6 mRNA is not constitutively expressed in the IEC-6 cells but is induced by endotoxin. From Meyer et al47 with permission.

The Significance of Intestinal Cytokines

207

reported. Thus, IL-6 production was not increased in cultured human intestinal epithelial cells following treatment with endotoxin in studies by Jung et al48 and Panja et al.49 In recent experiments in our laboratory, only a small increase in IL-6 production was noticed following treatment of cultured Caco-2 cells with endotoxin and this effect was not seen until the endotoxin concentration was raised to 10 g/ml.22 The reason for these contradictory results is not clear but may reflect species differences, the IEC-6 cell line being a rat cell line and the Caco-2 cell line being a human intestinal epithelial cell line. Because the mucosal production of multiple cytokines is increased during sepsis and endotoxemia, it is possible that they interact in the regulation of the production of inflammatory mediators by the enterocyte in a paracrine or autocrine manner. We recently tested the effects of different cytokines on the IL-6 production in cultured Caco-2 cells and found that IL-1 stimulated IL-6 production by these cells in a dose and time-dependent manner.22 In addition to stimulating IL-6 production, IL-1 upregulated IL-6 mRNA levels in the Caco-2 cells, suggesting that the increased IL-6 production was regulated at the transcriptional level. In the same experiments, IL-6 production by the Caco-2 cells was not influenced by TNF-∀ or IL-6. Similar results were reported by Eckmann et al46 who found that IL-8 production by Caco-2 cells was stimulated by IL-1 but not by TNF-∀. Thus, IL-1 may be a key mediator of enterocyte cytokine production, at a far as least IL-6 and IL-8 are concerned. In a recent study, Caco-2 cells were grown in a two compartment system, allowing for the separate study of the apical and basolateral membrane responses to stimulus.23 In these experiments, Caco-2 cells produced IL-6 and TNF in response to live E. coli and the magnitude of this response was polarized with a maximal response noticed on the side of the bacterial challenge. In other studies, exposure of the apical surface of cultured enterocytes to bacteria resulted in basolateral secretion of IL-8.50 These results further support the important role of the enterocyte in its unique position at the interface between the host and the environment in the intestinal lumen and suggest that exposure of the enterocyte to intestinal bacteria or bacterial products can signal cytokine production. Furthermore, stimulation of the enterocyte at the basolateral membrane may also result in cytokine production, which may be of importance during systemic sepsis and other inflammatory conditions

Molecular Mechanisms of Cytokine Production in Enterocytes In recent studies we have begun to explore the molecular mechanisms and intracellular pathways involved in enterocyte cytokine production. In particular, experiments have focused on the role of the nuclear factor kappa BNF-B in IL-6 production by Caco-2 cells. NF%-B belongs to the Rel family of transcriptional activator proteins and is involved in the activation of a large number of genes important for the inflammatory response.51 NF%- B is activated by a number of pathogens and agents, including IL-1, TNF and endotoxin. NF-%B exists in the cytoplasm as an inactive complex consisting predominantly of two DNA binding subunits, p50 and p65 and the inhibitory protein I%B.51 Upon stimulation, I%B is phosphorylated and ubiquitinated and subsequently dissociated from the p50 and p65 subunits, resulting in activated NF-%B. The activated p50/p65 heterodimer is translocated to the nucleus where it binds to NF-%B binding sites and activates gene transcription. A role of NF-%B in IL-6 production in B cells, T cells and monocytes has been documented in previous studies.52 In recent studies we have found evidence that

208

Cytokines and the Abdominal Surgeon

Fig 17.10. NF%B activity, determined with electrophoretic mobility shift assay, in nuclear extracts from Caco-2 cells treated with IL-1. IL-1 activated NF%B after 30 min. Unpublished results from our laboratory.

NF-%B may be involved in the regulation of IL-6 production in the enterocyte as well.53 In initial experiments, utilizing electrophoretic mobility shift assay, we found evidence that IL-1 activates NF-%B in the Caco-2 cell (Fig. 17.10). The increased NF-%B activity was associated with a rapid reduction of I B-%levels determined by Western blot, suggesting that breakdown of this inhibitory component is involved in the activation of NF-%B in IL-1%-stimulated Caco-2 cells. In contrast, the time course for the breakdown of I B-% suggested that this species of I B is not involved in the activation of NF-%B under these experimental conditions. When IL-1%-stimulated Caco-2 cells were treated with different NF-%B inhibitors, IL-6 production was reduced in a dose-dependent manner, supporting the concept that NF-%B regulates IL-6 production in human intestinal epithelial cells. The postulated role of NF- B in the IL-1%-induced IL-6 production in Caco-2 cells is summarized in Figure 17.11.

The Significance of Intestinal Cytokines Intestinal production of cytokines have several important biological and clinical implications (see Fig. 17.1). First, cytokines produced in the intestine may be released into the circulation and mediate some of the deleterious effects of systemic inflammation. Although increased bacterial translocation6 and activated neutrophils54 are probably important factors in the development of multiple organ failure, it may be speculated that cytokines originating in the gut may also participate in the development of multiple organ failure. Of particular interest is the fact that cytokines released from the intestine first reach the liver through the portal vein. It is possible that intestinal cytokines participate in the priming of Kupffer cells, making them more sensitive to the effects of endotoxin, or directly stimulate Kupffer cells to produce cytokines. The cytokines may also influence the hepatocytes and IL-6 released from the intestine together with IL-6 from the Kupffer cells may upregulate acute phase protein synthesis in the liver. Another example of a potential systemic effect of intestinal cytokines is the “gutneutrophil priming hypothesis” recently proposed by Moore et al.54 According to this hypothesis, inflammatory factors, including IL-6, produced in the gut activate

The Significance of Intestinal Cytokines

209 Fig 17.11. Simplified scheme of IL-6 production in the enterocyte. IL-1 activates NF%B, induces the expression of IL-6 transcripts and stimulates the production of IL-6 protein. Based on results from recent experiments in Caco-2 cells in our laboratory.22,53 In other experiments, the IL-1%-induced IL-6 production was blocked by various NF-%B inhibitors, further supporting the role of NF-%B in enterocyte IL-6 production.

neutrophils traversing the splanchnic circulation and these “intestinally primed” neutrophils contribute to distant organ injury, in particular lung injury. In addition to systemic effects, mucosal cytokines may have local effects on enterocytes and other cells in the mucosa. These effects of the cytokines may be exerted in an autocrine or paracrine manner. The increased IL-6 production in cultured Caco-2 cells following treatment with IL-1 described above offers one example of a potential local role of cytokines in the intestinal mucosa. Production of IL-8 following stimulation of cultured intestinal epithelial cells with IL-1 further supports the role of mucosal cytokines in the local production of intestinal cytokines.46 A better understanding of this cytokine network in the intestinal mucosa will be important for the development of new therapeutic modalities both in patients with local IBD and in patients with systemic inflammation. Mucosal cytokines may also be important for the local production of other proteins, most notably acute-phase proteins. Although the liver is the major source of acute-phase proteins, there is evidence that extrahepatic sites as well may contribute to acute-phase protein synthesis. Previous studies by Molmenti et al5 provided evidence that the enterocyte can synthesize several of the acute-phase proteins and that this biosynthetic activity is regulated by pro-inflammatory cytokines, including IL-6. Thus, in this respect, the enterocyte function is similar to that of the liver cell. Further evidence for a regulatory role of IL-6 in mucosal protein synthesis during endotoxemia was provided in a recent study from our laboratory.55 In that study, endotoxemia in mice was associated with increased mucosal protein synthesis, similar to previous reports of increased mucosal protein synthesis in septic and

210

Cytokines and the Abdominal Surgeon

endotoxemic rats.1,2 A similar response was seen when mice were injected with IL-6. Treatment of mice with anti-IL-6 antibody resulted in a paradoxical increase in mucosal IL-6 levels together with increased mucosal protein synthesis. In addition to regulating protein turnover, there is evidence that cytokines, in particular IL-1, may influence amino acid especially glutamine transport and metabolism in intestinal cells. For example, treatment of rats with IL-1 50 g/kg body weight resulted in a significant inhibition of glutamine uptake by the intestine.56 In the same report, administration of rTNF-∀ to rat sat doses of 50 or 150 g/kg did not influence intestinal glutamine uptake or mucosal glutaminase activity, suggesting that glutamine transport and metabolism in the intestine is more sensitive to IL-1 than TNF. Similar results were reported by Argiles et al57 who found that brushborder transport by system A, measured as absorption of 14C-amino-isobutyric acid, was inhibited by IL-1 but not by TNF. Cytokine-induced production of adhesion molecules with recruitment of inflammatory cells58,59 increased vascular permeability,60 stimulation of inflammatory and immunoactive cells in the mucosa, and enhanced production of mucosal prostaglandins and leukotriens61 are additional examples of potentially significant biological effects of mucosal cytokines. Furthermore, recent studies suggest that increased permeability is associated with high levels of IL-6.62 Thus, intestinal cytokines may be significant both for metabolic and functional changes in the mucosa.

References 1. von Allmen D, Hasselgren PO, Higashiguchi T et al. Increased intestinal protein synthesis during sepsis and following the administration of tumor necrosis factor or interleukin-1. Biochem J 1992; 286:585-5589. 2. Higashiguchi T, Noguchi Y, O’Brien W et al. Effect of sepsis on mucosal protein synthesis in different parts of the gastrointestinal tract in rats. Clin Sci 1994; 87:207-211. 3. Zamir O, Hasselgren PO, Higashiguchi T et al. Effect of sepsis or cytokine administration on release of gut peptides. Am J Surg 1992; 163:181-185. 4. Higashiguchi T, Noguchi Y, Noffsinger A et al. Sepsis increases production of total secreted proteins, vasoactive intestinal peptide and peptide YY in isolated rat enterocytes. Am J Surg 1994; 168:251-256. 5. Molmenti EP, Ziambaras T, Perlmutter DH. Evidence for an acute phase response in human intestinal epithelial cells. J Biol Chem 1993; 268:14116-14124. 6. Deitch EA. Bacterial translocation of the gut flora. J Trauma 1990; 30:S185-S189. 7. Fiocchi C. Cytokines in Inflammatory Bowel Disease. Austin: RG Landes Co, 1996. 8. Satsangi J, Wolstencroft RA, Cason J et al. Interleukin 1 in Crohn’s disease. Clin Exp Immunol 1987; 67:594-605. 9. Mazlam MZ, Hodgson HJ. Peripheral blood monocyte cytokine production and acute phase response in inflammatory bowel disease. Gut 1992; 33:773-778. 10. Nakamura M, Saito H, Kasanaki J et al. Cytokine production in patients with inflammatory bowel disease. Gut 1992; 33:933-937. 11. Mahida YR, Wu K, Jewell DP. Enhanced production of interleukin 1 by mononuclear cells isolated from mucosa with active ulcerative colitis or Crohn’s disease. Gut 1989; 30:835-838. 12. Ligumsky M, Simon PL, Karmeli F et al. Role of interleukin 1 in inflammatory bowel disease—enhanced production during active disease. Gut 1990; 31:686-689. 13. Brynskov J, Tvede N, Andersen CB et al. Increased concentrations of interleukin-1, interleukin-2, and soluble interleukin-2 receptors in endoscopical mucosal biopsy specimens with active inflammatory bowel disease. Gut 1992; 33:55-58.

The Significance of Intestinal Cytokines

211

14. Youngman KR, Simon PL, West GA et al. Localization of intestinal interleukin-1 activity and protein and gene expression to lamina propria cells. Gastroenterology 1993; 104:749-758. 15. Isaacs KL, Sartor RB, Haskill S. Cytokine messenger RNA profiles in inflammatory bowel disease mucosa detected by polymerase chain reaction amplification. Gastroenterology 1992; 103:1587-1595. 16. Casini-Raggi V, Kam L, Chong YTJ et al. Mucosal imbalance of IL-1 and IL-1 receptor antagonist in inflammatory bowel disease. A novel mechanism of chronic intestinal inflammation. J Immunol 1995; 154:2434-2440. 17. Suzuki Y, Saito H, Kasanuki J et al. Significant increase of interleukin 6 production in blood mononuclear leukocytes obtained from patients with active inflammatory bowel disease. Life Sci 1990; 47:2193-2197. 18. Mahida YR, Kurlac L, Gallagher A et al. High circulating concentrations of interleukin-6 in active Crohn’s disease but not ulcerative colitis. Gut 1991; 32:1531-1534. 19. Kusugami K, Fukuatsu A, Tanimoto M et al. Elevation of interleukin-6 in inflammatory bowel disease is macrophage- and epithelial cell-dependent. Dig Dis Sci 1995; 40:949-959. 20. Stevens C, Walz G, Singaram C et al. Tumor necrosis factor-∀, interleukin-1 , and interleukin-6 expression in inflammatory bowel disease. Dig Dis Sci 1992; 37:818-826. 21. Mitsuyama K, Toyonaga A, Sasaki E et al. Soluble form of interleukin-6 receptor in inflammatory bowel disease. Gastroenterology 1993; 104:A747. 22. Parikh AA, Salzman AL, Fischer JE et al. Interleukin-1 and interferon-! regulate interleukin-6 production in human intestinal epithelial cells. Submitted. 23. Michalsky MP, Deitch EA, Ding J et al. Interleukin-6 and tumor necrosis factor production in an enterocyte cell model Caco-2 during exposure to Escherichia coli. Shock 1997; 7:139-146. 24. Hyams JS, Treem WR, Eddy E et al. Tumor necrosis factor- is not elevated in children with inflammatory bowel disease. J Pediatr Gastroenterol Nutr 1991; 12:233-236. 25. Nielsen OH, Brynskov J, Bendtzen K. Circulating and mucosal concentrations of tumour necrosis factor and inhibitor(s) in chronic inflammatory bowel disease. Dan Med Bull 1993; 40:247-249. 26. Braegger CP, Nicholls S, Murch SH et al. TNF-∀ in stool as a marker of intestinal inflammation. Lancet 1992; 339:89-91. 27. Matsuura T, Kusugami K, Morise K et al. Interleukin-2 and interleukin-2 receptor in inflammatory bowel disease. In: Fiocchi C, ed. Cytokines in Inflammatory Bowel Disease. Austin: RG Landes Co, 1996:41-56. 28. James SP. Remission of Crohn’s disease after human immunodeficiency virus infection. Gastroenterology 1988; 95:1667-1669. 29. Sparano JA, Brandt LJ, Dutcher JP et al. Symptomatic exacerbation of Crohn disease after treatment with high-dose interleukin-2. Ann Intern Med 1993; 118:617-618. 30. Farrar MA, Schreiber RD. The molecular cell biology of interferon-gamma and its receptor. Annu Rev Immunol 1993; 11:571-611. 31. Breese E, Braegger CP, Corrigan CJ et al. Interleukin-2- and interferon-gammasecreting T cells in normal and diseased human intestinal mucosa. Immunology 1993; 78:127-131. 32. Pallone F, Fais S, Boirivant M. The interferon system in inflammatory bowel disease. In: Fiocchi C, ed. Cytokines in Inflammatory Bowel Disease. Austin: RG Landes Co, 1996:57-67.

212

Cytokines and the Abdominal Surgeon

33. van Tol EAF, Sartor RB. Cytokine networks in animal models of intestinal inflammation. In: Fiocchi C, Ed. Cytokines in Inflammatory Bowel Disease. Austin: RG Landes Co., 1996:203-224. 34. Cominelli F, Nast CC, Duchini A et al. Recombinant interleukin-1 receptor antagonist blocks the proinflammatory activity of endogenous interleukin-1 in rabbit immune colitis. Gastroenterology 1992; 103:65-71. 35. McCall RD, Haskill S, Zimmermann EM et al. Tissue interleukin 1 and interleukin1 receptor antagonist expression in enterocolitis in resistant and susceptible rats. Gastroenterology 1994; 106:960-972. 36. Kuhn R, Lohler J, Rennick D et al. Interleukin-10 deficient mice develop chronic enterocolitis. Cell 1993; 75:263-274. 37. Meyer TA, Wang JJ, Tiao GM et al. Sepsis and endotoxemia stimulate intestinal interleukin-6 production. Surgery 1995; 118:336-342. 38. Deitch EA, Xu D, Franko L et al. Evidence favoring the role of the gut as a cytokinegenerating organ in rats subjected to hemorrhagic shock. Shock 1994; 1:141-146. 39. Ogle CK, Mao JX, Wu JZ et al. The production of tumor necrosis factor, interleukin-1, interleukin-6 and prostaglandin E2 by isolated enterocytes and gut macrophages: Effect of lipopolysaccharide and thermal injury. J Burn Care Rehab 1994; 15:470-477. 40. Mester M, Tompkins RG, Gelfand JA et al. Intestinal production of interleukin-1 during endotoxemia in the mouse. J Surg Res 1993; 54:584-591. 41. Mester M, Tompkins RG, Burke JF et al. Lipopolysaccharide inducible IL-1 gene in the crypts of Lieberkuhn. Gastroenterology 1991; 100:A600. 42. Nathens AB, Ding JW, Marshall JC et al. The gut as a cytokine generating organ: small bowel TNF production during systemic endotoxemia. Presented at the 14th Annual Meeting of the Surgical Infection Society, Toronto, Ontario, Canada, April 29, 1994. 43. Beagley KW, Eldridge JH, Lee F et al. Interleukins and IgA synthesis. Human and murine interleukin-6 induce high rate IgA secretion in IgA committed B cells. J Exp Med 1989; 169:2133-2148. 44. Heinrich PC, Castell JV, Anders T. Interleukin-6 and the acute phase response. Biochem J 1990; 265:621-636. 45. Shirota K, LeDuy L, Yuan S et al. Interleukin-6 and its receptor are expressed in human intestinal epithelial cells. Virchows Archiv B Cell Pathol 1990; 58:303-308. 46. Eckmann L, Jung HC, Schürer-Maly A et al. Differential cytokine expression by human intestinal epithelial cell lines: regulated expression of interleukin-8. Gastroenterology 1993; 105:1689-1697. 47. Meyer TA, Noguchi Y, Ogle CK et al. Endotoxin stimulates interleukin-6 production in intestinal epithelial cells. A synergystic effect with prostaglandin E2. Arch Surg 1994; 129:1290-1295. 48. Jung HC, Eckmann L, Yang SK et al. A distinct array of proinflammatory cytokines is expressed in human colon epithelial cells in response to bacterial invasion. J Clin Invest 1995; 95:55-65. 49. Panja A, Siden E, Mayer L. Synthesis and regulation of accessory/proinflammatory cytokines by intestinal epithelial cells. Clin Exp Immunol 1995; 100:298-305. 50. Eckmann L, Kagnoff M, Fierer J. Epithelial cells secrete the chemokine interleukin8 in response to bacterial entry. Infect Immuno 1993; 61:4569-4574. 51. Baeuerle PA, Henkle T. Function and activation of NF%B in the immune system. Annu Rev Immunol 1994; 12:141-179. 52. Shimizu H, Mitomo K, Watanabe T et al. Involvement of a NF%B-like transcription factor in the activation of the interleukin-6 gene by inflammatory lymphokines. Mol Cell Biol 1990; 10:561-568.

The Significance of Intestinal Cytokines

213

53. Parikh AA, Salzman AL, Kane CD et al. IL-6 production in human intestinal epithelial cells following stimulation with IL-1 is associated with activation of the transcription factor NF- B. J Surg Res, in press. 54. Moore EE, Moore FA, Franciose RJ et al. The postischmeic gut serves as a primary bed for circulating neutrophils that provoke multiple organ failure. J Trauma 1994; 37:881-887. 55. Wang Q, Fischer JE, Hasselgren PO. Treatment of endotoxemic mice with antiinterleukin-6 antibody paradoxically increases interleukin-6 levels and stimulates mucosal protein synthesis. Arch Surg 1997; 132:82-88. 56. Austgen TR, Chen MK, Dudrick PS et al. Cytokine regulation of intestinal glutamine utilization. Am J Surg 1992; 163:174-179. 57. Argiles JM, Lopez-Soriano FJ, Wiggins D et al. Comparative effects of tumour necrosis factor-cachectin, interleukin-1 and tumour growth on amino acid metabolism in the rat in vivo. Biochem J 1989; 261:357-362. 58. Malizia G, Calabrese A, Cottone M et al. Expression of leukocyte adhesion molecules by mucosal mononuclear phagocytes in inflammatory bowel disease. Gastroenterology 1991; 100:150-159. 59. Fais S, Burgio V, Pallone F. Intercellular adhesion molecule-IICAM-I) expression by intestinal endothelium and peripheral monocyte recruitment into Crohn’s disease tissues. Gastroenterology 1993; 104:A697. 60. Martin S, Maruta K, Burkart V et al. IL-1 and IFN-gamma increase vascular permeability. Immunol 1988; 64:301-305. 61. Cominelli F, Nast CC, Dinarello CA et al. Regulation of eicosanoid production in rabbit colon by interleukin-1. Gastroenterology 1989; 97:1400-1405. 62. Janu P, Li J, Minard G et al. Systemic interleukin-6IL-6 correlates with intestinal permeability. Surg Forum 1996; 47:7-9.

Cytokines in Abdominal Vascular Surgery

215

CHAPTER 18

Cytokines in Abdominal Vascular Surgery C.V. Soong and B.J. Rowlands

Introduction

A

dvances in resuscitative technology have allowed surgeons to perform increasingly complex operations and revive trauma patients previously deemed unsalvagable. This ability to sustain life following major surgery and hemorrhagic shock has led to the recognition of a new phenomenon, multiple organ dysfunction syndrome (MODS), first described by Tilney in 1973.1 It is now realized that many of the effects of this syndrome are due to inflammatory mediators, e.g., cytokines and eicosanoids. The release of these potentially devastating agents have been observed following many forms of abdominal surgery such as liver resection and abdominal aortic aneurysm repair.2-10 Since the first abdominal aortic grafting was undertaken by Dubost et al in 1952,11 the number of cases performed per year has been increasing. Despite the initial improvement observed in morbidity and mortality with better surgical technique, graft materials and anaesthesia, the rate of complications following an operation on the abdominal aorta remains relatively high, especially in patients presenting with a ruptured abdominal aortic aneurysm. Cardiac complications have been implicated as major causes of morbidity and mortality following aortic surgery but more recently it has been recognized that multiple organ dysfunction syndrome (MODS) from visceral organ impairment may be the leading cause of death.12,13 MODS accounts for more than 90% of late deaths following ruptured abdominal aortic aneurysm, and those with more than two failed organs have a mortality of almost 100%.14 The mortality rate and incidence of MODS is low following elective aortic surgery but dysfunction of isolated organs may occur in up to 50% of these cases.13 The exact pathophysiological mechanisms leading to organ dysfunction remains uncertain but many factors contribute, with the role played by each differing in importance when comparing outcomes in elective aortic surgery and ruptured abdominal aneurysm repair. These include blood loss and transfusion, bowel ischemia, coexisting diseases such as ischemic heart disease, chronic lung disease, diabetes and hypertension, and ischemia-reperfusion injury to the bowel and lower limbs. These aetiological factors may appear different but most share a common feature of facilitating immune dysfunction of the host through the production of cytokines, activators and mediators of the systemic inflammatory response. Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

216

Cytokines and the Abdominal Surgeon

Evidence of Endotoxemia and Cytokine Production in Abdominal Vascular Surgery There is convincing evidence to demonstrate that endotoxemia occurs in patients after abdominal aortic surgery.3-8 Controversy remains about the interpretation of endotoxin data due to inconsistencies in its detection in these patients. The inability of Baigrie et al8 to detect endotoxin in this cohort of patients has fuelled scepticism about the role of endotoxin in this situation and it is recognized that detection of endotoxin is unpredictable and liable to environmental influence.15-18 A strict regimen for sampling and assay may minimize these problems but a more accurate and sensitive method of demonstrating endotoxemia in plasma is to measure the consumption of the antibodies to core glycolipid.19,20 These antibodies exist naturally and offer protection against some bacterial strains by blocking their endotoxin activities.21,22 An inverse relationship between endotoxin concentrations and these antibodies was found in patients with septic shock.20 A low antibody concentration was associated with a poor prognosis during endotoxemia.23-25 A significant decrease in the plasma concentration of endotoxin antibodies was demonstrated following abdominal aortic aneurysm repair suggesting consumption in an attempt by the host to neutralize circulating endotoxin.20 This decrease in concentration was most marked in those who died and probably reflected a more significant systemic endotoxemia in the nonsurvivors compared to survivors. It has been suggested that this may represent a hemodilutional effect but this is unlikely as there was no associated decrease in an antibody directed against a neutral antigen.26 It is now clear that endotoxin produces its systemic effects via cytokine mediators generated by immune cells e.g., macrophages.27-29 These polypeptides which include tumour necrosis factor (TNF), interleukin-1 (IL-1) and interleukin 6 (IL-6), are known to share and augment each others multibiological activities.30-32 Elevations in TNF, IL-1 and IL-6 concentrations have been found to occur following abdominal aortic surgery.3,7-10,33 By sampling early and frequently it is possible to demonstrate a clear pattern of cytokine generation in these patients. An early rise is observed in IL1 which peaks 2 hours following the surgical insult.8 This is followed by an elevation in IL-6 which reaches a maximum by 48 h with a significant difference observed between the concentrations in those who developed evidence of bowel ischemia compared to those who did not (Fig. 18.1). These concentrations fall towards baseline afterwards except in patients who develop bowel infarction or major complications when production may be sustained. These elevations of cytokine concentrations persist until the complication has resolved or the patient makes a full recovery following appropriate therapeutic intervention.8 Routine measurement of IL-6 concentrations may be of prognostic value in patients undergoing abdominal aortic surgery with high plasma concentrations of IL-6 indicating a poor outcome in septic patients.8,34,35 The detection of TNF in patients undergoing aortic surgery is more sporadic and intermittent and the presence of TNF in plasma even in septic patients may be unreliable and misleading.3,8,10,36 This may be related to sampling difficulties associated with the short half-life and mode of action which is predominantly as a paracrine and autocrine mediator.37,38 Care is also needed in the interpretation of TNF data as numerous extraneous influences e.g., the type of anesthesia, anaesthetic agent, and administration of heparin may effect serum concentrations, local paracrine and binding properties.39,40 Like so many other peptide mediators, the pleitropic activities of TNF are due to its binding to specific receptors on cell surfaces.41-44 These receptors

Cytokines in Abdominal Vascular Surgery

217

Fig. 18.1. The plasma concentration of IL-6 reached a maximum by 48 h following abdominal aortic aneurysm surgery with a significant difference observed between those who suffered some bowel ischemia (❏) compared with those who did not (&). * = p < 0.05 Mann-Whitney U test (Unpublished data).

are shed as soluble cytokine-binding proteins in various biological fluids. Their concentrations increase in diseases such as systemic lupus erythematosis, rheumatoid arthritis, acquired immunodeficiency syndrome due to HIV, trauma, endotoxemia and ascites secondary to infection or malignancy.45-52 They are also elevated under some physiological conditions e.g., pregnancy when high concentrations may be found in intrauterine fluid.53 Since they compete for binding to TNF with cell-surface bound receptors they act to block TNF activity, thus providing a protective mechanism.54-57 As TNF is one of the most potent stimulators of the shedding of soluble TNF receptors in serum, measurements of plasma TNF receptor concentrations may yield information on the presence of circulating systemic TNF.49,58,59 An elevation in the TNF receptor concentration has been found in patients following elective abdominal aortic aneurysm repair.26 This increase in concentration was consistently higher in nonsurvivors compared to survivors suggesting a greater stimulus for their shedding. A strong correlation was observed between the antibodies to endotoxin and TNF receptor concentrations, especially in those patients who survived. This suggests that small quantities of endotoxin modulate the production of TNF but in overwhelming endotoxemia this control is lost. This would support the hypothesis that a balance is required between pro-inflammatory and anti-inflammatory factors to maintain homeostasis and facilitate recovery of the patient.60 An imbalance may lead to MODS or an increase in susceptibility to infection.

The Mechanism of Production of Cytokines in Aortic Surgery The stimulus for cytokine production in aortic surgery remains undetermined (Fig. 18.2). A simplistic view would be that the development of bowel ischemia leads to disruption of the mucosal barrier function and this is the major source of endotoxemia and cytokine generation. In aortic surgery, the clamping and unclamping

218

Cytokines and the Abdominal Surgeon

Fig. 18.2. Cytokine generation following abdominal aortic surgery.

of the aorta and iliac vessels will cause an ischemia-reperfusion insult to the bowel and lower limbs. In the bowel, the ischemic injury is compounded by ligation of the inferior mesenteric artery and the occasional need to bypass the iliac system because of aneurysmal involvement or occlusive disease. If ischemia is prolonged severe bowel necrosis may ensue. The milder forms of ischemia may not be detected clinically but microscopic disruption of the mucosa may occur.61 Regardless of whether bowel infarction or more subtle changes develop, an increase in bowel permeability may occur which allows egress of bacterial products across the disrupted mucosal barrier, and the development of bacteremia and endotoxemia.62,63 The permeation of bacteria and their products into the systemic circulation may be direct, when bowel infarction occurs, or via the portal system in milder forms of ischemia where hepatic filtering acts to attenuate some of the effects of endotoxemia.64 In transient ischemia, the damage may be greater during reperfusion of the ischemic tissue due to the generation of oxygen-derived free radicals.65 These are cytotoxic species released principally via the xanthine oxidase system.66 Xanthine dehydrogenase occurs naturally, but during ischemia it is converted to xanthine oxidase by proteolysis. This enzyme catalyses the reaction which utilizes oxygen to convert hypoxanthine to uric acid with the production of superoxide. These free radicals cause damage directly via processes such as lipid peroxidation whereby the lipid membrane of cells is damaged by a series of reactions with the production of byproducts e.g., malondialdehyde.67 They may also trigger the activation of the eicosanoids and cytokine cascades.68-71 Some of the mediators in these cascades e.g., leukotriene B4 and IL-8 are potent chemoattractants which activate neutrophils. The latter infiltrate the site of reperfusion injury and remote organs.72,73 These activated neutrophils release more free radicals, cytokines and proteolytic agents causing further damage and create a self propagating cycle of adverse events. The majority of patients undergoing abdominal aortic surgery suffer some degree of bowel ischemia.4 It has been proposed that the endotoxemia and production of cytokines is driven primarily by events affecting the bowel wall which leads to an

Cytokines in Abdominal Vascular Surgery

219

Fig. 18.3. Greater concentrations of endotoxin, TNF and IL-6 were observed in patients undergoing abdominal aortic aneurysm surgery who developed evidence of bowel mucosal acidosis, an indication of ischemia, compared to those who did not suffer any ischaemic changes. * = p < 0.05 Mann-Whitney U test. Reproduced with permission from: Eur J Vasc Surg 1993; 7:534-39.

increase in bowel permeability.74 The increase in permeability may be related to splanchnic hypoperfusion. A greater concentration of endotoxin, TNF and IL-6 have been observed in those patients undergoing abdominal aortic aneurysm repair who develop evidence of bowel ischemia, compared to those who did not have ischemic changes (Fig. 18.3).3 This is supported by a significant correlation found between endotoxin and bowel perfusion (Fig. 18.4).4 Further evidence is provided by the detection of endotoxemia in ruptured aneurysms at presentation prior to resuscitation or surgery.5 This endotoxemia was worse during reperfusion with twice the number of positive endotoxin assays in comparison to samples taken when the aorta was cross-clamped.5,7 A significantly greater concentration of IL-6 and TNF were observed in the portal circulation than in the systemic circulation.7,75 The reperfusion of ischemic limbs also leads to a spectrum of complications ranging from local oedema to failure of one or more organs or systems.68,76-78 In addition to the generation of free radicals and their byproducts, an elevation in the concentration of cytokines may occur. An increase in plasma IL-6 concentration may follow reperfusion of ischemic limbs.78 The origin of this cytokine elevation is uncertain. Firstly, animal studies have shown an increase in bowel permeability following ischemia-reperfusion injury to the lower limbs.78,79 Morphological changes, characterized by a decrease in bowel mucosal thickness has been observed after reperfusion of ischemic limbs, associated with the development of endotoxemia and an increased mortality rate.78 The mechanism involved in this mucosal damage remains unclear. Despite an increase in the concentration of thromboxane A2, a potent chemoattractant and activator of neutrophils, no demonstrable infiltration of the intestinal mucosa

220

Cytokines and the Abdominal Surgeon

Fig. 18.4. A significant correlation was observed between endotoxin and the intramucosal pH of the bowel taken as a measure of splanchnic perfusion. rs = -0.54, p < 0.005. Spearman’s Rank Correlation test. Reproduced with permission from: Br J Surg 1994; 81:965-68.

by neutrophils has been reported.80 It may occur via a neutrophil independent process as studies have shown that these local and remote organ injuries can still occur following reperfusion of ischemic tissues in the absence of neutrophils.81,82 Data also exist demonstrating that cytokines may be released from ischemic skeletal muscle.68-70 In an isolated leg model, a rise in TNF, IL-1 and IL-6 was found following reperfusion of the ischemic limb.68,69 This was supported by the findings of Ascer et al70 who demonstrated that thromboxane A2 and IL-1 concentrations were significantly higher in the venous effluent of the gracilis muscle compared to the systemic concentrations. The production of these cytokines may be related to the generation of oxygen-derived free radicals by the endothelial cells which may be attenuated by pretreatment with superoxide dismutase and glutathione peroxidase.83 Cytokines and other acute phase protein concentrations were significantly higher in patients with critical limb ischemia compared to controls.84 The burst of TNF following reperfusion of such ischemic extremities may represent a washout of this peptide mediator produced during ischemia.69 Spillage of these cytokines and byproducts of free radicals into the systemic circulation may cause injury to organs such as the liver and lung remote from the site of injury.85-87 These injuries may reflect up-regulation of the neutrophil adherence receptors such as the CD18, as antibodies directed against them minimize the lung injury observed following ischemiareperfusion injury to the lower limbs.88,89 The presence of a neutrophil activator which is stable to storage and transfer following reperfusion injury has been demonstrated.90 Traction on the mesentery will also produce systemic effects due to release of vasoactive agents.91-94 The systemic reaction is typified by hypotension, tachycardia and cutaneous flushing.91 These hemodynamic effects may be reduced by the prior administration of ibuprofen or by an extraperioneal approach to the aorta which

