Blood Banking and Transfusion Medicine (Second Edition): Basic Principles and Practice

1600 John F. Kennedy Blvd. Ste 1800 Philadelphia, PA 19103-2899

BLOOD BANKING AND TRANSFUSION MEDICINE, Second Edition

ISBN-13: 978-0-443-06981-9 ISBN-10: 0-443-06981-6

Copyright © 2007, 2003 by Churchill Livingstone, an imprint of Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permissions may be sought directly from Elsevier’s Health Sciences Rights Department in Philadelphia, PA, USA: phone: (+1) 215 239 3804, fax: (+1) 215 239 3805, e-mail: [email protected]. You may also complete your request on-line via the Elsevier homepage (http://www.elsevier.com), by selecting ‘Customer Support’ and then ‘Obtaining Permissions’.

Cover photo copyright T & K Image/Photo Researchers; with permission.

Notice Knowledge and best practice in this field are constantly changing. As new research and experience broaden our knowledge, changes in practice, treatment and drug therapy may become necessary or appropriate. Readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of the practitioner, relying on their own experience and knowledge of the patient, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the Editors assume any liability for any injury and/or damage to persons or property arising out or related to any use of the material contained in this book. The Publisher

Library of Congress Cataloging-in-Publication Data Blood banking and transfusion medicine: basic principles & practice / Christopher D. Hillyer … [et al.].—2nd ed. p. cm. Includes bibliographical references and index. ISBN 0-443-06981-6 1. Blood—Transfusion. 2. Blood banks. I. Hillyer, Christopher D. II. Title. RM171.B583 2007 615′.39—dc22 2006048955

Acquisitions Editor: Dolores Meloni Developmental Editor: Kristina Oberle/Kim DePaul Project Manager: Bryan Hayward Design Direction: Steven Stave

Printed in United States of America

Last digit is the print number: 9 8 7 6 5 4 3 2 1

About the Editors

Dr. Hillyer is a tenured professor in the Departments of Pathology and Pediatrics, as well as the Division of Hematology/Oncology, Winship Cancer Institute, Emory University School of Medicine. He serves as director of the Transfusion Medicine Program at Emory and oversees the Emory University Hospital Blood Bank, the blood and tissue banks of Children’s Healthcare of Atlanta, and the Emory Center for the Advancement of International Transfusion Safety. He is an editor of three textbooks on transfusion medicine and an author of more than 120 articles and chapters pertaining to transfusion, HIV, cytokines, and herpesviruses (most notably CMV). Nationally recognized as an expert in hematology and blood transfusion, Dr. Hillyer is President of AABB (2006–2007) and is a Trustee of the National Blood Foundation (NBF). He has been awarded research funding from the NIH, CDC, NBF, and other agencies. He currently serves as principal investigator of a program project grant, several R-series awards, the Emory site of the NHLBI’s Transfusion Medicine/Hemostasis Clinical Trial Network, and REDS-II. He also is a co-principal investigator of AABB’s contract with HHS to provide technical assistance to six developing nations under the President’s Emergency Plan for AIDS Relief (PEPFAR). Dr. Hillyer is an associate editor of Transfusion and part-time medical director of the American Red Cross Southern Region. Dr. Hillyer is board certified in Transfusion Medicine, Hematology, Medical Oncology, and Internal Medicine. He received his BS from Trinity College and his MD from the University of Rochester School of Medicine, with postgraduate training and fellowships in hematology-oncology, transfusion medicine, and bone marrow transplantation at Tufts–New England Medical Center in Boston.

v Dr. Silberstein is a tenured professor in the Department of Pathology, Harvard Medical School, and a Senior Investigator at the CBR Institute for Blood Research. He serves as director of the Joint Program in Transfusion Medicine, with responsibility for the blood and tissue programs at Boston Children’s Hospital, the Brigham and Women’s Hospitals, and the Dana-Farber Cancer Institute. Dr. Silberstein has recently created the Center for Human Cell Therapy at Harvard Medical School. The goal of this innovative center related to transfusion medicine is to facilitate the translation of proof-of-principle discoveries to clinical applications. Dr. Silberstein is editor of several texts, including Hematology and the Handbook of Transfusion Medicine. He is a member of the editorial boards of Blood and Transfusion. Dr. Silberstein is a highly respected physician-scientist well known for his mentorship; he has trained more than 45 fellows with PhD and MD backgrounds in transfusion medicine-related research. A leader and expert in transfusion medicine and hematology, Dr. Silberstein’s research has focused on the immunology of B-cells and hematopoiesis, leading to the publication of more than 75 papers and numerous book chapters and reviews. Dr. Silberstein is board certified in Transfusion Medicine, Hematology, and Internal Medicine. He received his Baccalaureate and MD degrees from the University of Leiden, the Netherlands, and accomplished postgraduate training in Hematology/Oncology and Transfusion Medicine at Tufts–New England Medical Center in Boston.

ABOUT THE EDITORS

vi

Dr. Ness is director of the Transfusion Medicine Division at The Johns Hopkins Hospital and professor of Pathology, Medicine, and Oncology at The Johns Hopkins University School of Medicine. For many years he also acted as CEO and medical director of the Greater Chesapeake and Potomac Region of the American Red Cross Blood Services. Dr. Ness has served the AABB for a number of years and was President in 1999. He served on the editorial board of Transfusion until named Editor in 2003. Dr. Ness has been a member of the American Society of Clinical Pathologists Board of Registry Blood Bank examination committee, and the FDA Blood Products Advisory Committee, and he consults for many commercial and nonprofit organizations. He is the editor of several textbooks on transfusion medicine and has published more than 150 articles. Dr. Ness’ research focuses on transfusion-related complications and has been funded by the NIH and CDC. He was involved in the initial REDS program and now acts as consultant to REDS-II. He serves as principal investigator for the Johns Hopkins site of the Transfusion Medicine/ Hemostasis Clinical Trial Network, funded by NHLBI. Dr. Ness is co-principal investigator of a project funded by the REDS-II program to study donor virus epidemiology issues in China. He has worked extensively in international blood safety initiatives in China, Thailand, Vietnam, Botswana, and Nigeria. Dr. Ness received his undergraduate education at the Massachusetts Institute of Technology and his MD degree from the State University of New York at Buffalo. His postgraduate work includes residency in internal medicine at Johns Hopkins, fellowship training in hematology-oncology at the University of California, San Francisco, and a transfusion medicine fellowship at Irwin Memorial Blood Bank in San Francisco.

Dr. Anderson is the Kraft Family Professor of Medicine at Harvard Medical School and serves as chief of the Division of Hematologic Neoplasia, director of the Jerome Lipper Multiple Myeloma Center, and vice chair of the Joint Program in Transfusion Medicine at Dana-Farber Cancer Institute. Currently, Dr. Anderson is chair of the NCCN Multiple Myeloma Clinical Practice Guidelines Committee, is a Cancer and Leukemia Group B Principal Investigator, and is on the Board of Scientific Advisors of the International Myeloma Foundation. He has published more than 300 original articles and 200 book chapters, and has edited multiple textbooks on multiple myeloma and transfusion medicine. He is a Doris Duke Distinguished Clinical Research Scientist and has had long-term RO1, PO1, and SPORE funding from the NIH and other agencies. Dr. Anderson has received numerous awards, including the 2001 Charles C. Lund Award of the American Red Cross Blood Services, the 2003 Waldenstrom’s award for research in plasma cell dyscrasias, the 2004 Johnson & Johnson Focused Giving Award for Setting New Directions in Science and Technology, and the 2005 Robert A. Kyle Lifetime Achievement Award. Dr. Anderson graduated from Johns Hopkins Medical School, trained in internal medicine at Johns Hopkins Hospital, and completed hematology, medical oncology, and tumor immunology fellowships at the Dana-Farber Cancer Institute.

Dr. Roback is a tenured associate professor in the Department of Pathology and Laboratory Medicine at Emory University, associate director of the Emory Transfusion Medicine Program, and co-director of the Emory University Hospital Blood Bank and Stem Cell Processing Laboratory. Dr. Roback’s research focuses on human and animal models of CMV infection, emphasizing approaches to accelerate and improve the antiviral immune response following hematopoietic stem cell transplantation. He also is inventor or co-inventor of a number of novel devices and methodologies for rapid pretransfusion blood testing. Dr. Roback’s investigations have been funded by the NIH, CDC, NBF, and DOD. He is a co-principal investigator of the Emory site for REDS-II. Dr. Roback has authored 40 peer-reviewed publications and invited reviews, as well as 16 book chapters. He teaches medical, residency, and graduate school courses and was recognized for excellent clinical pathology teaching with the Golden Apple Award. An active member of a number of AABB committees, Dr. Roback is editor-in-chief of the 16th edition of the AABB’s Technical Manual, member of the editorial board for the journal Transfusion, and co-chair of the NHLBI’s Global Blood Safety and Availability task force on future transfusion medicine research initiatives. He is a Diplomate of the American Board of Pathology in Clinical Pathology and Blood Banking and Transfusion Medicine. Dr. Roback received his Baccalaureate degree from Johns Hopkins University and was awarded a PhD in experimental pathology and an MD from the University of Chicago. He completed a postdoctoral research fellowship and anatomic pathology residency training at Albert Einstein College of Medicine and subsequently completed clinical pathology and transfusion medicine training at Emory University.

Contributors

Sharon Adams, MT, CHS (ABHI)

Richard J. Benjamin, MS, MBChB, PhD

Supervisor, HLA Laboratory Department of Transfusion Medicine Warren G. Magnuson Clinical Center National Institutes of Health Bethesda, Maryland, USA

Chief Medical Officer American Red Cross Biomedical Services National Headquarters, Washington, D.C. Assistant Professor of Pathology Joint Program in Transfusion Medicine Harvard Medical School Boston, Massachusetts, USA

Barbara Alving, MD, MACP Professor of Medicine Uniformed Services University of the Health Sciences Bethesda, Maryland, USA

Kenneth C. Anderson, MD Chief, Division Hematologic Neoplasia Director, Jerome Lipper Multiple Myeloma Center Dana-Farber Cancer Institute Kraft Family Professor of Medicine Harvard Medical School Boston, Massachusetts

James P. AuBuchon, MD E. Elizabeth French Professor and Chair of Pathology Dartmouth-Hitchcock Medical Center Lebanon, New Hampshire, USA

Nicholas Bandarenko, MD Associate Professor of Pathology and Laboratory Medicine Transfusion Medicine Service University of North Carolina, Chapel Hill Chapel Hill, North Carolina, USA

Jon Barrett, MBBCh, FRCOG, MD, FRCSC Associate Professor, Department of Obstetrics and Gynecology, University of Toronto Senior Investigator Maternal and Infant Research Unit of Center for Research in Women’s Health Chief of Maternal Fetal Medicine, Sunnybrook and Women’s College Health Sciences Center Toronto, Ontario, Canada

Howard Benn, MD Chief Fellow, Department of Hematology/ Oncology Seton Hall University School of Graduate Medical Education South Orange, New Jersey, USA

Ginine M. Beyer, MD Department of Pathology University of Maryland School of Medicine Baltimore, Maryland, USA

Morris A. Blajchman, MD, FRCP Professor, Pathology and Molecular Medicine Head, Transfusion Medicine Services, Hamilton Regional Laboratory Medical Director, Canadian Blood Services Hamilton, Ontario, Canada

Neil Blumberg, MD Director, Clinical Laboratories Director, Transfusion Medicine Professor of Pathology and Laboratory Medicine, University of Rochester School of Medicine Rochester, New York, USA

Mark E. Brecher, MD Professor, Department of Pathology and Laboratory Medicine Director, Clinical Pathology University of North Carolina Chapel Hill, North Carolina, USA

vii

CONTRIBUTORS

Hal E. Broxmeyer, PhD Distinguished Professor, Chairman and Mary Margaret Walther Professor of Microbiology and Immunology Professor of Medicine Scientific Director of the Walther Oncology Center Indiana University School of Medicine Indianapolis, Indiana, USA

Chief Medical Officer and Vice President, Research and Medical Affairs Cerus Corporation Concord, California, USA

Robert L. Crookes, MBChB Medical Director South African National Blood Service Johannesburg, South Africa

Michael P. Busch, MD, PhD Vice President, Research, Blood Systems, Inc., Scottsdale, Arizona Director, Blood Systems Research Institute, San Francisco, California Adjunct Professor, Department of Laboratory Medicine University of California, San Francisco, California, USA

Jeannie L. Callum, BA, MD, FRCPC Assistant Professor, Department of Laboratory Medicine and Pathobiology University of Toronto Director, Blood and Tissue Banks Sunnybrook and Women’s College Health Sciences Center Toronto, Ontario, Canada

Sally A. Campbell-Lee, MD viii

Assistant Professor, Department of Pathology Associate Medical Director, Division of Transfusion Medicine Medical Director, Johns Hopkins Bayview Transfusion Medicine Service Baltimore, Maryland, USA

Jeffrey L. Carson, MD

Elizabeth E. Culler, MD Medical Director Blood Assurance, Inc. Chattanooga, Tennessee, USA

Melody J. Cunningham, MD Assistant Professor of Pediatrics Harvard Medical School Children’s Hospital Boston Boston, Massachusetts, USA

Richard J. Davey, MD Director, Transfusion Service The Methodist Hospital Houston, Texas, USA

Dana V. Devine, PhD Professor of Pathology and Laboratory Medicine Centre for Blood Research University of British Columbia Executive Director, Research and Development Canadian Blood Services Vancouver, British Columbia, Canada

Richard C. Reynolds Professor of Medicine Chief, Division of General Internal Medicine University of Medicine and Dentistry, New Jersey Robert Wood Johnson Medical School New Brunswick, New Jersey, USA

Roger Y. Dodd, PhD

Kenneth A. Clark, MD, MPH

Alexander Duncan, MD

Head, International Blood Safety Global AIDS Program Centers for Disease Control and Prevention Atlanta, Georgia, USA

Assistant Professor, Department of Pathology and Laboratory Medicine Emory University School of Medicine Director, Coagulation Laboratories Atlanta, Georgia, USA

Laurence Corash, MD Professor, Laboratory Medicine, University of California, San Francisco Attending Physician, Laboratory Medicine and Medicine-Hematology Division The Medical Center at the University of California, San Francisco

Vice President, Research and Development Director, Holland Laboratory American Red Cross, Biomedical Services Rockville, Maryland, USA

Walter H. Dzik, MD Co-Director, Blood Transfusion Service Massachusetts General Hospital Associate Professor of Pathology Harvard Medical School Boston, Massachusetts, USA

Joanna M. Heal, MRCP, MBBS

Director, Georgia Comprehensive Sickle Cell Center Grady Health System Professor of Hematology/Oncology and Medicine Winship Cancer Institute Emory University School of Medicine Atlanta, Georgia, USA

Associate Medical Director, American Red Cross Blood Services, New York-Penn Region Associate Clinical Professor of Medicine, Hematology-Oncology Unit University of Rochester School of Medicine Rochester, New York, USA

A. Bradley Eisenbrey, MD, PhD

Paul C. Hébert, MD, FRCPC, MHSc(Epid)

Chief, Transfusion Medicine Services William Beaumont Hospital, Royal Oak, Michigan HLA Laboratory Associate Director Gift of Life of Michigan, Ann Arbor, Michigan Assistant Professor of Pathology Wayne State University School of Medicine Detroit, Michigan, USA

Eberhard W. Fiebig, MD Associate Professor, Department of Laboratory Medicine Chief, Divisions of Hematology and Transfusion Medicine University of California San Francisco, California, USA

CONTRIBUTORS

James R. Eckman, MD

Vice-Chair, Department of Medicine Professor of Medicine and Epidemiology Chair in Transfusion and Critical Care Research Ottawa Health Research Institute and the University of Ottawa Ottawa, Ontario, Canada

Nancy Heddle, MSc, FCSMLS(D) Director, McMaster Transfusion Research Program Associate Professor, Department of Medicine McMaster University Hamilton, Ontario, Canada

John R. Hess, MD, MPH John M. Fisk, MD Clinical Instructor, Laboratory Medicine SUNY Upstate Medical University College of Medicine Assistant Director, Transfusion Medicine University Hospital of the State University of New York Syracuse, New York, USA

Terrence L. Geiger, MD, PhD Assistant Professor, Department of Pathology St. Jude Children’s Research Hospital Memphis, Tennessee, USA

Professor of Pathology and Medicine University of Maryland School of Medicine Baltimore, Maryland, USA

Christopher D. Hillyer, MD Director, Transfusion Medicine Program Professor, Department of Pathology and Laboratory Medicine and the Division of Hematology/Oncology, Winship Cancer Institute Emory University School of Medicine Atlanta, Georgia, USA

Krista L. Hillyer, MD Executive Medical Director Donor and Transplantation Services Canadian Blood Services Ottawa, Ontario, Canada

Chief Medical Officer, American Red Cross Blood Services, Southern Region Assistant Professor, Department of Pathology and Laboratory Medicine Emory University School of Medicine Atlanta, Georgia, USA

Shealynn B. Harris, MD

Paul V. Holland, MD

Assistant Medical Director American Red Cross Blood Services, Southern Region Atlanta, Georgia, USA

Clinical Professor of Medicine and Pathology UC Davis Medical Center, Sacramento, California Scientific Director, Delta Blood Bank Stockton, California, USA

Mindy Goldman, MD, FRCP(C)

ix

CONTRIBUTORS

Kim A. Janatpour, MD

Thomas J. Kunicki, PhD

Assistant Professor, University of California Davis School of Medicine, Department of Pathology and Laboratory Medicine Davis Medical Center Sacramento, California, USA

Associate Professor Division of Experimental Hemostasis and Thrombosis Department of Molecular and Experimental Medicine The Scripps Research Institute La Jolla, California, USA

Viviana V. Johnson, MD Transfusion Medicine Fellow Department of Pathology Georgetown University Hospital Washington, D.C., USA

Tzong-Hae Lee, MD, PhD Director, Molecular Biology Blood Systems Research Institute San Francisco, California, USA

Cassandra D. Josephson, MD Assistant Professor, Departments of Pathology and Pediatrics Emory University School of Medicine Assistant Director, Blood Banks and Transfusion Services Attending Pediatric Hematologist/Oncologist, Department of Pediatrics Children’s Healthcare of Atlanta Atlanta, Georgia, USA

Richard M. Kaufman, MD Medical Director, Adult Transfusion Service, Brigham and Women’s Hospital Assistant Professor of Pathology, Harvard Medical School Boston, Massachusetts, USA x

Thomas S. Kickler, MD Professor of Medicine and Pathology Johns Hopkins University School of Medicine Baltimore, Maryland, USA

Diane Killion, JD Staff Counsel AABB Bethesda, Maryland, USA

Karen Shoos Lipton, JD Chief Executive Officer AABB Bethesda, Maryland, USA

Lennart E. Lögdberg, MD, PhD Associate Professor, Department of Pathology and Laboratory Medicine Director, Crawford W. Long Hospital Transfusion Services Emory University School of Medicine Atlanta, Georgia, USA

Naomi L. C. Luban, MD Interim Executive Director, Center for Cancer and Blood Disorders Chair, Laboratory Medicine and Pathology Director, Transfusion Medicine/The Edward J. Miller Donor Center Vice Chair for Academic Affairs, Department of Pediatrics Children’s National Medical Center Professor, Pediatrics and Pathology The George Washington University Medical Center Washington, D.C., USA

Catherine S. Manno, MD Associate Medical Director Transfusion Medicine Division Johns Hopkins University School of Medicine Baltimore, Maryland, USA

Professor and Associate Chair for Clinical Affairs Department of Pediatrics Children’s Hospital of Philadelphia University of Pennsylvania School of Medicine Philadelphia, Pennsylvania, USA

Steven H. Kleinman, MD

Simon Mantha, MD

Kleinman Biomedical Research Victoria, British Columbia, Canada University of British Columbia Vancouver, British Columbia, Canada

Department of Laboratory Medicine Yale University School of Medicine Yale–New Haven Hospital New Haven, Connecticut, USA

Karen E. King, MD

Patricia T. Pisciotto, MD

Director, HLA and Immunogenetics Research Laboratory Department of Transfusion Medicine Warren G. Magnuson Clinical Center National Institutes of Health Bethesda, Maryland, USA

Professor, Laboratory Medicine University of Connecticut Health Sciences Center Director, Blood Bank, John Dempsey Hospital Farmington, Connecticut, USA

CONTRIBUTORS

Francesco M. Marincola, MD

Thomas H. Price, MD Bruce C. McLeod, MD Professor of Medicine and Pathology Director, Blood Center Rush University Medical Center Chicago, Illinois, USA

Executive Vice-President, Medical Division Medical Director, Puget Sound Blood Center Professor of Medicine, University of Washington Seattle, Washington, USA

Jayashree Ramasethu, MD Jay E. Menitove, MD Clinical Professor, Internal Medicine University of Kansas School of Medicine Kansas City, Kansas; Executive Director and Medical Director, Community Blood Center of Greater Kansas City Kansas City, Missouri, USA

Peter A. Millward, MD Assistant Professor, Clinical Pathology Milton S. Hershey Medical Center Pennsylvania State University Hershey, Pennsylvania, USA

Edward L. Murphy, MD, MPH Professor, Laboratory Medicine and Epidemiology/ Biostatistics University of California, San Francisco San Francisco, California, USA

Associate Professor of Clinical Pediatrics Division of Neonatology Department of Pediatrics Georgetown University Hospital Washington, D.C., USA

Sandra M. Ramirez-Arcos, MSc, PhD Associate Scientist, Canadian Blood Services Adjunct Professor, University of Ottawa Research and Development, Infectious Diseases Ottawa, Ontario, Canada

William Reed, MD Assistant Medical Director, Research Blood Systems Research Institute Clinical Associate Professor Department of Laboratory Medicine Medical Director, Human Islet and Cellular Therapy Laboratory University of California San Francisco, California, USA

Paul M. Ness, MD Director, Transfusion Medicine Division Johns Hopkins Medical Institutions Professor, Pathology, Medicine, and Oncology Johns Hopkins University School of Medicine Baltimore, Maryland, USA

Diane J. Nugent, MD Assistant Professor David Geffen School of Medicine at UCLA University of California, Los Angeles Director, Division of Hematology Children’s Hospital of Orange County General Pediatrics, Irvine Medical Center Los Angeles, California, USA

Peter L. Perrotta, MD Associate Professor of Pathology West Virginia University Morgantown, West Virginia, USA

Marion E. Reid, PhD Director, Immunohematology New York Blood Center New York, New York, USA

John D. Roback, MD, PhD Co-Director, Transfusion Medicine Program Associate Professor, Department of Pathology and Laboratory Medicine Emory University School of Medicine Atlanta, Georgia, USA

Scott D. Rowley, MD, FACP Chief, Adult Blood and Marrow Transplantation Program Hackensack University Medical Center Hackensack, New Jersey, USA

xi

CONTRIBUTORS

S. Gerald Sandler, MD

Leslie E. Silberstein, MD

Professor of Medicine and Pathology Georgetown University School of Medicine Director, Transfusion Medicine Department of Laboratory Medicine Georgetown University Hospital Washington, D.C., USA

Director, Joint Program in Transfusion Medicine Children’s Hospital Boston, Dana-Farber Cancer Institute, Brigham and Women’s Hospital Professor of Pathology, Harvard Medical School Boston, Massachusetts, USA

Audrey N. Schuetz, MD

Steven R. Sloan, MD, PhD

Department of Pathology and Laboratory Medicine Emory University School of Medicine Atlanta, Georgia, USA

Assistant Professor of Pathology Pediatrics Joint Program in Transfusion Medicine Harvard Medical School Boston, Massachusetts, USA

Eileen Selogie, MT(ASCP)SBB

Edward L. Snyder, MD

Consultant Compliance Officer Department of Pathology, Blood Donor Services Presbyterian Intercommunity Hospital Whittier, California, USA

Professor, Laboratory Medicine Yale University School of Medicine Director, Blood Bank, Yale–New Haven Hospital New Haven, Connecticut, USA

Beth Shaz, MD

Ronald G. Strauss, MD

Assistant Professor, Emory University School of Medicine Department of Pathology and Laboratory Medicine Director, Grady Memorial Hospital Blood Bank Atlanta, Georgia, USA

Professor of Pathology and Pediatrics University of Iowa College of Medicine Iowa City, Iowa, USA

David F. Stroncek, MD Technical Specialist, Transfusion Medicine Division, Johns Hopkins Hospital Baltimore, Maryland, USA

Chief, Laboratory Services Section Department of Transfusion Medicine Warren G. Magnuson Clinical Center National Institutes of Health Bethesda, Maryland, USA

Ira A. Shulman, MD

D. Michael Strong, PhD, BCLD(ABB)

Director of Transfusion Medicine Professor and Vice Chair of Pathology Keck School of Medicine of the University of Southern California Director of Laboratories and Pathology, LAC + USC Medical Center Los Angeles, California, USA

Executive Vice President, COO Puget Sound Blood Center Research Professor Department of Orthopaedics and Sports Medicine Department of Surgery University of Washington School of Medicine Seattle, Washington, USA

Suzanne Shusterman, MD

Leon L. Su, MD

Instructor, Harvard Medical School Department of Pediatric Oncology Dana-Farber Cancer Institute Children’s Hospital Boston Boston, Massachusetts, USA

Associate Medical Director Blood Systems, Inc. Assistant Medical Director United Blood Services, Arizona Scottsdale, Arizona, USA

R. Sue Shirey, MD, MT(ASCP) SBB xii

Associate Professor of Pathology and Medicine Director, Transfusion Medicine Service Dartmouth-Hitchcock Medical Center Lebanon, New Hampshire, USA

Gary E. Tegtmeier, PhD Scientific Director Community Blood Center of Greater Kansas City Kansas City, Missouri, USA

Alan Tinmouth, MD, FRCPC, MSc

Connie M. Westhoff, SBB, PhD Scientific Director, Molecular Blood Group and Platelet Antigen Testing Laboratory American Red Cross Adjunct Assistant Professor, University of Pennsylvania Department of Pathology and Laboratory Medicine Philadelphia, Pennsylvania, USA

CONTRIBUTORS

Zbigniew M. Szczepiorkowski, MD, PhD

Robert M. Winslow, MD President, Chairman and CEO, Sangart, Inc. Adjunct Professor, Department of Bioengineering University of California, San Diego San Diego, California, USA

Director, Adult Hemophilia and Bleeding Disorders Comprehensive Care Program Assistant Professor of Medicine Associate Scientist, Center for Transfusion Research Ottawa Health Research Institute and the University of Ottawa Ottawa, Ontario, Canada

Edward C. C. Wong, MD

Ena Wang, MD

Gary Zeger, MD

Staff Scientist, Immunogenetics Research Laboratory Department of Transfusion Medicine Warren G. Magnuson Clinical Center National Institutes of Health Bethesda, Maryland, USA

Associate Professor, Keck USC School of Medicine Co-Medical Director, USC University Hospital Clinical Laboratories Medical Director, Blood Bank, USC University Hospital Medical Director, USC Blood Donor Center Los Angeles, California, USA

Assistant Professor of Pediatrics and Pathology Department of Laboratory Medicine George Washington School of Medicine Director of Hematology, Associate Director of Transfusion Medicine Children’s National Medical Center Washington, D.C., USA

Kathryn E. Webert, MD, FRCPC

James C. Zimring, MD, PhD

Assistant Professor, Department of Medicine McMaster University Medical Consultant, Canadian Blood Services Hamilton, Ontario, Canada

Assistant Professor, Transfusion Medicine Program Department of Pathology and Laboratory Medicine Emory University School of Medicine Atlanta, Georgia, USA

xiii

Preface to the Second Edition

The editors are pleased to introduce the Second Edition of Blood Banking and Transfusion Medicine. Substantial modifications and additions have been made to the text, reflecting advancements in a number of areas, including cellular therapy, component preparation, infectious disease testing, and the underlying biology of transfusion therapy. In addition, we have continued to integrate elements of Anderson and Ness’s excellent textbook The Scientific Basis of Transfusion Medicine, which can be noted by the reader as a number of new chapters entitled “Principles of . . . .” We are grateful for the many suggestions offered by readers of the First Edition that led to additional improvements in the

text. We have made a concerted effort to ensure that each chapter includes the most up-to-date scientific underpinnings of transfusion biology as well as detailed information that can be applied to clinical transfusion practice. It is our goal that this textbook remain the definitive source of blood banking and transfusion medicine biology, technology, and practice for physicians, technologists, nurses, and administrative personnel, and we sincerely welcome readers’ observations, criticisms, and suggestions so that we can continue to work to improve this book. Finally, we thank you for your support of this text, the field of transfusion medicine, and the patients we serve. C. D. Hillyer L. E. Silberstein P. M. Ness K. C. Anderson J. D. Roback xv

Acknowledgments

We, the editors, would like to acknowledge the outstanding technical and professional support of Sue Rollins and the expertise, guidance, and friendship of Dolores Meloni. We would also like to thank our friends and families for their unconditional love and support, without which this edition could not have come to fruition. We thank especially Krista, Whitney, Peter, Margot,

Jackson, and James Hillyer; the family and friends of Les Silberstein; Barbara, Jennie, Steven, and Molly Ness; Cynthia, Emily, David, and Peter Anderson; and Linda, Evan, and Ethan Roback. Finally, we would like to acknowledge and thank the many mentors, physicians, and patients who have served as inspiration, colleagues, and friends.

xvii

Chapter 1

A Brief History of Blood Transfusion Kim A. Janatpour

●

Paul V. Holland

EARLY HISTORY Since the beginning of human history, blood has been recognized as a vital force, the essence of life. Prehistoric man created cave drawings showing individuals bleeding from traumatic wounds. In the Bible, Leviticus states “the life of the flesh is in the blood.” The Chinese Huang Di Nei Ching (770–221 bc) held that blood contained the soul. Blood played a central theme in ancient rituals. Egyptians and Romans took blood baths for physical and spiritual restoration,1 and Romans even drank the blood of fallen gladiators in the belief that the blood could transmit the gladiator’s vitality. Precolumbian North American Indians bled the body “of its greatest power” as self-punishment. In the Middle Ages, the drinking of blood was advocated as a tonic for rejuvenation and for the treatment of various diseases.2 Pope Innocent VIII drank the blood from three young boys in 1492. Unfortunately, the boys and the Pope died.2 The idea that infusion of blood could be beneficial did not emerge until the 17th century. From the time of Hippocrates (c. 450 bc), disease was believed to be caused by an imbalance of the four humours—blood, phlegm, yellow bile, and black bile. Of these, blood was the most important (Galen [130–201 ad] really advanced the humoural theory). The most popular treatment for most ailments, even as late as the 18th century, was blood letting (Fig. 1–1). Without the correct understanding of blood circulation, intravenous blood infusion could not even be imagined. This changed in 1628 with William Harvey’s description of the circulatory system. Harvey’s identification of separate yet connected arterial and venous systems in his De Motu Cordis paved the way for an entirely new arena of blood investigation.3 In 1656, Christopher Wren used a quill with an attached bladder to demonstrate that the intravenous injection of substances into animals had systemic effects.1,2 In 1666, Richard Lower successfully transfused blood from one dog to another, which led Samuel Pepys to speculate on the potential benefits of human transfusion, stating that “bad blood” might be mended by “borrowing” blood “from a better body.”3

THE FIRST ANIMAL-TO-HUMAN TRANSFUSIONS The first published animal-to-human transfusion was performed June 15, 1667, by Jean Baptiste Denis, a physician to Louis XIV, on a 16-year-old boy who had been “tormented

with a contumacious and violent fever.” The boy had been treated with multiple bleeds, following which “his wit seemed wholly sunk, his memory perfectly soft, and his body so heavy and drowsie that he was not fit for any thing.” Denis attributed these symptoms to the bloodletting he had received. As treatment, Denis exchanged 3 ounces of the boy’s blood for 9 ounces of lamb’s blood. Denis chose animal blood because he believed it purer than that of humans due to man’s “debauchery and irregularities in eating and drinking” and reasoned that if man could use animal milk as nutrient, animal blood would be safe. Following the infusion of lamb’s blood, the patient complained about “a great heat along his arm,” but otherwise suffered no ill effects. Denis subsequently performed such transfusions on three more patients, the last of which resulted in the first malpractice suit for blood transfusion.4 Antoine Mauroy was a 34-year-old madman who was brought to Denis after he was found wandering the streets of Paris in the winter of 1667. Mauroy had suffered for years from severe “phrensies,” during which he would beat his wife, strip off his clothes, and run through the streets, setting house fires. At this time, blood was believed to affect one’s temperament and character; therefore, it was reasoned that blood transfusion could be used to treat mental ailments. Denis’s patron, Monsieur de Montmort, proposed transfusing Mauroy to allay the “heat of his blood.”5 Denis transfused Mauroy with calf ’s blood, hoping that the calf ’s docile nature would be imparted to Mauroy. Although the patient complained of heat moving up his arm, he tolerated the transfusion well. A few days later, a second, larger transfusion was performed. This time, however, the patient complained “of great pains in his kidneys, and that he was not well in his stomack, that he was ready to [choak] unless they gave him his liberty.”6 The transfusion was quickly discontinued, after which the patient vomited and passed urine “black as soot.” Miraculously, the patient not only survived this hemolytic transfusion reaction, but also appeared to be cured, showing “a surprising calmness, and a great presence of mind … and a general lassitude in all his limbs.” In fact, upon seeing his wife a few days later, Mauroy greeted her tenderly, relating “with great presence of mind all that had befallen him.” Denis was astonished—the man who “used to do nothing but swear and beat his wife” had dramatically, almost magically, been cured.7 Also, later in 1667, Richard Lower successfully transfused a Cambridge University student described as “cracked a little in the head” with sheep’s blood.3,4 A bitter debate followed between Denis and Lower as to who could claim to have discovered blood transfusion.4

1 3

HISTORY

Figure 1–1 A collection of bloodletting instruments. (From Star D. Blood. An epic history of medicine and commerce. New York, HarperCollins Publishers, 2002. With permission.)

I 4

Although a select group of scientists was excited about the concept of transfusion, others were adamantly opposed to the practice. Denis, in particular, suffered harsh criticism from his peers. With this intense debate and criticism as the backdrop, Mauroy suffered a relapse; his wife begged Denis to transfuse her husband again. Denis found the patient to be very ill, so was hesitant to perform the transfusion, but reluctantly agreed. Before the transfusion began, however, Mauroy died and his widow refused to allow Denis to examine the body. The widow had been offered money from Denis’s rivals to charge him with murder; she offered to drop the matter if Denis would agree to support her financially. Denis refused, and the case went to court. Denis was exonerated when it was discovered that Mauroy had been poisoned with arsenic by his wife. Nonetheless, although Denis was acquitted of malpractice, the general opposition to transfusion ultimately led the French and English courts, and much of the rest of Europe, to ban all human transfusions.1,4,5,7,8

FIRST HUMAN-TO-HUMAN TRANSFUSION After being banned for more than 150 years, the use of blood transfusion was revived during the late 18th century. A footnote in an American journal indicates that the first humanto-human transfusion had been performed by Philip Syng Physick, the “Father of American Surgery,” in 1795, although this has never been confirmed.5,9 In 1816, John Henry Leacock, a Barbados physician, presented his dissertation “On the Transfusion of Blood in Extreme Cases of Haemorrhage.” Leacock subsequently performed and published a set of animal experiments that proved that the donor and recipient must be of the same species.10 Although Leacock apparently went no further with the experiments, his work inspired James Blundell, an obstetrician and physiologist at Guy’s Hospital in London, to carry out additional investigations. At the time, obstetricians could only stand by and watch helplessly as patients exsanguinated postpartum. Blundell was convinced that blood transfusion

DISCOVERY OF ABO BLOOD GROUPS Before 1901, the prevailing belief was that all human blood was the same. However, this changed in 1901 with Karl Landsteiner’s landmark discovery of ABO blood groups.12 Landsteiner, an Austrian immunologist, noticed that human blood mixed in test tubes with other specimens of human blood sometimes resulted in agglutination. By incubating red cells from some individuals with serum from others, he identified agglutination patterns, leading to the initial identification of three blood groups, A, B, and C (C was later renamed O).3,13 In 1902, Alfred Decastello and Adriano Sturli, two of Landsteiner’s former students, found the fourth blood group, AB.3 Landsteiner also contributed to forensic science by developing a method for blood typing of dried blood specimens.14 Interestingly, the importance of the blood groups was not immediately recognized; blood group typing did not become part of routine practice for several years. Richard Weil, a pathologist at the German Hospital in New York, was the first to perform ABO typing and began compatibility testing in 1907; he was also the first to suggest inheritance of ABO types.5 Also in 1907 and 1910, respectively, Jan Jansky of Czechoslovakia and Moss of the United States independently identified four human blood groups.3 However, the Roman numeral systems that Jansky and Moss each used for designating the four blood groups were completely reversed. Tremendous confusion ensued with the three different nomenclatures. Finally, in 1927, the American Association of Immunologists adopted a new classification scheme proposed by Landsteiner, the current ABO terminology.3

The discovery of blood groups led Ludvig Hektoen of Chicago to advocate selecting donors by blood group and crossmatching.8 In 1913, Reuben Ottenberg conclusively demonstrated the importance of compatibility testing in his report of 128 cases of transfusion.15 However, even as recently as 1937, some suggested that crossmatching was unnecessary if the selection of donors was restricted to individuals of the same blood group.5 The inheritance pattern of blood groups was finally proved by Felix Bernstein in 1924.3 Sadly, differences in race distribution of blood groups were manipulated and misused in Germany during World War I (WWI) and World War II (WWII), during which time blood group B was deemed a marker for Slavic or Jewish race, and blood group A was considered associated with intelligence and industry. In the 1950s in Louisiana, it was a misdemeanor for a physician to give blood from a black donor to a white person without consent. In the United States, segregation of blood by race existed until the 1960s.3

A BRIEF HISTORY OF BLOOD TRANSFUSION

could save patients’ lives. His extensive experimentation confirmed Leacock’s findings that blood could be used to treat hemorrhagic shock, but only blood from the same species could be used. Recognizing the potentially serious risks of transfusion, Blundell began attempting human-to-human transfusion in cases that were otherwise hopeless. Over a decade, he performed 10 such transfusions, all without success. However, in August 1825, Blundell successfully transfused a woman dying from postpartum hemorrhage with blood from her husband. Other successes followed, including three cases of postpartum hemorrhage, and a young boy who was hypovolemic following amputation of his leg.11 Subsequently, other reports of transfusion followed from Europe and then the United States, where it was reported that transfusion was used by the Union Army during the American Civil War.5,9 Significant progress in understanding the basis for the incompatibility between species was made by Emil Ponfick and Leonard Landois in the late 1800s.8 The first revelation came from Ponfick, who observed red cell lysis in the blood of a woman who died after receiving a transfusion of sheep blood. From animal experiments, Ponfick found that incompatible transfusions were associated with hemorrhage and “congestion” of the kidneys, lungs, and liver. He also recognized that the red urine that transfused animals excreted was caused by hemoglobinuria, not hematuria. Landois’s observation that human red cells would lyse when mixed in vitro with the sera of other animals set the stage for the study of the immunologic basis of blood incompatibility.8

DISCOVERY OF RH BLOOD GROUPS Although a major discovery in transfusion medicine, ABO blood group typing was not sufficient to prevent many fatal hemolytic transfusion reactions. In 1939 Philip Levine published a case report of post-transfusion hemolysis in a blood group O patient who received blood from her blood group O husband. Levine found that incubation of the patient’s serum with her husband’s red cells resulted in agglutination. Additionally, the woman’s serum was found to agglutinate 80 of 104 other samples of ABO-compatible blood. The name of the offending antibody came from parallel experiments conducted by Landsteiner and Alex Wiener in which antibodies produced by immunization of rabbits and guinea pigs with blood from rhesus monkeys caused red cell agglutination of 85% of humans tested. Those individuals whose red cells were agglutinated by these antibodies were classified as rhesus (Rh) positive.3 Levine was able to show that Rh antibodies were the main cause of serious hemolytic disease of the newborn (erythroblastosis fetalis).16 Later, it was appreciated that the Rh system is composed of numerous alleles. The current system of nomenclature—c, C, d, D, e, E—was proposed in 1944 by Cambridge geneticist Sir Ronald Fisher. Subsequent development of Rh immune globulin (RhIG) for prevention of hemolytic disease of the newborn was a major advance. The use of the antiglobulin test, first described by Carlo Moreschi in 1908 and rediscovered in 1945 by Robin Coombs, Rob Race, and Arthur Mourant, allowed the identification of many other blood group antigens in the decades that followed.3,17

BLOOD COAGULATION, PRESERVATION, AND STORAGE Despite some successes by Blundell and contemporaries, transfusions often failed to save lives, and remained a rarity until the early 20th century. Clotting remained a significant problem. A variety of devices, involving valves, syringes, and tubing, were invented to facilitate the collection and infusion of blood from one individual to another, including two invented by Blundell—the “Gravitator” and the “Impellor.” The impellor consisted of a double-walled funnel in which

1 5

HISTORY

Figure 1–2 A. Blundell’s “impellor.” (Modified from Jones HW, Mackmull G. The influence of James Blundell on the development of blood transfusion. Ann Med Hist 1928;10:242.) B, Sketch of Blundell’s gravitator. (Modified from Blundell J. Observations on transfusion of blood. Lancet 1828;2:321.)

I 6

the outer compartment was filled with warm water. The donor blood flowed into the funnel, was sucked into a syringe, and was forced along tubing into a cannula inserted into the patient’s vein by means of two oppositely acting spring valves below the funnel8 (Fig. 1–2). Gesellius used an equally complex device, in which the donor’s back was lanced multiple times and capillary blood extracted using suction cups5,8 (Fig. 1–3). James Aveling used a simpler method for direct blood transfusion from a donor using two silver cannulae, inserted into the recipient and donor, and connected by rubber tubing with a compressible bulb in the middle to promote and sustain flow.11 The Aveling device is featured in the first known photograph of an actual blood transfusion, taken at Bellevue Hospital in New York City in the 1870s9 (Fig. 1–4). In 1908, Alexis Carrel, a French researcher working at the Rockefeller Institute for Medical Research in New York, perfected a surgical technique for the direct anastomosis of donor artery to recipient vein.3 Although highly effective at providing blood to the patient without clotting, performance of this technique required tremendous skill. Further, it required donors willing to undergo the painful procedure. It was also impossible to accurately estimate the amount of blood passed from donor to recipient; donors often became hypotensive or recipients developed circulatory overload.3

CITRATE ANTICOAGULATION A chemical approach to anticoagulation was first attempted by Braxton Hicks, a 19th-century obstetrician, who experimented with phosphate of soda. Unfortunately, none of the four patients in whom it was used survived.8 Other substances used in anticoagulation attempts included sodium bicarbonate, ammonium oxalate, arsphenamine, sodium iodide, sodium sulfate, and hirudin.8 Surprisingly, these initial attempts did not include sodium citrate, which had long been used in laboratories as an anticoagulant.8 The 1% concentration of citrate commonly used in the laboratory, however, was toxic to humans.3 Nonetheless, in 1914 Albert Hustin reported the first human transfusion using citrated blood.8 In 1915, Richard Lewisohn of the Mount Sinai Hospital in New York proved that a 0.2% sodium citrate solution was effective as an anticoagulant for blood, while having no toxicity even when as much as 2500 mL of citrated blood were transfused.3 Also in 1915, Richard Weil, an American pathologist, found that citrated blood could be refrigerated for several days before use.18 Lewinsohn and Weil, as well as Rous and Turner, found that addition of dextrose to citrate would preserve blood for up to 2 weeks1,5 (Fig. 1–5). This permitted the first transfusion of stored blood in WWI by an American army physician, Oswald Robertson,

A BRIEF HISTORY OF BLOOD TRANSFUSION

Figure 1–3 Collection and transfusion of capillary blood by the method of Gesellius. (Modified from Gesellius F: Die Transfusion des Blutes. Leipzig, E. Hoppe, 1873.)

1 7

Figure 1–5 Lewisohn’s method of transfusion of citrated blood. Blood is collected in a graduated flask (A) and is promptly transfused to the patient (B). (Modified from Lewisohn R: The citrate method of blood transfusion after ten years. Boston Med Surg J 1924;190:733.)

Figure 1–4 Medical and nursing staff administering transfusion, Bellevue Hospital, 1876. Note that the Bellevue staff has placed both needles in the wrong orientation. (From Schmidt PJ. The first photograph of blood transfusion. Transfusion 2001;41:968–969.)

who transfused 20 casualties on 22 occasions during the battle of Cambrai in November 1917. Nine of the 20 recipients lived.19 The primary disadvantages of the Rous-Turner solution were that it was difficult to prepare and required a large volume of preservative solution in relation to the amount of blood. However, it remained the only anticoagulant–preservative solution available through most of WWII.5 Acid citrate dextrose (ACD), developed in 1943 by Loutit and Mollison, allowed for blood to be stored for up to 3 to 4 weeks,20 could be autoclaved, and had the advantage of being easier to prepare,

while requiring a smaller volume of solution relative to the amount of blood. Citrate phosphate dextrose (CPD) solution was subsequently adopted after studies showed blood could be stored for up to 28 days with better red cell survival than ACD.3,21 Long-term red cell preservation by freezing began in 1950, when Smith and colleagues showed that glycerol could prevent freeze–thaw damage.22

ADVENT OF BLOOD BANKS The first blood donor service was established in 1921 by Percy Oliver, Secretary of the Camberwell Division of the British Red Cross.3 At this time, donors generally came from an unreliable supply; most from the patient’s family and friends, or from indigents. Sometimes, no compatible donor could be found. On one such occasion, Oliver was contacted

&:: o

t:n

:r:

•

by a hospital with an urgent request for blood. Oliver and his coworkers rushed to the hospital to see if they could help. Blood from one individual in the group, a nurse, was found to be compatible with the patient. This experience gave Oliver the idea to create a stable supply of potential blood donors. In the program he developed, each potential blood donor underwent a physical examination, blood typing, and testing for syphilis before placement on the volunteer list. Because many physicians were still reluctant to use anticoagulation, the volunteer would go to the hospital and provide direct donation via venous cutdown (rendering the vein useless for subsequent donations). The service was supported entirely by donations, and the services were provided free. In 1922, the donor list consisted of 20 volunteers whose services were requested 13 times. By 1925, the number of requests had risen to 428, and it doubled the following yearY The increasing demand ultimately led to establishment of a new organization, the Greater London Red Cross Blood Transfusion Service? The first true predecessor to the modern blood bank was established in 1935 at the Mayo Clinic." Others credit the first blood bank to Bernard Fantus, who established a blood bank at Cook County Hospital in Chicago in 1937.24 In this latter facility, blood was collected into glass flasks containing sodium citrate, sealed, and stored refrigerated. Pilot tubes were prepared for typing and serology testing. Fantus was the first to coin the phrase "blood bank" for the operation because blood could be stored and saved for future use. 24 During the Spanish Civil War (1936-1939), Federico Duran-Iorda organized a highly successful mobile blood bank that could be transported wherever needed. Every donor was assessed with a questionnaire, physical examination, syphilis testing, and testing for blood type and red cell concentration. Only universal type 0 blood was collected. Using an entirely closed system of his own design, Duran-Iorda collected the blood into glass bottles containing a citrate and glucose solution.i-" Blood was then transported to front-line hospitals in vehicles fitted with refrigerators." At the height of fighting, DuranIorda's blood center in Barcelona was processing up to 75 blood donations per hour? When it became evident that the Nationalists would win the war, Duran-Iorda left Spain for England, where he assisted Janet Vaughan in establishing a blood bank at Hammersmith Hospital in 1938. Because war with Germany was imminent, the Medical Research Council supported Vaughan's proposal to establish four blood depots in London. In 1938, the War Office also created the army blood supply depot under the control of Lionel Whitby.' The Army's policy, to supply blood group 0 red cells at the battlefronts with blood that had been collected centrally rather than collected at the front from troops, proved to be highly successful.'

were eager to help, but knew that whole blood could not survive the long transatlantic journey?

Use of Plasma John Elliott, laboratory chief at Rowan Hospital in North Carolina, had been experimenting with methods of separating plasma from blood when a patient who had been stabbed in the heart presented at the emergency room. Because there was no time to obtain a blood sample for type and crossmatching, Elliott decided to try transfusing the patient with the plasma he had in the laboratory. The patient survived. Elliott found that, in addition to having many of the beneficial properties of whole blood transfusion, plasma retained its usefulness for months. He was convinced that plasma could be the best transfusion liquid available and became a relentless advocate for its widespread transfusion? Elliott's campaign efforts were successful. In August 1940, the Americans launched the Plasma for Britain program, headed by Charles Drew (Fig. 1-6). Drew was an African American surgeon whose 1938 doctoral thesis, "Banked Blood;' was considered the most authoritative work on the science of blood storage at the time." Under Drew's directorship, the Plasma for Britain program was a tremendous success, collecting blood from nearly 15,000 people, which produced 5500 vials of plasma (Fig. 1-7). Drew's contribution, however, extended beyond the sheer number of units collected. He was said to be the first to develop and implement strict procedures for blood collection and testing on a large, "industrial" scale? In 1941, Drew was appointed director of the first American Red Cross Blood Bank, in charge of blood for use by the U.S.Army and Navy.In 1942, however, he resigned his official posts following the armed forces' decision that the blood of African Americans would be accepted but would have to be stored separately from that of whites.V'

TRANSFUSION IN WWII Beginning in September 1940, London was relentlessly assaulted for months by nightly German bombs that caused tens of thousands of deaths and injuries. The efficient organization of the blood depots provided for rapid transfusion of the wounded. Through experience, it became clear that blood transfusion could be used to treat injuries other than hemorrhage, including traumatic shock, crush injuries, fractures, and burns, creating a need for an even greater blood supply than could be provided by Britain alone. Americans

Figure 1-6

Dr. Charles Drew. (From Star D. Blood. An epic history of medicine and commerce. New York, HarperCollins Publishers, 2002.)

of a variety of infectious diseases, including infectious hepatitis (hepatitis A), rubella, and measles, and in the treatment of hypogammaglobulinemia." Of particular significance, the isolation of Rh antibodies paved the way for discovery of methods to prevent most severe cases of hemolytic disease of the fetus and newborn.

» CD

::0:' en

-r-r

:r: V1

--I

o

~

o-r-r CD

c-

IMMUNOGLOBULIN IN THE PREVENTION OF RH-HEMOLYTIC DISEASE OF THE FETUS AND NEWBORN

o o o

--I

~

Z

V1

-r-r C

Figure 1-7 An American medic administers plasma to a wounded soldier in Sicily. (From Star D. Blood. An epic history of medicine and commerce. New York, HarperCollins Publishers, 2002.)

Eleanor Roosevelt took over his position. Drew became a professor of surgery at Howard University and trained numerous black surgeons. He died in an automobile accident in the late 1950s. It was widely alleged he didn't receive blood, because no "black blood" was available. This canard, however, is simply not true. Drew was transfused appropriately, but died as a result of injuries to his great vessels and his heart."

Blood Fractionation The benefits of plasma for treatment of injuries during WWII were evident, but like whole blood, plasma had its limitations. The protein-rich solution was highly prone to bacterial contamination. Freeze-dried plasma circumvented this problem, but was cumbersome and awkward to use on the battlefield. In 1940, Edwin Cohn, a Harvard chemist, isolated various fractions of plasma. Fraction V was found to be composed of albumin, which, in limited clinical studies of volunteers and accident victims, was found to restore circulatory collapse. Professor 1. S. Ravdin at the University of Pennsylvania established albumin's efficacy following the bombing of Pearl Harbor in 1942, where albumin was used to treat injuries of 87 patients? Most of these patients showed some clinical improvement, and only four suffered minor reactions.' Based on this success, the u.s. military added albumin production to efforts to produce plasma on a large scale? Albumin had a number of advantages over plasma. Because of the method of production, cold ethanol precipitation/fractionation, albumin is free of bacteria. In addition, because albumin is highly concentrated, it could be transported in small vials easily on the battlefield. However, production of a single unit of albumin required pooling of multiple blood donations, and, like plasma, production of albumin was difficult. During WWII, both plasma and albumin were produced in vast amounts. By the end of 1943, over 2.5 million packages of dried plasma and nearly 125,000 units of albumin had been sent to the u.s. military? Cohn's fractionation of plasma yielded numerous benefits beyond the purification of albumin. Fraction I was found to contain fibrinogen; fractions II and III contained immunoglobulins that proved effective in the temporary prevention

The first major advance, following Levine's discovery that maternal Rh antibodies were the cause of erythroblastosis fetalis, was the development of a practical procedure for umbilical vein exchange transfusion by a Boston pediatrician, Louis Diamond, in 1947. However, it took another 20 years before a treatment to prevent hemolytic disease of the newborn was discovered.' Initial experiments by Stern and colleagues in 1961 showed that passively administered "incomplete" anti-Rh antibody could interfere with primary Rh immunization.ls-f-" Because immunoglobulin M (IgM) antibodies would not cross the placenta, investigators initially attempted to use "complete" or IgM anti -Rh antibody to block Rh immunization in Rh-negative males, but found that antibody formation was enhanced rather than suppressed." Subsequently, suppression of Rh immunization by"incomplete" IgG anti-Rh antibody was demonstrated by Clarke and coworkers'v" and Freda and coworkers.F-" who were the first to use an immunoglobulin concentrate of IgG anti-Rh antibody given intramuscularly. Shortly afterward, it was realized that transplacental hemorrhage occurred chiefly at the time of delivery, and both groups showed that anti-Rh antibody given soon after delivery would suppress Rh immunization.l'vv"

BLOOD COMPONENTS Another advance in transfusion was development of the first cellseparator in 1951by Edwin Cohn; the cellseparator allowed blood to be separated into red cells, white cells, platelets, and plasma. Although in principle it was possible to harvest any particular component of the blood, in practice, satisfactory yields of only red cells or plasma could be obtained." Cohn's cell separator paved the way for component therapy, but the development of plastic containers made modern component therapy possible. Up until the 1950s, blood was collected through steel needles and rubber tubing into glass, rubber-stoppered bottles, which were reused following washing and sterilization. 1 Pyrogenic reactions and air embolism were known risks to blood collected using these materials. In 1952, Carl Walter, a researcher under Harvey Cushing, and William Murphy described a system in which the blood was collected into a collapsible bag of polyvinyl resin.l-" Plastic had the flexibility to permit the removal of plasma following sedimentation or centrifugation, techniques that became the foundations for component production. 1,37

FACTOR VIII CONCENTRATES The ability to separate blood into components resulted in major advances for the treatment of hemophilia. Initially, hemophiliacs were treated with fresh frozen plasma; however,

V1

oz

•

HISTORY

massive volumes were required to replenish the deficient factor VIII in these patients. A more concentrated form of the required factor VIII was found in the cryoprecipitated portion of plasma by Judith Pool in 1965.38 In 1968 Brinkhous and Shanbrom produced concentrated factor VIII by pooling hundreds to thousands of units of plasma. These factor concentrates could be carried by the patient and administered by self-injection at the earliest sign of bleeding.7,39 Factor VIII concentrates were a major advance in the treatment of hemophilia A, but they came with a very high risk of infectious disease transmission, first with hepatitis and later with human immunodeficiency virus (HIV).