Cytokines in Abdominal Vascular Surgery

221

avoids handling the bowel.93,94 Both strategies work by reducing the generation of prostaglandins. In addition, the eventration of the small bowel into a plastic bag may cause bacterial translocation demonstrated by the presence of bacteria in the fluid collected from the bag.95 This direct trauma to the bowel almost certainly causes cytokine release and may contribute to the development of graft infection. Atheromatous plaque is a rich source of cytokines and other inflammatory mediators. Explants from aortic aneurysms have significantly greater quantities of IL-6, IL-8, interferon-! and monocyte chemoattractant protein-1 in comparison to explants from occlusive or normal aortas.96,97 The magnitude of cytokine secretion by the aortic cells was related to the degree of atherosclerosis.98 Manipulation of the aorta and the aneurysm sac may cause a shower of these factors into the circulation and probably explains the elevation of TNF concentration observed in the endovascular repair of an abdominal aortic aneurysm which although less invasive, produces significantly greater systemic inflammatory response than open repair.33 It is tempting to suggest that the expansion of such a graft against the aneurysmal sac contents which has been left in situ will cause continuing release of these factors. Patients undergoing aortic surgery may loose considerable quantities of blood, especially if there is leakage or rupture of an abdominal aortic aneurysm. Many will experience transient periods of hypovolemic shock followed by normotension when they are successfully resuscitated. Hemorrhage and resuscitation may lead to the production of TNF, IL-1 and IL-6 which is associated with an acute lung injury.10,99,100 An increase in the concentration of IL-8 has been demonstrated when shock due to blood loss complicates rupture of an aneurysm. The acute lung injury may be caused by the sequestration of activated granulocytes in the pulmonary circulation.99,100 Blood loss may also alter the cytokine mRNA expression of pulmonary macrophages, cells of the gut associated lymphoid tissue and circulating immune cells.99,101,102 This change in expression of cytokine mRNA no doubt contributes to the damage observed in various organs following hemorrhage. Blood transfusion may lead to an elevation in cytokine concentrations, due to cytokines pre-existing in the stored blood products.103-105 The presence of these mediators has been implicated in the causation of some transfusion reactions observed in patients receiving blood.103 Even autologous blood may lead to an elevation in cytokines.106 Successful resuscitation of patients with hypovolemia and hypoperfusion will produce a global ischemia-reperfusion injury which initiates the free-radical-cytokine cascade. A fall in the concentration of ∀-tocopherol, the major lipid phase antioxidant, has been detected following aortic surgery suggesting consumption by free radical-induced lipid peroxidation.107 An increased plasma concentration of lipofuscin, a byproduct of lipid peroxidation, has been described recently in association with adult respiratory distress syndrome and MODS.108

The Effects of Elevated Cytokines Cytokines are polypeptides with a wide range of activities. IL-1, like TNF, is a pyrogen, which induces hypotension, activates and promotes adherence of neutrophils and causes an increase in acute phase protein synthesis.109-113 Although the role of some, e.g., IL-6, in septic shock remains uncertain, it can induce the synthesis of acute phase proteins and modulate the immune cells.114-117 The role of cytokines in the development of postoperative complications, SIRS and MODS following abdominal aortic surgery remains speculative and only by inference of data presently available. Patients who develop bowel ischemia suffer more severe organ impairment than those who do not have evidence of bowel underperfusion.3,4 Those who sustain an

222

Cytokines and the Abdominal Surgeon

ischemic insult to the bowel have higher concentrations of liver transaminases, cardiac complications and a greater hypoxemia than those with no evidence of ischemia.4 The greater organ impairment associated with splanchnic ischemia is correlated with a higher plasma concentration of endotoxin and TNF.3,4 The elevation in liver transaminases may be due to a larger load of endotoxin and cytokine in the portal circulation in those who develop bowel ischemia. The increased incidence of cardiac complications in bowel ischemia was postulated to be due to the higher concentrations of TNF which has myocardial depressant properties.110 Following infra-renal aortic aneurysm repair nonsurvivors have significantly higher concentrations of IL-6 compared to survivors.9 The higher concentration of IL-6 is associated with a complicated postoperative course.8,10 Greater concentrations of TNF and IL-1 were found in patients who developed ARDS and MODS. There is no correlation between postoperative infection and an increased bowel permeability following repair of ruptured abdominal aortic aneurysm.118 The release of these peptides may simply reflect the degree of trauma sustained by these patients, and the release of cytokines may be secondary to the development of complications.119 This may explain why patients presenting with ruptured abdominal aortic aneurysms have higher concentrations of TNF, IL-1 and IL-6 and why endovascular repair may abrogate some of these systemic responses.120

Potential Therapeutic Manipulation Therapeutic strategies aimed at reducing the production of cytokine and improving outcome must be viewed with caution. Firstly, endogenous cytokine production is part of the normal reaction to injury and inflammation. Together with other cascades they form part of the natural defences against invading microorganisms and are essential in the healing process following injury. Progressive organ impairment may occur when control is lost.60 Studies have shown that optimizing cardiac function can minimize morbidity and mortality following abdominal aortic surgery, but the use of certain inotropes, e.g., dopamine, may have paradoxical detrimental effects by decreasing the perfusion of intestinal mucosa.121,122 The importance of maintaining adequate oxygenation to the bowel to improve recovery and outcome in critically ill patients in the intensive care unit has been reported recently.123 As the gut appears to be the main source of bacteria and endotoxin, selective digestive decontamination may reduce the toxicity of these pathogens.124 A number of studies have demonstrated that enteral administration of antibiotics can improve infection related morbidity in patients receiving intensive therapy.125 Furthermore, cytokine release may be reduced by selectively decontaminating the intestine in cardiac surgery.126 Plasma concentrations of endotoxin and IL-6 were found to be significantly lower in patients given antibiotics compared to the controls. The role of anti-endotoxin and anticytokine antibodies remains controversial. Some studies have shown a reduction in mortality rate in critically ill patients using anti-endotoxin monoclonal antibody, but one study was abandoned due to a higher mortality in the treatment group.127-129 The efficacy of anticytokine therapy such as IL-1 receptor antagonist and monoclonal antibodies to TNF and IL-8 remains to be determined.130-132 The use monoclonal anti-TNF antibodies, soluble TNF receptor and IL-1 receptor antagonists can significantly diminish cytokine mRNA expression in cells within the pulmonary parenchyma but their ability to improve posthemorrhagic lung injury is uncertain.133-135 Pentoxifylline, a xanthine derivative, may reduce the systemic inflammatory response in patients suffering critical limb ischemia by preventing the expression of adhesion molecules and chemotactic effect of TNF

Cytokines in Abdominal Vascular Surgery

223

on neutrophils.84 It can also reduce vascular permeability and endotoxin-induced leukocyte adhesion in sepsis.136 The administration of antioxidants have shown promise in reducing morbidity following vascular procedures, such as aortic operations and lower limb bypass for ischemia.137-139 Unfortunately, their effect on the generation of cytokines and in improving mortality from MODS is uncertain. Despite the potential benefits shown by some of these therapies, the multitude of cascades activated in aortic surgery precludes a simple therapeutic strategy to prevent the development of an uncontrolled SIRS in high risk patients.

Conclusion Overwhelming evidence exists to demonstrate the release of cytokines following abdominal aortic surgery. Their role in the development of postoperative complications, sepsis, SIRS and MODS has not been fully defined. The part played by free radicals generated during reperfusion of the ischemic limbs and bowel will require further clarification. The complexity of the biological system and the interaction between different cascades will make it difficult to assess the contribution of the individual components to the host response following aortic surgery. The importance of the compensatory anti-inflammatory mediators in modulating the inflammatory response has only recently been appreciated. Vascular patients commonly have coexistent illnesses e.g., ischemic heart disease, pulmonary diseases and renal impairment. There is an increased risk of failure or deterioration of these major organ-systems. Dysfunction of one organ can pose additional strain on other organ-systems, especially if they were already compromised.1 Although the generation of cytokines may cause organ failures, a “failing body” can initiate and sustain a systemic response to critical illness. These interrelationships that govern homeostasis is supported by the elevation in plasma glutamine concentration prior to death even though glutamine levels normally fall following aortic surgery (Fig. 18.5). It may be that loss of cell membrane integrity allows the escape of

Fig. 18.5. An elevation in plasma glutamine concentration was observed just prior to death in the patient who suffered MODS (❍) in comparison to the fall in those whose recovery was unremarkable (❏).

224

Cytokines and the Abdominal Surgeon

intracellular contents which flood the systemic circulation signalling that demise is imminent. We can only speculate on the significance of cytokines in the outcome of abdominal aortic surgery when we do not know if their release is the cause or effect of organ impairment.

References 1. Tilney NL, Bailey GL, Morgan AP. Sequential systems failure after ruptured abdominal aortic aneurysm: An unresolved problem in postoperative care. Ann Surg 1973; 178:117-22. 2. Hewitt G, Halliday M, McCaigue M, Campbell G, Rowlands B, Diamond T. Mortality, endotoxaemia and cytokine expression after intermittent and continuous hepatic ischemia. Br J Surg 1995; 82:1424-26. 3. Soong CV, Blair PHB, Halliday MI, Hood JMH, Rowlands BJ, Barros D’Sa AAB. Endotoxaemia, the generation of the cytokines and their relationship to intramucosal acidosis of the sigmoid colon in elective abdominal aortic aneurysm repair. Eur J Vasc Surg 1993; 7:534-39. 4. Soong CV, Blair PHB, Halliday MI, Hood JMH, Rowlands BJ, Barros D’Sa AAB. Bowel ischemia and organ impairment in elective abdominal aortic aneurysm surgery. Br J Surg 1994; 81:965-68. 5. Roumen RMH, Frieling JTM, van Tits HWHJ, van der Vliet JA, Goris RJA. Endotoxaemia after major vascular operations. J Vasc Surg 1993; 18:853-57. 6. Welch M, Douglas JT, Smyth JV, Walker MG. Systemic endotoxaemia and fibrinolysis during aortic surgery. Eur J Vasc Endovasc Surg 1995; 9:228-32. 7. Cabie A, Farkas JC, Fitting C et al. High levels of portal TNF-alpha during abdominal aortic surgery in man. Cytokine 1993; 5:448-53. 8. Baigrie RJ, Lamont PM, Kwiatkowski D, Dallman MJ, Morris PJ. Systemic cytokine response after major surgery. Br J Surg 1992; 79:757-60. 9. Froon AH, Greve JW, van der Linden CJ, Buurman WA. Increased concentrations of cytokines and adhesion molecules in patients after repair of abdominal aortic aneurysm. Eur J Surg 1996; 162:287-96. 10. RoumenRMH, Hendriks T, van der Ven-Jongekrijg J et al. Cytokine patterns in patients after major vascular surgery, hemorrhagic shock, and sever blunt trauma. Ann Surg 1993; 218:769-76. 11. Dubost C, Allay M, Oeconomos N. Resection of the abdominal aorta. Arch Surg 1952; 64:405-8. 12. Huber TS, Harward TRS, Flynn TC, Albright JL, Seager JM. Operative mortality rates after elective infrarenal aortic reconstruction. J Vasc Surg 1995; 22:287-94. 13. Soreide O, Grimsgaard CHR, Myhre HO, Solheim K, Trippestad A. Time and course of death for 301 patients operated on for abdominal aortic aneurysm. Age and Aging 1982; 11:256-60. 14. Meesters RC, van der Graaf Y, Vos A, Eikelboom BC. Ruptured aortic aneurysm: early postoperative prediction of mortality using an organ system failure score. Br J Surg 1994; 81:512-16. 15. Caridis DT, Reihold RB, Woodruff PWH, Fine J. Endotoxaemia in man. Lancet 1972; ii:1381-86. 16. Danner RL, Elim RJ, Hosseini JM, Wesley RA, Reilly JM, Parillo JE. Endotoxaemia in human septic shock. Chest 1991; 9:169-75. 17. Fink PC, Lehr L, Urbascheck RM, Kozak J. Limulus amoebocyte lysate test for endotoxaemia investigations with a femtogram sensitive spectrophotometric assay. Clin Worchenschir 1981; 59:213-18. 18. Watson RWG, Redmond HP, McCarthy J, Burk P, Bouchier-Hayes D. Endotoxin regulates early host response to surgery. Br J Surg 1994; 81:757.

Cytokines in Abdominal Vascular Surgery

225

19. McCrory D, Weir C, Halliday MI et al. The reliability of endotoxin and cytokine measurements in polytrauma patients and their relationship to the development of sepsis. Clin Int Care 1994; 20:461. 20. Barclay GR, Scott BB, Wright IH, Rodgers PN, Smith DGE, Poxton IR. Changes in anti-endotoxin-IgG antibody and endotoxaemia in three cases of gram negative septic shock. Circ Shock 1989; 29:93-106. 21. Westphal O, Jann K, Himmelspach K. Chemistry and immunchemistry of bacterial lipopolysaccharides as cell wall antigens and endotoxins. Prog Allergy 1983; 33:9-39. 22. Freeman R, Gould FK. Prevention of fever and gram-negative infection after open heart surgery by anti-endotoxin. Thorax 1985; 40:846-48. 23. Brauner A, Svenson SB, Wretlind B. Antibody response to lipid A and core oligosaccharides in patients with bacteraemia. Serodiag Immunother 1987; 1:431-40. 24. Schedel I. New aspects in the treatment of gram-negative bacteraemia and septic shock. Infection 1988; 16:4-7. 25. Nys M, Joassin L, Somzee A, Demonty J. Enzyme-linked immunosorbent assay for immunoglobulin G subclass antibodies specific for enterobacterial re-core glycolipid in healthy individuals and in patients infected by gram-negative bacteria. J Clin Microbiol 1988; 26:857-62. 26. Soong CV, Halliday MI, Hood JM, Rowlands BJ, Barros D’Sa AAB. Bowel ischemia and the host response in elective abdominal aortic aneurysm repair. Crit Care Med 1997 (In press). 27. Michalek SM, Moore RN, McGhee JR, Rossenrtreich DL, Mengenhagen SE. The primary role of lymphoreticular cells in the mediation of host responses to bacterial endotoxin. J Infect Dis 1980; 141:55-63. 28. Beutler B, Cerami A. Cachectin: More than a tumour necrosis factor. N Engl J Med 1987; 316:379-85. 29. Beutler B. Cachectin in tissue injury, shock and related states. Crit Care Clin 1989; 5:353-67. 30. Beutler B, Milsark IW, Cerami AC. Passive immunization against cachectin/tumour necrosis factor protects mice from lethal effects of endotoxin. Science 1985; 229:869-71. 31. Dinarello CA. The biology of interleukin 1 and comparison to tumour necrosis factor. Immunol Lett 1987; 16:227-32. 32. Le J, Vilcek J. Interleukin-6: A multifunctional cytokine regulating immune reactions and the acute phase protein response. Lab Invest 1989; 61:588-602. 33. Swartbol P, Norgren L, Albrechtsson U et al. Biological responses differ considerably between endovascular and conventional aortic aneursym surgery. Eur J Vasc Endovasc Surg 1996; 12:18-25. 34. Calandra T, Gerain J, Henmann D, Baumgartner J-D, Glausner MP, Swiss-Dutch J5 Immunoglobulin Study Group. High circulating levels of interleukin-6 in patients with septic shock: Evolution during sepsis, prognostic value and interplay with other cytokines. Am J Med 1991; 91:23-29 35. Hack CE, DeGroot ER, Felt-Bersma JF et al. Increased plasma levels of interleukin6 sepsis. Blood 1989; 74:1704-10. 36. Debets JMH, Kampmeijer R, van der Linden MPMH, Buurman WA, van der Linden CJ. Plasma tumour necrosis factor and mortality in critically ill septic patients. Crit Care Med 1989; 17:489-94. 37. Metcalf D. The leukaemia inhibitory factor (LIF). Int J Cell Cloning 1991; 9:95-108. 38. Poo WJ, Conrad L, Janeway CA Jr. Receptor directed focusing of lymphokine release by helper T cells. Nature 1988; 332:378-80.

226

Cytokines and the Abdominal Surgeon

39. McBride WT, Armstrong MA, McMurray TJ. An investigation of the effects of heparin, low molecular weight heparin, protamine, and fentanyl on the balance of pro- and anti-inflammatory cytokines in in-vitro monocyte cultures. Anaesthesia 1996; 51:634-40. 40. Crozier TA, Muller JE, Quittkat D, Sydow M, Wuttke W, Kettler D. Effect of anaesthesia on the cytokine response to abdominal surgery. Br J Anaesthesia 1994; 72:280-85. 41. Hohmann H-P, Remy P, Brockhaus M, van Loon APGM. Two different cell types have different major recptors for human tumour necrosis factor (TNF-∀). J Biol Chem 1989; 264:14927-34. 42. Dembic Z, Loetscher H, Gubler U et al. Two human TNF-receptors have similar extracellular, but distinct intracellular domain sequences. Cytokine 1990; 2:231-37. 43. Tartaglia LA, Goeddel DV. Two TNF receptors. Immunol Today 1992; 13:151-53. 44. Tartaglia LA, Weber RF, Figari IS, Reynold C, Palladino MA Jr, Goeddel DV. The two different receptors for tumour necrosis factor mediate distinct cellular responses. Proc Natl Acad Sci USA 1991; 88:9292-96. 45. Aderka D, Wysenbeek A, Engelmann H et al. Correlation between serum levels of soluble tumour necrosis factor receptor and disease activity in systemic lupus erythematosus. Arthritis and Rheumatism 1993; 36:111-20. 46. Barrera P, Boerbooms AM, Jansson EM et al. Circulating soluble tumour necrosis factor recptors, interleukin-2 recptors, tumour necrosis factor alpha, and interleukin6 levels in rheumatoid arthritis. Longitudinal evaluation during methotrexate and azathioprine therapy. Arthritis and Rheumatism 1993; 36:1070-79. 47. Godfried MH, van der Poll T, Jansen J et al. Soluble receptors for tumour necrosis factor: a putative marker of progression in HIV infection. AIDS 1993; 7:33-36. 48. Gruss HJ, Dolken G, Brach MA, Mertelsmann R, Hermann F. The significance of serum levels of soluble 60kDa receptors for tumour necrosis factor in patients with Hodgkin’s disease. Leukaemia 1993; 7:1339-43. 49. Tan LR, Waxman K, Scannell G, Ioli G, Granger GA. Trauma causes early release of soluble receptors for tumour necrosis factor. J Trauma 1993; 34: 634-38. 50. Spinas GA, Keller U, Brockhaus M. Release of soluble receptors for tumour necrosis factor (TNF) in relation to circulating TNF during experimental endo-toxaemia. J Clin Invest 1992; 90:533-36. 51. Shapiro L, Clark BD, Orencole SF, Poutsiaka DD, Granowitz EV, Dinarello CA. Detection of tumour necrosis factor soluble receptor p55 in blood samples from healthy and endotoxaemic humans. J Infect Dis 1993; 167:1344-50. 52. Andus T, Gross V, Holstege A et al. High concentrations of soluble tumour necrosis factor receptors in ascites. Hepatology 1992; 16:749-55. 53. Austgulen S, Johnsen H, Kjollesdal AM, Liabakk NB, Espevik T. Soluble receptors for tumour necrosis factor: occurrence in association with normal delivery at term. Ostet Gynecol 1993; 82:343-47. 54. Bertini R, Delgado R, Faggioni R, Gascone MP, Ythier A, Ghezzi P. Urinary TNF binding protein (TNF soluble receptor) protects mice against the lethal effect of TNF and endotoxin shock. Eur Cytokine Network 1993; 4:39-42. 55. Schall TJ, Lewis M, Kollen KJ et al. Molecular cloning and expression of receptor for human tumour necrosis factor. Cell 1990; 61:361-70. 56. van ZeeKJ, Kohno T, Fischer E, Rock CS, Moldawer LL, Lowry SF. Tumour necrosis factor soluble receptors circulate during experimental and clinical inflammation and can protect against excessive tumour necrosis factor alpha in vitro and in vivo. Proc Natl Acad Sci USA 1992; 89:4845-49. 57. Porteu F, Nathan C. Shedding of tumour necrosis factor receptors by activated human neutrophils. J Exp Med 1990; 172:599-607.

Cytokines in Abdominal Vascular Surgery

227

58. Winzen R, Wallach D, Kemper O, Resch K, Holtmann H. Selective up-regulation of the 75kDa tumour necrosis factor (TNF) receptor and its mRNA by TNF and IL-1. J Immunol 1993; 150:4346-53. 59. Aderka D, Engelmann H, Ma ORY, Brackebusch C, Wallach D. Stabilization of the bioactivity of tumour necrosis factor by its soluble receptors. J Exp Med 1992; 175:323-29. 60. Bone RC. Sir Isaac Newton, sepsis, SIRS and CARS. Crit Care Med 1996; 24:1125-28. 61. Haglund U, Bulkley GB, Granger DN. On the pathophysiology of intestinal ischemic injury. Acta Chir Scan 1987; 153:321-24. 62. Langer JC, Sohal SS. Increased mucosal permeability after intestinal ischemiareperfusion injury is mediated by local tissue factors. J Ped Surg 1992; 27:329-32. 63. Granger DN, Sennet M, McElearney P, Taylor AE. Effect of local arterial hypotension on cat intestinal capillary permeability. Gastroenterology 1980; 79:474-80. 64. Bennion RS, Wilson SR, Williams RA. Early portal anearobic bacteramia in mesenteric ischemia. Arch Surg 1984; 119:151-55. 65. Parks DA, Granger DN. Contributions of ischemic and reperfusion to mucosal lesion formation. Am J Physiol 1986; 250:G749-53. 66. McCord JM. Oxygen-derived free radicals in postischemic tissue injury. N Engl J Med 1985; 312:159-63. 67. Kellog EW. Superoxide, hydrogen peroxide and singlet oxygen in lipid peroxi-dation by xanthine oxidase system. J Biol Chem 1975; 250:8812-17. 68. Seekamp A, Warren JS, Remick DG, Till GO, Ward PA. Requirements for tumour necrosis factor-alpha and interleukin-1 in limb ischemia/reperfusion injury and associated lung injury. Am J Pathol 1993; 143:453-63. 69. Sternbergh WC 3rd, Tuttle TM, Makhoul RG, Bear HD, Sobel M, Fowler AA 3rd. Postischemic extremities exhibit immediate release of tumour necrosis factor. J Vasc Surg 1994; 20:474-81. 70. Ascer E, Mohan C, Gennaro M, Cupo S. Interleukin-1 and thromboxane release after skeletal muscle ischemia and reperfusion. Ann Vasc Surg 1992; 6:69-73. 71. Klausner JM, Paterson IS, Kobzik L, Valeri CR, Shepor D, Hetchman HB. Oxygen free radicals mediate ischemia-induced lung injury. Surgery 1989; 105:192-99. 72. Grisham MB, Brock AF, Scafer AI. Leukotriene B4 stimulates polymorphonuclear leukocytes adhesion to cultured vascular endothelial cells. J Clin Invest 1984; 74:1522-25. 73. Oppenheim JJ, Zachariae COC, Mukaida N, Matsushima K. Properties of the novel proinflammatory supergene ‘intercrine’ cytokine family. Annu Rev Immunol 1991; 9:617-48. 74. Roumen RM, Hendriks T, Wevers RA, Goris JA. Intestinal permeability after severe trauma and hemorrhagic shock is increased without relation to septic complication. Arch Surg 1993; 128:453-7. 75. Baigrie RJ, Lamont PM, Whiting S, Morris PJ. Portal endotoxin and cytokine response during abdominal aortic surgery. Am J Surg 1993; 166:248-51. 76. Soong CV, Young IS, Blair P et al. Lipid peroxidation as a cause of lower limb oedema in femoro-popliteal bypass grafting. Eur J Vasc Surg 1993; 7:540-45. 77. Haimovici H, Moss CM, Veith FJ. Arterial embolectomy revisited. Surgery 1975; 78:409-10. 78. Yassin MM, Barros D’Sa AAB, Parks G, Abdulkadir AS, Halliday MI, Rowlands BJ. Mortality following lower limb ischemia-reperfusion: A systemic inflammatory response? World J Surg 1996; 20:961-66. 79. Corson RJ, Paterson IS, O’Dwyer ST et al. Lower limb ischemia and reperfusion alters gut permeability. Eur J Vasc Surg 1992; 6:158-63.

228

Cytokines and the Abdominal Surgeon

80. Corson RJ, Fisher M, Ward I, Haboubi NY, Little RA, McCOllum. Thromboxane antagonism prevents hind limb ischemia/reperfusion induced gut damage. Br J Surg 1993; 80:1464-64. 81. Deep GM, Grum CM, Lynch MJ et al. Neutrophils are not necessary for induction of ischemia-reperfusion lung injury. J Appl Physiol 1990; 68:374-81. 82. Steimle CN, Guynn TP, Morganroth ML, Bolling SF, Carr K, Deep GM. Neutrophils are not necessary for ischemia-reperfusion lung injury. Ann Thorac Surg 1992; 53:64-73. 83. Ala Y, Palluy O, Favero J, Bonne C, Modat G, Dornand J. Hypoxia/reoxygenation stimulates endothelial cells to promote interleukin-1 and interleukin-6 production. Effects of free radical scavengers. Agents and Actions 1992; 37:134-39. 84. Fossat C, Fabre D, Alimi Y et al. Leukocyte activation study during occlusive arterial disease of the lower limb: Effect of pentoxifylline infusion. J Cardiovasc Pharmacol 1995; 25 Suppl:S96-100. 85. Poggetti RS, Moore FA, Moore EE, Koeike K, Banerjee A. Simultaneous liver and lung injury following gut ischemia is mediated by xanthine oxidase. J Trauma 1992; 32:723-28. 86. Poggetti RS, Moore FA, Moore EE, Benard DD, Anderson BO, Banerjee A. Liver injury is a reversible neutrophil-mediated event following gut ischemia. Arch 1992; 127:175-79. 87. Goldman G, Welbourn R, Klausner J-M et al. Mast cells and leukotrienes mediate neutrophil sequestration and lung oedema after remote ischemia in rodents. Surgery 1992; 112:578-86. 88. Welbourn R, Goldman G, Kobzik L et al. Role of neutrophil adherence receptors (CD18) in lung permeability following lower torso ischemia. Circulation Res 1992; 71:82-86. 89. Hill J, Lindsay T, Valeri CR, Shepro D, Hetchaman HB. A CD18 antibody prevent lung injury but not hypotension after intestinal ischemia-reperfusion. J Appl Physiol 1993; 74:659-64. 90. Paterson IS, Smith FCT, Tsang GMK, Hamer JD, Shearman CP. Reperfusion plasma contains a neutrophil activator. Am Vasc Surg 1993; 7:68-75. 91. Seltzer JL, Ritter DE, Starsnic MA, Marr AT. The haemodynamic response to traction on the abdominal mesentery. Anesthesiology 1985; 63:96-99. 92. Gottlieb A, Skrinska VA, O’Hara P, Boutros AR, Melia M, Beck GJ. The role of prostacyclin in the mesenteric traction syndrome during anaesthesia for abdominal aortic reconstructive surgery. Ann Surg 1989; 209:363-67. 93. Hudson JC, Wurm WH, O’Donnell TF Jr et al. Hemodynamic and prostacyclin release in the early phases of aortic surgery: Comparison of transabdominal and retroperitoneal approaches. J Vasc Surg 1988; 7:190-98. 94. Hudson JC, Wurm WH, O’Donnell TF Jr et al. Ibuprofen pretreatment inhibits prostacyclin release during abdominal exploration in aortic surgery. Anesthesiology 1990; 72:443-49. 95. Ernst CB, Campbell HC, Daugherty ME, Sachetello CR, Griffen WO. Incidence and significance of intra-operative bacterial cultures during abdominal aortic aneurysmectomy. Ann Surg 1977; 185:626-33. 96. Szekanecz Z, Shah MR, Pearce WH, Koch AE. Human atherosclerotic abdominal aortic aneurysms produce interleukin (IL)-6 and interferon gamma but not IL-2 and IL-4: The possible role for IL-6 and interferon-gamma in vascular inflammation. Agent and Actins 1994; 42:159-62. 97. Koch AE, Kunkel SL, Pearce WH et al. Enhanced production of the chemotactic cytokines in terleukin-8 and monocyte chemoattractant protein-1 in human abdominal aortic aneurysms. Am J Pathol 1993; 142:1423-31.

Cytokines in Abdominal Vascular Surgery

229

98. Farber HW, Antonov AS, Romanov YA, Smirnov VN, Scarfo LN, Beer DJ. Cytokine secretion by human aortic endothelial cells is related to degree of atherosclerosis. Am J Physiol 1992; 262:H1088-95. 99. Shenkar R, Coulson WF, Abraham E. Hemorrhage and resuscitation induce alterations in cytokine expression and the development of acute lung injury. Am J Respir Cell Molecular Biol 1994; 10:290-97. 100. Reissman P, Manny N, Shapira SC, Shapira Y, Cotev S. Transfusion-related adult respiratory distress syndrome. Israel J Med Sci 1993; 29:303-7. 101. Abraham E, Chang YH. Hemorrhage-induced alterations in function and cytokine productin of T cells and T cell subpopulation. Clin Exp Immunol 1992; 90:497-502. 102. Shenkar R, Chang YH, Abraham E. Cytokine expression in Peyer’s patches following hemorrhage and resuscitation. Shock 1994; 1:25-30. 103. Kristiansson M, Soop M, Saraste L, Sundqvist KG. Cytokines in stored red blood cell concentrates: Promoters of systemic inflammation and stimulators of acute transfusion reactions? Acta Anaestheriol Scan 1996; 40:496-50. 104. Stack G, Snyder EL. Cytokine generation in store platelet concentrates. Transfusion 1994; 34:20-25. 105. Grunnenberg R, Kruger J. Analysis of cytokine profiles of stored CPDA-1 whole blood. Infusionstherapie Transfusion Medizine 1995; 22:292-94. 106. Southern EP, Huo MH, Mehta JR, Keggi KJ. Unwashed wound drainage blood. What are we giving our patients? Clin Orthorp Rel Res 1995; 320:235-46. 107. Murphy ME, Kolvenbach R, Aleksis M, Hansen R, Sies H. Antioxidant depletion in aortic cross clamping ischemia: increase of the plasma ∀-tocopheryl quinone/∀-tocopherol ratio. Free Radical Biol Med 1992; 13:95-100. 108. Roumen RMH, Hendriks T, de Man BM, Goris RJA. Serum lipofuscin as a prognostic indicator of adult respiratory distress syndrome and multiple organ failure. Br J Surg 1994; 81:1300-5. 109. Beutler B, Milsark IW, Cerami AC. Passive immunization against cachectin/tumour necrosis factor protects mice from lethal effects of endotoxin. Science 1985; 229:869-71. 110. Tracey KJ, Fong Y, Hesse DG et al. Anti-cachectin/TNF monoclonal antibodies prevent septic shock during lethal bacteraemia. Nature 1987; 330:662-64. 111. Weinberg JR, Wright DJM, Guz A. Interleukin-1 and tumour necrosis factor cause hypotension in the conscious rabbit. Clin Science 1988; 75:251-55. 112. Pohlman TH, Stannes KA, Beatty PG, Ochs HD, Harlan JM. An endothelial cell surface factor(s) induced in vitro by lipopolysaccharide, interleukin-1, and tumour necrosis factor-alpha increases neutrophil adherence by Cdw-18-dependent mechanism. J Immunol 1986; 136:4548-53. 113. Rosenblum MG, Donato NJ. Tumour necrosis factor alpha: A multifaceted peptide hormone. Crit Rev Immunol 1989; 9:21-44. 114. Hirano T, Yasukawa K, Harada et al. Complementary DNA for a novel human interleukin (BSF-2) that induces B lymphocytes to produce immunoglobulin. Nature 1986; 324:73-76. 115. Gauldie J, Richards C, Harnish D, Lansdorp PM, Baumann H. Interferon-2/B-cell stimulatory factor type 2 shares identity with monocyte-derived hepatocyte stimulating factor and regulates the major acute phase response in liver cells. Proc Natl Acad Sci USA 1987; 84:7251-255. 116. Garman RD, Jacobs KA, Clark SC, Raulet DH. B-cell stimulatory factor (beta-2interferon) functions as a second signal for interleukin-2 production by mature murine T cells. Proc Natl Acad Sci USA 1987; 84:7629-33.

230

Cytokines and the Abdominal Surgeon

117. Libert C, Vink A, Coulie P et al. Limited involvement of interleukin-6 in the pathogenesis of lethal septic shock as revealed by the effect of monoclonal antibodies against interleukin-6 or its receptor in various murine models. Er J Immunol 1992; 22:2625-30. 118. Roumen RM, Hendriks T, Wevers RA, Goris JA. Intestinal permeability after severe trauma and hemorrhagic shock is increased without relation to septic complicaiton. Arch Surg 1993; 128:458-57. 119. Bitterman H, Kinarty A, Lazorovich H, Lahat N. Acute release of cytokiness is proportional to tissue injury induced by surgical trauma and shock in rats. J Clin Immunol 1992; 11:184-92. 120. Thompson MM, Nasim A, Sayers RD et al. Oxygen free radicals and cytokine generation during endovascular and conventional aneurysm repair. Eur J Vasc Endovasc Surg 1996; 12:70-5. 121. Whittemore AD, Clowes AW, Hetchman HB, Mannick JA. Aortic aneurysm repair. Reduced operatiove mortality associated with maintenance of optimal cardiac performance. Ann Surg 1980; 192:414-21. 122. Soong CV, Halliday MI, Hood JM, Rowlands BJ, Barros D’Sa AAB. The effect of low dose dopamine on intramucosal acidosis of the sigmoid colon in patients undergoing elective abdominal aortic aneurysm repair. Br J Surg 1995; 82:912-15. 123. Guttierrez G, Palizas F, Doglio G et al. Gastric intramucosal pH as a therapeutic index of tissue oxygenation in critically ill patients. Lancet 1992; 339:195-99. 124. Blair P, Rowlands BJ, Lowry K, Amstrong P, Smilie J. Selective decontamination of the digestive tract: a stratified, randomised, prospective study in a mixed intensive care unit. Surgery 1991; 110:303-9. 125. Trialists’ Collaboration Group selective decontamination of the digestive tract. Metaanalysis of randomised controlled trials of selective decontamination of the digestive tract. Br Med J 1993; 307:525-32. 126. Martinez-Pellus AE, Merino P, Bru M et al. Can selective decontamination avoid the endotoxaemia and cytokine activation promoted by cardiopulmonary bypass? Crit Care Med 1993; 21:1684-91. 127. Greenman R, Schein R, Martin M et al. A controlled clinical trial E5 murine monoclonal IgM antibody to endotoxin in the treatment of gram-negative sepsis. The XOMA sepsis study group. JAMA 1991; 266:1097-1102. 128. Ziegler E, Fisher C, Sprung C et al. Treatment of gram-negative bacteraemia and septic shock with HA-1A human monoclonal antibody against endotoxin. N Engl J Med 1991; 324:429-36. 129. Dunn DL. Current status of anti-endotoxin monoclonal antibodies. In: Faist E, Baue AE, Schildberg FW (eds). The immune consequences of Trauma, Shock and Sepsis. Mechanisms and Therapeutic approaches. Pabst Science Publishers; Lengerch, Berlin, Dusseldorf, Riga, Scottsdale, Wien 1996:1084-91. 130. Granowitz EV, Porat R, Mier JW et al. Haematological and immunomodulatory effects of interleukin-1 receptor antagonist coinfusion during low-dose endotoxaemia in healthy humans. Blood 1993; 82:2985-90. 131. Fisher CR Jr, Opal SM, Dhainant J-F et al. Influence of an antitumour necrosis factor monoclonal antibody on cytokine levels in patients with sepsis. Crit Care Med 1993; 21:318-27. 132. Mulligan MS, Jones ML, Bolanowski MA et al. Inhibition of lung inflammatory reactions in rats by an antihuman IL-8 antibody. J Immunol 1993; 150:5585-95. 133. Abraham E, Jesmok G, Tuder R, Allbee J, Chang YH. Contribution of tumour necrosis factor-alpha to pulmonary cytokine expression and lung injury after hemorrhage and resuscitation. Crit Care Med 1995; 23:1319-26.

Cytokines in Abdominal Vascular Surgery

231

134. Abraham E, Coulson WF, Schwartz MD, Allbee J. Effects of therapy with soluble tumour necrosis factor receptor fusion protein on pulmonary cytokine expression and lung injury following haemorrhage and resuscitation. Clin Exp Immunol 1994; 98:29-34. 135. Abraham E, Allbee J. Effects of therapy with interleukin-1 receptor antagonist on pulmonary cytokine expression following haemorrhage and resuscitation. Lymphokine Cytokine Res 1994; 13:343-47. 136. Seiffge D, Bissinger T, Kremer E, Laux V, Schleyerbach R. Inhibitory effects of pentoxifylline on LPS-induced leukocyte adhesion and macromolecular extravasation in the microcirculation. Inflam Res 1995; 44:281-86. 137. Soong CV, Young IS, Lightbody JH et al. Reduciton of free radical generation minimises lower limb swelling following femoro-popliteal bypass surgery. Eur J Vasc Surg 1994; 8:435-40. 138. Oredsson S, Plate G, Qvartfordt P. The effect of mannitol on reperfusion injury and postischemic compartment pressure in skeletal muscle. Eur J Vasc Surg 1994; 8:326-31. 139. Paterson IS, Klausner JM, Goldman G et al. Pulmonary oedema after aneurysm surgery is modified by mannitol. Ann Surg 1989; 210:796-801.

Cytokines and Abdominal Organ Transplantation

233

CHAPTER 19

Cytokines and Abdominal Organ Transplantation Stephen W. Chung, Charles H. Scudamore and Reginald Gorczynski

Introduction

A

successful outcome for a solid organ transplantation is dependent upon many factors. These include the careful selection and maintenance of donor organs, the preservation of the graft during organ retrieval and subsequent cold storage, meticulous surgical technique at the time of the transplantation, and the ability to control the interaction between the transplanted graft and the recipient immune system. Significant progress has been made in many fields in transplantation resulting in 90% graft and patient survival in most solid organ transplants performed. However, despite these improved results, there remains significant morbidity and mortality, often associated with the use of immunosuppression. The “Holy Grail” of transplantation, dating back to the seminal studies by Billingham et al1 remains the induction of donor-specific unresponsiveness in the recipient immune system towards the transplanted organ. Donor-specific unresponsiveness is operationally defined, but in turn has come to be referred to as a state in the immune system known as tolerance. It is important to appreciate, however, that immunological unresponsiveness is itself a misleading term, since the induction of a tolerant state is likely achieved by a number of active immune processes.2-6 Before we consider how immunological tolerance can be achieved in clinically relevant scenarios, we will first briefly summarize the current state of understanding of the processes involved in the regulation of acute transplant rejection. Immune recognition for allograft rejection requires antigen recognition by T lymphocytes followed by a number of intracellular signalling events, culminating in gene transcription and the production of both a number of effector molecules (including cytokines) and various effector functions in the stimulated T cells.7-9 Included in these effector functions is a lytic activity, by which specifically activated T cells kill (or lyse) the appropriate target cell. Cytokines are believed to play a crucial role in cell:cell communication and activation within the immune system. Although there is redundancy in the activation and function of the immune response, some cytokines simultaneously promote and suppress different pathways of immunity. Cytokines are clearly involved in allograft rejection but measurements of serum cytokine levels do not always reflect the presence or absence of rejection.10,11 CD 4+ T lymphocytes Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

234

Cytokines and the Abdominal Surgeon

play a key role in the regulation of allograft rejection, and the pattern of cytokines produced by the cells is believed to play an important role in this function. The subdivision of CD 4+ T-helper cell subsets into T-helper type 1 (Th1) and T-helper 2 (Th2) populations is in turn based upon the production of specific cytokines (type 1 cytokines classically include interleukin IL-2 and interferon (IFN) gamma, whereas type 2 cytokines include IL-4 and has proven a useful concept in furthering our understanding of the regulation and progression of immunologically based pathologies.2,3 The Th1- and Th2-type T lymphocyte subsets may themselves be derived from a third (precursor) T cell subtype, the Th0-type T lymphocyte. The level of exogenous cytokines present during the initial stimulation of the precursor T cells dictates whether the Th0-type T lymphocyte will preferentially develop into a Th1type or Th2-type T cell population.2,3,12-16 Thus, the presence of the cytokines IL-12 and/or IFN-! biases development towards a Th1-type phenotype whereas IL-4 promotes a Th2-type phenotype. IL-12 is thus a T cell regulatory cytokine, though the primary source of IL-12 seems to be antigen presenting cells not T cells. However, IL-12 upregulates IFN-! production by T cells which may be one indirect mechanism by which IL-12 promotes polarization towards a Th1 phenotype. The concomitant presence of IL-12 and IFN-!, both of which antagonize the function of IL-4, provides a powerful inhibition of the maturation of Th2-type T cells, while promoting a balance toward Th1-type T cells. IL-10, again a cytokine produced predominantly by antigen presenting cells, and IL-4 (produced by T cells) together promote the development of Th2- type T cells. There is some recent evidence that IL-6 (like its counterpart IL-12, a molecule primarily produced by non-T cells) may play the key role in initiating the drive to IL-4 production with subsequent polarization to the Th2 pathway and inhibition of development along the Th1 pathway. Currently used immunosuppressive agents, including cyclosporine A, tacrolimus (FK506), corticosteroids, azathioprine, and mycophenolate mofetil have all been shown to inhibit the production of cytokines at various levels following antigen triggering of T cells.17 These drugs have been shown to inhibit cytokine production to varying degrees, but despite their powerful effects, clinical rejection episodes are common. Acute rejection occurs in approximately 30-60% of individuals undergoing solid organ transplantation.17 Furthermore, all of these immunosuppressive agents are relatively nonspecific (immunologically speaking) in nature, and thus it is not surprising that a common cause of morbidity and mortality post-transplantation is severe infection as a result of overimmunosuppression. An understanding of what cytokines are produced by the unmanipulated recipient immune system, and a comparison with those patterns of production in individuals with rejecting vs nonrejecting grafts, would be helpful in determining what contributes to the successes of the immunosuppression and might in turn provide some guidelines for further research aimed at targeting specific immunoregulatory cytokines. It is hoped that more precise determination of the functioning of the cytokine network during transplant rejection, and its regulation, will lead to selective immunosuppression based upon specific patterns of cytokine production in the recipient, and also lead to earlier detection of patients at risk for either severe rejection or infection.

The Cytokine Response in Acute Allograft Rejection In general, it is accepted that Th1-type T lymphocytes initiate and promote the development of acute allograft rejection. It has been consistently demonstrated that intragraft IL-2, IFN-!, and a variety of cytotoxic T cell markers (granzyme B, perforins)

Cytokines and Abdominal Organ Transplantation

235

can be detected in rejecting allografts both in experimental animal models and human transplants.18-21 Since, as noted above, the Th2-type T lymphocyte cytokines are antagonistic to the development of the Th1 lymphocytes and the function of the type 1 cytokines, it was thus thought that Th2-type T lymphocytes and their cytokines would preferentially lead to the promotion of allograft acceptance and a state of immunological tolerance. There are a variety of observations which challenge this simple hypothesis. IL-4 expression has not been consistently detected within allografts which have become tolerant, and conversely, IL-4 expression has been detected in the presence of an acute rejection episode. In a similar fashion, IL-2 has been shown to be strongly associated with the rejection process, and administration of IL-2 to animals undergoing various tolerance-inducing regimes has been shown to prevent the induction both of long term allograft acceptance and polarization to Th2-type cells. Nevertheless, experiments using IL-2 gene knock-out mice has demonstrated that these mice reject allografts in a time frame which is only somewhat more prolonged than allograft rejection in normal wild-type animals.22,23 CD4+ and CD8+ T cell lymphocytes are present within the rejecting grafts, and IL-4, IFN-!, and granzyme B are detectable in the absence of gene transcripts for IL-2. These observations are probably most easily understood in terms of a marked redundancy in cytokine function. IL-4, IL-7, IL-9 and IL-15, in addition to IL-2, have all been shown to have potent T-cell growth factor properties. Thus in the artificial absence of a major T cell growth factor (IL-2), as in the IL-2 knock-out mouse referred to above, other cytokines predominate in regulating T cell proliferation and contribute to allograft rejection. It is not clear that these pathways are necessarily key components of the rejection pathway under normal conditions, though it must be acknowledged that they may play a role when we attempt to impose regulation on the system in an artificial manner. Interferon gamma is thought to promote allograft rejection by the recruitment of macrophages into an allograft and their subsequent activation. In addition, IFN-! can enhance cytotoxic T cell activation by mechanisms which include the up-regulation of major histocompatibility complex (MHC) expression on cells within the allograft. As noted above, IFN-! transcripts are a consistent feature of allograft rejection. Nevertheless, IFN-! knock-out animals, like the IL-2 knock-outs referred to earlier, also reject islet cell and cardiac transplants in a similar time frame to that seen in wild-type animals.24 Further, the histological pattern of the rejection in the two animals is similar. The presence of other cytokines e.g., tumor necrosis factor (TNF) alpha, is unchanged in the IFN-! knock-out mice. Islet cell allograft rejection in an IFN-! receptor knock-out mouse similarly is a T cell-dependent phenomenon and occurs within the normal time frame of acute rejection. Thus a consistent story seems to emerge, which seems to indicate that individual (proinflammatory) cytokines are not necessarily indispensable to the process of allograft rejection because of their intrinsic redundancy and pleiotropic nature. Transforming growth factor (TGF) beta has generally been characterized as an immunosuppressive-type cytokine. This cytokine blocks IL-2 and IL-4-stimulated T cell proliferation, and in animals made tolerant by a variety of means T cells have been demonstrated to produce TGF beta in addition to IL-4 and IL-10.25,26 IL-10 is another cytokine with inhibitory properties.27-29 It inhibits IFN-! release and proliferation of Th1-type lymphocytes, downregulates MHC class II expression on macrophages and monocytes and inhibits IL-12 release from macrophages. IL-10, however, does not block alloreactivity induced by dendritic cells and B-cells, demonstrating that its effects are not the same on all antigen presenting cells.27 Clearly this has

236

Cytokines and the Abdominal Surgeon

Table 19.1. Donor-specific PV immunization alters lymphocyte cytokine production and renal allograft survival Treatment of recipientsa

CsA only CsA+pv cells CsA+iv cells

(day 5)b

Cytokine production after restimulation in vitro IL-2 1150±18 220±40 1410±205

IL-4 30±5 128±20 30±7

IL-10 14±3 83±10 15±4

IFN ! 145±26 32±8 175±28

TGF # 41±6 98±13 50±9

Survival (days)c

16±4 55±13 14±5

a Groups of 13 C3H mice received C57BL/6 renal transplants along with treatment as shown. The

dose of CsA used was 10mg/Kg (im on the day of transplantation and 36 hrs later). PV (iv) cells represents administration of 15x106 C57BL/6 bone-marrow derived dendritic cells infused via the portal vein at the time of transplantation. b Cytokine production in cultures of spleen cells from mice sacrificed 10 days after transplantation (see first column). Cells were pooled from 3 individual recipients/group and tested in triplicate cultures with 1x106 responder cells and 5x105 irradiated C57BL/6 spleen stimulator cells. Supernatants were harvested and assayed in ELISA assays, in triplicate, at 40 hr of culture. Data represent arithmetic means (±SD). IL-2 and IL-4 are expressed as pg/ml; the remaining cytokines are expressed as ng/ml. c Animal survival (10 mice/group) in days (arithmetic mean±SD).

important implications as dendritic cells are likely the most potent antigen presenting cell in allograft rejection. In some instances, IL-10 may exert proinflammatory effects.27 Thus, depending upon the type of antigen presenting cell and the environment of the inflammatory response, the effects of IL-10 may have opposing effects. For example, although IL-10 is generally considered to be an immunosuppressive cytokine, in experiments using transgenic mice expressing IL-10, islet cells from the transgenic mice transplanted into wild-type recipients were shown to undergo normal allograft rejection.30 Similarly, infusion of recombinant IL-10 did not delay the onset of allograft rejection.31 As emphasized above, the mechanisms by which cytokines may regulate immune responsiveness include those operating at the level of the antigen presenting cell. Cytokines regulate factor (cytokine) release from antigen presenting cell. However, they also regulate the expression of molecules believed to be important for the development of the immune response, including MHC molecules (T cells recognize antigen presented in association with MHC molecules), and costimulatory molecules (molecules important in providing signals required for optimal stimulation of T cells activated by antigen-MHC interactions). Amongst the more important costimulatory molecules on antigen presenting cells are members of the B7 family [B7-1, B7-2 (or CD80, CD86)], and CD40.32,33 IFN-! has been shown to upregulate expression of CD80, CD86 on antigen presenting cells, while IL-10 decreases expression of these same molecules.34

Experimental Models of Acute Rejection Poor initial graft function secondary to a preservation injury may increase post operative morbidity and also may increase the risk of early allograft rejection.35 A number of mediators, including proinflammatory cytokines may be released during

Cytokines and Abdominal Organ Transplantation

237

reperfusion in relation to the extent of the preservation, ischemia and reperfusion injury of the solid organ. This is analogous to the release of the inflammatory cytokines observed during cardiopulmonary bypass which can cause a postoperative systemic inflammatory response syndrome. While cytokine release can be triggered by many factors during cardiopulmonary bypass, ischemia/reperfusion likely plays the most important role. Release of the cytokines (tumor necrosis factor (TNF) alpha and interleukins -1, -6, and -8) is correlated with the duration of cardiac ischemia.36 Conversely, anti-inflammatory cytokines such as IL-10 may also play a role in limiting the inflammatory response.15 Steroid pretreatment is an effective intervention to inhibit the release of proinflammatory cytokines and enhances IL-10 production. In the liver, activation of Kupffer cells after ischemic injury during transplantation results in the production of a number of cytokines as well as the induction of cytokine-dependent costimulatory molecules. These molecules, including intracellular adhesion molecule (ICAM)-1 and lymphocyte function associated antigen (LFA)-1 and -3 may be expressed leading to increased leukocyte and platelet adherence which can result in allograft injury and predisposition to rejection.35,37 A number of studies have now shown an increase in the induction of these adhesion molecules following reperfusion injury and also the induction of these molecules during episodes of acute rejection. In a study comparing the production of inflammatory cytokines in organ donors compared to healthy controls, significant elevation in plasma concentrations of IL-6, IL-8, TNF and C-reactive protein were detected among the organ donors compared to the controls.38 Maximum serum levels of these cytokines, including IL-1, TNF alpha, IL-6 and IL-1ra, were maximally increased at 60 minutes following reperfusion, 120 minutes and 24 hours post reperfusion. In both human and experimental animal models, the serum levels of these cytokines decreased considerably by 24 hours and within 7 days following re-perfusion. The generation of these inflammatory cytokines as well as the subsequent expression of adhesion molecules subsequently predisposes the transplanted organ to acute rejection. More recently attention has focused on another family of inflammatory molecules, the chemokines, whose expression is also altered in a number of inflammatory states including transplant rejection. The chemokines are a family of structurally related proteins having the ability to induce migration of specific subsets of leucocytes, which in turn leads to their playing a critical part in the generation of cellular inflammation, including those responses implicated in providing protection from invading pathogens and in the pathological processes associated with infection and immune-mediated diseases.39 However, chemokines are more than simply chemotactic factors, being also implicated in leucocyte activation, ww, and antimicrobial functions. Upon stimulation (by micro-organisms, microbial products or other endogenous factors which include cytokines) macrophages can synthesize and release the large variety of cytokines referred to earlier, including IL-1, IL-1ra, IL-6, IL-8, IL-10, IL-12, TNF-∀, IFN-∀, IFN-!, TGF-#, C-CSF, GM-CSF, but also the chemokines monocyte-chemoattractant protein (MCP)-1, MCP-S, MIF, M-CSF, MIP-1, MIP-2.40 As noted, some of these cytokines can themselves upregulate the production of other cytokines/chemokines by macrophages (IL-3, GM-CSF IFN-!), while others can inhibit it (IL-4, IL-10, IL-13, TGF-#). While cytokines can modulate most macrophage functions and cell surface marker expression, the chemokines such as MCP-1,2,3, MIP-1,2 and RANTES seem to exert their major effects by regulating the recruitment of circulating monocytes within tissues.

238

Cytokines and the Abdominal Surgeon

Table 19.2. Donor-specific vs third-party stimulation of cytokine production in peripheral blood lymphocytes of transplant recipients Group under test Ratio of donor-specific cytokine production:nonspecific cytokine production

Rejecting Nonrejecting

IL-2 0.36±0.12 0.39±0.11

IL-4 1.0±0.21 2.3±0.66*

IFN 0.57±0.20 0.56±0.19

IL-10 0.90±0.19 2.5±0.82*

TGF 0.65±0.17 1.6±0.50*

Peripheral blood lymphocytes from individual subjects were stimulated in vitro with mitomycin-c treated donor-specific lymphocytes (thawed spleen cells), or a pool of 14 third-party donor spleen cells. Supernatants were pooled at 40h from replicate cultures and assayed for cytokine production. Graft rejection was assessed by histopathology by an observer blind to the data obtained by cytokine analysis. * p < 0.05, compared with data from peripheral blood lymphocytes of patients with rejecting grafts.

In an infectious disease model of granulomatous inflammation mediated by Th1and Th2-related cytokines, levels of MCP-1 were greater in the type 2 than in the type 1 response. In vivo depletion of IFN-! augmented type 2 inflammation and local MCP production, while IL-4 depletion had the opposite effect.41 These treatments had no significant effect on a type 1 response, Furthermore, treatment with anti-MCP-1, but not with anti-MIP-1 alpha, inhibited type-2 cytokine associated inflammation, though neither treatment affected the type 1 associated responses. MCP-1 was detected within lymph nodes using immunohistochemistry, and appeared to support IL-4-/IL-5-producing lymph node cells while directly inhibiting IL-12 production by inflammatory macrophages. Taken together these data suggest that MCP-1 (and perhaps other chemokines) contribute more to type 2 than to type 1 cytokine-mediated inflammation and suggest that the chemokines may have a broader role in regulating Th cell expression than previously anticipated.42 As noted above reports in a number of experimental and human transplant model systems suggest that specific cytokine profiles may be associated with an acute rejection episode or with prolongation and survival of the allograft. Highlighting the redundancy in the activation and function of the immune response, some cytokines can simultaneously promote and suppress different pathways of immunity. In experimental rodent models of liver transplantation,43-47 spontaneously accepting liver allografts and rejecting liver allografts were examined. Spontaneously accepted liver allografts initially showed upregulation of IL-2 and IFN-! (Th1-type cytokines), and IL-4 and IL-10 (Th2-type cytokines). These cytokines peaked at the sixth post operative day and then tapered off. However, both IFN-! and IL-10 persisted. This cytokine profile is consistent with transient intragraft inflammation associated with spontaneously resolving rejection. Recipients that rejected their grafts revealed a marked upregulation of IL-2 and IFN-!, but also demonstrated increased levels of IL-4 and IL-6. There was no difference in IL-10 levels between the rejecting and nonrejecting animals. Interferon gamma levels were, however, markedly increased in the rejecting livers. These results suggest an association between liver allograft rejection and an enhanced Th1-type cytokine immune response. In another study examining cytokine gene expression within liver allografts from tolerant rats compared to

Cytokines and Abdominal Organ Transplantation

239

isograft liver tissue, IL-10 was upregulated in tolerant animals compared to normal or rejecting livers. Interestingly, IFN-! was present in tolerant rats suggesting that the upregulation of IL-10 crossregulates the effector function of IFN-!. In a separate study also examining liver allograft rejection, marked inflammatory cell infiltrates could be seen in association with a decreased liver function, which was progressive in untreated recipients of the allograft until liver failure occurred. There was an associated increase in the inflammatory cytokines associated with nonspecific inflammation. In addition, when specific CD 4+ T-cell cytokine expression was analyzed, IL-4 and IL-10 were present in both rejecting and nonrejecting grafts; however, IL-2 and IFN-! were detected only in rejecting allografts. A similar pattern of cytokine expression could be observed in peripheral lymphoid organs. In a study examining tolerance to rat liver allografts, infiltrating cell populations as well as cytokine production and immunoglobulin deposition in liver allografts were examined. Increased levels of IL-2, IL-4, IL-6, IL-10, TNF-∀, TGF-# and IFN-! were observed with similar levels of expression in both tolerant and rejecting allografts. Cytokine mRNA expression in syngeneic grafts were not different from normal livers, except that the IL-6 and TGF-# were increased. There was, however, a major difference in the B lymphocyte infiltration with more B lymphocytes present in tolerant livers. As Th2-type T cells are associated with a humoral response to antigens, this provides indirect evidence of a predominance of Th2-type cytokine production in this model. In a mouse mode of islet cell transplantation from the spontaneously diabetic nonobese (NOD) mice.48 The destruction of islet cell grafts is associated with a Th1type T cell driven insulitis. Treatment of recipient mice with IL-4, however, could not prevent islet cell destruction whereas treatment of the recipients with a combination of IL-4 and IL-10 prevented the destruction of the islet cell grafts. Histological analysis of the islet cell grafts revealed less severely infiltrated islet cell grafts in the treated animals with well preserved islet cell architecture. Staining for lymphocytes and macrophages for TNF alpha did not show differences between the groups, but IFN-! was markedly decreased in the islets of animals treated with IL-4 and IL-10. In rats undergoing renal transplantation, IL-2, IL-2 receptor, IFN-!, and TNF alpha expression were expressed at low levels in stable allografts in comparison to rejecting grafts.49 However, these animals, who were on cyclosporine therapy, had a tendency to develop signs of chronic rejection, associated with an upregulation of TGF-#, HSP 70, and endothelin compared with controls. These data add a further complexity, indicating that cytokines can produce different pathological (inflammatory) changes depending upon the timing, duration, and level of expression within the allograft. Some of our own studies in two separate systems are shown in Tables 19.1 and 19.2. In the first (experimental) model system, we investigated renal allograft survival and donor-cell-stimulated recipient cytokine production in C3H mice receiving C57BL/6 kidney allografts. All recipients received cyclosporin A (10 mg/kg on the day of transplantation, and 36 hours later). In addition, groups of experimental animals received donor-specific cell infusion (10-day cultured C57BL/6 bone marrow dendritic cells) via the portal vein (pv), a route we have shown produced graft prolongation in a number of model systems, or via the lateral tail vein (iv). The remaining host kidney was removed 48 hours after transplantation. Animal survival and cytokine production (using cells pooled from 3 animals/group sacrificed at 10 days post transplant) are shown in Table 19.1. In this model system there is a clear correlation between increased graft survival and a polarization to type-2 cytokine production, as defined by increased IL-4, IL-10

240

Cytokines and the Abdominal Surgeon

and TGF-# production, with decreased IL-2 and IFN-! production. Interestingly, our data confirm that donor specific pv immunization may represent a potentially clinically useful means to achieve this goal (of polarization in cytokine production). In the second study shown, we studied patients undergoing orthotopic liver transplantation to determine if there was any predictive value (in relation to graft outcome) for measuring the cytokine production profile of peripheral blood lymphocytes taken from patients post transplant and restimulated with donor specific cells (vs third-party cells) in vitro. These data, shown in Table 19.2, suggest that a preferential polarization to production of type-2 cytokines (not type-1 cytokines) after donor-specific restimulation in vitro is indeed correlated with lack of evidence for inflammatory (rejection) responses in these recipients. Although controversial, graft rejection has been associated with the preferential activation of Th1-type T lymphocytes, whereas graft acceptance (tolerance) has been associated with preferential activation of Th2-type T lymphocytes; however, not all studies have demonstrated reliable cytokine patterns predictive of graft rejection or acceptance.50-54 Nevertheless, data in the experimental and clinical literature, suggest that regulation of cytokine production may indeed offer a useful mechanism both to manipulate graft outcome, and to monitor the same in an ongoing fashion.

Conclusion The available data cannot fully support the simple hypothesis that allograft rejection is only driven by Th1-type cytokines, nor is there enough evidence to suggest that production of Th2-type cytokines necessarily lead to transplantation tolerance. Depending upon the microenvironment and the inflammatory response in the local milieu, cytokines produced by either Th1- and/or Th2-type T lymphocytes may be beneficial, deleterious, and/or irrelevant to the development of allograft rejection. Nevertheless, one should interpret apparently conflicting data from the various animal model systems used with caution. The controversies may indeed be more apparent than real, and may reflect differences attributable to variations in methodology, type of transplanted organ being assessed, vascularization of the allograft and the presence or absence of ischemia/reperfusion injury. It is particularly difficult to interpret data from highly artificial immunological systems, i.e., those using knockout or transgenic animals, where the underlying immune system (in an otherwise healthy animal) can be expected to have adapted in some manner to the major (immunological) insult imposed. Cytokines are pleiotropic and redundant in nature, and primarily act locally but generally not systemically. Cytokine therapy in isolation to induce tolerance induction, has been inconsistent. In most models of allograft tolerance, important mechanisms operating to induce tolerance probably act at the level preventing optimal T cell activation during alloantigen recognition by native T lymphocytes. At this phase the role of cytokines in regulating not T cell growth and expansion, but in controlling expression of cell surface costimulatory molecules, is probably critical, although a relatively unexplored area of investigation. We can anticipate that in the next several years the answers to these and other questions concerning the role of cytokines in immunoregulation will become available. As they do, we may then be able to consider how we might use this information as a strategy to monitor graft outcome, and/or to develop a more rational plan for sophisticated interventions to modify graft rejection.

Cytokines and Abdominal Organ Transplantation

241

References 1. Billingham RE, Brent L, Medawar PB. Actively accepted tolerance of foreign cells. Nature 1953; 172:603. 2. Mosmann TR, Cherwinski H, Bond MW et al. Two types of murine helper T-cell clones. 1. Definition according to profiles of lymphokine activities and secreted proteins. J Immunol 1986; 136:2348. 3. Mosmann TR, Sad S. The expanding universe of T-cell subset: TH1, TH2, and more. Immunol Today 1996; 17:138. 4. Qin S, Cobbold SP, Pope H et al. Infectious transplantation tolerance. Science 1993; 259:974. 5. Lowry RP. The relationship of IL-4, IL-10, and other cytokines to transplantation tolerance. Transplant Sci 1993; 3:104. 6. Lenzavecchia E. Mechanisms of antigen uptake for presentation. Curr Op Immunol 1996; 8:348-354. 7. Germain RN, Marguiles DH. The biochemistry and cell biology of antigen processing and presentation. Annu Rev Immunol 1993; 11:403-450. 8. Seder RA, Paul WA. Acquisition of lymphokine producing phenotype by CD4+ T-cells. Annu Rev Immunol 1994; 12:635-673. 9. Mason DW, Morris TJ. Effective mechanisms in allograft rejection. Annu Rev Immunol 1986; 4:119-145. 10. Weintraub D, Masek M, Billingham WE. The lymphocyte subpopulations in cyclosporine-treated human hearts undergoing rejections. J Heart Transplant 1985; 4:213-222. 11. Gajewski TF, Joyce J, Fitch FW. Antiproliferative effect of interferon gamma in immune regulation. 3. Differential selection of TH1 and TH2 murine helper T-lymphocyte clones using recombinant IL-2 and recombinant interferon gamma. J Immunol 1989; 143:15. 12. Swain SL, Weinberg AD, English M, Huston G. IL-4 directs the development of TH2-like helper effectors. J Immunol 1990; 145:3796. 13. McKnight AJ, Jimmers GJ, Fogelman I et al. Effects of IL-12 on helper T-cell dependent immune responses in vivo. J Immunol 1994; 152:2172. 14. Marshall JD, Secrist H, DeKruyff RH et al. IL-12 inhibits the production of IL-4 and IL-10 in allergen specific human CD4+ T-lymphocytes. J Immunol 1995; 155:111. 15. Moore JW, O’Garra A, de Waal Malefyt R et al. Interleukin 10. Annu Rev Immunol 1993; 11:165. 16. Trinchieri G. Interleukin 12: A proinflammatory cytokine with immunoregulatory functions that bridge innate resistance and antigen specific adaptive immunity. Annu Rev Immunol 1995; 13:251. 17. Calne RY. Immunosuppression in liver transplantation. N Engl J Med 1994; 331:1154-1155. 18. O’Connell PJ, Tacheco-Silva A, Nickerson TW et al. Unmodified pancreatic islet allograft rejections results in the preferential expression of certain T-cell activation transcripts. J Immunol 1993;150:1093. 19. Takeuchi T, Lowry RP, Koneiczny V. Heart allografts in murine systems; The differential activation of TH2-like effector cells in peripheral tolerance. Transplantation 1992; 53:1281. 20. Dallman MJ, Larsen CP, Morris PJ. Cytokine gene transcription in vascularized organ grafts: analysis using semi-quantitative polymerase chain reaction. J Exp Med 1981; 174:1493. 21. Lippman ML, Stevens AC, Strom TB. Heightened intragraft cytotoxic T-lymphocyte gene expression in acutely rejecting renal allografts. J Immunol 1994; 152:5120.

242

Cytokines and the Abdominal Surgeon

22. Schorle H, Holtschke T, Hunig T et al. Development and function of T cells in mice rendered interleukin-2 deficient by gene targeting. Nature 1991; 352:621. 23. Steiger J, Nickerson TW, Steurer W et al. IL-2 knock-out recipient mice reject islet cell allografts. J Immunol 1995; 155:489. 24. Saleem S, Konieczny BT, Lowry RT et al. Acute rejection of vascularized heart allografts in the absence of interferon gamma. Transplantation 1996; 62;1908-1911. 25. Ruegemer J, Ho S, Augustine J et al. Regulatory effects of transforming growth factor beta on IL-2 and IL-4 dependent T-cell cycle progression. J Immunol 1990; 144:1767. 26. Chen Y, Kuchroo VK, Inobe JI et al. Regulation of T cell clones induced by oral tolerance: Suppression of autoimmune encephalomyelitis. Science 1994; 365:1237. 27. Macatonia SE, Doherty TM, Knight SC, O’Garra A. Differential effect of IL-10 on dendritic cell induced T-cell proliferation and interferon gamma production. J Immunol 1993; 150:3755. 28. de Waal Malefyt R, Haanen J, Spits H et al. Interleukin 10 and viral IL-10 strongly reduced antigen specific human T cell proliferation by diminishing the antigen presenting capacity of monocytes via downregulation of class II major histocompatibility complex expression. J Exp Med 1991; 174:915. 29. D’Andrea A, Aste-Amezaga M, Valiante NM et al. Interleukin 10 inhibits human lymphocyte interferon gamma production by suppressing natural killer cells stimulatory factor/IL-12 synthesis in accessory cells. J Exp Med. 1993; 178:1042. 30. Lee MS, Wogensen L, Shizuru J et al. Pancreatic islet production of murine interleukin 10 does not inhibit immune mediated tissue destruction. J Clin Invest 1994; 93:1332. 31. Zheng XX, Steele AW, Nickerson TW et al. Administration of noncytolytic IL-10/Fc in murine models of LPS induced septic shock and allogeneic islet transplantation. J Immunol 1995; 154:5590. 32. Bluestone JA. New perspectives of CD28/B7-mediated T cell costimulation. Immunity 1995; 2:255-259. 33. Banchereau J, Bazam F, Blanchard D et al. The CD40 antigen and its ligand. Annu Rev Immunol 1994; 12:881-922. 34. Ding L, Linsley TS, Huang LY et al. IL-10 inhibits macrophage costimulatory activity by selectively inhibiting the upregulation of B7 expression. J Immunol 1993; 151:1224. 35. Clavien TA, Harvey PRC, Strasberg SM. Preservation and reperfusion injuries in liver allografts. Transplantation 1992; 52:957-978. 36. Wan S, LeClerc JL, Vincent JL et al. Cytokine responses to cardiopulmonary bypass: Lessons learned from cardiac transplantation. Ann Thoracic Surgery 1997; 668:369-76. 37. Cywes R, Harvey PRC, Packham MA et al. The influence of prostaglandin E1 on platelet adherence and injury in preserved rat liver allografts. Liver Transplant Surgery 1996; 2:23-36. 38. Palombo JD, Birke PA, Moldawer LL et al. Assessment of the cytokine response in liver donors at the time of organ procurement and association with allograft function after orthotopic transplantation. J Am Col Surg 1994; 179:209-19. 39. Lider O, Hershkoviz R, Kachelsky SG. Interactions of migrating T lymphocytes, inflammatory mediators and the extracellular matrix. Crit Rev Immunol 1995; 15:271-283. 40. Cavaillon JM. Cytokines and macrophages. Biomed Pharm 1994; 48:445-453. 41. Chensue SW, Warmington KS, Ruth JH et al. Role of monocyte chemoattractant protein-1 (MCP-1) in Th1 (Mycobacterial) and Th2 (Schistosomal) antigen-induced granuloma formation. J Immunol 1996; 157:4602-4608.