INFECTIOUS DISEASE TRANSMISSION

I 10

The first hint that hepatitis could be caused by blood transfusion came during WWII following administration of yellow fever vaccines produced using human serum as a stabilizer.1 This was followed by a report in 1943 of a series of seven cases of jaundice following transfusion of whole blood or plasma.40 At the time, serum hepatitis (or hepatitis B) was assumed to be the major cause of transfusion-associated hepatitis. This began the awareness that blood transfusion could transmit potentially deadly viral diseases. In 1962, the connection between paying for units and an increased risk of post-transfusion hepatitis was made by J. Garrett Allen, a Stanford surgeon41; however, it wasn’t until a decade later that the National Blood Policy mandated a voluntary (unpaid) donation system in the United States.1 Subsequently, in 1975 transfusion-associated hepatitis was shown to be primarily due to non-A, non-B hepatitis, or what became identified later as hepatitis C.42 Despite the risk of hepatitis transmission, blood utilization continued to increase. In the United States, for example, blood utilization doubled between 1971 and 1980.1 However, the emergence of acquired immunodeficiency syndrome (AIDS) changed this trend. The first case of AIDS was reported in 1981. This mysterious disease initially occurred only in gay men. However, within a few years, it became clear that AIDS was caused by a blood-borne virus. The first reported case of transfusionassociated AIDS occurred in a 20-month-old infant who had received multiple transfusions for hemolytic disease of the newborn.43 Hemophiliacs, dependent on lifelong infusions of factor VIII concentrates, were particularly devastated by the disease. In 1985, the first serologic test to detect HIV was implemented by blood banks to protect the blood supply. However, the possibility that other, as yet unknown, pathogens could also be transmitted by blood transfusion resulted in a new awareness that blood should be used judiciously. Multiple serologic tests are now routinely employed on every blood component for a variety of infectious agents, including hepatitis B, hepatitis C, HIV, cytomegalovirus (CMV), syphilis, and human T-lymphotropic virus (HTLV). Despite the improvement in blood safety that serologic testing provides, the risk of infectious disease transmission is not completely eradicated from the blood supply due to the window period in which donors are infectious, but have not yet developed detectable antibody. The development of nucleic acid amplification technology (NAT) has limited the risk of infectious disease transmission even more by allowing the direct detection of even small quantities of pathogen. Currently, NAT testing is widely employed for HIV and hepatitis C detection, and is used, in certain areas, for detection of hepatitis B virus

and West Nile virus. Some type of pathogen inactivation applied to blood components might obviate the need for yet more testing of known and unknown pathogens that might be transfusion-transmissible.

NONINFECTIOUS COMPLICATIONS OF TRANSFUSION In addition to infectious diseases, other risks of transfusion became apparent as transfusions increased. Transfused leukocytes were found to have a number of undesirable effects (e.g., graft-versus-host disease [GVHD] and febrile reactions).8 In 1970, Graw and colleagues demonstrated that GVHD could be prevented by the irradiation of blood components.44 In 1962, the first-generation leukocyte filter was shown to be effective in the prevention of febrile transfusion reactions.45 The additional benefits of leukoreduction, including reduction of leukotropic viruses, such as HTLV 1/2 and CMV, and minimization of the risk of human leukocyte antigen (HLA) sensitization, have resulted in the widespread adoption of leukoreduction of cellular blood components.

THE MODERN ERA Transfusion medicine continues to evolve in the modern era. New, problematic infectious agents emerge, such as West Nile virus, which is usually transmitted to humans by mosquitoes. West Nile virus was first reported to be transmitted by transfusion and organ transplantation in 2002. As detection and prevention of transfusion-transmitted viral infections has improved, bacterial contamination has evolved into one of the most significant causes of transfusion-transmitted infectious disease. Bacterial detection systems are being used on platelet components with some success. Fortunately, new methodologies designed to inactivate bacteria have been developed and are undergoing evaluation.46 Alternatives to human blood for transfusion, so called “blood substitutes,” continue to be actively investigated but have thus far had limited success. Current “blood substitutes” are limited to substances designed to carry oxygen. Unfortunately, these substances have limited application and are associated with a unique set of risks and potential complications.47 The history of transfusion medicine parallels mankind’s understanding of physiology, immunology, chemistry, infectious diseases, and advances in technology. What began as a belief that blood carries important healing properties has been validated by science. However, despite numerous advances over the centuries, blood remains an indispensable, life-giving force. Interestingly, some of the more important advances in blood banking/transfusion medicine have occurred as the result of wars.

SUMMARY REMARKS Blood transfusions today are an indispensable part of many medical and surgical therapies. The use of blood and its components temporarily replaces what may be lost or not produced before, during, or after a disease process and/or its treatment. The benefits of transfusion today far outweigh

Acknowledgments The authors wish to acknowledge Drs. Leo McCarthy and Paul Schmidt for their generous contributions of photos, information, and review of this chapter. REFERENCES 1. Rossi E, Simon T, Moss G (eds). Principles of Transfusion Medicine. Baltimore, Williams & Wilkins, 1991. 2. Zmijewski CM. Immunohematology, 3rd ed. New York, Appleton-Century-Crofts, 1978. 3. Giangrande PL. The history of blood transfusion. Br J Haematol 2000;110:758–767. 4. Myhre BA. The first recorded blood transfusions: 1656 to 1668. Transfusion 1990;30:358–362. 5. Petz L, Swisher S, Kleinman S, Spence R (eds). Clinical Practice of Transfusion Medicine, 3rd ed. New York, Churchill Livingstone, 1996. 6. Denis J. A letter concerning a new way of curing sundry diseases by transfusion of blood. Philos Trans R Soc Lond [Biol] 1667;2:489. 7. Star D. Blood. An epic history of medicine and commerce. New York, HarperCollins Publishers, 2002. 8. Greenwalt TJ. A short history of transfusion medicine. Transfusion 1997;37:550–563. 9. Schmidt PJ. The first photograph of blood transfusion. Transfusion 2001;41:968–969. 10. Schmidt PJ, Leacock AG. Forgotten transfusion history: John Leacock of Barbados. BMJ 2002;325:1485–1487. 11. Baskett TF. James Blundell: the first transfusion of human blood. Resuscitation 2002;52:229–233. 12. Landsteiner K. Uber Agglunationserscheinungen normalen menschlichen Blutes. Wein Klin Wschr 1901;14:1132–1134. 13. Watkins WM. The ABO blood group system: historical background. Transfus Med 2001;11:243–265. 14. Levine P. A review of Landsteiner’s contributions to human blood groups. Transfusion 1961;1:45–52. 15. Ottenberg R, Kaliski DJ. Accidents in transfusion: their prevention by preliminary blood examination: based on an experience of 128 transfusions. JAMA 1913;61:2138–2140. 16. Mollison PL. Blood Transfusion in Clinical Medicine, 7th ed. Oxford, Blackwell Scientific Publications, 1983. 17. Moreschi C. Neue Tatsachen Uber die Blutkorperchen-agglutination. Zentralbl Bakteriol Parasitenkd Infektkr 1908;1 Originale 46:49–51. 18. Weil R. Sodium citrate in the transfusion of blood. JAMA 1915;64: 425–426. 19. Hess JR, Schmidt PJ. The first blood banker: Oswald Hope Robertson. Transfusion 2000;40:110–113. 20. Loutit JF, Mollison PL. Advantages of a disodium-citrate-glucose mixture as a blood preservative. BMJ 1943;2:744–745. 21. Gibson JG, Gregory CB, Button LN. Citrate-phosphate-dextrose solution for preservation of human blood: a further report. Transfusion 1961;1:280–287. 22. Smith AU. Prevention of haemolysis during freezing and thawing of red blood cells. Lancet 1950;2:910–911.

23. Moore SB. A brief history of the early years of blood transfusion at the Mayo Clinic: the first blood bank in the United States (1935). Transfus Med Rev 2005;19:241–245. 24. Telischi M. Evolution of Cook County Hospital blood bank. Transfusion 1974;14:623–628. 25. Jorda FD. The Barcelona blood transfusion service. Lancet 1939;1:773– 776. 26. www.pbs.org/wnet/redgold 27. www.cdrewu.edu 28. Schmidt PJ. Charles Drew, a legend in our time. Transfusion 1997;37:234–236. 29. Stern K, Goodman HS, Berger M. Experimental isoimmunization to hemoantigens in man. J Immunol 1961;87:189–198. 30. Clarke CA, Donohoe WT, McConnell RB, et al. Further experimental studies on the prevention of Rh haemolytic disease. BMJ 1963;5336:979– 984. 31. Clarke CA, McConnell RB. Prevention of Rh-hemolytic disease. Springfield, Ill., Charles C Thomas, 1972. 32. Freda VJ, Gorman JG, Pollack W. Successful prevention of experimental Rh sensitization in man with an anti-Rh gamma2globulin antibody preparation: A preliminary report. Transfusion 1964;77:26–32. 33. Freda VJ, Gorman JG, Pollack W. Rh factor; prevention of immunization and clinical trial on mothers. Science 1966;151:828–830. 34. [No authors listed]. Prevention of Rh-haemolytic disease: results of the clinical trial. A combined study from centres in England and Baltimore. BMJ 1966;2:907–914. 35. Pollack W, Gorman JG, Freda VJ, et al: Results of clinical trials of RhoGAM in women. Transfusion 1968;8:151–153. 36. Walter CW, Murphy WP. A closed gravity technique for the preservation of whole blood in ACD solution utilizing plastic equipment. Surg Gynecol Obstet 1952;95:113–119. 37. Sack T, Gibson JG, Buckley ES. The preservation of whole ACD blood collected stored and transfused in plastic equipment. Surg Gynecol Obstet 1952;95:113–119. 38. Pool JG, Shannon AE. Production of high-potency concentrates of antihemophilic globulin in a closed-bag system. N Engl J Med 1965;273:1443–1447. 39. Brinkhous KM, Shanbrom E, Roberts HR, et al. A new high-potency glycine-precipitated antihemophilic factor (AHF) concentrate. Treatment of classical hemophilia and hemophilia with inhibitors. JAMA 1968;205:613–617. 40. Beeson PB. Jaundice occurring one to four months after transfusion of blood or plasma. JAMA 1943;121:1332–1334. 41. Allen JG, Sayman WA. Serum hepatitis from transfusion of blood. JAMA 1962;180:1079–1085. 42. Feinstone SM, Kapikian AZ, Purcell RH, et al. Transfusion-associated hepatitis not due to viral hepatitis type A or B. N Engl J Med 1975;292:767–770. 43. Ammann AJ, Cowan MJ, Wara DW, et al. Acquired immunodeficiency in an infant: possible transmission by means of blood products. Lancet 1983;1:956–958. 44. Graw RG Jr, Buckner CD, Whang-Peng J, et al. Complication of bone-marrow transplantation. Graft-versus-host disease resulting from chronic-myelogenous-leukaemia leucocyte transfusions. Lancet 1970;2:338–341. 45. Greenwalt TJ, Gajewski M, McKenna JL. A new method for preparing buffy coat-poor blood. Transfusion 1962;2:221–229. 46. Lin L, Hanson CV, Alter HJ, et al. Inactivation of viruses in platelet concentrates by photochemical treatment with amotosalen and long-wavelength ultraviolet light. Transfusion 2005;45:580–590. 47. Klein HG. Blood substitutes: how close to a solution? Dev Biol (Basel) 2005;120:45–52.

A BRIEF HISTORY OF BLOOD TRANSFUSION

their minute (yet real) risks with all the current safeguards to select donors, test blood, and ensure that compatible blood is transfused to the correct patient. Transfusion medicine has come a long way due to multiple pathfinders, adventurous physicians, and courageous donors and patients, especially in the last half century. We owe much to these pioneers.

1 11

A. Immunohematology i. Basic Principles Chapter 2

Principles of the Immune System Central to Transfusion Medicine Terrence L. Geiger

INTRODUCTION The histories of immunology and transfusion medicine intertwine, and efforts in each field have revealed fundamental truths about the other. One of the early leaders in both areas was Paul Ehrlich.1,2 At the turn of the 19th century, Ehrlich observed that goats were able to produce an immune substance—antibodies—which could destroy red blood cells (RBCs) derived from other goats. Interestingly this substance did not affect a goat’s own RBCs. From these findings Ehrlich developed the concept of horror autotoxicus, that in order to avoid “autotoxicity,” or self-harm, immune responses are directed against foreign and not self compounds. Karl Landsteiner expanded on Ehrlich’s work.3 He demonstrated that anti-red cell agglutinins found in human serum, as in goats, were never directed against self, or autologous, RBCs. However, he also observed that some serum specimens were able to agglutinate samples of homologous RBCs, while other serum specimens could not. He categorized these agglutinins in analyses that led to the description of the ABO blood group system and thereby formed a scientific basis for modern transfusion medicine. One hundred years after the pioneering studies of Ehrlich and Landsteiner, we still struggle to understand the immune mechanisms they described. The human immune system, functioning as a barrier against pathogen attack, can prevent the transfusion of cells and proteins and the transplantation of tissues. These cells and tissues, which are foreign to our bodies, contain antigens, or chemical structures recognized by the immune system. As Ehrlich observed, our immune system generally does not respond to self-antigens, antigens that are intrinsic to our bodies. In contrast, it can and often will respond against foreign antigens. In the case of transfusion, these foreign antigens result from differences between the structures of proteins and sugars present in transfused blood and those native to the recipient. The recognition of antigens within transfused blood by the immune system may have several consequences. It may lead to the destruction of transfused cells and the neutralization of transfused proteins. This can occur through

the generation of specific antibodies or less frequently through the activation of cellular immunity. Immune engagement may also inactivate the immune system. This is called immune tolerance. If tolerance develops against antigens on transfused or transplanted cells, the immune system will not respond when it encounters the same antigens again, even if they are present on different cells or tissues. Finally, the immune system may neither elaborate an effective response against an antigen nor become tolerant of it, an event sometimes called immune ignorance.4,5 In each of these circumstances a transfusion is not passively witnessed by the immune system. Whether or not we observe a clinical consequence of immune engagement after a transfusion, the immune system is continuously active, identifying antigens and determining whether and how to respond. The immune system consists of two lines of defense. The first, termed the innate immune response, is triggered by substances common to many pathogens, such as bacterial lipopolysaccharide or viral nucleic acid. These substances indicate potential danger to a host. Innate immunity is immediate. It is found in most multicellular organisms, even those with very short life spans, and it rapidly neutralizes the offending agent. A second, more complex immune response may follow the innate response. This response, which specifically targets antigens present in the immunizing substance, is called the adaptive immune response. Adaptive immunity requires days to weeks to develop and is only found in complex organisms. It is characterized by the development of antigen-specific antibody and T-cell responses.

THE INNATE IMMUNE SYSTEM The innate immune system includes an amalgamation of interacting pathways that use invariant proteins to protect our bodies from specific types of assault. It continuously monitors our internal environment for the presence of pathogens and other hazards. It also defines the type of danger present, initiates a protective response, and establishes an environment conducive to the formation of effective adaptive immunity.6

2 15

BLOOD BANKING

The innate immune system includes dedicated cell types that lack antigen-specific receptors yet are able to recognize and neutralize pathogens and infected cells. These cells can destroy pathogens through phagocytosis, or ingestion, or through the production of cytocidal compounds such as reactive oxygen species and nitric oxide. They can also sound an alarm, recruiting other inflammatory cells to sites of infection and initiating inflammatory reactions. Examples of innate immune cells include macrophages, mast cells, neutrophils, eosinophils, and natural killer (NK) cells. Innate immunity is, however, not limited to specific cell types. Elements of innate immunity are present in and around virtually all cells in our body. Some of these elements are produced within our cells, such as the cytoplasmic antiviral proteins that are produced in response to viral infections. Others, such as secreted antimicrobial peptides called defensins and cathelicidins, are exclusively extracellular.7 Activation of innate immunity begins with the recognition of pathogen motifs by specific receptors.8 These motifs tend to be conserved within a class of organisms. An example is gram-negative bacteria, whose double membrane contains endotoxin, a unique complex of lipid and polysaccharide also called lipopolysaccharide (LPS). LPS binds to an LPS-specific receptor present on innate immune cells called Toll-like receptor 4 (Tlr4).9 Tlr4 is one of a group of similar receptors that are expressed on the cell surface or in endosomes (internalization vacuoles) and signal when danger is present.10 Other Toll-like receptors recognize motifs specific to gram-positive bacteria, viruses, or other pathogens (Table 2–1). When a Toll-like receptor is activated, it induces signaling molecules that coordinate the early response against an

offending pathogen. These signaling molecules are called biological response modifiers (BRMs). BRMs include lipids, such as leukotrienes, and other small signaling molecules, such as histamine and serotonin. They also include a group of messenger proteins called cytokines that are critical in orchestrating the innate immune response (Table 2–2). There are several classes of cytokines with distinct roles in innate immunity. The interaction of viral DNA or RNA with specific Toll-like receptors induces the synthesis of Type I interferons (IFNs), a class of cytokines that binds IFN receptors on cells and promotes the formation of proteins that block virus replication. Another class of cytokines is called chemokines. The word chemokine is a conjunction of chemotactic, the promotion of migration along a chemical gradient, and cytokine. As chemokines diffuse from their site of production, they form a gradient along which immune cells migrate. An example is interleukin-8 (IL-8), which recruits inflammatory cells to the site of an infection.11,12 Other released cytokines promote inflammation. For example, tumor necrosis factor (TNF) activates endothelial cells, aiding in the diapedesis of reactive cells into tissue, and increases the production of pathogen-neutralizing effector molecules such as reactive oxygen species and nitric oxide.13 Yet other cytokines, such as IL12, primarily link innate and adptive immunity. IL-12 promotes T-lymphocyte differentiation into helper T cells (Th1 cells) capable of recruiting and activating macrophages that destroy pathogens.14 Cumulatively, the cytokines and other BRMs released during the innate immune response form a communication network between cells that identifies the pathogen class and severity of the infection and establishes an initial response to it.

II 16

Table 2–1 Recognition of Pathogen-Specific Motifs by Toll-like Receptors Receptor

Recognition Motif

Pathogen

Tlr1 (associates with Tlr2) Tlr2 Tlr3 Tlr4 Tlr5 Tlr6 (associates with Tlr2) Tlr7

Triacyl lipoproteins Peptidoglycan, lipoproteins, lipopeptides Double-stranded RNA Lipopolysaccharide Flagellin Diacetylated lipoproteins Viral single-stranded RNA; imidazoquinalines and other guanosine nucleotide-related synthetic compounds Viral single-stranded RNA Unmethylated CpG DNA

Gram-negative bacteria Gram-positive and other bacteria, parasites, fungi Viruses Gram-negative bacteria Bacteria Mycobacteria Viruses; synthetic agents

Tlr8 Tlr9

Viruses DNA viruses, prokaryotes

Table 2–2 Some Important Cytokines and Their Functions Cytokine

Function

IL-1 IL-2 IL-4 IL-5 IL-10 IL-12 IL-15 IFN-γ IFN-α, β TNF-α TGF-β

T-cell and APC stimulatory, fever T-cell expansion and apoptosis sensitivity B-cell stimulation, IgE production, Th2 differentiation Eosinophil expansion and development Suppression of macrophage and DC function Th1 differentiation, NK activation NK cell and T-cell expansion and survival Macrophage activation, enhancement of antigen presentation, promotion of IL-12 production Antiviral Inflammation, sepsis Anti-inflammatory, T-cell inhibition, IgA production

influence the magnitude and quality of the adaptive immune response against co-administered antigens.22 Thus, the innate immune context in which a transfusion is administered may have a significant effect on the immunologic outcome of that transfusion.

THE ADAPTIVE IMMUNE SYSTEM The adaptive immune system selectively neutralizes pathogens and nonpathogenic foreign antigens. It recognizes these through specific antigen receptors. Whereas receptors of the innate immune system detect just a handful of pathogenspecific chemical motifs, antigen receptors of the adaptive immune system are able to selectively identify billions of distinct antigenic structures. The 3 billion base pair long human genome would be insufficient to encode adequate numbers of proteins to accomplish this feat if each receptor was a separate gene, as is the case for Toll-like receptors. To circumvent this problem, antigen receptors are built up from modules that allow the incorporation of variability as they form (Fig. 2–1).23 Each polypeptide of an antigen receptor is composed of three or four fragments of DNA. These fragments are called variable (V), diversity (D), junctional (J), and constant (C) fragments. More than a single piece of DNA can encode for each of these fragments, and their arbitrary linkage results in combinatorial variation. For example, if an organism has 20 V regions, 2 D regions, 10 J regions, and 1 C region for a receptor in its genome, random recombination would allow the formation of 20×2×10×1 or 400 unique receptors from the 20+2+10+1, or 33, gene segments. Two additional features promote variation. As the receptor fragments recombine, enzymes randomly insert or remove nucleotides at the recombination sites. The sites where these alterations are made correspond to parts of the receptor that directly contact antigen and are called hypervariable regions.24

Figure 2–1 Antigen receptor gene rearrangement. The germline genetic sequence consists of multiple variable (V), diversity (D), and junctional (J) segments. Rearrangement of a D and J segment with excision of intervening DNA sequence is followed by that of the DJ sequence to a V region. Completion of rearrangement juxtaposes sequences necessary for RNA transcription. The RNA transcript is spliced to create an mRNA that is translated into an antigen receptor chain. The diagram is illustrative and does not indicate the full complexity of the T-cell receptor or B-cell receptor loci.

PRINCIPLES OF THE IMMUNE SYSTEM CENTRAL TO TRANSFUSION MEDICINE

Although critical for innate immune pathogen detection, Toll-like receptors are not exclusive in this ability. Other proteins have complementary functions.15 A cytoplasmic protein called retinoic acid inducible gene 1 binds viral RNA and induces production of IFN-α and IFN-β. Other proteins that include nucleotide-binding oligomerization domains have been found to recognize intracellular bacteria. Mannosebinding lectins in the extracellular fluid bind the surfaces of pathogens and then activate complement. Recent data have also documented a completely new manner through which our bodies can recognize pathogens. Very short RNA sequences, microRNAs, are able to specifically recognize viral genetic material and in so doing inhibit viral replication within a cell.16 Our ability to withstand the onslaught of potential invaders surrounding us is critical to our survival, and our bodies invest heavily in our innate immune system as a first line of defense. The activation of innate immunity through pathogen-specific receptors may seem irrelevant to immune responses to blood transfusions, which to the best of our abilities lack contaminating pathogens. However, the innate immune system establishes a basal inflammatory state. Transfused RBC or plasma antigens are presented to the adaptive immune system in the context of this state, which may influence the immune response to them. Although a direct link between a specific type of inflammatory response and the type of response generated against transfused allogeneic blood products has not been demonstrated, associations between infection and other alloimmune responses, particularly graft rejection, have been documented.17,18 Furthermore, specific immune reactivity against “bystander” antigens during responses against pathogens has been described in other circumstances and is therefore likely present in the context of transfusion.19–21 Indeed, establishing such a bystander response forms the scientific rationale for the adjuvants present in modern vaccines. Adjuvants, by stimulating specific innate immune receptors,

2 17

\CJ Z

;;;;; Z

«co o

o o---'

co

•

Thus antigen receptors are most variable at critical antigenbinding sites. Further, antigen receptors are heterodimers composed of two polypeptides, a V-D-J-C chain, and a V-D-J chain. Each of these receptor components uses an independent set of V, D, J, and C fragments. The chains recombine independently but recognize antigen together. Thus billions of distinct antigen receptors may form from a relatively small number of gene fragments. Antigen receptors are expressed on only two classesof cells, B lymphocytes and T lymphocytes. On B cells they form the B-cell receptor (BCR), which is a cell surface form of immunoglobulin. On T cells they form the T-cell receptor (TCR). Each T cell or B cell expresses only a single (or in some rare casestwo) immune receptors on their cellsurface." Therefore, each T cell or B cell is essentially a clone recognizing a distinct piece of an antigen, called an epitope. Because of the diversity of immune receptors, the frequency of T cells or B cells able to respond to any single antigenic protein is extremely low, typically ranging from 1110,000 to 11100,000 cells." Receptor engagement activates a lymphocyte and may lead to the selection and rapid outgrowth of rare antigen-specific clones. Cell activation by antigen may also induce differentiation into effector lymphocyte forms designed to participate in distinct types of immune responses, such as allergic responses or delayed type hypersensitivity responses.

The Trail of Antigen Foreign antigens can enter the body through different routes. In the case of infection, entry is generally through the skin or mucosal tissues. In the case of transfusion, direct intravenous inoculation occurs. In both situations, antigens make their way to lymphoid tissue. Antigens in the skin are picked up by specialized resident skin dendritic cells called Langerhans cells and are then carried to draining lymph nodes." Antigens in the blood are picked up by cells in the spleen. Antigens in the gut are taken up by cells in the gut-associated lymphoid tissue." They are then presented to T lymphocytes and B lymphocytes by specialized antigen-presenting cells (APCs), also called professional APes, in the lymphoid tissues. The lymphoid tissues are specifically designed to facilitate the

A Figure 2-2

interaction between rare antigen-specific T cells and B cells and APCs bearing their cognate antigen (Fig. 2-2). The cell carrying antigen to lymphoid tissue may transfer that antigen to professional APCs within that tissue." Alternatively the courier cell, often a resident tissue dendritic cell, may serve as the APC within the lymphoid tissue, directly presenting antigen to lymphocytes. Conversion of such courier cells to effective professional APCs requires their maturation under the influence of cytokines or Tolllike receptor signals that are present in the context of inflammation and innate immune activation. These stimuli induce the formation of signaling molecules on the surface of the APC that can bind and activate lymphocyte receptors. They also induce the secretion of cytokines and other BRMs by the APCs.29,3o Such supplemental, or costimulatory, signals can both modify and augment the signal received by a lymphocyte through its antigen receptor. In the presence of costimulatory signaling, a lymphocyte may become more fully activated, permitting its expansion and differentiation into an effector cell. In contrast, the absence of costimulatory signals indicates to a lymphocyte that antigen is not being presented in the context of inflammation or host danger. Signals through antigen receptor in this circumstance lead to incomplete stimulation that may promote the development of immune tolerance, sometimes through the death of the responding antigen-specific lymphocyte.v" As selfderived antigens will tend not to be presented in the context of inflammation, this form of tolerance is one route through which the immune system rids itself of self-reactive lymphocytes that may cause autoimmunity.

Antigen Presentation to T Cells The TCR is a complex that includes six different polypeptides, most commonly ex, ~,y, 8, E, and ~ (Fig. 2_3).33,34 The ex and ~ chains, which include recombined VJC and VDJC regions, respectively, are directly involved in antigen recognition. The remaining polypeptides do not recognize antigen but transmit signals into T cells after antigen engagement occurs. The y, 8, and E signaling chains are collectively referred to as the CD3 complex.

B

Schematic depiction of the structure of a lymph node (A) and an area of the splenic white pulp (B). Lymphocytes enter through the high endothelial venule (HEV) or central arteriole and then migrate to T- or B-rich areas where they may interact with antigen-presenting cells. T cells and B cells may also engage each other, often at the interface between the B- and T-cell zones of the spleen or lymph node. (From Mondino A, Khoruts A, Jenkins MK. The anatomy of T cell activation. Proc Natl Acad Sci USA 1996;93:2246.)

N N N

C

β2m

Figure 2–3 The TCR–CD3 complex consists of eight polypeptide chains. Specificity is determined by the αβ chains or, in some T cells, by analogous γδ TCR chains. Expression of the CD3 γ, δ, and ε chains, as well as ζ or its splice variants, are necessary for surface expression and signaling.

T cells do not see antigen alone through their TCR. They see it only in the context of major histocompatibility complex, or MHC, proteins. The significance of the MHC, also called the human leukocyte antigen (HLA) complex in humans, was first recognized through the work of Peter Gorer and George Snell in the 1930s.2 Gorer and Snell were studying tumor graft rejection and determined that this genetic locus was absolutely critical in determining whether grafted cells were rejected. We now know that the MHC serves as a window through which T cells see the world around them. T lymphocytes recognize antigen only in the context of MHC molecules. Recognition is remarkably specific. Even subtle allelic variations in a single MHC molecule are sufficient to abrogate T-cell recognition.35 MHC proteins fall into two categories, class I and class II molecules.36 Class I MHC molecules are heterodimers consisting of a larger α or “heavy” chain, which is membrane bound, and a smaller soluble protein called β2-microglobulin. Class II MHC molecules are heterodimers of two membrane-bound polypeptides, an α chain and a β chain of roughly equal size. How the MHC presents antigen to T cells was not clear until the late 1980s when x-ray crystallographic structures of MHC molecules were solved (Fig. 2–4).37–39 Although the amino acid sequences of class I and class II molecules are distinct, they form similar three-dimensional structures. The upper surface of the MHC molecule forms a platter, called a β sheet, on top of which lie two parallel tubular coils of peptide, called α helices. Between the α helices lies a groove into which various protein fragments, or peptides, may bind. These peptides, presented together with their restricting MHC molecules, are the antigenic structures recognized by T lymphocytes. Class I MHC molecules bind peptides approximately 8 to 10 amino acids in length, whereas class II MHC molecules bind longer peptides, approximately 13 to 24 amino acids. Only a few interactions between peptide amino acids and the MHC are needed to stabilize binding.40,41 This allows for the formation of large numbers of different peptide–MHC complexes. A typical antigenic protein will have one or more peptides able to bind any specific MHC molecule. The MHC therefore serves as a platform that displays the antigenic universe to T cells. The T cells view these peptides through their TCR, which recognize and bind the peptide–MHC complex.

A

C

α3 B

Figure 2–4 Structure of a class I MHC molecule. A, Three-dimensional ribbon model of the extracellular portion of the molecule. The peptide-binding groove lies between the helices at the top of the molecule. B, Top view. (From Bjorkman PJ et al Structure of the human class I histocompatability antigen, HLA-A2. Nature 1987;329:506.)

It could be imagined that a pathogen may readily evade the immune system by simply designing proteins that do not contain MHC-binding motifs. However, this is not possible for two reasons. First, we express several different class I and class II MHC molecules, three of each for the major, or classical, MHC forms on each chromosome. These are called HLA-A, -B, and -C for the human class I locus, and HLA-DR, -DQ, and -DP for the class II locus. Therefore, an individual heterozygous for each locus may have a total of six different class I molecules, and, because the α and β chains of the class II molecule encoded on each chromosome may heterodimerize either in cis or in trans, an even larger number of distinct class II MHC molecules. If a pathogen’s proteins failed to bind one HLA protein, it would likely bind one of the others. More importantly, the MHC genes are remarkable among all genes for their polymorphism, or allelic variability. Because of this variability, most of us have different sets of MHC genes. Indeed, no other human genetic locus is as polymorphic as the MHC.42 Sequence variation of the allelic variants is particularly prominent at the peptide-binding groove, and different alleles of even a single MHC gene therefore often have distinct peptide-binding and antigen-presentation properties. MHC polymorphism ensures that a pathogen that attempted to evade the immune system of one individual would not be able to do so for others, thereby protecting the population as a whole. The polymorphism and polygenicity of the MHC also presents a problem in transfusion and transplantation. T lymphocytes respond particularly vigorously against allogeneic (allelically distinct) MHC molecules. Each of these allogeneic MHC molecules can bind thousands of different peptides. Peptides derived from self-antigens, to which a person’s immune system is normally tolerant, when combined with allogeneic MHC molecules, form complexes that are foreign to the host immune system.43 It is estimated that roughly 1% to 10% of a person’s T cells will respond to another person’s allogeneic MHC combined with the array of self-peptides that may bind in its groove. For stem cell transplant recipients, this recognition may be manifested by the development of graft-versus-host disease (GVHD) or graft rejection. In the case of transfusion, it may be made apparent by the development of immune responses resulting in transfusion-associated GVHD or platelet refractoriness.

PRINCIPLES OF THE IMMUNE SYSTEM CENTRAL TO TRANSFUSION MEDICINE

α2

α1

2 19

BLOOD BANKING

Specialized Functions of Class I and Class II MHC Molecules Despite their similar structures, class I and class II MHC molecules have distinct roles. Key to this distinction is the repertoire of peptides that each MHC molecule binds. Class I MHC binds peptides derived from within a cell and thus displays a cell’s intracellular milieu. Class II MHC displays antigens acquired from the extracellular space. Thus, each set of MHC proteins is telling T cells a different tale. In the event of a viral infection, presentation of viral antigens on a cell’s class I MHC would indicate to a T cell that the cell is infected. Presentation on class II MHC molecules would instead indicate that virus is present in the environment, but not necessarily within that cell. The distinct presentation capabilities of class I and class II MHC results from the different mechanisms through which each protein acquires its peptides. Within the cytoplasm of the cell proteins are continuously being degraded by a large proteolytic complex called the proteasome.44 The peptide fragments formed by the proteasome are perfectly sized for binding class I MHC molecules. The production of antigenic peptides may be enhanced and the sites of proteolysis modified by cytokines produced during an immune

response, most notably interferon-γ.45 As class I MHC molecules are synthesized, they are extruded into a membranous compartment within the cell called the endoplasmic reticulum. Appropriately sized, proteasomally derived peptide fragments are continuously transported into the endoplasmic reticulum by a special peptide transporter, called transporter of antigenic peptides, or TAP1/2.46,47 The transported peptides can then bind to the nascent class I MHC molecules (Fig. 2–5).48 MHC molecules bound to peptide are then transported to the cell surface. Class II MHC molecules take a different route to the cell surface. They are also extruded into the endoplasmic reticulum. However, unlike class I MHC molecules, they bind to a protein called the invariant chain as they are synthesized (Fig. 2–6). A peptide segment of the invariant chain folds into the peptide-binding groove of the class II molecule, blocking any other peptides from binding. Rather than going directly to the cell surface like class I MHC molecules, the class II molecules are first diverted to specialized vesicles, or membrane-contained compartments, within the cell where they meet up with extracellularly derived antigens.49–51 Cells may take up external antigens through a variety of processes, including phagocytosis, or wholesale consumption of cells or particles; pinocytosis, or internalization of small

20S proteasome

TAP1

ER membrane

␣

II

TAP2

␤

20

ATP ␤

PA28 ␣ Peptide

Peptide

A

B

Calnexin

Calreticulin Heavy chain

Heavy chain– ␤2m–peptide

Tapasin

␤2m

C

TAP1

TAP2

Figure 2–5 A, Schematic drawing of a cut-open 20S proteasome generating peptide fragments from an unfolded polypeptide chain threaded into the narrow entry formed by an outer ring of β-type subunits. Cleavage occurs in the central cavity formed by two seven-membered rings of β-type subunits. A PA28 regulator complex, inducible by IFN-γ is bound at the bottom of the cylinder. B, Hypothetical cross-section of TAP1/TAP2 heterodimer embedded into the endoplasmic reticulum membrane. The adenosine triphosphate (ATP)-binding domain, which extends into the cytoplasm, is shown. ATP cleavage provides energy for peptide translocation. C, Maturation of class I molecules in the endoplasmic reticulum. Nascent class I heavy chains initially associate with calnexin. On binding of β2-microglobulin, the heterodimer dissociates from calnexin and forms a complex with calreticulin, tapasin, and TAP1/2. After peptide binding, the heavy-chain β2-microglobulin complexes are released and exit to the cell surface. (From Koopman JO, Hammerling GJ, Momburg F. Generation, intracellular transport, and loading of peptides associated with MHC class I molecules. Curr Opin Immunol 1997;9:81.)

PRINCIPLES OF THE IMMUNE SYSTEM CENTRAL TO TRANSFUSION MEDICINE

Figure 2–6 Intracellular pathways of class I and class II MHC molecules. Class I MHC molecules acquire antigenic peptides, generated by the proteasome in the cytosol, that are translocated into the endoplasmic reticulum by the TAP molecules (bottom of figure). The class I MHC–peptide complex is transported through the Golgi complex directly to the cell surface for presentation to CD8 T cells (A). In contrast, class II MHC molecules acquire antigenic peptides derived from antigens that are internalized in the endocytic pathways (B). Class II MHC heterodimers associate in the endoplasmic reticulum (bottom of figure) with invariant chains to form nonameric αβ–Ii complexes. At the trans-Golgi network (TGN), these complexes are targeted to class II MHC compartments (MIIC) in the endocytic pathway as a result of targeting signals within the Ii cytoplasmic tail (not shown). There the class II MHC-associated Ii is degraded in distinct steps, at least partially by cathepsin proteases (C), leaving a peptide from Ii in the class II MHC–peptide-binding groove. This peptide is exchanged for antigenic peptides in a process catalyzed by HLA-DM molecules. Peptide-loaded class II MHC complexes are then transported to the plasma membrane for presentation to CD4 T cells (D). (From Pieters J. MHC class II restricted antigen presentation. Curr Opin Immunol 1997;9:91.)

amounts of the extracellular fluid; and receptor-mediated internalization, which may occur through bound immunoglobulin or sugar-binding receptors such as the mannose receptor.52–54 This material is broken down by enzymes within the endocytic, or internalization, pathway. Vesicles containing endocytosed and degraded material merge with the class II MHC containing vesicles. The invariant chain is degraded by proteases at this point, and vesicular peptides can then bind within the class II MHC-binding groove. The class II MHC, with externally derived bound peptide, then makes its way to the cell surface. T-Cell Co-receptors and the Recognition of Class I and Class II MHC Molecules MHC molecules present the antigenic universe to T lymphocytes. Considering the different types of antigens presented by class I and class II MHC, intracellular versus extracellular, it is not surprising that these molecules will engage different types of T cells and that this engagement will result in different types of responses. To understand how this occurs requires some background in how T cells recognize antigen. The variable α and β chains of the TCR recognize peptide–MHC complexes. However, their affinity for peptide– MHC is poor, approximately 100- to 1000-fold lower than

the binding affinity, or interaction strength, of antibodies for their target antigens.55,56 For most TCRs, this interaction is too weak and too transient to generate a signal capable of fully activating a T cell. The TCR is able to signal because it also receives assistance from co-receptors called CD4 and CD8. CD4 binds to class II MHC molecules and CD8 binds to class I MHC molecules. When the TCR engages cognate peptide–MHC, CD4 or CD8 co-associates. CD4 and CD8 are transmembrane proteins that perform two tasks.57 First, because they both associate with the TCR and bind MHC at a site different from where the TCR binds peptide–MHC, they increase the binding affinity of the TCR complex for peptide–MHC. Second, and more importantly, they carry bound to their intracellular domain an enzyme of the src family, called lck (Fig. 2–7).58 Lck is a type of enzyme called a tyrosine kinase. It adds phosphates to certain tyrosine residues present in proteins. Tyrosine phosphorylation of the γ, δ, ε, and ζ chains of the TCR is the first step in generating a signal once a TCR recognizes antigen. The complex of CD4 or CD8 co-receptor with the TCR and peptide–MHC aligns the lck kinase with specific conserved tyrosines present within the signaling chains of the TCR called immunoreceptor tyrosine-based activation motifs, or ITAMs.59 Once phosphorylated, these ITAMs

2 21

BLOOD BANKING

Ag/MHC

Ag/MHC

Ag/MHC

TCR

TCR

TCR CD4

CD4

CD4 PO4

PO4 PO4

Lck

SH2

Lck

SH2

PO4

Lck

PO4 SH2

SH2

SH2

ZAP–70

SH2

ZAP–70 Cellular substrates

Figure 2–7 Model of TCR, Lck and ζ-associated protein (ZAP)-70 interactions during antigen recognition by a CD4 T cell. The phosphorylation of tyrosine residues of the immunoreceptor tyrosine-based activation motifs within the ζ and CD3 chains has been simplified. (From Weiss A. T-cell antigen receptor signal transduction: A tale of tails and cytoplasmic protein kinases. Cell 1993;73:211.)

II 22

serve as docking sites for additional proteins, particularly ZAP-70, another tyrosine kinase that is in turn phosphorylated and activated by lck. This triggers a cascade of signaling inside the cell as additional proteins are recruited to these phosphorylated molecules. In a manner analogous to how the coagulation cascade may be triggered by single event, such as exposure to tissue factor, a low amplitude signal that initially results from TCR–MHC engagement is enzymatically amplified within a T cell. The impact of CD4 and CD8 is dramatic, and it has been estimated that these molecules enhance signaling through the TCR by a factor of 100.57 Through its sophisticated signaling apparatus, T cells may be discriminately stimulated by 100 or fewer antigen–MHC complexes on the surface of an APC.60 Mature T cells express either CD4 or CD8 and do not co-express both of these. This contrasts with their progenitors that develop in the thymus, which co-express CD4 and CD8. During development, these CD4+D8+ thymocytes are tested.61 Cells that interact strongly with self-antigen–MHC complexes in the thymus are deleted in a process termed negative selection. Negative selection is a major mechanism through which the adaptive immune repertoire is depleted of self-antigen–specific T cells. Cells that do not interact with MHC present in the thymus do not receive necessary survival signals and undergo death through neglect. In contrast, in a process termed positive selection, cells with receptors that interact loosely with class I MHC molecules receive survival signals, downmodulate CD4, and are left expressing only CD8. Thymocytes that interact with class II MHC downmodulate CD8 and become CD4-expressing T cells. Thus, T cells can be divided into two major groups based on their interaction with either class I or class II MHC molecules, and these groups can be identified by the alternative expression of CD4 or CD8. Further, CD4 and CD8 define classes of T cells with distinct functional potentials. Cells That Present Antigen Because class I MHC presents antigens derived from intracellular pathogens, like viruses, or that result from intrinsic abnormalities, such as malignant transformation, it is important that all cells capable of self-propagation express class I MHC. CD8+ T cells recognizing class I MHC-derived antigens may eliminate these aberrant or infected cells. The

majority of nucleated cells indeed express class I MHC constitutively, though class I MHC expression can be enhanced by cytokines, such as in the setting of inflammation.62 Not all cells need to express class II MHC, which presents external antigens. Class II MHC expression is limited to specialized cells that provide T cells with environmental information. These include professional APCs, dendritic cells, macrophages, B cells, and endothelial cells, though other cell types may be induced to express class II MHC in the context of inflammation. Of the professional APCs that present antigen to T cells on class II MHC, dendritic cells are particularly efficient. They reside in tissues in a quiescent state. When activated by any of a large variety of inflammatory stimuli, such as cytokines or Toll-like receptor signals, they sample antigens from their environment, which they present on cell surface class II MHC molecules.30 After accumulating large quantities of antigen, the dendritic cells cease internalizing antigen and migrate to lymphoid tissues, particularly lymph nodes, where they present antigen to lymphocytes. There are several types of dendritic cells.63 Some, the follicular dendritic cells, are particularly adept at presenting whole antigen to B lymphocytes.64 This is important because B cells recognize whole antigens or large pieces of antigens through their immunoglobulin receptors. In contrast, most dendritic cells, including myeloid, lymphoid, and plasmacytoid dendritic cells, digest antigens into small fragments and present these to T lymphocytes. Because of the scarcity of lymphocytes specific for any single antigen, presentation in specialized lymphoid tissue is essential. The professional APCs structure themselves into a reticular network through which lymphocytes continuously and rapidly percolate.65 After a dendritic cell is stimulated through its Toll-like receptor or other receptors, it may secrete chemokines that will attract T lymphocytes to it and costimulatory molecules that can activate them. The lymphocytes scan these dendritic cells with their antigen receptor for the presence of cognate antigen. With recognition through their antigen and costimulatory receptors, the lymphocytes pause, alter their metabolism, and modulate gene expression. They may then expand and differentiate into effector forms. Other professional APCs are macrophages and B lymphocytes. Macrophages are potent phagocytic cells when activated

Functionally and Phenotypically Distinct Classes of T Cells T cells are categorized into discrete groups based on their phenotype, function, and differentiation potential. The broadest categorization relates to the type of TCR expressed. The large majority of T cells express the αβ TCR described above, though a small number of cells express an alternative receptor composed of a parallel heterodimer of chains called γ and δ. These γ and δ chains recognize antigen–MHC complexes and are distinct from the signaling γ and δ chains that are part of the CD3 complex. Indeed, γδ TCR associate with the CD3 and ζ signaling chains, forming a signaling complex identical to that of αβ receptors. The function of γδ T cells is unclear. Some of these cells reside in lymphoid tissue and have a highly diverse receptor repertoire. Other γδ T cells reside in intraepithelial spaces. These cells have limited receptor diversity within a specific location. This suggests that they recognize relatively invariant molecules, either pathogen or host derived.66 Indeed, some γδ T cells recognize nonclassical, nonpolymorphic class I MHC molecules called class Ib MHC.67 Recent data indicates that they may also efficiently present antigen to αβ T cells and may promote the rapid induction of αβ T-cell responses after they are activated.68 Therefore, these cells may act more as facilitators and regulators rather than mediators of adaptive immunity. The more common αβ class of T cells is typically divided into helper and effector classes. Helper T cells integrate the environmental cues they receive and then signal to other cells, effector cells, orchestrating their response. Helper T-cell signals may include the release of cytokines, such as IL-2, IL-4, and IFN-γ. They also may signal by interacting with other cells through cell surface receptors. Helper T cells fall largely in the CD4 class of T cells. As a consequence, often the terms helper T cell and CD4 T cell are used interchangeably. On a biological level, this is inaccurate and should be avoided. CD4 T cells in some circumstances have effector functions and CD8 T cells may have helper activities. Nevertheless, it should not be surprising that CD4 T cells tend to be dominantly helper cells by nature. CD4 T cells probe the antigenic environment through class II MHC. Their response to environmental cues that may suggest danger should not be directed against the APC, which is merely a messenger carrying antigen to the T cell. They must choreograph a broader response that will directly target harmful antigens. When a freshly formed CD4 T cell exits the thymus it is “naive,” never having seen its cognate antigen. The cell will home to lymphoid tissues, occasionally circulating through the bloodstream to a different lymphoid location.69 As it does this, it searches for antigen able to stimulate its TCR, which is delivered to it by APCs present within lymphoid tissue. Once stimulated this T cell may differentiate into one of a variety of forms. Two of the best studied types of CD4 T cells are Th1 and Th2 cells.70,71 Cytokines secreted by Th1 cells, such as IFN-γ and TNF-β, are responsible for delayed-type hypersen-

sitivity (DTH) reactions.72,73 Th2 cells, which secrete IL-4, IL5, IL-10, and IL-13, are responsible for allergic responses.71,74 In the DTH response, Th1 T cells stimulate B cells to produce specific antibodies that efficiently activate complement after antigen binding. The Th1 cells also migrate to sites of antigen presentation in tissues. Once there, they secrete cytokines, such as IFN-γ and TNF, and chemokines that promote the recruitment and activation of macrophages, additional lymphocytes, and other inflammatory cells.75,76 Activation of endothelial cells leads to a loss of vascular integrity and the development of induration, a hallmark of the DTH response. The activated macrophages locally present antigen and amplify the immune response. Macrophages also serve as primary effector cells, ingesting and eliminating pathogens, and producing reactive oxygen species that can kill microorganisms. If this response is unable to clear the inciting agent, granuloma formation and scarring may wall it off. In Th2 responses, CD4 T cells stimulate the production of antibodies that may be neutralizing or promote antibodydependent cell-mediated cytotoxicity. Cytokines produced by the Th2 cells activate and promote the expansion of eosinophils and other cell types. Th2 T cells prominently support the production of IgE, which binds specific high-affinity receptors on mast cells and basophils.77 Antigen crosslinks these antibodies, thereby activating the mast cells and basophils. This induces the release of histamine and other BRMs that are associated with allergic or anaphylactic symptoms, such as vascular dilatation, mucus production, urticaria, and diarrhea. We commonly associate these symptoms with pathologic responses to allergens and not with the clearance of infections. However, Th2 responses appear to be important for the clearance of some parasitic infections.78,79 The final form a naïve T cell differentiates into is determined by its microenvironment after antigen stimulation.80,81 Th1 cell differentiation is promoted by the presence of IL-12 and IFN-γ. Th2 cell differentiation is promoted by the presence of IL-4 and prostaglandin E2. These stimuli may be produced after innate immune signaling in cells. IL-4 is produced by activated mast cells and basophils.82–84 Activated dendritic cells may produce IL-12 and the closely related IL-23.82,85 NK cells may produce IFN-γ.86 These differentiation cytokines may also be produced by T cells themselves. Th1 cells produce IFN-γ, which further promotes Th1 differentiation and inhibits Th2 differentiation. Th2 cells produce IL-4, which promotes Th2 and inhibits Th1 development. Thus Th1 or Th2 antigen-specific T cells that developed in prior immune responses or in an earlier phase of an active immune response may polarize a nascent immune response, promoting the development and expansion of T cells of their own class while inhibiting those of other classes. This ensures an unadulterated response to a particular pathogen type, DTH or allergic in the case of Th1 and Th2 cells, respectively. CD8 T cells are typically effector in nature. When activated they differentiate into a class of cells called cytotoxic T lymphocytes (CTLs). In many circumstances this differentiation depends on signals provided by CD4 helper T cells.87,88 One manner by which stimulated CD4 T cells do this is through the upregulation of a TNF-family protein, CD40L, on their cell surface.89 CD40L interacts with CD40 on the surface of APCs, signaling into the APC. This signal is important to “license” the APC, that is, induce in the APC production of cytokines and signaling molecules that are needed to adequately stimulate the CD8 T cell. If a CD8 T cell does not receive these supplemental signals, it will not appropriately

PRINCIPLES OF THE IMMUNE SYSTEM CENTRAL TO TRANSFUSION MEDICINE

and may internalize a variety of different pathogens as well as dead and dying cells. Macrophages may present the antigen acquired at the site of tissue injury or within lymphoid organs and are important in guiding some types of T-cell responses. B cells primarily internalize antigens through their BCR, surface immunoglobulin. The high affinity of the BCR for specific antigens allows them to focus small quantities of antigen present in the environment and to display this to cognate antigen-specific T cells.

2 23

BLOOD BANKING

expand and differentiate in response to antigen and may die after antigen encounter. When a CTL’s TCR encounters and engages its target MHC–antigen complex on a cell, it extrudes preformed granules that contain lytic proteins.90 These proteins include perforin, which forms pores in the membranes of the target cells, and granzymes, enzymes that kill targets through a preprogrammed cell death pathway called apoptosis. They also include a protein called fasL, which binds to a receptor called fas on many cell types, also inducing apoptosis. The absence of adequate lytic activity can have severe consequences. Patients with mutations in perforin develop a disease called hemophagocytic lymphohistiocytosis.91,92 Failure to control infections results in an unrestrained activation of the immune system that is most often fatal.

Regulation of T-Cell Responses

II 24

The effector responses that T cells coordinate can cause substantial bystander damage to host tissues if unrestricted. Because of this, homeostatic mechanisms that limit immune responses are essential. There are several such mechanisms. One mechanism is the apoptosis, or programmed cell death, of lymphocytes. After immune activation, T cells proliferate profusely. This is physically manifested by the splenomegaly and adenopathy commonly identified with infection. Failure to eliminate this expanded cell population would result in a plethora of activated immune effector cells. Most of these cells, however, die through apoptosis.93–95 Members of the TNF family of proteins are particularly important for this. For example, in many circumstances activated T cells will express a pair of TNF family molecules, fas and fasL. When fasL engages fas, it stimulates the fas receptor. This activates a set of cellular proteases called caspases that can cause a loss of mitochondrial and ultimately cellular integrity. In the absence of molecules that inhibit apoptosis, caspase activation ultimately leads to cell death. Dysfunctional apoptosis mechanisms in lymphocytes can lead to autoimmunity by allowing the persistence of self-reactive T cells that normally would be eliminated. Indeed, patients with mutations in fas or apoptotic signaling develop a disease called autoimmune lymphoproliferative syndrome.96 Cytokines and other BRMs may also restrict immune responses. Many cytokines have dual properties, both enhancing and suppressing immune effector mechanisms even within a single type of immune response. For example, IFN-γ activates macrophages and other effector cells and promotes Th1-cell differentiation. It also inhibits proliferation and may in some circumstances promote apoptosis.86 A good example of the dual effects of IFN-γ is observed in an experimental murine model of multiple sclerosis, an autoimmune disease that is predominantly Th1 in nature. IFN-γ produced early in the pathologic response promotes the induction of Th1-polarized T cells and therefore disease development. Later in the disease course, IFN-γ is protective, and its neutralization exacerbates disease.97,98 Although many cytokines have mixed pro- and antiinflammatory effects, some cytokines have dominantly inhibitory roles.99–101 For example, transforming growth factor-β (TGF-β) strongly suppresses T-cell activity. Mice genetically modified so that their T cells are unable to respond to this cytokine develop a spontaneous lethal autoinflammatory condition. IL-10, while able to activate B cells, is also a critical downregulatory cytokine for activated macrophages.

Mice lacking IL-10 develop spontaneous inflammatory bowel disease.102 Negative signaling is not limited to cytokines but can also occur through cell associated proteins. As mentioned above, professional APCs provide T cells with access to costimulatory proteins. These generally promote signaling in the context of TCR activation.29 The proteins B7.1 and B7.2 are prime examples.103 These are upregulated on APC after exposure to Tlr signaling or CD40L, and bind to CD28 on T cells. In the absence of adequate costimulatory signaling a T cell will become unresponsive to its cognate antigen, a condition called anergy, or may die.31 The B7 proteins are only one of a large group of essential costimulatory molecules expressed on APC. However, costimulatory molecules need not transduce a positive signal into a T cell. Examples include PD-1L and B7h3, which when expressed on an APC downmodulate T-cell signaling and T-cell responses.104,105 T cells may also express variant ligands for a single costimulatory molecule. Thus CD28 transduces a strong growth and survival signal when it engages B7 ligand. But a similar molecule, CTLA-4, when expressed on T cells, binds the same B7 ligands but downmodulates T-cell activation.106,107 Mice that lack CTLA4 develop uncontrolled lymphoproliferation. Temporal regulation of CD28 and CTLA-4 on a T cell may specify the magnitude of the T-cell response. Whereas most T cells express CD28 constitutively, they only express CTLA-4 after activation. Thus initial antigen exposure provides a strong signal to T cells. This signal is downmodulated by CTLA-4 if the exposure persists. Not all T cells are involved in choreographing and mediating immune effector functions. The dominant role of several classes of T cells is the suppression of immune responses. These antagonize the pro-inflammatory activities of effector T cells. One important category of regulatory T cells co-expresses CD4 and the high-affinity receptor for IL-2, CD25.108,109 Development of these cells is defined by the expression of a DNA-binding protein, or transcription factor, called FoxP3. Patients with the IPEX syndrome (immune polyendocrinopathy, X-linked) have mutations in this transcription factor and lack these regulatory T cells.110 Overwhelming multiorgan autoimmunity develops early in life. When stimulated by antigen, FoxP3-expressing regulatory T cells inhibit T-cell responses both through direct cell–cell contact effects and through the production of cytokines, particularly IL-10 and TGF-β. Other classes of regulatory cells, which have been given different names by investigators, including Th3 and Tr1 cells, primarily inhibit immune responses through the production of regulatory cytokines.99,101,111 Again, TGF-β and IL-10 seem to be critical. The complexity to immune regulation parallels the complexity of immune activation. At any point in time the integration of many different signals, some promoting activation and others signaling restraint, will define the ultimate direction taken.

B-Lymphocyte Responses B lymphocytes are critical effector cells for immune protection. They act through the production of immunoglobulin, or antibodies. Antibodies specifically bind to target antigens. By so doing, they may sterically block the function of some antigens, such as toxins, enzymes, and receptors. Antibody binding may promote complement fixation and thereby the lysis or phagocytosis of cells. Antibody–antigen complexes may crosslink

complex. This will be degraded and presented by the B cell on the surface of its class II MHC molecules. Because B cells internalize antigens almost exclusively through their BCR, they are able to selectively capture and concentrate their specific antigens from the extracellular environment. If a T cell that has been stimulated by its cognate antigen on a dendritic cell or other APC then discovers the same antigen on a B cell, the cells will transiently join together and stimulate each other.118 The B lymphocyte along with the T lymphocyte will proliferate, expanding typically at the interface between the T and B regions of the lymph node. The epitope recognized by a B lymphocyte’s antigen receptor need not be the same as that recognized by the TCR of the T cell that stimulates it. When the BCR binds to and endocytoses antigen, it may internalize whole proteins; complexes containing multiple proteins, DNA, lipids, and polysaccharides; or even whole organisms. All of this will be digested and presented to T cells on the B cell’s class II MHC. The antigenic epitope recognized by the T cell therefore does not need to be identical to that recognized by the B cell, but it must be co-internalized and therefore linked to it. The help provided to B cells by T cells is thus often termed linked recognition.119 Stimulated B cells that secrete antibodies are called plasma cells. These cells functionally transform themselves, upregulating protein synthetic machinery so as to be able to produce large quantities of antibodies. They localize to the medulla, or central core of the lymph nodes, or specific regions of the spleen or bone marrow, where they secrete copious amounts of specific antibodies.120,121 Some stimulated B cells will migrate into the B-cell rich follicles of the lymph node or spleen and form what is called a germinal center.122 Germinal centers consist of large numbers of proliferating and differentiating antigen-specific B cells accompanied by smaller numbers of supporting T lymphocytes. Within the germinal centers the B cells are continuously stimulated by antigen presented by follicular dendritic cells. They receive further help from T cells, which promotes their survival, growth, and differentiation. Under the influence of these signals, the B cells may isotype switch, or replace the constant region of their immunoglobulin receptor to allow production of antibody isotypes other than IgM. Isotype switching involves the excision of DNA sequence encoding upstream antibody constant domains, placing the rearranged VDJ region in proximity with a different C region. It is fostered by the expression of CD40L on activated T cells, which binds CD40 on B cells. Cytokines, such as IL-4 and IL-6, also promote B-cell differentiation and isotype switching. B cells from patients lacking CD40L or an enzyme required for DNA rearrangement, activation-induced cytidine deaminase, fail to isotype switch and exclusively produce IgM.123,124 B lymphocytes are capable of an additional trick unavailable to T cells. B cells in the germinal center can actually mutate their BCR, a process termed somatic hypermutation.125,126 The mutations are focused in the antigen-binding region of the receptor. Within the germinal center a process of selection occurs for B cells based on their affinity for antigen. As antigen recedes, the rare B cells in which immunoglobulin mutations have increased their affinity for antigen will survive best, whereas B cells that less effectively bind, internalize, and concentrate antigen will fail to receive necessary survival signals and will perish. Due to somatic hypermutation, B cells may acquire mutated antibodies that have 100-fold or

PRINCIPLES OF THE IMMUNE SYSTEM CENTRAL TO TRANSFUSION MEDICINE

immunoglobulin receptors, Fc receptors, on the surface of effector cell types, such as NK cells, neutrophils, macrophages, and mast cells.112,113 If the Fc receptor is an activating form of the receptor, it will stimulate functions capable of clearing or neutralizing the antigen through a variety of mechanisms, including phagocytosis, antibody-dependent cell-mediated cytotoxicity (ADCC), cytotoxic granule release, and the generation of reactive oxygen and other cytocidal compounds. The antibody, or humoral, immune response is therefore essential to host defense. In its absence, there is a dramatic increase in the susceptibility to and severity of infections.114 Like the T-cell repertoire, the B-cell repertoire is richly diverse, with billions of possible unique receptors. Like T cells, B cells only express a single receptor on their surface. And like T cells, B cells are not constitutively activated, secreting random immunoglobulins. Rather, they produce soluble immunoglobulins only after being activated through their antigen receptor. Unlike T cells, B cells do not recognize antigen complexed with MHC; they recognize whole antigens. B cells can differentiate to secrete antibodies in the absence of T cells. These responses are called T-independent or helper-independent responses.115,116 There are two types of Tindependent humoral immune responses. In T-independent1 responses, the antigen is able to simultaneously stimulate the B lymphocyte through the BCR and other activating receptors. For example, B cells specific for LPS will receive stimulatory signals both through their antigen receptor and through their LPS-specific Toll-like receptor. This may occur in the setting of a gram-negative bacterial infection. Whereas high concentrations of LPS will polyclonally stimulate all B cells through Tlr4, low doses will only simulate those T-independent-1 B cells that are able to concentrate the LPS with their BCR. The combined Toll-like receptor/BCR signal promotes antibody secretion. In T-independent-2 responses, the antigen is not intrinsically stimulatory, but acts by crosslinking a threshold number of BCRs on the B lymphocyte. Too strong a BCR stimulus can turn off a B lymphocyte, and too small a stimulus will not adequately activate a B cell. However, intermediate levels of stimulus may be sufficient to activate subgroups of B lymphocytes, most prominently B cells that express the CD5 antigen (B-1 B cells) and marginal zone B cells. Often the immune response to T-independent antigens is directed against polysaccharides and other polyvalent structures present in the cell walls of pathogens. The T-independent response may allow an early antibody response against these antigens. This antibody may promote internalization of infecting organisms by phagocytes and antigen presentation to T lymphocytes. Activation of the T-cell response may then further drive the B-cell response through the provision of T-cell help in the form of cytokines and other signaling molecules. Patients with Wiskott-Aldrich syndrome fail to produce T-independent responses to bactetrial antigens and consequentially are at great risk for infection with encapsulated bacteria.117 T-independent immunoglobulin responses are primarily of the IgM class of antibodies (described in the following section). Further differentiation of the B cell is required to produce other types of antibody, and T-cell help must be provided for this. Generally, T-cell help is available to promote B-cell responses. As B cells enter the spleen or lymph node from the circulation, they initially enter T-cell rich areas. If the B cell has encountered its antigen, either in the circulation or in the lymphatic tissue, it will internalize the BCR–antigen

2 25

BLOOD BANKING

higher affinity for antigen than their antecedent antibodies. Affinity-matured B cells may then form plasma cells, secreting high-affinity antibody. Alternatively, high-affinity B cells may become quiescent, forming what is called a memory B cell. Memory cells are long-lived formerly activated lymphocytes, either T or B, capable of rapid response on restimulation.127 Because of the presence of memory cells, the antibody response on secondary stimulation is of a higher magnitude and affinity than the primary response that develops after initial exposure to antigen. Antibodies Antibodies come in several forms, each of which has the same basic structure, consisting of two linked dimers of a heavy and light chain. Each of the chains includes tandem globular domains called immunoglobulin domains that are conserved within the immunoglobulin superfamily of proteins (Fig. 2–8).128 The heavy chain is composed of a variable region encompassing the V, D, and J segments, and a C region with three or four immunoglobulin domains depending on the antibody isotype. This heavy chain is linked through a disulfide bond (two cysteines bound together through their sulfur moiety) to a light chain. The light chain consists of V and J segments linked to a constant region with a single immunoglobulin domain. Two identical heavy-light chain dimers are fastened by disulfide linkages between the two heavy chain genes adjacent to the second immunoglobulin domain. Each heavy and light chain pair may independently bind antigen, and a single immunoglobulin monomer thus has a valency for antigen of 2.

II 26

VL

CL

Figure 2–8 Structure of a prototypical, monomeric, secreted antibody molecule. Ovals represent immunoglobulin domains. Heavy chains are indicated by white fill, light chains by gray fill. Polypeptide linkages are indicated by lines. A single representative disulfide linkage between the hinge regions of the heavy chain is indicated; other interchain and intrachain disulfide bonds are not shown.

Different functional regions are encoded within immunoglobulin molecules. The Fv consists of the paired heavy and light chain variable domains. Fv are not independently stable when separated from the rest of the immunoglobin molecule; however, engineered single chain Fv in which the heavy and light variable domains are genetically linked have been created and are capable of binding antigen. These engineered scFv may become increasingly common as the use of engineered antibodies for therapy and diagnostics becomes more routine.129–131 The addition of the first constant region domain to the Fv domain incorporates a disulfide linkage between the heavy and light chains and stabilizes the Fv. These Fab (fraction antibody binding) retain the antigen-binding properties of the parent immunoglobulin. Fab fragments may be released from immunoglobulin molecules by protease digestion. The final two or three constant regions of the heavy chain form the Fc domain. Where the Fab domains connect to the Fc stalk, there is a flexible hinge that allows each Fab domain to independently bend. This helps the antibody bind to multivalent antigens, where the spacing between individual epitopes may vary. The Fc binds to immunoglobulin-binding receptors on cells, Fc receptors, and is therefore responsible for ADCC and other cell-mediated activities. The Fc also binds complement and participates in the activation of this effector pathway. Fab and Fv antibody fragments lack the Fc domain and therefore lack these activities. Immunoglobulin heavy chains come in five basic varieties—μ, γ, δ, ε, and α. When linked to either of the two varieties of light chain—κ or λ—these form IgM, IgG, IgD, IgE, and IgA, respectively, the different isotypes of immunoglobulin. Each of the immunoglobulin isotypes come in a membranebound form expressed on the surface of B cells. A transmembrane domain pierces the cell membrane, attaching the immunoglobulin to the cell surface. Transmembrane forms of immunoglobulin associate with cell membrane proteins, called Igα and Igβ, which allow BCR signaling in a manner analogous to the CD3 chains of the TCR. All isotypes except IgD may also be secreted, in which case the antibody protein is synthesized without the transmembrane anchor. The role of IgD is poorly characterized. Cells expressing IgD also express an IgM version of the same antibody. The secreted forms of different antibodies differ structurally. Secreted IgM consists of five immunoglobulin molecules that are linked together by a junctional chain, forming a pentamer with 10 antigen-binding sites. IgA may be secreted as a single immunoglobulin, and this is its most common form in plasma. However, IgA produced by plasma cells underlying epithelia in the gut, respiratory, or reproductive tracts typically is dimeric, and therefore antibodies secreted into these locations have four antigen-binding sites.132 The remaining immunoglobulin isotypes are only produced as monomers with two antigen-binding sites. Soluble antibody can bind cognate antigen in solution. The binding affinity an antibody has for its antigen is related to the chemical energy of binding of the Fv domain with its antigen. When an antigen is polyvalent, however, the ability of immunoglobulin to multimerize with its respective antigen will significantly increase its ability to bind the antigen. In this circumstance dissociation of the antibody from antigen would not simply require the dissociation of a single interaction, but would require all Fv regions to simultaneously disconnect from the antigenic particle. This increases net binding capacity, or avidity, of immunoglobulin molecules.

Ig Class IgM IgG IgG1 IgG2 IgG3 IgG4 IgA IgE

Conc (mg/dL)

No. of Ab

% Intravascular

Half-life (d)

Complement Fixation (− to ++++)

Agglutination Capacity

Placental Transport

Epithelial Transport

120–400 800–1600

5 1 1 1 1 1 1–2 1

41 48

5–6 18–23

++++

++++ +/−

− +

+ −

++ −

− −

++++ −

40–220 17–450 ng/mL

76 51

5–6.5 2.3

Increased avidity is particularly important for IgM. IgM is formed early in the course of an immune response, and because it will not have undergone somatic hypermutation, is typically of low affinity. This is compensated for by its high valency and thereby enhanced avidity. As described, naive B cells secrete IgM when stimulated to produce antibody. A minimal requirement for B cells to switch isotypes is the engagement of cell surface CD40 by CD40L on T cells. The isotype the B cell converts to depends on the inflammatory environment at the time of isotype switching. Cytokines are particularly important. IFN-γ will promote formation of subclasses of IgG that efficiently bind complement. IL-4 will promote IgE formation. TGF-β promotes the formation of IgA.133–135 These different antibody isotypes and even subclasses within isotypes have distinct functional properties, including complement fixation, Fc-receptor binding, and transport across epithelial barriers or the placenta. Some of these properties are listed in Table 2–3. All antibodies are not equivalent, and the properties an antibody possesses will define its functional capabilities. These properties include the antibody isotype and valency, affinity for antigen, temperature dependence in binding, and titer. For instance, IgM cold agglutinins generally bind poorly at physiologic temperatures and require high titers for clinically significant effects. Some antibody isotypes are most efficient at fixing complement, and anti-RBC alloantibodies of these isotypes are of most concern for intravascular hemolytic reactions. Only IgG isotype anti-RBC antibodies pass the placenta and cause hemolytic disease of the newborn. IgE antibodies bind to specific IgE receptors on mast cells and when crosslinked activate the mast cells to secrete histamine and other BRMs that generate allergic symptoms. Sometimes we can take advantage of the properties of an antibody. For example, the rate of IgM diffusion into the extravascular space is low compared with IgG. This facilitates IgM removal by plasmapheresis.

CONCLUSION The immune system is a dynamic web of interacting cells and molecules that protect our bodies from harm. In the practice of transfusion medicine, we may only become aware of the immune system when our patients manifest symptoms, such as when they become alloimmunized or develop allergic reactions. However, underlying these responses are not only endeffector molecules, cytokines, or antibodies, but also complex pathways of innate and adaptive immunity involving many

+++ +/− +++ − +/− −

cell types and continuous intercellular dialogue. This dialogue occurs even when we fail to clinically observe an immunologic consequence of transfusion. It even occurs within stored blood components, which respond to their storage conditions by producing cytokines and other BRMs that are then infused into patients.136,137 The complexity of the immune response leads to unpredictability in the effects of a transfusion. We still cannot predict who will or will not have an adverse outcome to a specific transfusion event. Nevertheless, a continuous conscientiousness toward the nature of the immune response, how the immune response may perturb the intended outcomes of our transfusions, and how transfusion may perturb the immune system itself, can only help inform the clinical decisions we make.

PRINCIPLES OF THE IMMUNE SYSTEM CENTRAL TO TRANSFUSION MEDICINE

Table 2–3 Properties of Immunoglobulin Classes138

REFERENCES 1. Silverstein AM. Autoimmunity versus horror autotoxicus: the struggle for recognition. Nat Immunol 2001;2:279–281. 2. Silverstein AM. The History of immunology. In Paul WE (ed). Fundamental Immunology. New York, Raven Press, 1989, pp 21–38. 3. Owen R. Karl Landsteiner and the first human marker locus. Genetics 2000;155:995–998. 4. Ohashi PS, Oehen S, Buerki K, et al. Ablation of tolerance and induction of diabetes by virus infection in viral antigen transgenic mice. Cell 1991;65:305–317. 5. Soldavila G, Geiger T, Flavell R. Breaking immunologic ignorance to an antigenic peptide of simian virus 40 large T antigen. J Immunol 1995; 155:5590–5600. 6. Krutzik SR, Modlin RL. The role of Toll-like receptors in combating mycobacteria. Semin Immunol 2004;16:35–41. 7. Selsted ME, Ouellette AJ. Mammalian defensins in the antimicrobial immune response. Nat Immunol 2005;6:551–557. 8. Janeway CA Jr, Medzhitov R. Innate immune recognition. Annu Rev Immunol 2002;20:197–216. 9. Miller SI, Ernst RK, Bader MW. LPS, TLR4 and infectious disease diversity. Annu Rev Immunol 2005;3:36–46. 10. Iwasaki A, Medzhitov R. Toll-like receptor control of the adaptive immune responses. Nat Immunol 2004;5:987–995. 11. Murphy PM. The molecular biology of leukocyte chemoattractant receptors. Annu Rev Immunol 1994;12:593–633. 12. Hedrick JA, Zlotnik A. Chemokines and lymphocyte biology. Curr Opin Immunol 1996;8:343–347. 13. Munoz-Fernandez MA, Fernandez MA, Fresno M. Synergism between tumor necrosis factor-α and interferon-γ on macrophage activation for the killing of intracellular Trypanosoma cruzi through a nitric oxidedependent mechanism. Eur J Immunol 1992;22:301–307. 14. Trinchieri G. Interleukin-12 and the regulation of innate resistance and adaptive immunity. Nat Rev Immunol 2003;3:133–146. 15. Takeda K, Akira S. Toll-like receptors in innate immunity. Int Immunol 2005;17:1–14. 16. Lecellier CH, Dunoyer P, Arar K, et al. A cellular microRNA mediates antiviral defense in human cells. Science 2005;308:557–560. 17. Gao LH, Zheng SS. Cytomegalovirus and chronic allograft rejection in liver transplantation. World J Gastroenterol 2004;10:1857–1861.

2 27

BLOOD BANKING

II 28

18. Cainelli F, Vento S. Infections and solid organ transplant rejection: a cause-and-effect relationship? Lancet Infect Dis 2002;2:539–549. 19. Falcone M, Bloom BR. A T helper cell 2 (Th2) immune response against non-self antigens modifies the cytokine profile of autoimmune T cells and protects against experimental allergic encephalomyelitis. J Exp Med 1997;185:901–907. 20. Stohlman SA, Pei L, Cua DJ, et al. Activation of regulatory cells suppresses experimental allergic encephalomyelitis via secretion of IL-10. J Immunol 1999;163:6338–6344. 21. Aharoni R, Teitelbaum D, Sela M, Arnon R. Bystander suppression of experimental autoimmune encephalomyelitis by T cell lines and clones of the Th2 type induced by copolymer 1. J Neuroimmunol 1998;91:135–146. 22. Marciani DJ. Vaccine adjuvants: Role and mechanisms of action in vaccine immunogenicity. Drug Discov Today 2003;8:934–943. 23. Toyonaga B, Mak T. Genes of the T-cell antigen receptor in normal and malignant T cells. Annu Rev Immunol 1987;5:585–620. 24. Chothia C, Boswell DR, Lesk AM. The outline structure of the T cell αβ receptor. EMBO 1988;7:3745–3755. 25. Malissen M, Trucy J, Jouvin-Marche E, et al. Regulation of TCR alpha and beta gene allelic exclusion during T-cell development. Immunol Today 1992;13:315–322. 26. Tse H, Schwarz R, Paul W. Cell–cell interactions in the T cell proliferative response. J Immunol 1980;125:401. 27. Carbone FR, Belz GT, Heath WR. Transfer of antigen between migrating and lymph node-resident DCs in peripheral T-cell tolerance and immunity. Trends Immunol 2004;25:655–658. 28. Man AL, Prieto-Garcia ME, Nicoletti C. Improving M cell mediated transport across mucosal barriers: do certain bacteria hold the keys? Immunology 2004;113:15–22. 29. Frauwirth KA, Thompson CB. Activation and inhibition of lymphocytes by costimulation. J Clin Invest 2002;109:295–299. 30. Guermonprez P, Valladeau J, Zitvogel L, et al. Antigen presentation and T cell stimulation by dendritic cells. Annu Rev Immunol 2002;20:621–667. 31. Schwartz RH. T cell anergy. Annu Rev Immunol 2003;21:305–334. 32. Heath WR, Carbone FR. Cross-presentation, dendritic cells, tolerance and immunity. Annu Rev Immunol 2001;19:47–64. 33. Malissen B, Ardouin L, Lin SY, et al. Function of the CD3 subunits of the pre-TCR and TCR complexes during T cell development. Adv Immunol 1999;72:103–148. 34. Davis MM, Bjorkman PJ. T-cell antigen receptor genes and T-cell recognition. Nature 1988;334:395–402. 35. Sprent J. T lymphocytes and the thymus. In Paul W (ed). Fundamental Immunology. New York, Raven Press, 1989, pp 69–93. 36. Campbell R, Trowsdale J. Map of the human MHC. Immunol Today 1993;14:349–352. 37. Bjorkman PJ, Saper MA, Samraoui B, et al. The foreign antigen binding site and T cell recognition regions of class I histocompatibility antigens. Nature 1987;329:512–518. 38. Bjorkman PJ, Saper MA, Samraoui B, et al. Structure of the human class I histocompatibility antigen, HLA-A2. Nature 1987;329:506–512. 39. Fremont D, Hendrikson W, Marrack P, Kappler J. Structures of an MHC class II molecule with covalently bound single peptide. Science 1996;272:1001–1004. 40. Falk K, Rotzsche O, Stevanovic S, Jung G, Rammensee H. Allele specific motifs revealed by sequencing of self peptides eluted from MHC molecules. Nature 1991;351:290–296. 41. Hunt D, Henderson R, Shabanowitz J, et al. Characterization of peptides bound to the class I molecule HLA-A2.1 by mass spectrometry. Science 1992;255:1261–1263. 42. Parham P, Ohta T. Population biology of antigen presentation by MHC class I molecules. Science 1996;272:67–74. 43. Sherman LA, Chattopadhyay S. The molecular basis of allorecognition. Annu Rev Immunol 1993;11:385–402. 44. Groettrup M, Soza A, Kuckelkorn U, Kloetzel P. Peptide antigen production by the proteasome: complexity provides efficiency. Immunol Today 1996;17:429–435. 45. Groettrup M, Khan S, Schwarz K, Schmidtke G. Interferon-gamma inducible exchanges of 20S proteasome active site subunits: why? Biochimie 2001;83:367–372. 46. Lehner P, Cresswell P. Processing and delivery of peptides presented by MHC class I molecules. Curr Opin Immunol 1996;8:59–67. 47. Koopmann J, Post M, Neefjes J, et al. Translocation of long peptides by transporters associated with antigen processing (TAP). Eur J Immunol 1996;26:1720–1728. 48. Sadasivan B, Lehner P, Ortmann B, et al. Roles for calreticulin and a novel glycoprotein, tapasin, in the interaction of class I MHC molecules with TAP. Immunity 1996;5:103–114.

49. Bakke O, Doberstein B. MHC class II associated invariant chain contains a sorting signal for endosomal compartments. Cell 1990;63:707–716. 50. Amigorena S, Drake J, Webster P, Mellman I. Transient accumulation of new MHC molecules in a novel endocytic compartment in B lymphocytes. Nature 1994;369:113–120. 51. Castellino F, Germain R. Extensive trafficking of MHC class II invariant chain complexes in the endocytic pathway and appearance of peptide loaded class II in multiple compartments. Immunity 1995;2:73–88. 52. Sallusto F, Sella M, Danieli C, Lanzavecchia A. Dendritic cells use macropinocytosis and the mannose receptor to concentrate macromolecules in the major histocompatibility complex class II compartment: downregulation by cytokines and bacterial products. J Exp Med 1995;182:389–400. 53. Bonnerot C, Lankar D, Hanau D, et al. Role of B cell receptor Ig α and β subunits in MHC class II-restricted antigen presentation. Immunity 1995;3:335–347. 54. Lanzavecchia A. Mechanisms of antigen uptake for presentation. Curr Opin Immunol 1996;8:348–354. 55. Sykulev Y, Brunmark A, Jackson M, et al. Kinetics and affinity of reactions between an antigen-specific TCR and peptide–MHC complexes. Immunity 1994;1:15–22. 56. Matsui K, Boniface J, Steffner P, et al. Kinetics of T cell receptor binding to peptide/IEk complexes: correlation of the dissociation rate with Tcell responsiveness. Proc Natl Acad Sci USA 1994;91:12862–12866. 57. Janeway CA Jr. The T cell receptor as a multicomponent signalling machine: CD4/CD8 coreceptors and CD45 in T cell activation. Annu Rev Immunol 1992;10:645–674. 58. Li QJ, Dinner AR, Qi S, et al. CD4 enhances T cell sensitivity to antigen by coordinating Lck accumulation at the immunological synapse. Nat Immunol 2004;5:791–799. 59. Weiss A, Littman D. Signal transduction by lymphocyte antigen receptors. Cell 1994;76:263–274. 60. Demotz S, Grey HM, Sette A. The minimal number of class II MHC– antigen complexes needed for T cell activation. Science 1990;249: 1028–1030. 61. Starr TK, Jameson SC, Hogquist KA. Positive and negative selection of T cells. Annu Rev Immunol 2003;21:139–176. 62. Boehm U, Klamp T, Groot M, Howard JC. Cellular responses to interferon-γ. Annu Rev Immunol 1997;15:749–795. 63. Wilson HL, O’Neill HC. Murine dendritic cell development: difficulties associated with subset analysis. Immunol Cell Biol 2003;81:239–246. 64. Park CS, Choi YS. How do follicular dendritic cells interact intimately with B cells in the germinal centre? Immunology 2005;114:2–10. 65. Huang AY, Qi H, Germain RN. Illuminating the landscape of in vivo immunity: insights from dynamic in situ imaging of secondary lymphoid tissues. Immunity 2004;21:331–339. 66. Kabelitz D, Holtmeier W. γδ T cells link innate and adaptive immune responses. Chem Immunol Allergy 2005;86:151–183. 67. Shin S, El-Diwany R, Schaffert S, et al. Antigen recognition determinants of γδ T cell receptors. Science 2005;308:252–255. 68. Brandes M, Willimann K, Moser B. Professional antigen-presentation function by human γδT cells. Science 2005;309:264–268. 69. Mackay CR. Migration pathways and immunologic memory among T lymphocytes. Semin Immunol 1992;4:51–8. 70. Mossman T, Coffman R. Th1 and Th2 cells: different patterns of lymphokine secretion lead to different functional properties. Annu Rev Immunol 1989;7:145–173. 71. Abbas A, Murphy M, Sher A. Functional diversity of helper T lymphocytes. Nature 1996;383:787–793. 72. Romagnani S. Type 1 T helper and type 2 T helper cells: functions, regulation and role in protection and disease. Int J Clin Lab Res 1991;21:152–158. 73. Cher D, Mossman T. Two types of murine helper T cell clone. II. Delayed type hypersensitivity is mediated by Th1 clones. J Immunol 1987;138:3688–3694. 74. Coffman R, Seymour B, Lebman D, et al. The role of helper T cell products in mouse B cell differentiation and isotype regulation. J Immunol 1988;102:5–28. 75. Bernhagen J, Bacher M, Calandra T, et al. An essential role for macrophage migration inhibitory factor in the tuberculin delayed-type hypersensitivity reaction. J Exp Med 1996;183:277–282. 76. Grabbe S, Schwarz T. Immunoregulatory mechanisms involved in elicitation of allergic contact hypersensitivity. Immunol Today 1998; 19:37–44. 77. Siraganian RP. Mast cell signal transduction from the high-affinity IgE receptor. Curr Opin Immunol 2003;15:639–46. 78. Capron A, Dessaint JP. Immunologic aspects of schistosomiasis. Annu Rev Med 1992;43:209–218.

109. Wraith DC, Nicolson KS, Whitley NT. Regulatory CD4+ T cells and the control of autoimmune disease. Curr Opin Immunol 2004;16: 695–701. 110. Bennett CL, Christie J, Ramsdell F, et al. The immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) is caused by mutations of FOXP3. Nat Genet 2001;27:20–21. 111. Roncarolo MG, Bacchetta R, Bordignon C, et al. Type 1 T regulatory cells. Immunol Rev 2001;182:68–79. 112. Ravetch JV, Bolland S. IgG Fc receptors. Annu Rev Immunol 2001; 19:275–290. 113. Ravetch JV. Fc receptors. Curr Opin Immunol 1997;9:121–125. 114. Conley ME. Early defects in B-cell development. Curr Opin Allergy Clin Immunol 2002;2:517–522. 115. Fagarasan S, Honjo T. T-Independent immune response: new aspects of B cell biology. Science 2000;290:89–92. 116. Mond JJ, Lees A, Snapper CM. T cell-independent antigens type 2. Annu Rev Immunol 1995;13:655–692. 117. Rijkers GT, Sanders LA, Zegers BJ. Anti-capsular polysaccharide antibody deficiency states. Immunodeficiency 1993;5:1–21. 118. Heyzer-Williams LJ, Heyzer-Williams MG. Antigen-specific memory B cell development. Annu Rev Immunol 2005;23:487–513. 119. Parker DC. T cell-dependent B cell activation. Annu Rev Immunol 1993;11:331–360. 120. Cyster JG. Homing of antibody secreting cells. Immunol Rev 2003; 194:48–60. 121. Kunkel EJ, Butcher EC. Plasma-cell homing. Nat Rev Immunol 2003; 3:822–829. 122. Cozine CL, Wolniak KL, Waldschmidt TJ. The primary germinal center response in mice. Curr Opin Immunol 2005;17:298–302. 123. Durandy A, Revy P, Imai K, Fischer A. Hyper-immunoglobulin M syndromes caused by intrinsic B-lymphocyte defects. Immunol Rev 2005;203:67–79. 124. Fuleihan RL. The X-linked hyperimmunoglobulin M syndrome. Semin Hematol 1998;35:321–331. 125. Maizels N. Immunoglobulin gene diversification. Annu Rev Genet 2004;39:22–46. 126. Wu X, Feng J, Komori A, et al. Immunoglobulin somatic hypermutation: double-strand DNA breaks, AID and error-prone DNA repair. J Clin Immunol 2003;23:235–246. 127. Gourley TS, Wherry EJ, Masopust D, Ahmed R. Generation and maintenance of immunological memory. Semin Immunol 2004;16: 323–333. 128. Williams AF, Barclay AN. The immunoglobulin superfamily—domains for cell surface recognition. Annu Rev Immunol 1988;6:381–405. 129. Bilbao G, Contreras JL, Curiel DT. Genetically engineered intracellular single-chain antibodies in gene therapy. Mol Biotechnol 2002;22: 191–211. 130. Hombach A, Heuser C, Abken H. The recombinant T cell receptor strategy: insights into structure and function of recombinant immunoreceptors on the way towards an optimal receptor design for cellular immunotherapy. Curr Gene Ther 2002;2:211–226. 131. Adams GP, Schier R. Generating improved single-chain Fv molecules for tumor targeting. J Immunol Methods 1999;231:249–260. 132. Mestecky J, Lue C, Russell MW. Selective transport of IgA. Cellular and molecular aspects. Gastroenterol Clin North Am 1991;20:441–471. 133. Oettgen HC. Regulation of the IgE isotype switch: new insights on cytokine signals and the functions of epsilon germline transcripts. Curr Opin Immunol 2000;12:618–623. 134. Purkerson J, Isakson P. A two-signal model for regulation of immunoglobulin isotype switching. FASEB J 1992;6:3245–3252. 135. Lebman DA, Edmiston JS. The role of TGF-β in growth, differentiation, and maturation of B lymphocytes. Microbes Infect 1999; 1:1297–1304. 136. Stack G, Baril L, Napychank P, Snyder EL. Cytokine generation in stored, white cell-reduced, and bacterially contaminated units of red cells. Transfusion 1995;35:199–203. 137. Stack G, Snyder EL. Cytokine generation in stored platelet concentrates. Transfusion 1994;34:20–25. 138. Bawa N, Tomar RH. Laboratory evaluation of immunoglobulin function and humoral immunity. In Henry JB (ed). Clinical Diagnosis and Management by Laboratory Methods. Philadelphia, W.B. Saunders, 1996, pp 913–927.

PRINCIPLES OF THE IMMUNE SYSTEM CENTRAL TO TRANSFUSION MEDICINE

79. Grencis RK. Th2-mediated host protective immunity to intestinal nematode infections. Philos Trans R Soc Lond B Biol Sci 1997;352:1377–1384. 80. Kamogawa Y, Minasi L, Carding S, et al. The relationship of IL-4 and IFN gamma producing T cells studied by lineage ablation of IL-4 producing cells. Cell 1993;75:985–995. 81. Constant S, Bottomly K. Induction of Th1 and Th2 responses: the alternative approaches. Annu Rev Immunol 1997;15:297–322. 82. Dvorak AM. New aspects of mast cell biology. Int Arch Allergy Immunol 1997;114:1–9. 83. Yoshimoto T, Paul W. CD4pos, NK1.1pos T cells promptly produce interleukin 4 in response to in vivo challenge with anti-CD3. J Exp Med 1994;179:1285–1295. 84. MacDonald HR. Development and selection of NKT cells. Curr Opin Immunol 2002;14:250–254. 85. Langrish CL, McKenzie BS, Wilson NJ, et al. IL-12 and IL-23: master regulators of innate and adaptive immunity. Immunol Rev 2004; 202:96–105. 86. Billiau A, Heremans H, Vermeire K, Matthys P. Immunomodulatory properties of interferon-γ. An update. Ann NY Acad Sci 1998;856: 22–32. 87. Smith CM, Wilson NS, Waithman J, et al. Cognate CD4+ T cell licensing of dendritic cells in CD8+ T cell immunity. Nat Immunol 2004;5:1143– 1148. 88. Behrens G, Li M, Smith CM. Helper T cells, dendritic cells and CTL immunity. Immunol Cell Biol 2004;82:84–90. 89. van KC, Banchereau J. CD40-CD40 ligand. J Leukoc Biol 2000;67:2–17. 90. Berke G. The CTL’s kiss of death. Cell 1995;81:9–12. 91. Ueda I, Morimoto A, Inaba T. Characteristic perforin gene mutations of haemophagocytic lymphohistiocytosis patients in Japan. Br J Haematol 2003;121:503–510. 92. Grunebaum E, Roifman CM. Gene abnormalities in patients with hemophagocytic lymphohistiocytosis. Isr Med Assoc J 2002;4:366–369. 93. Russell JH. Activation-induced death of mature T cells in the regulation of immune responses. Curr Opin Immunol 1995;7:382–388. 94. de Alborán, IM, Robles MS, Bras A, et al. Cell death during lymphocyte development and activation. Semin Immunol 2003;15:125–133. 95. Green DR, Droin N, Pinkoski M. Activation-induced cell death in T cells. Immunol Rev 2003;193:70–81. 96. Lenardo M, Chan KM, Hornung F, et al. Mature T lymphocyte apoptosis—immune regulation in a dynamic and unpredictable antigenic environment. Annu Rev Immunol 1999;17:221–253. 97. Ferber IA, Brocke S, Taylor-Edwards C, et al. Mice with a disrupted IFN-γ gene are susceptible to the induction of experimental autoimmune encephalomyelitis (EAE). J Immunol 1996;156:5–7. 98. Willenborg DO, Fordham SA, Staykova MA, IFN-γ is critical to the control of murine autoimmune encephalomyelitis and regulates both in the periphery and in the target tissue: a possible role for nitric oxide. J Immunol 1999;163:5278–5286. 99. Battaglia M, Gianfrani C, Gregori S, Roncarolo MG. IL-10-producing T regulatory type 1 cells and oral tolerance. Ann NY Acad Sci 2004;1029:142–153. 100. Grutz G. New insights into the molecular mechanism of interleukin10-mediated immunosuppression. J Leukoc Biol 2005;77:3–15. 101. O’Garra A, Vieira P. Regulatory T cells and mechanisms of immune system control. Nat Med 2004;10:801–805. 102. Rennick DM, Fort MM. Lessons from genetically engineered animal models. XII. IL-10-deficient (IL-10(-/-) mice and intestinal inflammation. Am J Physiol Gastrointest Liver Physiol 2000;278:G829–G833. 103. Lenschow DJ, Walunas TL, Bluestone JA. CD28/B7 system of T cell costimulation. Annu Rev Immunol 1996;14:233–258. 104. Loke P, Allison JP. Emerging mechanisms of immune regulation: the extended B7 family and regulatory T cells. Arthritis Res Ther 2004;6:208–214. 105. Greenwald RJ, Freeman GJ, Sharpe AH. The B7 family revisited. Annu Rev Immunol 2005;23:515–548. 106. Nakaseko C, Miyatake S, Iida T, et al. Cytotoxic T lymphocyte antigen 4 (CTLA-4) engagement delivers an inhibitory signal through the membrane-proximal region in the absence of the tyrosine motif in the cytoplasmic tail. J Exp Med 1999;190:765–774. 107. Oosterwegel MA, Greenwald RJ, Mandelbrot DA, et al. CTLA-4 and T cell activation. Curr Opin Immunol 1999;11:294–300. 108. Fontenot JD, Rudensky AY. A well adapted regulatory contrivance: regulatory T cell development and the forkhead family transcription factor Foxp3. Nat Immunol 2005;6:331–337.

2 29

Chapter 3

Principles of the Complement System Central to Transfusion Medicine Dana V. Devine

II 30

The existence of the complement system has great impact on the practice of transfusion medicine. In its normal immune activities, complement functions to kill pathogens, mediate inflammation, maintain the solubility of immune complexes, promote a normal adaptive immune response, and opsonize particles for phagocytosis. However, complement can also mediate pathogenic processes, including anaphylaxis, intravascular hemolysis of transfused blood cells, leukocyte mobilization in transfusion-related acute lung injury (TRALI), and activation of platelets. The group of proteins known to constitute the complement system was first recognized in the 1880s as the labile bactericidal activity in serum.1 Paul Ehrlich coined the term for the phenomenon by proposing a model of antibodymediated cytotoxicity in which a serum factor complements the bactericidal activity of antibody. A detailed understanding of the biochemistry of the complement system required the development of techniques that would permit the isolation of individual complement proteins; this technology finally appeared in the 1960s. With the explosion of research activity in the complement field in the 1970s came the recognition that this system was much more complex than had been imagined. This complexity is amply demonstrated by the fact that at least 25 complement proteins are involved in the activation and regulation of that activity known to Ehrlich simply as complement. The understanding of the complement system is made easier by setting a proper context. Like coagulation, the complement system is an activated enzyme cascade. In such cascades, proteins normally circulate in an inactive form, the zymogen. When the pathway is initiated, the first protein in the sequence is converted from a zymogen to an activated enzyme, which acts on the next protein zymogen in the cascade. Such pathways are amplifying, because each enzyme molecule generated can act on multiple substrate molecules. Activated enzyme pathways are also characterized by the presence of regulatory proteins, both humoral and cellular, that prevent the activated enzymes from converting all available substrate. The nuances of the complement system have long struck fear into the hearts of basic scientist and clinician alike. Clinical aspects of complement biology in the pathophysiology of disease have been reviewed by Morgan2 and more recently by others.3,4 This chapter describes the central role of complement in many physiologic processes, including those associated with the use of blood components.

BASIC BIOCHEMISTRY OF THE COMPLEMENT SYSTEM Classical Pathway Activation Activation of the complement system occurs via two pathways; with the activation of C3 (the third component of complement), these pathways join to form a common pathway that completes the cascade (Fig. 3–1). The primary function of both pathways is the generation of enzyme complexes that activate C3 by cleaving it to C3b. The antibody-mediated activation of complement occurs by the classical pathway, so called because it was the first pathway recognized. Activators of the classical pathway include not only antibody molecules, but also several nonimmunoglobulin proteins (Table 3–1). Only immunoglobulins (Ig) of the M and G isotypes activate complement by the classical pathway. In humans, IgG3 and IgG1 are strong complement activators, but IgG2 is a poor activator, and IgG4 does not activate complement. These differences result fro m variation in the ability of the different IgG subclasses to bind the first component of complement, C1. The ability of an antibody to activate complement, with the accompanying opsonization and perhaps lysis of the cell, parallels the opsonic potential of the IgGs themselves. The varying risk of phagocytic destruction by crystallizable fragment (Fc) receptor-mediated recognition of IgG is an important feature in distinguishing clinically significant antibodies from those with less destructive potential. C1 is a multisubunit complex that contains the initial antibody-binding subunits, C1q, as well as two types of zymogen subunits, C1r and C1s, that acquire serine protease activity on activation of the complex. Each molecule of C1 contains six C1q subunits, and two each of the C1r and C1s subunits. The fixation of a C1 molecule to the surface of the cell by the C1q subunits requires a minimum of one molecule of IgM or at least two molecules of IgG for efficient activation. C1q itself contains six identical subunits composed of a triple helical region with homology to collagen and a globular domain at the distal end. The proximal end of C1q is associated with the other subunits of the C1 complex, C1r and C1s, in a calcium-dependent manner. Two of the six C1q subunits must be bound by antibody to effect activation. Although a single molecule of IgM is capable of activating complement, it must be bound to antigen, where it assumes a staple-shaped conformation. Fluid-phase or planar IgM does not activate complement, inasmuch as the C1q-binding sites are exposed only in the staple form.5

PRINCIPLES OF THE COMPLEMENT SYSTEM CENTRAL TO TRANSFUSION MEDICINE

Figure 3–1 The activation pathway of complement. The activation of complement may proceed by the classical pathway or the alternative pathway. Inactive (zymogen) forms of complement proteins are cleaved by activated proteins that have serine protease activity (heavy arrows). Once cleaved, a substrate acquires enzymatic activity and acts on the next substrate in the pathway. The membrane attack complex forms by the assembly of the individual components in an enzyme-independent manner. (From Anderson KC, Ness PM. Scientific Basis of Transfusion Medicine, 2nd ed. Philadelphia, Saunders, 2000.)

3 31

Once two subunits of C1q are bound to antibody, the molecular conformation changes, in that the angle between the subunits is greatly reduced; the resulting stress on the molecule facilitates the autocatalysis of the C1r subunits.6 On autoactivation, C1r acquires serine protease activity, which is directed against the C1s subunits. Once C1s is cleaved by activated C1r, it also acquires serine protease activity. This

activated C1 complex is the initiation complex of the classical pathway. Although the majority of classical pathway activation is antibody mediated, antibody-independent activation of the classical pathway has been described in several situations (see Table 3–1). Such activation may involve the direct binding of C1q to a surface, or it may be mediated by other plasma

Table 3–1 Complement Activators Lectin Pathway

Classical Pathway

Alternative Pathway

Surfaces containing N-acetylglucosamine and mannose

IgG1 and IgG3; IgG2 weakly; IgM Some negatively charged surfaces Crystalline cholesterol

Some dialysis membranes, especially cuprophane Desialated erythrocytes Surfaces that promote the binding of factor B IgA complexes Protein aggregates Microbial pathogens Tumor cells

BLOOD BANKING

II

proteins, such as C-reactive protein (CRP) or mannosebinding lectin. Activated C1s has the next two proteins in the pathway, C4 and C2, as its substrates. The C4 molecule is cleaved by C1s to C4a and C4b. Some of the C4b molecules bind covalently to the cell surface through a reactive thiol ester bond that is exposed when the C4a fragment is removed; most C4b molecules are inactivated by hydrolysis and never bind to a cell surface. In a magnesium-dependent reaction, a C2 molecule binds to a molecule of C4b; the C2 is then also cleaved by C1s. This cleavage results in the generation of the active serine protease, C2a, which remains associated with C4b. Clearly, the cleavage of both C4 and C2 by the same molecule of C1s is sensitive to the local geometry; deposition of the C4b molecule far from the activated C1 molecule results in the termination of the activation pathway. The bimolecular complex, C4b2a, is the C3-converting complex of the classical pathway. The cleavage of C3 by C4b2a results in the formation of two fragments, C3a and C3b. C3b, like C4b, contains a reactive thiol ester bond that enables it to bind covalently to the cell surface. The intramolecular thiol ester present within C3 is formed by a transacylation reaction between the thiol group of 988Cys and the gamma amide group of 991Gln.7 The exposure of this reactive center results in the interaction with cell surface moieties through the formation of an ester or an amide bond. As indicated earlier, the activation of the classical pathway does not require immunoglobulin. The most recent developments in complement biochemistry have included an appreciation for the role of molecules of the pentraxin and lectin families in the activation of complement. Pentraxins, a family of molecules named for their cyclic pentameric subunit structure, include two proteins that have been implicated in complement activation: CRP and serum amyloid P.8–10 Both proteins bind C1q, thereby initiating the classical pathway without a requirement for antibody.

32

Lectin Pathway Activation The carbohydrate-binding properties of plant-derived lectins are well known in the field of blood banking, in which they are useful phenotyping reagents. Lectins or lectin-like proteins are also found in mammals. One of the C-type animal lectins, mannose-binding lectin (MBL), belongs to a group of soluble pattern recognition receptors. These molecules and their membrane-bound cousins the Toll-like receptors, recognize pathogen-associated molecular patterns located on infectious organisms. MBL recognizes mannose-containing carbohydrates and also binds N-acetylglucosamine. MBL binding to its substrates has been reported to mediate complement activation.11 The hexameric form of MBL is structurally similar to C1q and will bind the C1r2s2 complex. MBL is found in association with one of three forms of a serine protease, MBL-associated protease (MASP1, MASP2, or MASP3). Complement activation may be mediated by either C1r2s2 or by the direct action of MASPs. The relative biological significance of the pentraxin and lectin pathways is not fully understood.12 However, both CRP and MBL are acutephase reactants and may reach significant serum concentration during infection or inflammation.

Alternative Pathway Activation A different C3-converting complex is generated by the activation of the alternative pathway of complement. The term

alternative arises from the fact that this mechanism of C3 cleavage was discovered many decades after the classical pathway. It should in no way be considered a secondary pathway of complement activation, because it is the only pathway of complement that can respond to microorganisms in the absence of specific antibody. It is, therefore, a front-line host defense. Activators of the alternative pathway include a broad spectrum of substances, from renal dialysis membranes to immune complexes to microorganisms (Table 3–2). The initial step in the activation of the alternative pathway is the generation of a partially activated molecule of C3, which has been described as C3b-like.13 This molecule is presumably generated by the low-grade, spontaneous hydrolysis of C3 that occurs in the body and is not the result of the presence of the activator substance per se. C3(H2O) has the important characteristic of being able to interact in the fluid phase with the complement protein factor B in a magnesium-dependent manner. Once factor B has associated with C3(H2O), it is cleaved by factor D, a circulating serine protease that has specificity for factor B bound to C3(H2O). This cleavage results in the formation of two fragments, Ba and Bb. The Bb fragment has serine protease activity and remains associated with C3(H2O). The bimolecular complex C3(H2O)Bb has C3 as its substrate. The C3(H2O)Bb complex cleaves C3 in the same way as C4b2a, and the resulting C3b molecules can bind covalently to the cell surface. If C3b binds to an activator surface, factor B associates with it and is cleaved by factor D. This results in the formation of an alternative pathway C3-converting complex on the surface of the activator. Because this complex has C3 as its substrate, it produces a feedback amplification loop for the deposition of C3b onto the activator. The binding of a C3b molecule in the immediate vicinity of a C4b2a or C3bBb enzyme complex produces a trimolecular complex that is capable of cleaving C5. The C3b molecule has a binding site for C5; if the geometry is appropriate, C5 is presented to the enzyme complex, where it is cleaved by C2a or Bb serine proteases. The cleavage produces the fragments C5a and C5b. Although there is a great deal of homology among C3, C4, and C5, there is no reactive thiol ester in the C5 molecule. Therefore, the cleavage of C5 does not generate a fragment capable of binding covalently to a cell surface. The C5b molecule remains briefly associated with C3b before its association with the terminal proteins of the complement pathway.

Formation of the Membrane Attack Complex Membrane attack complex of complement refers to the association of the complement proteins C5, C6, C7, C8, and C9 to form a potentially cytolytic complex. When C5 is activated in either the classical or alternative pathway, the resulting C5b molecule contains binding sites for the next components in the pathway. This part of the complement pathway is still a cascade, but the activation of a component results in the exposure of binding sites for other terminal complement proteins rather than in the acquisition of enzymatic activity. C5b, but not C5, contains a binding site for C6, which becomes bound while the C5b molecule is associated with its C3b tether. The C5b6 complex may be released from C3b, or it may remain anchored until it binds a molecule of C7. Once C7 is attached, the trimolecular complex undergoes a transition in which the normal hydrophilic character of the

Pathway

Protein

Molecular Weight (kDa)

Structure

Average Serum Concentration (mg/mL)

Activation complex classical pathway

C1q

Subunit A: 27 Subunit B: 27 Subunit C: 27 Complex: 465 83

6 each of subunits A, B, and C

80

If full complex, C1q + 2 C1r + 2 C1; if no C1q, then 2 C1r + 2 C1s

50

C1r C1s C4

Activation complex alternative pathway Common pathway

Factor B Factor D C3 C5 C6 C7 C8

C9

83 Subunit a: 97 Subunit b: 75 Subunit c: 33 Complex: 205 92 24 Subunit a: 110 Subunit b: 75 Complex: 185 Subunit a: 115 Subunit b: 75 Complex: 190 120 110 Subunit a: 64 Subunit b: 64 Subunit c: 22 Complex: 150 70

individual members of the complex is lost and a transient hydrophobic character is acquired. At this time, C5b-7 may associate with a membrane through the hydrophobic region; if it fails to interact with a membrane, the complex inactivates by self-aggregation or by interaction with the inhibitors described later. Once C5b-7 is membrane associated, a binding site for C8 is exposed in C5b. The binding of C8 causes the complex to insert more deeply into the membrane. C9 binds to C8 and undergoes significant conformational changes. Not only does it acquire considerable hydrophobic character, pushing the membrane attack complex deeper into the membrane, but it also gains affinity for other molecules of C9, which polymerize into the complex. Electron microscopic studies suggest that the C5b-9(n) complex resembles a hollow cylinder. This cylinder is formed by the polymerized C9 molecules, which number between 12 and 18 in each complex. The C5b-8 is thought to play little active role in the structure of the cylindrical pore; it is, however, the essential catalyst for the pore’s formation. The biochemical characteristics of the complement proteins are given in Table 3–2.

The Regulation of Complement Activation Plasma Proteins As discussed previously, the activation of the complement system is expressed as the activation of several enzymes of the serine protease group. As with any zymogen-to-enzyme conversion, there must be a way of regulating the activity of the enzyme, or else it will cleave all available substrate molecules. With the complement proteins, this regulation is achieved by multiple mechanisms. The geometric arrange-

One each of subunits a, b, and c synthesized from a single precursor

50 600

Single chain

210

Single chain One each of subunits a and b synthesized from a single precursor One each of subunits a and b synthesized from a single precursor Single chain Single chain One each of subunits a, b, and c

20% overall identity Share overall homology to one another ranging from 16% to 33% C6 and C7 share greatest homology All undergo amphipathic conversion to acquire membrane-binding character All function to regulate complement activation All have multiple (60–70) amino acid short consensus repeats

C2, factor B

C3, C4, C5

C6, C7, C8, C9

CR1, CR2, factor H, DAF, MCP, C4 binding protein

DAF, delay accelerating factor; MCP, membrane cofactor protein.

THE PHYSIOLOGIC RESPONSE TO COMPLEMENT ACTIVATION Anaphylatoxins The cleavage of C3, C4, or C5 by the enzyme complexes of the alternative or classical pathways results in the formation of two fragments. The fate of the C3b, C4b, and C5b fragments was discussed previously. The other fragments, C3a, C4a, and C5a, are known as the anaphylatoxic fragments of complement. These very small fragments of complement can produce a very large and potentially life-threatening physiologic response. The C3bBb and C4b2a enzyme complexes recognize an Arg-X sequence near the amino terminus of C3, C4, and C5. Cleavage of this bond results in the formation of a small N-terminal peptide of 77 amino acids for C3a and C4a and a small N-terminal peptide of 74 amino acids for C5a, each with a C-terminal arginine residue. Although these peptides are very similar in structure, their potency in mediating cellular responses varies considerably. C5a is the most potent of the complement-derived anaphylatoxins, and C3a is the least potent. These complement fragments exert their anaphylatoxic effects by interacting with specific cellular receptors present on the surface of mast cells and basophils. The occupation of the receptor triggers the release of histamine and serotonin from intracellular granules. These two soluble factors cause contraction of smooth muscle cells and increased vascular permeability of blood vessels. Neutrophils, monocytes, macrophages, and platelets also bind anaphylatoxic fragments of complement; the occupation of these receptors activates the cell and, in the case of neutrophils, induces chemotaxis toward the site of complement activation. Although the anaphylatoxic fragments of C3, C4, and C5 are potent biological response modifiers, they are rapidly inactivated in plasma by the action of a pair of carboxypeptidases. The best known, serum carboxypeptidase N, removes the C-terminal arginine residue from the peptide. Another

Figure 3–2 The domain structure of the terminal complex proteins. This figure illustrates the relative homologies among the proteins of the membrane attack complex. The highly homologous regions are found within cysteine-rich domains types I, II, and III (according to Morgan2). In addition, C6 and C7 contain cysteine-rich short consensus repeats (SCRs). (From Anderson KC, Ness PM. Scientific Basis of Transfusion Medicine, 2nd ed. Philadelphia, Saunders, 2000.)

PRINCIPLES OF THE COMPLEMENT SYSTEM CENTRAL TO TRANSFUSION MEDICINE

cysteine-rich regions. From the amino terminus, using the nomenclature of Morgan,2 the four types of cysteine-rich domains are known as type I, type II, type III, and the SCR (Fig. 3–2). The terminal complement proteins all contain type I cysteine-rich domains of approximately 60 amino acids, as well as other cysteine-rich domains of approximately 40 amino acids that share sequence homology to those in the low-density lipoprotein receptor (types II and III). These cysteine-rich domains, which greatly contribute to the tertiary structure of the protein, are clearly important for function. The SCRs are present in at least 12 of the complement proteins, and 8 proteins contain the other types of cysteinerich domains. C6, C7, and C9 are single-polypeptide chains encoded by single genes; C8 is composed of three polypeptide chains, each of which is encoded by its own gene. The C8 α and β proteins are highly homologous, but the C8 γ chain shows no homology to any complement protein; it shows homology to α1-microglobulin. The proteins involved in the regulation of complement activation are also structurally related. In addition, the genes encoding the proteins that modulate C3 and C4 activation are grouped in one region of the long arm of chromosome 6 known as the regulators of complement activation (RCA) cluster.40 The RCA group includes factor H, C4-binding protein, DAF, MCP, and the complement receptors CR1 and CR2. These proteins contain a variable number of SCRs of 60 to 70 amino acids. The chromosomal assignments of the complement proteins are given in Table 3–5. Molecular biology has also proved invaluable in the identification of polymorphisms of complement proteins that had previously been noted as variants in electrophoretic mobility or antigenicity. Polymorphisms have been described for most of the complement proteins.41,42 Some of the complement protein polymorphisms are associated with a loss or decrease of complement activity. Others demonstrate some degree of disease association; however, no disease associations have been established for any polymorphisms of the terminal complex proteins.

3 37

BLOOD BANKING

Table 3–5 Chromosomal Assignments of the Complement Proteins Protein(s)

Chromosomal Assignment

C1q C1r/C1s C2, C4A, C4B, factor B C1 inhibitor, CD59 C4-binding protein, factor H, CR1, CR2, MCP, DAF Factor D Factor I Properdin C3 C5 C6/C7 C8 C9

A, B and C chains on 1 Closely linked on 12 In MHC locus on 6 11 In RCA cluster on 1 19 4 X 19 9 Closely linked on 5 α and β chains on 1; γ chain on 9 5

DAF, delay accelerating factor; MCP, membrane cofactor protein; MHC, major histocompatibility complex; RCA, regulators of complement activation.

II 38

enzyme, carboxypeptidase R, has been identified as the primary inactivator of kinin as well as anaphylatoxin peptides.43 Unlike carboxypeptidase N, carboxypeptidase R is itself rapidly inactivated under normal purification conditions. After removal of the arginine, the peptide acquires the designation “des arg.” C3a and C4a are rapidly and completely inactivated by the two carboxypeptidases; C5a is somewhat more resistant to inactivation. In addition, neutrophils can still respond to C5a des arg, albeit with some 103 weaker affinity; the removal of arginine from C3a and C4a results in complete loss of biological activity. The relative resistance to carboxypeptidases and retention of bioactivity make C5a the most physiologically important of the anaphylactic complement peptides. Generation of C5a causes many of the effects seen in inflammation that are mediated by neutrophils. After activation by the binding of C5a into specific receptors, neutrophils bind to the capillary endothelium and migrate through the vessel wall, after the concentration gradient of the C5a. Once they are in contact with the higher concentrations of C5a present at the site of complement activation, neutrophils release granule contents and reactive metabolites, including lysozyme, reactive oxygen species, and eicosanoids. Although this is part of the normal mechanism of response to tissue injury or infection, the generation of large amounts of C5a or its presence in inappropriate locations can cause significant damage to uninvolved tissues.