Cytokines and Abdominal Organ Transplantation

243

42. O’Hehir RE, Lake RA, Schall TJ et al. Regulation of cytokine and chemokine transcription in a human Th2 type T cell clone during the induction phase of anergy. Clin Exp Allergy 1996; 26:20-28. 43. Alfrey EJ, Most D, Wang X et al. Interferon gamma and IL-10 messenger RNA are upregulated after orthotopic liver transplantation in tolerant rats: Evidence for cytokine mediated immune disregulation. Surgery 1995; 118:399-404. 44. Thai ML, Fu S, Qian S et al. Cytokine mRNA profiles in mouse orthotopic liver transplantation. Graft rejection is associated with augmented TH1 function. Transplantation 1995; 59:274-81. 45. Sun J, McCaughan GW, Matsumoto Y et al. Tolerance to rat liver allografts: Diffference between tolerance and rejection are more marked in the B cell compared with T cell or cytokine response. Transplantation 1994; 57:1349-57. 46. Shirwan H, Cosenza CA, Wang HK et al. Prevention of orthotopic liver allograft rejection in rats with short term brequinar sodium therapy: Analysis of intragraft cytokine gene expression.Transplantation 1994; 57:1072-1080. 47. Egawa H, Martinez OM, Quinn MB et al. Acute liver allograft rejection in the rat. An analysis of the immune response. Transplantation 1995;59:97-102. 48. Faust A, Rothe H, Schade U et al. Primary nonfunction of islet grafts in autoimmune diabetic nonobese mice is prevented with interleukin-4 and interleukin-10. Transplantation 1996; 62:648-52. 49. Nadeau KC, Azuma H, Tilney NL. Sequential cytokine expression in renal allografts in rats immunosuppressed with maintenance cyclosporine or mycophenolate mofetil. Transplantation 1996; 62:1362. 50. Gorczynski RM, Adams RB, Levy GA, Chung SW. Correlation of peripheral blood lymphocyte and intragraft mRNA expression with rejection in orthotopic liver transplantation. Surgery 1996; 120:496-502. 51. Cosenza CA, Shirwan H, Cramer DV et al. Intragraft cytokine expression in human liver allografts. Liver Transplant Surg 1995; 1:16-22. 52. Martinez OM, Krams SM, Sterneck M et al. Intragraft cytokine profile during human liver allograft rejection. Transplantation 1992; 52; 449-456. 53. Platz KP, Mueller AR, Rossaint R et al. Cytokine pattern during rejection and infection after liver transplantation—Improvements in postoperative monitoring? Transplantation 1996; 62:1441-1450. 54. Strom TB, Roy-Chaudry P, Manfro R et al. The Th1/Th2 paradigm and the allograft response. Curr Op Immunol 1996; 8:688-693.

Anticytokine Strategies in Peritonitis

245

CHAPTER 20

Anticytokine Strategies in Peritonitis: Is Local, IntraAbdominal, Therapy Possible? Alfred Ayala and Irshad H. Chaudry

Introduction

S

epsis induced multiple organ failure is reported to be responsible for upwards of 60% of the deaths which occur in the surgical intensive care unit.1,2 This occurs despite the availability of not only specific antibiotics, aggressive operative intervention, and nutritional support, but also the use of new systemically administered therapeutic agents such as antibodies/antagonists to endotoxin and the pro-inflammatory cytokines.3,4 In light of these results, it is even more critical to determine the precise etiology contributing to the development of this state if we are to design better/novel treatments or maximize standard therapies for sepsis. In this chapter we will attempt to describe those anticytokine approaches that have been attempted (most of which thus far have been systemic-mono-therapeutic approaches) and examine the basis and/or plausibility of localized anticytokine therapy with respect to peritonitis/sepsis. A number of studies have suggested that the link between cell and organ dysfunction associated with multiple organ failure lies in the initial presentation of sepsis/peritonitis (gram-negative, gram-positive, and/or fungal in nature).2,5-9 For the purpose of this article we will utilize the definition of Fink and Heard10 who described sepsis as a “constellation of clinical and laboratory findings indicative of a generalized inflammatory response that is otherwise unexplained and is accompanied by acute organ system dysfunction and is often, but not invariably, associated with the presence of a serious bacterial, fungal or viral infection”. Peritonitis as such might be thought of as a subgroup within the definition of sepsis, in which the typically infectious inflammatory stimulant is initially restricted to the peritoneal and/or abdominal cavity. It must, however, be kept in mind that while peritonitis as such is initially thought to be the result of a localized abdominal insult, such as perforated and/or necrotic bowel, pancreatitis, peritoneal abscess, bacterial translocation, appendicitis, etc., the response to this insult, or evidence of immune/inflammatory reaction to such a challenge, typically also exhibits a systemic component, such as the release of neuroendocrine mediators, altered cardiovascular/metabolic response, altered immune mediator/cell profiles as well as Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

246

Cytokines and the Abdominal Surgeon

changes in distant tissue/organ function.11,12 Therefore, in discussing the background information, which will serve as our basis for consideration of potential intra-abdominal pro- or anti-inflammatory cytokine intervention, we will also consider not only the potential candidate cytokine(s) but also their cellular/tissue source(s) while trying to maintain a perspective on these mediators’ systemic effects.

Models of Sepsis/Peritonitis Before one can assess the intra-abdominal, as opposed to systemic, approaches to treatment of peritonitis, we have to first clarify the experimental models which have and continue to be used to examine this issue. The discussion of models is, nonetheless, abbreviated in order to confine to the limits of this chapter and is meant only as an introductory overview. For greater detail on these experimental models of peritonitis, we suggest the reader also see chapter 10 of this book as well as several other extensive reviews.10,13,14 Clinically, peritoneal sepsis (peritonitis) commonly manifests itself as a two phase process in which the patient initially goes through a hyperdynamic-hypermetabolic phase (increased cardiac output, fever, enhanced metabolic rate) and subsequently through the hypodynamic-hypometabolic state.15,16 The latter stage of this process is typically referred to as “septic shock” in which the circulatory system is compromised, organ function becomes more depressed, and the animal or patient eventually dies. It has been postulated that the presence of large numbers of microbes and/ or their products, either due to the physical breech of the normal mechanical barriers (due to trauma, i.e., mechanical or thermal injury) or due to bacterial translocation from the gut,17-19 serve as a stimulant of the systemic host inflammatory response and may be responsible for causing the eventual septic shock state. In this regard, there have been a number of studies which implicate endotoxin (a component of the gram negative bacterial cell wall, i.e., lipopolysaccharide, LPS) as one of the primary agents responsible for many of the events observed in the septic state. When endotoxin is infused in mg/kg body weight (BW) doses, it produces a shocklike state comparable to that seen in late sepsis.20-23 Alternatively, endotoxin has also been administered in a chronic low dose manner (∋ mg/kg BW) which induces a hyperdynamic-hypermetabolic (pro-inflammatory) state comparable to that seen in early sepsis.24-27 However, with neither situation is the progression of changes from hyperdynamic to hypodynamic state seen. While endotoxin has been detected in many clinical situations in the critically ill patient, it is frequently present in amounts of only ng/L of blood,28-31 a concentration well below that necessary to induce an acute septic shock-like state. However, most models of endotoxemia rarely exhibit both these two states.10,13,14 In this regard, the murine model of cecal ligation and puncture (CLP) produces both the early hyper- and late hypo-dynamic metabolic phases of sepsis-peritonitis.10,13,14 The CLP model, as used in our hands, also produces a more chronic disease as mortality develops over days (not hours). This delayed onset of pathology makes this more comparable to the clinical conditions associated with the sepsis syndrome32 and allows for extended/delayed therapeutic interventions. However, while this model emulates many of the phenomena associated with the sepsis syndrome in the patients’ progression to septic shock and death32 it is realized that other etiologies may contribute to this condition such as wound infection, pancreatitis, endocarditis, etc. However, we would suggest that these states all share a component of inflamed/necrotic/injured tissue which may similarly contribute to the eventual shock state in a manner comparable to that seen in CLP.

Anticytokine Strategies in Peritonitis

247

Differential Effects of Sepsis/Peritonitis of Immune Cells Since it is not within the realm of this article to provide a detailed background on the “typical” host immune response, the reader is directed to salient texts/reviews on the subject.33,34 Similarly, as detailed discussion of the biochemistry and generalized function of the pro- and/or anti-inflammatory cytokine has already been provided in the earlier chapters of this book we will not extensively cover this topic in the confines of this chapter. Table 20.1 provides a descriptive listing of the cytokines which may have relevance to possible intra-abdominal cytokine approaches or represent cytokines which have been implicated in the response to peritonitis/sepsis. We have provided them in two generalized but not exclusive categories. The first being the pro-inflammatory cytokines. This is a family of protein mediators which have been identified primarily with the augmentation/potentiation of the inflammatory response and/or the stimulation of cell-mediated immunity. Alternatively, those protein mediators which appear to primarily contribute to attenuating/blunting the inflammatory and/or cell-mediated immune response have typically been considered as anti-inflammatory cytokines. However, with respect to immune cells and/or the cytokines listed here it is important to keep several general tenants in mind when considering them in therapeutic approaches of any form to sepsis/peritonitis and/or associated multiple organ failure. These agents (cytokines) have disparate effects on systemic (endocrine), paracrine and autocrine levels. Also in the past the approach has been to arbitrarily lump these mediator(s) (e.g., the pro-inflammatory cytokines) as agents whose presence is considered essentially a “bad” event in the course of immune response to peritoneal/ septic insult. This was based largely on the results of studies in which typically abnormal high doses of pro-inflammatory cytokines were administered to produce a “septic shock”-like state.11,13,20-23 As can be seen from Table 20.1 and as we will further discuss, this is a gross over-simplification which ignores the reality that all mediators produced by immune cells play important roles in mounting a competent response to foreign pathogens. Similarly, we typically tend to think of the response of a given wing of the immune system, such as T lymphocyte or macrophage, based on data often limited to a representative cell population or cell line to which access is most convenient (e.g., blood monocytes or lymphocytes, peritoneal macrophage or splenic lymphocytes from mice, Jurkat T cell line, etc.). However, it is important to remember that differential cell responsiveness/activation of specific tissue immune cell populations typically takes place in response to septic stimuli by production of either pro- and/or anti-inflammatory cytokines. This is important not only in understanding the pathophysiology of the response to peritonitis but also when considering possible cellular targets for intra-abdominal therapeutic interventions. For example the elimination of such cellular sources as resident macrophage, while ablating the particular cytokines’ release by them, may actually further compromise the host phagocytic defense.35-37 With respect to the host’s defense against peritoneal microbial challenge, studies from a number of laboratories indicate that the immune system appears to exhibit a biphasic response, comparable to the metabolic and circulatory responses seen in peritonitis/sepsis.6,15,38 Initially there is evidence of an exaggerated systemic inflammatory response thought to be due to immune cell (monocyte/macrophage) stimulation by microbes and/or their products released following the onset of sepsis.39-45 A number of investigators have suggested that mediators released during this early phase, in response to trauma, shock and/or sepsis, may indeed be the nidus for the hyperdynamic-hypermetabolic response.46-48 Over time, however, this stage gives way to what appears to be a state of generalized immune cell hyporesponsiveness.31,49-55

Interleukin-12

Interleukin-11

(T/B cell proliferation; (B cell differentiation; (Monocyte/

Interleukin-6

Tc1 (CD8 T cell sub-type) phenotype commitment while

(T cell differentiation/stimulates Th1 (CD4 T cell sub-type)/

macrophage cytotoxicity, pro-inflammatory cytokine release, prostanoid release; (hepatic type II acute protein production; (endothelial cell tissue factor release, ICAM-1 expression, prostanoid release (T-proliferation; (Granulocyte hematopoiesis,; (hepatic type II acute protein production; (intestinal stem/progenitor cell proliferation

release; protects gut mucosa from inflammatory/chemotherapeutic injury ; (hematopoiesis Potential antitumor agent based on enhanced NK, T cell activating capacity as well as (IFN-! release. This latter

(fever; ( type II acute phase protein

(fever; ( type II acute phase protein release;(vasodilation; hypotension; anorexia; (hematopoiesis

(response to high dose IL-2 most likely mediated by TNF, IL-1)

(fever; nausea; capillary leak syndrome

(T cell activation/differentiation/proliferation, (growth factor receptor expression, (IFN-! release, (cytotoxicity, (cell mediated immunity; (B cell activation; (Macrophage activation, (IL-1 release, (cytotoxicity, (phagocytosis; (Natural killer cell activity, (IFN-!, (cytotoxicity release (via (IL-12 release);

Interleukin-2

pro-inflammatory cytokine release, prostanoid release; (hepatic type I acute protein production; (endothelial cell tissue factor release, ICAM-1 expression, prostanoid release

(fever; ( type I acute phase protein release;(vasodilation; hypotension; anorexia; (bone and cartilage resorption; (hematopoiesis

Systemic Effects

(T/B cell proliferation; (Monocyte/macrophage cytotoxicity,

Pro-inflammatory: Interleukin (IL)-1 (∀/#)

Cellular (Local) Effects

Table 20.1. Cytokines of general interest and/or potential tools/targets for localized intervention in abdominal sepsis

248 Cytokines and the Abdominal Surgeon

Continued...

Chemokines (C-X-C): IL-8, MIP-2, KC/Gro, CTAP, IP-10, ENA-78 (C-C): MCP-1,RANTES,MIP-1,TCA-3 Interferon-! (IFN-!)

Tumor Necrosis Factor-∀ (TNF-∀)

Interleukin-15

Family of chemotactic/haptotactic cytokines separated into 2 families based on presence/absence of amino acid between 1st two cysteine residues (C-X-C vs. C-C), cell-specific chemotaxins dependent on chemokine (lymphocytes vs. PMN vs. monocytes vs. fibroblasts), some activate target cell or (adhesion. (T cell differentiation/stimulates Th1 (CD4 T cell sub-type)/ Tc1 (CD8 T cell sub-type) phenotype commitment while suppressing Th2/Tc2 phenotype, (cell mediated immunity; (macrophage and natural killer cell activation, cytotoxicity; (macrophage pro-inflammatory cytokine response; (macrophage antigen presentation; (macrophage inducible nitric oxide production

suppressing Th2/Tc2 phenotype; (IFN-! release growth factor receptor expression, (cell mediated immunity; (B-cell differentiation; (Macrophage activation (via IFN-!), (Natural killer cell activation/differentiation, (IFN-!, (cytotoxicity; (T cell activation/differentiation/proliferation, (growth factor receptor expression, (cell mediated immunity; (B cell activation; (Macrophage activation, (IL-1 release, (cytotoxicity, (phagocytosis; (Natural killer cell activity, (IFN-!, (cytotoxicity release (via (IL-12 release); (T/B cell proliferation; (Monocyte/ macrophage cytotoxicity, pro-inflammatory cytokine release, prostanoid release; (hepatic type I acute protein production; (endothelial cell tissue factor release, ICAM-1 expression, prostanoid release

Potential antitumor agent based on enhanced NK, macrophage, and/or T cell activating capacity; given in high doses potentiates deleterious effects of systemic inflammation via release of TNF and IL-1

hypotension; )myocardial contractility; capillary leak; anorexia/cachexia; acidosis; )hematopoiesis ( IL-8 in fluids (blood/synovia) of patients with rheumatoid arthritis or septic shock

(fever; ( type I acute phase protein release;(vasodilation;

effect through induction of IFN-! release may mediate pro-inflammatory components (TNF and/or NO) of innate and cell-mediated immune response. Comparable effects to those seen with high-doses IL-2 assumed.

Anticytokine Strategies in Peritonitis 249

(T cell differentiation/stimulates Th2 (CD4 T cell sub-type)/Tc2

Interleukin-4

(CD8 T cell sub-type) phenotype commitment while

Competitive inhibitor of IL-1 ∀ and # binding to IL-1 cellular receptor

(granulocyte hematopoiesis

Anti-inflammatory: Interleukin-1 receptor antagonist (IL-1ra)

Granulocyte-colony stimulating factor (G-CSF)

(granulocyte and macrophage activation, cytotoxicity,

Granulocyte-macrophage-colony stimulating factor (GM-CSF)

antigen presentation, potentiates pro-inflammatory cytokine release; (granulocyte and macrophage hematopoiesis

Has many overlapping properties with IL-12, but is a biochemically distinct protein which requires intracellular protease processing prior to activation. However, unlike IL-12 induces Fas antigen expression.

Cellular (Local) Effects

IFN-!-Inducing Factor (IGIF, IL-18)

Table 20.1. (Continued)

Mediates IgE directed immune response; some selected antitumor

Directly antagonizes IL-1 mediated effects

Potential antitumor agent based on enhanced NK, T cell activating capacity as well as (IFN-! release. This latter effect through induction IFN-! release may mediate pro-inflammatory components (TNF and/or NO) of innate and cell-mediate immune response. However, also may mediated cytotoxic effects by induction of apoptosis (via. FasL-Fas system) (granulocyte and macrophage hematopoiesis; given in high doses potentiates deleterious effects of systemic inflammation via TNF and IL-1 release and/or via IFN-! (granulocyte hematopoiesis

Systemic Effects

250 Cytokines and the Abdominal Surgeon

Soluble Cytokine Receptors/ Decoy Receptor

Transforming Growth Factor-#

Interleukin-13

Interleukin-10

suppressing Th1/Tc1 phenotype, )cell mediated immunity; (B-cell activation/ differentiation/proliferation; )macrophage and natural killer cell activation, cytotoxicity; )macrophage pro-inflammatory cytokine response, (IL-1ra secretion; (macrophage antigen presentation; )macrophage inducible nitric oxide production (T cell differentiation/stimulates Th2 (CD4 T cell sub-type)/Tc2 (CD8 T cell sub-type) phenotype commitment while suppressing Th1/Tc1 phenotype, )cell mediated immunity; (Bcell differentiation; )macrophage and natural killer cell activation, cytotoxicity; )macrophage pro-inflammatory cytokine response, (IL-1ra secretion; (macrophage antigen presentation; )macrophage inducible nitric oxide production )macrophage and natural killer cell activation, cytotoxicity; )macrophage pro-inflammatory cytokine response, (IL-1ra secretion; (macrophage antigen presentation; )macrophage inducible nitric oxide production; No effect on T-/B-cells in humans )T cell proliferation while (T cell differentiation towards Th3 (CD4 T cell sub-type) phenotype commitment; (B-cell differentiation, )B-cell proliferation; )macrophage and natural killer cell activation, cytotoxicity; )macrophage/ (monocyte pro-inflammatory response, (IL-1ra macrophage secretion, (macrophage prostanoid release; )macrophage antigen presentation; )macrophage inducible nitric oxide production, chemoattractant for fibroblasts, macrophage, PMNs, Tlymphocytes; (localized PMN activation Bind cytokine and typically inhibit interaction with functional cell-surface receptor (e.g., TNF-RI(p55), TNFRII(p75), IL-2R, etc) or presence of nonfunctional decoy receptor (e.g., IL-8) competes cytokine. Directly/indirectly antagonizes particular pro-inflammatory cytokine mediated effects.

(Wound healing and w; )in vivo inflammatory response of IL-2, TNF and IL-1; immunosuppressant; )lymphopoiesis and hematopoiesis; (bone and cartilage resorption; (vasoconstriction

Potentiates IL-4 mediated IgE immune response; antagonizes cell mediated immune response/ delay type hypersensitivity.

effects, antagonizes selected aspects of cell mediated immune response/ delay type hypersensitivity. Anticytokine Strategies in Peritonitis 251

252

Cytokines and the Abdominal Surgeon

Regarding the early stage of the response to experimental sepsis/peritonitis, this is characterized by the detection in the circulation of pro-inflammatory cytokines, such as TNF, IL-1, IL-6, and chemokines, typically of macrophage derivation.39-45 In this regard, we have detected elevated circulating levels of TNF and IL-6 in both the mouse and rat model of polymicrobial sepsis, i.e., CLP.49,56,57 However, this early pro-inflammatory mediator release in response to an abdominal stimulus, such as CLP, is a result of selective tissue macrophage activation and is not the product of nonspecific/generalized monocyte/macrophage stimulation. In this respect, we have reported that endogenous stimulation of IL-1, IL-6 and TNF release are evident in peritoneal macrophage and Kupffer cells following CLP but is not detected in macrophage harvested from the spleen or lung49,58(also see Table 20.2). Hadjiminas et al,59 using a chronic model of sepsis, further documented that the enhanced TNF-∀ release observed following sepsis was not only a reflection of increasing secretion but also elevated transcriptional activity of the gene early following CLP. We have also found that both endotoxin-tolerant C3H/HeJ mice and endotoxin-sensitive C3H/ HeN mice exhibit a similar systemic pro-inflammatory cytokine response (increased TNF and IL-6) as well as show comparable rates of mortality following CLP.50 These findings not only demonstrate the ability of this murine polymicrobial sepsis model to mimic/produce this early systemic inflammatory response but suggest that sensitivity and/or responsiveness to endotoxin does not markedly alter this response. With the progression of experimental sepsis/peritonitis, i.e., >12 h after CLP, there is a marked decline in the capacity of macrophages (M) as well as lymphocytes to respond to stimulation. These cells, therefore, appear to become dysfunctional (anergic) or immune compromised. Recent findings in our laboratory55,60 as well as others53 indicate that, at least with respect to splenic lymphocytes, the decrease in immune responsiveness appears to be related to a concomitant decrease in cellular ATP stores55 and an increase in cellular calcium levels.60,61 In this respect, studies by Choudhry et al53 indicated that splenocyte depression of IL-2 release and mitogen responsiveness following gram-negative bacteremic insult in the rat was associated with increased basal intracellular calcium as well as changes in calcium mediated signal transduction. We have also reported that by 24 h post-CLP there is a marked reduction in IL-2 and IFN-! release capacity by mouse lymphocytes31 (see Table 20.2). Regarding M responsiveness, Gallinaro et al62 reported that following CLP, there is a marked decrease in the capacity of peritoneal and splenic M to present the antigen conalbumin to cloned D10.G4.1 T cells. This was associated with a marked reduction in the expression of the MHC class II antigen on those cells. However, a number of questions remain as to the nature of the defect in M antigen presenting capacity. These include whether there are alterations in the expression of membrane associated co-stimulants63,64 such as, ICAM-1, B7-1 and/or 2, membrane associated cytokines, etc. In our own laboratory, we have observed that by 24h following CLP there is a marked decrease in the capacity of M harvested from the spleen, liver, peritoneum or lungs of septic (CLP) mice to respond to in vitro stimulation by the elaboration of cytokines27,49 or their RNA messages.65 However, the mechanism behind the induction of immune suppression and/or cell dysfunction in late sepsis is not entirely understood. A number of investigators have suggested that irrespective of the nidus of stimulation, it is the systemic pro-inflammatory mediator response that is the motor for developing the host cell dysfunction. In this respect, the results of studies from several laboratories indicate that pro-inflammatory cytokines (of which macrophages are a key source), particularly TNF, IL-1 and IL-6, might be the agents which initiate

Anticytokine Strategies in Peritonitis

253

the development of cell and organ dysfunctions associated with sepsis and multiple organ failure.41 Support for this suggestion came from early studies by Waage et al42,66 Dammas et al,43 Marks et al,44 and Marano et al45 in which they reported that morbidity and/or mortality in humans due to sepsis and/or traumatic injury was associated with the marked elevation of TNF, IL-1 and/or IL-6. Further support for the role of M derived cytokines in the pathophysiology of the septic state comes from studies which have shown that when these agents are administered intravenously, they are capable of producing a shock-like state (circulatory decompensation)67,68 as well as organ dysfunction69 similar to that seen in septic shock. However, the majority of these studies in animals require relatively high doses or extended infusion of these cytokines which are not typical of the levels detected in CLP35,49,56,59 and/or septic patients.45,70 Inasmuch, while we have demonstrated49,56 that there is a marked early (~1 h peak) elevation in circulating bioactive TNF, followed by IL-6 (~4 through 24 h) and later by TGF-# (24 h and thereafter), the levels of TNF are markedly lower than that required to produce shock.11 What then might be the significance of this early low-level TNF response which is encountered in CLP? Studies by Eskandari et al35 indicate that when mice were pretreated with antibodies to TNF prior to CLP, a significant decrease in their survival rate was observed. Further, Hadjimias et al71 also reported that passive immunization with antibodies against either TNF-∀ or IL-1b did not inhibit sequestration of neutrophils in the lung following CLP. It may well be that inhibition of the localized release of these agents further depresses the host’s microbiocidal response (either by a direct effect on these cytokine mediated events or by indirect effects such as suppression of the acute phase response), thus making the animal more susceptible to the morbidity and mortality associated with sepsis. Similarly Bagby et al72 observed that passive pre-immunization of rats against TNF-alpha in a E. coli peritonitis model did not have the protective effects which were seen in a model of intravenous lethal endotoxemia. Thus suggesting that the early low-level systemic TNF response may be indicative of host mounted responses to the pathogen(s) but may not be of sufficient concentration to produce the septic shock response. Thus the early pro-inflammatory mediator response does not appear to be necessarily out of control (beyond normal limits) but may reflect simply the animals initial attempts to contain the infection. However, in the animal/patient when a chronic peritoneal/septic insult persists, we suggest there is a sustained antiinflammatory response (i.e., IL-1ra, IL-4, IL-6, IL-10, IL-13, TGF-#, TNF-RI/-RII, etc.) at a local and/or systemic level which eventually pushes the immune system towards suppression. To the extent that CLP might induce excessive anti-inflammatory mediator release, studies from our laboratory31 have shown that in late sepsis the decrease in IL-2 and IFN-! release capacity is associated with an elevated ability to produce the anti-inflammatory cytokines IL-4 and IL-10. However, in these studies neither the contribution of IL-4 nor IL-10 was directly linked to the observed suppression of Th1-lymphocyte lymphokine release. In this respect, Walley et al73 have recently determined that the systemic release of IL-10 following CLP is a delayed response (peaks at 12 h) and is not associated with pro-inflammatory mediator release. Nonetheless, the alterations in CD28, CTLA-4, and CD40 antigens on T cells as well as the release of CD40 ligand which effect T cell activation64 have not been assessed during peritonitis. Furthermore, the contribution of IL-12 (a stimulant of cell-mediated immunity)74,75 produced by macrophages as well as IL-18 (a potent stimulant of IFN-! release and cell-mediated immune amplification in response to endotoxin)76,77 release by macrophages from septic mice has not been determined. Interestingly, studies

254

Cytokines and the Abdominal Surgeon

carried out by Ertel et al78 indicate that IL-10, when administered following the induction of CLP in mice, has the capacity to suppress the early pro-inflammatory cytokine response but it decreased the animals’ overall survival and suppressed their Th1 lymphokine response. This observation indirectly provides evidence of a potential role of agents such as IL-10 in septic host immune suppression. Similarly studies by Hauser et al79,80 comparing peripheral blood monocytes to peritoneal exudate macrophages obtained from septic/traumatized patients versus control (elective hernia repair) demonstrated that not only do these two populations exhibit a differential cytokine response (similar to that seen following CLP in mice), but also that the peritoneal cells exhibit augmented IL-10 expression associated with decreased immune cell function. Studies from our laboratory50 have also demonstrated that immune suppression during sepsis is associated with the systemic release of the anti-inflammatory cytokine TGF-#.81,82 The role of TGF-# and IL-6 (a direct/indirect inducer of TGF-# release) as potential anti-inflammatory agents in sepsis, as well as in trauma, has been documented by the work of Miller-Graziano et al83 and Zhou et al.84,85 In light of this, one can envision a network of systemic effects mediated by agents which induces the release of IL-6, leading to increased TGF-# levels which then directly or indirectly induces the suppression of host responses seen during sepsis. Support for this hypothesis comes from our recent finding which showed that the Kupffer cells appear to be a significant source of systemic IL-6 during CLP. Furthermore, the depletion of Kupffer cells by GdCl2 pretreatment not only markedly attenuates the rise in blood IL-6 levels but preserves splenic lymphokine release capacity.86 Outside of the anti-inflammatory cytokines, one of the more important mediators which acts on and is released by macrophages are the prostaglandins, particularly PGE2. Their production can be induced not only by microbial components such as endotoxin, but also by complement, platelet and leukocyte aggregation. There is also evidence that cytokines released such as TNF, IL-1, and TGF-#, may also be direct/indirect stimulants for the release of PGE2.41,54,83,87 This relationship between TNF/IL-1 and prostaglandins also appears to have an autocrine component, since PGE2 can down-regulate the production of both of these cytokines.88-90 One could, therefore, speculate that it is the release of the prostanoids that down-regulates the host’s capacity to clear and destroy the microbes present during acute sepsis. Much of the evidence for this hypothesis has been garnered from the use of cyclooxygenase inhibitors, either in an in vivo or in vitro setting.88,89,91-96 Studies by Ertel et al57 in rats subjected to CLP indicated not only that elevated IL-6 and TNF levels were detected in circulation but also a marked increase in plasma PGE2 levels was observed. Choudhry et al97 have recently reported evidence that prostaglandins appear to mediate lymphocyte dysfunction in the rat sepsis model. However, the role of prostaglandins in the induction of M hyporesponsiveness seen during late sepsis is not established. Another potentially important immune suppressive agent which may contribute to the suppression of both lymphocytic and macrophage mediated immune response following sepsis is nitric oxide (NO). With respect to the macrophage elevated NO release, this has been associated with depressed macrophage antigen presentation as well as a decrease in MHC class II expression.98,99 However, few studies have examined the role of this mediator or its inhibitors with respect to immune dysfunction associated with polymicrobial sepsis. Thus, its contribution to septic immune depression remains to be established.

ETXpump +* _ _ _ _ _

CLP +* _ _ _ _ _

+++ +++

+++ +++

++ +++

Sham

+++ +++

+++ +++

++ +++

Salpump

+++ ++++*

+++ +*

+* +*

CLP

Inducible Cytokine/Lymphokine Release@

(, relative increase reported *, reported to be significant from equivalent Sham/Sal-pump groups. -, either no or background levels of cytokine reported. #, lymphokine/cytokine release reported in the absence of LPS/Concanavalin A (indicative endogenous [in vivo] stimulation). @, lymphokine/cytokine release reported in the presence of LPS/Concanavalin A (an indicator of cell function)

SalTime Sham pump Peritoneal Macrophage IL-1/IL-6/TNF (27,49): 1h _ _ 24h _ _ Splenocyte IL-2/IFN-! (31,50): 1h _ _ 24h _ _ Splenocyte IL-4/IL-10 (31): 1h _ _ 24h _ _

Endogenous Cytokine/Lymphokine Release#

+++ +++

+++ +++

++ +++

ETXpump

Table 20.2. Summary of observed changes in the capacity of macrophage or splenocytes to release cytokines/lymphokines in the absence or presence of exogenous stimulants (LPS for macrophage and concanavalin A for splenocytes) harvested from mice early (1h) or late (24h) following the onset of polymicrobial sepsis (CLP) as opposed to Sham or animals exposed to chronic low dose endotoxin (ETX-pump) or saline (Sal-pump) (27,31,49,50)

Anticytokine Strategies in Peritonitis 255

256

Cytokines and the Abdominal Surgeon

Predisposition of the animal/patient prior to peritoneal/septic insult also has been examined by several laboratories. In this respect, studies from a number of laboratories including our own have documented that traumatic injury or shock markedly decreases the capacity of the animal to survive subsequent sepsis.100-108 These findings would suggest that following trauma or shock the animal which is subsequently exposed to bacterial challenge is not capable of mounting an adequate immune/ cytokine/inflammatory response. This conclusion is based on observation of depressed immune/cytokine/inflammatory responsiveness seen in immune cell populations harvested from animals previously exposed to shock and/or trauma.100-108 Nonetheless, these findings are supported by comparable findings of generalized immune suppression in trauma patients.12,109-113 However, with respect to the animal studies none of these have examined the local and/or systemic cytokine or the immune cell response after the onset of sepsis if sepsis was preceded by shock or trauma. Similarly, in the clinical setting the precise identification of sepsis in the patient has remained unclear, making it difficult to separate traumatized/shocked patients from those which are classically defined as septic or having the “sepsis syndrome”.114 Until recently little or no attention has been paid to the impact of gender or age with respect to the immune cell and/or cytokine mediated response. Zellweger et al115 has recently reported that female (8-12 week old) mice in the pro-estrus state appear to survive septic challenge better than age-matched males and that this is associated with improved lymphocyte responsiveness. However, it was not established if this difference (advantage) was due to related changes in macrophage (liver and/or peritoneal) pro-inflammatory cytokine release or anti-inflammatory mediator production. Also while this study suggested that the salutary effects seen in proestrus female on survival following sepsis might be due to elevated immune enhancing agents such as prolactin, it did not rule out a role for estrogen and/or the lack of immune suppressive androgens.116 Taken together, the information provided above indicates several potential intraabdominal targets, both pro- and anti-inflammatory cytokines and their cellular sources, as potential therapeutic agents.

Potential Intra-Abdominal Pro- or Anti-Inflammatory Cytokine Therapies? Thus far there have been few therapeutic or treatment studies that have been directed solely (or had as their primary objective) at local intra-abdominal pro- or anticytokine release/production. The great majority of studies which have examined sepsis or endotoxic shock experimentally have used intravenous or subcutaneous administration of either the therapeutic/treatment agent(s) and/or the microbes/ microbial toxin(s). Inasmuch as these are adequately addressed in many other reviews as well as chapters in this text we have purposely restricted our discussion to those experimental studies utilizing the localized intra-abdominal treatments in peritonitis models. Of these the best studied was initially the intraperitoneal injection of mice with lethal doses of endotoxin or monospecific Escherichia coli injection35,117-130 (see Table 20.3). Until recently treatment interventions utilized in these studies were primarily carried out to address the hypothetical mechanisms involving the cause and effect relationships. These typically involved pretreatment of the animal with either the cytokine or its antagonist/antibody prior to the administration of endotoxin or E. coli. The initial emphasis was directed at the inhibition of pro-inflammatory

Anticytokine Strategies in Peritonitis

257

cytokine(s) which have been previously identified as agents produced in response to endotoxin or were capable by themselves of producing a “shock-like” state. In this respect, TNF and IL-1 were the initial targets of many of these peritoneal interventions.117-120,124-127,131 Later studies using localized peritoneal administration of agents such as IL-10, IL-11, pentoxifyline, chlorpromazine, dexamethasone or even antibiotics utilized a less mediator selective approach.73,120-123,132 These studies chose instead to use antagonists/inhibitors which were directed at suppressing the effects of multiple pro-inflammatory mediators, cytokines and/or cell populations.73,120-123,132 At face value, these studies typically indicated either suppression of the target proinflammatory cytokine(s) response/release and/or improved the survival of the animal to subsequent lethal endotoxemic/monospecific bacterial challenge. Alternatively, only a few studies have attempted any form of immune-enhancing/supportive localized peritoneal administrative approaches. In this regard, i.p. preor post-treatment with GM-CSF of rodents subjected to either CLP or monospecific Streptococcal infection have shown improvement in both phagocytic capacity and increased overall survival.133,134 Similarly, i.p. pretreatment with G-CSF also improved survival of subsequent lethal E. coli challenge.128 Along these lines, seven day pretreatment of mice with recombinant IL-2 also has been shown to have both a restorative effect on cell-mediated and humoral responsiveness as well as improved survival to monospecific E.coli challenge.135,136 Interestingly, several studies have demonstrated not only that the localized i.p. injection of antibodies against TNF can actually reduce the animals survival when subjected to CLP.35 Additional studies indicate that providing nontoxic (nonshock inducing) doses of TNF and/or IL-1 i.p. before lethal endotoxin/E.coli also improves survival.129,130 These latter findings demonstrate the important immune stimulatory role that these same agents also play which is vital in fighting infectious agents, as opposed to the shock response associated with high doses of cytokines. The above mentioned studies, however, did have several limitations which have largely been ignored. First, such an approach assumes monospecific etiology for sepsis/peritonitis. Second, these studies typically lack an associated inflamed/necrotic tissue site which may actively contribute to the host’s cytokine/immune cell response in a manner not accounted for by endotoxin or mono-bacterial infection alone. Furthermore, the dosage of endotoxin used (see earlier discussion of models) is often very high. Moreover, these models all utilize otherwise healthy young male subjects not predisposed to factors such as age, nutritional status, gender, or prior exposure to a traumatic or stressful event. In as much, the absence of these components in the utilization of the animal models such as endotoxemia may have in part contributed to the misinterpretation and/or inappropriate extrapolation of data from such animal/volunteer patient studies to the setting of clinical peritoneal sepsis. Irrespective of whether the treatment target was endotoxin or a pro-/anti-inflammatory cytokine, studies in which antibody/antagonist was administered to animals intraperitoneally did provide an initial basis for considering local intra-abdominal therapeutic approaches. However, the major drawback with the potential clinical application of these approaches is that the majority of these therapies have been directed against the pro-inflammatory cytokine response and have been primarily of a pretreatment nature (see Table 20.3). While this list is not inclusive of all localized sepsis-treatment models which may have been attempted, it is evident that no more than 10-15% of them utilized a post-treatment approach.125,131,133 Another potential disadvantage (which was not a concern nor a typical objective of these studies) is

Peritonitis Model

both CPX & DEX )plasma TNF but only DEX also )IL-1/IL-6, CPZ but not SEX ( IL-10 production, survival not assessed.

Endotoxemia (mouse/i.p.)

Both TNF-RI and IL-10 gene transfer

IL-10: (survival, )plasma TNF, Anti-IL-10: )survival, (plasma TNF

(survival, )plasma TNF

(survival, )serum TNF

survival not assessed

)anorexia/et. loss, )IL-6, )acute phase proteins but no effect on ETX ) of albumin synthesis,

(survival

(survival

Outcome(s)

Lethal Endotoxemia (rat/i.p.)

Howard et al (122) IL-10 (0.5-10 ∝g/mouse/i.p.) pretreatment (30 min) Lethal Endotoxemia (1 mg/mouse/i.p.) Walley et al (73) IL-10 (5 ∝g/i.p.) vs. anti-IL-10 antibody CLP (mouse) (i.p.) pretreatments Rogy et al (123) Cationic liposomal human TNF-RI or human Lethal endotoxemia (250 ng ETX &

Schade (120) Pentoxifylline (50 mg/kg BW/i.p) pretreatment (1 h) Mengozzi et al (121) Chlorpromazine (CPZ)(i.p.) vs. dexamethasone DEX)(i.p.) pretreatment (30 min or 24 h).

Localized Anticytokine or Anti-inflammatory Cytokine Treatments: Beutler et al (117) Anti-TNF antibody (i.p.) pretreatment (5 h) Lethal Endotoxemia (400 ∝g/mouse/i.p.) Nassif et al (118) Anti-TNF antibody (i.p.) pretreatment Lethal N. meningitidis (rat/i.p.) challenge Sharma et al (119) Anti-TNF antibody (i.p.) pretreatment (4 h) Lethal Endotoxemia (rat/i.p.)

Local/Intra-abdominal Treatment:

Inhibition of Pro-inflammatory Cytokine(s) Response:

Table 20.3. Experimental local/intra-abdominal treatment studies/approaches in abdominal sepsis/peritonitis.

258 Cytokines and the Abdominal Surgeon

Continued...

Austin et al (134)

(recruitment.

Application of Immune Enhancing/Hematopoietic Stimulants: Wheeler & Givner (133) Human GM-CSF (i.p.) post-treatment (7-19 h) Lethal group B Streptococcus (newborn rats/i.p.) challenge

IL-10 gene transfection (200 ∝g/i.p.) 18 D-galactosamine) )as pretreatment (48 h))plasma TNF and lung MPO. Fletcher et al (124) Tetra peptide inhibitor of IL-1# converting Endotoxemia (mouse/i.p.) enzyme (L-709,049)(i.p.) pretreatment Luheshi et al (125) IL-1 receptor antagonist (IL-1ra)(i.p.) Endotoxemia (100 ∝g/lg BW, rat, i.p.) 1-2 h post-ETX-treatment McNamara et al (126) Anti IL-1 receptor antibody (IL-1rab; 100 to 200 Lethal endotoxemia (30-40 mg/kg ∝g/mouse/i.p.) pretreatment (1-6 h). BW/mouse/i.p.) challenge. Combined IL-1rab and anti-TNF- antibody pretreatment Chang et al (132) Human IL-11 (250 ∝g/i.p.) pretreatment (11 d) Lethal group B Streptococcus (newborn rats/i.p.) challenge Acton et al (127) Transgenic (knock-out) mice deficient in Lethal live E. coli (mouse/i.p.) challenge either TNF-type I (RI, p55) receptor or the vs. lethal endotoxemia (mouse/i.v.) p80 IL-1 receptor (equivalent to pretreatment).

(survival, no change in blood leukocyte number or oxidate burst capacity but possible priming of localized peritoneal phagocytic cells

Both transgenic mouse strains exhibit ( survival of i.v. ETX, only p80 IL-1 receptor deficient mouse (survival but not TNF-RI deficient mice.

(survival, (recruitment, speculated effects on gut.

IL-1rab alone: (survival, )plasma IL-6 Combined IL-1rab/anti-TNF no improvement in survival

)fever, )plasma IL-6 bioactivity

)plasma IL-1# without effecting IL-1∀ or IL-6 plasma levels, survival not assessed.

(survival. IL-10 but not TNF-RI gene transfer

Anticytokine Strategies in Peritonitis 259

Fraker et al (129) TNF-∀ (25 or 100 ∝g/kg BW, i.p.) pretreatment (5 d) or tolerization Cross et al (130) TNF-∀ (1 or 10 ∝g/kg BW, i.p.) and/or IL-1∀ (500 or 1000 U/kg BW, i.p.) pretreatment (just before ETX) Eskandari et al (35) Anti-TNF antibody (i.p.) pretreatment (4 h)

Weyand et al (135,136) Mouse IL-2 (i.p.) pretreatment (7 d) Assessment of Deleterious Effects of Pro-Inflammatory Mediator Inhibition/loss: Bagby et al (72) Anti-TNF antibody (i.p.) pretreatment (1-2 h)

(survival, )histopathology

(survival

)plasma TNF after CLP or ETX, )survival after CLP or ETX Did not inhibit CLP associate ( lung MPO/PMN

Lethal E. coli (mouse/i.p.) challenge

CLP or Lethal Endotoxemia (mouse/i.v or i.p.)

Unlike protection seen against i.v. ETX no protection seen in fecal peritonitis model.

Lethal E. coli Fecal peritonitis (rat/i.v or i.p.) compared to i.v ETX Lethal Endotoxemia (rat/i.p.)

(survival; (humoral (serum IgG) response

bactericidal activity

(survival, (blood PMN number as well as delayed (peritoneal PMN number,( PMN

cell functions.

(survival, (peritoneal cell influx, (phagocytic

Outcome(s)

Lethal E. coli (mouse/i.p.) challenge

Lethal E. coli (mouse/i.p. implantation of bacterial contaminated agar pellet) challenge

CLP (mouse)

GM-CSF (i.p.) pretreatment (5 d)

Dunne et al (128) Human G-CSF (i.p.) pretreatment

Peritonitis Model

Local/Intra-abdominal Treatment:

Table 20.3. (Continued)

260 Cytokines and the Abdominal Surgeon

Assessment of Intra-abdominally Administered Antibiotics on Septic Pro-Inflammatory Response: Rosman et al (131) Imipenem or Taurolidine (i.p.) post-treatment. Bi-specific lethal entero-bacterial (rat/i.p.) challenge

Deletion of Cellular Sources of Cytokines: Wiktor-Jedrzejczak et al (37) Macrophage deficient (less than 5% normal Lethal Fecal (autoclaved feces & E. coli ) complement of peritoneal macrophage) Peritonitis (mouse/i.p.) Osteopetrotic op/op mice (equivalent to pretreatment) Callery et al (36) CLP (mouse) Depletion of Kupffer cells by prior GdCl2 pretreatment)survival Ayala et al (86) Depletion of Kupffer cells by prior GdCl2 CLP (mouse) pretreatment (Splenocyte IL-2 release )Splenocyte IL-4 release,

Survival highest in imipenem group, TNF in plasma and abdominal fluid level in the Taurolidine treated group was lower than sham, while TNF levels were increased by Imipenem

survival was not assessed

)plasma IL-6 but no change in TNF

(Splenocyte proliferation & IL-2 release

infiltration

)survival of op/op mice, )peritoneal PMN

Anticytokine Strategies in Peritonitis 261

262

Cytokines and the Abdominal Surgeon

that proteinaceous agents used in these treatments did not remain confined to the abdominal/peritoneal cavity and hence affected extra-visceral targets. However, as there are systemic sequella often associated with peritoneal sepsis this may be more of a perceived theological drawback then an actual problem. With the advent of gene therapy one of the more novel approaches to intra-abdominal cytokine/anticytokine intervention has been recently advanced by the work of Rogy et al.123 If one ignores the two marked limitations of this study (those being that it is a mechanistic pretreatment based study and that the model is one of lethal endotoxemia [see earlier discussions]), the investigators have clearly demonstrated that in a situation where a known inflammatory stimulant such as i.p. endotoxin is administered that the effects of this form of shock can be inhibited by prior (48 h prior) transient transfection with cationic liposome suspension containing either a vector with the human TNF-RI(p55) or human IL-10 gene inserted. Furthermore, the transfected gene expression at 48 h after i.p. administration was restricted to visceral sites, such as the liver, kidney and spleen, but was not detectable in peripheral blood leukocytes. There was, however, some evidence of transfected gene expression in the lungs which the investigators suggested was due to lymphatic drainage to these distal sites. However, this study does not give insight as to the actual phenotype (immune/nonimmune) of the cells which expressed the transfected gene. Also of interest is the transient nature of this form of transfection (rapid decline in gene expression by 96 h after transfection). This may also be another potential advantage of this form of site/tissue directed therapy, as it reduces the chances of long term unintended effects of these forms of cytokine therapy. In light of this, transient transfection may provide a possible intra-abdominal therapeutic delivery system. To the extent that the cell/tissue/organ sites/sources of cytokine release have been the targets of intra-abdominal therapies, several investigators have carried out, primarily mechanistically driven studies (where therapeutic intervention was not the goal) which indirectly shed light on this possibility. Callery et al36 documented that Kupffer cells appear to be a key source of the inflammatory cytokine IL-6 following CLP in rats. Furthermore, intravenous administration of gadolinium chloride (GdCl2) (a rare earth element which can be used to inhibit tissue phagocyte function and kill them; and deplete Kupffer cell numbers when given i.v.) attenuated the rise in systemic IL-6 levels. We have made a comparable observation in mice pretreated with i.v. GdCl2 to significantly decrease Kupffer cell number prior to CLP and found that not only were IL-6 levels depressed systemically but distant splenocyte immune responses were protected.86 Interestingly peritoneal macrophages responses were not effected by i.v. injection of GdCl2. We would speculate that intra-peritoneal administration of GdCl2 should have a more restrictive intra-abdominal effect. This, however, has not been determined. In this respect, other agents such as desulfated-carrageenan137 or liposome-encapsulated dichloromethylene diphosphonate138,139 have been utilized to deplete animals of macrophages. While the studies with both these agents efficiently depleted the animals peritoneum of macrophage, their effects were not restricted to the peritoneal cavity, as macrophage in the liver, spleen and thymus were also effected to some extent. Also while these are novel methods of cell depletion, no functional assessment of cytokine responsiveness nor the response to septic challenge was examined in such mice. Moreover, as eluded to earlier the aims of these studies were to define the source of systemic pro-inflammatory cytokines such as IL-6 in response to septic insult and not to address therapeutic efficacy. Importantly, the potential limitations of such cell ablative approach were also shown in a

Anticytokine Strategies in Peritonitis

263

subsequent study of Callery et al36 in which GdCl2 pretreated rats were subjected to subsequent CLP. They reported that the mortality in these GdCl2 pretreated animals was markedly increased. This may be a reflection of a trade-off for diminished cytokine release at the expense of the loss of the capacity of the hepatic reticuloendothelial systems ability to clear and destroy the microbial pathogens. Support for this view also comes from the work of Wiktor-Jedrzejczak et al37 who examined the survival of macrophage deficient op/op knockout mice. They found not only that a decrease in neutrophil/granulocyte infiltration of the peritoneal cavity (an event mediated in large part by cytokines), but also a marked decline in the survival of animals subjected to lethal fecal peritonitis.

Summary The available information indicates that there are several directions which have and continue to be pursued, which either directly or indirectly, have attempted to address the plausibility of localized anticytokine therapy in intra-abdominal sepsis/ peritonitis. We have discussed studies which have moved from the application of monospecific passive immunization against specific pro-inflammatory cytokines (such as antibodies to either TNF and IL-1) in simple models of lethal intra-peritoneal endotoxemia, through the use of transgenic (knock-out) mice (deficient in a given pro- or anti-inflammatory cytokine), to the more recent application of gene therapy (transient transfection) to not only increase our understanding of the pathophysiology of sepsis/peritonitis but as possible localized therapeutic approaches. While each approach has certain merits, they also must be considered within the limitations of the experimental models which are employed if they are to be salutary in patients. We would also speculate that as our understanding of the pathophysiology of abdominal sepsis and its link to multiple organ failure evolves we will move from the present state of all-or-none cytokine stimulation /inhibition, to one of selective immune cell population activation or suppression as required. However, to do this we will require not only a more complete knowledge of the developing immune/ cytokine response to various forms of peritoneal insult (the teleology of cytokine gene/protein expression/release; the nature of the cellular source(s) as well as their nidus for stimulation and/or suppression; but also knowledge of their role in the developing host immune response and their interactions with other mediators), but we will need to ask these questions in models which reflect the diversity of predispositions encountered in the patient populations in which we wish to apply these cytokine and/or anticytokine therapies. Moreover, while we see post-treatment protocols as the potentially most clinically applicable type of investigations, it is still important not to loose sight of information that has and still is provided from pretreatment models such as per-immunization or knock-out animals. These tools of investigation provide novel insight into the pathobiology of this condition, which at present is not provided in any other way. If the problem is approached in this fashion it should be possible develop intra-abdominal therapeutic interventions which may then become a plausible treatment option to the patient.

References 1. Baue AE. Multiple Organ Failure. Patient Care and Prevention. St. Louis: Mosby, 1990:336-363. 2. Goris RJA. Sepsis and multiple organ failure: The result of whole body inflammation. In: Faist E, Meakins J, Schildberg FW, eds. Host Defense Dysfunction in Trauma, Shock and Sepsis, Berlin-Heidelberg: Springer-Verlag, 1993:161-170.

264

Cytokines and the Abdominal Surgeon

3. Suffredini AF. Current prospects for the treatment of clinical sepsis. Crit Care Med 1994; 22:S12-S18. 4. Barriere SL, Lowry SF. An overview of mortality risk prediction in sepsis. Crit Care Med 1995; 23:376-393. 5. Marshall J, Sweeney D. Microbial infection and the septic response in critical surgical illness. Arch Surg 1990; 125:17-23. 6. Marshall JC. Immunologic dyshomeostasis in multiple organ failure: The gut-liver axis. In: Faist E, Meakins J, Schildberg FW, eds. Host, Dysfunction in Trauma, Shock and Sepsis, Berlin-Heidelberg: Springer-Verlag, 1993:243-252. 7. Baue AE. The horror autotoxicus and multiple-organ failure. Arch Surg 1992; 127:1451-1462. 8. Deitch EA. Multiple organ failure: Pathophysiology and potential future therapy. Ann Surg 1992; 216:117-134. 9. Friese RS, Rehring TF, Wollmering M et al. Trauma primes cells: Editorial review. Shock 1994; 1:388-394. 10. Fink MP, Heard SO. Laboratory models of sepsis and septic shock. J Surg Res 1990; 49:186-196. 11. Schlag G, Redl H. Pathophysiology of Shock, Sepsis and Organ Failure. Berlin: Springer-Verlag 1993:1-1142. 12. Faist E, Meakins J, Schildberg FW. Host Defense Dysfunction in Trauma, Shock and Sepsis. Berlin-Heidelberg: Springer-Verlag, 1993:1-1296. 13. Chaudry IH, Ayala A, Singh G, Wang P, Hauptman JG. Rodent models of endotoxemia and sepsis. In: Schlag G, Redl H, eds. Pathophysiology of Shock, Sepsis and Organ Failure. Berlin: Springer-Verlag, 1993:1048-1059. 14. Wichterman Y, Baue AE, Chaudry IH. Sepsis and septic shock—A review of laboratory models and a proposal. J Surg Res 1980; 29:189-201. 15. Cerra FB. Hypermetabolism-organ failure syndrome: A metabolic response to injury. Crit Care Clinics 1989; 5:289-302. 16. Groeneveld AB, Bronsveld W, Thijs LG. Hemodynamic determinants of mortality in human septic shock. Surgery 1986; 99:140-153. 17. Deitch EA. Intestinal permeability is increased in burn patients shortly after injury. Surgery 1990; 107:411-415. 18. Border JR, Hasset J, LaDuca J, Seibel R, Steinberg S, Mills B, Losi P, Border D. The gut origin septic states in blunt multiple trauma (ISS=40) in the ICU. Ann Surg 1987; 206:427-448. 19. Schlag G, Redl H, Davies J, van Vuuren CJJ, Smuts P. Bacterial Translocation During Traumatic Shock in Baboons. In: Schlag G, Redl H, eds. Pathophysiology of Shock, Sepsis, and Organ Failure, Berlin Heidelberg: Springer-Verlag, 1993:279-291. 20. Mathison JC, Virca GD, Wolfson E, Tobias PS, Glaser K, Ulevitch RJ. Adaptation to bacterial lipopolysaccharide controls lipopolysaccharide-induced tumor necrosis factor production in rabbit macrophages. J Clin Invest 1990; 85:1108-1118. 21. D’Orio V, Wahlen C, Rodriguez LM, Fossion A, Juchmes J, Halleux J, Marcelle R. A comparison of Escherichia coli endotoxin single bolus injection with low-dose endotoxin infusion on pulmonary and systemic vascular changes. Circ Shock 1987; 21:207. 22. Zuckerman SH, Bendele AM. Regulation of serum TNF in glucocorticoid-sensitive and -resistant rodent endotoxin shock models. Infect Immun 1989; 57:3009-3013. 23. Roth BL, Neilsen TB, McKee AE. Molecular and cellular mechanisms of septic shock. New York: Alan R. Liss, Inc. 1988.

Anticytokine Strategies in Peritonitis

265

24. Fink MP, Morrissey PE, Stein KL, Clement RE, Fiallo V, Gardiner WM. Systemic and regional hemodynamic effects of cyclooxygenase and thromboxane synthetase inhibition in normal and hyperdynamic endotoxemic rabbits. Circ Shock 1988; 26:41-57. 25. Johnston TD, Hampton WH, Fry DE. E. coli endotoxin produces hyperdynamic hemodynamics in the rat. Circ Shock 1989; 27:320. 26. Hesse DG, Tracey KJ, Fong Y, Manogue KR, Palladino MA, Jr., Cerami A, Shires GT, Lowry SF. Cytokine appearance in human endotoxemia and primate bacteremia. Surg Gynecol Obstet 1988; 166:147-153. 27. Ayala A, Deol ZK, Lehman DL, Herdon CD, Chaudry IH. Does endotoxin play a major role in inducing the depression of macrophage function during polymicrobial sepsis? Arch Surg 1995; 130:1178-1185. 28. Aasen AO. LPS plasma levels in patients. In: Schlag G, Redl H, eds. Pathophysiology of Shock, Sepsis and Organ Failure. Berlin: Springer-Verlag, 1993:417-426. 29. Danner RL, Elin RJ, Hosseini JM, Wesley RA, Reilly JM, Parillo JE. Endotoxemia in human septic shock. Chest 1991; 99:169-175. 30. Van Deventer SJH, Buller HR, Ten Cate JW, Struk A, Pauw W. Endotoxemia: an early predictor of septicaemia in febrile patients. Lancet 1988; 1:605-609. 31. Ayala A, Deol ZK, Lehman DL, Herdon CD, Chaudry IH. Polymicrobial sepsis but not low dose endotoxin infusion causes decreased splenocyte IL-2/IFN-gamma release while increasing IL-4/IL-10 production. J Surg Res 1994; 56:579-585. 32. Piper RD, Cook DJ, Bone RC, Sibbald WJ. Introducing critical appraisal to studies of animal models investigating novel therapies in sepsis. Crit Care Med 1996; 24:2059-2070. 33. Paul WE. Fundamental Immunology. New York: Raven. 1993. 34. Roitt IM. Essential Immunology. Oxford, UK: Blackwell. 1988. 35. Eskandari MK, Bolgos G, Miller C, Nguyen DT, DeForge LE, Remick DG. Antitumor necrosis factor antibody therapy fails to prevent lethality after cecal ligation and puncture or endotoxemia. J Immunol 1992; 148:2724-2730. 36. Callery M, Kamei T, Flye MW. Kupffer cell blockade increases mortality during intra-abdominal sepsis despite improving systemic immunity. Arch Surg 1990; 125:36-41. 37. Wiktor-Jedrzejczak W, Dzwigala B, Szperl M, Maruszynski M, Urbanowska E, Szwech P. Colony-stimulating factor 1-dependent resident macrophages play a regulatory role in fighting Escherichia coli fecal peritonitis. Infection & Immunity 1996; 64:1577-1581. 38. Lewis RA, Austen KF. Leukotrienes. In: Gallin JI Goldstein IM, Snyderman R, eds. Inflammation: Basic Principles and Clinical Correlates, New York: Raven Press, 1988:121-128. 39. Hamilton TA, Adams DO. Molecular mechanisms of signal transduction in macrophages. Immunol Today 1987; 8:151-158. 40. Border J. Hypothesis: Sepsis, multiple systems organ failure, and the macrophage. Arch Surg 1988; 123:285-286. 41. Beutler B, Cerami A. The biology of cachectin/TNF—a primary mediator of the host response. Ann Rev Immunol 1989; 7:625-655. 42. Waage A, Brandtzaeg P, Halstensen A, Kierulf P, Espevik T. The complex pattern of cytokines in serum from patients with meningococcal septic shock. Association between interleukin 6, interleukin 1, and fatal outcome. J Exp Med 1989; 169:333-338. 43. Damas P, Reuter A, Gysen P, Demonty J, Lamy M, Franchimont P. Tumor necrosis factor and interleukin-1 serum levels during severe sepsis in humans. Crit Care Med 1989; 17:975-978.

266

Cytokines and the Abdominal Surgeon

44. Marks JD, Marks CB, Luce JM, Montgomery B, Turner J, Metz CA, Murray JF. Plasma tumor necrosis factor in patients with septic shock. Am Rev Respir Dis 1990; 141:94-97. 45. Marano MA, Fong Y, Moldawer LL, Wei H, Calvano SE, Tracey KJ, Barie PS, Manogue K, Cerami A, Shires GT, Lowry SF. Serum cachectin/tumor necrosis factor in critically ill patients with burns correlates with infection and mortality. Surg Gynecol Obstet 1990; 170:32-38. 46. van der Poll T, Lowry SF. Tumor necrosis factor in sepsis: Mediator of multiple organ failure or essential part of host defense? Shock 1995; 3:1-12. 47. Maier RV. The “angry” macrophage and its impact on host response mechanisms. In: Faist E, Meakins J, Schildberg FW, eds. Host Defense Dysfunction in Trauma, Shock and Sepsis, Berlin-Heidelberg: Springer-Verlag, 1993:191-197. 48. Marshall JC. Immunologic dyshomeostasis in multiple organ failure: The gut-liver axis. In: Faist E, Meakins J, Schildberg FW, eds. Host Defense Dysfunction in Trauma, Shock and Sepsis, Berlin-Heidelberg: Springer-Verlag, 1993:245-252. 49. Ayala A, Perrin MM, Kisala JM, Ertel W, Chaudry IH. Polymicrobial sepsis selectively activates peritoneal but not alveolar macrophage to release inflammatory mediators (IL-1, IL-6 and TNF). Circ Shock 1992; 36:191-199. 50. Ayala A, Knotts JB, Ertel W, Perrin MM, Morrison MH, Chaudry IH. Role of interleukin 6 and transforming growth factor-beta in the induction of depressed splenocyte responses following sepsis. Arch Surg 1993; 128:89-95. 51. Munoz C, Misset B, Fitting C, Bleriot JP, Carlet J, Cavaillon JM. Dissociation between plasma and monocyte-associated cytokines during sepsis. Eur J Immunol 1991; 21:2177-2184. 52. Munoz C, Carlet J, Fitting C, Misset B, Bleriot J, Cavaillon J. Dysregulation of in vitro cytokine production by monocytes during sepsis. J Clin Invest 1991; 88:1747-1754. 53. Choudhry MA, Ahmad S, Thompson KD, Sayeed MM. Erratum: T-lymphocyte Ca2+ signalling and proliferative responses during sepsis. Shock 1:466-471, 1994. 54. Miller-Graziano CL, Szabo G, Kodys K, Mehta B. Interactions of immunopathological mediators (tumor necrosis factor-∀, TGF-#, prostaglandin E2) in traumatized individuals. In: Faist E, Meakins JL, Schildberg FW, eds. Host Defense Dysfunction in Trauma, Shock and Sepsis: Mechanisms and Therapeutic Approaches, Berlin. Springer-Verlag, 1993:637-650. 55. Meldrum DR, Ayala A, Chaudry IH. Energetics of lymphocyte “burnout” in late sepsis: Adjuvant treatment with ATP-MgCl2 improves energetics and decreases lethality. J Surg Res 1994; 56:537-542. 56. Ayala A, Kisala JM, Felt JA, Perrin MM, Chaudry IH. Does endotoxin tolerance prevent the release of inflammatory monokines (IL-1, IL-6, or TNF) during sepsis? Arch Surg 1992; 127:191-197. 57. Ertel W, Morrison MH, Wang P, Ba ZF, Ayala A, Chaudry IH. The complex pattern of cytokines in sepsis—Association between prostaglandins, cachectin and interleukins. Ann Surg 1991; 214:141-148. 58. Ayala A, Perrin MM, Wang P, Chaudry IH. Sepsis induces an early increased spontaneous release of the hepatocellular stimulatory factor (IL-6) by Kupffer cells in both endotoxin tolerant or intolerant mice. J Surg Res 1992; 52:635-641. 59. Hadjiminas DJ, McMasters KM, Peyton JC, Cheadle WG. Tissue tumor necrosis factor mRNA expression following cecal ligation and puncture or intraperitoneal injection of endotoxin. J Surg Res 1994; 56:549-555. 60. Ayala A, Knotts JB, Lehman DL, Pierson CR, Patel P, Chaudry IH. Early administration of diltiazem is insufficient to block the late changes in splenocyte lymphokine release and total cell calcium induced by sepsis. FASEB J 1993; 7:A593.(Abstract)

Anticytokine Strategies in Peritonitis

267

61. Knotts J, Ayala A, Chaudry IH. Increased splenocyte [Ca+2] during polymicrobial sepsis. The FASEB J 1991; 5:A1102. 62. Gallinaro RN, Naziri W, McMasters KM, Peyton JC, Cheadle WG. Alteration of mononuclear cell immune-associated antigen expression, interleukin-1 expression, and antigen presentation during intra-abdominal infection. Shock 1994; 1:130-134. 63. Ayala A, Ertel W, Chaudry IH. Trauma-induced suppression of antigen presentation and expression of major histocompatibility class II antigen complex. Shock 1996; 5(2):79-90. 64. June CH, Bluestone JA, Nadler LM, Thompson CB. The B7 and CD28 receptor families. Immunol Today 1994; 15:321-331. 65. Zhu X, Zellweger R, Wichmann MW, Ayala A, Chaudry IH. Effects of prolactin and metoclopramide on macrophage cytokine gene expression in late sepsis. Cytokine 1997; 9:437-446. 66. Waage A, Halstensen A, Espevik T. Association between tumor necrosis factor in serum and fatal outcome in patients with meningococcal disease. Lancet 1987; i:355-357. 67. Okusawa S, Gelfand JA, Ikejima T, Connolly RJ, Dinarello CA. Interleukin-1 induces a shock like state in rabbits. Synergism with tumor necrosis factor and the effects of cyclooxygenase inhibition. J Clin Invest 1988; 81:1162-1172. 68. Tracey KJ, Beutler B, Lowry SF et al. Shock and tissue injury induced by recombinant human cachectin. Science 1986; 234:470-474. 69. Schirmer WJ, Schirmer JM, Fry DE. Recombinant human tumor necrosis factor produces hemodynamic changes characteristic of sepsis and endotoxemia. Arch Surg 1989; 124:445-448. 70. Deitch EA. Multiple organ failure: Summary and overview. In: Deitch EA ed. Multiple Organ Failure, New York: Thieme Medical Publishers, 1990:285-299. 71. Hadjiminas DJ, McMasters KM, Peyton JC, Cook MD, Cheadle WG. Passive immunization against tumor necrosis factor and interleukin-1 fails to reduce lung neutrophil sequestration in chronic sepsis. Shock 1994; 2:376-380. 72. Bagby GJ, Plessala KJ, Wilson LA, Thompson JJ, Nelson S. Divergent efficacy of antibody to tumor necrosis factor-alpha in intravascular and peritonitis models of sepsis. J Infect Dis 1991; 163:83-88. 73. Walley KR, Lukacs NW, Standiford TJ, Strieter RM, Kunkel SL. Balalnce of inflammatory cytokines related to severity and mortality of murine sepsis. Infect Immun 1996; 64:4733-4738. 74. Trinchieri G. Interleukin-12 and its role in the generation of TH1 cells. Immunol Today 1993; 14:335-338. 75. Biron CA, Gazzinelli RT. Effect of IL-12 on immune responses to microbial infections: a key mediator in regulating diseases outcome. Curr Opin Immunol 1995; 7:485-496. 76. Kohno K, Kataoka J, Ohtsuki T, Suemoto Y, Okamoto I, Usui M, Ikeda M, Kurimoto M. IFN-gamma-inducing factor (IGIF) is a costimulatory factor on the activation of Th1 but not Th2 cells and exerts its effect independently of IL-12. J Immunol 1997; 158:1541-1550. 77. Tsutsui H, Matsui K, Kawada N, Hyodo Y, Hayashi N, Okamura H, Higashino K, Nakanishi K. IL-18 accounts for both TNF-∀ and fas ligand-mediated hepatotoxic pathways in endotoxin-induced liver injury inmice. J Immunol 1997; 159:3961-3967. 78. Ertel W, Scholl F, Steckholzer U, Trentz O. In vivo administration of Interleukin10 attenuates the release of proinflammatory cytokines but causes suppression of cellular immunity in experimental sepsis. Arch Surg 1996.(In Press) 79. Hauser CJ. Regional macrophage activation after injury and the compartmentalization of inflammation in trauma. New Horizons 1996; 4:235-251.