Other Complement Activation Peptides Other peptides generated during the course of complement activation have been identified as inducing cell activation or chemotaxis. Factor Ba, the peptide cleaved from factor B by factor D, is a weak chemotactic factor. The cleavage products of factor B—Ba and Bb—have been reported to inhibit and stimulate B-cell proliferation, respectively. Peptides with functions similar to those of C3a and C5a can be generated by the action of noncomplement enzymes, particularly plasmin, on C3 and C5. These C3a-like and C5a-like peptides may play a role in the activation of both platelets and white cells under existing blood storage protocols. The use of negatively charged leukoreduction filters may significantly affect the levels of various contact activation peptides in blood products.44 In general, negatively charged artificial membrane surfaces promote complement activation; however,

some types of filters appear to remove C3a that is generated during processing and storage.45 The membrane attack complex itself causes significant activation in a wide variety of cell types. Because most studies of the C5b-9 complex have focused on its lytic effect on erythrocytes, the functional effect on nucleated cells and platelets has lately gained appreciation. Although there are some reports in the literature that C5b-7 has chemotactic activity for neutrophils, it is the fully formed C5b-9 that has the greatest effect on these cells. In comparison with erythrocytes, nucleated cells are very resistant to C5b-9 lysis. In response to C5b-9, nucleated cells and platelets coalesce the C5b-9 complexes and bud them off in vesicles of plasma membrane. In nucleated cells, especially neutrophils, this process is accompanied by the activation of the cell and the release of enzymes, leukotrienes, prostaglandins, thromboxanes, and reactive oxygen metabolites from the cell.46 Platelets interact with the activated proteins of the complement system at several levels.47 Specific platelet receptors for C1q have been identified.48–50 Although the precise role of C1q receptors remains to be determined, one such receptor has been reported to play a role in phagocytosis in other cell types.51 The exposure of human platelets to C3a alters the response to physiologic agonists but does not induce platelet aggregation. The effect of C5a on human platelets has not been investigated. Membrane attack complex can also affect platelet function.52 C5b-9 can induce the formation of platelet membrane vesicles (or microparticles), increase the procoagulant activity of the platelet, and cause some degree of arachidonic acid generation.53 Studies of the effects of C5b-9 on platelets have been carried out in purified protein systems; the significance of these observations for the whole plasma system remains to be determined. The platelet may modify the effects of activated complement fragments through the action of its surface regulatory proteins, DAF, MCP, C8-binding protein, and CD59. Platelets also contain internal pools of vitronectin, which may contribute to the local modulation of complement. The significant effects of activated complement fragments on the physiologic processes of platelets must be considered in the setting of platelet concentrate storage. Complement is activated under storage conditions,54 with or without prestorage leukoreduction. Activated complement fragments as well as C5b-9 may contribute to the platelet storage lesion.55

Cytotoxic Effects In the transfusion medicine setting, the effects of complement activation are graphically illustrated by the acute intravascular transfusion reaction. The generation of C3a and C5a produces bronchospasm and hypotension. C5a can also stimulate the production of interleukin-1 from macrophages, thereby causing fever.56 The recruitment of large numbers of neutrophils to the lung by C5a generation produces ventilation–perfusion abnormalities. This effect may have devastating consequences for the recipients of transfusion products who develop TRALI subsequent to the fixation of complement by anti-HLA antibodies. Two other hallmarks of the acute intravascular transfusion reaction, hemoglobinemia and hemoglobinuria, result from the activation of complement on the red cell surface in sufficient amounts to overwhelm the cellular and plasma control proteins, thereby lysing the cell. The generation of activated complement proteins affects more than red cell survival. The activation of cells with the concomitant release of enzymes contributes to the activation of the coagulation system, as does the action of complement proteins on coagulation substrates and endothelial cells. In vitro studies suggest that these processes may trigger the disseminated intravascular coagulation seen in severe cases of hemolytic transfusion reaction. Evidence for the direct cytolysis of platelets or granulocytes during transfusion is not abundant, perhaps because of the difficulty in constructing an adequate study design. However, complement activation may be directly linked to platelet destruction in the setting of paroxysmal nocturnal hemoglobinuria, sepsis, and thrombotic thrombocytopenic purpura/hemolytic uremia syndrome. Each of these conditions is associated with complement activation and platelet dysfunction. The complement-mediated destruction of antibody-coated platelets can be experimentally induced. Antibody to the human platelet antigen P1A1 (HPA-1) fixes sufficient complement to lyse target platelets.57 In vitro platelet lysis can also be induced by cold agglutinin anti-I antibodies58; antibodies of this specificity have been proposed to mediate the thrombocytopenia associated with Epstein-Barr virus infection.

Opsonic Effects The opsonic effects of complement activation result in the accelerated clearance of particles or cells that bear C3 as well as IgG. Although complement activation is the primary mechanism of cell destruction in the acute hemolytic transfusion reaction, the extravascular destruction of erythrocytes does not have the absolute requirement of complement activation. The primary clearance mechanism is through the phagocyte Fc receptor with recognition of the IgG present on the cell surface.59 Opsonization of cells by complement, rather than through lysis, means that the cell has been able to regulate complement effectively to prevent the assembly of cytolytic membrane attack complexes. This regulation may reflect the titer, avidity, affinity, or thermal amplitude of the antibody in its interaction with the cell target. In addition, as described previously, only IgG1 and IgG3 are efficient complement activators. The presence of C3b (or its degradation products) on the cell target in addition to antibody accelerates the clearance by the reticuloendothelial

system. If a macrophage is already activated, it will bind and ingest cells that bear only C3b. Normal, resting macrophages require that IgG also be present on the erythrocyte surface. This experimental result is supported by the in vivo observation that the erythrocytes of patients with cold agglutinin disease circulate through the spleen bearing C3b but no IgG and are not sequestered. That is not to say that C3 has no in vivo role in clearance. Bacteria coated with C3 only are efficiently phagocytized by macrophages, resting or activated, even in the absence of IgG. Macrophages in the liver (Kupffer cells) are capable of clearing erythrocytes coated with C3b only. In addition, persons who are genetically deficient in one of the components of the early classical pathway are unable to bind C3 to anti-D-coated erythrocytes; those target cells are cleared from the circulation much more slowly than in persons with intact complement systems.60 The opsonic and cytotoxic effects of complement are important in the ex vivo setting of blood storage. Many types of bacteria that have been implicated in transfusionmediated sepsis are either lysed or opsonized by the complement proteins in the blood unit. These bacteria are then engulfed by phagocytes also present in the bag. The removal of phagocytes by leukodepletion within 24 hours of collection results in the removal of contaminating bacteria as well. Leukodepletion more than 24 hours after collection may fail to sterilize the unit, because white cell breakdown may result in the release of viable bacteria from phagolysosomes.

PRINCIPLES OF THE COMPLEMENT SYSTEM CENTRAL TO TRANSFUSION MEDICINE

EFFECTS OF COMPLEMENT ACTIVATION ON CELL SURVIVAL

The Role of Complement Receptors in Cell Clearance and in Maintenance of the Immune Response The removal of complement-coated target cells is mediated by specific receptors for C3 and its activation and degradation fragments. As previously discussed, the activation of C3 produces two fragments, C3a, the anaphylatoxin, and C3b, which is covalently bound to the cell surface. In the presence of factor H, MCP, or the complement receptor CR1, C3b is rapidly cleaved by factor I to an inactivated form, iC3b (Fig. 3–3). This cleavage is estimated to occur in vivo within 5 minutes of the generation of C3b. Inactivated C3b undergoes another, slower interaction with factor I that results in a second cleavage on the opposite side of the thiol ester bond from the initial cleavage. This second cleavage occurs within 30 minutes of generation of C3b and results in the generation of the C3c fragment, which is no longer tethered to the cell, and the C3dg fragment, which contains the thiol ester bond and remains bound to the cell surface. Under laboratory conditions, the C3dg fragment can be further cleaved by trypsin to leave the C3d fragment on the cell surface; the frequency of generation of this fragment in vivo is uncertain. Several complement receptors have been identified to date. They have overlapping cell distribution (Table 3–6). The specificity of these receptors for individual fragments of C3 is reflected in the biological response to receptor occupation. The central function of the complement receptor CR1 is clearance of immune complexes from the circulation through its interaction with C3b contained in the complex.61 The bulk of the total CR1 in the circulation is present on red cells, because they provide the greatest mass of cells in the peripheral blood. The erythrocyte is therefore essential in transporting immune complexes from the plasma to the resident macrophages in the spleen.

3 39

BLOOD BANKING

II 40

Figure 3–3 Degradation fragments of C3. Native C3 is cleaved by C4b2a or C3bBb, resulting in the exposure of a reactive thiol ester in the α chain of C3 and the generation of two fragments, C3a and C3b. C3b is inactivated by factor I in the presence of factor H or CR1 to iC3b, which remains bound to the surface by the thiol ester. A further cleavage by factor I on the other side of the covalent bond from the first cleavage results in the release of the C3c fragment; the C3dg fragment remains surface bound. In vitro, the C3dg fragment may be further degraded to C3d, C3e, and C3g. The C3d fragment, a terminal breakdown product of C3, remains covalently bound to the surface. (From Anderson KC, Ness PM. Scientific Basis of Transfusion Medicine, 2nd ed. Philadelphia, Saunders, 2000.)

The further processing of C3b to iC3b enables the complement fragment to interact with the complement receptor CR3, also known as CD11b/CD18. This receptor, with its distribution on monocytes, macrophages, and neutrophils, plays a major role in immune-mediated phagocytosis. CR3 belongs to a family of related receptor molecules that have the β chain CD18. The pairing of the CD11a α chain to CD18 defines the leukocyte function-associated antigen 1, which is important in mediating killing by T lymphocytes. CD18 may also pair with the α chain CD11c, forming the complex known as p150,95. This molecule has been unofficially designated CR4 in recognition of its binding affinity for iC3b. The importance of the complement receptors CR3 and CR4 is indicated by the severity of the deficiency state, in which patients have defective phagocytic function and are susceptible to recurrent infections.62 The degradation of iC3b by factor I produces C3dg, the principal ligand for the receptor CR2. The expression of CR2 is restricted to B lymphocytes, on which the binding of C3dg triggers Bcell activation and proliferation. In addition to its role as a C3dg-binding protein, CR2 is the cellular receptor through which the Epstein-Barr virus gains entry to B lymphocytes. The identification and characterization of the complement receptors has led to the development of new therapeutic modalities. For example, a recombinant CR1 has been engineered with the transmembrane region missing. This molecule retains the complement-regulatory ability of CR1 but is soluble in plasma. In animal models, this protein, sCR1, has proved efficacious in reducing the complement activation seen during thrombolytic therapy.63 Fragments of C3 are important in maintaining the immune response.64 Animals with an experimental depletion of C3 fail to mount a normal IgG response to secondary immunization. This observation suggests that C3 is important in the development of immunologic memory, but such a role for C3 has not been confirmed in humans. It has also been established that the complement receptors CD21 and CD35 are important in the regulation of B-cell immunity while the complement regulatory proteins CD46 and CD55 have an additional role in T-cell function through the regulation of cytokine production.65,66

Therapeutic Complement Inhibition The great advances in our understanding of both complement system biochemistry and the role of complement in

Table 3–6 Structure and Distribution of Complement Receptors Receptor

Molecular Structure

Cellular Distribution

C1q

65 kDa, single chain

C3a C5a CR1 (CD35) CR2 (CD21) CR3 (CD11b/CD18)

Single chain protein predicted to have 7 transmembrane regions 45 kDa, single chain Four allotypes ranging from 160 to 250 kDa 145 kDa, single chain Two chains, 165 kDa and 95 kDa

CD11c/CD18 (CR4)*

Two chains, 150 kDa and 95 kDa

(CR5)*

Unknown

Platelets, monocytes, macrophages, B lymphocytes, endothelial cells Mast cells, monocytes, macrophages, neutrophils, basophils, T cells Monocytes, macrophages, neutrophils, mast cells Erythrocytes, monocytes, neutrophils, B lymphocytes B cells, follicular dendritic cells Neutrophils, monocytes, macrophages, follicular dendritic cells, natural killer cells Neutrophils, monocytes, macrophages, natural killer cells Neutrophils, platelets

*

CR designation is unofficial.

LABORATORY ANALYSIS OF COMPLEMENT Measurement of Cell-Associated Complement The methods available for the measurement of cell-associated immunoglobulins are readily adaptable to the detection of cell-bound complement. Traditionally, the presence of cell-associated C3b and its cleavage products is detected by means of an aggregating anti-C3d antibody (C3 Coombs’ test). The presence of small amounts of C4d antigen on the erythrocyte is identified in the blood bank as the ChidoRodgers blood group antigen system. The antigenic difference is actually determined by the C4A and C4B alleles, with Rodgers specificity found in the C4A isotypes and Chido specificity found in the C4B isotypes. Techniques developed to measure cell-associated immunoglobulin can all be adapted to measure cell-bound complement, especially C3. These methods include fluorescence flow cytometry, radioimmunoassay, and enzyme-linked immunosorbent assay (ELISA).70,71 In addition, the development of monoclonal antibodies that can distinguish among the various fragments of C3 may prove useful in laboratory diagnosis.

Measurement of Complement Activation or Deficiency States Until the mid-1980s, there were no readily accessible tests of complement activation. The activation of complement was implied by a decrease in the level of an individual component, most often C4, in a clinical setting in which activation was suspected. Although this approach may be of some utility in the outpatient setting, it is of little use in the investigation of very sick patients, especially those receiving blood products. The problem with this diagnostic approach is twofold. First, reduction in the C4 level can occur from decreased synthesis of the protein in the liver rather than from consumption. Second, because of the presence of null genes for C4, the normal range for C4 is very wide; a person with a C4 level that usually is at the high end of the normal range may activate more than 50% of C4 before the measured level exceeds the normal range. Only laboratories with specialized complement expertise had methods permitting the detection of activation peptides of complement, generally by gel electrophoresis techniques or by radioimmunoassay of small complement fragments remaining in solution after precipitation of native protein and large fragments. With the advent of monoclonal antibody technology, antibodies have been made that are capable of recognizing neoantigens expressed in the activation peptides of complement. These antibodies have been used in the formulation of commercial ELISA kits. ELISA tests that detect the activa-

tion peptides of either the classical (C4d) or alternative (factor Bb) pathway or both (C3a, C5a, iC3b, and C5b-9) are currently available. In addition, descriptions of many other monoclonal antibody-based activation peptide assays can now be found in the literature. These activation-dependent assays enable definite differentiation between patients with complement activation and those with decreased production of complement proteins. Until recently, the assessment of patients with suspected complement deficiencies has relied on somewhat cumbersome gel methods. However, a new ELISA-based procedure has been developed for detection of complement deficiencies, including those due to a loss of proteins in the lectin pathway.72 The continuing improvement of testing methods should make these assays more accessible to nonspecialists. In summary, the burgeoning of research activity on the complement system has produced a clearer understanding of the biochemistry of these important proteins. This knowledge has led to the development of better diagnostic tools, to an increased clarity around the role of complement in disease pathophysiology, as well as to the invention of therapeutic modalities to control unwanted complement activation. REFERENCES 1. Ross GD. Immunobiology of the Complement System. Orlando, Fla., Academic Press, 1986, pp 1–19. 2. Morgan BP. Complement: Clinical Aspects and Relevance to Disease. New York, Academic Press, 1990. 3. Sjoholm AG, Jonsson G, Braconier JH, et al. Complement deficiency and diseases: an update. Mol Immunol 2006;43:78–85. 4. Thiel S, Frederiksen PD, Jensenius JC. Clinical manifestations of mannan-binding lectin deficiency. Mol Immunol 2006;43:86–96. 5. Sim RB, Reid KBM. C1: Molecular interactions with activating systems. Immunol Today 1991;12:307–311. 6. Perkins SJ, Nealis AS. Solution structure of human and mouse immunoglobulin M by synchrotron X-ray scattering and molecular graphics modelling. J Mol Biol 1991;221:1345–1366. 7. Levine RP, Dodds AW. The thiolester bond of C3. Curr Top Microbiol Immunol 1989;153:73–82. 8. Szalai AJ, Briles DE, Volanakis JE. Role of complement in C-reactive protein-mediated protection of mice from Streptococcus pneumoniae. Infect Immunol 1996;64:4850–4853. 9. Bristow CL, Boackle RJ. Evidence for the binding of human serum amyloid P component to C1q and Fab. Mol Immunol 1986;23:1045–1052. 10. Pepys MB, Blatz ML. Acute phase proteins with special reference to Creactive protein and related proteins (pentraxins) and serum amyloid A protein. Adv Immunol 1983;34:141–212. 11. Lu JH, Thiel S, Wiedemann H, et al. Binding of the pentamer/hexamer forms of mannan-binding protein to zymosan activates the proenzyme C1rC1s2 complex of the classical pathway of complement, without involvement of C1q. J Immunol 1990;144:2287–2294. 12. Fujita T, Matushita M, Endo Y. The lectin-complement pathway—its role in innate immunity and evolution. Immunol Rev 2004;198:195–202. 13. Lachmann PJ, Hughes-Jones NC. Initiation of complement activation. Springer Semin Immunopathol 1984;7:143–162. 14. Devine DV. The regulation of complement on cell surfaces. Trans Med Rev 1991;5:123–131. 15. Donaldson VH, Bissler JJ. C1 inhibitors and their genes: an update. J Clin Lab Med 1992;119:330–222. 16. Hillarp A, Dahlback B. The protein S-binding site localized to the central core of C4b-binding protein. J Biol Chem 1987;262:11300–11307. 17. Hammer CH, Jacobs RM, Frank MM. Isolation and characterization of a novel plasma protein which binds to activated C4 of the classical complement pathway. J Biol Chem 1989;264:2283–2291. 18. Farries TC, Lachmann PJ, Harrison RA. Analysis of the interaction between properdin and factor B, components of the alternative pathway C3 convertase of complement. Biochem J 1988;253:667–675. 19. Murphy BF, Kirszbaum L, Walker ID, et al. SP-40,40, a newly identified normal human serum protein found in the SC5b-9 complex of complement and in the immune deposits in glomerulonephritis. J Clin Invest 1988;81:1858–1864.

PRINCIPLES OF THE COMPLEMENT SYSTEM CENTRAL TO TRANSFUSION MEDICINE

pathophysiology have led to the creation of strategies to control complement activation in order to minimize its deleterious effects.67 Several promising compounds are in preclinical development; one monoclonal antibody against C5 has been used successfully in a clinical trial in the treatment of PNH to reduce ongoing hemolysis.68 Although complement inhibitors have been proposed for use in acute hemolytic transfusion reactions, no clinical studies have yet been performed.69

3 41

BLOOD BANKING

II 42

20. Choi NH, Mazda T, Tomita M. A serum protein Sp-40,40 modulates the formation of membrane attack complex of complement on erythrocytes. Mol Immunol 1989;26:835–840. 21. Jenne DE, Lowin B, Peitsch MC, et al. Clusterin (complement lysis inhibitor) forms a high density lipoprotein complex with apolipoprotein A-I in human plasma. J Biol Chem 1991;266:11030–11036. 22. Nicholson-Weller A, Burge J, Fearon DT, et al. Isolation of a human erythrocyte membrane glycoprotein with decay accelerating activity for C3 convertases of the complement system. J Immunol 1982;129:184–189. 23. Medof ME, Kinoshita T, Nussenzweig V. Inhibition of complement activation on the surface of cells after incorporation of decay-accelerating factor (DAF) into their membranes. J Exp Med 1984;160:1558–1578. 24. Pangburn MK. Differences between the binding sites of the complement regulatory proteins DAF, CR1, and factor H on C3 convertases. J Immunol 1986;136:2216–2221. 25. Kinoshita T, Medof ME, Nussenzweig V. Endogenous association of decay-accelerating factor (DAF) with C4b and C3b on cell membranes. J Immunol 1986;136:3390–3395. 26. Berger M, Medof ME. Increased expression of complement decay accelerating factor during activation of human neutrophils. J Clin Invest 1987;79:214–220. 27. Seya T, Atkinson JP. Functional properties of membrane cofactor protein of complement. Biochem J 1989;264:581–588. 28. Schoermark S, Rauterberg, EW, Shin ML, et al. Homologous species restriction in lysis of human erythrocytes: a membrane-derived protein with C8-binding capacity functions as an inhibitor. J Immunol 1986;136:1772–1776. 29. Zalman LS, Wood LM, Muller-Eberhard JH. Isolation of a human erythrocyte membrane protein capable of inhibiting expression of homologous complement transmembrane channels. Proc Nat Acad Sci USA 1986;83:6975–6979. 30. Holguin MH, Frederick LR, Bernshaw NJ, et al. Isolation and characterization of a membrane protein from normal human erythrocytes that inhibits reactive lysis of the erythrocytes of paroxysmal nocturnal hemoglobinuria. J Clin Invest 1989;84:7–17. 31. Sugita Y, Nakamo Y, Tomita M. Isolation from human erythrocytes of a new membrane protein which inhibits the formation of complement transmembrane channels. J Biochem (Tokyo) 1988;104:633–637. 32. Okada N, Harada R, Fujita T, et al. A novel membrane glycoprotein capable of inhibiting membrane attach by homologous complement. Int Immunol 1989;1:205–208. 33. Stefanova I, Hilgert I, Kirstofova H, et al. Characterization of a broadly expressed human leukocyte antigen MEM-43 anchored in membrane through phosphatidylinositol. Mol Immunol 1989;26:153–161. 34. Lockert DH, Kaufman KM, Chang CP, et al. Identity of the segment of human complement C8 recognized by complement regulatory protein CD59. J Biol Chem 1995;270:19723–19728. 35. Husler T, Lockert DH, Kaufman KM, et al. Chimeras of human complement C9 reveal the site recognized by complement regulatory protein CD59. J Biol Chem 1995;270:3483–3486. 36. Parker CJ, Omine M, Richards S, et al. Diagnosis and management of paroxysmal nocturnal hemoglobinuria. Blood 2005;106:3699–3709. 37. Telen MJ, Green AM. The Inab phenotype: characterization of the membrane protein and complement regulatory defect. Blood 1989;74: 437–441. 38. Merry AH, Rawlinson VI, Uchikawa M, et al. Studies on the sensitivity to complement-mediated lysis of erythrocytes (Inab phenotype) with a deficiency of DAF (decay accelerating factor). Br J Haematol 1989;73:248–253. 39. Yamashina M, Ueda E, Kinoshita T, et al. Inherited complete deficiency of 20-kilodalton homologous restriction factor (CD59) as a cause of paroxysmal nocturnal hemoglobinuria. N Engl J Med 1990;323:1184–1189. 40. Farries TC, Atkinson JP. Evolution of the complement system. Immunol Today 1991;12:295–300. 41. Marcus D, Alper CA. Methods for allotyping complement proteins. In Rose N, Manual of Clinical Immunology. Washington, D.C., American Society for Microbiology, 1986, pp 185–196. 42. Winkelstein JA, Colten HR. Genetically determined disorders of the complement system. In The Metabolic Basis of Disease. St. Louis, McGraw-Hill, 1987, pp 2711–2737. 43. Campbell W, Okada H. An arginine specific carboxypeptidase generated in blood during coagulation or inflammation which is unrelated to carboxypeptidase N or its subunits. Biochem Biophys Res Comm 1989;162:933–939. 44. Shiba M, Tadokoro K, Sawanobori M, et al. Activation of the contact system by filtration of platelet concentrates with a negatively charged white

45. 46. 47. 48. 49.

50. 51. 52. 53. 54. 55. 56. 57. 58.

59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72.

cell-removal filter and measurement of venous blood bradykinin level in patients who received filtered platelets. Transfusion 1997;37:457–462. Snyder EL, Mechanic S, Baril L, et al. Removal of soluble biologic response modifiers (complement and chemokines) by a bedside white cell-reduction filter. Transfusion 1996;36:707–713. Morgan BP. Complement membrane attack on nucleated cells: Resistance, recovery and non-lethal effects. Biochem J 1989;264:1–14. Devine DV. The effects of complement activation on platelets. Curr Top Microbiol Immunol 1992;178:101–113. Peerschke EIB, Ghebrehiwet B. Human blood platelets possess specific binding sites for C1q. J Immunol 1987;138:1537–1541. Nepomuceno RR, Henschen-Edman AH, Burgess WH, et al. cDNA cloning and primary structure analysis of C1qR(P), the human C1q/MBL/ SPA receptor that mediates enhanced phagocytosis in vitro. Immunity 1997;6:119–129. Nepomuceno RR, Tenner AJ. C1qRP, the C1q receptor that enhances phagocytosis, is detected specifically in human cells of myeloid lineage, endothelial cells, and platelets. J Immunol 1998;160:1929–1935. Butko P, Nicholson-Weller A, Wessels MR. Role of complement and complement receptor C1qR in the antibody-independent killing of group B streptococcus. Adv Exp Med Biol 1997;418:941–943. Sims PJ, Wiedmer T. The response of human platelets to activated components of the complement system. Immunol Today 1991;12:338–342. Wiedmer T, Esmon CT, Sims PJ. Complement proteins C5b-9 stimulate procoagulant activity through platelet prothrombinase. Blood 1986; 68:875–880. Bode AP, Miller DT, Newman SL, et al. Plasmin activity and complement activation during storage of citrated platelet concentrates. J Lab Clin Med 1989;113:94–102. Gyongyossy-Issa MIC, McLeod E, Devine DV. Complement activation in platelet concentrates is surface-dependent and modulated by the platelets. J Lab Clin Med 1994;123:859–868. Dalmasso AP. Complement in the pathophysiology and diagnosis of human disease. CRC Crit Rev Clin Lab Sci 1986;24:123–283. Cines DB, Schreiber RD. Effect of anti-P1A1 antibody on human platelets: I. The role of complement. Blood 1979;53:567–577. Dixon RH, Rosse WF. Mechanisms of complement-mediated activation of human blood platelets in vitro: comparison of normal and paroxysmal nocturnal hemoglobinuria platelets. J Clin Invest 1977;59: 360–368. Lutz HU. Innate immune and non-immune mediators of erythrocyte clearance. Cell Mol Biol 2004;50:107–116. Schreiber AD. An experimental model of immune hemolytic anemia. Ann Intern Med 1977;87:211–217. Schifferli JA, Ng YC, Peters DK. The role of complement and its receptor in the elimination of immune complexes. N Engl J Med 1986;315:488–495. Anderson DC, Springer TA. Leukocyte adhesion deficiency: an inherited defect in the MAC-1, LFA-1, and p150,95 glycoproteins. Annu Rev Med 1987;38:175–194. Weisman HF, Bartow T, Leppo MK, et al. Soluble complement receptor type 1: in vivo inhibitor of complement suppressing post-ischemic myocardial inflammation and necrosis. Science 1990;249:146–151. Erdei A, Fust G, Gergely J. The role of C3 in the immune response. Immunol Today 1991;12:332–337. Liu J, Miwa T, Hilliard B, et al. The complement inhibitory protein DAF (CD55) suppresses T cell immunity in vivo. J Exp Med 2005;201: 567–577. Wagner C, Hansch GM. Receptors for complement C3 on T-lymphocytes: relics of evolution or functional molecules? Mol Immunol 2006;43:22–30. Mollnes TE, Kirschfink M. Strategies of therapeutic complement inhibition. Mol Immunol 2006;43:107–121. Hill A, Hillmen P, Richards SJ, et al. Sustained response and long-term safety of eculizumab in paroxysmal nocturnal hemoglobinuria. Blood 2005;106:2559–2265. Yazdanbakhsh K, Kang S, Tamasauskas D, et al. Complement receptor 1 inhibitors for prevention of immune-mediated red cell destruction: Ppotential use in transfusion therapy. Blood 2003;15:5046–5052. Garratty G. The significance of IgG on the red cell surface. Trans Med Rev 1987;1:47–57. Schwartz KA. Platelet antibody: review of detection methods. Am J Hematol 1988;29:106–14. Seelen MA, Roos A, Wieslander J, et al. Functional analysis of the classical, alternative and MBL pathways of the complement system: standardization and validation of a simple ELISA. J Immunol Meth 2005;296:187–198.

Chapter 4

Principles of Red Blood Cell Allo- and Autoantibody Formation and Function James C. Zimring

INTRODUCTION Humoral immunity represents the main barrier to transfusion of red blood cells (RBCs) in humans. Crossmatching units of RBCs for the naturally occurring antibodies against the ABO carbohydrate antigens is an absolute requirement to avoid acute hemolysis. However, the majority of immunohematology focuses on the identification of acquired antibodies to RBC antigens, which are typically generated after previous exposure to foreign RBCs. Several hundred different RBC antigens have now been described (see Chapters 5, 6, 7, and 8). These antigens can consist of carbohydrates, linear peptides, and tertiary confirmations dependent on proper folding and membrane insertion. Different blood group antigens have distinct immunogenicities, because RBC antigens vary in their likelihood of inducing an antibody response. In addition, antibodies to some blood group antigens frequently cause hemolysis (clinically significant), whereas others cause no deleterious effect (clinically insignificant). The current understanding of the basic science regarding immunogenicity of transfused RBCs and the immune-mediated mechanisms of RBC destruction are detailed in this chapter.

BASIC SCIENCE OF ALLOANTIBODY FORMATION Most anti-RBC antibodies detected in the practice of transfusion medicine are humoral responses to alloantigens encountered during previous exposures to foreign RBCs, typically via transfusion or pregnancy. A great deal is now understood about the general mechanics of humoral immunization to soluble and particulate antigens (see Chapter 2). However, although sheep red blood cells have been used as a model antigen for many years, the current paradigms of immunology have been largely developed outside the context of immunization by RBC transfusion. Transfused RBCs appear to be only weakly immunogenic, and it is currently unclear if immune responses to RBC transfusion are qualitatively different from general mechanisms of immune response to foreign antigen. However, it is safe to say that there are several notable differences between traditional microbial or protein immunogens and transfusion of sterile blood, including route of immunization, potential absence of inflammation and danger signals, dose and kinetics of antigenic exposure, type of antigen-presenting cell involved, and the molecular nature of the antigen. The exact role and importance of these

variables in RBC transfusion is currently unknown, but the theoretical potential for their importance is discussed in the following sections.

Frequency of Alloimmunization to RBC Transfusion: Immunity and Tolerance Despite there being numerous foreign epitopes on essentially all transfusions of nonautologous RBCs, transfusion is not a highly immunogenic stimulus. Even in response to multiple transfusions, alloimmunization to alloantigens on transfused RBCs has an overall frequency of approximately 2% to 6%.1–3 However, it has been hypothesized that frequency of alloimmunization may vary with the underlying pathophysiology of the transfused patient. One study revealed alloimmunization rates as follows, based on underlying disease: lymphocytic leukemia 0%, gastrointestinal bleed 11%, aplastic anemia 11%, renal failure 14%, myelogenous leukemia 16%, and hemoglobinopathy 29%.4 These findings indicated a trend, with a clear decrease in lymphocytic leukemia, presumably due to immunosuppression. There is also an apparent increase in other disease states, but these differences were not found to be statistically significant. It has been proposed that alloimmunization is unusually high in some disease states, such as sickle cell anemia, myelodysplastic anemia, and autoimmune hemolytic anemia. However, rates of alloimmunization can vary widely from study to study. For example, the range of alloimmunization in adult sickle cell patients ranges from 18.6%5 to 47%,6 depending on the study, with an average rate of 25%.7 The presence of alloantibodies in patients with autoimmune hemolytic anemia, when excluding autoantibodies that mimicked alloantibodies, ranges from 12%8 to 40%,9 with an average frequency of 32%.10 Likewise, two different groups reported alloimmunization rates in patients with myelodysplastic syndrome as 21% and 58.6%, respectively.11,12 Analysis of the above literature demonstrates considerable variation in the rates of alloimmunization reported by different groups. A number of factors may lead to varying results, including differences in patient demographics, differences in diagnostic criteria or subclasses of disease, and differences in methodology of alloantibody detection. In addition, it has been reported that up to 40% of alloantibodies subsequently become undetectable,13,14 raising the possibility that the time after transfusion that specimens are collected may influence outcome. It has also been reported that leukoreduction may decrease immunogenicity of transfused RBCs.15 Since use of leukoreduced blood varied in the above studies, this may be

4 43

BLOOD BANKING

II 44

a relevant issue. A further complicating factor is that most studies are retrospective and do not compare pretransfusion specimens to post-transfusion specimens and it is unclear if the presence of an alloantibody is necessarily the result of an antecedent transfusion. The precise mechanism by which different patient populations may differ in their alloimmunization rates is unclear. However, a number of potential factors may be involved. First, the volume and frequency of transfusion are important variables, because they reflect the magnitude and kinetics of antigen dose. Second, the extent to which donor units differ phenotypically from recipient RBCs will always be an important factor. Thus, one must consider the genetic similarity of the donor and recipient populations. A third variable is the extent to which the underlying pathophysiology of the condition necessitating transfusion may perturb immune function. For example, systemic immunosuppression linked to the pathophysiology likely contributes to a low rate of seroconversion in patients with lymphocytic leukemia.4 In contrast, the pathophysiology of other disease states may lead to immune dysregulation resulting in increased alloimmunization. Given that only a small percentage of individuals make detectable alloantibodies, despite multiple transfusion, it appears as though RBCs may represent a fairly weak immunogen. Of course, this notion is biased by the fact that RBCs are typically matched a priori for the most immunogenic antigens (i.e., Rh0D). Immunization rates to Rh0D have been estimated at 30% to 90%,16–23 depending on the study. Still, even for the Rh0D antigen, up to 20% of Rh-negative individuals do not mount an anti-Rh response, despite repeated exposure to Rh-positive blood. The nonresponse of such individuals may in part be due to Rh actually being self in the case of recipients who are weak D and do not type D+ by routine methods. In addition, nonresponders may have genetic factors such that they lack the ability to respond to the Rh antigen. However, this pattern is also consistent with initial exposure to Rh-positive RBCs inducing a state of active tolerance against Rh antigens, although such has not been demonstrated. Thus, overall, transfused RBCs are strongly immunogenic in some recipients, only weakly immunogenic or nonimmunogenic in other patients, and the theoretical potential for tolerance exists.

VARIABLES ASSOCIATED WITH RBCs AS AN IMMUNOGEN The context in which the immune system encounters antigens on transfused RBCs is associated with a number of characteristics that may contribute to the relatively weak immunogenicity of RBCs in some settings.

Route of Immunization Unlike subcutaneous and intramuscular routes, which are generally immunogenic, intravenous administration is among the most tolerogenic methods of antigen administration.24–26 Because transfusion of a sterile unit of RBCs involves introduction of intravenous antigen, one might predict that no immune response should occur to RBC antigens. However, the induction of tolerance to intravenous antigen has typically been described with soluble proteins, and aggregated protein complexes are typically immunogenic. Although blood group antigens are polymeric as a result of

being in a membrane, the molecules that carry blood group antigens are neither free-floating soluble proteins nor are they insoluble protein aggregates. Thus, it is unclear how this molecular form fits into the paradigm of soluble versus aggregate. It is worth noting that although intraperitoneal injection of soluble protein is not immunogenic, injection of the same protein chemically coupled to autologous RBCs results in an antibody response equivalent to immunization with the antigen in Freund’s complete adjuvant, which is among the strongest known methods for immunization.27,28 The same observation has recently been made regarding both humoral and cellular immunity using the intravenous route.29,30 Thus, the intravenous route through which blood is given may influence the immunogenicity of antigens on RBCs, but the particulars of this are unclear given the form in which the immune system encounters RBC antigens.

Absence of a Danger Signal It has been argued that without an inflammatory stimulus (also called a danger signal), immunity will not typically occur.31,32 Danger signals are not present on healthy self-tissues and typically need to be introduced by microbial infections or inflammation.33–37 In support of this concept, even a subcutaneous injection into the footpad of mice, which is typically the best route to induce humoral immunity, can induce tolerance when performed in the absence of an inflammatory adjuvant.24 Properly processed units of RBCs should not contain any microbial products. However, it is possible that processing and transfusion of blood introduces danger signals from nonmicrobial sources. It is also possible that the danger signal is provided by ongoing inflammation in the transfusion recipient. Thus, the role of inflammation and/or danger signals during RBC transfusion is unclear. However, whereas danger signals would be typically present during infection and immunization, they may be altered or absent during transfusion, which could influence the nature of anti-RBC immune responses.

Dose of Antigen Another important difference between an RBC transfusion and other immune stimuli is the amount of antigen present. The dose of foreign antigen received during the early stages of an infection can be quite low, especially when the infection begins as a small inoculum. As the infection proceeds, levels of antigen will rise rapidly and then decline within several weeks in the case of adequate clearance. Alternatively, in a chronic infection, the antigen will often still decline, but will persist at some level long term. In contrast, transfusion of blood results in a sudden burst of high-level antigen that persists for a moderate period of time (several months). In the case of fresh blood, transfused RBCs have a life span of approximately 120 days. Starting with a transfusion of a single 200-mL unit of packed RBCs, approximately 1.67 mL of packed RBCs are consumed per day, which translates to consumption of about 12.8 million RBCs per minute. Because these cells are removed by macrophages of the reticuloendothelial system, RBC transfusion essentially represents a selective targeting of large doses of foreign antigen to professional antigen-presenting cells (APCs). Transfusion of a stored unit of blood will result in a slightly different pattern of consumption, because RBCs continue to senesce during storage of blood products.38 Transfusion of a 200-mL bag of blood that has been stored for 4 weeks may result in more

Antigen-Presenting Cells Involved in Processing and Presenting RBC Antigens In order for a primary immune response to occur, a professional APC (typically a macrophage or dendritic cell) must process and present the antigen. Although macrophages are not required for immune responses to antigens in tissues,43–45 evidence suggests that they are necessary for generating immune responses to transfused RBCs.43–45 Thus, the type of APC required for a primary immune response to transfused RBCs may differ from the APC required for responses to other better-studied antigens. Evidence also suggests that the lymphatic compartment, and thus the immunologic microenvironment, varies for antigens injected into tissues and transfused RBCs. Tissueinjected antigens are consumed by resident dendritic cells, which then migrate to draining lymph nodes and accumulate in areas of naive T-cell activation, whereas macrophages are excluded from these areas.44 Likewise, lessons can be learned from solid tissue transplantation, where immune responses can be generated via two mechanisms. The direct pathway involves donor dendritic cells migrating to recipient lymph nodes and priming recipient T cells. In the indirect pathway, however, recipient dendritic cells enter the transplanted tissues, consume donor antigens, and return to recipient lymph nodes to prime T cells with donor antigens on recipient MHC. In both cases, the predominant APC is a dendritic cell in the specialized microenvironment of a lymph node. In contrast, the majority of transfused RBCs are consumed by macrophages in the spleen and liver, and there is no discrete tissue parenchyma where recipient dendritic cells can enter.

Polymeric Nature of Antigens on RBCs Erythrocyte surface antigens are physically linked via their mutual association with the cellular membrane and are thus polymeric in nature. As a group, polymeric antigens tend to be T-cell independent antigens that directly activate B cells46 to secrete immunoglobulin M (IgM) without the requirement of T-cell help. However, responses to T-cell independent antigens typically fail to class switch to high-affinity IgG. Thus, it is possible that transfused RBCs activate T-cell

independent pathways of humoral immunization. In support of this notion, it has been shown that initial anti-RBC IgM responses can be T-cell independent.46 However, in animal models, subsequent class switching to anti-RBC IgG depends on T-cell help.46 In addition, human patients with defective helper T-cell function secondary to infection with human immunodeficiency virus have significantly decreased rates of alloimmunization to transfused RBCs.47 Thus, as an immunogen, RBCs appear to have a unique combination of early T-cell independence for IgM synthesis followed by T-cell dependence for class switching and memory.

BASIC SCIENCE OF ANTI-RBC AUTOANTIBODY FORMATION Loss of immunologic tolerance to self-tissues can result in autoimmune processes targeted against specific organs or broadly reactive with self-tissues. Among the known autoimmune pathologies is the generation of autoantibodies against self-RBC antigens. This process can result in a clinically silent autoagglutinin that is detected incidentally by a positive direct antiglobulin test (DAT) during alloantibody screening. However, if the autoantibodies promote erythrocyte destruction, the clinical manifestation of autoimmune hemolytic anemia (AIHA) can result. AIHA can range in severity from a transient mild clinical course to massive lethal hemolysis. Although the exact etiology of AIHA is unclear, it can occur either spontaneously or in close temporal association to transfusion of RBCs. Because the basic processes that lead to AIHA in these two settings are likely different, they are considered separately in the following discussion.

AIHA Not Associated with a Transfusion (Spontaneous AIHA) Although the loss of tolerance to self-tissues is a general feature of all AIHA, it is unlikely that a single mechanism accounts for the process by which anti-RBC autoantibodies are generated. In the general field of autoimmune pathophysiology, a number of mechanisms may explain how loss of self-tolerance occurs. Because there is evidence for several of these mechanisms playing a potential role in the generation of AIHA in different settings, each will be considered here. Central and Peripheral Tolerance to Blood Group Antigens The fine-tuning of the immune system such that it can respond to a wide variety of microbial antigens without recognizing self-tissues is essential to both maintaining immune competence and avoiding autoimmunity. The process begins when T-cell receptors and B-cell receptors recombine randomly and generate a highly diverse repertoire of specificities. Then, in a process referred to as central tolerance, immature autoreactive T cells are deleted in the thymus, and autoreactive B cells either become anergic or are deleted, first in the bone marrow and then in the periphery. Such deletion events depend on the exposure of these lymphocytes to self-antigens during development. Although a great variety of antigens are expressed in the thymus, and the majority of autoreactive T cells are deleted, some autoreactive T cells escape this thymic education. If RBC-specific T cells and B cells survive deletion, AIHA could develop. This theory has been formally demonstrated

PRINCIPLES OF RED BLOOD CELL ALLO- AND AUTOANTIBODY FORMATION AND FUNCTION

rapid removal of packed RBCs due to senescence of stored blood.38 Thus, in contrast to the patterns seen with infectious pathogens, transfusion of RBCs results in a sudden introduction of large amounts of antigen (or very large amounts in the case of stored blood), followed by persistence of the antigen for several months. In the case of chronically transfused patients, exposure to common blood group antigens may be extended to a considerably longer period of time, depending on the frequency of transfusion. Dose alterations can profoundly affect immune responses. It has been known for decades that moderate doses of antigen generally promote immune responses, whereas both low levels and very high levels of antigen lead to tolerance.39 Duration of antigen persistence also has a significant effect on the nature of the immune response. Although CD4+ T cells expand and differentiate into helper T (Th) cells on initial exposure to antigen, extended exposure to antigens during chronic infection results in the inactivation and downregulation of viral antigen-specific CD4+ T cells.40,41 Moreover, the level of antigen present during chronic exposure can regulate immunity versus tolerance at the level of CD4+ Th cells.42

4 45

BLOOD BANKING

II 46

in an animal model. In the NZB mouse, which spontaneously develops AIHA, CD4+ helper T cells specific for peptides from RBC autoantigens survive thymic education, whereas in other strains that do not develop AIHA, the cells are deleted.48 Moreover, it has recently been shown that autoreactive T cells specific for peptides from Rh antigens are detectable in human patients with measurable anti-Rh autoantibodies.49 It thus seems likely that escape of some autoreactive CD4+ T cells may be involved in generation of AIHA. However, it has also been reported that γ/δ T cells, which are not specific for RBC antigens, can provide help to anti-RBC autoreactive B cells.50 Thus, it is also possible that T cells of other specificities can give bystander help to antiRBC B cells. In addition, T cells specific for cryptic epitopes may also be involved. Thus, it appears that although T cells are likely involved in helping autoreactive B cells differentiate into plasma cells, the exact nature of such help may vary. The activation of autoreactive B cells specific for RBC antigens is an absolute requirement for the generation of AIHA. Like thymic education of T cells, education of autoreactive B cells requires that immature B cells encounter the self-antigen, which results in either deletion or the induction of anergy.51 Although deletion of autoreactive B cells can occur in the periphery,52 B cells that encounter the antigen, recognized by their rearranged immunoglobulin gene, while still in the bone marrow undergo arrested development and death.53 In addition, B cells are more easily tolerized to membrane-bound surface antigens than to soluble antigens.52 Because erythroid development occurs in the bone marrow where immature B cells appear to first undergo education, and because blood group antigens are membrane-bound proteins, it seems likely that deletion of autoreactive B cells that recognize RBC antigens is a highly efficient process. Nonetheless, it has been demonstrated that while autoreactive B cells specific for RBC antigens are mostly deleted or rendered anergic, there is sufficient survival of autoreactive B cells to cause AIHA in some settings.54,55 The autoreactive B cells that survive in this setting appear to reside predominantly in gut tissues, and their activation can be significantly influenced by the presence of enteric pathogens or inflammation.55 If autoreactive cells manage to escape central tolerance, they can still be deleted or inhibited via peripheral tolerance, a second process that prevents autoreactive cells from differentiating into mature effector cells that may cause autoimmunity. The distinction between central and peripheral tolerance, which is clear for T cells (thymic versus extrathymic), is less well defined for B cells. Mechanisms of peripheral tolerance include induction of apoptosis, anergy, and active inhibition by suppressor/regulatory T cells. This combination of central tolerance, which eliminates most autoreactive cells, and peripheral tolerance, which inhibits those few autoreactive cells that escape, is generally effective; AIHA is a relatively rare event. However, the presence of autoreactive lymphocytes essentially constitutes a pre-existing risk factor for AIHA if the normal factors that keep autoreactive lymphocytes at bay are sufficiently perturbed. Several events can lead to the inappropriate activation of peripheral autoreactive cells. Infection with a pathogen that carries epitopes similar to blood group antigens (molecular mimicry) can provide sufficient stimulation of autoreactive cells in an inflammatory environment such that peripheral tolerance is overcome. Alternatively, general immune dysregulation and polyclonal activation of lymphocytes may non-

specifically activate T and/or B cells that happen to be specific for RBC antigens. In support of this concept, development of AIHA is associated with chronic inflammation and lymphoproliferative disorders in both human and animal models56,57 and is also associated with a variety of infections, including Mycoplasma pneumoniae, Epstein-Barr virus, cytomegalovirus, and rubella.58 Cytokine dysregulation may also play a role, as a polarization to a Th2-type profile in AIHA has been observed in both humans and in the NZB/W murine model of AIHA.59,60 Additional mechanisms include a loss of regulatory T cells, as CD4+ CD25+ T reg levels appear to be decreased in AIHA.61 Thus, the breakdown of peripheral tolerance by a variety of mechanisms may allow the few autoreactive lymphocytes that escape central tolerance to differentiate into mature effectors that ultimately result in the formation of plasma cells that secrete antibodies specific for blood group antigens. An additional mechanism of loss of tolerance involves the generation of cryptic peptide epitopes. For any given protein, only a limited number of peptides are normally presented by MHC molecules. The repertoire of peptides presented by each individual is largely a function of their HLA type, because peptides differ in their affinities for the grooves of different MHC variants. The range of peptides presented is also a function of how a protein is processed and broken down by a system of protease complexes called the proteosome. It has been demonstrated that a given protein may contain peptide sequences that fit well into an MHC molecule but are not typically presented on MHC because that particular peptide fragment is not generated by proteolysis of the protein in question. Such a peptide is called a cryptic epitope. Because cryptic epitopes will not be processed and presented by normal thymic tissues, T cells that recognize cryptic epitopes will not be deleted in the thymus. Such T cells are technically not autoreactive, because the cryptic epitopes are also not presented on peripheral tissues and thus do not constitute a self-antigen. However, abnormal processing of self-proteins can result in presentation of cryptic epitopes on self-tissues. For example, ongoing inflammation and microbial infection leads to the release of microbial proteases and a large-scale breakdown of self-proteins by inflammatory proteases released from leukocytes. Infection of APCs by microbes may also change the processing of self-antigens. Furthermore, cryptic epitopes may be generated by exposure to caustic stimuli that denature or destroy proteins, such as burns or chemicals. The generation of such peptides may lead to activation of T cells specific for the cryptic epitope, which would then be capable of providing T-cell help to a B cell presenting the same cryptic epitope. Once a B cell has differentiated into an antibody-secreting plasma cell, it no longer requires immune stimulation to produce antibodies. Thus, in theory, a sustained autoantibody response can be generated during a transient exposure to cryptic epitopes. Direct experimental evidence demonstrates that cryptic epitopes of Rh antigens are recognized by T cells in patients with AIHA of anti-Rh specificity.62–65

Transfusion-Associated AIHA A considerable percentage of AIHA patients develop an antiRBC autoantibody subsequent to transfusion with allogeneic RBCs. Several studies have examined the association between transfusion and the development of anti-RBC autoantibodies. The phenomenon has been observed with considerable

Mechanisms of Autoimmunization by Transfusion The mechanisms by which allogeneic RBC transfusion induces an anti-RBC autoantibody are undetermined. However, as with the development of spontaneous autoantibodies, it is likely that several mechanisms are involved. First, any of the aforementioned immunologic perturbations involving loss of tolerance, which may contribute to the spontaneous generation of AIHA, may also contribute to transfusion-associated AIHA. However, in addition to the general breakdown of immune tolerance, several mechanisms unique to transfusion-induced AIHA require special mention here. Linked Recognition of Foreign T-Cell Epitopes and Self B-Cell Epitopes One detriment of the human immune system’s use of linked recognition of T-cell and B-cell epitopes is that slightly altered self-antigens run a considerable risk for inducing humoral autoimmunity. For example, there are cases in which, although thymic education is complete, some B cells with immunoglobulins that recognize self-antigens persist. This appears to be the case for B cells specific for RBC antigens.54,55 When these autoreactive B cells encounter the antigen recognized by their rearranged immunoglobulins, the antigen is phagocytosed and peptides from that antigen are presented on class II MHC molecules of the B cell. In this scenario, because all T cells that are capable of recognizing such peptide–MHC complexes have been deleted in the thymus, humoral tolerance is maintained. However, this mecha-

nism of tolerance fails if the self-protein recognized by an autoreactive B cell has a foreign T-cell epitope attached to it. In this setting, the B cell now presents a foreign peptide–class II MHC epitope and receives the help required to differentiate into an antibody-secreting plasma cell. Thus, the linkage of a foreign T-cell epitope to a self B-cell epitope may result in humoral autoimmunity. For such a mechanism to have physiologic relevance, one must ask in what situations would the immune system encounter a self B-cell epitope linked to a foreign T-cell epitope. One scenario is the exposure to foreign human tissues during transfusion. Many genes have numerous allelic variants throughout the human population. This is perhaps best demonstrated in the extensive catalog of human blood group antigens that have now been described. Although the described allelic variations of blood group antigens have been predominantly detected serologically, and are thus best known as B-cell epitopes, they can also serve as T-cell epitopes if the HLA of a given individual can present peptides containing the amino acid variation. Moreover, there may be a considerable number of polymorphisms that do not alter B-cell epitopes and have thus not been detected serologically, but can still constitute a variant of a T-cell epitope presented by class II MHC molecules. For example, a donor and recipient may both be positive for the Rh D epitope. However, the donor may have an amino acid difference in an internal portion of the Rh molecule. If the peptide containing this variation can be presented by the recipient HLA, then this constitutes the linkage of a foreign T-cell epitope to a self B-cell epitope. This is precisely the situation in which T-cell tolerance would be circumvented. This mechanism is hypothetical in the context of RBC transfusion and has not been formally tested. However, it is consistent with wellestablished experimental immunology in which linkage of foreign T-cell epitopes to self B-cell epitopes can lead to autoantibodies.68 ”Autoantibodies” of Donor Origin: RBC-Specific Humoral Graft-versus-Host Disease When an individual’s immune system generates antibodies against their own tissue, it is generally assumed that their own B cells are the origin of the antibody. However, transfusion of blood results in the introduction of foreign B cells into the recipient. Because transfused leukocytes have the capacity to proliferate, transfusion of very few cells can be sufficient to generate a chimeric state. Indeed, it has been reported that even in the case of leukoreduced blood, sufficient leukocytes are transfused to cause microchimerism in some patients.69 Thus, transfusion of RBCs introduces donor lymphocytes into the recipient immune system. Aside from special populations at risk for transfusionassociated graft-versus-host disease (GVHD), including immunosuppressed patients and related individuals, the clinical consequences of transfusing foreign leukocytes are seldom significant. Transfusion of nonleukoreduced blood is routinely performed without any observable negative consequences. However, there are exceptions. It has been reported that in one patient transfusion-associated anti-RBC “autoantibodies” had immunoglobulin allotypes that were not part of the patient’s genome.70 Although one cannot rule out the possibility of spontaneous mutation of a patient’s allotypes, the most likely explanation is that the antibodies were produced by B cells from a transfusion donor. A certain number of transfusion-associated anti-RBC autoantibodies

PRINCIPLES OF RED BLOOD CELL ALLO- AND AUTOANTIBODY FORMATION AND FUNCTION

frequency in patients with sickle cell anemia. For example, one study reported that 8% of pediatric patients and 9.7% of adults with sickle cell anemia who received transfusions developed anti-RBC autoantibodies. In a separate retrospective study, of 2618 patients with a positive DAT or indirect antiglobulin test, 121 (4.6%) were reported to also have autoantibodies. Of the patients with autoantibodies, 10% also had an identifiable alloantibody and generated both the alloantibody and autoantibody subsequent to transfusion.66 Due to the inability to obtain transfusion records on all patients, this frequency of autoimmunization secondary to transfusion may be an underestimate. Although the generation of new anti-RBC autoantibodies can have a strong temporal association to antecedent RBC transfusion, such observations constitute only a correlation and do not establish causality. Moreover, as transfusion recipients clearly have an underlying pathology that necessitates the transfusion in the first place, it is practically impossible to isolate the transfusion as the only variable. However, several settings do allow independent investigation of the role of transfusions on autoantibody development. For example, normal Rh-negative volunteers can be transfused with Rhpositive blood to generate a source of anti-Rh immunoglobulin.17 In one study involving 34 volunteers, two participants developed a positive DAT on their own RBCs during the immunization protocol.17 No clinical hemolysis was observed in these two individuals, but RBC survival studies that could detect low-level hemolysis were not carried out. Additionally, in an animal model of transfusion-induced AIHA, repeated transfusion of xenogeneic RBCs (rat into mouse) results in the generation of autoantibodies.67 Taken together, these observations indicate that generation of autoantibodies can be a direct sequelae of transfusion of RBCs.

4 47

BLOOD BANKING

may therefore not actually be “autoantibodies,” but instead may be alloantibodies synthesized in situ by donor B cells, thus actually representing humoral GVHD. Typically, transfused leukocytes are rejected by the recipient immune system as foreign. It is thus unclear why a transfused B cell would survive and differentiate into a plasma cell. In addition, it seems very unlikely that alloreactive B cells would be transfused, because the precursor frequency of donor B cells specific for recipient RBC antigens is expected to be low and relatively few B cells are transfused even in nonleukoreduced blood. However, it has been pointed out that up to 10% of transfusion donors were previously transfusion recipients.71 In addition, any female who has carried a child may be alloimmunized to paternal RBC antigens. Given these facts, donor precursor frequencies of anti-RBC B cells may be higher than expected. Because these RBC-specific B cells would likely be memory cells, their activation requirements would be lower, which may make engraftment and antibody production more likely. The frequency by which this phenomenon occurs is undetermined.