268

Cytokines and the Abdominal Surgeon

80. Hauser CJ, Lagoo S, Lagoo A, Hale E, Hardy KJ, Barber WH, Bass JD, Poole GV. Tumor necrosis factor alpha gene expression in human peritoneal macrophages is suppressed by extra-abdominal trauma. Arch Surg 1995; 130:1186-1191. 81. Massague J. The transforming growth factor-# family. Annu Rev Cell Biol 1990; 6:597-641. 82. Wahl SM, McCartney-Francis N, Mergenhagen SE. Inflammatory and immunomodulatory roles of TGF-#. Immunol Today 1989; 10:258-261. 83. Miller-Graziano CL, Szabo G, Griffey K, Metha B, Kodys K, Catalano D. Role of elevated monocyte transforming growth factor # (TGF-#) production in posttrauma immunosuppression. J Clin Immunol 1991; 11:95-102. 84. Zhou D, Munster AM, Winchurch RA. Inhibitory effects of interleukin 6 on immunity. Possible implications in burn patients. Arch Surg 1992; 127:65-69. 85. Zhou D, Munster A, Winchurch RA. Pathologic concentrations of interleukin 6 inhibit T cell responses via induction of activation of TGF-# . FASEB J 1991; 5:2582-2585. 86. Ayala A, O’Neill PJ, Uebele SA, Herdon CD, Chaudry IH. Mechanism of splenic immunosuppression during sepsis: key role of kupffer cell mediators. J Trauma 1997; 42:882-888. 87. Sherry B, Cerami A. Mini-Review. Cachectin/TNF exerts endocrine, paracrine, and autocrine control of inflammatory responses. J Cell Biol 1988; 107:1269-1277. 88. Kunkel SL, Wiggins RC, Chensue SW, Larrick J. Regulation of macrophage tumor necrosis factor production by prostaglandin E2. Biochem Biophys Res Comm 1986; 137:404-410. 89. Kunkel SL, Chensue SW, Phan SH. Prostaglandins as endogenous mediators of interleukin-1 production. J Immunol 1986; 136:186-192. 90. Kunkel SL, Remick DG, Strieter RM, Larrick JW. Mechanisms that regulate the production and effects of tumor necrosis factor-alpha. Crit Rev Immunol 1989; 9:93-117. 91. Ertel W, Morrison MH, Ayala A, Dean RE, Chaudry IH. Interferon-gamma attenuates hemorrhage-induced suppression of macrophage and splenocyte functions and decreases susceptibility to sepsis. Surgery 1992; 111:177-187. 92. Stephan RN, Conrad PH, Saizawa M, Dean RE, Chaudry IH. Prostaglandin E2 depresses antigen-presenting cell function of peritoneal macrophages. J Surg Res 1988; 44:733-739. 93. Zlotnik A, Shimonkevitz R, Kappler J, Marrack P. Effect of prostaglandin E2 on interferon-! induction of antigen presenting ability of P388D1 cells and on IL-2 production by T cell hybridoma. Cell Immunol 1985; 90:154-166. 94. Snyder DS, Beller DI, Unanue ER. Prostaglandins modulate macrophage Ia expression. Nature 1982; 299:163-165. 95. Garcia-Penarrubia P, Bankhurst AD, Koster FT. Prostaglandins from human T-suppressor/cytotoxic cells modulate natural killer antibacterial activity. J Exp Med 1989; 170:601-606. 96. Metz CA, Sheagren JN. Ibuprofen in animal models of septic shock. J Crit Care 1990; 5:206-212. 97. Choudhry MA, Ahmad S, Sayeed MM. Role of Ca2+ in prostaglandin E2-induced T-lymphocyte proliferative suppression in sepsis. Infect Immun 1995; 63:3101-3105. 98. Albina JE, Cui S, Mateo RB, Reichner JS. Nitric oxide-mediated apoptosis in murine peritoneal macrophages. J Immunol 1993; 150:5080-5085. 99. Sicher SC, Vazquez MA, Lu CY. Inhibition of macrophage Ia expression by nitric oxide. J Immunol 1994; 153:1293-1300. 100. Chaudry IH, Ayala A. Immunological Aspects of Hemorrhage. Austin, TX: Medical Intelligence Unit; R.G. Landes Company, 1992:1-132.

Anticytokine Strategies in Peritonitis

269

101. Chaudry IH, Ayala A. Mechanism of increased susceptibility to infection following hemorrhage. Am J Surg 1994; 165:S59-S67. 102. Alexander J. Nutrition and resistance to sepsis. In: Faist E Ninnemann J, Green D eds. Immune Consequences of Trauma, Shock, and Sepsis: Mechanisms and Therapeutic Approaches, Berlin: Springer-Verlag, 1989:375-382. 103. Alexander JW. Mechanism of immunologic suppression in burn injury. J Trauma 30:S70-S75, 1990. 104. Herndon DN, Zeigler ST, Rutan TC et al. Bacterial translocation following major thermal injury in rats, sheep, and mini-pigs. In: Faist E, Meakins J, Schildberg FW, eds. Host Defense Dysfunctions in Trauma, Shock and Sepsis, Berlin-Heidelberg: Springer-Verlag, 1993:893-897. 105. Zellweger R, Ayala A, Zhu X, Holme KR, DeMaso CM, Chaudry IH. A novel nonanticoagulant heparin improves splenocyte and peritoneal macrophage immune function after trauma-hemorrhage and resuscitation. J Surg Res 199559:211-218. 106. O’Suilleabhain C, O’Sullivan ST, Kelly JL, Lederer J, Mannick JA, Rodrick ML. Interleukin-12 treatment restores normal resistance to bacterial challenge after burn injury. Surgery 1996; 120:290-296. 107. Hill AD, Naama H, Shou J, Calvano SE, Daly JM. Antimicrobial effects of granulocyte-macrophage colony-stimulating factor in protein-energy malnutrition. Arch Surg 1995; 130:1273-1277. 108. Redmond HP, Gallagher HJ, Shou J, Daly JM. Antigen presentation in proteinenergy malnutrition. Cell Immunol 1995; 163:80-87. 109. Cheadle WG, Hershman MJ, Wellhausen SR, Polk HC, Jr. Role of monocyte HLADR expression following trauma in predicting clinical outcome. In: Faist E, Ninnemann J, Green D, eds. Immune Consequences of Trauma, Shock, and Sepsis: Mechanisms and Therapeutic Approaches, Berlin. Springer-Verlag, 1989:119-122. 110. Livingston DH, Apel SH, Wellhausen SR, Sonnenfeld G, Polk HC. Depressed interferon-gamma production and monocyte HLA-DR expression after severe injury. Arch Surg 1988; 123:1309-1312. 111. Christou NV, MacLean L, Meakins JL. Host defense in blunt trauma: interrelationships of kinetics of anergy and depressed neutrophil function: Nutritional status and sepsis. J Trauma 1980; 20:833-841. 112. Christou NV, Tellado JM. The impact of preexisting disease conditions for host defense integrity in traumatized and critically ill patients. In: Faist E, Meakins J, Schildberg FW, eds. Host Defense Dysfunction in Trauma, Shock and Sepsis, Berlin-Heidelberg: Springer-Verlag, 1993:73-82. 113. Miller-Graziano CL, Szabo G, Kodys K, Mehta B. Interactions of immunopathological mediators (tumor necrosis factor-∀, TGF-#, prostaglandin E2) in traumatized individuals. In: Faist E, Meakins J, Schildberg FW, eds. Host Defense Dysfunction in Trauma, Shock and Sepsis, Berlin-Heidelberg: Springer-Verlag, 1993:637-650. 114. Bone RC, Balk RA, Cerra FB et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 1992; 101:1644-1655. 115. Zellweger R, Wichmann MW, Ayala A, Stein S, DeMaso CM, Chaudry IH. Females in proestrus state maintain splenic immune functions and tolerate sepsis better than males. Crit Care Med 1997; 25:106-110. 116. Wichmann MW, Zellweger R, DeMaso CM, Ayala A, Chaudry IH. Mechanism if immunosuppression in males following trauma-hemorrhage: Critical role of testosterone. Arch Surg Arch Surg 1996; 131:1186-1192. 117. Beutler B, Milsark IW, Cerami A. Passive immunization against cachectin/tumor necrosis factor protects mice from lethal effect of endotoxin. Science 1985; 229:869-871.

270

Cytokines and the Abdominal Surgeon

118. Nassifk X, Mathison JC, Wolfson E, Koziol JA, Ulevitch RJ, So M. Tumour necrosis factor alpha antibody protects against lethal meningococcaemia. Molecular Microbiology 1992; 6:591-597. 119. Sharma RJ, Macallan DC, Sedgwick P, Remick DG, Griffin GE. Kinetics of endotoxin-induced acute-phase protein gene expression and its modulation by TNF∀ monoclonal antibody. Am J Physiol 1992; 262:R786-R793. 120. Schade UF. Pentoxifylline increases survival in murine endotoxin shock and decreases formation of tumor necrosis factor. Circ Shock 1990; 31:171-181. 121. Mengozzi M, Fantuzzi G, Faggioni R, Marchant A, DeMaria R, Orencole S, Clark BD, Sironi M, Testi R, Ghezzi P. Chlorpromazine specifically inhibits peripheral and brain TNF production, and up-regulates IL-10 production, in mice. Immunology 1994; 82:207-210. 122. Howard M, Muchamuel T, Andrade S, Menon S. Interleukin 10 protects mice from lethal endotoxemia. J Exp Med 1993; 177:1205-1208. 123. Rogy MA, Auffenberg T, Espat NJ, Philip R, Remick D, Wollenberg GK, Copeland III EM, Moldawer LL. Human tumor necrosis factor receptor (p55) and interleukin 10 gene transfer in the mouse reduces mortality to letal endotoxemia and also attenuates local inflammatory responses. J Exp Med 1995; 181:2289-2293. 124. Fletcher DS, Agarwal L, Chapman KT, Chin J, Egger LA, Limjuco G, Luell S, MacIntyre DE, Peterson EP, Thornberry NA. A synthetic inhibitor of interleukin-1 beta converting enzyme prevents endotoxin-induced interleukin-1 # production in vitro and in vivo. Interferon & Cytokine Research 1995; 15:243-248. 125. Luheshi G, Miller AJ, Brouwer S, Dascombe MJ, Rothwell NJ, Hopkins SJ. Interleukin-1 receptor antagonist inhibits endotoxin fever and systemic interleukin6 induction in the rate. Am J Physiol 1996; 270:E91-E95. 126. McNamara MJ, Norton JA, Nauta RJ, Alexander HR. Interleukin-1 receptor antibody (IL-1rab) protection and treatment against letal endotoxemia in mice. Surgical Research 54:316-321, 1993. 127. Acton RD, Dahlberg PS, Uknis ME, Klaerner HG, Fink GS, Norman JG, Dunn DL. Differential sensitivity to Escherichia coli infection in mice lacking tumor necrosis factor p55 or interluekin-1 p80 receptors. Arch Surg 1996; 131:1216-1221. 128. Dunne JR, Dunkin BJ, Nelson S, White JC. Effects of granulocyte colony stimulating factor in a nonneutropenic rodent model of Escherichia coli peritonitis. Surgical Research 1996; 61:348-354. 129. Fraker DL, Stovroff MC, Merino MJ, Norton JA. Tolerance to tumor necrosis factor in rats and the relationship to endotoxin tolerance and toxicity. J Exp Med 1988; 168:95-105. 130. Cross AS, Sadoff JC, Kelly N, Bernton E, Gemski P. Pretreatment with recombinant murine tumor necrosis factor ∀/cachectin and murine interleukin-1∀ protects mice from lethal bacterial infection. J Exp Med 1989; 169:2021-2027. 131. Rosman C, Westerveld GJ, van Oeveren W, Kooi K, Bleichrodt RP. Effect of intraperitoneal antimicrobials on the concentration of bacteria, endotoxin, and tumor necrosis factor in abdominal fluid and plasma in rats. Eur Surg Res 1996; 28:351-360. 132. Chang M, Williams A, Ishizawa L, Knoppel A, van de Ven C, Cairo MS. Endogenous interleukin-11 (IL-11) experession is increased and prophylactic use of exogenous IL-11 enhances platelet recovery and improves survival during thrombocytopenia associated with experimental group B streptococcal sepsis in neonatal rats. Blood Cells, Molecules, & Diseases 1996; 22:57-67. 133. Wheeler JG, Givner LB. Therapeutic use of recombinant human granulocyte-macrophage colony-stimulating factor in neonatal rats with type III group B streptococcal sepsis. J Infect Dis 1992; 165:938-941.

Anticytokine Strategies in Peritonitis

271

134. Austin OM, Redmond HP, Watson WG, Cunney RJ, Grace PA, Bouchier-Hayes D. The beneficial effects of immunostimulation in posttraumatic sepsis. Surgical Research 1995; 59:446-449. 135. Weyand C, Goronzy J, Fathman CG, O’Hanley P. Administration in vivo of recombinant interleukin 2 protects mice against septic death. J Clin Invest 1987; 79:1756-1763. 136. Goronzy J, Weyand C, Quan J, Fathman CG, O’Hanley P. Enhanced cell-mediated protection against fatal Escherichia coli septicemia induced by treatment with recombinant IL-2. J Immunol 1989; 142:1134-1138. 137. Ishizaka S, Kuriyama S, Tsujii T. In vivo depletion of macrophages by desulfated lcarrageenan in mice. J Immunol Methods 1989; 124:17-24. 138. Naito M, Nagai H, Kawano S, Umezu H, Zhu H, Moriyama H, Yamamoto T, Takatsuka H, Takei Y. Liposome-encapsulated dichloromethylene disphosphonate induces macrophage apoptosis in vivo and in vitro. J Leukocyte Biol 1996; 60:337-344. 139. Fryer J, Grant D, Jiang J, Metrakos P, Ozcay N, Ford C, Garcia B, Behme R, Zhong R. Influence of macrophage depletion on bacterial translocation and rejection in small bowel transplantation. Transplantation 1996; 62:553-559.

The Antibiotic-Induced Cytokine Response

273

CHAPTER 21

The Antibiotic-Induced Cytokine Response Timothy G. Canty Jr., Edward Boyle Jr., and E. Patchen Dellinger

Introduction

S

uccessful treatment of intra-abdominal sepsis requires a combination of surgical interventions to irradicate focal infection and potent broad spectrum antibiotics to treat systemic bacteremia. The timely utilization of both these therapies decreases mortality and improves outcome. Even with advances in life-support and surgical technique, however, greater than 100,000 people die each year as a result of gram negative sepsis. With little change in the mortality of established septic shock in the past 20 years, investigators continue to search for new therapeutic modalities as well as critically examine existing therapies.

Antibiotics, Bacterial Modulins, and Cytokines Antibiotics act to inhibit or kill growth of microorganisms, and for this reason are the cornerstone in the treatment of systemic infection and sepsis. Recently, however, antibiotics have been shown to both directly and indirectly modulate, in desirable and undesirable ways, the host immune response to infection (see Fig. 21.1). Antibiotics themselves have been shown to directly inhibit components of the host immune response including chemotaxis, phagocytosis, and lymphocyte transformation. In addition, some antibiotics have a direct effect on the regulation of cytokine synthesis. It is the indirect inflammatory effect of antibiotics that has been the focus of investigation over the last decade. Antibiotics kill bacteria and, in doing so, cause the release of toxic, pro-inflammatory bacterial byproducts, most notably endotoxin. Endotoxin incites intravascular inflammation through the generation of cytokines. To characterize these proinflammatory bacterial components such as endotoxin, Henderson et al coined the term bacterial “modulin” to describe a class of bacterial molecules that modulate cell activity with pathologic consequences. This chapter will examine the effects of antibiotics on the immunologic and inflammatory host response in the face of infection. More specifically it will focus on the mechanisms by which antibiotics induce cytokine release. Utilizing endotoxin as a paradigm, we will review data from existing in vitro studies, animal studies, and human studies to address several questions: Do antibiotics promote the release of proinflammatory bacterial byproducts and do those byproducts in turn induce the release of inflammatory Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

274

Cytokines and the Abdominal Surgeon

Fig. 21.1. Antibiotic modulation of the immunologic inflammatory host response to infection.

cytokines? Do different antibiotics release more or less cytokines? and Does antibiotic induced release of bacterial modulins have clinical implications for the practicing surgeon?

Antibiotic Induced Release of Bacterial Modulins: History The theory that the treatment of infections with antimicrobial agents might worsen a patient’s condition due to the release of toxic microbial breakdown products is greater than a century old. In 1895, Jarisch and Herximeimer reported a syndrome of marked pyrexia, hypotension, and rigors following the application of mercurial antibiotics in the treatment of syphilis. In 1949, Patel reported fatal vasomotor collapse following the loading dose of chloramphenicol in the treatment of typhoid fever. He and others went on to postulate that this phenomena was due in part to antibiotic induced release of a massive and fatal dose of “endotoxin”. In 1985 the Lancet published an anonymous editorial letter stating the following: “There is a long standing and persistently disturbing suspicion, fueled more by anecdote than by fact, that antibiotics might aggravate rather than alleviate the development of shock in the bacteremic patient... Antibiotics, especially those that act on the cell wall do have the potential to exacerbate endotoxin shock.” As the molecular mechanisms of the septic response became clear a potential explanation for these anecdotal observations began to emerge. Antibiotics kill bacteria by attacking and breaking down their outer membrane. Bacterial byproducts from the outer membrane, most notably endotoxin, were shown to reproduce the diffuse intravascular inflammation seen in sepsis. Thus, a hypothesis developed that the bactericidal action of antibiotics may result in a “bolus” of toxic bacterial components into the bloodstream, exacerbating and fueling the inflammatory response. The high mortality seen in septic patients in the face of appropriate antibiotic therapy caused clinicians to question and study the role of antibiotics in the treatment of sepsis.

The Antibiotic-Induced Cytokine Response

275

Fig. 21.2. Effect of endotoxin on various cells.

Background Advances in cellular and molecular biology have demonstrated that bacterial byproducts, most notably endotoxin, induce the transcription and translation of a variety of pro-inflammatory (TNF, IL-1 IL-6, and IL-8), procoagulant (Tissue Factor), and vasoactive substances (endothelin and nitric oxide) from a variety of cells (see Fig. 21.2). These mediators, in turn induce the synthesis of chemokines (C5a, IL-8, and LTE4) and adhesion molecules (E-selectin, ICAM, VCAM) which propagate the inflammatory response. This complex and redundant host response produces the recalcitrant hypotension, the consumptive coagulopathy, and the end organ damage which clinically characterizes gram negative septic shock.

Antibiotic Induced Release of Bacterial Modulins: Endotoxin The ability to isolate and purify endotoxin, a substance which alone can reproduce the pathophysiology of sepsis, provided researchers a tool to examine the role of antibiotic induced inflammation. Endotoxin, also referred to as lipopolysaccharide (LPS), is a amphiphilic macromolecule located in the outer membrane of all gram-negative bacteria composed of three distinct regions; the variable antigenic O side chain, the highly conserved central core called the R region, and the toxic, immuno-stimulatory portion of the molecule known as the Lipid A moiety. While the exact mechanisms involved in the LPS induced host response are still not fully known, it is generally believed that LPS must be detached from the outer membrane of bacteria to exerts its full effect. Liberation of LPS does occur with antibiotic induced cell death, however, free LPS is also generated during normal log phase growth of gram negative bacteria and through complement mediated lysis. Free intravascular LPS forms a complex with an LPS-binding protein (LBP). The interaction of LPS with LBP increases the LPS potency 1000-fold. Next, LPS/LBP

276

Cytokines and the Abdominal Surgeon

complexes bind to a surface glycoprotein CD14 that is usually found on monocytes and macrophages. This CD14 interaction initiates a signal transduction pathway that results in the rapid production of potent inflammatory cytokines which initiate and propagate intravascular inflammation. Endotoxin acts on a variety of cells including macrophages, monocytes, and endothelial cells. The LPS-host interaction, however, is complex and variable. This is due, in part, to several host substances that can modulate the bioavailability and potency of LPS. One such molecule is high density lipoprotein, HDL, that interacts with the Lipid A moiety and is thought to attenuate LPS effects. Another protein, called bactericidal permeability-increasing protein (BPI) is secreted by activated PMN’s and has been shown to strongly attenuate LPS effects. Thus, both natural bacterial proliferation and antibiotic induced cell death result in endotoxin release. Host proteins mediate the bio-availability and potency of intravascular endotoxin. Different bacteria have different growth rates and respond differently to different antibiotics. It is easy to appreciate the complexity and variability inherent in the study of antibiotic induced LPS release.

Antibiotics and Endotoxin (In Vitro) Initial in vitro studies to investigate the phenomena of antibiotic induced endotoxin release involved exposing cultures of gram negative bacteria to various bactericidal antibiotics. Antibiotics caused the bacterial count to fall rapidly, however, the endotoxin level in the surrounding media significantly increased with bacterial death. One study by Dofferhof et al showed that antibiotic exposure resulted in greater than 100-fold increase in measured endotoxin compared with control cultures. To assay the biologic activity of antibiotic induced endotoxin, investigators utilized filtrates of bacterial cultures exposed to antibiotics. The filtrates containing free endotoxin were then placed on human monocyte or whole blood preps and the cytokine response was measured. Simon and Arditi showed that antibiotic exposed bacterial filtrates generated a marked TNF-∀ response in human white blood cells. This cytokine response could be blocked by monclonal antibodies against endotoxin as well as the LPS binding agent, polymixin B, demonstrating that cytokine induction was specific to endotoxin activity.

Antibiotics and Endotoxin (Animal Studies) In parallel to in vitro findings, antibiotic therapy in rat, rabbit, and pig E. coli sepsis models results in a rapid serum increase in endotoxin coupled with a decreasing bacterial count. Three additional animal studies documented no difference in endotoxin levels between control and antibiotic treated animals. These studies did, however, show that the endotoxin/bacteria count ratio increased in the antibiotic treated animals, indicating a liberation of free endotoxin into the circulation. To assay the cytokine response generated by antibiotic induced endotoxin release in vivo, investigators documented temporal increases of TNF-∀, IL-1, and IL-6 following initial antibiotic administration. While the observations that antibiotics cause the intravascular release of endotoxin during bacterial cell death in vitro and in vivo are provocative, the real question is whether or not this antibiotic-induced endotoxin and subsequent cytokine bolus results in clinical deterioration or adverse outcome. Andersen et al showed that benzylpenicillin resulted in significant endotoxin release and demonstrated a higher mortality in a murine model of Neisseria meningitidis infection. Rokke et al demonstrated significant hemodynamic instability following administration of antibiotics

The Antibiotic-Induced Cytokine Response

277

in a porcine sepsis model associated with a rise in both LPS and TNF-∀. In a model of meningitis Tauber and colleagues showed an increase in brain edema following antibiotics. Overall, data from animal studies are in agreement with data from in vitro work; endotoxin is released with antibiotic therapy from dying bacteria which in turn generates a pro-inflammatory cytokine response. In addition, there are a few animal studies which demonstrate that endotoxin release is paralleled by a deterioration in the markers of disease severity. This data, however, needs to be taken in context. The pulse of endotoxin and cytokines measured following antibiotic administration occurs over several hours. Followed over a longer period of time the cytokine and endotoxin levels in antibiotic treated animals fall sharply to below pretreatment levels. In contrast, untreated septic animals have levels of cytokines and endotoxin significantly higher than the levels seen following antibiotic therapy.While there is no question that antibiotics improve survival in the face of sepsis, the clinical consequences of an antibiotic induced endotoxin “bolus” remain less clear.

Antibiotics and Endotoxin (Clinical Studies) There are a limited number of clinical studies addressing the phenomena of antibiotic induced endotoxin release. In studies by Dofferhof and Shenep endotoxin levels were serially measured in patients with suspected gram negative sepsis before and after appropriate antibiotic therapy. Shenep showed that 7 of 11 patients with positive blood cultures demonstrated a 2- to 50-fold increase in free endotoxin following antibiotic therapy, while Dofferhof reported a 2- to 15-fold increase in endotoxin levels in 3 out of 10 bacteremic patients given antibiotics. In those three patients, higher endotoxin levels were associated with a decrease in blood pressure and a rise in serum lactate. Meningitis studies in the pediatric population support the phenomena of antibiotic induced endotoxin release. Mustafa compared intrathecal antibiotics to IV antibiotics in children with documented meningitis. Repeat CSF analysis showed a marked increase in CSF endotoxin and IL-1 in the intrathecally treated children compared with those treated using the IV route. Children with Haemophilus influenzae meningitis treated with ceftriaxone were also studied by Arditi et al. Eight patients in the study underwent a second lumbar puncture. The CSF in all eight of the children showed decreased bacterial count and a decrease in total endotoxin. There was, however, a definite shift from cell bound to free endotoxin accompanied by an increase in lactate and LDH with a decrease in glucose, suggesting an increased inflammatory response. Hurley studied antibiotic induced endotoxin release in chronically bactiuric patients. Patients who received antibiotics demonstrated a significant decrease in bacterial count along with an increase in total urinary endotoxin level. In nontreated patients endotoxin and bacterial levels remained unchanged. Maximal bacterial lysis preceded the maximal increase in endotoxin by 1 hour.

Do Antibiotics Differ in Potential for Endotoxin Release? In vitro studies clearly show that antibiotic induced bacterial cell death results in the release of biologically active endotoxin. There is also evidence that this occurs in vivo. There is, however, no question that antibiotics reduce morbidity and mortality in the face of infection. The question that faces the clinician treating bacteremia and sepsis is, therefore, whether equally effective antibiotics differ in the amounts of endotoxin release and whether that difference has an impact on clinical outcome.

278

Cytokines and the Abdominal Surgeon

The amount of endotoxin released from bacteria following treatment with antibiotics is not only a function of the concentration and effectiveness of a given antibiotic, but also a function of how the antibiotic kills the bacteria. This is best illustrated by examining the antibiotics that are classified as cell wall inhibitors, such as the penicillins, the cephalosporins, and the carbapenems. Cell wall inhibitors, bind to and inhibit the enzymatic penicillin binding proteins (PBP) in the bacterial plasma membrane. This results in cell wall disruption and bacterial death. There are several different isoforms of PBPs. Cell wall antibiotics have variable affinities for the different PBP isoforms. Inhibition of different PBP isoforms results in different mechanisms of bacterial death and different amounts of endotoxin release. For example, antibiotics selective for PBP 1 proteins such as cephaloridine and cefsoludin, cause rapid bacterial lysis and death through cell wall degradation. Antibiotics which have an affinity for PBP 2 proteins, such as imipenem and meropenem, convert bacilli to round spheroplasts that result in the loss of viability, but do not cause rapid degradation. It is these antibiotics which have consistently been shown to cause the least amount of endotoxin release in vitro. Antibiotics which select for PBP 3 proteins such as aztreonam, mezlocillin, cephalexin, cefotaxime, and ceftazidime, inhibit bacterial septation and induce filamentous changes causing death without lysis. These nonviable filaments appear to continue to release endotoxin over time. The PBP 3 antibiotics, in general, cause more endotoxin release in vitro. Extensive review of available data can be found in the reviews by Hurley and Prins. Other antibiotics which kill bacteria through mechanisms that do not involve cell wall synthesis also demonstrate variable endotoxin releasing properties. Review of the literature reveals several trends. The aminoglycosides, like the PBP2 antibiotics, induce small amounts of endotoxin release. In the case of aminoglycosides, this effect may be due to the ability of antibiotics, such as gentamicin and amikacin to bind and neutralize endotoxin. An overview table of the variable endotoxin releasing properties of common antibiotics is shown in Table 21.1. The biological significance of the varying endotoxin-releasing properties of different antibiotics is less clear. Shenep, using a model of rabbit E. coli sepsis, demonstrated that moxalactam induced more endotoxin release than gentamicin, but there was no difference in survival. Tauber studied E. coli meningitis in rabbits and also reported significantly more endotoxin in the CSF using cefotaxime compared to chloramphenicol. He also observed less brain edema in the chloramphenicol treated animals. In contrast, Dofferhoff showed no difference in endotoxin release between aztreonam, ceftazadime, and imipenem in a rat sepsis model. Studies investigating the clinical implications of differential endotoxin release by antibiotics are rare. Mock et al examined critically injured trauma patients admitted to the ICU. In a large retrospective review they looked to see if PBP3 antibiotics were detrimental to patients as compared with PBP2 antibiotics. They did not measure endotoxin or cytokine levels, but instead attempted to correlate the antibiotic regimen to outcome. While the study is limited, they do show some evidence that ICU patients that received PBP2 antibiotics did better than those who received PBP3 antibiotics. In this report antibiotics were not documented and were simply assumed to be present if antibiotics were administered. Within the last year Holzheimer examined endotoxin release and IL-6 levels in surgical ICU patients. Taking serial measurements of endotoxin and IL-6 they showed reduced levels of both with imipenem compared to cefotaxime and ceftriaxone. In their study quinolones demonstrated intermediate levels of endotoxin and IL-6.

The Antibiotic-Induced Cytokine Response

279

Table 21.1. Antibiotic induced endotoxin release** AntibioticClass

Quinolones

PBP 3 Antibiotics

Aztreonam Ciprofloxacin Ofloxacin Ceftazidime Cefotaxime Mezlocillin Ticarcillin Cefoperazone Ampicillin Chloramphenicol Amoxacillin

Aminoglycosides

PBP 2 Antibiotics

EndotoxinRelease

Tetracycline Amikacin Tobramycin Gentamicin

+++

++

+

Imipenem Mero penem

** Data summary of Hurley J et al. Clin Infect Dis 1992; 15:840-854. Prins J et al. Antimicrob Agents and Chemother 1994; 38:1211-1218.

Cytokine Stimulating Microbial Components Other Than LPS As discussed above, most research dedicated to examining the potential deleterious effect of antibiotic therapy has focused on antibiotic induced endotoxin release. This chapter has reviewed in vitro, in vivo, and clinical data on antibiotic induced endotoxin release to demonstrate a concept. The ability of endotoxin to reproduce the biological and clinical features of sepsis has been instrumental in the development of cytokine biology. Because a single substance, such as endotoxin, can elicit a sepsis host response does not, however, preclude other bacterial components from having biological significance. Until recently the ease of isolation of LPS and the epidemiology of sepsis has focused efforts on the molecular mechanism of gram negative sepsis. There is, however, recent evidence that other components of gram negative bacteria, gram positive bacteria, and fungal elements released during cell death can generate an inflammatory response. With a resurgence in gram positive and fungal infections investigators have broadened their scope of research to examine other potential microbial elements that may contribute to the host inflammatory response. Table 21.2., which provides an overview of the bacterial products that have been shown to elicit a cytokine response, illustrates these efforts.

280

Cytokines and the Abdominal Surgeon

Table 21.2. Cytokine-inducing bacterial components other than endotoxin BacterialComponent

Porins LAP Fimbrial proteins Surface-associated proteins Protein A Heat shock proteins Lipoproteins Glycoproteins Lipids Cell surface polysaccharides Peptidoglycan Peptidoglycan fragments Teichoic acids LAM Exotoxins Superantigens

Lowestconc.incducing cytokinesynthesis(ng/ml) 10 1 10 0.01 1,000 10 0.5 0.05 0.1 2,500 100 50 1,000 100 0.03 10

Mechanism**

NL NL NL NL ? ? ? ? ? NL/L ?/L NL NL L NL NL

NL=mechanism of cytokine stimulation is clearly different from that of endotoxin; L=evidence suggests that component stimulates cytokine synthesis by an endotoxin-like mechanism (i.e., CD14 dependent); NL/L=some reports demonstrate that component acts in a CD-14 independent manner; ?=unknown mechanism of cytokine induction. Reproduced with permission from: Henderson B et al. Microbiol Reviews, 1996; 60:316-341.

Today 30-50% of cultures from septic patients reveal gram positive bacteria as the source. While LPS is the central virulence factor for gram negative infections, study of the gram positive cell wall has also revealed bioactive materials. The major components of gram positive bacterial walls are highly cross-linked peptidoglycan (PG) interspersed to a variable extent with lipoteichoic acid (LTA) polymers. These gram positive components can now be isolated and have been shown to stimulate in vitro production of TNF, IL-1, IL-6, and IL-8 in human monocytes and macrophages. The central difference between these gram positive bacterial components and LPS is potency. While anywhere from 10-100 microgram per milliliter doses of LTA or PG are required for cytokine production, LPS elicits cellular responses in the picogram to nanogram per milliliter range. How and to what extent antibiotics against gram positive bacteria induce the release of bioactive products is still under investigation. There is some evidence that suggest that gram positive components act through the same CD14 signaling pathways that LPS utilizes. This suggests that foreign material may activate the immune inflammatory response through a common pathway which may be a potential therapeutic target. It may be more likely, however, that gram positive and fungal elements stimulate a host response through unique mechanisms which will require unique therapies. As the clinical significance of gram positive infections, fungal infections, and mixed infections grows, the question of whether antibiotics cause release of other toxic cellular components may have clinical implications.

The Antibiotic-Induced Cytokine Response

281

Cytokine Synthesis and Antibiotics While little is published in the area, the direct immunologic and inflammatory effects of antibiotics provides another mechanism by which antibiotics may influence the cytokine response in septic patients. Bailey et al demonstrated that high concentrations of quinolones decrease endotoxin induced TNF production. Lower concentrations seem to stimulate more IL-1 and IL-2 production from human monocytes. Takizawa reported that erythromycin supressed IL-6 production in bronchial epithelial cells. Just recently Galley showed that ciprofloxacin supressed IL-1 and stimulation of human endothelial cells, suggesting an anti-inflammatory action of this common antibiotic. Direct effects of antibiotics on the immune response have also been documented. There are reports showing that gentamicin inhibits chemotaxis and sulfa drugs supress lymphocyte transformation. While potentially clinically relevant, the immunologic effects of most antibiotics have not been studied.

Conclusion Antibiotics and surgical intervention are currently the most effective therapies in the battle against microbial infection. There are no data to support not using antibiotics in cases of life threatening systemic infections. Eradication of the microbial threat is the central tenet of antimicrobial therapy, not only in intra-abdominal sepsis, but also in any life threatening infection. Review of the current literature on the subject of antibiotic induced release of cytokine stimulating bacterial products, however, reveals evidence that this is a real phenomena. Bacterial and fungal breakdown products generated in antibiotic induced cell death are able to initiate the inflammatory cytokine cascade thought to be responsible for the pathophysiology seen in septic shock. In the case of endotoxin, there are some data, in animal and human studies, to show that antibiotic induced release may have adverse clinical consequences. There are, however, also firm data to suggest that withholding antibiotics in systemic infection results in profound shock and death. Thus, antibiotics are the only tool currently available to control and eradicate disseminated infection, even though the use of these may, in some situations, cause temporary exacerbation of intravascular inflammation. Reconciliation of this dichotomy may lie in the fact that different antibiotics work through different mechanisms. These differences can result in variability of endotoxin release. As shown above, certain classes of antibiotics result in more endotoxin release than others in the treatment of a given infection. While in most patients with systemic infection the “bolus” of endotoxin release upon administration of antibiotics may have a transitory effect, there are potentially patients in which any further inflammatory insult could result in refractory deterioration. Severe trauma patients, immunosupressed patients, nutritionally depleted patients, or elderly patients may experience a second hit phenomenon from antibiotic induced endotoxin that pushes them over the edge. In these patients it may be appropriate to consider the propensity for endotoxin release by a given antibiotic as one criteria in the selection process. Regardless of this, the antimicrobial susceptibility spectrum of the antibiotics chosen will continue to be the most important factor to consider. Antibiotics and surgery are still and most likely will continue to be the arsenal abdominal surgeons use to treat intra-abdominal sepsis. It will take prospective multicenter studies to answer the question as to whether a certain antibiotic should be employed over another antibiotic because of its endotoxin releasing properties. As the management of sepsis enters the molecular age with various cytokine

282

Cytokines and the Abdominal Surgeon

modulators and immunotherapies, there will be new therapeutic modalities available to decrease the host inflammatory response. It is logical to consider the use of these new anti-inflammatory drugs in combination with traditional antibiotics to simultaneously eradicate bacterial infection and inhibit the inflammatory response generated from inflammatory bacterial products. Clearly the current mortality seen with intra-abdominal sepsis is unacceptable and any means of improvement deserves careful analysis.

References 1. Kreger B, Craven D, McCabe W. Gram negative bacteremia. IV. Re-evaluation of clinical features and treatment in 612 patients. Am J Med 1980; 68:344-355. 2. Nogare D. Southwestern internal medicine conference: Septic shock. Am J Med Sci 1991; 302:50-65. 3. Young LS, Martin WJ, Meyer RD. Gram-negative rod bacteremia: Microbiologiimmunologic, and therapeutic considerations. Ann Intem Med 1977; 86:456. 4. Hauser W, Remington J. Effect of antibiotics on the immune response. Am J Med 1982; 72:711-716. 5. Galley H, Nelson S, Dubbels A. Effect of ciprofloxacin on the accumulation of interleukin 6, interleukin 8, nad nitrite from a human endothelial cell model of sepsis. Crit Care Med 1997; 25:1392-1395. 6. Cohen J, McConnell J. Antibiotic induced endotoxin release. Lancet 1985; 2:1069-1070. 7. Prins J, van Deventer S, Kuijper E, Speelman P. Clinical relevance of antibioticinduced endotoxin release. Antimicrob Agents Chemother 1994; 38:1211-1218. 8. Mustafa M, Mertsola J, Ramilo O et al. Increased endotoxin and interleukin-IB concentrations in cerebrospinal fluid of infants with coliform menigitis and ventriculitis with intraventricular gentamicin therapy. J Infect Dis 1989; 160:891-895. 9. Shenep J, Flynn P, Barrett G et al. Serial quantitation of endotoxemia and bacteremia during therapy for gram-negative bacterial sepsis. J Infect Dis 1988; 157:565-568. 10. Hurley J. Antibiotic-induced release of endotoxin: A reappraisal. Clin Infect Dis 1992; 15:840-854. 11. Henderson B, Poole S, Wilson M. Bacterial Modulins: A novel class of virulence factors which cause host tissue pathology by inducing cytoldne synthesis. Microbiol Rev 1996; 60:316-341. 12. Gelfand JA, Elin RJ, Berry FW. Endotoxemia associated with the Jarisch-Herheimer. N Engl J Med 1976; 295:211-13. 13. Jarisch A. Therapeufische veruche bei Syphilis. Wein Med Wochenschr 1895; 45:721-24. 14. Patel JC, Banker DD, Modi CJ. Chloramphenicol in typhoid fever. BMJ 1949; 2:908-909. 15. Anonymous. A nasty shock from antibiotics? Lancet 1985; 2:594. 16. Fong Y, Moldawer L, Marano M et al. Endotoxemia elicits increased circulating B2-INF/IL-6 in man. J Immunol 1989; 142:2321-2324. 17. Van Deventer S, Buller H, ten Cate J et al. Endotoxaemia: An early predictor of septicemia in febrile patients. Lancet 1988; 1:605-609. 18. Van Deventer S, Buller H, ten Cate J et al. Experimental endotoxinen-in humans: Analysis of cytokine release and coagualtion, fibrinolytic, and complement pathways. Blood 1990; 76:2520-2526. 19. Ulevitch RJ, Tobias PS. Recognition of endotoxin by cells leading to transmembrane signaling. Curr Opin Immunol 1994; 6(l):125-30.

The Antibiotic-Induced Cytokine Response

283

20. Martich G, Danner R, Ceska M, Suffrendini A. Detection on interleukin 8 and tumor necrosis factor in non-nal humans after intravenous endotoxin: The effect of antiinflammatory agents. J Exp Med 1991; 173:1021-1024. 21. Schletter J, Holger H, Ulmer A, Rietschel E. Molecular mechanisms of endotoxin activity. Arch Microbiol 1995; 164:383-389. 22. Burrell R. Human response to bacterial endotoxin. Circulatory Shock 1994; 43:137-153. 23. Schletter J, Holger H, Ulmer A, ET R. Molecular mechanisms of endotoxin activity. Arch Microbiol 1995; 164:383-389. 24. Hurley J. Antibiotic-induced release of endotoxin. Drug Safety 1995; 12:183-195. 25. Cohen J, MsConnell J. Release of endotoxin from bacteria exposed to ciprofloxacin and its prevention with polymixin B. Eur J Clin Microbiol 1986; 5:13-17 26. Dofferhoff A, Nijland J, de Vries-Hospers H et al. Effects of different types and combinations of antimicrobial agents on endotoxin release from gram negative bacteria: An in vitro and in vivo study. Scand J Infect Dis 1991; 23:745-754. 27. Mellado M, Rodriguez-Contreras R, Mariscal A et al. Effect of penicillin and chloramphenicol on the growth and endotoxin release by N. meningitides. Epidemiol Infect 1991; 106:283-288. 28. Bingen E, Gouty V, HB et al. Bactericidal activity of beta-lactams against Haemophilus influenzae: effect on endotoxin release. J Antimicrobiol Hemotherap 1992; 30:165-172. 29. Van Den Berg C, De Neeling A, CS S et al. Delayed antibiotic-induced lysis of Escherichia coli in vitro is correlated with endotoxemia. Scand J Infect Dis 1992; 24:619-627. 30. Simon D, Koenig G, Trenholme G. Differences in release of tumor necrosis factor from THP-1 cellls stimulated by filtrates of antibiotic-killed Escherichia coli. J Infect Dis 1991; 164:800-802. 31. Arditi M, Zhou J, Dorio R et al. Endotoxin-mediated endothelial cell injury and activation: role of soluble CD 14. Infect Immun 1993; 61:3149-56. 32. Shenep J, Bartton R, Mogan K. Role of antibiotic class in the rate of liberation of endotoxin during therapy for experimental gam-negative bacterial sepsi. J Infect Dis 1985; 151:1012-1018. 33. Shenep J, Morgan K. Kinetics of endotoxin release during antibiotic therapy for experimental gram-negative sepsis. J Infect Dis 1984; 150:380-388. 34. Rokke 0, Revhaug A, Osterud B, Giercksky K. Increased plasma levels of endotoxin and corresponding changes in circulatory performance in a porcine sepsis model: the effect of antibiotic administration. Prog Clin Biolo Res 1988; 272:247-262. 35. Almdahl S, Ostrud B. Effect of antibiotics on gram-negative sepsis in the rat. Acta Chir Scand 1987; 153:283-286. 36. Dofferhoff A, Bartels H, de Vrioes-Hosper Nijland H et al The release of endotoxin and interleukin-6 during treatment with different antibiotics or in combination with taurolidine of rats with experimental gram-negative sepsis. Nucroecol Ther 1994; 28:399-406. 37. Natanson C, Danner R, Reiley J et al. Antibiotics versus cardiovascular support in canine model of human septic shock. Am J Physiol 1990;59:Hl4440-Hl447. 38. Friedland 1, Jafari H, Ehrett S et al. Comparison of endotoxin release by different antimicrobial agents and the effect on inflammation in experimental Escherichia coli meningitis. J Infect Dis 1993; 168:657-662. 39. Healy D, VerstBrasch C, Clendening C et al. Influence of drug class and dose size on antibiotic-induced endotoxin/IL-6 release and impact on efficacy of anti-endotoxin antibody. J Endotox Reser 1996; 3:219-227.

284

Cytokines and the Abdominal Surgeon

40. Andersen B, Solberg S. Effect of benzylpenicillin in mice infected with endotoxin liberating or nonendotoxin liberating strains of Neisseria meningitides. Scand J Infect Dis 1984; 16:257-266. 41. Tauber M, Shibl A, Hackbarth C et al. Antibioitic therapy, endotoxin concentration in cerebrospinal fluid, and brain edema in experimental Escherichia coh meningitis in rabbits. J Infect Dis 1987; 156:456-462. 42. Mertsola J, Ramilo 0, Mustafa M et al. Release of endotoxin after antibiotic treatment of gram-negative meningitis. Pediatr J Infect Dis 1989; 8:904-906. 43. Tomasz A. Penicillin-binding proteins and the antibacterial effectiveness of B-lactam antibiotics. Rev Infect Dis 1986; 8(supple 3):S262-S278. 44. Spratt BG, Cromie KD. Penicillin-binding proteins of gam-negative bacteria. Rev Infect Dis 1988; 10:699-711. 45. Goto H, Nakamura S. Liberation of endotoxin from Escherichia coli by addition of antibiotics. Jpn J Exp Med 1980; 50:35-43. 46. Tuomanen E, Gilbert K, Tomasz A. Modulation of bacteriolysis by cooperative effects of penicillin-binding proteins la and 3 in Escherchia coli. Antimicrob Agents Chemotherap 1986; 30:659-663. 47. Neu HC. Penicillin-binding proteins and role of aindinocillin in causing bacterial cell death. Am J Med 1983; 75(Suppl 2A):9-20. 48. Mock C, Jurkovich G, Dries D, Maier R. Clinical significance of antibiotic endotoxin-releasing properties in trauma patients. Arch Surg 1995; 130:1234-1243. 49. Holzheimer R, Hirte J, Reith B et al. Different endotoxin release and IL-6 plasma levels after antibiotic administration in surgical intesive care patients. J Endotox Reser 1996; 3:261-267. 50. Bailey S, Fay M, Roche Y. Effects of quinolones on tumor necrosis factor production by human monocytes. Int J Immunopharmacol 1990; 12:31-36. 51. Takizawa H, Desaki M, Ohtoshi T. Erythromycin surpresses interleukin 6 expression by human bronchial cells: A potential mechanism of its anti-inflammatory action. Biochem Biophys Res Commun 1995; 210:781-786. 52. Montravers P, Gauzit R, Muller C. Emergence of antibiotic-resistant bacteria in cases of peritonitis after intraabdominal surgery affects the efficacy of emperical antimicrobial therapy. Clin Infect Dis 1996; 23:486-494. 53. Mosdell D, Morris D, Voltura A. Antibiofic treatment for surgical peritonitis. Ann Surg 1991; 214:543-552. 54. Leibovici L, Paul M, Poznanski O. Monotherapy versus beta-lactarnaminoglycoside combination treatment for Gram-negative bacteremia: A prospective observational study. Antimicrob Ag Chemother 1997; 41:1127-1133.

Cytokines and Pentoxifylline in Surgical Sepsis

285

CHAPTER 22

Cytokines and Pentoxifylline in Surgical Sepsis J. Schröder, K.H. Staubach, F. Stüber and P. Zabel

Introduction

S

econdary peritonitis and acute necrotizing pancreatitis are the most frequent intra-abdominal conditions which are still associated with a high mortality rate. In addition to surgical intervention a calculated antimicrobial initial treatment can contribute to improve the patient´s prognosis in peritonitis.1 In necrotizing pancreatitis the therapeutic strategy changed to an initial conservative antibiotic therapy followed by operative procedures in infective necrosis.2 Despite advances in operative, antibiotic and intensive care management of these patients, the prognosis of complicated peritonitis or necrotizing pancreatitis has not decreased in the last decade.3 Complicated intra-abdominal infections are characterized by an uncontrolled systemic response to an initially localized peritoneal or pancreatic process which has been termed as sepsis.4 In the complex endogenous mediator cascade of sepsis and subsequent multiple organ dysfunction, proinflammatory cytokines play a key role.5 Tumor necrosis factor alpha (TNF) is regarded as the central mediator and many attemps have been made to modulate the systemic cytokineinduced inflammation with release of TNF and other proinflammatory cytokines like interleukin (IL-1, IL-6 and IL-8). In clinical trials, however, none of the cytokine-blocking substances revealed clinical benefit as far as survival was concerned.6 Compounds with properties to reduce proinflammatory cytokine production and the potential of long-term therapy may be more promising treatment modalities in sepsis compared to cytokine-blocking monoclonal antibodies. Total blockade of TNF as central mediator has even led to increased mortality in human studies7 because TNF to some extent is essential for the endogenous defense response to infection.8 Pentoxifylline (POF) is known to oppose the proinflammatory effects of TNF with the potential of long-term therapy. Pentoxifylline is a methyl-xanthine derivate which has been used for the treatment of chronic occlusive arterial disease because of its rheological actions.9 POF is known to inhibit phosphodiesterase, thereby increasing cytoplasmatic cyclic adenosine monophosphate (cAMP) levels. Elevation of cAMP strongly inhibits TNF gene transcription10 associated with downregulation of TNF production in animal experiments and in humans challenged with endotoxin.11,12 In intra-abdominal sepsis in animals POF improved survival and reduced Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

286

Cytokines and the Abdominal Surgeon

Table 22.1. Severity of sepsis at enrollment and mortality*

APACHE II score MOD-score Septic shock (number of pts) 28-day mortality Hospital mortality

POF

Placebo

17.1±0.9 11.0±0.8 23/27 30% 41%

18.8±1.0 11.8±1.0 21/24 33% 54%

The values are mean + SEM. POF, pentoxifylline; MOD-score, multiple organ dysfunction score.23 * Differences between the study groups were not significant.

blood concentrations of TNF.13-15 Based on the beneficial effects in experimental endotoxic shock and intrabdominal infection, clinical trials with POF remained warranted. Short-term pilot studies of POF in humans failed to demonstrate any adverse effects with large doses in respect to pulmonary and hemodynamic parametes.16-18

Long-Term Administration of Pentoxifylline in Surgical Sepsis To evaluate the effect of high dose and long-term administration of pentoxifylline as adjunctive therapy in surgical sepsis a prospective, randomized, double-blind and placebo-controlled study was performed. Surgical intensive care patients were included if they fulfilled the criteria of severe sepsis or septic shock according to the definition of Bone et al.19 Sepsis was characterized by clinical evidence of infection; temperature of >38.5°C or <35.6°C; tachycardia of >90 beats/min; tachypnea of >20 breaths/min or mechanical ventilation and evidence of altered organ function including mental disorientation; oliguria (<0.5 mL/kg bw/h); hypoxemia (PaO2 <75 torr or PaO2/FiO2 <250); disseminated intravascular coagulation (platelets <100x109/L, decrease in platelets >50% or >20% prolongation in prothrombin time); cardiac index (CI) >4.0 L/min/m2 with systemic vascular resistance (SVR) <800 dyne x s/cm5; metabolic acidosis (pH <7.30) or elevated lactate levels. Septic shock was defined as a sustained decrease in systolic blood pressure to <90 mm Hg; or a decrease of 40 mm Hg from baseline in the absence of antihypertensive agents despite adequate fluid resuscitation. Patients received full intensive care management for sepsis including fluid resuscitation, vasopressors, ventilatory support, hemofiltration, antibiotics and surgical procedures as required. The entire management of the patients was performed by physicians who were not involved in the study. Patients were excluded for any of the following reasons: <18 years of age; suspected pregnancy; uncontrolled hemorrhage; cardiogenic shock; burns; severe, preexisting, parenchymal liver disease with clinically significant portal hypertension; former therapy with glucocorticoids, methylxanthine derivates, amrinone or nonsteroidal anti-inflammatory drugs; former irradiation or chemotherapy; participation in ongoing or previous clinical trials using anti-inflammatory agents. Organ transplant recipients and those patients with hematological malignancies as acute underlying condition were not enrolled.

Cytokines and Pentoxifylline in Surgical Sepsis

287

Fig. 22.1. Changes in mean PaO2/FiO2-ratio throughout the study period for patients with pentoxifylline (solid squares) or placebo (open squares). * p<0.05, t-test with Bonferroni's correction.

Fifty-one surgical patients were randomly assigned within 12 hours after diagnosis of sepsis to receive either a continuous intravenous infusion of 1 mg/kg body weight per hour (bw/h) of pentoxifylline (Rentyllin®, Rentschler, Germany), maximum 1800 mg per day, or normal saline as placebo over 28 days or until patients were discharged from ICU or died. The majority of patients with surgical sepsis had intraabdominal infections. Eighteen patients in the POF-group had peritonitis compared to twelve patients in the placebo group. Eight patients in the placebo group and four in the POF-group had necrotizing pancreatitis. Other causes of sepsis included severe soft tissue infection (2 patients), postoperative sepsis (4 pts), mediastinitis (2 pts) and ARDS (1 pt). The primary source of sepsis was documented by the investigators at enrollment. Causative microorganisms were identified from blood or body fluid and adequate antimicrobial therapy was determined by an infectious disease specialist not involved in the study. For both treatment groups combined, 22% had Gram-negative, 24% had Gram-positive and 37% had mixed Gram-positive and -negative bacteria found as causative microorganisms. Six patients had a combination with fungi and 3 patients had no documented bacteria or fungi. Microorganisms causing infections were equally distributed in the treatment groups. In this prospective clinical trial patient groups who were assigned to receive pentoxifylline or placebo besides standard therapy of sepsis were well matched in terms of severity. At enrollment nearly 90% of patients in both groups had septic shock (Table 22.1). No difference was found with regard to parameters for organ dysfunction, for multiple organ dysfunction score according to Marshall20 and for the Acute Physiology and Chronic Health Evaluation (APACHE) II score.21

288

Cytokines and the Abdominal Surgeon

Table 22.2. Catecholamine administration in patients with severe sepsis assigned to receive pentoxifylline (POF) or placebo

Dopamin (n / %) Dobutamin (n / %) Epinephrin (n / %) Norepinephrin (n / %)

POF

Placebo

27 (100) 16 (59) 13 (48) 14 (52)

24 (100) 15 (63) 11 (46) 12 (50)

Differences among the study groups were not significant.

Effects of Pentoxifylline on Multiple Organ Dysfunction

Organ dysfunction parameters were used as valid endpoints22 besides mortality which was not restricted to the conventional 28-day mortality. Hospital mortality was considered as an additional parameter demonstrating an increased mortality compared to the 28-day rate which was not different between the groups (30 vs 33%). The mortality rate of the control group is consistent with studies of other anti-inflammatory agents indicating a standardized clinical trial protocol with adequate antibiotic and intensive care management. One third of all nonsurvivors died after end of the study period with a 13% reduction of mortality rate in the POF-treated group which was not significant in this small trial (Table 22.1). There has been a lack of formal consensus on criteria used to define the degree of organ dysfunction.23 Based on the studies of Marshall et al20 the optimal descriptors for multiple organ dysfunction were used in this study to characterize the degree of single organ dysfunction. Pulmonary dysfunction was characterized by the worst Pa02/Fi02-ratio, cardiovascular dysfunction was monitored by pressure-adjusted heart rate (central venous pressure (CVP) x heart rate (HR) / mean arterial pressure (MAP). Liver dysfunction was described by total bilirubin, renal dysfunction was characterized by serum creatinine concentration and hematological dysfunction was described by thrombocyte count. PaO2 /FiO2-ratio as parameter for pulmonary dysfunction increased from day 7 after diagnosis of sepsis in patients with POF compared to patients receiving placebo (Fig. 22.1). The difference was statistically significant on day 14 and 17 after diagnosis (p<0.05, t-test with Bonferroni's correction for multiple comparisons). Pressure-adjusted heart rate (CVP x HR / MAP) characterizing cardiac function was significantly different on day 6 after diagnosis (p<0.05) and remained decreased compared to the placebo-group. Myocardial depression with reduction of mean arterial pressure could be presented early in the course of septic shock related directly to the severity of the septic challenge.24 This effect clearly demonstrated that in patients treated with placebo could be attenuated with POF during the initial phase of sepsis (Fig. 22.2). The beneficial effects of TNF blocking agents on cardiovascular function indicate that hemodynamic alterations may be mediated, at least in part, by TNF induction and release.25 Studies on both ventricle and isolated cardiac myocytes demonstrated that TNF exerted a negative inotropic effect in a dose- and time-dependent manner. This effect was fully reversible by removal of TNF or was prevented by pretreatment with TNF blocking agents.26-28

Cytokines and Pentoxifylline in Surgical Sepsis

289

Fig. 22.2. Mean arterial pressure from diagnosis until day 7 after diagnosis of sepsis in patients with pentoxifylline (solid squares) or placebo (open squares). * p<0.05, t-test with Bonferroni's correction.

Increase in mean arterial pressure is the most striking hemodynamic change related to good prognosis in septic patients. Increased MAP in septic patients, mostly with septic shock, is related to both an increase in vascular tone and an improvement in ventricular function,24 which could be achieved within one week with standard treatment of sepsis in this study but earlier in patients with additional administration of POF. In guinea pigs intravenous POF reversed hypotension during TNF infusion, indicating that TNF is involved in the systemic hypotension during sepsis. In rats a hemodynamic benefit by POF administration could be confirmed with reversal of hypotension and improvement of cardiac function.29-31 The mechanism by which POF attenuated the decrease in MAP during the first days of severe sepsis in this study, however, could not be differentiated. POF may directly improve cardiac function or may affect local cytokine production with reduction of TNF production and indirect improvement of organ function. In patients with severe sepsis oxygenation or cardiac function may be effected by mechanical ventilation or variability in the use of inotropic agents. However, no differences in parameters of mechanical ventilation, the number of patients who received inotropes (Table 22.2) or the concentration of inotropic agents could be found. This indicates that the improvement of cardiopulmonary function was effected by pentoxifylline, resulting in a decreased multiple organ dysfunction score. The score was used to describe the multiple organ dysfunction syndrome as clinical endpoint of the septic process. At baseline, the multiple organ dysfunction score was not different between groups. From day 4 after diagnosis of sepsis, the score decreased in patients with POF compared to placebo-infused patients (Fig. 22.3). This difference reached significance on day 14 after diagnosis of sepsis (p < 0.05). The mechanism to improve pulmonary dysfunction in sepsis characterized by arteriolar vasoconstriction and pulmonary artery hypertension33 may be the downregulatory effect of POF on neutrophils with attenuation of alveolar microvascular leakage.32 TNF-induced neutrophil adherence, degranulation, superoxide production

290

Cytokines and the Abdominal Surgeon

Fig. 22.3. Mean Multiple organ dysfunction (MOD)-score according to Marshall20 from surgical patients with pentoxifylline (solid squares) vs placebo (open squares). * p<0.05, t-test with Bonferroni's correction.

and expression of the adhesion molecule could be suppressed by pentoxifylline as a mechanism of improved alveolar leakage.32,34 However, based on experimental studies, it can only be suggested that POF reduces local TNF concentration by inhibition of TNF transcription10 and therefore improved lung injury. The rheological and antithrombotic effects of POF may contribute to the reduced microvascular leak and lung edema.35,36 Improvement of pulmonary leakage may be the beneficial mechanism in this study, which lead to significantly increased the PaO2/FiO2-ratio in POFtreated patients. However, it is not known why this effect does not occur in the early phase but in the later stage of sepsis. Total bilirubin as parameter for hepatic dysfunction, serum creatinine as parameter of renal dysfunction and thrombocyte count as parameter of hematological dysfunction did not show any difference throughout the evaluation in this study.

Cytokines and Pentoxifylline A systemic microcirculatory injury is considered a fundamental problem in the development of organ dysfunction and sepsis. The excessive production of various mediators released by neutrophils and macrophages is known to induce an increased endothelial permeability with organ edema and depressed organ function. Meanwhile clinical data confirm experimental evidence that TNF∀ has to be considered as the central mediator of sepsis.37 No effect on circulating TNF could be demonstrated in contrast to experimental human studies12 and animal studies where a dose-dependent effect on TNF activity was shown.38 No significant difference for TNF bioactivity could be demonstrated between the two study groups throughout the evaluation and during the first two weeks with regard to survivors and nonsurvivors in patients with POF or placebo (Fig. 22.4). According to Zeni et al18 a significant reduction of TNF immunoreactivity in humans was found with a higher dose of POF (1.5 mg/kg bw/h) associated with 2-fold increased plasma concentration of POF after 24 hours. This study again highlighted the difficulties of designing anti-TNF strategies based on circulating TNF levels because it has been difficult to correlate serum

Cytokines and Pentoxifylline in Surgical Sepsis

291

Fig. 22.4. Serial median tumor necrosis factor-∀ bioactivity during the initial two weeks after diagnosis of sepsis. Serum bioactivity in nonsurvivors (nonsurv) or survivors (surv) did not differ significantly between patients with pentoxifylline (POF) or placebo (PLA).

or plasma levels of a particular proinflammatory cytokine with the overall extent of tissue damage.39 The clinical significance of TNF detection in critically ill patients varies in different series because of the short half-life of TNF,40 the timing of measurement, the method used or the influence of circulating TNF inhibitors.41 TNF is usually produced locally within organs and tissues acting more as a paracrine or autocrine mediator. Therefore serum levels of TNF alpha may not reflect TNF bioactivity.42 The effect of POF on downregulation of IL-6 has been demonstrated inconsistently in animal and human studies. While IL-6 release was downregulated in experimental sepsis in rats13 no effect was found in human volunteers12 and in this study with severe sepsis measuring IL-6 bioactivity43 (Fig. 22.5).

Perspectives Pentoxifylline did not effect serum concentration of TNF as the central mediator in sepsis, but the results in terms of pulmonary and cardiac function may suggest beneficial local effects within different organ systems. In experimental intra-abdominal sepsis enhanced survival was associated with increased cell recruitment into the peritoneum without inhibiting TNF synthesis.44 Although the peritoneal response to infection is comparable to the systemic inflammatory response, immune responses occur in functionally separated compartments.4 Intraperitoneal levels of cytokines are higher than systemic concentrations measured simultaneously, indicating a compartmentalized inflammatory immune response.45 Intraperitoneal administration of POF improved survival after cecal ligation and puncture in burned mice through the down-regulation of proinflammatory cytokines46 comparable to intravenous administration. However, the role of localized effects of anti-inflammatory agents especially in relation to the systemic response are not fully known and require further studies.

292

Cytokines and the Abdominal Surgeon

Fig. 22.5. Serial median interleukin-6 bioactivity during the initial two weeks after diagnosis of sepsis. Serum bioactivity in nonsurvivors (nonsurv) or survivors (surv) did not differ significantly between patients with pentoxifylline (POF) or placebo (PLA).

In experimental studies the timing of POF administration was important for the therapeutic effect and adverse events.47 In the clinical setting of this study the administration of POF to septic patients started after the initial insult of organ dysfunction without adverse effects, especially in terms of cardiopulmonary function. Pentoxifylline as an inexpensive, readily available agent has the potential of longterm administration with no adverse effects in patients with sepsis. Because of the beneficial cardiopulmonary effects POF may become a promising agent in the treatment of sepsis. The clinical efficacy of POF in terms of organ dysfunction and survival response, however, has to be evaluated in further studies to prove the hypothesis that modulation of local cytokine-induced inflammation by POF may influence uncontrolled systemic mediator disease in surgical and intraabdominal sepsis.

References 1. Wittmann DH, Schein M, Condon RE. Management of secondary peritonitis. Ann Surg 1996; 224:10-18. 2. Ho HS, Frey CF. The role of antibiotic prophylaxis in severe acute pancreatitis. Arch Surg 1997; 132:487-493. 3. Ayres SM. SCCM's new horizons conference on sepsis and septic shock. Crit Care Med 1985; 13:864-866. 4. Schein M, Wittmann DH, Wise L et al. Abdominal contamination, infection and sepsis, a continum. Br J Surg 1997; 84:269-272. 5. Fong Y, Lowry SF. Tumor necrosis factor in the pathophysiology of infection and sepsis. Clinic Immunol Immunopathol 1990; 55:157-170. 6. Christman JW, Holden EP, Blackwell TS. Strategies for blocking the systemic effects of cytokines in the sepsis syndrome. Crit Care Med 1995; 23:955-963. 7. Fisher CJ, Agosti JM, Opal SM et al. Treatment of septic shock with the tumor necrosis receptor:Fc fusion protein. N Engl J Med 1996; 334:1697-1702. 8. Bone RC. Monoclonal antibodies to tumor necrosis factor in sepsis: Help or harm? Crit Care Med 1993; 21:311-312.

Cytokines and Pentoxifylline in Surgical Sepsis

293

9. Dettelbach HR, Aviado DM. Clinical pharmacology of pentoxifylline with special reference to its hemorrheologic effect for the treatment of intermittent claudication. J Clin Pharmacol 1985; 25:8-26. 10. Doherty GM, Jensen JC, Alexander HR et al. Pentoxifylline suppression of tumor necrosis factor gene transcription. Surgery 1991; 110:192-198. 11. Noel P, Nelson S, Bokulie R et al. Pentoxifylline inhibits lipopolysaccharide-induced serum tumor necrosis factor and mortality. Life Sciences 1990; 47:1023-1029. 12. Zabel P, Wolter DT, Schönharting MM et al. Oxpentifylline in endotoxaemia. Lancet 1989; 2:1474-1477. 13. Lundblad R, Ekstrom P, Giercksby KE. Pentoxifylline improves survival and reduces tumor necrosis factor, interleukin-6, and endothelin-1 in fulminant intraabdominal sepsis in rats. Shock 1995; 3:210-215. 14. Chalkiadakis GE, Kostakis A, Karayannacos PE et al. Pentoxifylline in the treatment of experimental peritonitis in rats. Arch Surg 1985; 120:1141-1144. 15. Harada H, Ishizaka A, Yonemaru M et al. The effects of aminophylline and pentoxifylline on multiple organ damage after Escherichia coli sepsis. Am Rev Respir Dis 1989; 140:974-980. 16. Boldt J, Müller M, Heyn S et al. Influence of long-term continuous intravenous adminstration of pentoxifylline on endothelial-related coagulation in critically ill patients. Crit Care Med 1996; 24:940-946. 17. Montravers P, Fagon JY, Gilber C et al. Pilot study of cardiopulmonary risk from pentoxifylline in adult respiratory distress syndrome. Chest 1993; 103:1017-1022. 18. Zeni F, Pain P, Vindimian M et al. Effects of pentoxifylline on circulating cytokine concentrations and hemodynamics in patients with septic shock: results from a double-blind, randomized, placebo-controlled study. Crit Care Med 1996; 24:207-214. 19. Bone RC, Sprung CL, Sibbald WJ. Definitions for sepsis and organ failure. Crit Care Med 1992; 20:724-726. 20. Marshall JC, Cook DJ, Christou NV et al. Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med 1995; 23:1638-1652. 21. Knaus WA, Draper EA, Wagner DP et al. Apache II: A severity of disease classification system. Crit Care Med 1985; 13:818-829. 22. Sibbald WJ, Vincent JL. Round table conference on clinical trials for the treatment of sepsis. Crit Care Med 1995; 23:394-399. 23. Members of the ACCP/SCCM Consensus Conference Committee: Bone RC, Balk RA, Cerra FB et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapy in sepsis. Crit Care Med 1992; 20:864-874. 24. Abel FL. Myocardial function in sepsis and endotoxin shock. Am J Physiol 1989; 257:1265-1281. 25. Yao Y-M, Bahrami S, Redl H et al. Monoclonal antibody to tumor necrosis factora attenuates hemodynamic dysfunction secondary to intestinal ischemia/reperfusion in rats. Crit Care Med 1996; 24:1547-1553. 26. Yokoyama T, Vaca L, Rossen RP et al. Cellular basis for the negative inotropic effects of tumor necrosis factor-alpha in the adult mammalian heart. J Clin Invest 1993; 92:2302-2312. 27. Weisensee D, Breitner-Hahn J et al. Effects of cytokines on the concractility of cultured cardiac myocytes. Int J Immunopharmacol 1993; 15:581-587. 28. Vincent J-L, Bakker J, Marecaux G et al. Administration of anti-TNF antibody improves left ventricular function in septic shock patients: Results of a pilot study. Chest 1992; 101:810-815. 29. Lilly CM, Sandhu JS, Ishizaka A et al. Pentoxifylline prevents tumor necrosis factor-induced lung injury. Am Res Respir Dis 1989; 139:1358-1361.