CELLULAR IMMUNIZATION IN RESPONSE TO RBC TRANSFUSION

II 48

Typically and traditionally, clinical monitoring of immunization to RBCs is carried out by screening for antibodies that are detected by an agglutination reaction. This process of mounting an antibody response to transfused RBCs is known to involve activation and differentiation of CD4+ helper T cells that recognize peptides from RBC antigens presented on class II MHC molecules of the recipient’s APCs. The activated helper T cells then give the required signals to antigen-specific B cells to allow differentiation into antibody-secreting plasma cells. However, in addition to helper T cells, it has also been appreciated that transfusion of blood products can result in the generation of class I MHC-restricted CD8+ T cells with lytic activity. Except in exceptionally rare cases of autoimmunity, cellular immunity is not known to be capable of contributing to hemolysis of transfused RBCs.72 Nonetheless, to the extent that multiply transfused patients ultimately receive bone marrow or organ transplants, the generation of cellular immunity is relevant in that it can contribute to subsequent transplant rejection. Traditionally, units of packed RBCs have not been stringently leukoreduced. This results in the transfusion of a small but significant number of leukocytes of donor origin. Because leukocytes express class I MHC and can also serve as APCs, direct alloimmunization to donor MHC molecules results in activation of recipient T cells specific for donor human leukocyte antigens (HLAs). This can contribute to rejection of donor organs and bone marrow if they share the same HLA. The widespread implementation of leukoreduction filters for units of RBCs has significantly limited the exposure of transfusion recipients to donor leukocytes. Because human RBCs do not express MHC (with the possible exception of the Bg blood group), direct alloimmunization of recipient T cells is highly unlikely. However, it has been reported that minor histocompatibility antigens on donor RBCs are efficiently crosspresented into the class I MHC of recipient APCs.30 This results in expansion of recipient CD8+ T

cells specific for the minor histocompatibility antigen. In the context of transplantation of HLA-matched tissues, this immunization to minor histocompatibility antigens may contribute to rejection if the organ donor and transfusion donor have the same minor histocompatibility antigens. Thus, although leukoreduction of blood products may significantly decrease rates of alloimmunization to major histocompatibility antigens, transfusion of RBCs themselves has the potential to immunize against minor histocompatibility antigens. As previously discussed, this has little relevance to transfusion, because cellular responses do not typically lyse RBCs. However, it is relevant in the setting of the patient receiving transfusions as support prior to transplantation.

BASIC SCIENCE OF RBC DESTRUCTION Nonimmunologic Mechanisms of Hemolysis Hemolysis of transfused RBCs is the major sequelae prevented by crossmatching blood prior to transfusion. Despite this precaution, occasionally one observes the hemolysis of crossmatch-compatible blood. In some cases, hemolysis of crossmatch-compatible blood is an immune-mediated mechanism that involves undetected antibodies. However, mechanisms exist by which transfused RBCs can be hemolyzed in the absence of an alloantibody. To begin with, nonimmune-based physical factors can lead to hemolysis. For example, erythrocytes are highly sensitive to osmotic damage. Thus, if RBCs are transfused through an intravenous line that is simultaneously delivering hypotonic saline, direct lysis can occur. This can lead to massive hemolysis that presents clinically as an acute hemolytic transfusion reaction. Alternatively, RBC hemolysis can occur in the recipient due to nonantibody-based factors. For example, it has been reported that transfusion of RBCs from donors deficient in glucose-6-phosphate dehydrogenase can result in considerable hemolysis, especially in infants.73–76 There are also antibody-dependent mechanisms of RBC destruction that do not require antibody binding to the RBCs. For example, large-scale complement activation by immune complexes not associated with RBCs can lead to complement sensitization of nearby RBCs that leads to “bystander hemolysis” of RBCs not coated with antibodies.77 In addition, cellular immunity can lead to hemolysis in rare settings, and it has been reported that natural killer cells can lyse RBCs in DAT-negative AIHA.72 Despite these nonimmunologic mechanisms of RBC destruction, one must take care not to mistakenly exclude antibody-mediated hemolysis on the basis of a negative antibody screen. Although an anti-RBC antibody is usually detectable with immune-mediated destruction of transfused RBCs, there are exceptions. It has been observed that Rh-positive RBCs have a decreased life span in Rh-negative individuals previously exposed to Rh-positive blood, even if no anti-Rh is detectable.78 Such individuals typically proceed to generate detectable anti-Rh antibodies on subsequent transfusion with Rh-positive blood. Thus, there appears to be a level of anti-RBC antibody that is capable of causing hemolysis but is below the threshold of detection by agglutination-based assays. In addition, not all antibodies that bind RBCs in vivo are detected in vitro. Also, anamnestic responses

Antibody-Mediated Hemolysis Although certain antibodies, such as anti-A and anti-B IgM, cause rapid intravascular hemolysis through complement fixation, the majority of clinically significant anti-RBC antibodies lead to delayed hemolytic transfusion reactions, in which hemolysis occurs over a matter of days to weeks. In this process, RBCs are opsonized by IgG, which can be augmented by complement proteins and results in removal of the RBCs by cells of the reticuloendothelial system. This occurs in the extravascular space and is mediated predominantly by tissue macrophages in the spleen and liver. In addition to hemolysis as a result of RBC phagocytosis, RBCs can also be lysed by the direct release of lysosomal proteases.77 However, in either case, the process is facilitated by anti-RBC immunoglobulin. There has been considerable investigation into the isotype and subclass of antibodies involved in hemolysis of RBCs. The majority of this work has been performed in the setting of AIHA, but likely applies to hemolysis of incompatible transfusions as well. The predominant antibody involved in delayed hemolytic transfusion reaction is of the IgG type. The presence of IgG on the RBC surface effectively opsonizes the RBC through interaction with Fc receptors on phagocytic macrophages predominantly in the spleen and liver. Although anti-RBC IgM typically results in complement-mediated intravascular hemolysis, IgM has been reported to promote a delayed hemolytic reaction in some settings.77 This likely occurs through deposition of complement proteins on the RBC surface, which opsonizes RBCs via complement receptors on macrophages, because macrophages do not express IgM-binding Fc receptors. In support of this notion, it has been reported that IgM-coated RBCs have a normal life span in humans with complement deficiencies.77 Although very uncommon, isolated anti-RBC IgA molecules have also been reported in patients with hemolysis.77 However, the conclusion that the IgA is responsible for hemolysis in this setting is predicated on the assumption that no IgG was present. As it has been shown that the level of IgG necessary to induce hemolysis can be below the level of detection by standard antiRBC IgG assays, the legitimacy of this assumption is unclear. The potential of different IgG subclasses to induce hemolysis has been studied in both humans and mice. Several groups have performed large-scale analysis of DATpositive patients with or without clinically evident AIHA in an attempt to identify the IgG subclass dependence of hemolysis.79–82 Garratty performed an in-depth comparison of IgG subclasses involved in DAT-positive individuals with or without AIHA.82 This analysis of 78 patients showed IgG1 to be the sole IgG subclass present in the majority of AIHA cases. However, IgG1 was also found on 72% of patients without clinical AIHA, and the presence of IgG1 does not predict whether a DAT-positive individual will experience hemolysis. In an analysis of 304 patients, Sokol and colleagues reported similar findings, with anti-RBC IgG1 present in 98% of cases and IgG1 as the sole subclass in 64% of cases.80 A smaller analysis of 34 patients also had similar findings concerning the predominance of IgG1.80 Garratty’s analysis also demonstrated that some patients had IgG2 alone or IgG3 alone, but that neither was more prevalent in patients with AIHA compared with patients with no hemolysis. Isolated IgG4

was found only in five normal blood donors and none of the AIHA patients analyzed. The above findings demonstrate that although IgG1 is the most common anti-RBC IgG subclass, isolated IgG1, IgG2, and IgG3 can lead to hemolysis. However, the detection of an isolated IgG subclass does not predict the likelihood of hemolysis. In contrast, the presence of multiple simultaneous subclasses clearly shows an increased risk for hemolysis.82 Although this could represent a cooperativity of subclass, the presence of multiple IgG subclasses resulted in a larger amount of overall IgG on the RBC surface.80 Indeed, it has been reported that the overall quantity of IgG on the RBC surface predicts the likelihood of hemolysis in vivo.77,79,80,83–87 Considerable evidence exists suggesting that additional genetic and environmental factors may play a role in the question of IgG subclass and hemolysis. For instance, in contrast to the above conclusions, it has been reported that IgG3 is more potent than other subclasses at promoting RBC phagocytosis by monocytes,88 and it has also been reported that AIHA with IgG3 involvement may be less responsive to treatment.89 However, a direct comparison in the clearance of Rh-positive cells from Rh-negative volunteers given monoclonal anti-Rh of the IgG1 or IgG3 type showed that RBCs were cleared from IgG1 recipients at a significantly higher rate than in IgG3 recipients.90 Interestingly, volunteers receiving IgG3 had a wide distribution of responses, with rapid clearance in some individuals and slow clearance in others. Subsequent studies by Kumpel and colleagues demonstrated that naturally occurring allelic variants in the FcγRIIIa receptor alter the efficiency with which IgG3-coated RBCs are cleared.91 In addition, an animal model of IgGinduced hemolytic anemia demonstrated that concurrent viral infection significantly enhances disease with a hemolytic IgG2a antibody but not with a hemolytic IgG1 antibody due to an increase in erythrophagocytic capacity of macrophages.92 Thus, genetic variation in accessory molecules other than IgG may alter the IgG subclass dependence of AIHA in different patient populations. Likewise, environmental factors and patient-specific pathology that leads to inflammation and macrophage activation may significantly affect hemolysis by RBC-binding antibodies. Although animal models are biologically distinct from human physiology, they represent a setting in which highly controlled studies can be carried out on genetically identical subjects in a controlled environment. Izui and colleagues and others have made extensive use of a murine model of AIHA to dissect out the biology of IgG-mediated RBC hemolysis.92–98 In this model system, both high-affinity and lowaffinity monoclonal antibodies against the RBC autoantigen have been isolated. Similar to human IgG-mediated hemolysis, complement does not appear to be required for RBC removal in the murine model, because hemolytic anemia still occurs in mice lacking C5 or C3.93 However, both Fc receptor-dependent phagocytosis (mostly in the liver) and RBC sequestration of RBCs (mostly in the spleen) contribute to decreased hematocrit in response to autoantibodies.93 To control for intrinsic differences in the epitope specificity and affinity of different monoclonal antibodies, recombinant genetics have been employed to create artificial monoclonal antibodies of each murine IgG subclass that have precisely the same epitope-binding domains. IgG1, IgG2a, and IgG2b were each hemolytic, whereas IgG3 did not promote significant hemolysis.96 IgG2a was 20 to 100 times more potent in inducing hemolysis than IgG1 or IgG2b. Gene deletion

PRINCIPLES OF RED BLOOD CELL ALLO- AND AUTOANTIBODY FORMATION AND FUNCTION

may result in an initial negative screen followed by a delayed hemolysis and conversion to a positive screen.

4 49

BLOOD BANKING

studies demonstrated that the FcγRIII receptor was responsible for hemolysis by IgG1, whereas both FcγRI and FcγRIII were involved in hemolysis by IgG2a and IgG2b. FcγRIIB did not appear to contribute to hemolysis, consistent with its generally understood role as an inhibitory molecule. Recently, a new FcγR has been identified (FcγRIV), which binds both IgG2a and IgG2b and can contribute to antibody-mediated thrombocytopenia.99 The role of FcγRIV has not yet been formally tested in antibody-mediated hemolysis.

Antibody-Induced Antigen Loss/Suppression

II 50

Although many anti-RBC antibodies can result in either intravascular or extravascular hemolysis of the RBCs that they bind, hemolysis is not the inevitable result of antibody binding. There are certainly a considerable number of individuals who have strongly positive DATs without hemolysis. Moreover, numerous antibodies have been described against “clinically insignificant” blood group antigens for which crossmatch-incompatible blood can be transfused without ill effect. Additionally, there is a somewhat obscure (but well described) phenomenon in which the RBC phenotype will convert from positive to negative for a given antigen after encountering immunoglobulins that bind to the antigen. This phenomenon has been referred to as weakened antigenicity, antigen reduction, antigen suppression, or acquired loss and has been reported to occur for many blood group antigens, including Rh (D and e), Kell, Kidd, Duffy, Lutheran, LW, Co, Ge, En(a), AnWj, and Sc1.77,100–112 Although the mechanisms of antigen loss have not been fully elucidated, publications documenting the phenomenon have proposed a number of possible explanations. It has been suggested that the antibodies to a given RBC antigen are toxic to newly formed RBCs and that the presence of such antibodies alters hematopoiesis such that only antigen-negative RBCs escape the marrow intact.102,105 However, this proposed mechanism cannot account for the observation that antigen loss can also be seen in the setting of transfusion of crossmatch-incompatible RBCs, which have already synthesized the antigen being recognized.101,107 It has also been suggested that antigens might be lost due to factors other than immunoglobulins, such as destruction by microbial enzymes. However, suppression of antigen correlates more with the presence of antibody and, in most cases, antigen suppression does not involve a documented infection. Nonetheless, the presence of subclinical infections cannot be ruled out. It has also been suggested that antibody binding results in the shedding of antigen from mature erythrocytes through formation of antigen–antibody immune complexes.102 Insights into the mechanisms of antibody-induced antigen loss have been gained from an animal model of this phenomenon. It has been reported that presynthesized antigen is shed from transfused RBCs while leaving the RBC intact and without decreasing circulatory life span.113 This process requires the presence of both IgG and FcγRIII, but it occurs normally in splenectomized animals. Thus, in this animal model, antigen loss involves neither bone marrow suppression of RBC antigen synthesis nor microbial destruction. In contrast, antigen appears to be lost through interactions between RBCs coated with IgG and FcγRIII-bearing cells. However, the exact mechanism in humans remains to be determined, and antigen loss may occur as a result of distinct mechanisms in different settings.

It is unclear why mechanisms that result in antibodyinduced antigen loss may evolve. In this context, it is worth noting that biological mechanisms have been described by which RBCs shed whole antigens as part of normal RBC biology. CD35, which is also known as the CR1 complement receptor and the Knops blood group antigen, is a receptor for the C3b fragment of complement.114 In what has been termed the immune-adherence phenomenon, circulating immune complexes or antibody-bound bacteria bind to RBCs via interactions of CD35 on RBC and C3b on the immune complex.115,116 When a RBC encounters hepatic macrophages, the immune complex is transferred to the phagocyte. Like antibody-induced antigen loss, RBCs delivering immune complexes to macrophages lose their CD35 surface antigen but are not destroyed.117–119 Because this process appears to depend on the Fc portion of the antibodies, interactions with Fc receptors may also be involved.120 Furthermore, because immune complex transfer occurs in the liver,120,121 an intact spleen is presumably not required. Whether loss of CD35 is due to complete removal of the molecule or due to proteolytic fragmentation by Fc-bearing cells is unresolved. However, it has been demonstrated that no residual cytoplasmic domain of CD35 remains after loss of CD35 from podocytes during immune complex–based pathophysiology of lupus, a situation that resembles CD35 loss from RBC delivering immune complexes to phagocytes.122 Thus, despite the fact that CD35 is a transmembrane protein, complete extrusion of the CD35 molecule seems a distinct possibility. Overall, the natural process of RBC-mediated delivery of immune complexes to the liver resembles antibody-mediated antigen loss in several regards and suggests that antibody-induced antigen loss may be functioning through existing pathways that evolved to deliver immune complexes to hepatic macrophages.

REFERENCES 1. Heddle NM, Soutar RL, O’Hoski PL, et al. A prospective study to determine the frequency and clinical significance of alloimmunization post-transfusion. Br J Haematol 1995;91:1000–1005. 2. Hoeltge GA Domen RE, Rybicki LA, Schaffer PA. Multiple red cell transfusions and alloimmunization. Experience with 6996 antibodies detected in a total of 159,262 patients from 1985 to 1993. Arch Pathol Lab Med 1995;119:42–45. 3. Seyfried H, Walewska I. Analysis of immune response to red blood cell antigens in multitransfused patients with different diseases. Materia Medica Polona 1990;22:21–25. 4. Blumberg N Peck K, Ross K, Avila E. Immune response to chronic red blood cell transfusion. Vox Sang 1983;44:212–217. 5. Rosse WF. Gallagher D, Kinney TR, et al. Transfusion and alloimmunization in sickle cell disease. The Cooperative Study of Sickle Cell Disease. Blood 1990;76:1431–1437. 6. Aygun B, Padmanabhan S, Paley C, Chandrasekaran V. Clinical significance of RBC alloantibodies and autoantibodies in sickle cell patients who received transfusions. Transfusion 2002;42:37–43. 7. Garratty G. Severe reactions associated with transfusion of patients with sickle cell disease. Transfusion 1997;37:357–361. 8. Issitt PD, Combs MR, Bumgarner DJ, et al. Studies of antibodies in the sera of patients who have made red cell autoantibodies. Transfusion 1996;36:481–486. 9. Leger RM, Garratty G. Evaluation of methods for detecting alloantibodies underlying warm autoantibodies. Transfusion 1999;39: 11–16. 10. Branch DR Petz LD. Detecting alloantibodies in patients with autoantibodies.. Transfusion 1999;39:6–10. 11. Stiegler G, Sperr W, Lorber C, et al. Red cell antibodies in frequently transfused patients with myelodysplastic syndrome. Ann Hematol 2001;80:330–333. 12. Novaretti MC, Sopelete CR, Velloso ER, et al. Immunohematological findings in myelodysplastic syndrome. Acta Haematol 2001;105:1–6.

46. Mond JJ, Lees A, Snapper CM. T cell-independent antigens type 2. Annu Rev Immunol 1995;13:655–692. 47. Boctor FN, Ali NM, Mohandas K, Uehlinger J. Absence of D-alloimmunization in AIDS patients receiving D-mismatched RBCs. Transfusion 2003;43:173–176. 48. Perry FE, Barker RN, Mazza G, et al. Autoreactive T cell specificity in autoimmune hemolytic anemia of the NZB mouse. Eur J Immunol 1996;26:136–141. 49. Hall AM, Vickers MA, McLeod E, Barker RN. Rh autoantigen presentation to helper T cells in chronic lymphocytic leukemia by malignant B cells. Blood 2005;105:2007–2015. 50. Watanabe N, Ikuta K, Fagarasan S, et al. Migration and differentiation of autoreactive B-1 cells induced by activated γ/δ T cells in antierythrocyte immunoglobulin transgenic mice. J Experiment Med 2000;192:1577–1586. 51. Goodnow CC, Crosbie J, Adelstein S, et al. Altered immunoglobulin expression and functional silencing of self-reactive B lymphocytes in transgenic mice. Nature 1988;334:676–682. 52. Hartley SB, Crosbie J, Brink R, et al. Elimination from peripheral lymphoid tissues of self-reactive B lymphocytes recognizing membranebound antigens. Nature 1991;353:765–769. 53. Hartley SB, Cooke MP, Fulcher DA, et al. Elimination of self-reactive B lymphocytes proceeds in two stages: Arrested development and cell death. Cell 1993;72:325–335. 54. Okamoto M, Murakami M, Shimizu A, et al. A transgenic model of autoimmune hemolytic anemia. J Experiment Med 1992;175:71–79. 55. Murakami M, Nakajima K, Yamazaki K, et al. Effects of breeding environments on generation and activation of autoreactive B-1 cells in anti-red blood cell autoantibody transgenic mice. J Experiment Med 1997;185:791–794. 56. Stellrecht KA, Vella AT. Evidence for polyclonal B cell activation as the mechanism for LCMV-induced autoimmune hemolytic anemia. Immunol Lett 1992;31:273–277. 57. De Rossi G, Granati L, Girelli G, et al. Incidence and prognostic significance of autoantibodies against erythrocytes and platelets in chronic lymphocytic leukemia (CLL). Nouv Rev Franc Hematol 1988;30:403–406. 58. Roelcke D. Cold agglutination. Transfus Med Rev 1989;3:140–166. 59. Fagiolo E. Immunological tolerance loss vs. erythrocyte self antigens and cytokine network disregulation in autoimmune hemolytic anaemia. Autoimmunity Rev 2004;3:53–59. 60. Toriani-Terenzi C, Fagiolo E. Th2 cytokine role in autoimmune haemolytic anaemia (AIHA) pathogenesis. Panminerva Medica 2001;43:1–5. 61. Mqadmi A, Zheng X, Yazdanbakhsh K. CD4+CD25+ regulatory T cells control induction of autoimmune hemolytic anemia. Blood 2005;105:3746–3748. 62. Barker RN, Elson CJ. Multiple self epitopes on the Rhesus polypeptides stimulate immunologically ignorant human T cells in vitro. Eur J Immunol 1994;24:1578–1582. 63. Fagiolo E, Toriani-Terenzi C. Mechanisms of immunological tolerance loss versus erythrocyte self-antigens and autoimmune hemolytic anemia. Autoimmunity 2003;36:199–204. 64. Barker RN, Hall AM, Standen GR, et al. Identification of T-cell epitopes on the Rhesus polypeptides in autoimmune hemolytic anemia. Blood 1997;90:2701–2715. 65. Elson CJ, Barker RN, Thompson SJ, Williams NA. Immunologically ignorant autoreactive T cells, epitope spreading and repertoire limitation. Immunol Today 1995;16:71–76. 66. Young PP, Uzieblo A, Trulock E, et al. Autoantibody formation after alloimmunization: Are blood transfusions a risk factor for autoimmune hemolytic anemia? [see comment]. Transfusion 2004;44:67–72. 67. Cox KO, Keast D. Erythrocyte autoantibodies induced in mice immunized with rat erythrocytes. Immunol 1973;25:531–539. 68. Dalum I, Jensen MR, Hindersson P, et al. Breaking of B cell tolerance toward a highly conserved self protein. J Immunol 1996;157:4796–4804. 69. Lee TH, Paglieroni T, Utter GH, et al. High-level long-term white blood cell microchimerism after transfusion of leukoreduced blood components to patients resuscitated after severe traumatic injury. Transfusion 2005;45:1280–1290. 70. Ishikura H, Endo J, Saito Y, et al. Graft-versus-host antibody reaction causing a delayed hemolytic anemia after blood transfusion. Blood 1993;82:3222–3223. 71. Garratty G. Autoantibodies induced by blood transfusion [comment]. Transfusion 2004;44:5–9. 72. Gilsanz F, De La Serna J, Molto L, Alvarez-Mon M. Hemolytic anemia in chronic large granular lymphocytic leukemia of natural killer cells: Cytotoxicity of natural killer cells against autologous red cells is associated with hemolysis. Transfusion 1996;36:463–466.

PRINCIPLES OF RED BLOOD CELL ALLO- AND AUTOANTIBODY FORMATION AND FUNCTION

13. Lostumbo MM, Holland PV, Schmidt PJ. Isoimmunization after multiple transfusions. N Engl J Med 1966;275:141–144. 14. Schonewille H, Haak HL, van Zijl AM. Alloimmunization after blood transfusion in patients with hematologic and oncologic diseases. Transfusion 1999;39:763–771. 15. Blumberg N, Heal JM,Gettings KF. WBC reduction of RBC transfusions is associated with a decreased incidence of RBC alloimmunization. Transfusion 2003;43:945–952. 16. Frohn C, Dumbgen L, Brand JM, et al. Probability of anti-D development in D− patients receiving D+ RBCs. Transfusion 2003;43:893–898. 17. Cook IA. Primary rhesus immunization of male volunteers. Br J Haematol 1971;20:369–375. 18. Freda VJ, Gorman JG, Pollack W, et al. Prevention of Rh isoimmunization. Progress report of the clinical trial in mothers. JAMA 1967;199:390–394. 19. Gunson HH, Stratton F, Cooper DG, Rawlinson VI. Primary immunization of Rh-negative volunteers. BMJ 1970;1:593–595. 20. Hattersley P. Two popular fallacies regarding Rh. J Lab Clin Med 1947;32:423. 21. Waller RK. Intentional isoimmunizations against the antigen D(Rh0). J of Lab Clin Med 1949;34:270. 22. Wiener AS. Further observations on isosensitization to the Rh factor. Proc Soc Exper Biol Med 1949;70:576. 23. Clarke CA, Donohoe WT, Mc C R, et al. Further experimental studies on the prevention of Rh haemolytic disease. BMJ 1963;5336:979. 24. Azar MM, Wyche AA. Route of antigen administration for tolerance production. Life Sciences 1974;14:2151–2157. 25. Endres RO, Grey HM. Antigen recognition by T cells. II. Intravenous administration of native or denatured ovalbumin results in tolerance to both forms of the antigen. J Immunol 1980;125:1521–1525. 26. Weigle WO. Immunological unresponsiveness. Adv Immunol 1973; 16:61–122. 27. Magnani M, Chiarantini L, Vittoria E, et al. Red blood cells as an antigen-delivery system. Biotech Appl Biochem 1992;16:188–194. 28. Dominici S, Laguardia ME, Serafini G, et al. Red blood cell-mediated delivery of recombinant HIV-1 Tat protein in mice induces anti-Tat neutralizing antibodies and CTL. Vaccine 2003;21:2073–2081. 29. Zimring JC, Hair GA, Anderson KM, et al. Use of red blood cells as a vaccine vector to induce humoral and cellular immunity. Transfusion 2005;45S:24A. 30. Zimring JC, Hair GA, Deshpande SS, Horan JT. Immunization to minor histocompatibility antigens on transfused RBC through crosspriming into recipient MHC class I pathways. Blood 2005;107:187–189. 31. Matzinger P. The danger model: a renewed sense of self. Science 2002;296:301–305. 32. Matzinger P. Tolerance, danger, and the extended family. Annu Rev Immunol 1994;12:991–1045. 33. Akira S, Takeda K. Toll-like receptor signalling. Nature Rev Immunol 2004;4:499–511. 34. Takeda K, Kaisho T, Akira S. Toll-like receptors. Annu Rev Immunol 2003;21:335–376. 35. Imler JL, Hoffmann JA. Toll and Toll-like proteins: An ancient family of receptors signaling infection. Rev Immunogenetics 2000;2:294–304. 36. Murillo LS, Morre SA, Pena AS. Toll-like receptors and NOD/CARD proteins: Pattern recognition receptors are key elements in the regulation of immune response. Drug Today 2003;39:415–438. 37. Modlin RL. Mammalian Toll-like receptors. Ann Allergy Asthma Immunol 2002;88:543–547. 38. Mollinson PL, Engelfriet CP, Contreras M. Blood Transfusion in Clinical Medicine. Oxford, Blackwell Science, 1997, pp 288–291. 39. Mitchison NA. Induction of immunological paralysis in two zones of dosage. Proc R Soc Lond B Biol Sci 1964;161:275–292. 40. Ciurea A, Hunziker L, Klenerman P, et al. Impairment of CD4+ T cell responses during chronic virus infection prevents neutralizing antibody responses against virus escape mutants. J Experiment Med 2001;193:297–305. 41. Fuller MJ, Zajac AJ. Ablation of CD8 and CD4 T cell responses by high viral loads. J Immunol 2003;170:477–486. 42. Singh NJ, Schwartz RH. The strength of persistent antigenic stimulation modulates adaptive tolerance in peripheral CD4+ T cells. J Experiment Med 2003;198:1107–1117. 43. Delemarre FG, Kors N, van Rooijen N. Elimination of spleen and of lymph node macrophages and its difference in the effect on the immune response to particulate antigens. Immunobiol 1990;182:70–78. 44. Trombetta ES, Mellman I. Cell biology of antigen processing in vitro and in vivo. Annu Rev Immunol 2005;23:975–1028. 45. Guermonprez P, Valladeau J, Zitvogel L, et al. Antigen presentation and T cell stimulation by dendritic cells. Annu Rev Immunol 2002;20:621–667.

4 51

BLOOD BANKING

II 52

73. Kumar P, Sarkar S, Narang A. Acute intravascular haemolysis following exchange transfusion with G-6-PD deficient blood. Eur J Pediatr 1994;153:98–99. 74. Mimouni F, Shohat S, Reisner SH. G6PD-deficient donor blood as a cause of hemolysis in two preterm infants. Isr J Medical Sci 1986;22:120–122. 75. Gulati S, Singh S, Narang A, Bhakoo ON. Exchange transfusion with G-6-PD deficient donor blood causes exaggeration of neonatal hyperbilirubinemia. Ind Pediatr 1989;26:499–501. 76. Shalev O, Bogomolski-Yahalom V, Sharon R. Hemolysis following transfusion of erythrocytes from a donor with G6PD deficiency and betathalassemia minor. Isr J Medical Sci 1993;29:214–216. 77. Petz LD, Garratty G. Immune Hemolytic Anemias. Philadelphia, Churchill Livingstone, 2004. 78. Petz LD, Swisher SN, Kleinman S, et al. Clinical Practice of Transfusion Medicine. New York, Churchill Livingstone, 1996, p 42. 79. Dubarry M, Charron C, Habibi B, et al. Quantitation of immunoglobulin classes and subclasses of autoantibodies bound to red cells in patients with and without hemolysis. Transfusion 1993;33:466–471. 80. Sokol RJ, Hewitt S, Booker DJ, Bailey A. Erythrocyte autoantibodies, subclasses of IgG and autoimmune haemolysis. Autoimmunity 1990;6:99–104. 81. Sokol RJ, Hewitt S, Booker DJ, Bailey A. Red cell autoantibodies, multiple immunoglobulin classes, and autoimmune hemolysis. Transfusion 1990;30:714–717. 82. Garratty G. Factors affecting the pathogenicity of red cell auto- and alloantibodies: immune destruction of red blood cells. Arlington, Va., American Association of Blood Banks, 1989. 83. Lynen R, Neuhaus R, Schwarz DW. Flow cytometric analyses of the subclasses of red cell IgG antibodies. Vox Sang 1995;69:126–130. 84. Garratty G, Nance SJ. Correlation between in vivo hemolysis and the amount of red cell-bound IgG measured by flow cytometry. Transfusion 1990;30:617–621. 85. Merry AH, Thomson EE, Rawlinson VI, Stratton F. Quantitation of IgG on erythrocytes: Correlation of number of IgG molecules per cell with the strength of the direct and indirect antiglobulin tests. Vox Sang 1984;47:73–81. 86. Merry AH, Thomson EE, Rawlinson VI, Stratton F. A quantitative antiglobulin test for IgG for use in blood transfusion serology. Clin Lab Haematol 1982;4:393–402. 87. van der Meulen FW, de Bruin HG, Goosen PC, et al. Quantitative aspects of the destruction of red cells sensitized with IgG1 autoantibodies: an application of flow cytofluorometry. Br J Haematol 1980;46:47–56. 88. Zupanska B, Brojer E, Thomson EE, et al. Monocyte–erythrocyte interaction in autoimmune haemolytic anaemia in relation to the number of erythrocyte-bound IgG molecules and subclass specificity of autoantibodies. Vox Sang 1987;52:212–218. 89. Li Z, Shao Z, Xu Y, et al. Subclasses of warm autoantibody IgG in patients with autoimmune hemolytic anemia and their clinical implications. Chin Medical J 1999;112:805–808. 90. Kumpel BM, Goodrick MJ, Pamphilon DH, et al. Human Rh D monoclonal antibodies (BRAD-3 and BRAD-5) cause accelerated clearance of Rh D+ red blood cells and suppression of Rh D immunization in Rh D− volunteers. Blood 1995;86:1701–1709. 91. Kumpel BM, De Haas M, Koene HR, et al. Clearance of red cells by monoclonal IgG3 anti-D in vivo is affected by the VF polymorphism of Fcγ RIIIa (CD16). Clin Exper Immunol 2003;132:81–86. 92. Meite M, Leonard S, Idrissi ME, et al. Exacerbation of autoantibody-mediated hemolytic anemia by viral infection. J Virol 2000;74:6045–6049. 93. Shibata T, Berney T, Reininger L, et al. Monoclonal anti-erythrocyte autoantibodies derived from NZB mice cause autoimmune hemolytic anemia by two distinct pathogenic mechanisms. Int Immunol 1990;2:1133–1141. 94. Azeredo da Silveira S, Kikuchi S, Fossati-Jimack L, et al. Complement activation selectively potentiates the pathogenicity of the IgG2b and IgG3 isotypes of a high affinity anti-erythrocyte autoantibody. J Exper Med 2002;195:665–672. 95. Fossati-Jimack L, Reininger L, Chicheportiche Y, et al. High pathogenic potential of low-affinity autoantibodies in experimental autoimmune hemolytic anemia. J Exper Med 1999;190:1689–1696. 96. Fossati-Jimack L, Ioan-Facsinay A, Reininger L, et al. Markedly different pathogenicity of four immunoglobulin G isotype-switch variants of an antierythrocyte autoantibody is based on their capacity to interact in vivo with the low-affinity Fcγ receptor III. J Experiment Med 2000;191:1293–1302.

97. Izui S, Fossati-Jimack L, da Silveira SA, Moll T. Isotype-dependent pathogenicity of autoantibodies: analysis in experimental autoimmune hemolytic anemia. Springer Semin Immunopathol 2001;23:433–445. 98. Meyer D, Schiller C, Westermann J, et al. FcγRIII (CD16)-deficient mice show IgG isotype-dependent protection to experimental autoimmune hemolytic anemia. Blood 1998;92:3997–4002. 99. Nimmerjahn F, Bruhns P, Horiuchi K, Ravetch JV. FcγRIV: A novel FcR with distinct IgG subclass specificity. Immunity 2005;23:41–45. 100. Beck ML, Marsh WL, Pierce SR, et al. Auto anti-Kpb associated with weakened antigenicity in the Kell blood group system: A second example. Transfusion 1979;19:197–202. 101. Brendel WL, Issitt PD, Moore RE, et al. Temporary reduction of red cell Kell system antigen expression and transient production of antiKpb in a surgical patient. Biotest Bull 1985;2:201–206. 102. Williamson LM, Poole J, Redman C, et al. Transient loss of proteins carrying Kell and Lutheran red cell antigens during consecutive relapses of autoimmune thrombocytopenia. Br J Haematol 1994;87:805–812. 103. Seyfried H, Gorska B, Maj S, et al. Apparent depression of antigens of the Kell blood group system associated with autoimmune acquired haemolytic anaemia. Vox Sang 1972;23:528–536. 104. Puig N, Carbonell F, Marty ML. Another example of mimicking antiKpb in a Kp(a+b−) patient. Vox Sang 1986;51:57–59. 105. Marsh WL, Oyen R, Alicea E, et al. Autoimmune hemolytic anemia and the Kell blood groups. Am J Hematol 1979;7:155–162. 106. Manny N, Levene C, Sela R, et al. Autoimmunity and the Kell blood groups: Auto-anti-Kpb in a Kp(a+b-) patient. Vox Sang 1983;45:252–256. 107. Vengelen-Tyler V, Gonzalez B, Garratty G, et al. Acquired loss of red cell Kell antigens. Br J Haematol 1987;65:231–234. 108. Koscielak J. Session on blood group substances and red cell membrane receptors. Vox Sang 1980;39:289. 109. Issitt PD, Gruppo RA, Wilkinson SL, Issitt CH. Atypical presentation of acute phase, antibody-induced haemolytic anaemia in an infant. Br J Haematol 1982;52:537–543. 110. Ganly PS, Laffan MA, Owen I, Hows JM. Auto-anti-Jka in Evans’ syndrome with negative direct antiglobulin test. Br J Haematol 1988; 69:537–539. 111. Issitt PD, Obarski G, Hartnett PL, et al. Temporary suppression of Kidd system antigen expression accompanied by transient production of anti-Jk3. Transfusion 1990;30:46–50. 112. Garratty G. Target antigens for red-cell-bound autoantibodies. In Nance SJ (ed). Clinical and Basic Science Aspects of Immunohematology. Arlington, Va., American Association of Blood Banks, 1991, p 33–72. 113. Zimring JC, Hair GA, Chadwick TE, et al. Nonhemolytic antibodyinduced loss of erythrocyte surface antigen. Blood 2005;106:1105–1112. 114. Fearon DT. Identification of the membrane glycoprotein that is the C3b receptor of the human erythrocyte, polymorphonuclear leukocyte, B lymphocyte, and monocyte. J Exper Med 1980;152:20–30. 115. Nelson RA Jr. The immune-adherence phenomenon: a hypothetical role of erythrocytes in defence against bacteria and viruses. Proc Royal Soc Med 1956;49:55–58. 116. Nelson RA Jr. The immune-adherence phenomenon: an immunologically specific reaction between microorganisms and erythrocytes leading to enhanced phagocytosis. Science 1953;118:733–737. 117. Cosio FG, Shen XP, Birmingham DJ, et al. Evaluation of the mechanisms responsible for the reduction in erythrocyte complement receptors when immune complexes form in vivo in primates. J Immunol 1990;145:4198–4206. 118. Cornacoff JB, Hebert LA, Smead WL, et al. Primate erythrocyteimmune complex-clearing mechanism. J Clin Invest 1983;71:236–247. 119. Davies KA, Hird V, Stewart S, et al. A study of in vivo immune complex formation and clearance in man. J Immunol 1990;144:4613–4626. 120. Nardin A, Lindorfer MA, Taylor RP. How are immune complexes bound to the primate erythrocyte complement receptor transferred to acceptor phagocytic cells? Molec Immunol 1999;36:827–835. 121. Nardin A, Schlimgen R, Holers VM, Taylor RP. A prototype pathogen bound ex vivo to human erythrocyte complement receptor 1 via bispecific monoclonal antibody complexes is cleared to the liver in a mouse model. Eur J Immunol 1999;29:1581–1586. 122. Moll S, Miot S, Sadallah S, et al. No complement receptor 1 stumps on podocytes in human glomerulopathies. Kidney Int 2001;59:160–168.

ii. Red Blood Cell, Platelet, and Leukocyte Antigens and Antibodies Chapter 5

Membrane Blood Group Antigens and Antibodies Marion E. Reid

●

Connie M. Westhoff

Erythrocyte blood group antigens are polymorphic, inherited carbohydrate or protein structures located on the outside surface of the red blood cell (RBC) membrane. Our ability to detect and identify blood group antigens and antibodies has contributed significantly to the current safe, supportive blood transfusion practice and to the appropriate management of pregnancies at risk for hemolytic disease of the fetus and newborn (HDN). Exposure to erythrocytes carrying an antigen lacking on the RBCs of the recipient can elicit an immune response in some individuals. Thus, blood group antigens are clinically important in allogeneic blood transfusions, maternofetal blood group incompatibility, and organ transplantation. By virtue of their relative ease of detection and generally straightforward mode of inheritance, blood group antigens have been used in genetic, forensic, and anthropologic investigations. The polymorphisms of blood groups have been exploited as a tool to monitor in vivo survival of transfused RBCs, to monitor engraftment of bone marrow transplants, and to monitor for blood doping by allogeneic transfusion in sports and athletics. Antigen profiles have been used to predict inheritance of diseases encoded by a gene in close proximity to the gene encoding the blood group antigen (e.g., the association of gene deletions causing chronic granulomatous disease with the loss of expression of Kell). Blood group antigens have also contributed to our understanding of cell membrane structure. The lack of Kell/Kx expression is associated with acanthocytes, Rh-deficient RBCs are stomatocytic, and mild elliptocytosis occurs in RBCs lacking Gerbich glycophorin C (GPC) and/or glycophorin D (GPD). More recently, in the postgenomic era, knowledge about the molecular basis associated with blood group antigens and phenotypes is being applied to the detection of single nucleotide polymorphisms (SNPs) associated with blood group antigens. Microarray and microchip technologies hold promise for transfusion medicine testing. This chapter reviews the scientific basis of molecules that express RBC blood group antigens and their relevance and application to the practice of transfusion medicine. One of the most important clinical concerns in contemporary transfusion medicine occurs with the discovery that a patient

requiring transfusion has an unexpected blood group antibody. Although the laboratory can characterize an antibody in terms of specificity, immunoglobulin class, and in vitro characteristics, it cannot always predict the antibody’s clinical significance. Therefore, when a blood group antibody is detected, an important, although often overlooked, first step is to review all available pertinent patient information (Table 5–1). One can then assess the potential for adverse effects by correlating the serologic information with the patient history and also with historical clinical experience in other patients. A patient who has not been exposed to RBCs by transfusion or pregnancy is unlikely to have a clinically significant alloantibody.

ERYTHROCYTE MEMBRANE The erythrocyte membrane consists of lipids, proteins, and carbohydrates, which interact to form a dynamic and fluid structure. By dry weight, the ratio of protein-to-lipid-tocarbohydrate in the RBC membrane is 49:43:8. The RBC membrane behaves as a semisolid, with elastic and viscous properties that are not observed with simple lipid vesicles. These properties are critical for the RBC to survive in the circulation for approximately 120 days during numerous cycles (approximately 75,000) and passages through narrow veins and sinusoids in the spleen. The RBC accomplishes this goal without intracellular machinery to repair damage. The multiple connections between the membrane skeleton and the lipid bilayer cause the bilayer to follow the contours of the membrane skeleton. Together, the membrane skeleton and lipid bilayer give the erythrocyte shape and resilience.1,2

Lipids The lipids in the RBC membrane form a bilayer, with the hydrophobic tails on the inside and the hydrophilic polar head groups to either the outside (extracellular) or the inside (cytoplasmic) surface (Fig. 5–1). The following three types of lipids occur in the RBC membrane: phospholipid (50%), cholesterol (40%), and glycolipids (10%). The arrangement

5 53

BLOOD BANKING

Table 5–1

Information for Problem-Solving in Immunohematology

Available Information

Considerations

Patient demographics

Diagnosis, age, sex, ethnicity, transfusion and/or pregnancy history, drugs, intravenous fluids (lactated Ringer’s solution, IVIgG, antilymphocyte globulin), infections, malignancies, hemoglobinopathies, stem cell transplantation ABO, Rh, direct antiglobulin test, phenotype, antibody detection, autologous control, crossmatch Hemoglobin, hematocrit, bilirubin, lactate dehydrogenase, reticulocyte count, haptoglobin, hemoglobinuria, albumin:globulin ratio, red blood cell (RBC) morphology Site and technique of collection, age of sample, anticoagulant, hemolysis, lipemic, color of serum/plasma, agglutinates/aggregates in the sample Check records in current and previous institutions for antibodies to blood group antigens Autologous control, phase of reactivity, potentiator (saline, albumin, low-ionic-strength solution, polyethylene glycol), reaction strength, effect of chemicals on antigen (proteases, thiol reagents), pattern of reactivity (single antibody or mixture of antibodies), characteristics of reactivity (mixed field, rouleaux), hemolysis, preservatives/antibiotics in reagents, use of washed RBCs

Initial serologic results Hematology/chemistry values Sample characteristics Other Antibody identification

fluid and flexible. The RBC membrane skeleton is associated with the lipid bilayer through specific interactions with transmembrane proteins. Specific protein components of the RBC membrane skeleton, which is associated with the inner leaflet of the lipid bilayer, interact with the cytoplasmic domains of some antigen-carrying transmembrane proteins.3–5 Two major and well-defined interactions are ankyrin, which binds to spectrin in the membrane skeleton and the cytoplasmic domain of the multipass transmembrane protein, band 3 (anion exchanger, AE1),6–14 and protein 4.1, which provides a link between spectrin, actin, and p55 in the membrane skeleton to the single-pass transmembrane proteins GPC and GPD.15–22 Some integral transmembrane proteins interact with other transmembrane proteins, forming either small or large macromolecular complexes (e.g., glycophorin A (GPA) with glycophorin B (GPB)2; band 3 with GPA23–27; Kell with Kx28,29; RhD, RhCE, RhAG,

of phospholipids in the bilayer is asymmetrical. The outer leaflet predominantly contains the neutral phospholipids (phosphatidylcholine and sphingomyelin), and the inner leaflet predominantly contains the aminophospholipids (phosphatidylethanolamine and phosphatidylserine). The presence of phosphatidylserine, which is negatively charged, on the inner monolayer results in a significant difference in charge between the two sides of the bilayer. The lipid molecules can diffuse rapidly within their own monolayer, but they rarely “flip-flop,” maintaining membrane “sidedness.”1

Proteins II 54

Peripheral proteins form a meshwork under the lipid bilayer that is called the membrane skeleton. This name implies a relatively rigid structure; however, the meshwork is actually

SINGLE-PASS PROTEINS

ABO H Le Ii

MNSs Gerbich Indian Knops

GPI-LINKED PROTEINS

NH2

COOH

NH2

P GLOB

MULTI-PASS PROTEINS

NH2 Extracellular

Lutheran LW Xg Ok

Kell

Rh* (12) Kx* (10)

Duffy (7)

Diego (14) Colton (7) Kidd (7)

Cromer Yt Dombrock JMH EMM Cytoplasmic

COOH

COOH

NH2

N-glycan

O−glycan

GPI-linkage

COOH

NH2

* Not glycosylated

(#) represents predicted number of membrane passes Figure 5–1 Diagram of a cross-section of the red blood cell membrane lipid bilayer and various membrane components that carry blood group antigens. GPI, glycosylphosphatidylinositol.

lack an antigen is exposed to RBCs that possess that antigen, the person may mount an immune response and produce an antibody that reacts with the antigen. Depending on the immunoglobulin class of the antibody and, to some extent, on the number and topology of antigens on the RBC membrane, the interaction between antibody and antigen may be detected by agglutination or hemolysis or may require an antiglobulin reagent. In human blood grouping, most tests involve agglutination, in which clumping of the RBCs serves as the detectable end point.33 The polymorphic erythrocyte marker is given a blood group antigen name after genetic and family studies.

Carbohydrates

Terminology

Carbohydrates are essentially restricted to the extracellular surface of the RBC membrane, where they collectively form a negatively charged environment that is largely responsible for keeping the RBCs from adhering to one another and to the endothelium. The majority of carbohydrates are attached to lipids on ceramide and to proteins by attachment to asparagine (N-linked) or to serine or threonine (O-linked) during passage through the lumen of the Golgi.32 Some blood group antigens are determined by the terminal carbohydrate residue (e.g., A and B) whereas others require the presence of a chain of carbohydrate residues (e.g., Leb and I). The carbohydrates form the glycocalyx, a negatively charged barrier approximately 10 Å thick around the outside of the RBC membrane. This barrier can keep immunoglobulin G (IgG) antibodies, particularly those recognizing antigens that reside close to the lipid bilayer, from readily interacting with the corresponding antigen. Thus, the glycocalyx affects the ability of an IgG antibody to cause direct agglutination.

A working party on terminology for RBC surface antigens, sanctioned by the International Society for Blood Transfusion (ISBT), has categorized the blood group antigens into four classifications34: the genetically discrete blood group systems (Table 5–3 summarizes the name, ISBT system number, chromosome location, gene name, associated antigens, component name, and possible functions for each system); blood group collections that consist of serologically, biochemically, or genetically related antigens; 700 series of low-incidence antigens; and 901 series of high-incidence antigens (Table 5–4 contains the ISBT name, number, and associated antigens. The chromosome location for these antigens has not yet been determined). As RBC antigens were discovered, notations were devised to describe them. The terminology used is inconsistent: a single letter (e.g., A, D, K), a symbol with a superscript (e.g., Fya, Jkb, Lua), or a numerical notation (e.g., Fy3, Lu4, K12) is used. Even within the same blood group system, antigens have been named with different schemes, resulting in a cumbersome terminology for describing phenotypes. The use of the same symbol with a different superscript letter (e.g., Fya and Fyb) indicates products of alleles (antithetical antigens).34 Some confusing terminology persists; for example, the P1 antigen is the sole antigen in the P blood group system, and the P antigen is the only antigen in the GLOB system. There is also a GLOB collection, with Pk and LKE antigens. In clinical practice, the traditional terminology is still extensively used; this is the terminology that will be used throughout this chapter for the discussion of red cell antigens. Collectively, there are over 260 antigens recognized by the ISBT. We restrict discussion to the more commonly encountered antibodies and antigens.

RBC BLOOD GROUP ANTIGENS Figure 5–1 depicts the membrane components that are known to carry blood group antigens. Some antigens are carbohydrates attached to lipids or to proteins, some are protein, and some require both protein and carbohydrates. Although the blood group antigens do not themselves have a function, the molecules on which they are carried do. The function of proteins carrying blood group antigens has been determined through observation of the morphology of RBCs that lack the protein (Table 5–2), direct experimentation, or comparison of the predicted protein sequence with protein databases to identify similar proteins whose function is known, even if in other tissues or organisms. Recognition of a blood group antigen begins with discovery of an antibody. When an individual whose RBCs

Inheritance Most blood group antigens are encoded by genes on autosomes.35 Most are codominant (e.g., M/N, E/e, K/k, Fya/Fyb),

Table 5–2 Red Blood Cell Morphology and Associated Phenotype Shape

Clinical Manifestation

Protein

Phenotype

Stomatocyte Elliptocyte Acanthocyte

Mild hemolytic anemia Slightly reduced red blood cell survival Elevated creatinine phosphokinase, muscular choreiform movements, neurologic defects, reduced RBC survival

Absent Rh or RhAG (Rh50) Absent GPC and GPD Absent Kx

Rh-null, Rh-mod Ge2, Ge3, Ge4 (Leach) McLeod

MEMBRANE BLOOD GROUP ANTIGENS AND ANTIBODIES

LW (ICAM4), CD47, and GPB).4,30 The recent discovery that the Rh–RhAG complex interacts with cytoskeletal ankyrin31 suggests that there are additional attachment sites between integral membrane proteins and the membrane skeleton and provides an possible explanation for the stomatocytosis associated with Rh-null erythrocytes. Many blood group antigens are carried on transmembrane proteins or glycoproteins; however, a few antigens are carried on glycosylphosphatidylinositol (GPI)-linked proteins (see Fig. 5–1). Antigens in two blood group systems (Lewis and Chido-Rodgers) are carried on proteins that are adsorbed onto RBCs from the plasma.

5 55

BLOOD BANKING

II 56

Table 5–3 Genetically Discrete Blood Group Systems Name

ISBT Chromosome Number Location

Gene Name ISBT ISGN

Associated Antigens

Component Name

Possible Function

ABO MNS

001 002

9q34.2 4q28.2–q31.1

ABO MNS

ABO GYPA, GYPB

Carbohydrate GPA; GPB

Carrier of sialic acid

Carbohydrate RhCE; RhD

Structural/Transport

ACHE

A, B, A,B, A1 M, N, S, s, U, He Mia, Vw +35 more P1 D, C, E, c, e, f, Cw and more Lua, Lub, Lu3, Lu4 and more K, k, Kpa, Kpb, Ku, Jsa and more Lea, Leb, Leab, Lebh and more Fya, Fyb, Fy3, Fy4 and more a Jk , Jkb, Jk3 Dia, Dib, Wra, Wrb, Wda and more Yta, Ytb

P Rh

003 004

22q11.2-qter 1p36.13-p34.3

P1 RH

Lutheran

005

19q13.2

LU

P1 RHD, RHCE LU

Kell

006

7q33

KEL

KEL

Lewis

007

19p13.3

LE

FUT3

Duffy Kidd Diego

008 009 010

1q22–q23 18q11-q12 17q21-q22

FY JK DI

DARC SLC14A1 SLC4A1

Yt

011

7q22

YT

Xg Scianna Dombrock

012 013 014

Xp22.32 1p35-p32 12p13.2–p12.1

XG SC DO

XG SC DO

Xga Sc1, Sc2, Sc3 Doa, Dob, Gya, Hy, Joa

Colton LandsteinerWiener ChidoRodgers Hh Kx Gerbich

015 016

7p14 19p13.3

CO LW

AQP1 LW

Coa, Cob, Co3 LWa, LWab, LWb

017

6p21.3

CH/RG

C4A, C4B

018 019 020

19q13.3 Xp21.1 2q14–q21

H XK GE

FUT1 XK GYPC

Cromer Knops

021 022

1q32 1q32

CROM KN

DAF CR1

Indian Ok Raph

023 024 025

11p13 19pter-p13.2 11p15.5

IN OK MER2

CD44 CD147 MER2

Ch1, Ch2, Rg1 and more H Kx Ge2, Ge3, Ge4, Wb and more Cra, Tca, Tcb, Tcc and more Kna, Knb, McCa, Sla, Yka Ina, Inb OKa MER2

JMH I

026 027

15p22.3–q23 6p24.2

JMH IGNT

CD108 IGNT

JMH I

Globoside

028

3q26.1

GIL

029

9p13.1

βGalN β3GALT AcT1 GIL AQP3

P GIL

Lutheran Kell glycoprotein

Enzymatic

Carbohydrate Fy glycoprotein Kidd glycoprotein Band 3

Chemokine receptor Urea transport Anion transport

Acetylcho linesterase Xga glycoprotein Sc glycoprotein Do glycoprotein CD297 Channel-forming LW glycoprotein

Enzymatic

Water transport Adhesion

C4A; C4B

Complement

Carbohydrate Kx glycoprotein GPC; GPD CD55 (DAF) CD35 (CR1) CD44 CD147 Tetraspanin (CD151) CD108 N-Acetylglucosaminyltransferase A N-Acetylgalactosaminyltransferase AQP3

Adhesion Enzymatic

Transport Interacts with protein 4.1 and p55 Complement Complement Adhesion Adhesion Adhesion Adhesion Glycocalyx Glycocalyx Glycerol/water/ urea transport

B-CAM, cell adhesion molecule; GP, glycophorin; ISBT, International Society of Blood Transfusion; ISGN, International Society for Gene Nomenclature.

Table 5–4 Other Blood Groups Whose Chromosome Locations Have Not Yet Been Determined Name Collections Cost i Er Globoside Unnamed Series Low-incidence antigens High-incidence antigens

International Society of Blood Transfusion Number

Associated Antigens

205 207 208 209 210

Csa, Csb i Era, Erb P, Pk, LKE Lec, Led

700 901

By, Chra, Bi, Bxa, Rd, Pta + 15 more Vel, Lan, Ata, Jra, JMH, Emm, AnWj, Sda, Duclos, PEL, ABTI, MAM

Expression Some blood group antigens are also found on nonerythroid cells. Some examples are A, B, H, Kna (CD35), Ina (CD44), Oka (CD147), and Cromer-related antigens (CD55), which have a wide tissue distribution.36,37 The ability to culture stem cells and sort erythroid lineages and the availability of monoclonal antibodies have enabled the estimation of the timing of expression of blood group antigens during in vitro erythroid maturation. The blood group antigens appear in the following order: GPC, Kell, RhAG, LW, RhCE, GPA, band 3, RhD, Lutheran, and Duffy.38,39

Maturation Several blood group antigens are not expressed or are only weakly expressed on fetal RBCs and do not reach adult levels until a person is approximately age 2. Cord RBCs do not express Lea, Sda, Ch, Rg, or AnWj antigens. Antibodies to these antigens are unlikely to cause HDN, because RBC expression is a prerequisite for HDN. Expression of A, B, H, P1, I, Leb, Lua, Lub, Yta, Vel, Doa, Dob, Gya, Hy, Joa, Xga, Kn, and Bg is greatly reduced on RBCs from cord blood compared to adult RBCs. In contrast, the i and LW antigens are more strongly expressed on RBCs from cord blood than on RBCs from adults. Furthermore, although adults express more LW on D-positive RBCs than on D-negative RBCs, cord RBCs express LW antigens equally regardless of D type. This information makes the testing of cord RBCs useful in antibody identification studies.