294

Cytokines and the Abdominal Surgeon

30. Wang P, Ba ZF, Zhou M et al. Pentoxifylline restores cardiac output and tissue perfusion after trauma-hemorrhage and decreases susceptibility to sepsis. Surgery 1993; 114:352-359. 31. Chick TW, Scotto P, Icenogle MV et al. Effects of pentoxifylline on pulmonary hemodynamics during acute hypoxia in anesthetized dogs. Am Rev Respir Dis 1988; 137:1099-1103. 32. Carter MB, Wilson MA, Wead WB et al. Pentoxifylline attenuates pulmonary macromolecular leakage after intestinal ischemia-reperfusion. Arch Surg 1995; 130:1337-1344. 33. Villar J, Blzquez MA, Lubillo S et al. Pulmonary hypertension in acute respiratory failure. Crit Care Med 1989; 17:523-526. 34. Lo SK, Everitt J, Gu J et al. Tumor necrosis factor mediates experimental pulmonary edema by ICAM-1 and CD-18-dependant mechanisms. J Clin Invest 1992; 89:981-988. 35. Tighe D, Moss R, Heath MF et al. Pentoxifylline reduces pulmonary leucostasis and improves capillary patency in a rabbit peritonitis model. Circ Shock 1989; 28:159-164. 36. Vesconi S, Rossi GP, Pesenti A et al. Pulmonary microthrombosis in severe adult respiratory distress syndrome. Crit Care Med 1988; 16:111-113. 36. Espevik T, Nissen-Meyer J. A highly sensitive cell line, WEHI 164 subclone 13, for measuring cytotoxic factor/tumor necrosis factor from human monocytes. J Immunol Methods 1986; 95:99-105. 37. Cohen J, Carlet J. Intersept: An international, multicenter, placebo-controlled trial of monoclonal antibody to human tumor necrosis factor-∀ in patients with sepsis. Crit Care Med 1996; 24:1431-1440. 38. Schade UF. Pentoxifylline increases survival in murine endotoxin shock and decreases formation of tumor necrosis factor. Circ Shock 1990; 31:171-181. 39. Suffredini AF. Current prospects for the treatment of clinical sepsis. Crit Care Med 1994; 20:S12-S18. 40. Michie HR, Manogue KR, Spriggs DR et al. Detection of circulating tumor necrosis factor after endotoxin administration. N Engl J Med 1988; 318:1481-1486. 41. Van Zee KJ, Kohno T, Fischer E et al. TNF soluble receptors protect against excessive TNF∀ during infection and injury. Proc Natl Acad Sci USA 1992; 80:4845-4849. 42. Fong Y, Moldawer LL, Shires GT et al. The biological characteristics of cytokines and their implications in surgical injury. Surg Gynecol Obstet 1990; 170:363-378. 43. Aarden LA, de Groot ER, Schaap OL et al. Production of hybridoma growth factor by human monocytes. Eur J Immunol 1987; 17:1411-1416. 44. Hadjiminas DJ, McMasters KM, Robertson SE et al. Enhanced survival from cecal ligation and puncture with pentoxifylline is associated with altered neutrophil trafficking and reduced interleukin-1b expression but not inhibition tumor necrosis factor synthesis. Surgery 1994; 116:348-355. 45. Holzheimer RE, Schein M, Wittmann DH. Inflammatory response in peritoneal exudate and plasma of patients undergoing planned relaparotomy for severe secondary peritonitis. Arch Surg 1995; 130:1314-1320. 46. Holzheimer RG, Molloy RG, O´Riordain DS et al. Long-term immuntherapeutic intervention with pentoxifylline in a mouse model of thermal injury and infection. J Trauma 1995; 38:757-762. 47. Ridings PC, Windsor ACJ, Sugerman HJ et al. Beneficial cardiopulmonary effects of pentoxifylline in experimental sepsis are lost once septic shock is established. Arch Surg 1994; 129:1144-1152.

Epilogue

295

CHAPTER 23

Epilogue Moshe Schein and Leslie Wise “Now this is not the end. This is not even the beginning of the end. But it is, perhaps, the end of the beginning.” Winston Churchill, 1942

Introduction

M

atters were much more simpler for us—clinical surgeons—only a few years ago. Postoperative or posttraumatic fever, raised white cell count, sliding-down organ-system function, with or without “shock” meant for us only one thing—“sepsis”. And “sepsis” meant “infection”, usually bacterial in nature, necessitating antibiotic therapy. So we administered the “strongest”, ever-changing, antimicrobial agents available on the market, we looked for pus-draining it whenever present, and we prayed for the “infection” to subside. Some of our patients, however, continued to deteriorate, dying slowly from respiratory and/or renal failure. We buried them, blaming the death on an “intractable sepsis”, which in our minds always signified an infection—“somewhere” in their blood, abdomen, urine or lungs. Then, in the early 1980s, when our supportive care and reoperative efforts became more aggressive-resulting in prolonged survival, we begun to note that many of our patients were dying a “septic” death in the absence of “infection”; we did not understand why. Towards the second half of the 1980s a rapidly developing field of molecular biology has started “bombarding” us with a huge volume of studies mentioning a new term—cytokines. Initially, we paid little attention, doubting that these obscure molecules have anything to do with our patients. But gradually, persuaded by the burden of evidence accumulated by basic scientists and brighter-than us surgeons, we begun to understand that the buzz word to explain a lot of what we see in clinical practice is not “sepsis” or “infection” but inflammation-which in turn is fueled by pro-inflammatory mediators-cytokines. The knowledge so well summarized in the preceding pages totally changed the way we began to look at the surgical patient. We see him being inflamed by operative trauma as well as by the postoperative complications and their therapies. But before we go further, discussing what we-clinicians can and should do about all the data presented in this book, we need to clarify a few issues in terminology.

Infection vs Inflammation and Sepsis The evolving concept that the term infection is not synonymous with sepsis, and that sepsis is one form of inflammation—i.e., SIRS (systemic inflammatory response syndrome) but that SIRS also includes noninfective causes, may appear complicated but was developed to simplify and explain matters. That some surgeons use the terms Cytokines and the Abdominal Surgeon, edited by Moshe Schein and Leslie Wise. © 1998 R.G. Landes Company.

296

Cytokines and the Abdominal Surgeon

infection and sepsis interchangeably is evident in the literature where abdominal sepsis and wound sepsis seem to mean for many abdominal and wound infection, respectively. It has been said that “changing terminology results from the evolution of patient care, and from attempts to clarify or better describe what is happening...”1 This has occurred in the last decade during which an immense body of new datapartially summarized in the preceding chapters, has increased our understanding of the biologic events that generate the systemic inflammatory response syndrome (SIRS), now a common denominator for sepsis, septic shock, organ dysfunction and, eventually, death in critically ill medical, surgical and severely traumatized patients.2 Until the past decade it was presumed that the physiologic response to infection represented a microorganism related consequence of their proliferation or toxin production. This understanding still prevails in updated medical dictionaries in which ‘septic’ is defined as “pertaining to an infection with pyogenic microorganisms”3 and ‘sepsis’ is described as “ the presence of pathogenic microorganisms or their toxins in the blood or other tissues and, the condition associated with such presence.”4,5 Nothing is mentioned about the host’s response to the inflammatory stimulus and about differences between localized and systemic events. In the late 1970s, the absence of infective foci in some patients with intra-abdominal infection and “dying a septic death” led to the evolution of the concept of “sepsis without infection” or tertiary peritonitis. Meakins et al termed this phenomenon “nonbacteremic clinical sepsis;” 6 while others named it “sepsis without bacteria,”7 “hypermetabolic organ failure complex”8 or the “systemic septic response.”9 Gradually, it has been recognized that clinical sepsis, manifesting all the signs that traditionally were associated with infection such as tachycardia, fever, increased respiratory rate, altered mentation, hyperdynamic circulation, hypermetabolic ‘autocannibalistic’ response, and, at the cellular level, defects in oxygenation, represents a nonspecific systemic host determined inflammatory-septic response. This septic response can follow any infection irrespective of the specific causative bacterial, fungal or viral microorganism,10 and may persist for a variable period of time despite control of the primary infection. Notably, a response identical to the septic response may be triggered by multiple noninfective conditions such as trauma,11 burns,12 acute pancreatitis,13 aspiration pneumonia,14 necrotic tissue15 and tissue inflammation16 as recently reviewed.17 Marshall and Sweeney studied the differential roles of infection and sepsis, and demonstrated that the magnitude of the host response, independent of the presence of bacteria, bacteriologic characteristics, or control of infection, is the major determinant of outcome in critical surgical patients.7 Moreover, microorganisms cultured from patients with advanced sepsis may represent a consequence of the physiologic derangements rather than a cause.18 A hypothesis has matured to suggest that sepsis represents an exaggerated immune response of the host leading to generalized autodestructive inflammation.19 This concept has been substantiated by the advances in molecular biology that unraveled the biological mechanisms producing clinical sepsis and tissue damage through the common final pathway of endogenous proinflammatory mediators, and activation of leukocytes at the endothelial level.20

Current Terminology in Clinical Sepsis The recognition that clinical sepsis can arise in the absence of infection has created confusion in the mind of those using this term to imply infection. New terminology was thus required to ‘cover’ the newly accumulated perception. The term

Epilogue

297

systemic inflammatory response syndrome (SIRS) was coined to encompass the noninfectious situations producing clinical sepsis, replacing the terms ‘septic state’ or ‘septic response.’21 Sepsis was redefined as the systemic response to infection, consisting of SIRS with microbiological evidence of infection (Sepsis = inflammation (SIRS) + infection). In another words, SIRS and sepsis represent an identical hostdetermined response, the former in culture negative patients and the latter when infection is documented. Both manifest a continuum of clinical and pathophysiologic severity.22 The term septicaemia has been defined in the past as a systemic disease caused by the presence of microorganisms or their toxins in the blood,4 but has been used clinically to cover a spectrum of SIRS/sepsis-associated conditions. To reduce confusion the presence of viable microorganisms in the blood is preferably termed bacteraemia, viraemia or fungaemia, according to the specific causative organism independent of clinical symptoms.21 Recently, the term bloodstream infection, describing signs of systemic infection when a pathogen is isolated from the blood, was proposed to replace septicemia.23

Local Inflammation vs Contamination, Infection or Sepsis The aforementioned arguments which clarified the nomenclature of systemic infection/sepsis applies also to local surgical conditions. The terms intra-abdominal infection and intra-abdominal sepsis, for example, are used interchangeably by some surgeons. Current wisdom implies that the term abdominal sepsis as used today is confusing and not specific enough because it describes the systemic inflammatory response only and does not specifically address local events within the peritoneal cavity. The term abdominal sepsis has been used to emphasize the systemic response to an initially localized infectious process. Now, as we have learned that sepsis is only one component of the SIRS and SIRS is defined as a systemic rather than localized phenomenon, the term abdominal sepsis becomes obsolete because it does not address the important initial local inflammatory response. Furthermore, it does not address the noninfectious local inflammatory response seen in peritonitis. Analogous to systemic sepsis, abdominal sepsis represents the nonspecific peritoneal inflammatory response of the host to various noxious causes. Although the peritoneal inflammatory response is comparable to the systemic inflammatory response and utilizes identical mechanisms of humoral and cellular interactions, the two responses occur in two functionally separate peritoneal and systemic compartments. Studies in experimental primary, and secondary bacterial peritonitis have demonstrated that bacterial invasion of the peritoneum is associated with a tremendous local outpouring of pro-inflammatory cytokines, with their intraperitoneal concentrations inversely correlating with survival. Peritoneal levels of cytokines in peritonitis are manyfold higher than the simultaneously measured systemic levels, indicating a compartmentalized inflammatory process with plasma levels representing only the tip of the iceberg.24 As discussed by the various authors in this book, noninfectious abdominal conditions also can trigger an identical cytokine-mediated response. Levels of IL-1 and IL-6 are elevated following major abdominal operations, the extent of the response increasing with the magnitude of the operative trauma, and with portal vein levels higher than those measured systemically.25 Also, acute pancreatitis represents a local, intra-abdominal, ‘mediator disease’ with portal vein levels of TNF-∀ higher than in

298

Cytokines and the Abdominal Surgeon

the hepatic vein (chapter 7). Hence, it appears that the term ‘abdominal sepsis’ represents a local inflammatory response syndrome (LIRS)-an intra-abdominal analog of SIRS caused by microorganisms. It has been mentioned that SIRS caused by infection with a microorganism is sepsis. In analogy, is the presence of bacteria in the peritoneal cavity enough to call it intra-abdominal infection or “sepsis”? Not always. The early invasion of the peritoneal cavity with microorganisms indicates an abdominal contamination. It occurs during the period in which tissue invasion has yet to occur, as evidenced by lack of a local inflammatory response.26 A common example is a traumatic colonic perforation with peritoneal soiling. Intra-abdominal infection, on the other hand, is an inflammatory response to microorganisms or the invasion of normal sterile peritoneal tissues by these organisms. Surgeons still mix cases of abdominal contamination with those of infection or sepsis, lumping them together in studies entitled “severe intraabdominal sepsis”. Consequently, many clinical trials are ‘diluted’ with low mortality penetrating trauma cases that represent contamination rather than infection with a low mortality risk. For example, the average mortality rate in the antibiotic studies reviewed by Solomkin et al27 was only 3.5%. This sharply contrasts with the 32% and 29% mortality rates reported with nonselected intra-abdominal infections.28,29

Contamination, Infection, Sepsis: A Continuum Clearly, local contamination, infection and sepsis mean disparate things. These specific conditions, however, may co-exist in the same patient, developing simultaneously or consecutively. The installation of feces into the peritoneal cavity may involve a continuum of local and systemic conditions ranging from local contamination to septic shock. Untreated or neglected abdominal contamination ends in intra-abdominal infection and the latter is invariably associated with a systemic septic response. More significantly, abdominal inflammation may persist after the infection is cured. Surgical and antimicrobial therapy addresses the infectious component but does not always halt the inflammation. Two or three planned re-operations supplemented with a short course of antibiotics are sufficient to sterilize the peritoneum in cases of most severe peritonitis but the inflammatory process may continue.30,31 A scenario in which bacterial peritonitis is cured, but local and systemic inflammation persists, represents “tertiary peritonitis”, a term coined to describe that subgroup of patients who develop multiple organ dysfunction and die despite “successful” operations, effective antibiotic treatment and maximal supportive therapy.32 After adequate treatment of less advanced abdominal infections, residual abdominal inflammation is self limited. It may manifest postoperatively as persistent fever or leukocytosis and usually reflects local residual cytokine-mediated inflammation or LIRS rather than continuing infection.

Clinical Relevance “Our current diagnostics are inadequate to define when the critical threshold of local inflammation has been transgressed and the transition to systemic inflammation has occured” writes Donald Fry adding that “in retrospect, ten years from now, I am certain that we will laugh at the primitive criteria that we have employed for the entry of patients into “sepsis trials”.33 The previous chapters in this book helped us understand that things are not so simple anymore and that the chief driving powers of morbidity and mortality in our surgical patients are inflammatory. To complicate matters further the late Roger Bone emphasized that in response to LIRS and SIRS the body mounts a local and systemic

Epilogue

299

anti-inflammatory response, mediated by anti-inflammatory cytokines-termed by him CARS (compensatory anti-inflammatory response syndrome). When SIRS and CARS are balanced, Bone hypothesized, homeostasis occurs; when SIRS predominates apoptosis (cell death) and organ dysfunction is the result; when CARS prevails immunoparesis is the consequence.34 As the search for the magic bullet to arrest the cascades of LIRS, SIRS and to modulate CARS continues, we begin to understand the immense complexity of the patient’s response to disease and surgery. As with many other essential things in life, too much may be harmful and too little may not be satisfactory. The same probably is true for the inflammatory and anti-inflammatory responses, which in a certain phase and magnitude are beneficial but when out of control are harmful. The timing and extent of interference with the mechanisms of inflammation are therefore crucial for therapeutic success, but as yet remain undefined. But meanwhile, until effective anti-LIRS/SIRS and CARS management becomes available, is there anything we can do for these patients based on all this new knowledge? Certainly; first, we need to use terms accurately, distinguishing between local inflammation and infection, between SIRS and systemic sepsis. In abdominal surgery, for example accurate usage of the terms contamination, infection and sepsis is of importance in the study and treatment of these conditions since they may represent a continuum but are approached differentially. Abdominal contamination is controlled by the local peritoneal defense mechanisms, assisted by operative peritoneal toilet and prophylactic antibiotics. Infection is treated by the surgical control of its source and therapeutic antibiotics. The term sepsis should be used only to connote the systemic inflammatory response that is often seen in intra-abdominal infections. Persistence of SIRS and LIRS after infection is cured should be managed expectantly and without antibiotics.35 Second, we must avoid adding fuel to the inflammatory fire, appreciating that what, how much and how we do things matter. “Greater” operation and rough handling of tissues means more inflammation—more LIRS and SIRS. Unnecessary and poorly-timed re-interventions may produce a “second hit” in a previously primed host.36 Ongoing sources of LIRS and SIRS, infective and noninfected alike, should be dealt with early. Iatrogenic contributors to LIRS and SIRS should be minimized: the patient must not be continuously injured with indiscriminate insertion of catheters, tubes and pipes. Blood products may be harmful and should be used only as necessary. Antibiotics represent a double-edged sword and may in fact increase SIRS by various mechanisms. It is impossible to prove that each of the above measures decrease SIRS/MODS; but proper management as a whole is the mainstay of prevention of the so-called horror autotoxicus.37 The vast information presented in this book affirms us that matters are not simple, in fact, occasionally, with increasing knowledge they may become more and more complicated. In historical terms, however, our increasing knowledge may perhaps be “the end of the beginning” (Winston Churchill, 1942).

References 1. Baue AE. What’s in a name? An acronym or a response? Am J Surg 1993; 165:299-301. 2. Bone RC. Sepsis and its complications: The clinical problem. Crit Care Med 1994; 22(Suppl):S8-S11. 3. Anderson KN, Anderson LE, Glanze WD (ed). Mosby’s Medical, Nursing and Allied Health Dictionary, 4th edition, Mosby-year Book, 1994.

300

Cytokines and the Abdominal Surgeon

4. Dorland I, Newman WA, eds. Dorland’s Illustrated Medical Dictionary, 28th edition Philadelphia: WB Saunders Company, 1994. 5. Brown MJ, ed. Miller-Keane, Encyclopedia & Dictionary of Medicine, Nursing, & Allied Health, 5th edition, Philadelphia: W.B Saunders Company, 1992. 6. Meakins JL, Wicklund B, Forse RA, McLean PH. The surgical intensive care unit: Current concepts in infection. Surg Clin N Amer 1980; 60:117-132. 7. Marshall J, Sweeney D. Microbial infection and the septic response in critical surgical illness. Arch Surg 1990; 125:17-23. 8. Cerra FB. Hypermetabolism, organ failure, and metabolic support. Surgery; 1987; 101:1-14. 9. Wiles JB, Cerra FB, Siegel JH, Border JR. The systemic sepstic response: Does the organism matter? Crit Care Med 1980; 8:55-60. 10. Ayres SM. SCCM’s new horizon conference on sepsis and septic shock. Crit Care Med 1985; 13:864-866. 11. Nuytinck HKS, Offermans XJ, Kubat K, Goris RJA. Whole body inflammation in trauma patients. Arch Surg 1988; 123:1519-1524. 12. Marano MA, Fong Y, Moldawer LL, Wei H, Calvano SE, Tracey KJ et al. Serum cachectine/tumor necrosis factor in critically ill patients with burns correlates with infections and mortality. Surg Gyencol Obstet 1990; 170:32-38. 13. Grewal HP, Kotb M, El Din D, Ohman M, Salem A, Gaber L, Gaber AO. Induction of tumor necrosis factor in severe acute pancreatitis and its subsequent reduction after hepatic passage. Surgery 1994; 115:213-221. 14. Goldman G, Welbourn R, Kobzik L, Valerie CR, Shepro D, Hechtman HB. Tumor necrosis factor mediates alpha acid aspiration-induced systemic organ injury. Ann Surg 1990; 212:513-520. 15. Trunkey DL. Inflammation and trauma. Arch Surg 1998; 123:1517-1518. 16. Lalonde C, Demling RH, Pecquet Goad ME. Tissue inflammation without bacteria produece increased oxygen consumption and distant organ peroxidation. Surgery 1988; 104:49-56. 17. Najma AA, Christou NV, Meakins JL. The systemic inflammatory response syndrome and the critically ill surgical patient. Current Opin Critical Care 1995; 1:290-305. 18. Rotstein, OD, Pruett, TL, Sommons RL. Microbiologic features and treatment of persistant peritonitis in the intensive care unit. Can J Surg 1986; 29:247-251. 19. Goris RJA, Boekhorst TPA, Nuytinck JKS, Gimbrere JSF. Multiple Organ Failure. Generalized autodestructive inflammation? Arch Surg 1985; 120:1109-1115. 20. Livingston DH, Mosenthal AC, Deitch EA. Sepsis and multiple organ dysfunction syndrome: A clinical-mechanistic overview. New Horizons 1995; 3:257-266. 21. Bone RC and Members of the American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference Committee: Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992; 20:864-874. 22. Rangel-Frausto MS, Pittet D, Costigan M, Hwang T, Davis C, Wenzel RP. The natural history of the Sytemic Inflammatory Response Syndrome (SIRS). JAMA 1995; 273:117-123. 23. NNIS manual. National Nosocomial Infections Surveillance System, U.S Department of Health and Human Services Public Health Service. Center for Disease Control Atlanta, Georgia, 1988. 24. Schein M, Wittmann DH, Holzheimer R, Condon RE. Hypothesis: Compartmentalization of cytokines in intra-abdominal infection. Surgery 1996; 119:694-700.

Epilogue

301

25. Riché F, Dosquet C, Panis Y, Valleur P, Lainsné MJ, Briard C, Wautier JL. Levels of portal and systemic blood cytokines after colectomy in patients with carcinoma or Crohn’s disease. J Am Coll Surg 1995; 180:718-724. 26. Bohnen JMA, Solomkin JS, Dellinger EP, Bjornson HS, Page CP. Guidelines for Clinical Care: Anti-Infective Agents for Intra-abdominal Infection: A Surgical Infection Society Policy Statement. Arch Surg 1992; 127:83-89. 27. Solomkin JS, Meakins JL, Dellinger EP. Antibiotics trials in intra-abdominal infections: a critical evaluation of study design and outcome reporting. Ann Surg 1984; 201:29-39. 28. Wittmann DH, Walker AP, Condon RE. Peritonitis, intra-abdominal infection, and intra-abdominal abscess. In: Schwartz SI, Shires GT, Spencer FC, eds. Principles of Surgery. 6th ed. New York: McGrawHill, 1993:1449-1484. 29. Schein M. Management of severe intra-abdominal infection. Surg Annual 1992; 24:47-68. 30. Aprahamian C, Schein M, Wittmann DH. Cefotaxime and metronidazole in severe intra-abdominal infection. Diagn Microbiol Infect Dis 1995; 22:183-188. 31. Van Goor H, Hulsebos R, Bleichrodt R. Complications of planned relaparotomies in patients with severe generalized peritonitis. Curr Opinion Surg Infect 1995; 3(Suppl):28(abstract). 32. Rotstein OD, Meakins JL. Diagnostic and therapeutic challenges of intraabdominal infections. World J Surg 1990; 14:159-166. 33. Fry DE. Inflammation, infection and organ failure—Invited comment. In: Crucial Controversies in Surgery-1988. Schein M, Wise L, eds. Krager Landes Systems, 1997:8C. 34. Bone RC. Sir Isaac Newton, sepsis, SIRS, and CARS. Crit Care Med 1996; 24:1125-1128. 35. Schein M, Wittmann DH, Lorenz W, eds. Duration of antibiotic treatment in surgical infections of the abdomen. Eur J Surg 1996; 162(supp576):1-75. 36. Meldrum DR, Cleveland JC, Moore EE, Patrick DA, Banerjee A, Harken AH. Adaptive and maladaptive mechanisms of cellular priming. Ann Surg 1997; 226:587-598. 37. Baue AE. The horror autotoxicus and multiple organ failure. Arch Surg 1992; 127: 1451-1452.

Index A Adenoviruses 12 Adhesions 1, 53-59 Alcoholic liver disease 146 Anesthesia 27, 29, 100, 101, 216 Angiogenesis 13, 37, 38, 40-42, 44, 58 Antibiotics 30, 107, 117, 123-126, 128-134, 151, 154-156, 163, 165-167, 222, 245, 257, 261, 273, 274, 276-282, 286, 298, 299 Anticytokine therapy 2, 80, 222, 245, 263 Antisense oligonucleotides 13 Aortic surgery 2, 26, 185, 215-218, 221-224 Apoptosis 8-10, 12, 22, 25, 27, 77, 122, 133, 186, 187, 250, 299 ARDS 26, 29, 75, 78, 79, 222, 287 Arginine 28, 84, 90, 91, 204 Ascitic fluid (AF) 118, 143-149

B Bax 10 Blunt abdominal trauma 63, 66, 68 BRCA1 10

C Cancer 5, 7, 8, 10-13, 21, 25, 27, 28, 64, 84, 87, 90, 124, 127 CD18 164, 172, 220 Chronic ambulatory peritoneal dialysis (CAPD) 147 Complement 22, 23, 98, 111, 143, 144, 148, 151, 172, 254, 261 C-reactive protein (CRP) 67, 98-100, 102, 126, 134, 146-148

E EGF 40, 42, 43 ELISA 236 Endoscopic retrograde cholangiopanreatography (ERCP) 99

Endotoxin 19, 20, 22, 24, 27-29, 84-89, 92, 98, 108, 110, 112, 117, 119, 120, 123, 132, 147, 151-156, 163-166, 171-176, 179-184, 188, 203-208, 216, 217, 219, 220, 222, 223, 245, 246, 252-257, 262, 273-281, 285 Endotoxin neutralizing protein (ENP) 153 Enteral nutrition (EN) 86 Enterocyte 9, 84, 183, 198, 201, 203-205, 207-209 Enzyme-linked immunoabsorbent assay (ELISA) 7, 8 Epidermal growth factor (EGF) 43, 58 Epithelialization 38, 40-44, 54 Extracellular matrix 39, 58

F Fas 9, 22, 27, 250 FGF 35, 37, 40, 41 Fibronectin 38-41, 123 Filgrastim 122-124, 126, 133, 134

G Gene therapy 3, 4, 11-13, 262, 263 Granulocyte colony stimulating factor (G-CSF) 2, 40, 119, 120, 122, 123, 126, 129, 131-135, 153, 250, 257, 260 Granulocyte-macrophage colony stimulating factor (GM-CSF) 18, 19, 21, 40, 42, 120, 122, 134, 148, 153, 167, 237, 250, 257, 259 Growth hormone (GH) 84, 90-92, 101

H Hemorrhage 22, 25, 63, 66, 67, 184, 185, 190, 221 Human genome project 5

304

I ICE family proteases 9 IFN-! 19, 23, 35, 37-39, 41, 42, 65-67, 83, 84, 87-91, 112, 146, 201, 202, 234-240, 248, 249, 252, 253, 255 IGF-1 90-92 Inducible nitric oxide synthase (iNOS) 38, 39 Inflammatory bowel disease (IBD) 26, 197, 199, 200, 202 Insulin-like growth factor 1 84, 91 Interferon (IFN) 19, 21, 23-25, 35, 37, 38, 63, 101, 201, 202, 221, 234, 235, 238, 249 Interleukin-1 9, 19, 20, 107, 146, 203, 204, 216, 250 Interleukin-2 58, 147, 248 Interleukin-3 63, 64, 66, 112, 120, 237 Interleukin-4 251 Interleukin-6 (IL-6) 18, 21, 23-29, 38, 39, 41, 54-57, 63-68, 74-78, 83-92, 120, 126, 185, 98-103, 107, 110-112, 146-149, 205, 248, 292 Interleukin-8 24, 54, 188 Interleukin-10 24, 251 Interleukin-12 25, 248 Interleukin-13 251 Ischemia/reperfusion 3, 180, 237, 240

Cytokines and the Abdominal Surgeon Macrophage colony stimulating factor (M-CSF) 119, 120, 134, 135, 237 Macrophage inflammatory proteins (MIP-1) 37, 237, 238, 249 Major histocompatibility complex (MHC) 19, 23, 26, 64, 65, 122, 135, 164, 201, 235, 236, 252, 254 Megakaryocyte growth and development factor (MGDF) 120, 134, 135 Mesenteric ischemia 2, 179, 180, 182, 184, 188 Minimally-invasive surgery 29 Monocyte chemotactic factor-1 (MCF-1) 36 Multiple organ dysfunction syndrome (MODS) 215, 217, 221-223, 289 Multiple organ failure (MOF) 17, 19, 24, 26, 87, 91, 107, 113, 118, 125, 135, 163, 171-176, 179-182, 184, 185, 187-189, 197, 208, 245, 247, 253, 263

N Neutrophil-activating peptide-2 (NAP-2) 37 Nitric oxide (NO) 38, 39, 41, 42, 75, 77, 90, 204, 249-251, 254, 275 Northern blot 6, 201 Nutrition 1, 3, 27, 28, 83-88, 91

K

O

Kupffer cells 20, 22, 25, 27, 85, 86, 88, 111, 208, 237, 252, 254, 261, 262

“one-hit” model 172

L Laparoscopic cholecystectomy 29, 97, 99-101, 166, 185 Laparoscopic surgery 29, 97, 98-101, 103 Leukotriene (LT) 88, 218 LPS 20, 37, 42, 85, 86, 88-92, 111, 112, 120, 135, 151-154, 165, 181, 182, 205, 206, 246, 255, 275-277, 279, 280

M Macrophage 18, 20, 21, 35-43, 54, 57, 58, 64-68, 78, 84-86, 88, 89, 91, 98, 99, 110-112, 120, 130, 144, 147, 148, 151-155, 164, 166, 171, 173, 179, 182, 184, 186, 201, 216, 221, 235, 237-239, 247-256, 261-263, 276, 280, 290

P p53 10, 13 Pancreatitis 1, 9, 73-81, 125, 166, 245, 246, 285, 287, 296, 297 PCR 4, 7, 201, 205, 206 Pentoxifylline 2, 67, 164, 166, 190, 222, 258, 285-292 Peritonitis 2, 24, 25, 27, 28, 86-89, 92, 100, 109, 111, 113, 124-126, 143-147, 245-247, 252, 253, 256-258, 260, 263, 285, 287, 296-298 Platelet 21, 36, 37, 41-43, 58, 65, 75, 77, 107, 111, 119, 120, 186, 237, 286 Platelet factor 4 35, 40 Platelet-derived growth factor (PDGF) 35, 36-42, 58 Polymorphonuclear neutrophils (PMNs) 54, 180-183, 185-188, 190

Index Polyunsaturated fatty acids (PUFA) 67, 84, 88 Programmed cell death (apoptosis) 8, 22, 25, 186 Proinflammatory cytokines 19, 27, 65, 67, 97, 107, 109-113, 120, 123, 132, 146, 151, 155, 156, 163, 165, 184, 236, 237, 285, 291 Prostaglandin (PG) 18, 20, 36, 56, 64, 65, 68, 88, 122, 236, 280

R RANTES 237, 249 rhG-CSF (filgrastim) 122, 130-134 RT-PCR 7

S Sepsis 1-4, 17, 20, 21, 25, 27, 28, 38, 39, 41, 43, 66, 67, 75, 78-80, 86, 90, 101, 107-113, 117-119, 123, 125-129, 131-135, 151-155, 163, 165, 166, 171-174, 197, 203-205, 207, 223, 245-248, 252-258, 262, 263, 273-279, 281, 282, 285-292, 295-299 SIRS (systemic inflammatory response syndrome) 117 Southern blot 6 Splenocyte 66, 90, 91, 112, 153, 255, 261 Spontaneous bacterial peritonitis (SBP) 2, 143-149

305 Stem cell factor (SCF) 120, 134, 135 Systemic inflammatory response syndrome (SIRS) 163, 171-173, 180, 181, 187, 221, 223, 295-297

T T cells 19-21, 25, 57, 64, 65, 112, 201, 207, 233-236, 239, 252, 253 Thromboxane A2 65, 219, 220 TPN 84-87, 89, 90 Transforming growth factor-beta (TGF-#) 35, 253, 254, 237, 239, 240 Tumor necrosis factor (TNF) 9, 18-21, 23-29, 35, 37-40, 42, 43, 54-57, 63-65, 67, 68, 74-79, 83-92, 98-102, 107, 110-113, 117, 119, 120, 122, 123, 132, 135, 146-149, 181-186, 188, 190, 201, 203, 205, 207, 210, 216, 217, 219-222, 235, 237, 239, 248-255, 257-259, 262, 263, 275-277, 280, 281, 285, 286, 288-291, 297 Tumor necrosis factor-alpha (TNF-∀) 35, 146 “two-hit” model 172

W Western blot 8, 208 Wound contraction 42