Molecular Genetic Basis Of the 29 genes (or gene families) that encode blood group antigens, 28 have been cloned and sequenced (only the gene encoding P remains to be clarified),40–44 and the genetic basis of many blood group antigens has been determined.45–48 This knowledge is being applied to transfusion medicine and is discussed later in the chapter. Relevant details about blood groups and molecular knowledge are given in Chapters 6 through 8. Information concerning the genetic basis of blood group antigens can also be obtained from the Blood Group Antigen Mutation Database: www.bioc.aecom.yu.edu/bgmut/index.php.

Evolution It has been known for many years from agglutination with human sera that blood group antigens have homologues in anthropoid apes.49 The availability of monoclonal antibodies and molecular analysis of the gene homologues from nonhuman primates have contributed to defining the epitopes of human blood group antigens.26,49–52 Additionally, because of the conserved nature of proteins, protein sequence comparisons are contributing to predictions about the function of human proteins (see Table 5–3). The sequencing of the genomes of many model organisms has allowed researchers to make test-

able inferences about protein function. The understanding of blood groups has benefited from this revolution.

Natural Knockouts The detection of an alloantibody to a high prevalence antigen during compatibility testing or prenatal testing has led to the discovery of RBCs with null phenotypes. Such RBCs lack the specific carbohydrate or carrier protein and therefore, all blood group antigens within that system. Thus, these RBCs serve as natural knockout models and these null phenotypes have provided insights into the function of the carrier proteins.

Function In general, the polymorphisms that we recognize as blood group antigens and that have significance in transfusion medicine do not appear to alter the function of the carrier molecule. The predicted functions are often based on the function of closely related proteins in other tissues; however, their role in the mature RBC may not be the same as in other cells, altered forms may function as recognition signals in senescent RBCs, or they may have an important role during earlier stages of erythroid development.5 The possible functions of the various components carrying blood group antigens can be divided into the following broad categories: contributors to membrane structural integrity, transport proteins, receptors for extracellular ligands, adhesion proteins, extracellular enzymes, complement regulatory proteins, and maintainers of surface charge in the glycocalyx (see Table 5–3).4,45,46,53–55 Specific details can be found in the chapters that follow.

BLOOD GROUP ANTIBODIES An antigen can be defined as a substance that will induce the production of antibodies, although there are certain properties of a molecule, such as foreignness, size, and chemical complexity, that are associated with increased immunogenicity. Proteins usually induce the most vigorous immune responses, followed by carbohydrates, whereas lipids and nucleic acids are usually not strong immunogens, although clinically significant antibodies specific for these types of molecules do exist (e.g., antiphospholipid antibodies and anti-DNA antibodies found in some autoimmune diseases). Most antigens require helper activity, usually in the form of cytokines, from helper T cells to induce strong antibody responses, and this helper activity is required for production of antibody classes other than IgM. Some antigens, usually carbohydrate in nature, can induce antibodies in the absence of helper T-cell activity, but these responses are primarily IgM, with little, if any, antibodies of other classes produced. Antibodies are produced by B lymphocytes, also known as B cells. The basic antibody molecule consists of four polypeptide chains—two heavy chains of 50 to 70 kilodaltons, depending on the antibody class, and two light chains of approximately 25 kilodaltons. The two heavy chains produced by a B cell are identical, as are the two light chains. In each polypeptide chain, whether it is a heavy chain or a light chain, approximately the first 100 amino acids are known as the variable region, and the remainder of the polypeptide chain constitutes the constant region, which is identical in all antibodies of the same class. The antigen-binding site is formed by association of the variable regions of one heavy

MEMBRANE BLOOD GROUP ANTIGENS AND ANTIBODIES

but some appear to be dominant if an antibody to the presumed antithetical antigen has not been discovered (e.g., Ula, Cra). Some rare phenotypes appear to be inherited in a dominant manner, for example, dominant type Lu(a−b−), or in a recessive manner, for example, recessive type Lu(a−b−). Xga and Kx are encoded by genes on the X chromosome and are inherited in a classic X-linked manner. See Table 5–3 for the chromosomal location of genes encoding blood groups.

5 57

BLOOD BANKING

chain and one light chain, which means that each four-chain unit has two identical antigen-binding sites. However, in most immune responses, a large number of B cells are stimulated by an antigen, and each antigen-specific B cell will produce a unique antibody, resulting in a heterogeneous response consisting of many different antibody molecules directed toward multiple sites on the antigen.

Immunogenicity Several factors influence the ability to stimulate antibody production, including antigen size, complexity, and dose as well as host human leukocyte antigen genotype and other, as yet unidentified, susceptibility factors. Most carbohydratebased RBC antigens are T independent and therefore tend to elicit an IgM response. The protein-based antigens usually are T dependent and induce an IgM primary response that progresses to IgG.56 Antigen exposure usually occurs by transfusion of products containing RBCs or during pregnancy (immune antibodies) or by exposure to microbes (apparently, naturally occurring antibodies). Table 5–5 summarizes the usual type of immunoglobulin response and the potential clinical significance, in transfusion or in HDN, of selected blood group antibodies, which are listed in order of clinical significance.36

Clinical Significance

II

Antibodies recognizing antigens in the ABO blood group system are by far the most clinically significant. This is because they occur naturally in people whose RBCs lack the corresponding antigen. Other clinically significant antibodies occur in the following order, from the most commonly to the least commonly encountered in transfusion practice: anti-D, antiK, anti-E, anti-c, anti-Fya, anti-C, anti-Jka, anti-S, anti-Jkb. All

other clinically significant antibodies occur with an incidence of less than 1% of immunized patients.57–59 Antibodies that are considered clinically insignificant unless the antibody reacts in tests performed strictly at 37°C are anti-P1, anti-M, antiN, anti-Lua, anti-Lea, anti-Leb, and anti-Sda. Other clinically insignificant antibodies that react at 37°C in the indirect antiglobulin test (IAT) are those of the Knops and Chido-Rodgers systems and anti-JMH (Table 5–6). The incidence of a blood group antibody depends on both the prevalence in the population and the immunogenicity of the antigen. Immunized patients frequently produce multiple antibodies, and the more antibodies present, the more difficult they are to identify.

Detection and Identification Compatibility testing (testing patient’s serum against donor’s RBCs) still uses techniques that were described 100 years ago for direct agglutination and 50 years ago for indirect agglutination. Even today, with our detailed understanding of blood group antigens, we have no single technical procedure able to detect all known blood group antibodies. The hemagglutination technique is simple and inexpensive, does not require sophisticated equipment, and when done correctly is sensitive and specific in terms of clinical relevance. Agglutination should be graded according to the strength of reaction, and an evaluation of the reaction strength can aid in identification of antibodies, especially when multiple antibodies are present in a serum.33 The first blood group antigens to be identified were those that could be agglutinated by the alloantibodies when antigen-positive RBCs were suspended in a saline medium (direct agglutination). This direct agglutination reflects the fact that these antibodies are usually IgM and detect carbohydrate antigens (ABO, P1, Le, and H antigens). Although

58

Table 5–5 Characteristics of Some Blood Group Alloantibodies Clinical Antibody Specificity

IgM (Direct)

IgG (Indirect)

Transfusion Reaction

HDN

ABO

Most

Some

Immediate; mild to severe

Rh Kell Kidd Duffy M N S s U P1 Lutheran Lea Leb Diego Colton Dombrock LW Yta I Chido-Rodgers JMH Knops Xga

Some Some Few Rare Some Most Some Rare Rare Most Some Most Most Some Rare Rare Rare Rare Most Rare Rare Rare Rare

Most Most Most Most Most Rare Most Most Most Rare Most Few Few Most Most Most Most Most Rare Most Most Most Most

Immediate/delayed; mild to severe Immediate/delayed; mild to severe Immediate/delayed; mild to severe Immediate/delayed; mild to severe Delayed (rare) None Delayed; mild Delayed; mild Immediate/delayed; mild to severe None (rare) Delayed Immediate (rare) None Delayed; none to severe Delayed; mild Immediate/delayed; mild to severe Delayed; none to mild Delayed (rare); mild None Anaphylactic (several cases) Delayed (rare) None None

Common; mild to moderate Common; mild to severe Sometimes; mild to severe Rare; mild Rare; mild Rare; mild None Rare; mild to severe Rare; mild to severe Rare; severe None Rare; mild None None Mild to severe Rare; mild to severe Rare; mild Rare; mild None None None None None None

Clinically Significant?

Alloantibodies

Always

A and B H in Oh Rh Kell Duffy Kidd Diego S, s, U P, PP1, Pk Vel Ata Colton Cromer Dombrock Gerbich Indian Jra JMH Lan LW Scianna Yt A1 H Lea Lutheran M, N P1 Sda Chido-Rodgers Cost Knops Leb Xga

Sometimes

Not unless reactive at 37°C

Not usually

Locating Antigen-Negative Blood

anti-A and anti-B are highly clinically significant, antibodies to the other carbohydrate antigens generally are not. Most of the other antigens (e.g., Rh, Kell, Duffy, and Kidd) are proteins and are detected by antibodies that react in the IAT. This test detects IgG antibodies, complement attached to RBCs, or both. The sensitized RBCs are washed to remove unattached immunoglobulin (which would inhibit the antiglobulin reagent), and anti-human immunoglobulin is added. The specimen is centrifuged and examined for agglutination. The antiglobulin test can be direct or indirect. The direct antiglobulin test is used to detect RBCs sensitized in vivo, such as alloantibodies causing transfusion reactions or

Once a patient is actively immunized to an RBC antigen and produces a clinically significant alloantibody, the patient is considered immunized for life and must be transfused with antigen-negative RBCs, even if the antibody is no longer detectable. Patients with passively acquired antibody (e.g., neonates and recipients of plasma products, Rh immune globulin, or IVIgG) are not actively immunized and may be transfused with antigen-negative RBCs only while the passive antibody is still present. Selection of blood for transfusion of a patient with blood group alloantibodies is the joint responsibility of the staff at the transfusion service, the donor center, and the patient’s physician. Thus, it is very important that there be communication regarding the number of units of RBC products required and the time frame involved. Figure 5–2 is a flow chart that outlines the process for locating blood. Table 5–8 lists the prevalence of donors whose RBCs lack selected antigens. To locate antigen-negative blood for transfusion, it is not necessary to identify an antibody to a low-prevalence antigen detected in the recipient’s blood during compatibility testing, because another unit of donor blood is highly unlikely to be positive for the same uncommon antigen. In contrast, if an antibody to a low-prevalence antigen is detected in the serum of a pregnant woman, identification of the antibody or determination of its reactivity with paternal RBCs is required to predict the likelihood and severity of HDN in the baby. Locating blood for exchange transfusion is not difficult, for the reason previously described. If a patient’s serum contains alloantibodies to a highprevalence antigen, blood for transfusion may be very difficult to locate. Whether the investigation is for transfusion purposes or for prediction of HDN, the antibody should be identified. The identification aids in both the assessment of its clinical significance and the location of appropriate blood for transfusion. Family members, in particular siblings, are

Table 5–7 Reactivity of Antigen-Positive Red Blood Cells after Treatment with Ficin/Papain or DTT Ficin/Papain

Dithiothreitol (200 mM)

Possible, Antibody Specificity

Negative Negative Positive Variable Positive Positive

Positive Negative Weak Negative Negative Positive

Positive

Enhanced

M,N,S,s†; Fya,Fyb; Ge2,Ge4; Xga; Ch/Rg Indian; JMH Cromer; Knops (weak in ficin), Lutheran; Dombrock; AnWj; MER2 Yta Kell; LW; Scianna A, B; H; P1; Rh; Lewis; Kidd; Fy3; Diego; Co; Ge3; Oka, I, i; P, LKE; Ata; Csa; Emm; Era; Jra; Lan; Vel, Sda, PEL Kx

†

variable with ficin/papain.

MEMBRANE BLOOD GROUP ANTIGENS AND ANTIBODIES

HDN and autoantibodies in autoimmune hemolytic anemia or cold agglutinin disease. The IAT is used to detect RBCs sensitized in vitro; for instance, antigen typing, antibody detection and identification, and compatibility testing. To identify antibodies, serum is tested against RBCs of known phenotype by various techniques. It is sometimes helpful to treat antigen-positive test RBCs with proteolytic enzymes and chemical agents and to compare the antibody reactivity in tests against treated and untreated RBCs to aid antibody identification (Table 5–7).47 Brief, but informative, descriptions of the technical and clinical aspects of most blood group antibodies can be found elsewhere.33,60

Table 5–6 Clinical Significances of Some Alloantibodies to Blood Group Antigens

5 59

BLOOD BANKING

Patient has clinically significant antibody?

No

Yes

Use routine crossmatch procedure

Calculate incidence Multiply negative frequency for each antigen ⫻ 100 Example: 0.91 (K) ⫻ 0.48 (S) ⫽ 0.436 ⫻ 100 ⫽ 44% Calculate number of antigen-negative donors: 100 ⫼ 44 ⫽ 1 in 3 units will lack K and S Antigen negative available from hospital inventory?

Yes

No Screen inventory at donor center Test siblings/family Call in phenotyped donors Check frozen inventory Examine rare donor file (last resort) Use autologous

Crossmatch antigen-negative units Figure 5–2 Flow chart for locating antigen-negative blood.

II 60

the first source to investigate for antigen-negative units. Most national blood donor centers can often assess national and international Rare Donor Registries and can help provide the appropriate antigen-negative blood.

COMPATIBILITY PROCEDURES Most laboratories follow recommendations made by the American Association of Blood Banks,61 and all must comply with regulations of state and federal agencies. Routine approaches to compatibility testing involve testing a blood sample from a prospective recipient for ABO and D blood groups and for the presence of blood group antibodies. In most cases, no unexpected antibodies are detected, and donor RBCs of appropriate ABO and D blood type are selected for transfusion. A sample of the donor’s RBCs may be tested (crossmatched) with the patient’s serum by means of either an immediate spin procedure or the IAT. Alternatively, the blood can be issued without direct crosstesting if a computer check is performed.33 Any blood typing problems should be resolved, and when antibodies are detected, they should be identified. Knowing the specificity of an antibody helps establish whether it is likely to be clinically significant and determine the approach necessary to provide an adequate supply of compatible blood. Once an antibody has been identified, antigen-negative RBC products selected for transfusion should be tested with the patient’s serum by the IAT to ensure compatibility.

It is important not to delay surgery or transfusion unnecessarily by attempting to obtain antigen-negative RBCs for patients with clinically insignificant antibodies. Also, in hemolytic anemia due to warm-reactive autoantibodies, compatibility may be difficult to demonstrate. The important issue is to be sure that there are no underlying, clinically significant alloantibodies. This fact can be determined with autologous or homologous autoabsorptions,33 which require extra time and, possibly, the services of an immunohematology reference laboratory. It is helpful to remember that patients without a history of immunization (transfusion or pregnancy) are unlikely to have underlying alloantibodies. Unfortunately, a patient’s transfusion history is not always reliable, because many patients are unaware or forget that they have received transfusions.62

CHOICE OF ANTIGEN-NEGATIVE BLOOD FOR DISEASES REQUIRING LONG-TERM TRANSFUSION THERAPY Transfusion management of patients who require long-term transfusion therapy, in particular patients with sickle cell anemia, has been the subject of controversy and debate.63,64 There is still no consensus as to the best and most practical approach, although the goal is to provide blood with maximal survival. In addition to the traditional practice of providing antigen-negative blood only after the patient has made an antibody, approaches that have been adopted

Incidence of Antigen Negativity System

Antigen

Rh

D C E C E F CW V VS CE cE Ce M N S S M,S M,s N,S N,s P1 Lea Leb Lua Lub K K Kpa Kpb Jsa Jsb Fya Fyb Jka Jkb Doa Dob Coa Cob

MNS

P Lewis Lutheran Kell

Duffy Kidd Dombrock Colton

White

approach. Indeed, in some regions, it is a challenge to provide antigen-negative blood to patients who have already made the antibodies. Extensive screening for antigen-negative donors may be realized with the advent of microarray technology.

Black

0.15 0.32 0.71 0.20 0.02 0.35 0.98 >0.99 >0.99 25% estimated blood volume Platelets Blood platelets < 50 to < 100×109/L and significant bleeding Blood platelets < 50 × 109/L and invasive procedure Blood platelets < 20 × 109/L prophylaxis and clinically stable Blood platelets < 50 to < 100 × 109/L prophylaxis and clinically unstable Neutrophils Blood neutrophils < 3 × 109/L and fulminant sepsis during first week of life Blood neutrophils < 1 × 109/L and fulminant sepsis after first week of life *

Words in italics must be defined according to local practices. Hct, hematocrit.

Another physiologic factor to be considered in the transfusion decision is the use of circulating RBC volume rather than blood Hct or Hb level.11,34,37 Although circulating RBC volume is a potentially useful index of the blood’s oxygen-carrying capacity, it cannot be predicted accurately by measurement of the Hct in infants.38 Low circulating RBC volume identifies, better than Hb or Hct, those infants who will respond to transfusion with a decrease in cardiac output.39 At present, circulating RBC volume measurements are not widely available. However, promising techniques using nonradioactive biotin to tag RBCs have been adapted for infant studies. Strauss and coworkers were able to demonstrate with this technique that the post-transfusion recovery and in vivo survival of donor RBCs, stored for up to 42 days, confirmed earlier studies that defined the efficacy and safety of stored allogeneic RBCs for small-volume transfusions to neonates. Specifically, no significant differences were observed between allogeneic RBCs stored on days 1 to 21 compared to days 22 to 42.39–41

TRANSFUSION OF NEONATES AND PEDIATRIC PATIENTS

that a low Hct contributes to tachypnea, dyspnea, apnea, and tachycardia or bradycardia because of decreased oxygen delivery to the respiratory center of the brain. If this theory is true, transfusions of RBCs should decrease the number of apneic spells by improving oxygen delivery to the central nervous system. However, results of clinical studies have been contradictory.7,29 A recent study by Bell and colleagues provides evidence to the contrary.31 In this study, 100 preterm infants, weighing from 500 to 1300 grams, were randomized to either a liberal or restrictive RBC transfusion regimen (i.e., a relatively low or high pretransfusion Hct level, respectively). The investigators found that the mean number of RBC transfusions was higher in the liberal group than the restrictive group; however, because a “single donor/unit program” was used, the number of donor exposures was not statistically significantly different. Furthermore, the most surprising finding was that those infants in the more restrictive group were more likely to have severe grades of intraparenchymal brain hemorrhage or periventricular leukomalacia as well as having increased frequency of apnea. The authors suggest that more restrictive transfusion practices may be detrimental to preterm infants and hypothesize that decreased oxygen delivery to brain tissue may be the pathophysiology of the brain injury and increased frequency of apnea—a speculation supported by another study in which high cerebral fractional oxygen extraction was found in infants with hemorrhagic parenchymal infarction as a manifestation of possible low cerebral blood flow.31,32 Very importantly, a multicenter study of similar design with only preliminary reports to date33 found results contradictory to Bell and colleagues. In the randomized study, 451 preterm infants were transfused according to either relatively low or high pretransfusion blood Hb concentrations. In contrast to the findings of Bell and colleagues, no difference in morbidity and mortality were found—favoring acceptance of relatively low pretransfusion blood Hb or Hct levels in clinical practice. Because the experimental design and outcomes were not identical in the two studies, there is no definitive recommendation possible for transfusion practices at this time. However, the potential for harm due to “undertransfusion,” if very low pretransfusion Hb or Hct levels are accepted, must be considered when transfusion decisions are made. Generally in practice, the decision whether to transfuse RBCs is based on the desire to maintain the Hct at a level judged to be most beneficial for the infant’s clinical condition. Investigators who believe that this “clinical” approach is too imprecise have suggested the use of “physiologic” criteria for transfusions, such as RBC mass,34 available oxygen,35 mixed venous oxygen saturation, or measures of oxygen delivery and utilization,36 to develop guidelines for transfusion decisions. In one study of 10 infants with severe (oxygendependent) bronchopulmonary dysplasia, improvement in physiologic end points (increased systemic oxygen transport and decreased oxygen use) was shown to be a consequence of small-volume RBC transfusions.36 However, these promising but technically demanding methods are, at present, difficult to apply in the day-to-day practice of neonatology, and studies conducted directly in infants are needed. Application of data obtained from studies of animals and adult humans that correlate tissue oxygenation with the clinical effects of anemia and the need for RBC transfusions is confounded by the differences between infants and adults in Hb oxygen affinity, ability to increase cardiac output, and regional patterns of blood flow.

Selecting an RBC Product to Transfuse an Infant The RBC products usually chosen for small-volume transfusions given to infants are RBCs suspended either in citrate-phosphate-dextrose-adenine solution (CPDA) or in extended storage media (AS-1, AS-3, AS-5) at a Hct ranging from 55% to 70%. Some centers prefer to centrifuge RBC aliquots before transfusion, to prepare packed RBCs at a Hct of 80% to 90%. Most RBC transfusions are infused slowly over 2 to 4 hours at a dose of about 15 mL/kg body weight. Because of the small quantity of RBC preservative fluid infused and the slow rate of transfusion, the type of anticoagulant/preservative medium selected is believed not to pose risks for the majority of premature infants given smallvolume transfusions.42 Accordingly, the traditional use of relatively fresh RBCs (25 mL/kg), in which larger doses of K+ may be harmful, especially if the infusion is rapid. Cardiac surgeons and other pediatric surgeons who may transfuse large RBC volumes expectedly or unexpectedly need to be especially cognizant of this point. As for the second objection, 2,3-diphosphoglycerate is totally depleted from RBCs by 21 days of storage, and

37 513

TRANSFUSION MEDICINE

Table 37–3

Formulation of Anticoagulant-Preservative Solutions Present in Blood Collection Sets

Constituent

CPDA

AS-1

AS-3

AS-5

Volume (mL) Sodium chloride (mg) Dextrose (mg) Adenine (mg) Mannitol (mg) Trisodium citrate (mg) Citric acid (mg) Sodium phosphate (monobasic) (mg)

63* None 2000 17.3 None 1660 206 140

100† 900 2200 27 750 None None None

100† 410 1100 30 None 588 42 276

100† 877 900 30 525 None None None

* Approximately 450 mL of donor blood is drawn into 63 mL of CPDA. A unit of red blood cells (hematocrit, approximately 70%) is prepared by centrifugation and removal of most plasma. † When AS-1 or AS-5 is used, 450 mL of donor blood is first drawn into 63 mL of CPD, which is identical to CPDA except that it contains 1610 mg of dextrose per 63 mL and has no adenine. When AS-3 is used, donor blood is drawn into CP2D, which is identical to CPD except that it contains double the amount of dextrose. After centrifugation and removal of almost all plasma, red blood cells are resuspended in 100 mL of the additive solution (AS-1, AS-3, or AS-5) at a hematocrit of approximately 55% to 60%. AS-1, AS-3, and AS-5, extended storage media; CPDA, citrate-phosphate-dextrose-adenine solution. Data from Luban NLC, Strauss RG, Hume HA. Commentary on the safety of red blood cells preserved in extended storage media for neonatal transfusions. Transfusion 1990;30:229.

III 514

this is reflected by a decrease in the oxygen half-saturation pressure (P50) from about 27 mmHg in fresh blood to 18mmHg at the time of outdate. The last value of older transfused RBCs corresponds to the “physiologic” P50 obtained from the RBCs of many normal preterm infants at birth, reflecting the relatively high affinity for oxygen normally exhibited by infant RBCs. Therefore, the P50 of older transfused RBCs is no worse than that of RBCs produced endogenously by the infant’s own bone marrow. Moreover, these older adult RBCs provide a benefit to the infant because the 2,3-diphosphoglycerate and the P50 of transfused RBCs (but not of endogenous infant RBCs) increase rapidly after transfusion. Regarding the third objection, the quantity of additives present in RBCs stored in extended-storage media is believed not to be dangerous to neonates given smallvolume transfusions (≤15 mL/kg).42 A comparison of CPDA with three types of extended-storage media is presented in Table 37–3. Regardless of the type of suspending solution, the quantity of additives is quite small in the clinical setting, in which infants are given small-volume transfusions of RBCs transfused over 2 to 4 hours, and it is far lower than doses believed to be toxic (Table 37–4). Importantly, the efficacy and safety of these theoretical calculations have been confirmed by clinical experience. In addition, many investigators have reported the successful transfusion of stored, rather than fresh, RBCs for small-volume transfusions in infants.43–46 The small-volume nature of these

Table 37–4

transfusions cannot be overemphasized. Therefore, these same rules do not apply to massive transfusion situations, which many pediatric surgeons and neonatologists may encounter knowingly or unknowingly and for which they need to be prepared accordingly. This information is especially important for surgeons and neonatologists who are recommending that the parents, relatives, or friends of their patients perform directed RBC donation. Communication between the physicians and blood bank is encouraged for many reasons: to confirm the child’s blood type so typecompatible RBCs can be donated; to help with the collection center’s choice of anticoagulant-preservative solution into which the units will be drawn (i.e., CPDA and/or AS units), which might differ if a massive transfusion situation seems likely; and to coordinate timing of the surgery so that tested units are in hospital inventory prior to the surgical date and time. Recombinant Erythropoietin in the Anemia of Prematurity Recognition of the low plasma EPO levels in preterm infants provides a rational basis for the use of recombinant human EPO as therapy for the anemia of prematurity. More than 20 controlled trials have tested several doses and treatment schedules in preterm infants, and results are mixed, making consensus impossible on the optimal role of recombinant human EPO treatment in the anemia of prematurity.7,47,48

Constituents Infused (mg/kg) in 10 mL/kg Red Blood Cells (Hematocrit, 60%)

Additive

CPDA

AS-1

AS-3

Toxic Dose*

NaCl Dextrose Adenine Citrate Phosphate Mannitol

0 13 0.2 12 9 0

28 86 0.4 6.5 1.3 22

5 15 0.4 8.4 3.7 0

137 mg/kg/day 240 mg/kg/hr 15 mg/kg/dose 180 mg/kg/hr >60 mg/kg/day 360 mg/kg/day

* Actual toxic dose is difficult to predict accurately because the infusion rate usually is slow, permitting metabolism and distribution from blood into extravascular sites, and dextrose, adenine, and phosphate enter red blood cells and are somewhat sequestered. Potential toxic doses are based on Luban, Strauss, and Hume (1990).32 AS-1 and AS-3, extended storage media; CPDA, citrate-phosphate-dextrose-adenine solution. Data from Luban NLC, Strauss RG, Hume HA. Commentary on the safety of red blood cells preserved in extended storage media for neonatal transfusions. Transfusion 1990;30:229.

PLATELET TRANSFUSIONS Older Children and Adolescents Guidelines for platelet support of children and adolescents with quantitative and qualitative platelet disorders are similar to those for adults (see Table 37–1), in whom the risk of life-threatening bleeding that occurs after injury or spontaneously can be related to the severity of thrombocytopenia (when low blood platelet counts are caused by diminished marrow production). Thrombocytopenia caused by accelerated turnover (e.g., immune thrombocytopenia) usually is not treated with platelet transfusions. Platelet transfusions

should be given to patients with platelet counts lower than 50 × 109/L due to marrow failure if they are bleeding or are scheduled for an invasive procedure. Studies of patients with thrombocytopenia caused by poor marrow production indicate that spontaneous bleeding increases markedly when platelet levels fall to less than 20 × 109/L, particularly in patients who are ill with infection, anemia, or dysfunction of the liver, kidneys, or lungs. For this reason, many pediatricians recommend prophylactic platelet transfusions to maintain the platelet count at greater than 20 ×109/L in children with thrombocytopenia due to bone marrow failure. This threshold has been challenged, and some favor a platelet transfusion trigger of 5 to 10 × 109/L for patients with uncomplicated conditions. Many pediatric hematology/oncology and stem cell transplantation physicians have lowered their prophylactic platelet transfusion trigger to 10 × 109/L, extrapolating from the adult acute myeloid leukemia studies by Rebulla and colleagues in 1997 and Wandt and coworkers in 1998.53,54 These prospective clinical trials, among others, demonstrated that nonbleeding stable thrombocytopenic patients without coexisting symptoms can be managed safely with a more restrictive platelet transfusion trigger, 10 × 109/L, and fewer platelet transfusions were given. However, those with fever, active bleeding, and/or a coagulation disorder had a higher trigger of at least 20 × 109/L. Qualitative platelet disorders may be inherited or acquired (e.g., in advanced hepatic or renal insufficiency, after cardiopulmonary bypass procedures). In such patients, platelet transfusions are justified only if significant bleeding occurs. Because individuals with platelet dysfunction may have intermittent bleeding episodes throughout their life necessitating repeated transfusions, which may lead to alloimmunization and refractoriness, prophylactic platelet transfusions are rarely justified unless an invasive procedure is planned. In such cases, a bleeding time greater than twice the upper limit of laboratory normal may be taken as diagnostic evidence that platelet dysfunction exists, but this test is poorly predictive of hemorrhagic risk or the need to transfuse platelets. The bleeding time has been supplanted by other platelet function assays and is rarely performed anymore, especially in children. In these patients, alternative therapies, particularly subcutaneous or intranasal desmopressin acetate, should be considered to avoid platelet transfusions.

Infants and Younger Children Pathophysiology of Neonatal Thrombocytopenia Blood platelet counts of 150 × 109/L or greater are present in normal fetuses (=17 weeks of gestation) and neonates. Lower platelet counts indicate potential problems, and preterm infants exhibit thrombocytopenia commonly.55,56 In one neonatal intensive care unit, 22% of infants had blood platelet counts lower than 150 × 109/L during hospitalization.55 Although multiple pathogenic mechanisms probably are involved in these sick neonates, a predominant one is accelerated platelet destruction, as shown by shortened platelet survival time, increased platelet-associated immunoglobulin G, increased platelet volume, a normal number of megakaryocytes, and an inadequate increment in blood platelet values after platelet transfusion.55,57 Another major mechanism contributing to neonatal thrombocytopenia is diminished platelet production, as evidenced by decreased

TRANSFUSION OF NEONATES AND PEDIATRIC PATIENTS

Unquestionably, proper doses of recombinant human EPO and iron effectively stimulate erythropoiesis in preterm infants, as evidenced by increased marrow erythroid activity and blood reticulocyte counts. However, the efficacy of recombinant human EPO in substantially diminishing the number of RBC transfusions—the major goal for which it is prescribed—has not been convincingly demonstrated for all groups of preterm infants.48 In many trials, the subjects were relatively large preterm infants and those in stable clinical condition; such infants currently receive few RBC transfusions when given only standard supportive care (i.e., not given recombinant human EPO).27,49 Currently, even without use of recombinant human EPO, fewer than 50% of infants with birth weights greater than 1.0 kg receive RBC transfusions. Almost all infants weighing less than 1.0 kg at birth are given RBCs, and most transfusions are given during the first 3 to 4 weeks of life.49 To illustrate the difficulty of avoiding RBC transfusions, a multicenter, randomized North American trial, in which infants received either recombinant human EPO or placebo during a 6-week study period, reported a statistically significant difference but only modest success.50 Although significantly fewer RBC transfusions were given to infants treated with recombinant human EPO during the study phase (1.1 transfusion vs 1.6 for placebo), all infants required multiple transfusions during the 3-week prestudy phase. Therefore, recombinant human EPO exerted only a modest effect on total RBC transfusions given throughout the entire study (4.4 for recombinant human EPO vs 5.3 for placebo) and did not resolve the problem of severe neonatal anemia.50 Physicians wishing to prescribe recombinant human EPO are faced with a dilemma. The relatively large or stable preterm infants who respond best to recombinant human EPO plus iron are given relatively few RBC transfusions and, accordingly, have little need for recombinant human EPO to avoid transfusions. Extremely preterm infants, who are sick and have the greatest need for RBC transfusions shortly after birth, have not consistently responded to recombinant human EPO plus iron, again questioning the efficacy of recombinant human EPO therapy.47 However, extremely preterm infants are being evaluated in therapeutic trials of recombinant human EPO and iron, both given intravenously shortly after birth.51,52 Although preliminary review of these promising studies suggests success in avoiding transfusions early in life, the data are limited and are insufficient to clearly establish efficacy or to detect potential toxicity. Therefore, firm guidelines for the use of recombinant human EPO in the treatment of the anemia of prematurity cannot be offered at this time.

37 515

TRANSFUSION MEDICINE

numbers of clonogenic megakaryocyte progenitors58,59 and relatively low levels of thrombopoietin59,60 in response to thrombocytopenia, when compared with children and adults. Similar to the situation with EPO and the anemia of prematurity, thrombopoietin is produced by thrombocytopenic preterm infants, but at relatively low levels. Controlled clinical trials are needed to determine the possible role and potential toxicity of recombinant thrombopoietin therapy in infants. Blood platelet counts lower than 100 × 109/L pose significant clinical risks for premature neonates. In one study, infants with birth weights lower than 1.5 kg and blood platelet counts lower than 100 × 109/L were compared with nonthrombocytopenic control infants of similar size.56 The bleeding time was prolonged when platelet counts were lower than 100 × 109/L, and in many infants platelet dysfunction was suggested by bleeding times that were disproportionately long for the degree of thrombocytopenia present. Hemorrhage was more frequent in the thrombocytopenic infants, with the incidence of intracranial hemorrhage being 78% in those weighing less than 1.5 kg at birth, compared with 48% for nonthrombocytopenic infants of similar size. Moreover, the extent of hemorrhage and neurologic morbidity was greater in the group of thrombocytopenic infants.56 Recommendations for Platelet Transfusions during Infancy

III 516

The use of prophylactic platelet transfusions in an attempt to prevent bleeding in preterm neonates has been studied systematically.61 However, no randomized clinical trials have been reported examining therapeutic platelet transfusions in bleeding thrombocytopenic neonates. Therefore, basic questions regarding the relative risks of different degrees of thrombocytopenia in various clinical settings during infancy remain largely unanswered. However, it seems logical to transfuse platelets to thrombocytopenic infants, and guidelines acceptable to many neonatologists are listed in Table 37–2. Two firm indications for neonatal platelet transfusions are to treat hemorrhage that has already occurred and to prevent hemorrhage from complicating an invasive procedure. Little disagreement exists regarding the use of a blood platelet count lower than 50 × 109/L as a “transfusion trigger” in these instances. However, platelet transfusions are given to infants by some physicians to treat bleeding that occurs at higher platelet counts (between 50 and 100 × 109/L) or to diminish the threat of intracranial hemorrhage in high-risk preterm infants whenever the platelet count is lower than 100 × 109/L.56 Prophylactic platelet transfusions can be given under two circumstances: to prevent bleeding when severe thrombocytopenia is present and poses a risk of spontaneous hemorrhage and to maintain the presence of a normal platelet count to prevent the infant from slipping into high-risk situations. Regarding the first circumstance, most agree that it is reasonable to give platelets to any neonate whose blood platelet count is lower than 20 × 109/L. There is broad acceptance that spontaneous hemorrhage is a risk with platelet counts below this level. Also, severe thrombocytopenia occurs most commonly in sick infants who, because of their illnesses, receive medications that may further compromise platelet function. Because all of these factors are pronounced in extremely preterm infants, some neonatologists favor prophylactic platelet transfusion whenever the platelet count falls to less than 50 × 109/L, or even 100 × 109/L, in critically ill infants.56

Regarding the second circumstance, the need to maintain a completely normal platelet count (≥150 × 109/L) or even higher in preterm infants without bleeding is unproven. Intracranial hemorrhage occurs commonly in sick preterm infants, and, although the etiologic role of thrombocytopenia and the therapeutic benefit of platelet transfusions have not been conclusively established in this disorder, it seems logical to presume that thrombocytopenia is a risk factor.62 However, a randomized trial designed to address this issue—in which transfusion of platelets whenever the platelet count fell to less than 150 × 109/L so as to maintain the average platelet count at greater than 200 × 109/L was compared with transfusion of platelets only when the platelet count fell to less than 50 × 109/L—did not detect a difference in the incidence of intacranial hemorrhage (28% vs 26%, respectively).61 Therefore, there is no documented benefit to transfusing “prophylactic platelets” to maintain a completely normal platelet count, compared with transfusing “therapeutic platelets” in response to thrombocytopenia when it actually occurs. Currently, there are no alternatives to platelet transfusions to treat thrombocytopenia in neonates. Recombinant thrombopoietin (i.e., c-mpl ligand or megakaryocyte growth and differentiation factor) and interleukin-11 are promising agents. However, neither is recommended for use during infancy, and both have potential toxicities that might preclude their use in sick preterm infants. Clearly, they must not be prescribed at present, except in experimental settings. Selecting a Platelet Product to Transfuse an Infant or Younger Child The ideal goal of most platelet transfusions is to raise the platelet count to greater than 50 × 109/L or, for sick preterm infants, to greater than 100 × 109/L. This goal can be achieved by the infusion of 5 to 10 mL/kg of standard (i.e., unmodified) platelet concentrates, collected by centrifugation of fresh units of whole blood or from a plateletpheresis unit collected by automated plateletpheresis. The platelet dose should be transfused as rapidly as the overall condition permits, certainly within 2 hours. Routinely reducing the volume of platelet concentrates for infants by additional centrifugation steps is both unnecessary and unwise. Transfusion of 10 mL/kg platelet concentrate provides approximately 10 × 109/L platelets. Assuming that the estimated blood volume of an infant is 70 mL/kg body weight, the platelet dose of 10 mL/kg will increase the platelet count by 143 × 109/L. This calculated increment is consistent with the observed increment after this dose reported in clinical studies.61 In general, 10 mL/kg is not an excessive transfusion volume, provided that the intake of other intravenous fluids, medications, and nutrients is monitored and adjusted. It is desirable that the infant and the platelet donor be of the same ABO blood group, and it is important to minimize repeated transfusions of group O platelets to group A or B recipients, because large quantities of passive anti-A or antiB can lead to hemolysis, resulting in severe morbidity and mortality in children.63–66 Although proven methods exist to reduce the volume of platelet concentrates when truly warranted (i.e., when many transfusions are anticipated in which multiple doses of passive anti-A or anti-B might lead to hemolysis or when there is failure to respond to 10 mL/kg of unmodified platelet concentrate), additional processing (i.e., plasma volume reduction) should be performed with great care because of

NEUTROPHIL TRANSFUSIONS Older Children and Adolescents Several methodologic advances—in particular, the use of recombinant granulocyte colony-stimulating factor (GCSF) to stimulate donors—have made it possible to collect extraordinarily large numbers of normal neutrophils (polymorphonuclear neutrophils, PMNs) for transfusion into neutropenic patients who have life-threatening infections. Because larger doses of PMNs can be transfused, renewed interest has arisen in the use of PMN (granulocyte) transfusions to treat adult oncology patients and hematopoietic progenitor cell transplant recipients, in whom neutropenia complicated by severe infections persists as a significant problem despite combination antibiotic therapy, recombinant cytokines, myeloid growth factors, and use of mobilized peripheral blood progenitor cells to minimize neutropenic infections. If children are suffering significant morbidity and mortality from neutropenic infections despite modern supportive care, it is logical to explore the efficacy, potential toxicity, and cost effectiveness of tranulocyte transfusion therapy through properly designed, randomized clinical trials performed in pediatric subjects.67 Serious and repeated infections with bacteria, yeast, and fungi are a consequence of severe neutropenia and PMN dysfunction in some settings. In the multicenter Trial to Reduce Alloimmunization to Platelets (TRAP) study, 7% of adult patients with acute nonlymphocytic leukemia died from infection during first-remission induction therapy, despite the use of modern antibiotic therapy.68 In another study of patients given intensive chemotherapy, some of whom also received transfusions of autologous hematopoietic stem cells, 7.6% of patients experienced systemic fungal infection.69 Unless severe neutropenia is reversed fairly quickly in adult patients, the mortality of systemic fungal infections approaches 100%. Therefore, modern “high-dose” granulocyte transfusion therapy is considered by some experts to be very promising for adult oncology and transplantation patients.70,71 However, contrary data suggest that, with appropriate anti-infective therapy, serious infections are rare in patients who are transplanted with adequate numbers of peripheral blood progenitor cells.72 Because of these controversial views, pediatricians must survey the outcome of life-threatening infections with bacteria, yeast, and fungi in children who are undergoing intense chemotherapy or hematopoietic progenitor cell transplantation in their own institutions to determine whether there is a need for therapeutic granulocyte transfusion. If infections in neutropenic children respond promptly to antibiotics plus standard supportive care and survival approaches 100%, granulocyte transfusion is unnecessary. Moreover, it should not be used if there is no apparent need, because the lack of demonstrable benefit would not outweigh the potential risks. However, if significant numbers of infected high-risk patients fail to respond to antibiotics alone, or if the intensity of therapy is compromised because it is limited by fear of neutropenia, the addition of granulocyte transfusion should be considered, along with other modifications of therapy intended to reduce infections, such as selection of different

antibiotics, closer monitoring of antibiotic blood levels, and use of intravenous immunoglobulin (IVIG), G-CSF, other recombinant cytokines, and immune-modulating agents. The role of granulocyte transfusion added to antibiotics for patients with severe neutropenia (50%) homologous at the nucleic acid sequence level, and they crossreact immunologically to a great extent (particularly the core and polymerase antigens). For this reason, up to 90% of sera from HIV-2– infected persons have been found to test positive with FDAlicensed anti-HIV-1 assays, with variable reactivity on HIV-1 Western blot analysis.29,30 This crossreactivity undoubtedly prevented transfusion transmission of HIV-2 by blood screened for anti-HIV-1 in areas where the type 2 virus was present before implementation of combination HIV-1/ HIV-2 screening tests. This high-level crossreactivity has also facilitated surveillance for HIV-2 in regions where it was rare or absent.31,32 From 1987, when the first case of HIV-2 infection was diagnosed in the United States, to 1998, a total of 79 persons with HIV-2 infections were documented, of which approximately two thirds had been born in West Africa.33 Combination HIV-l/HIV-2 assays were developed in the late 1980s,34–36 and mandatory implementation of either a combination test or a separate anti-HIV-2 test in the donor screening setting was required by the FDA effective June 1, 1992.37 From the implementation of HIV-2 screening in 1992 through 1996, three prospective U.S. blood donors were found to be HIV-2 positive at the time of attempted donation.38 One was a U.S.-born woman without identifiable risks for HIV infection; the other two were men, born in France and Liberia, West Africa respectively, who had resided for years in West Africa. These three cases of HIV-2 infection were detected from screening of more than 50 million whole blood donations, indicating that the prevalence of HIV-2 is very low (less than 1 in 15 million screened donations). A more recent study identified a single HIV-2–infected donor among 7.2 million donations at 18 U.S. blood centers from 1997 through mid2000.18 As of 2001, no cases of HIV-2–infected transfusion recipients had been reported in the United States.

Clinical HIV-1 Disease In approximately 60% of acute HIV infections newly infected persons experience a nonspecific flulike illness, usually within 2 to 4 weeks of exposure.39 This retroviral syndrome coincides with appearance of antibody to HIV antigens and is thought to represent a reaction of the immune system to the rapidly

proliferating virus. In most cases signs and symptoms of acute HIV disease subside within weeks to a few months, giving way to a clinically silent period that can last several years. The lack of clinical symptoms during this stage of the disease is deceptive, however, because viremia persists in untreated infection and the number of CD4+ lymphocytes, the primary target of HIV, is gradually declining.40 Paralleling the loss of CD4+ lymphocytes are clinical disease manifestations, such as opportunistic infections, numerous conditions related to suppression and dysfunction of the immune system, direct viral effects on multiple organ systems, and ultimately death after a median survival of 8 to 10 years from exposure. Although it appeared initially that disease manifestations and clinical course of transfusion-transmitted HIV infection follows a more rapid course than HIV infection due to other routes of transmission, subsequent analysis did not confirm this impression.40 Also not confirmed were concerns that allogeneic transfusions—via immune stimulation by donor leukocytes— may result in accelerated disease progression and shortened survival in HIV-infected transfusion recipients. The Viral Activation by Transfusion Study (VATS), a large multicenter U.S. clinical trial, specifically addressed this issue and found no evidence of significant activation of HIV replication or accelerated disease progression in HIV-1–infected recipients of either leukoreduced or nonleukoreduced transfusions.41 Widespread availability of potent antiretroviral therapy in western countries since the late 1990s has changed the course of HIV disease to that of a manageable chronic condition with reduction in opportunistic infections and other HIVrelated complications and has markedly prolonged survival. Unfortunately, current treatment regimens are not capable of eradicating HIV from cell- and tissue-based sanctuaries throughout the body, and therapy-related side effects and development of resistant viral strains add to the problems of managing HIV-infected patients.42

Efficiency of HIV Transmission by Blood and Blood Products and Transmission Risk Prior to Blood Donor Screening Human immunodeficiency virus is sensitive to drying43 but survives refrigeration of blood and freezing of plasma. Virus

45 603

604

2

Risk of HIV per unit (%)

TRANSFUSION MEDICINE

III

present in the bloodstream of an undetected infected donor is therefore readily passed on by transfusion. Data from the Transfusion Safety Study (TSS), which traced and enrolled recipients of retrospectively identified HIV-1–seropositive units,44,45 demonstrated HIV transmission in 111 of 124 (89.5%) enrolled recipients.45 Neither characteristics of the donor’s infection nor inherent recipient susceptibility factors significantly influenced transmission of HIV-1 by transfusions.46 Variables that have been identified to correlate with likelihood of HIV-1 transmission are the type of blood component transfused and its duration of storage.46,47 Washed red blood cell (RBC) units and RBC units stored more than 26 days had lower transmission rates than other components. This observation, as well as experimental evidence,48 suggests that component manipulations that reduce the number of viable leukocytes, free virus, or both may reduce but not eliminate infectiousness. With regard to recipients of clotting factor concentrates, an average of approximately 50% of hemophiliacs treated with factor VIII in the early 1980s experienced seroconversion.49,50 However, the rate of seroconversion to anti-HIV-l positivity in hemophiliacs who were treated with very high doses of factor VIII (>500,000 units) approached 100%.49 This finding indicates that, perhaps with the rare exception of persons lacking HIV-1 co-receptors required for infection,51 virtually no one is resistant to HIV-1 infection, given a large enough inoculum and repeated exposures. Given the high efficiency of HIV transmission by intravenous administration of infected blood and blood products, it is not surprising that the majority of HIV transmissions in the United States occurred before discovery of the virus and institution of blood donor screening. A principal lesson learned from the HIV-1 epidemic is how an infectious agent with a prolonged interval between infection and manifestation of clinical disease can spread silently within a population (and its blood donor base) for years before recognition.52 In the United States, this early phase was localized primarily to homosexual and bisexual men, groups who were eligible blood donors at the time. Although we know now that the virus began to spread at an exponential rate in these men in the late 1970s, it was not until 1981 that clusters of Kaposi sarcoma and Pneumocystis pneumonia were first recognized among homosexual men in New York and Los Angeles. In late 1982, descriptions of AIDS-like illnesses in hemophiliacs and recipients of blood components first appeared.53,54 With these reports, a blood-borne infectious etiology for AIDS became probable, and efforts were initiated to exclude from blood donation persons with symptoms of or risk factors associated with AIDS.55 Approximately a year later, HIV-1 was discovered,56 an event soon followed by development of diagnostic anti-HIV-1 tests, which were licensed for donor screening in March 1985. Epidemiologic evidence showed that the risk of transfusion-associated HIV-1 infection in the San Francisco Bay Area, one of the epicenters of the early HIV epidemic in the United States, rose rapidly from its first occurrence in 1978 to a peak risk of approximately 1.1% per transfused unit in late 1982 (Fig. 45–5). Beginning in early 1983 and continuing through implementation of anti-HIV-1 screening, there was a marked, progressive decline in risk as a result of diminishing numbers of blood donations from at-risk or infected individuals. This decline was directly attributable to growing awareness of the infectious nature of AIDS in the homosexual community and to implementation and refinement of

1.5

1

First TA-AIDS case reported: High-risk donor deferral initiated

First AIDS cases reported

HIV discovered: Progressive impact of high-risk donor education Anti-HIV screening implemented

0.5

0 1978

1979

1980

1981 1982 1983 Year of transfusion

1984

1985

1990

Figure 45–5 Risk of human immunodeficiency virus (HIV-1) infections from transfusions in San Francisco before anti-HIV screening. Solid line represents estimated risk of recipient infection per unit transfused. Dashed line indicates what the risk would have been if high-risk donor deferral measures had not been implemented. The risk in the United States as a whole probably trailed that in San Francisco by approximately 1 year, and the peak risk was of lower magnitude. (Modified from Busch MP, Young MJ, et al. Risk of human immunodeficiency virus (HIV) transmission by blood transfusions prior to the implementation of HIV-1 antibody screening. Transfusion 1991;31:4-11.)

donor education and deferral measures by blood banks.57,58 In fact, it was estimated that approximately 90% of highrisk men in San Francisco who were donors between 1980 and 1982 deferred themselves from blood donation before implementation of the anti-HIV-1 test in early 1985.57 These data demonstrate the effectiveness of donor education and self-deferral measures. Despite successful interventions like these, it has been estimated that more than 12,000 transfusion recipients in the United States were infected with HIV from transfusions administered before implementation of blood donor screening for HIV infection.59 Hemophiliacs, who are exquisitely vulnerable to transfusion-transmitted diseases because of exposure to factor concentrates derived from plasma pooled from thousands of blood donors, were hit especially hard by the early epidemic of transfusion-transmitted AIDS.50 Incidence and prevalence data from 1978 through 1990 for 16 hemophiliac cohorts in the United States and Europe show that infections began in 1978, peaked in October 1982 at 22 infections per 100 person-years at risk, and declined to 4 per 100 person-years by July 1984. Few new infections occurred among hemophiliacs after 1986, but by that time, 50% of subjects in the combined cohort were infected. The decline in incidence from the infection peak in late 1982 coincided with health interventions introduced to reduce transmission, including recommendations for reduction in use of factor concentrates in 1982, voluntary donor deferral in 1983, and the availability of heat-treated factor VIII in 1984 and of HIV-1 screening in 1985.

Blood Donor Testing for HIV-1 and HIV-2 Antibody The development of a sensitive and specific test for antibodies to HIV that was suitable for diagnosis (with confirmation by immunoblot analysis) and large-scale screening60 represented a major breakthrough in combating the evolving pandemic in the early 1980s. Mandatory screening of all blood, plasma, and platelet donor samples for HIV with the antibody test began in the United States shortly after the assay became commer-

Testing for HIV Antigen Testing of blood donor samples for HIV-1 antigens (in practice p24 antigen) was implemented in the United States in 1996 in addition to HIV antibody testing. This decision was based on studies showing that HIV-1 antigen appears in blood early in the course of infection, approximately 1 week before antibody is detectable. Transmission of HIV has been reported from transfusion of seronegative blood that was later shown to contain p24 antigen; the donors subsequently showed seroconversion.66 Mathematical models, constructed with findings from geographic areas with very high incidence of new HIV infections, suggested that routine antigen testing would detect 1 antigen-positive/antibody-negative donation in every 1.6 million tested.67 In reality, this projection was not achieved. Only seven donations from infected window-phase (p24 antigen-positive/anti-HIV-negative) volunteer donors were intercepted on the basis of p24 antigen testing alone in the first 5 years of screening, for an observed yield of 1 in 10 million donations.68 A possible explanation for the underperformance of HIV-1 antigen testing in detecting HIV-1 antibody-negative window period infections is that prospective

blood donors in the early phase of HIV-1 infection who have HIV-1 antigen only may be experiencing symptoms from acute retroviral infection that deter them from blood donation or result in their rejection during the predonation history and physical examination. After licensing of several nucleic acid amplification testing (NAT) assays suitable for blood and source plasma donor screening beginning in September 2001, the FDA allowed discontinuation of HIV-1 p24 antigen testing for blood donor samples screened by a licensed NAT assay.69 In practice, this has resulted in virtual disappearance of p24 antigen testing from blood donor screening algorithms in the United States. However, antigen/antibody combination tests are employed in many developing countries where NAT screening is not feasible.

HIV, HTLV, AND OTHER RETROVIRUSES

cially available in March 1985.61 Initial testing for antibodies to HIV-1 and HIV-2 is performed on serum or plasma from pilot tubes collected at the time of donation. If the initial test result is reactive, duplicate retesting is performed. If results of one or both second tests are reactive, the unit is designated repeat reactive and discarded. A process is also initiated to identify and quarantine all in-date components from any prior donations from that donor and to notify recipients who received blood components from prior donations.62 Developments in test methodology have occurred at a rapid rate, and blood banks have implemented improved assays as soon as possible after FDA licensure. The evolution of antibody assays has been achieved through improvements in the antigens on the solid phase and in assay formats. From crude viral lysates, production moved to viral lysate assays spiked with purified HIV-1 antigens, and then on to the use of cloned (recombinant DNA [rDNA]-derived) and synthetic peptide antigens. Although concerns have been raised about the sensitivity of rDNA-derived and peptide antigen-based enzyme immunoassays (EIAs) to immunologically variant strains,22,23 these assays have been “engineered” to include highly selected antigenic regions of HIV-1 and HIV-2 that maximize sensitivity to infected subjects from around the world. Consequently, rDNA-derived and synthetic peptide antigen-based assays are now in wide use in U.S. and non-U.S. blood banks. Test manufacturers have also developed new assay formats with greater capacity to detect low-titer immunoglobulin M (IgM) antibody produced during early seroconversion. For example, the recombinant “antigen sandwich” EIA format allows use of lower dilutions of donor sera and detects IgM with better sensitivity.34–36 The progressive introduction of assays with increased sensitivity to early infection has led to significant narrowing of the seroconversion window period and concomitant reductions in the HIV transmission risk associated with receiving screened blood.63,64 The realization that HIV-1 subgroup and variant strains may not be reliably detected by current HIV-1/HIV-2 assays has led to demands by the FDA and efforts by kit manufacturers to reengineer standard assay formats to improve detection of unusual viral strains without compromising their sensitivity to main group HIV-1/HIV-2 strains.65

Nucleic Acid Testing Detection of genomic HIV RNA sequences in plasma by NAT represents the most sensitive method for identification of newly infected persons who have not yet developed antibody to the virus (i.e., are in the preseroconversion window period of the infection).70,71 An unrecognized new HIV infection in a healthy-appearing blood donor is by far the greatest single risk factor for transmission of the virus via transfusion,72 development of suitable NAT assays for large-scale blood donor screening in the late 1990s was an important milestone toward the ultimate goal of complete prevention of HIV transmission from screened blood units. NAT assays began to be used in Europe for regional blood donor screening in 1997.73 In the United States, routine testing of blood donor samples with combination NAT assays for HIV and hepatitis C virus (HCV) began in the spring of 1999 in the form of large-scale clinical trials conducted under investigational new drug (IND) applications to the FDA. Despite the research aspect, NAT quickly became the standard of practice, and by July 2001 essentially the entire U.S. blood supply was screened for HIV (and HCV) with the new technology. NAT is currently performed on pools of 16 to 24 blood donor samples, known as minipool NAT. The pooling strategy lowers the sensitivity of the test but allows faster turnaround times, which is critically important for on-time release of cellular blood components. In 3 years of practical experience in the United States from March 1999 to April 2002, HIV NAT detected 10 HIV-1 RNA-positive, HIV-1 EIA-negative, p24 antigen-negative donors out of 37 million donations tested (approximately 1 in 4 million) and missed no p24 antigen-positive, HIV-1 EIA-negative donations.70 These results are consistent with a modest reduction in the infectious window period of HIV, averaging approximately 10 to 12 days against a sensitive HIV-1 antibody and 3 days against p24 antigen testing. Rare, unfortunate cases of HIV transmission from minipool NATscreened units have occurred, demonstrating the potential infectiousness of such units.62,74–77 The two NAT test systems for blood donor screening were licensed by the FDA in 2002, allowing facilities that incorporated licensed HIV NAT systems to discontinue HIV-1 antigen testing.

Supplemental Testing Because of the exquisite sensitivity of HIV EIAs used in blood donor screening and the low pretest probability of HIV infection among low-risk blood donors, the majority of positive screening results are false positive, despite the excellent specificity of the tests. Supplemental assays are therefore

45 605

TRANSFUSION MEDICINE

III 606

essential to confirm positive screening results before donor counseling. Supplemental testing must use FDA-licensed reagents and must rule out both HIV-1 and HIV-2 infection.37 Although combination HIV-1/HIV-2 supplemental assays using rDNA-derived or synthetic peptide antigens that appear to accurately detect and discriminate anti-HIV1 and anti-HIV-2 have been developed,78 they have never been approved by the FDA. Therefore, current confirmatory algorithms in U.S. blood banks employ HIV-1 viral lysatebased Western blots or immunofluorescence assays, in combination with a licensed anti-HIV-2 EIA and an unlicensed HIV-2 supplemental assay.37 Interpretive criteria for Western blots have evolved over time as tests have improved and our understanding of the meaning of various banding patterns has grown. For currently licensed assays, a positive interpretation requires antibody reactivity to two of the following three HIV antigens: p24 (the major gag protein), gp41 (transmembrane env protein), and gp120/160 (external env protein/env precursor protein). Although these criteria are generally accurate, it is clear that some donors who show antibody reactivity only to the envelope glycoproteins are not infected with HIV-1, resulting in false-positive classifications by Western blots.79 It is important, therefore, that all initial positive Western blot results are confirmed by testing of a separate follow-up sample, both to rule out specimen mix-up or testing errors and to discriminate nonspecific patterns from early seroconversion. A negative test result by Western blot is, by definition, the absence of any bands. Any other pattern of reactivity that does not meet criteria for a positive result is classified as indeterminate. Only a very small proportion of donors who test indeterminate on Western blot are infected with HIV-1.80 Holodniy and Busch81 provide a detailed discussion of interpretive criteria and review of developments in HIV supplemental test technology. Because HIV NAT testing results are now routinely available on essentially all U.S. blood donations, this information can provide valuable additional clues regarding the infectious status of a donor with inconclusive serologic HIV test results. Donors with true positive repeat reactive HIV EIA results are expected to be reactive on minipool NAT testing82; a negative NAT test result thus suggests a false-positive HIV EIA result. It should be noted, however, that NAT has not been approved for supplemental HIV testing and is not yet incorporated in official confirmatory testing algorithms. Recipients of prior donations from donors with confirmed positive antibody or NAT results are traced in a process called lookback.62,83 Seronegative donors whose supplemental test results are negative (or persistently indeterminate and NAT-negative) are eligible for possible reentry into the donor pool according to an FDA-specified protocol,37 although in practice, logistical and legal considerations have generally prevented reinstatement of such donors.

Risk of Infection from Screened Donations Implementation of anti-HIV-1 screening in 1985 resulted in a marked decline of virus transmission by transfusion over the first several years of screening.84,85 However, initial hopes that the newly developed antibody test would lead to eradication of HIV transmission by blood donor screening were not fulfilled. In the late 1980s, well-documented cases of HIV-1 transmission from screened blood transfusions were reported,86 and 14 of 106 (13%) cases initially reported as due to infections

from screened blood transfusions could be confirmed. It was therefore clear that a small number of transmissions continued, despite screening with antibody tests. After adjusting for uninvestigated cases, and assuming an equal distribution of infections from 1986 through 1991, investigators from the Centers for Disease Control and Prevention (CDC) estimated that during that period, approximately five transfusion-associated AIDS cases due to infections from anti-HIV-screened blood transfusions had occurred per year.87 In the United States, two prospective studies estimated the risk of HIV transmission from antibody-screened donations at approximately 1 in 60,000 units between 1985 and 1991.88,89 These studies were discontinued in 1992 because of their high cost. Since then, an alternative approach has been developed for estimating the risk of HIV infection from transfusions, now known as the incidence-window period model.67 It is based on the premise that the risk of viral transmission by blood transfusion in a given geographic area is primarily a function of the incidence rate— that is, the number of newly infected blood donors per person-years of observation—and the length of the window period—defined as the time from infectiousness to antibody seroconversion. This concept of a window period can also be applied to other HIV tests, such as p24 antigen test, HIV1 RNA NAT, Western blot, and “detuned” HIV-1 antibody assays, where results with a sensitive antibody test are compared with a modified less sensitive assay to discriminate recent from well-established infections.90 Sequential emergence of assay reactivity allows classification of primary HIV-1 infection into distinct laboratory stages (Fig. 45–6).82 Determination of the length of window periods associated with HIV-1 assays enables estimation of residual risk of HIV-1 transmission from repeat donors and, with some adjustments, from first-time donors. Window period estimates for HIV-1 antibody tests in use in the United States dropped from a median of 45 days (95% confidence interval, 34–55 days) for the overall period from 1985 to 1990 to approximately 22 to 25 days after routine introduction in 1992 of new format anti-HIV-1/HIV-2 combination tests, which detect HIV-specific IgM antibody 10 to 15 days earlier than previously available assays.63. By combining the 25-day window period estimate with data on the frequency of HIV seroconversions in large U.S. donor populations, two independent studies derived point estimates for the risk for HIV transmission during the period 1992 through 1995 as 1:450,00091 and 1:495,000.92 Introduction of HIV-1 antigen screening in 1995 further reduced the risk, but because fewer than expected HIV-1 antigen-positive HIV-1/2 antibodynegative units were intercepted once the test was introduced, the test’s contribution to risk reduction seems to have been rather limited. In contrast, incorporation of minipool NAT into routine blood donor screening93 was able to reduce the risk of HIV transmission to approximately 1 in 2 million blood component units transfused.70 Multiplying the perunit risk estimates by approximately 20 million components transfused to 5 million recipients per year in the United States leads to the expectation that fewer than 10 recipients per year are transfused with HIV-1-infected blood. The risk of HIV-2 transmission has been estimated at less than 1 in 15 million,38 with other rare subtypes (e.g., HIV-1 subtype O)22 being of even lesser concern. These combined risks are more than 10,000-fold lower than the risk existing at the peak of the transfusion-associated AIDS epidemic between 1982 and 1984.57

I

P*

P

WB Ab

RNA

LS-Ab LS-Ab OD cutoff

HIV, HTLV, AND OTHER RETROVIRUSES

N

p24 Ag 0 I II III IV V ecl. Recent infection

VI

0

100

25

50

75

Early chronic infection 125

150

175

200

Days from HIV exposure Figure 45–6 Schematic, semi-quantitative display of the progression of HIV markers (from top to bottom: WB, western blot; Ab, HIV antibody; RNA, HIV RNA; LS-Ab, HIV antibody determined by sensitive/less sensitive EIA testing strategy90; p24 Ag, HIV p24 antigen), from time of exposure (day 0) through the first 200 days of infection. As each of the markers appears in the bloodstream, the infection is assigned a new stage from 0, characterized by undetectable viral markers in blood samples, except for perhaps low-level viral blips96; through stage I (definitive HIV RNA viremia), stage II (p24 antigenemia), stage III (HIV EIA Ab reactive), stage IV (WB indeterminate, “I”), stage V (WB positive without p31 band, “P*”), and stage VI (WB positive, “P” with p31 band). Stages I–VI were derived from analysis of 435 serial samples from 51 plasma donors with new HIV-1 infection.82 Incorporation of LS Ab EIA testing would allow further characterization of stage VI samples as representing recent vs. early chronic infection, i.e., infection that occurred within vs. beyond approximately 6 months from Ab seroconversion by an IgM-sensitive EIA.90 The standardized optical density (OD) cutoff for the LS Ab EIA may be varied with recommended cutoffs from 0.5–1.0. Shown in the figure are the results for an OD cutoff of 0.75, as chosen in the original publication.90 Cutoffs of 0.5 and 1.0 would result in average demarcations of recent from early chronic infection at 124 and 186 days, respectively. (Reproduced with permission from Fiebig EW et al. Dynamics of HIV viremia and antibody seroconversion in plasma donors: Implications for diagnosis and staging of primary HIV infection. AIDS 2003;17:1871–1879.)

45

Investigators have frequently used calculations based on the incidence-window period model to project the risk of transmission of HIV and other viruses by transfusion in different geographic areas and at various periods of the epidemic. Experience over time has confirmed the validity of the approach, and the risk projections derived from the model are generally accepted as meaningful and reliable estimates of the true risk.94,95 Decreasing projections of the magnitude of risk in the same population over time reflect removal of high-risk donors and improvements in blood donor screening. Recently, widespread use of NAT assays in blood donor screening led to application of a new strategy for estimating risk of HIV transmission based on rates of detection of recently infected donors.72 The method takes into account the expected increase of viral genomic equivalents (viral load) in blood donor samples during the initial phase of viral replication before appearance of HIV antibodies.72,82 By back-extrapolation of the viral load from a reference point such as appearance of p24 antigen or detection of HIV infection by minipool NAT to a minimal viral concentration that is thought to be infective (e.g., 1 viral copy per 20 mL plasma), the window period between the infectious threshold and the reference point can be projected. By substituting the reference point with a theoretical threshold such as assay sensitivity of a new NAT, the window period for that assay can be estimated. A practical application of the new method is the projection of risk reduction for transfusion transmission of HIV by testing individual donor samples rather than minipools of 16 to 24 samples.

The result suggests a rather modest decrease in risk, from approximately 1 per 2 million component units transfused with MP-NAT, to 1 in 3 to 4 million units with ID-NAT,82 at considerably higher cost and potentially delayed availability of transfusable cellular blood components. Not addressed by the novel strategy of estimating residual transmission risk of HIV based on viral replication dynamics is the finding of low-level (estimated 1 to 10 copies/mL) viral “blips” in the 2- to 3-week long period between infection and ramp-up phase of viremia, sometimes referred to as the eclipse phase of HIV infection (see Fig. 45–6). Such “blips” have been observed at intervals of approximately 1 to 3 weeks before HIV ramp-up viremia in 6 of 15 plasma donors with newly acquired HIV infection.96 The added risk from blood donations during the eclipse phase is unknown, but might contribute slightly to the overall risk estimate associated with current NAT blood donor screening.96

HUMAN T-CELL LYMPHOTROPIC VIRUS Although eclipsed by concerns about HIV, HTLV is nonetheless relevant to the safety of blood transfusion. Both HTLV-1 and HTLV-2 are transmitted by blood transfusion, cause chronic retroviral infection of humans, and are associated with serious disease outcomes. Serologic testing for HTLV-1 in U.S. blood donors has been in place since 1988. This policy has led to the unexpected discovery that at least half of blood donors testing positive for HTLV-1 are in fact

607

Virology Both HTLV-1 and HTLV-2 are human retroviruses of the oncornavirus class. As previously mentioned, the first report of HTLV-1 discovery was published in 1980 and that of HTLV-2 in 1982.5,7 With RNA genomes of approximately 8000 nucleotides in length, the genetic organizations of HTLV-1 and HTLV-2 are similar to those of other retroviruses (Fig. 45–7).100,101 The HTLV genome contains gag regions coding for viral core proteins, pol regions coding for viral reverse transcriptase, protease and integrase env regions coding for the viral envelope proteins, and finally the tax or px region (analogous to the HIV tat gene), which is responsible for transcriptional regulation of HTLV-1 and HTLV-2. HTLV-1 and HTLV-2 have approximately 60% nucleotide sequence homology. Many of their viral proteins crossreact serologically, although peptides eliciting specific immune responses and allowing differential serologic diagnosis of HTLV-1 versus HTLV-2 have been discovered. HTLV-1 predominantly infects CD4+ T lymphocytes, whereas HTLV-2 has a broader tropism with preference for CD8+ lymphocytes and, to a lesser degree, CD4+ lymphocytes, B lymphocytes, and macrophages.102,103 Neither virus has high levels of cell-free viremia, perhaps accounting for lower transmissibility than either hepatitis B virus or HIV. Also in contrast to HIV, there is relatively little active replication of HTLV-1 and HTLV-2 in infected humans. Instead, most expansion of the pool of infected lymphocytes appears to occur by lymphocytic division and the proliferation of clones of infected lymphocytes.104 Corollaries of these observations are that descriptions of HTLV viral load refer to measurements of proviral DNA in the cellular compartment. There has been only one report of measurable HTLV RNA in cell-free plasma105; however, further research will be necessary to exclude the possibility of a significant viremia in cell-free plasma.

1000

0

2000

3000

4000

5000

6000

7000

8000

Non-translated region

U3 R U5

9032

U3 R U5

env

gag pro 9( M p2 A) 4( p1 CA 5( ) N C )

608

HTLV-1 is endemic in sub-Saharan Africa; the Caribbean region and parts of South America, including Colombia, Peru, and Brazil as a result of the slave trade; southwestern Japan; and parts of Melanesia and Australia.106 HTLV-1 has also been reported from Iran, India, and Taiwan as well as in countries containing significant immigrant populations from the main endemic areas. Given the high HTLV-1 prevalence in southwestern Japan, the absence of HTLV-1 infection in mainland China and in Korea as well as other parts of Southeast Asia poses an epidemiologic puzzle. Molecular epidemiologic studies have indicated the presence of at least two viral subtypes in Africa; a cosmopolitan subtype found in Africa, Japan, the Caribbean, and other more recent foci of HTLV-1 infection; and a distantly related Melanesian subtype.107 HTLV-2 has also been found to be endemic in a number of Amerindian tribes throughout South, Central, and North America.108–110 Not all tribes are infected, and tribes with the least intermingling with Western settlers, such as the Kayapo of Brazil, appear to have the highest prevalence rate.108 The other endemic focus of HTLV-2 infection appears to be among Pygmies in sub-Saharan Africa, again in tribes having relatively little contact with European settlers.111 HTLV-2 infection is also prevalent among injection drug users and their sexual partners in the United States, Brazil, and Europe.112 Given the relative recency of injection drug use behavior, there appears to have been an epidemic spread of HTLV-2 over the past 40 or 50 years as a result of the introduction of endemic HTLV-2 into the injection drug use population and subsequent spread through sharing of contaminated needles.113 In endemic populations, both HTLV-1 and HTLV-2 show age-specific seroprevalence rates that rise steadily with age, from 1% or less among infected children to 5% to 10% of older individuals.113–116 Female seroprevalence is generally greater than male seroprevalence, presumably owing to more efficient sexual transmission from males to females than vice versa. In one large study of blood donors in the United States in the 1990s, HTLV-1 seroprevalence was approximately 10 per

gp46(SU) p21(TM)

pol

p1

TRANSFUSION MEDICINE

III

Epidemiology and Modes of Transmission

infected with HTLV-2, for which disease outcomes are less well described.97–99

tax rex

mRNAs: gag, pro, pol

An

env

An

tax, rex

An

HTLV-II RxRE HTLV-I/II CRS

HTLV-I RxRE

Figure 45–7 Structure and organization of the HTLV genome. The HTLV provirus genome is shown at the top of the figure. Positions of known genes are indicated. Sizes and positions of the proteins encoded by the provirus are shown beneath the provirus. The structure of the three messenger RNA species produced are shown at the bottom of the figure. (Reproduced with permission from Fields BN, Knipe DM, Howley PM (eds). Fields Virology, 3rd ed. Philadelphia, Lippincott-Raven, 1996, p 1850.)

HTLV-II Prevalence per 100,000

180 160 140 120 100 80 60 40 20 0 ⬍30

A

30–39

40–49

50–59

60⫹

50–59

60⫹

Age group

HTLV-II Prevalence per 100,000

180 160 140 120 100 80 60 40 20 0

B

⬍30

30–39

40–49 Age group

Figure 45–8 Age-specific seroprevalence of human T-cell lymphotropic virus type 2 among female (A) and male (B) blood donors in five U.S. cities participating in the multicenter Retrovirus Epidemiology Donor Study, stratified by two western (solid circle) and 3 eastern or central (solid diamond) blood centers. Seroprevalence is expressed per 100,000 donors.

readily transmitted than HTLV-2 may not be true.120 The most important factors determining whether transmission will occur between HTLV-serodiscordant couples (i.e., one seropositive and one seronegative for HTLV) are the total duration of the sexual relationship and the HTLV proviral load in the male.121 The frequency or type of sexual activities performed appears to have little effect on the risk of transmission, although the predominant behavior in this study was heterosexual vaginal intercourse. The CDC recommends prevention of sexual transmission through the use of condoms by HTLV-serodiscordant couples as well as by HTLV-seropositive single individuals with multiple sexual partners.121 HTLV-1 and HTLV-2 are also transmitted efficiently by the sharing of contaminated needles or other parenteral exposures, leading to high seroprevalence among injection drug users.112,122 Data on transmission by blood transfusion are provided in more detail later. There is no data supporting the casual transmission of HTLV-1 or HTLV-2 within households or between individuals who do not have any of the preceding risk factors for transmission.

HIV, HTLV, AND OTHER RETROVIRUSES

100,000 overall, rising steadily with age, and was twice as prevalent in females as in males.113 In contrast, HTLV-2 was found among the same U.S. blood donors at a rate of 20 per 100,000 overall.113 Other studies of U.S. donors have shown similar overall prevalence, but a more equal ratio of HTLV2 to HTLV-1.114 Although there is at least a twofold female excess of infection, the age prevalence curve peaks in persons aged 30 to 49 and falls in older individuals (Fig. 45–8). This apparent birth cohort effect is consistent with the hypothesis of an epidemic of HTLV-2 transmitted by injection drug use and secondary sexual transmission beginning in the late 1960s and 1970s.113 HTLV-1 and HTLV-2 have similar modes of transmission. Mother-to-child transmission is important for both viruses, and between 20% and 30% of infants born to infected mothers become infected themselves.117 In contrast to HIV, however, the great majority of such transmission appears to occur by means of breast-feeding. Public health intervention studies in Japan have documented a decrease in the transmission rate from 30% to 3% when infected mothers substituted bottle-feeding for breast-feeding.117 Sexual transmission also occurs for both HTLV-1 and HTLV-2, presumably by means of infected lymphocytes or other cells contained in seminal fluid.118,119 The earlier impression that transmission from males to females is more efficient than vice versa and that HTLV-1 is more

Disease Outcomes Soon after its discovery, HTLV-1 was associated with a CD4+ T-lymphocytic lymphoma, ATL.123,124 This lymphoma, which has a leukemic phase in one fourth to one third of infected individuals, is also associated with hypercalcemia, skin lesions, and hepatosplenomegaly.125 Atypical malignant lymphocytes with convoluted nuclei, referred to as flower cells, are seen in a high number of patients with leukemic-phase ATL and may also be seen in low numbers in asymptomatic persons who are seropositive for HTLV1.126,127 The prognostic implications of having flower cells are not well defined, although individuals with either very high HTLV proviral loads or with relatively high numbers of circulating flower cells appear to be predisposed to a higher risk of ATL.128 An individual infected with HTLV-1 at birth has an estimated 4% lifetime risk of ATL, and the risk is presumably lower for those infected sexually as adults.129 There are two case reports of ATL occurring after apparent transfusion-related HTLV-1 infection.130 Chemotherapy is often less effective for ATL than for other lymphomas and leukemia, and a high mortality is associated with this hematologic malignancy.131,132 Although both HTLV-1 and HTLV-2 are known to cause spontaneous lymphocytic proliferation in vitro,133–135 HTLV-2 does not appear to cause hematologic malignancy. Even though HTLV-2 was initially isolated from two cases of atypical T-lymphocytic hairy cell leukemia, subsequent epidemiologic studies of hairy cell leukemia have not revealed an association with HTLV-2 infection.136–139 Similarly, initial case reports of mycosis fungoides and large granular cell leukemia among HTLV-2-infected persons have not led to confirmed associations after further investigation.140–142 The other major disease association of both HTLV-1 and HTLV-2 is HAM/TSP. Initially described by Gessain and colleagues143 in Martinique, the association with HAM/TSP was soon confirmed by investigators in Japan and others.144–147 HAM/TSP is a slowly progressive myelopathy characterized by spastic paraparesis of the lower extremity, hyperreflexia, bowel and bladder symptomatology, and relative sparing of both upper extremity strength and cognitive function.148

45 609

TRANSFUSION MEDICINE

III 610

The disease course is slow but progressive, with 10 years often elapsing between the first signs and severe paraplegia necessitating the use of a wheelchair. There is no definitive treatment for HAM/TSP, although the use of systemic steroids, immunosuppression with azathioprine, and the use of the androgenic agent danazol have had transient success.148 Clinical trials are planned to evaluate the efficiency of various treatment regimens. A cohort study has estimated the risk of HAM/TSP to be about 2% in HTLV-1–positive persons.149 Sexual acquisition of HTLV-1 may be a risk factor for HAM/TSP150; however, the incubation period may be shortest after transfusion-acquired infection.151–153 Cases of typical HAM/TSP have been reported in HTLV-2–infected humans, although the risk appears to be slightly less than or equal to that with HTLV-1.149 In contrast to some initial reports, there is no association of either HTLV-1 or HTLV-2 and classic multiple sclerosis.154–156 Likewise, the implication of HTLV-1 and HTLV-2 in other neurologic syndromes is controversial, because individuals infected with HTLV-2 may have genetic or environmental factors that also cause neurologic disease.157 Other immunologic diseases and phenomena have been reported in relation to HTLV-1 and HTLV-2. HTLV-1 has been associated with lymphocytic pneumonitis and uveitis.158,159 HTLV-1–positive cases of polymyositis have also been reported, and both HTLV-1 and HTLV-2 may be associated with a higher incidence of arthritis.159–161 Of particular interest, a cohort study has shown that HTLV-1, and particularly HTLV-2, may be associated with a higher incidence of other infections, including pneumonia, acute bronchitis, and urinary tract infection.162 This finding is consistent with the association of HTLV-1 and infective dermatitis in children.163 The mechanism of these disease associations has not yet been described and may be due to either subclinical immune dysfunction or increased inflammation. It is not clear whether either HTLV-1 or HTLV-2 contributes to an increased incidence of other types of malignancy; a study in Japan has reported an increased risk of “viral-associated” malignancy such as hepatoma.164 However, a study in the United States showed no increase of any nonhematologic malignancy with HTLV-1 or HTLV-2.162 Intriguingly, infection with HTLV-2 but not HTLV-1 was found recently to be associated with increased mortality in blood donors.165 In another recent study on the impact of co-infection with HTLV-1/2 in HIVinfected patients, co-infection with HTLV-2 was associated with a higher rate of clinical complications but paradoxically also with longer survival and delayed progression to AIDS.166 However, a well-designed meta-analysis, which accounted for date of HIV infection, showed no effect of HTLV-2 co-infection on HIV disease progression.167

Transfusion-Related Transmission Soon after its discovery, HTLV-1 was shown to be transmitted by blood transfusion, with transmission efficiency in Japan of at least 50%.168,169 A cohort study in Jamaica in the late 1980s demonstrated a 25% to 30% risk of HTLV-1 infection after transfusion of one unit of HTLV-1-infected blood, with a 50-day mean latency period between the implicated transfusion and the development of de novo anti-HTLV antibodies.170 Results of the TSS in the United States, with data from the mid-1980s, showed a much lower rate of transmission, only 10% to 20% for HTLV-1 and HTLV-2.171 Differences between these transmission rates may be due to the requirement that

HTLV-infected, viable lymphocytes must be present in the transfused unit for it to be classified as infected. A significant inverse correlation between the duration of refrigerated storage (presumably related to lymphocyte viability) and the risk of HTLV transmission was demonstrated in the TSS. Although such data were not reported, it is conceivable that blood unit storage times were shorter in the Japanese and Jamaican studies than in the U.S. study, and therefore viable infected lymphocytes were more frequently transfused. A later study of recipients of large-volume transfusions in the United States showed estimated risks of HTLV transmission to be 12 per 100,000 units before and 1.4 per 100,000 units after the 1988 institution of universal screening for HTLV antibodies.88 The screening was introduced in that year after consideration of early data on HTLV-1 prevalence in blood donors, its transfusion transmissibility, and diseases associated with the infection.114 Studies estimating the current risk of HTLV-1 by blood transfusion, similar to those of other low-frequency agents, are hindered by the large sample sizes required to measure low-frequency events. Nonetheless, the Retrovirus Epidemiology Study and the American Red Cross (ARC) have measured the incidence of HTLV-1 and HTLV-2 infection among serial blood donors and have used these data, along with estimates of the window period between infection and the development of antibodies, to model the residual risk of HTLV-1 infection. The most recent published projections, based on voluntary donations to the ARC system from 2000 to 2001, estimated a residual risk of approximately 1 per 3 million transfused units for HTLV-1, which is lower than the current risk of HIV and hepatitis B or C infection, and evidence of further dramatic risk reduction since the early 1990s. The risk of transmission of HTLV-1 by cell-free plasma is controversial. The Jamaican study showed no episodes of transfusion transmission in individuals receiving only plasma transfusions,170 and laboratory studies have shown that cell-free transmission of HTLV-1 is difficult. A report on the detection by RT-PCR of HTLV-1 viral RNA in cellfree plasma is of concern, if its findings can be replicated.105 Nonetheless, transfusion of blood products containing residual leukocytes appears to carry a higher risk of HTLV transmission than transfusion of products free of leukocytes.90 Also of note in regard to transfusion safety is the definite association between HTLV-2 infection and previous injection drug use or sexual relations with an injection drug user. Because injection drug users are more effectively screened out of the blood donor population, the typical HTLV-2-infected individual is a middle-aged woman who reports remote sexual exposure to an injection drug user. These data highlight the need to continue refining behavioral screening criteria, including consideration of banning donors who admit any prior sexual contact with injection drug users.172

Laboratory Diagnosis Screening for HTLV-1 and HTLV-2, analogous to HIV screening, is generally performed with EIAs. The earliest assays included only HTLV-1 native viral proteins. It soon became apparent, however, that these assays were deficient in the native HTLV envelope proteins because of the viral purification process used to produce the proteins. This problem led to the subsequent addition of recombinant envelope proteins to the EIA antigen mix to improve the test’s sensitivity. Because HTLV-1 and HTLV-2 have 60% nucleotide sequence

supplemental assays that would differentiate HTLV-1 and HTLV-2 infections. Less clear is the need for improving the sensitivity of HTLV blood donor screening by implementing NAT-based HTLV testing. Lower incidence rates of infection than seen for the other major transfusion-transmitted viruses, at least partial protection of blood recipients from HTLV transmission by widespread use of leukoreduced cellular blood components, and low disease penetrance in the 5% range argue against such a proposition, especially if cost effectiveness is considered. On the other hand, tolerance of even a low risk of retroviral infection has not been acceptable in a political environment that demands zero risk. Before large-scale blood donor screening with HTLV NAT assays could be introduced, however, technical issues would also have to addressed. Current NAT systems do not have the capacity to extract cell-associated nucleic acids, so detection of HTLVs (and other cell-associated pathogens, including parasites) by NAT will have to await a new generation of high-throughput screening systems.

HIV, HTLV, AND OTHER RETROVIRUSES

homology, there is significant cross-reaction of HTLV-2 with HTLV-1 EIAs.101 However, because 10% to 15% of HTLV-2 infections were missed by earlier HTLV-1 EIAs,173 subsequent EIAs utilized a recombinant transmembrane envelope glycoprotein, rgp21, that had greater sensitivity for both HTLV-2 and HTLV-1, and later generations of EIAs use recombinant proteins from both HTLV-2 and HTLV-1. Thus, the FDA has permitted the labeling of single assays for the combined detection of HTLV-1 and HTLV-2. Supplemental testing for HTLV-1 and HTLV-2 has continued to be more problematic. The earliest and most common supplemental test is the Western blot. Because of a deficiency in viral envelope protein, however, Western blots using only HTLV-1 native viral proteins are relatively nonspecific and may also be insensitive to HTLV-2. Western blots supplemented with rgp21, which have greater sensitivity to HTLV-2 because of its crossreactivity with rgp21, soon became available. However, false-positive reactivity to rgp21 in combination with nonspecific reactivity with the gag p19 or p24 bands occasionally led to false-positive Western blot interpretations.174 Radioimmunoprecipitation has been used in research laboratories over the years to more specifically confirm the presence of antibodies to envelope proteins of HTLV-1 and HTLV-2. However, this assay is very labor intensive and so is not suitable for use as a routine supplemental test in blood banks. Second-generation Western blots have been developed that contain native HTLV-1 proteins supplemented with a refined version of rgp21, to decrease nonspecific reactions, in addition to recombinant peptides specific to both HTLV-1 (rgp46-I) and HTLV-2 (rgp46-II).175,176 In the research setting, these assays have proved to be both sensitive and specific for the diagnosis of HTLV-1 and HTLV-2. Because of the presence of the type-specific epitopes, they allow the differentiation of HTLV-1 from HTLV-2 infection in the great majority of cases.177,178 HTLV type-specific peptides are also available in the EIA format; however, these tests should be used for typing specimens that have already been confirmed by other supplemental assays.177–179 This issue is important, because counseling of the infected donor should be specific to the disease outcomes to be expected with either HTLV-1 or HTLV-2, and risk factor profiles have been shown to be specific to either HTLV-1 or HTLV-2.122 Few of the HTLV supplemental tests described have been licensed by the FDA, however, so they have been used only under provisions for research. Thus, the counseling of infected donors based on the results of such tests has carried potential regulatory risks. In addition, because of the complexity of some of these supplemental assays, proficiency of the laboratories performing them depended on the local expertise and the volume of supplemental testing performed. This situation has led the FDA to ban the use of nonlicensed supplemental assays in the diagnosis of blood donors who test positive on the HTLV EIA. Although the regulatory reasons for this decision are understandable, it has significantly reduced the specificity of the diagnosis of HTLV-infected blood donors, because only 10% to 25% of EIA-positive blood donors are in fact seropositive for HTLV-1 or HTLV-2, and donors with indeterminate EIA results are unlikely to be infected with HTLV in the absence of definite risk factors.180 The lack of accurate licensed supplemental tests means that a large number of HTLV EIA-positive donors are falsely being informed that they are potentially seropositive for HTLV. Thus, there is an urgent need for the development of licensed

ADDITIONAL RETROVIRUSES Besides HIVs and HTLVs, there may be additional human retroviruses that have not as yet been discovered. It is also now well established that retroviruses from other species can be transmitted to humans and subsequently pose a risk to the blood supply. These concerns led some to call for active surveillance for new and emerging retroviruses with potential relevance for transfusion safety. Indeed two new HTLV variants, tentatively named HTLV-3 and HTLV-4, have been isolated from a few human Pygmies and Bantu living in remote areas of sub-Saharan Africa.181,182 Because the infected humans hunted and butchered monkeys or kept them as pets, the authors suggest that recent cross-species transmission may have occurred. Interestingly, it is known that 5% to 10% of the genomes of humans and many other species is comprised of endogenous retroviral sequences. The origin and functions of human endogenous retroviral (HERV) sequences remain unclear, although it is generally believed that these sequences represent relics of ancestral retroviral infections that have been conserved through evolution for functional properties, including protection from exogenous retroviral infection. To date there is no convincing evidence that HERVs are associated with any disease or that they can be transmitted horizontally by blood transfusion or other routes. However, there is concern that endogenous retroviruses in the genomes of other species could be activated into exogenous retroviruses and be transmitted to humans.183,184 This was highlighted recently by demonstration of in vitro transmission of porcine endogenous retrovirus sequences to human cells.183 The potential of cross-species infection is of particular concern with animal-derived blood constituents (e.g., porcine-derived clotting factors, bovine albumin, equine immunoglobulins) or organs in xenotransplantation protocols. Reassuringly, recent studies to examine the issue of whether endogenous retroviruses from other species may have infected humans exposed to blood products or organs from these species have failed to document any cases of infection using sensitive serologic and molecular assays.184,185 A further concern that has been raised regarding crossspecies transmission of exogenous retroviruses is the potential that such transmissions could result in accelerated

45 611

TRANSFUSION MEDICINE

transmission patterns and increased pathogenicity with humans, similar to what likely happened with SIV/HIV. For example, a clinical syndrome called idiopathic (HIV-negative) CD-4 lymphocytopenia, described in the early 1990s, was speculated to result from a novel retrovirus infection. However, subsequent studies both in high-risk and blood donor populations failed to confirm a retroviral or other infectious etiology, with the exception of a subset of cases due to group O HIV-1.186 There has also been speculation over the years that donors with indeterminate seroreactivity for HIV or HTLV may harbor crossreactive primate or ungulate retroviruses.187 However, most studies that have probed samples from donors with indeterminate test results have failed to identify evidence for infection based on type-specific serologic or PCR analyses.187 Over the past several years novel assays have been developed that allow for investigation of donor and other specimens for evidence of unknown retroviruses. These include amplified reverse transcriptase assays (Amp-RT), also called productenhanced RT assays (PERT), and generic PCR assays that target conserved regions of polymerase and gag genes that are common among highly divergent retroviral families. Continued application of these tools to samples from blood donor–recipient repositories, both in developed and developing countries, will be important to reassure the public that additional retroviral agents do not pose a risk to blood safety.

APPROACHES FOR FURTHER REDUCING THE RISK OF RETROVIRAL TRANSMISSION BY TRANSFUSIONS III 612

Twenty years of blood donor screening for HIV and HTLV, including utilization of sensitive NAT technology for HIV RNA detection since the late 1990s, have greatly contributed to improved blood safety with regard to retroviral transmission in high-income nations that can afford and support the required infrastructure for sophisticated blood screening programs. Yet, despite the now exceedingly small likelihood of becoming infected with HIV or HTLV from a screened blood transfusion, there continues to be the call for and expectation of a blood supply that is risk free in regard to HIV and other infectious disease transmission. Principal avenues that are actively being pursued in the effort to eliminate transfusion-transmitted retroviral infections are the selection of the safest possible donor populations, continuing improvements in blood donor testing, and introduction of safe, effective, and affordable viral inactivation methods into transfusion practice. Because of the disproportionate impact of the HIV epidemic on poor urban racial minorities and the known higher prevalence of HIV among minority donors, some have argued for what has been termed demographic recruitment. However, this concept fails to recognize the critical need for a genetically diverse donor base to support the transfusion needs of a similarly diverse recipient population. Others have proposed expanding or modifying the risk factor interview process by, for example, adding questions on recent heterosexual activity, using cartoon depictions of risk behaviors to improve understanding by less-educated donors, and implementing computer-based interview strategies to enhance confidentiality.188 Convincing data to support these proposals have not been generated, however.189 Risk factor profiles of seropositive donors have changed over time. For example, Petersen

and associates190 compared HIV risk factors among 508 seropositive donors identified at 20 blood centers between May 1988 and August 1989 with those of 472 seropositive donors identified from January 1990 through May 1991. The overall rate of seropositive donations declined slightly over time (from 0.021% to 0.018%), primarily as a result of a decrease in infected donations by homosexual and bisexual men. In contrast, the rate of infected donations by persons probably exposed through heterosexual contact remained stable. Toward the end of the observation period, 56% of infected female donors and 12% of infected male donors had seropositive heterosexual partners; another 41% of infected female donors and 29% of infected heterosexual male donors were probably infected by heterosexual contact (on the basis of serologic studies for sexually transmitted disease markers), even though specific infected sex partners could not be identified. These researchers rejected the option of deferring donors on the basis of recent or lifetime number of heterosexual partners, because numeric cutoffs that would identify only half of the seropositive heterosexual donors would also result in a loss of 7% to 13% of all donations. This study illustrates the significant tradeoffs required to maintain an adequate as well as safe blood supply. Incentives for blood donors pose a similar dilemma. Paid blood donors have long been identified as having a higher risk for transmissible viral diseases, and a volunteer-only blood donor system is considered a cornerstone in ensuring blood safety.191 Nonmonetary incentives are widely used to attract volunteer donors, however, and surveys have shown that some of these incentives may also entice high-risk donors to give blood.192 Since implementation of NAT testing in the so-called mini-pool format, strategies to further improve blood donor screening for HIV focus on introduction of individual donation (ID) testing. As studies have shown, however, the expected additional risk reduction from ID-NAT testing would be modest,72 and complete elimination of transmission risk is unlikely to be achieved by blood screening alone. An added layer of safety is provided by viral inactivation technology, which has been used for almost 2 decades in the manufacturing of plasma derivatives and almost as long for treatment of pooled plasma in many European countries.193 Various methods are being used, with solvent-detergent treatment being the most common principle employed.194 Effectiveness of the solvent-detergent method is very high for lipid-enveloped viruses, including HIV and HTLV, but the procedure does not reliably inactivate nonenveloped viruses.195 The safety of the solvent-detergent treatment process for fresh frozen plasma came under further scrutiny by the finding of preferential inactivation of antithrombotic proteins in treated plasma that could result in a prothrombotic state in large-volume recipients. Reports of venous thrombotic events (VTE) in patients receiving exchange transfusions with solvent-detergent–treated fresh frozen plasma,196,197 and several fatalities associated with VTEs in liver transplant patients who received solvent-detergent– treated fresh frozen plasma, contributed to the discontinuation of use of the product in the United States, although it is still available in Europe.195 Newer chemical, photodynamic, and photochemical methods capable of inactivating a wide range of pathogens, including known transfusion-transmissible viruses, bacteria, and parasites, have been developed. These work with single-donation platelets and fresh frozen plasma components, thereby

REFERENCES 1. Zanotto PM, Gibbs MJ, Gould EA, Holmes EC. A reevaluation of the higher taxonomy of viruses based on RNA polymerases. J Virol 1996;70:6083–6096. 2. Levy J. HIV and the Pathogenesis of AIDS. Washington, D.C., ASM Press, 1998. 3. Rowland-Jones SL. Timeline: AIDS pathogenesis: What have two decades of HIV research taught us? Nat Rev Immunol 2003;3:343–348. 4. Bishop JM. The molecular genetics of cancer. Science 1987;235: 305–311. 5. Poiesz BJ, Ruscetti FW, Cazdar AF, et al. Detection and isolation of type C retrovirus particles from fresh and cultured lymphocytes of a patient with cutaneous T-cell lymphoma. Proc Natl Acad Sci USA 1980;77:7415–7419 6. Blattner WA (ed). Human Retrovirology: HTLV. New York, Raven Press, 1990. 7. Kalyanaraman VS, Sarngadharan MG, Robert-Guroff M, et al. A new subtype of human T-cell leukemia virus (HTLV-II) associated with a T-cell variant of hairy cell leukemia. Science 1982;218:571–573. 8. Rosenblatt JD, Chen IS, Golde DW. HTLV-II and human lymphoproliferative disorders. Clin Lab Med 1988;8:85–95. 9. Gao F, Bailes E, Robertson DL, et al. Origin of HIV-1 in the chimpanzee Pantroglodytes troglodytes. Nature 1999;397:436–444. 10. Zhu T, Korber BT, Nahmias AJ, et al. An African HIV-1 sequence from 1959 and implications for the origin of the epidemic. Nature 1998;391: 594–597. 11. UNAIDS. Joint United Nations Programme on HIV/AIDS. 2006 Report on the Global AIDS Epidemic: Executive Summary. UNAIDS, Geneva, Switzerland, May 2006. Available at http://www.unaids.org/ en/publications. Accessed June 30, 2006. 12. Roberts JD, Bebenek K, Kunkel TA. The accuracy of reverse transcriptase from HIV-1. Science 1988;242:1171–1174. 13. Vartanian J-P, Meyerhans A, Asjo B, et al. Selection, recombination, and G-A hypermutation of HIV-1 genomes. J Virol 1991;65:1779–1788. 14. Zhu T, Mo H, Wang N, et al. Genotypic and phenotypic characterization of HIV-1 in patients with primary infection. Science 1993;261: 1179–1181. 15. Zhu T, Wang N, Carr A, et al. Evidence for coinfection by multiple strains of human immunodeficiency virus type 1 subtype B in an acute seroconverter. J Virol 1995;69:1324–1327. 16. Diaz R, Sabino EC, Mayer A, et al. Dual human immunodeficiency virus type 1 infection and recombination in a dually-exposed transfusion recipient. J Virol 1995;69:3273–3281. 17. Perrin L, Kaiser L, Yerly S. Travel and the spread of HIV-1 genetic variants. Lancet Infect Dis 2003;3:22–27. 18. Delwart EL, Orton S, Parekh B, et al. Two percent of HIV-positive U.S. blood donors are infected with non-subtype-B strains. AIDS Res Hum Retrovir 2003;19:1065–1070.

19. Nkengasong JN, Janssens W, Heyndrickx L, et al. Genotypic subtypes of HIV-1 in Cameroon. AIDS 1994;8:1405–1412. 20. Peeters M, Gueye A, Mboup S, et al. Geographical distribution of HIV-1 group O viruses in Africa. AIDS 1997;11:493–498. 21. Pau CP, Hu HDJ, Spruill C, et al. Surveillance for human immunodeficiency virus type 1 group O infections in the United States. Transfusion 1996;36:398–400. 22. Loussert-Ajaka I, Ly TD, Chaix ML, et al. HIV-1/HIV-2 seronegativity in HIV-1 subtype O infected patients. Lancet 1994;343:1393–1394. 23. Schable C, Zekeng L, Pau CP, et al. Sensitivity of United States HIV antibody tests for detection of HIV-1 group O infections. Lancet 1994;344:1333–1334. 24. U.S. Food and Drug Administration. Interim Recommendations For Deferral of Donors at Increased Risk for HIV-1 Group O Infection. Rockville, Md., Center for Biologics Evaluation and Research (CBER), 1996. 25. Clavel F, Guetard D, Brun-Vezinet F, et al. Isolation of a new human retrovirus from West African patients with AIDS. Science 1986;233: 343–346. 26. O’Brien TR, George JR, Holmberg SD. Human immunodeficiency virus type 2 infection in the United States: epidemiology, diagnosis, and public health implications. JAMA 1992;267:2775–2779. 27. Marlink R, Kanki P, Thior I, et al. Reduced rate of disease development after HIV-2 infections as compared to HIV-1. Science 1994;265: 1587–1590. 28. Poulsen AG, Aaby P, Larsen O, et al. 9-year HIV-2-associated mortality in an urban community in Bissau, West Africa. Lancet 1997;349:911–914. 29. George JR, Rayfield MA, Phillips S, et al. Efficacies of U.S. Food and Drug Administration licensed HIV-1-screening enzyme immunoassays for detecting antibodies to HIV-2. AIDS 1990;4:321–326. 30. de Cock KM, Porter A, Kouadio J, et al. Cross-reactivity on Western blots in HIV-1 and HIV-2 infection. AIDS 1991;5:859–863. 31. Busch MP, Petersen L, Schable C, Perkins HA. Monitoring blood donors for HIV-2 infection by testing anti-HIV-1 reactive sera. Transfusion 1990;30:184–187. 32. Centers for Disease Control. Surveillance for HIV-2 infection in blood donors—United States, 1987–1989. MMWR Morb Mortal Wkly Rep 1990;39:829. 33. Centers for Disease Control and Prevention. Human Immunodeficiency Virus Type 2, updated October 1998. Available at http://www. cdc.gov/hiv/pubs/facts/hiv2.htm. Accessed July 7, 2005. 34. Parry JV, McAlpine L, Avillez MF. Sensitivity of six commercial enzyme immunoassay kits that detect both anti-HIV-l and anti-HIV-2. AIDS 1990;4:355–360. 35. Gallarda JL, Henrard DR, Liu D, et al. Early detection of antibody to HIV-1 using an antigen conjugate immunoassay correlates with the presence of IgM antibody. J Clin Microbiol 1992;30:2379–2384. 36. Zaaijer HL, v Exel-Oehlers P, Kraaijeveld T, et al. Early detection of antibodies to HIV-1 by third-generation assays. Lancet 1992;340:770–772. 37. U.S. Food and Drug Administration. Revised Recommendations for the Prevention of Human Immunodeficiency Virus (HIV) Transmission by Blood and Blood Products. Rockville, Md., Center for Biologics Evaluation and Research (CBER), 1992. 38. Sullivan MT, Guido EA, Metler RP, et al. Identification and characterization of an HIV-2 antibody-positive blood donor in the United States. Transfusion 1998;38:189–193. 39. Kahn JO, Walker BD. Acute human immunodeficiency virus type 1 infection. NEJM 1998;339:33–39. 40. Operskalski EA, Busch MP, Mosley JW, Stram DO. Comparative rates of disease progression among persons infected with the same or different HIV-1 strains. The Transfusion Safety Study Group. J Acquir Immune Defic Syndr Hum Retrovirol 1997;15:145–150. 41. Collier AC, Kalish LA, Busch MP, et al. Leukocyte-reduced red blood cell transfusions in patients with anemia and human immunodeficiency virus infection: the Viral Activation Transfusion Study: A randomized controlled trial. JAMA 2001;285:1592–1601. 42. Wood E, Hogg RS, Harrigan PR, et al. When to initiate antiretroviral therapy in HIV-1-infected adults: a review for clinicians and patients. Lancet Infect Dis 2005;5:407–414. 43. Tjotta E, Hungnes O, Grinde B. Survival of HIV-1 activity after disinfection, temperature and pH changes, or drying. J Med Virol 1991;35: 223–227. 44. Kleinman SH, Niland JC, Azen SP, et al, and the Transfusion Safety Study Group.Prevalence of antibodies to human immunodeficiency virus type 1 among blood donors prior to screening: the Transfusion Safety Study/NHLBI donor repository. Transfusion 1989;29:572–580. 45. Donegan E, Stuart M, Niland JC, et al, and the Transfusion Safety Study Group. Infection with human immunodeficiency virus type I

HIV, HTLV, AND OTHER RETROVIRUSES

avoiding the risks associated with pooling of blood components.193 One photochemical inactivation system was licensed in Europe in early 2002; however, licensing in the United States has been delayed for now. Systems of pathogen inactivation of red cell products are at an earlier stage of development and clinical study and are therefore further away from possible implementation.195 An impediment to the introduction of any new viral inactivation method for blood components is the concern that these methods may themselves pose adverse consequences for transfusion recipients, potentially offsetting the benefits of the new treatment method in comparison with existing strategies. Pathogen inactivation methods are discussed in more detail in Chapter 27 . An ongoing struggle is the prevention of transfusiontransmitted retroviral and other infections in low-income and developing nations. Burgeoning prevalence and incidence rates as well as lack of resources make effective blood donor screening in these areas a challenge that has only been partially met. It is hoped that blood safety in developing countries can be improved gradually through establishment of sufficiently funded national and regional programs that incorporate establishment of donor recruitment networks and high-quality screening capacity.198

45 613

TRANSFUSION MEDICINE

46. 47.

48. 49. 50.

51. 52. 53. 54. 55.

56. 57.

III

58. 59.

614 60. 61. 62.

63.

64.

65.

66.

67.

68.

(HIV-1) among recipients of antibody-positive blood donations. Ann Intern Med 1990;113:733–739. Busch MP, Operskalski EA, Mosley JW, et al. Factors influencing human immunodeficiency virus type 1 transmission by blood transfusion. Transfusion Safety Study Group. J Infect Dis 1996;174:26–33. Donegan E, Lenes BA, Tomasulo RA, Mosley JW, and the Transfusion Safety Study Group. Transmission of HIV-l by components type and duration of shelf storage before transfusion. Transfusion 1990;30: 851–852. Rawal BD, Busch MP, Endow R, et al. Reduction of human immunodeficiency virus-infected cells from donor blood by leukocyte filtration. Transfusion 1989;26:460–462. Kim HC, Nahum K, Raska K Jr, et al. Natural history of acquired immunodeficiency syndrome in hemophilic patients. Am J Hematol 1987;24:168–176. Kroner BL, Rosenberg PS, Aledort LM, et al, for the Multicenter Hemophilia Cohort Study. HIV-1 infection incidence among persons with hemophilia in the United States and western Europe, 1978–1990. J AIDS Hum Retrovirol 1994;7:279–286. O’Brien SJ, Moore JP. The effect of genetic variation in chemokines and their receptors on HIV transmission and progression to AIDS. Immunol Rev 2000;177:99–111. Kilbourne ED. New viral diseases: a real and potential problem without boundaries. JAMA 1990;264:68–70. Centers for Disease Control. Update on acquired immune deficiency syndrome (AIDS) among patients with hemophilia A. MMWR Morb Mortal Wkly Rep 1982;31:844. Centers for Disease Control and Prevention. Possible transfusionassociated acquired immune deficiency syndrome (AIDS)—California. MMWR Morb Mortal Wkly Rep 1982;31: 652. American Association of Blood Banks (AABB), Council of Community Blood Centers (CCBC), American Red Cross (ARC). Joint statement on acquired immune deficiency syndrome (AIDS) related to transfusion. Transfusion 1983;23:87–88. Gallo RC. Historical essay. The early years of HIV/AIDS. Science 2002; 298:1728–1730. Busch MP, Young MJ, Samson SM, et al, and the Transfusion Safety Study Group: risk of human immunodeficiency virus transmission by blood transfusions prior to the implementation of HIV antibody screening in the San Francisco Bay Area. Transfusion 1991;31:4–11. Perkins HA, Samson S, Busch MP. How well has self-exclusion worked? Transfusion 1988;28:601–602. Centers for Disease Control and Prevention. Human immunodeficiency virus infection in transfusion recipients and their family members. MMWR Morb Mortal Wkly Rep 1987;36:137–114. Sarngadharan MG, Popovic M, Bruch L, et al. Antibodies reactive with human T-lymphotropic retroviruses (HTLV-III) in the serum of patients with AIDS. Science 1984;224:506–508. Code of Federal Regulations 610.40. Title 21. Food and Drug Administration. U.S. Department of Health and Human Services, National Archives and Records Administration, 2005. U.S. Food and Drug Administration. Current good manufacturing practices for blood and blood components: notification of consignees receiving blood and blood components at increased risk for transmitting HIV infection. Fed Regist 1996;61:47413–47423. Busch MP, Lee LLL, Satten GA, et al. Time course of detection of viral and serological markers preceding HTLV- I seroconversion: Implications for blood and tissue donor screening. Transfusion 1995;35:91– 97. Lindback S, Thorstensson R, Karlsson AC, et al. Diagnosis of primary HIV-1 infection and duration of follow-up after HIV exposure. Karolinska Institute Primary HIV Infection Study Group. AIDS 2000;14:2333–2339. Dorn J, Masciotra S, Yang C, et al. Analysis of genetic variability within the immunodominant epitopes of envelope gp41 from human immunodeficiency virus type 1 (HIV-1) group M and its impact on HIV-1 antibody detection. J Clin Microbiol 2000;38:773–780. U.S. Food and Drug Administration. Memorandum: Recommendation for Donor Screening With A Licensed Test for HIV-1 Antigen. Rockville, Md., Center for Biologics Evaluation and Research (CBER), 1995. Kleinman S, Busch MP, Korelitz JJ, Schreiber GB. The incidence/window period model and its use to assess the risk of transfusion-transmitted human immunodeficiency virus and hepatitis C virus infection. Transfus Med Rev 1997;11:155–172. Busch MP, Stramer SL. The efficiency of HIV p24 antigen screening of US blood donors: Projections versus reality. Presentation at Satellite symposium Sensitivity and Validity of HIV Screening Methods,

69.

70. 71. 72. 73. 74.

75.

76. 77. 78. 79. 80.

81. 82. 83. 84. 85.

86. 87. 88.

89. 90.

Joint Congress of the International Society of Blood Transfusion and the German Society for Transfusion Medicine and Immunohematology, Frankfurt, Oct. 1997. Infusionther Transfusionmed 1998; 25:5–8. U.S. Department of Health and Human Services. FDA, Center for Biologics Evaluation and Research. Guidance for Industry. Use of Nucleic Acid Tests on Pooled and Individual Samples from Donors of Whole Blood and Blood Components (Including Source Plasma and Source Leukocytes) to Adequately and Appropriately Reduce the Risk of Transmission of HIV-1 and HCV. Rockville, Md., Center for Biologics Evaluation and Research (CBER), 2004. Stramer SL, Glynn SA, Kleinman SH, et al. Detection of HIV-1 and HCV infections among antibody-negative blood donors by nucleic acid-amplification testing. NEJM 2004;351:760–768. Pilcher CD, Fiscus SA, Nguyen TQ, et al. Detection of acute infections during HIV testing in North Carolina. NEJM 2005;352:1873–1883. Busch MP, Glynn SA, Stramer SL, et al. A new strategy for estimating risks of transfusion-transmitted viral infections based on rates of detection of recently infected donors. Transfusion 2005;45:254–264. Roth WK, Weber M, Buhr S, et al. Yield of HCV and HIV-1 NAT after screening of 3.6 million blood donations in central Europe. Transfusion 2002;42:862–868. Ling AE, Robbins KE, Brown TM, et al. Failure of routine HIV-1 tests in a case involving transmission with preseroconversion blood components during the infectious window period. JAMA 2000; 284:210–214. Dreier J, Gotting C, Wolff C, et al. Recent experience with human immunodeficiency virus transmission by cellular blood products in Germany: antibody screening is not sufficient to prevent transmission. Vox Sang 2002;82:80–83. Stramer SL, Chambers L, Page PL, et al. Third reported US case of breakthrough HIV transmission from NAT screened blood. (Abstract) Transfusion 2003;43(9 Suppl):40a–41a. Delwart E, Kalmin N, Jones T, et al. First case of HIV transmission by an RNA screened blood donation. Vox Sang 2004;86:171–177. Phelps R, Robbins K, Liberti T, et al. Window-period human immunodeficiency virus transmission to two recipients by an adolescent blood donor. Transfusion 2004;44:929–933. Kleinman S, Busch MP, Hall L, et al for the Retrovirus Epidemiology Donor Study. False-positive HIV-1 test results in a low-risk screening setting of voluntary blood donation. JAMA 1998;280:1080–1085. Busch MP, Kleinman SH, Williams AE, et al, and the Retrovirus Epidemiology Donor Study (REDS). Frequency of human immunodeficiency virus (HIV) infection among contemporary anti-HIV-1 and anti-HIV-1/anti-HIV-2 supplemental test-indeterminate blood donors. Transfusion 1996;36:37–44. Holodniy M, Busch MP. Establishing the diagnosis of HIV infection. In Dolin R, Masur H, Saag MS (eds). AIDS Therapy, 2nd ed. New York,Churchill Livingstone (Elsevier), 2002, pp 3–20. Fiebig EW, Wright DJ, Rawal BD, et al. Dynamics of HIV viremia and antibody seroconversion in plasma donors: implications for diagnosis and staging of primary HIV infection. AIDS 2003;17:1871–1879. Busch MP. Let’s look at human immunodeficiency virus lookback before leaping into hepatitis C virus lookback! Transfusion 1991;31:655–661. Leitman SF, Klein HG, Melpolder JJ, et al. Clinical implications of positive tests for antibodies to human immunodeficiency virus type I in asymptomatic blood donors. NEJM 1989;321:917–924. Busch MP, Operskalski EA, Mosley JW, et al, and the Transfusion Safety Study Group. Epidemiologic background and long-term course of disease in human immunodeficiency virus type 1-infected blood donors identified before routine laboratory screening. Transfusion 1994;34: 858–864. Ward JW, Holmberg SD, Allen JR, et al. Transmission of human immunodeficiency virus (HIV) by blood transfusions screened as negative for HIV antibody. NEJM 1988;318:473–478. Selik RM, Ward JW, Buehler JW. Trends in transfusion-associated acquired immune deficiency syndrome in the United States, 1982 through 1991. Transfusion 1993;33:890–893. Nelson KE, Donahue JG, Munoz A, et al. Transmission of retroviruses from seronegative donors by transfusion during cardiac surgery: A multicenter study of HIV-1 and HTLV-I/II infections. Ann Intern Med 1992;117:554–559. Busch MP, Eble BE, Khayam-Bashi H, et al. Evaluation of screened blood donations for human immunodeficiency virus type I infection by culture and DNA amplification of pooled cells. NEJM 1991;325:1–5. Janssen RS, Satten GA, Stramer SL, et al. New testing strategy to detect early HIV-1 infection for use in incidence estimates and for clinical and prevention purposes. JAMA 1998;280:42–48.

116. Vitek CR, Gracia FI, Giusti R, et al. Evidence for sexual and mother to child transmission of human T lymphotropic virus type II among Guayami Indians, Panama. J Infect Dis 1995;171:1022–1026. 117. Hino S, Katamine S, Miyata H, et al. Primary prevention of HTLV1 in Japan. Leukemia 1997;11(Suppl 3):57–59. 118. Murphy EL, Figueroa JP, Gibbs WN, et al. Sexual transmission of human T-lymphotropic virus type I. Ann Intern Med 1989;111:555–560. 119. Kaplan J, Khabbaz RF, Murphy EL, et al, and the Retrovirus Epidemiology Donor Study Group.Male to female transmission of human T lymphotropic virus types I and II: association with viral load. J AIDS Hum Retrovirol 1996;12:193–201. 120. Roucoux DF, Wang B, Smith D, et al. A prospective study of sexual transmission of human T lymphotropic virus (HTLV)-I and HTLV-II. J Infect Dis 2005;191:1490–1497. 121. Centers for Disease Control and Prevention (CDC) and the U.S. PHS Working Group. Guidelines for counseling persons infected with human T-lymphotropic virus type I (HTLV-l) and type II (HTLV-II). Ann Intern Med 1993;118:448–454. 122. Feigal E, Murphy EL, Vranizan K, et al. HTLV-I/II in intravenous drug users in San Francisco: Risk factors associated with seropositivity. J Infect Dis 1991;164:36–42. 123. Hinuma YK, Nagata K, Hanoaka M, et al. Adult T-cell leukemia: Antigen in an ATL cell line and detection of antibodies to the antigen in human sera. Proc Natl Acad Sci USA 1981;78:6476–6480. 124. Hinuma YK, Komoda H, Chosa T, et al. Antibodies to adult T-cell leukemia-virus-associated antigens (ATLA) in sera from patients with ATL and controls in Japan: A nationwide seroepidemiologic study. Int J Cancer 1982;29:631–635. 125. Bunn PA Jr, Schechter GP, Jaffe E, et al. Clinical course of retrovirus-associated adult T-cell lymphoma in the United States. NEJM 1983;309:257–264. 126. Seiki M, Eddy R, Shows TB, Yoshida M. Non-specific integration of the HTLV provirus into adult T-cell leukemia cells. Nature 1984;309: 640–642. 127. Kinoshita K, Amagasaki T, Ikeda S, et al. Preleukemic state of adult T cell leukemia: Abnormal T lymphocytosis induced by human adult T cell leukemia-lymphoma virus. Blood 1985;66:120–127. 128. Tachibana N, Okayama A, Ishihara S, et al. High HTLV-I proviral DNA level associated with abnormal lymphocytes in peripheral blood from asymptomatic carriers. Int J Cancer 1992;51:593–595. 129. Murphy EL, Hanchard B, Figueroa JP, et al. Modeling the risk of adult T-cell leukemia/lymphoma in persons infected with human T-lymphotropic virus type I. Int J Cancer 1989;43:250–253. 130. Chen Y-C, Wang C-H, Su I-J, et al. Infection of human T-cell leukemia virus type I and development of human T-cell leukemia/lymphoma in patients with hematologic neoplasms: a possible linkage to blood transfusion. Blood 1989;74:388–394. 131. Prince H, Kleinman S, Doyle M, et al. Spontaneous lymphocyte proliferation in vitro characterizes both HTLV-I and HTLV-II infection. J AIDS Hum Retrovirol 1990;3:1199–1200. 132. Wiktor SZ, Jacobson S, Weiss SH, et al. Spontaneous lymphocyte proliferation in HTLV-II infection. Lancet 1991;337:327–328. 133. Portis T, Harding JC, Ratner L. The contribution of NF-kappa B activity to spontaneous proliferation and resistance to apoptosis in human T-cell leukemia virus type 1 Tax-induced tumors. Blood 2001; 98:1200–1208. 134. Lal RB, Rudolph DL, Dezzutti CS, et al. Costimulatory effects of T cell proliferation during infection with human T lymphotropic virus types I and II are mediated through CD80 and CD86 ligands. J Immunol. 1996;157:1288–1296. 135. Prince HE, York J, Owen SM, Lal RB. Spontaneous proliferation of memory (CD45RO+) and naive (CD45RO−) subsets of CD4 cells and CD8 cells in human T lymphotropic virus (HTLV) infection: distinctive patterns for HTLV-I versus HTLV-II. Clin Exp Immunol 1995;102: 256–261. 136. Rosenblatt JD, Giorgi JV, Golde DW, et al. Integrated human T-cell leukemia virus II genome in CD8+ T cells from a patient with “atypical” hairy cell leukemia: Evidence for distinct T and B cell lymphoproliferative disorders. Blood 1988;71:363–369. 137. Rosenblatt JD, Gasson JC, Glaspy J, et al. Relationship between human T-cell leukemia virus-II and atypical hairy cell leukemia: a serologic study of hairy cell leukemia patients. Leukemia 1987;1:397–401. 138. Lion T, Razvi N, Golomb HM, Brownstein RH. B-lymphotropic hairy cells contain no HTLV-II DNA sequences. Blood 1988;7 2:1428–1430. 139. Katayama I, Maruyama K, Fukushima T, et al. Cross-reacting antibodies to human T cell leukemia virus-I and -II in Japanese patients with hairy cell leukemia. Leukemia 1987;1:401–404.

HIV, HTLV, AND OTHER RETROVIRUSES

91. Lackritz EM, Satten GA, Aberle-Grasse J, et al. Estimated risk of transmission of the human immunodeficiency virus by screened blood in the United States. NEJM 1995;333:1721–1725. 92. Schreiber GB, Busch MP, Kleinman SH, Korelitz JJ. The risk of transfusion-transmitted viral infections. The Retrovirus Epidemiology Donor Study. NEJM 1996;334:1685–1690. 93. Busch MP, Kleinman SH, Jackson B, et al. Committee report. Nucleic acid amplification testing of blood donors for transfusion-transmitted infectious diseases: Report of the Interorganizational Task Force on Nucleic Acid Amplification Testing of Blood Donors. Transfusion 2000;40:143–159. 94. Glynn SA, Kleinman SH, Wright DJ, et al. International application of the incidence rate/window period model. Transfusion 2002;42: 966–972. 95. Coste J, Reesink HW, Engelfriet CP, Laperche S, et al. ISBT International Forum: Implementation of Donor Screening for Infectious Agents Transmitted by Blood by Nucleic Acid Technology: Update through 2003. Vox Sang 2005;88:289–303. 96. Fiebig EW, Heldebrant CM, Smith RI, et al. Intermittent low-level viremia in very early primary HIV-1 infection. J AIDS 2005;39:133–137. 97. Lee HH, Swanson P, Rosenblatt JD, et al. Relative prevalence and risk factors of HTLV-I and HTLV-II infection in US blood donors. Lancet 1991;337:1435–1439. 98. Taylor PE, Stevens CE, Pindyck J, et al. Human T-cell lymphotropic virus in volunteer blood donors. Transfusion 1990;30:783–786. 99. Eble BE, Busch MP, Guiltinan A, et al. Determination of human T-lymphotropic virus type by PCR and correlation with risk factors in northern California blood donors. J Infect Dis 1993;167:954–957. Erratum in: J Infect Dis 1993;168:262. 100. Cann AJ, Chen ISY. Human T-cell leukemia virus types I and II. In Fields BN, Knipe DM (eds). Virology. New York, Raven Press, 1990. 101. Hall WW, Ishak R, Zhu SW, et al. Human T lymphotropic virus type II (HTLVII): Epidemiology, molecular properties, and clinical features of infection. J AIDS Hum Retrovirol 1996;13(Suppl 1):s204–s214. 102. Lal RB, Owen SM, Rudolph DL, et al. In vivo cellular tropism of human T lymphotropic virus type II is not restricted to CD8+ cells. Virology 1995;210:441–447. 103. Casoli C, Cimarelli A, Bertazzoni U. Cellular tropism of human T cell leukemia virus type II is enlarged to B lymphocytes in patients with high proviral load. Virology 1995;206:1126–1128. 104. Wattel E, Vartanian JP, Pannetier C, Wain-Hobson S: Clonal expansion of human T cell leukemia virus type I infected cells in asymptomatic and symptomatic carriers without malignancy. J Virol 1995;69:2863–2868. 105. Rios M, Pombo de Oliveira MS, Correa RB, et al. HTLV-I viremia in infected individuals with and without disease [abstract ME 19]. Presented at the Eighth International Conference on Human Retrovirology: HTLV, Rio de Janeiro, June 9–13, 1997. 106. Manns A, Hisada M, La Grenade L. Human T lymphotropic virus type I infection. Lancet 1999;353:1951–1958. 107. Gessain A, Mahieux R, de The G. Genetic variability and molecular epidemiology of human and simian T cell leukemia/lymphoma virus type I. J AIDS Hum Retrovirol 1996;13(Suppl 1):s132–s145. 108. Black FL, Biggar RJ, Neel JV, et al. Endemic transmission of HTLV type II among Kayapo Indians of Brazil. AIDS Res Hum Retrovir 1994;10: 1165–1171. 109. Reeves WC, Cutler JR, Gracia F, et al. Human T cell lymphotropic virus infection in Guayami Indians from Panama. Am J Trop Med Hyg 1990;43:410–418. 110. Hjelle B, Mills R, Swenson S, et al. Incidence of hairy cell leukemia, mycosis fungoides, and chronic lymphocytic leukemia in first known HTLV-II-endemic population. J Infect Dis 1991;163:435–440. 111. Gessain A, Mauclere P, Froment A, et al. Isolation and molecular characterization of a human T-cell lymphotropic virus type II (HTLV-II), subtype B, from a healthy Pygmy living in a remote area of Cameroon: An ancient origin for HTLV-II in Africa. Proc Natl Acad Sci USA 1995;92: 4041–4045. 112. Khabbaz RF, Onorato IM, Cannon RO, et al. Seroprevalence of HTLV-I and HTLV-II among intravenous drug users and persons in clinics for sexually transmitted diseases. NEJM 1992;326:375–380. 113. Murphy EL, Watanabe K, Nass CC, et al. Evidence among blood donors for a 30-year old epidemic of HTLV-II infection in the United States. J Infect Dis 1999;180:1777–1783. 114. Williams AE, Fang CT, Slamon DJ, et al. Seroprevalence and epidemiological correlates of HTLV-I infection in U.S. blood donors. Science 1988;240:643–646. Erratum in: Science 1989;244:757. 115. Murphy EL, Figueroa JP, Gibbs WN, et al. Human T-lymphotropic virus type I (HTLV-I) seroprevalence in Jamaica. I: Demographic determinants. Am J Epidemiol 1991;133:1114–1124.

45 615

TRANSFUSION MEDICINE

III 616

140. Zucker-Franklin D, Coutavas EE, Rush MG, Zouzias DC. Detection of human T-lymphotropic virus-like particles in cultures of peripheral blood lymphocytes from patients with mycosis fungoides. Proc Natl Acad Sci USA 1991;88:7630–7634. 141. Busch MP, Murphy EL, Nemo G: More on HTLV tax and mycosis fungoides. NEJM 1993;329:2035–2036. 142. Loughran TP Jr, Coyle T, Sherman MP, et al. Detection of human T cell leukemia/lymphoma virus, type II, in a patient with large granular lymphocyte leukemia. Blood 1992;80:1116–1119. 143. Gessain A, Vernant JC, Sonan T, et al. Antibodies to human T-lymphotropic virus type-I in patients with tropical spastic paraparesis. Lancet 1985;2(8452):407–410. 144. Osame M, Usuku K, Izumo S, et al. HTLV-I associated myelopathy, a new clinical entity. Lancet 1986;1(8488):1031–1032. 145. Maloney EM, Cleghorn FR, Morgan OS, et al. Incidence of HTLV-I associated myelopathy/tropical spastic paraparesis (HAM/TSP) in Jamaica and Trinidad. J AIDS Hum Retrovirol 1998;17:167–170. 146. Jacobson S, Raine CS, Mingioli ES, McFarlin DE. Isolation of an HTLVl-like retrovirus from patients with tropical spastic paraparesis. Nature 1988;331:540–543. 147. Bhagavati S, Ehrlich G, Kula R, et al. Detection of human T-cell lymphoma/leukemia virus type 1 DNA and antigen in spinal fluid and blood of patients of chronic progressive myelopathy. NEJM 1988;318: 1141–1147. 148. Gessain A, Gout O. Chronic myelopathy associated with human T-lymphotropic virus type I (HTLV-I). Ann Intern Med 1992;117: 933–946. 149. Murphy EL, Fridey J, Smith JW, et al. HTLV associated myelopathy in a cohort of HTLV-I and HTLV-II infected blood donors. The REDS Investigators. Neurology 1997;48:315–320. 150. Kramer A, Maloney EM, Morgan OSC, et al. Risk factors and cofactors for HAM/TSP in Jamaica. Am J Epidemiol 1995;142:1212–1220. 151. Gout O, Baulac M, Gessain A, et al. Rapid development of myelopathy after HTLV-I infection acquired by transfusion during cardiac transplantation. NEJM 1990;322:383–388. 152. Kurosawa M, Machii T, Kitani T, et al. HTLV-I associated myelopathy (HAM) after blood transfusion in a patient with CD2+ hairy cell leukemia. Am J Clin Pathol 1991;95:72–76. 153. Osame M, Janssen R, Kubota H, et al. Nationwide survey of HTLV-Iassociated myelopathy in Japan: association with blood transfusion. Ann Neurol 1990;28:50–56. 154. Reddy EP, Sandberg-Wollheim M, Mettus R, et al. Amplification and molecular cloning of HTLV-I sequences from DNA of multiple sclerosis patients. Science 1989; 243:529–533. Erratum in: Science 1989; 246:246. 155. Madden DL, Mundon FK, Tzan NR, et al. Serologic studies of MD patients, controls, and patients with other neurologic diseases: antibodies to HTLV-I, II, III. Neurology 1988;38:81–84. 156. Richardson JH, Wucherpfennig KW, Endo N, et al. PCR analysis of DNA from multiple sclerosis patients for the presence of HTLV-I. Science 1989;246:821–824. 157. Hjelle B, Appenzeller O, Mills R, et al. Chronic neurodegenerative disease associated with HTLV-II infection. Lancet 1992;339:645–646. 158. Sugimoto M, Nakashima H, Watanabe S, et al. T-lymphocyte alveolitis in HTLV-I-associated myelopathy. Lancet 1987;2(8569):1220. 159. Mochizuki M, Watanabe T, Yamaguchi K, et al. HTLV-I uveitis: a distinct clinical entity caused by HTLVI. Jpn J Cancer Res 1992;83:236–239. 160. Morgan OS, Rodgers-Johnson P, Mora C, Char G. HTLV-I and polymyositis in Jamaica. Lancet 1989;2(8673):1184–1187. 161. Kitajima I, Maruyama I, Maruyama Y, et al. Polyarthritis in human T lymphotropic virus type I-associated myelopathy. Arthritis Rheum 1989;32:1342–1344. 162. Murphy EL, Glynn SA, Fridey J, et al. Increased incidence of infectious diseases and neurologic abnormalities during prospective follow-up of HTLV-II and -I infected blood donors. Arch Intern Med 1999;159: 1485–1491. 163. LaGrenade L, Hanchard B, Fletcher V, et al. Infective dermatitis of Jamaican children: a marker for HTLV-I infection. Lancet 1990;336: 1345–1347. 164. Stuver SO, Okayama A, Tachibana N, et al. HCV infection and liver cancer mortality in a Japanese population with HTLV-I. Int J Cancer 1996;67:35–37. 165. Orland JR, Wang B, Wright DJ, et al. Increased mortality associated with HTLV-II infection in blood donors: a prospective cohort study. Retrovirology 2004;1:4–12. Available at http://www.retrovirology. com/content/1/1/4. 166. Beilke MA, Theall KP, O’Brien M, et al. Clinical outcomes and disease progression among patients coinfected with HIV and human T lymphotropic virus types 1 and 2. Clin Infect Dis 2004;39:256–263.

167. Hershow RC, Galai N, Fukuda K, et al. An international collaborative study of the effects of coinfection with human T-lymphotropic virus type II on human immunodeficiency virus type 1 disease progression in injection drug users. J Infect Dis 1996;174:309–317. 168. Okochi K, Sato H, Hinuma Y. A retrospective study on transmission of adult T cell leukemia virus by blood transfusion: seroconversion in recipients. Vox Sang 1984;46:245–253. 169. Kamihira S, Nakasima S, Oyakawa Y, et al. Transmission of human T-cell lymphotropic virus type I by blood transfusion before and after mass screening of sera from seropositive donors. Vox Sang 1987; 52:43–44. 170. Manns A, Wilks RJ, Murphy EL, et al. A prospective study of transmission by transfusion of HTLV-I and risk factors associated with seroconversion. Int J Cancer 1992;51:886–891. 171. Donegan E, Lee H, Operskalski EA, et al. Transfusion transmission of retroviruses: human T-lymphotropic viruses types I and II compared with human immunodeficiency virus type 1. Transfusion 1994;34:478–483. 172. Kleinman S: Donor selection and screening procedures. In Nance SJ (ed). Blood Safety: Current Challenges. Bethesda, Md., American Association of Blood Banks, 1992, pp 169–200. 173. Hjelle B, Wilson C, Cyrus S, et al. Human T-cell leukemia virus type II infection frequently goes undetected in contemporary U.S. blood donors. Blood 1993;81:1641–1644. 174. Kleinman S, Kaplan J, Khabbaz, et al. Evaluation of a p21e-spiked Western blot (Immunoblot) in confirming human t-cell lymphotropic virus type I and II infection in volunteer blood donors. J Clin Microbiol 1994;32:603–607. 175. Lal RB, Brodine S, Kuzura J, et al. Sensitivity and specificity of a recombinant transmembrane glycoprotein (rgp21)-spiked Western immunoblot for serologic confirmation of human T-cell lymphotropic virus type I and type II infection. J Clin Microbiol 1992;30:296–299. 176. Brodine SK, Kaime EM, Roberts C, et al. Simultaneous confirmation and differentiation of human T-lymphotropic virus types I and II infection by modified Western blot containing recombinant envelope glycoproteins. Transfusion 1993;33:925–929. 177. Lal RB, Heneine W, Rudolph DL, et al. Synthetic peptide-based immunoassays for distinguishing between human T-cell lymphotropic virus type I and type II infections in seropositive individuals. J Clin Microbiol 1991;29:2253–2258. 178. Lal RB, Rudolph DL, Lairmore MD, et al. Serologic discrimination of human T cell lymphotropic virus infection by using a synthetic peptide-based enzyme immunoassay. J Infect Dis 1991;163:41–46. 179. Viscidi RP, Hill PM, Li S, et al. Diagnosis and differentiation of HTLV-I and HTLV-II infection by enzyme immunoassays using synthetic peptides. J AIDS Hum Retrovirol 1991;4:1190–1198. 180. Busch MP, Laycock M, Kleinman SH, et al. Accuracy of supplementary serological testing for human T-lymphotropic virus (HTLV) types I and II in US blood donors. The Retrovirus Epidemiology Donor Study. Blood 1994;83:1143–1148. 181. Calattini S, Chevalier SA, Duprez R, et al. Discovery of a new human T-cell lymphotropic virus (HTLV-3) in Central Africa. Retrovirology 2005;2:30. 182. Wolfe ND, Heneine W, Carr JK, et al. Emergence of unique primate T-lymphotropic viruses among central African bushmeat hunters. Proc Natl Acad Sci USA 2005;102:7994–7999. 183. Patience C, Takeuchi Y, Weiss RA. Infection of human cells by an endogenous retrovirus of pigs. Nat Med 1997;3:282–286. 184. Heneine W, Tibell A, Switzer WM, et al. No evidence of infection with porcine endogenous retrovirus in recipients of porcine islet-cell xenografts. Lancet 1998;352:695–699. 185. Heneine W, Switzer WM, Soucie JM, et al. Evidence of porcine endogenous retroviruses in porcine factor VIII and evaluation of transmission to recipients with hemophilia. J Infect Dis 2001;183:648–652. 186. Busch MP, Holland PV. Idiopathic CD4+ T-lymphocytopenia (ICL) and the safety of blood transfusions: What do we know and what should we do? Transfusion 1992;32:800–804. 187. Busch MP, Switzer WM, Murphy EL, et al. Absence of evidence of infection with divergent primate T-lymphotropic viruses in United States blood donors who have seroindeterminate HTLV test results. Transfusion 2000;40:443–449. 188. Mayo DJ, Rose AM, Matchett SE, et al. Screening potential blood donors at risk for human immunodeficiency virus. Transfusion 1991; 31:466–474. 189. Johnson ES, Doll LS, Satten GA, et al. Direct oral questions to blood donors: the impact on screening for human immunodeficiency virus. Transfusion 1994;34:769–774. 190. Petersen LR, Doll LS, White CR, et al. Heterosexually acquired human immunodeficiency virus infection and the United States blood supply:

192. 193. 194. 195.

196. Flamholz R, Jeon HR, Baron JM, et al. Study of three patients with thrombotic thrombocytopenic purpura exchanged with solvent/detergent-treated plasma: Is its decreased protein S activity clinically related to their development of deep venous thromboses? J Clin Apher 2000; 15:169–172. 197. Yarranton H, Cohen H, Pavord SR, et al. Venous thromboembolism associated with the management of acute thrombotic thrombocytopenic purpura. Br J Haematol 2003;121:778–785. 198. Clark KA, Kataaha P, Mwangi J, et al. Predonation testing of potential blood donors in resource-restricted settings. Transfusion 2005;45: 130–132.

HIV, HTLV, AND OTHER RETROVIRUSES

191.

considerations for screening of potential blood donors. HIV Blood Donor Study Group. Transfusion 1993;33:552–557. Eastlund T: Monetary blood donation incentives and the risk of transfusion-transmitted infection. Transfusion 1998;38:874–882. Munsterman KA, Grindon AJ, Sullivan J, et al. Assessment of motivations for return donation among deferred blood donors. Amer Red Cross ARCNET Study Group. Transfusion 1998;38:45–50. Council of Europe. Expert Committee in Blood Transfusion. Study Group on Pathogen Inactivation of Labile Blood Components. Pathogen inactivation of labile blood products. Transfus Med 2001;11:149–175. Klein HG, Dodd RY, Dzik WH, et al. Current status of solvent/detergent-treated frozen plasma. Transfusion 1998;38:102–107. Allain JP, Bianco C, Blajchman MA, et al. Protecting the blood supply from emerging pathogens: the role of pathogen inactivation. Transfus Med Rev 2005;19:110–126.

45 617

Chapter 46

Human Herpesvirus Infections John D. Roback

INTRODUCTION

III 618

The Herpesviridae is a family of approximately 100 viruses with common structural features. Each virus has a linear 120- to 230-kb double-stranded DNA genome maintained in a toroidal conformation and surrounded by an icosadeltahedral nucleocapsid. The 100 nm diameter capsid, composed of 162 capsomeres, is encompassed by a dense tegument or matrix and an outer trilaminar lipid envelope that contains proteins of both viral and host cell origins.1 Herpesviruses share a number of biologic characteristics as well, including expression of viral enzymes that participate in DNA synthesis and nucleic acid metabolism, confinement of viral DNA synthesis and packaging to the host cell nucleus, destruction of the infected cell during active viral replication, and capacity to remain in a latent state indefinitely.1 Of the herpesviruses, eight are known to infect humans (Table 46–1). Members of the human herpesvirus (HHV) family are categorized into three subfamilies based on biologic properties including cell tropism, genome structure, and sequences of conserved open reading frames (ORFs). Most have a commonly employed name, such as cytomegalovirus (CMV); each is also known by an accompanying HHV designation, according to guidelines from the International Committee on Taxonomy of Viruses (e.g., CMV is HHV-5).2 This chapter focuses primarily on CMV, the herpesvirus with most clinical relevance to transfusion medicine, along with discussion of other leukocytotropic herpesviruses (EpsteinBarr virus [EBV], HHV-6, HHV-7, and HHV-8) that may contaminate blood components.

CYTOMEGALOVIRUS (HHV-5) Molecular and Cellular Virology of CMV Viral Structure CMV was the first identified betaherpesvirus and remains the prototype of this group. The CMV virion contains a linear double-stranded DNA genome of approximately 230 kbp in length, the largest of the herpesviruses.3,4 The genome is divided into unique long (UL) and unique short (US) segments, each flanked by a pair of inverted repeat regions.4 The UL and US segments can each independently invert with respect to one another, yielding four different genomic isomers. After infection, the termini of the linear genome are joined to produce a circular, or concatemeric, replicative form.5 Circularized CMV genomes have also been identified in latently infected peripheral blood CD14+ leukocytes.6 At least 200 ORFs have been identified within the CMV genome,

many of which encode viral proteins of still unknown function.4 After translation, viral proteins may undergo modifications, including phosphorylation, glycosylation, and cleavage. Mature virions range from 150 to 200 nm in diameter7 and contain approximately 30 viral proteins distributed in the capsid, tegument, and envelope. Recently, viral RNA transcripts have also been identified packaged with viral DNA in mature virions.8 This appears to be a relatively nonspecific process, mediated by viral proteins, including the tegument protein pp28, because cellular RNA transcripts are also incorporated into the virions.9,10 The function of these packaged RNA transcripts is unclear, but they may allow viral protein translation to begin immediately on infection.8 Additionally, in analogy with structural studies of retroviruses,11 packaged RNA transcripts may also be important to the integrity of the viral particles. Biology of Infection VIRAL–CELLULAR INTERACTIONS

CMV can infect a range of cell types, including those of endothelial, epithelial, mesenchymal, hematopoietic, and neuronal lineages, frequently causing characteristic cellular enlargement (cytomegalia).4,12 Infection appears to involve three sequential steps: viral attachment to the target cell, fusion of the viral and cellular membranes, and penetration of the viral capsid into the cell. Each step may require interactions between multiple viral coat proteins and cellular membrane proteins, the latter serving the role of viral receptors. The observation that the CMV genome contains at least 54 ORFs that encode putative viral membrane glycoproteins13 hints at the potential complexity of these processes. Viral attachment, the formation of a low-affinity dissociable viral–cellular interaction, involves interactions between viral proteins and cell surface receptors. One receptor class consists of ubiquitously expressed cell surface heparan sulfate proteoglycans, which appear to interact with the viral glycoprotein M/N (gM/gN) and/or gB envelope complexes and are required but not sufficient for infection.14–18 The viral gB complex (sometimes called gcI, UL55, or gp130/55) is a disulfide-linked complex composed of 120- to 130-kDa and 55- to 60-kDa viral glycoproteins. The most abundant component in the viral envelope, gB is also a prominent target of neutralizing antibodies produced during natural infection.15 gB binds to multiple cellular receptors. The epidermal growth factor receptor (EGFR) appears to be one member of this group of receptors.19 Viral gB stably docks with EGFR and can displace EGF from its binding site. CMV infection leads to EGFR phosphorylation, Akt activation, and calcium mobilization. Furthermore, cell lines that did not express

Subfamily

HHV Designation

Common Name

Alphaherpesvirinae

HHV-1 HHV-2 HHV-3 HHV-5 HHV-6A, -6B HHV-7 HHV-4 HHV-8

Herpes simplex-I (HSV-1) Herpes simplex-2 (HSV-2) Varicella-zoster virus (VZV) Cytomegalovirus (CMV) — — Epstein-Barr virus (EBV) Kaposi’s sarcoma-associated herpesvirus (KSHV)

Betaherpesvirinae

Gammaherpesvirinae

EGFR were resistant to CMV infection.19 Because hematopoietic target cells do not express EGFR, other unidentified docking receptors for gB are probably involved. gB also activates Toll-like receptors, conserved cellular membrane proteins involved in initiating antipathogen responses.20 Although attachment complexes are initially dissociable, they can be stabilized by membrane fusion. After attachment, interactions between gB and its receptors may initiate fusion. In addition, the viral protein complex gp86 (gcIII, UL75:UL115, or gH:gL) appears to be involved in these processes.21 Evidence indicates that gp86 is composed of the viral gH and gL proteins, as well as a third component not yet identified.22 The cell membrane receptor for gp86 has been identified as a 92.5-kD constitutively phosphorylated glycoprotein.23,24 Binding of gB and gp86 to the target cell initiates signaling processes that may be important to CMV infection. Interactions between gB and its cellular receptor(s) activates signal transduction through the interferon-response pathway, leading to induction of the interferon-responsive genes OAS and ISG54,25 and binding of gp86 to the cellular 92.5-kD protein can alter intracellular calcium concentration.26 After fusion, the viral capsid penetrates into the host cell and releases viral DNA. The molecules mediating these processes have not been identified. Additional proteins have also been hypothesized to mediate viral infection. For example, cellular human leukocyte antigen (HLA) class I proteins may promote viral attachment through interactions with β2microglobulin attached to the virion membrane.27 VIRAL LIFE CYCLE

Active and Latent Infections After the steps of viral attachment, fusion, and penetration, active infection occurs if the target cell is permissive for the complete sequence of viral gene expression, viral genome replication, and production of progeny virions. During active infection, viral genes are expressed in coordinated waves. Three distinct kinetic classes of viral genes have been identified: immediate early (α), followed by delayed early (β) and then late (γ).4 α-Class gene transcription is controlled by a combination of constitutively expressed host cell proteins and viral proteins present in the infecting virion. Thus, α genes can be transcribed in the presence of pharmacologic inhibitors of protein synthesis. Viral α proteins, in turn, are required for expression of viral genes of the β and γ classes.28–30 The β protein products perform viral DNA replication and metabolic functions; the γ genes encode structural proteins required for assembly of progeny virions. Finally, mature virions are transported through the Golgi apparatus and are released

from infected cells by exocytosis,31 eventually resulting in host cell destruction. Cytomegalovirus may also assume a latent state when it infects target cells that are not permissive for viral replication. Latency, the presence of viral DNA in an infected cell in the absence of active viral replication, may persist indefinitely because the host cell is not destroyed by the virus. The latent CMV genome retains the capacity to reactivate viral gene expression, produce infectious virions, and enter lytic growth at a later time. Studies with human and murine CMV have demonstrated that latency can be established in hematopoietic cells, primarily those of the granulocyte-monocyte lineage, as well as in endothelial cells.32–38 The possibility of CMV latency in other cell types has not been excluded. The molecular mechanisms regulating CMV latency and reactivation from latency have not been completely elucidated. In the herpesvirus EBV, a clearly defined set of viral proteins have been identified that control latency and reactivation.39 Similarly, CMV latencyassociated transcripts (LATs) have been detected in 0.01% to 0.001% of sorted CD33+ lineage-committed hematopoietic progenitors from the peripheral blood of naturally infected individuals.36,40 CMV LATs are transcribed from the viral IE gene locus and encode immunogenic viral proteins that are targets of naturally arising antibodies in CMV-seropositive individuals.32,40,41 The function of LATs is currently unknown. Circularization of the CMV genome is associated with latency in CD14+ peripheral blood monocytes in CMV-seropositive individuals,6 a phenomenon also seen during latent infections with other herpesviruses.42–44 The possibility of persistent viral infections, an intermediate state between active and latent infection in which low levels of virus are produced, remains a topic of considerable debate.45,46 Viral Genes Important to Pathogenesis In addition to viral proteins that control expression of viral genes and virion assembly and provide structural support for the viral particle, CMV also encodes proteins that favor viral replication at the expense of host cell metabolism and disrupt the host’s ability to combat viral infection. For example, CMV infection alters the expression, accumulation, and activity of the cellular tumor suppressor proteins, cyclins, and cyclin-associated kinases. These alterations in the cellcycle machinery act to simultaneously promote progression toward the G1/S transition but prevent cellular DNA synthesis and cell division, resulting in cell-cycle arrest and cellular aneuploidy. It has been hypothesized that in the arrested state cellular DNA synthesis would be blocked but the cellular milieu would contain abundant nucleotides and other metabolic precursors that could support viral replication.47–50 One viral protein involved in this process is the immediate-early IE1

HUMAN HERPESVIRUS INFECTIONS

Table 46–1 Human Herpesviruses

46 619

TRANSFUSION MEDICINE

III 620

72-kDa protein, which complexes with the cellular Rb-related protein p107 and blocks its ability to repress E2F-responsive promoters.51 The IE1-mediated derepression at the level of E2F, in turn, allows expression of cellular genes that promote cell-cycle progression.51 CMV infection also activates cyclindependent kinase 2 (CDK2), a cellular protein that controls progression through the G1 and S phases of the cell cycle. The importance of CDK2 to viral replication was illustrated by blocking CDK2 activity with either a dominant-negative mutant or the pharmacologic inhibitor roscovitine. In both cases, inhibition of CDK2 activity prevented CMV replication and production of progeny virus.50 Cytomegalovirus has also developed mechanisms to interfere with antiviral immune function (reviewed in Hengel and colleagues52 and Reddehase53). The UL37 ORF of CMV encodes vMIA, an anti-apoptotic protein that localizes to the mitochondria and protects infected cells from immune-mediated apoptosis by blocking the effects of Fas, tumor necrosis factor receptor-1 (TNFR-1), and granzyme B.54 Monocytes are a prominent site of CMV latency, and monocyte-derived macrophages can support active CMV replication. During differentiation to macrophages, CMV in monocytes displays delayed replication kinetics and viral particles are retained in the Golgi apparatus, which may facilitate immune evasion until sufficient progeny virions have been produced.55 In contrast, in patients with compromised antiviral immunity, CMV can replicate with rapid kinetics, displaying viral doubling times approaching 24 hours.56 Despite sophisticated viral mechanisms of immune evasion, clinical and experimental evidence demonstrates that the competent immune system can effectively suppress viral replication. For example, the murine CMV gp40 and gp48 glycoproteins (ORFs m152 and m06, respectively) can decrease expression of cellular class I major histocompatibility complex (MHC) proteins during infection of fibroblasts and thus decrease CMV antigen presentation to CD8+ T cells. However, the significance of these mechanisms during natural infection are unclear because CMV infection does not disrupt class I MHC expression and antigen presentation in macrophages, the professional antigen-presenting cell most important in initiating the anti-CMV immune response.57 Furthermore, although human CMV has also evolved mechanisms to interfere with antigen presentation by infected cells,58,59 the immune system circumvents these obstacles by utilizing structural proteins in the infecting viral particle as immunodominant epitopes for an immune response.60,61 Thus, the immune response can be initiated before expression of antiviral proteins that halt antigen presentation by the infected cell. CMV has also developed strategies to interfere with interferon-γ (IFN- γ) signals that normally upregulate MHC expression during viral infection.62 Interestingly, downregulation of class I MHC cell surface expression by CMV should lead to destruction of the infected cells by host natural killer (NK) cells.63 However, expression of the viral class I MHC homologue m144 by murine CMV decreases the susceptibility of the infected cell to NK-mediated lysis.64

CMV Infection, Immune Response, and Diagnosis Transmission, Prevalence, and Epidemiology During CMV infection, active viral replication results in shedding of infectious virions into plasma and bodily fluids,

including saliva, tears, breast milk, urine, stool, and semen. Community-acquired CMV infection is usually the result of close contact with a person shedding CMV. The incidence of community-acquired CMV infection varies with the study population. For example, the yearly CMV seroconversion rate in health care workers has been estimated at 0.6% to 3.3%,65 similar to rates of 2.0% to 6.3% reported in middleclass women during and between pregnancies.66 In contrast, rates as high as 13% per year have been observed in adolescents.67 In blood donors, the CMV seroconversion rate is estimated at approximately 1% per year.68 Most studies have shown that 50% to 80% of the population is CMV seropositive,4 although the incidence can be higher in some urban populations and lower in some groups of blood donors.69 Most individuals contracting community-acquired CMV infection are immunocompetent, and the infection is often asymptomatic. However, a mild self-limited infectious mononucleosis syndrome can occur, with symptoms that include fever, malaise, hepatosplenomegaly, and a rash.70 CMV can be isolated from bodily secretions during the symptomatic phase. The infected individual mounts both a humoral and cell-mediated immune response and viral symptoms rapidly resolve, leading to a complete recovery. However, despite effective control of CMV infection by the competent host immune system, the virus is not completely eliminated but instead becomes latent. Transplacental transmission of CMV to a developing fetus is an important viral cause of birth defects.71,72 Fetal infection occurs in 40% to 50% of cases in which a seronegative mother contracts a primary CMV infection during pregnancy.73,74 CMV disease occurs in 5% to 15% of the infected infants, presenting most often with intrauterine growth retardation, deafness, mental retardation, blindness, and thrombocytopenic bleeding.72,73 However, when mothers are seropositive before pregnancy, maternal antiviral immunity can limit congenital CMV infection and disease. For example, in one study of seropositive mothers the rate of vertical transmission was approximately 1%.73 Furthermore, no cases of symptomatic CMV infections were seen in 27 congenitally infected infants born to seropositive mothers.73 Cytomegalovirus can also be transmitted by blood transfusion or transplantation of hematopoietic stem cells and solid organs from infected donors. When the recipients are immunocompromised, CMV transmission through these mechanisms can produce serious clinical consequences. Prevention of CMV infection is an important concern in transfusion medicine. CMV Infection of Peripheral Blood and Marrow Cells CELL TROPISM

From the perspective of transfusion medicine, the most important target cells of CMV infection are peripheral blood leukocytes and their progenitors. Under appropriate conditions, these cell types can either harbor latent CMV or allow active viral replication, and thus are well-suited to mediate transfusion-transmitted CMV infection. CMV infection of bone marrow hematopoietic progenitor cells likely occurs during primary infection75 (Fig. 46–1). Most evidence suggests that these cells restrict viral replication but support viral latency,32,33 although some studies have shown low levels of CMV replication in bone marrow– derived cells in culture.75,76 CMV DNA has been identified

Allogeneic cellular interactions

CD34⫹ self-renewing stem cells

CD33⫹ lineagecommitted myeloid progenitor

Cytokines

Tissue macrophage

Monocyte

HUMAN HERPESVIRUS INFECTIONS

Figure 46–1 Hypothetical model integrating long-term latency of cytomegalovirus (CMV) in the hematopoietic compartment with transmission of CMV by blood transfusion and hematopoietic stem cell transplantation. CMV infects CD34+ multipotent progenitors during primary infection, and latently infected cells retain the viral genome during selfrenewal. Committed progenitors in the marrow may also be directly infected with CMV. Either cell type may transmit CMV to a seronegative transplant recipient. CMV remains latent as CD33+ progenitors differentiate into circulating peripheral blood monocytes. Latently infected monocytes subsequently differentiate into tissue macrophages, either in the original host or after transfusion into a recipient. The allogeneic or cytokine-mediated signals monocytes encounter during differentiation render them permissive to CMV reactivation and viral replication. See text for detailed discussion.

Latent CMV

CMV

CMV CMV

VASCULATURE BONE MARROW

CMV reactivation PERIPHERAL TISSUES

CMV

by polymerase chain reaction (PCR) in sorted multipotent CD34+ progenitor cells from the bone marrow of seropositive, and in some cases seronegative, donors.33,36,77,78 Because of their capacity for self-renewal, latently infected hematopoietic progenitor cells represent a potential long-term reservoir of latent virus. In fact, when CMV-infected CD34+ cells are grown in suspension cultures, transfer of CMV DNA to progeny cells during mitosis has been demonstrated.34 CMV DNA has also been identified in myeloid-lineage-committed CD33+ progenitor cells in the marrow or mobilized into the peripheral blood by granulocyte-macrophage colony-stimulating factor. Lineage-committed progenitors appear to be latently infected, as indicated by the presence of CMV LATs in 0.01% to 0.001% of sorted CD33+CD14+ or CD33+CD15+ bone marrow cells from seropositive donors.36 These findings support a model for latency in which early hematopoietic progenitor cells are latently infected during primary CMV infection and thereafter serve as viral reservoirs. Furthermore, latently infected marrow progenitor cells are a likely vector for transmission of CMV infection by hematopoietic stem cell transplantation. As CD33+ progenitors continue to differentiate they enter the peripheral blood. Monocytes appear to retain latent virus, but as they differentiate into macrophages CMV replication with production of progeny virus has been observed35,79 (see Fig. 46–1). Cells of the monocytic lineage have been hypothesized to mediate transfusion-transmitted CMV, but the prevalence of latently infected monocytes in the peripheral blood appears to be low. It has been estimated that 0.004% to 0.01% of peripheral blood mononuclear cells (PBMCs) mobilized by granulocyte colony-stimulating factor38 and 0.01% to 0.12% of PBMCs from healthy seropositive blood donors35 contain CMV DNA, with a range of 2 to 13 viral genomes per infected cell.38 Because approximately 5% of PBMCs are monocytes, latently infected monocytes may comprise only 1 to 25 of every million peripheral blood white blood cells (WBCs). The low numbers of latently

infected leukocytes in transfused blood components may contribute to the variable incidence of transfusion-transmitted CMV observed clinically. Cytomegalovirus is also found associated with other cell types in the peripheral blood and marrow. In immunocompromised patients with CMV infections, polymorphonuclear leukocytes (PMNs) phagocytose and contain large amounts of virus.80 Although PMNs do not appear to support the complete viral replication cycle, they can retain CMV in a viable and infectious form under experimental circumstances.81 CMV can also productively infect megakaryocytic precursors and mature megakaryocytes.82 Plasma free virus appears to be less stable than intracellular virus, and the presence of free virus in plasma is usually transient.83 For example, in one study of recently infected adolescents, only a minority (25% to 40%) had plasma viremia, which was rarely identified more than 4 months after seroconversion.67 Based on the available evidence, these other peripheral blood sources of CMV are unlikely to be as important as latently infected monocytes to the pathogenesis of transfusion-transmitted CMV. VIRAL LOADS

Quantitation of peripheral blood CMV load is clinically useful in immunocompromised patients, where viral loads can reach 108 copies per milliliter of plasma or greater,84 and correlate with the severity of viral disease.85–90 For example, in liver transplant recipients a 50% probability of CMV disease was associated with a viral load of 105.1 genomes/mL of blood, and a 90% probability with 105.5 genomes/mL.87 Infectious virus can also be cultured readily from the blood of immunocompromised patients with active infections.91 In contrast, the peripheral blood viral load in CMVinfected but otherwise healthy individuals is much lower and rarely quantitated. For example, in a series of published studies CMV could be cultured from only 2 of over 1500 buffy coat samples from healthy blood donors.70,92,93 Nonetheless,

46 621

TRANSFUSION MEDICINE

a consideration of CMV loads in healthy individuals who are potential blood donors is useful in understanding the biology of transfusion-transmitted CMV (Fig. 46–2). WBCassociated viremia is often detectable for 6 months after infection,67,94 in contrast to plasma free virus, which often disappears by 4 months. In recently seroconverted adolescents, 75% to 80% of WBC samples were positive for CMV DNA by PCR within the first 4 months of infection, compared with 25% to 40% of plasma samples within this same period.67 A study of 98 seroconverting blood donors likewise identified a low frequency of transient plasma viremia.83 In recently infected pregnant women, WBC-associated CMV DNA was detected in 100% of samples during the first month of infection using PCR and in 90% of samples during the second month of infection. None of the samples were positive after 6 months of infection.94 WBC-associated viral load decreased substantially during early infection. During the first month 60% of positive samples had viral loads of more than 10 CMV genome-equivalents (GE)/105 WBCs (range of 10 to 398), whereas only 3.3% of positive samples during the second month of infection had more than 10 GE/105 WBCs.94 Thus, shortly after infection of the immunocompetent host, the patient’s viral load peaks. The subsequent decline in viral load correlates with the development of host anti-CMV immunity. Host Immune Response HUMORAL IMMUNITY

CMV infection initiates both a humoral and cellular immune response in the host, although anti-CMV antibodies exert

III WBC CMV DNA

622

Anti-CMV IgG Anti-CMV CTL

Plasma CMV DNA Anti-CMV IgM

Window phase

Infection

2 months

4 months

6 months

Figure 46–2 Temporal relationships between detection of plasma and white blood cell (WBC)-associated cytomegalovirus (CMV) DNA and the development of an immune response after primary infection. CMV DNA can be detected by polymerase chain reaction in both the plasma and WBC peripheral blood compartments during the first month of infection, and subsequently declines to undetectable levels over 4 to 6 months. The curves are not meant to be quantitative, but rather to illustrate that WBC-associated virus is more frequently detectable than plasma CMV during primary infection and also persists for a longer time. A humoral response is usually detectable by 8 weeks postinfection, and persists indefinitely, along with a cytotoxic T-lymphocyte response. The window phase represents the period between the initial presence of CMV in the peripheral blood and the first serologic evidence of CMV infection. Seronegative blood components obtained from donors in the window phase may explain some episodes of breakthrough transfusion-transmitted CMV in patients transfused exclusively with seronegative blood. See text for detailed discussion.

limited control over CMV infection and disease. Antibody expression is typical of other humoral responses, with transient anti-CMV IgM synthesis followed by persistent expression of antiviral immunoglobulin G (IgG) (see Fig. 46–2). In a study of recently infected adolescents, anti-CMV antibodies were usually detectable by 6 to 8 weeks postinfection, a time of high peripheral blood viral loads.67 However, despite the fact that anti-CMV, anti-gB, as well as viral neutralizing antibodies could be detected during the humoral response, they were insufficient to produce a precipitous decline in CMV DNA in either the plasma (free virus) or WBCs. Plasma and WBC-associated viral DNA were present in 25% to 40% and 75% to 80% of individuals, respectively, during the first 16 weeks of infection, and could still be detected in some individuals at 48 weeks of infection. Development of an antibody response likewise failed to immediately suppress viral shedding because CMV could be isolated from 59% of urine specimens during the first 80 weeks of infection.67 Consistent with these observations, infectious virus has also been identified in saliva and cervical secretions of remotely infected seropositive individuals.95,96 These findings indicate that anti-CMV antibodies, including those with neutralizing activities in vitro, may not completely prevent viral infectivity in vivo. Nonetheless, anti-CMV antibodies can protect against sequelae of CMV infections in some circumstances. In a study of neonatal CMV infection, 10 of 10 (50%) infants born to seronegative mothers who subsequently contracted transfusion-transmitted CMV developed serious or fatal CMV disease. In contrast, 32 infants born to seropositive mothers contracted CMV infections, but none developed CMV disease, suggesting that passively acquired maternal CMV antibodies abrogated disease severity.97 Therapeutic administration of antiviral antibodies, such as those present in intravenous immunoglobulin (IVIg) preparations, can also be efficacious in altering CMV disease course in some circumstances.98 Although anti-CMV antibodies generated during natural infection display a spectrum of specificities, among the more important targets is viral gB. Radioimmunoprecipitation assays using recombinant gB demonstrated the presence of anti-gB antibodies in the serum of all 48 seropositive donors tested.99 Furthermore, the anti-gB antibodies were a significant component of the CMV-neutralizing activity in the serum samples. When anti-gB antibodies were absorbed with recombinant gB protein, viral neutralizing titer in the serum was reduced an average of 48%.99 Similarly, other studies have demonstrated that 40% to 88% of serum CMVneutralizing activity in naturally infected donors is directed against gB.100,101 These results lend support for the use of recombinant gB as a subunit CMV vaccine. CELLULAR IMMUNE RESPONSE

Cellular immune responses play an important role in the control of CMV infection. In bone marrow transplant (BMT) recipients, a patient population at high risk for CMV infection, development of a class I HLA-restricted CD8+ cytotoxic T-lymphocyte (CTL) response to CMV was significantly associated with the effective control of CMV infections.102–106 In a study of 58 allogeneic BMT recipients with low or absent anti-CMV CTL activity at enrollment, 43 developed CMV infections, which were lethal in 12 of the patients.102 Detectable anti-CMV CTL activity developed in all survivors of CMV infection, but in only 2 of the 12 patients

Laboratory Diagnosis of CMV Infection Accurate detection of CMV infection enables the identification of transfusion recipients at risk for CMV infection, as well as blood donors whose components are potentially infectious. Furthermore, quantitation of the degree of viral replication is important for guiding appropriate use of antiviral therapies, such as ganciclovir, cidofovir, and foscarnet, in immunocompromised patients. The standard approach for identifying a previously infected individual is through detection of anti-CMV antibodies (Table 46–2). Serologic

assays have been developed in multiple configurations, including indirect hemagglutination, complement fixation, solid phase fluorescence immunoassay, enzyme immunoassay (EIA), latex or particle agglutination, and solid phase red cell adherence, although the first three of these techniques are no longer frequently used.116–120 These assays detect antiCMV antibodies of the IgG, and in some cases IgM and IgA, classes. Direct comparisons of the sensitivities and specificities of the latter four methodologies are difficult to perform owing to the lack of good standards. Some EIAs have stated sensitivities and specificities of approximately 99% and because of their objective readouts may have advantages over techniques such as latex agglutination. However, antiCMV antibodies may not be detected by serology until 6 to 8 weeks after primary infection,67 and serology cannot accurately identify or quantitate the extent of active CMV infection. Although viral culture can be used for these purposes, conventional tube cultures can require 2 weeks or more to yield results, and the more recently implemented shell vial methodology may still require 24 to 48 hours to detect the presence of infectious CMV.121,122 Furthermore, these assays are only quantitative in a limiting-dilution format, which is labor intensive and not suited to routine clinical use. The CMV antigenemia assay (which uses immunostaining to identify and quantitate peripheral blood leukocytes that contain CMV proteins) and CMV PCR have solved some of these problems.123–127 The antigenemia assay can be used for both viral diagnosis and surveillance. Significantly, this methodology is sensitive enough for early quantitative detection of CMV infections, allowing the institution of preemptive (presymptomatic) antiviral therapies.128,129 Qualitative PCR allows even earlier detection of CMV infection than does antigenemia.86,130–132 However, due to its sensitivity, in some earlier studies PCR displayed poor positive predictive value for identifying patients at risk for CMV disease because some patients with low but detectable viral loads did not develop disease.86,129,131 The recent introduction of quantitative PCR assays for CMV may provide a more rapid, sensitive, and specific predictor of patients at risk for CMV disease.133 For example, the results obtained with a moderately sensitive (400 copies of CMV DNA/mL) quantitative CMV PCR assay strongly correlated with results from the antigenemia method and with development of CMV disease.134 Advantages of the PCR method included reduced turnaround time, smaller sample requirements (200 μL plasma versus 3 to 5 mL blood), simplified specimen processing, improved stability of specimens before processing, and ability to test samples from patients with leukopenia.

Table 46–2 Routine Diagnostic Laboratory Methods for CMV Method

Rapid

Quantitative

Detects Active Infection

Detects Latent Infection

Differentiates Active from Latent Infection

Serology Culture Antigenemia PCR

++ No + ++

No No Yes Yes*

Yes Yes Yes Yes

Yes No No Yes

No Yes Yes Yes*

++, < 6 hours turnaround time. +, < 24 hours turnaround time. * Depending on assay; see text for details. PCR, polymerase chain reaction.

HUMAN HERPESVIRUS INFECTIONS

who succumbed to CMV disease. NK cell activity was also depressed in the patients with fatal CMV infection but not in those who controlled their infections.102,103 In another study, 10 of 20 recipients of allogeneic BMT developed CMVspecific CTL activity by 3 months post-transplantation. Six of the 10 patients who failed to develop anti-CMV CTL activity died of CMV pneumonitis, and all 10 patients with a detectable CMV CTL response were protected.104 Similar conclusions regarding the protective effects of CMV CTLs were reached with CMV-seropositive patients who underwent autologous peripheral blood stem cell or bone marrow transplantation. In these patients, whose preexisting CMV-specific CTLs were suppressed by the preparative regimen, the reappearance of anti-CMV CTL activity was positively correlated with control of CMV infections (p = 0.002).105 The investigators also noted that a CD4 T-helper cell response to CMV always preceded the reappearance of anti-CMV CTLs, and appeared to be obligatory for the CTL response.105,106 Using a murine model, investigators specifically depleted CD4+ and CD8+ T cells from the animals before experimental CMV infection to determine the contribution of each subset to antiviral immunity. These studies showed that CMV-primed CD8+ CTLs were capable of controlling CMV infections in the absence of CD4+ cells, except in salivary glands.107,108 Furthermore, the mice deficient in CD4+ cells did not make high levels of anti-CMV antibodies, indicating that when the antiviral CTL response is intact a humoral response is unnecessary for effective control of CMV infection.107 In the murine model, NK cells are also important for control of acute CMV infection.64,109 IFN-γ production appears to be one of the principal mechanisms through which CD8+ and NK cells exert this effect.109–114 The observation that cellular immunity plays a critical role in controlling CMV infection has led to successful earlystage clinical trials in which CMV-immune CTLs were adoptively transferred into immunocompromised BMT recipients.115

46 623

TRANSFUSION MEDICINE

Turnaround time for PCR testing may be reduced even further by adopting assays to the real-time format.84 Alternative nucleic acid testing methodologies, such as nucleic acid sequence-based amplification of CMV IE mRNA, may also be useful in CMV diagnosis.135,136

Transfusion-Transmitted CMV Infections Background

III 624

By the mid-1960s, a number of investigators had described an illness with clinical similarities to infectious mononucleosis occurring in patients who were exposed to blood products during cardiopulmonary bypass for open-heart surgery.137,138 Patients typically presented with fever, splenomegaly, and atypical lymphocytosis within 3 to 8 weeks of surgery, but had a negative heterophile antibody test and did not experience exudative pharyngitis or lymphadenopathy. Klemola and Kaariainen subsequently demonstrated an increase in the titer of complement-fixing anti-CMV antibodies concurrent with the illness, suggesting that the etiology was CMV infection acquired from transfused blood products.139,140 Further support for this hypothesis was provided by culturing CMV shed into the urine or blood of patients experiencing what was then called cytomegalovirus mononucleosis.141 In the ensuing years, the disease has come to be known as transfusion-transmitted cytomegalovirus infection and has been described in a wide variety of clinical circumstances, although it is of most significance in immunocompromised transfusion recipients.142–147 In most cases the diagnosis of transfusion-transmitted CMV is correlative: evidence of primary CMV infection in a previously seronegative patient who received a cellular blood component (red cells, platelets, or granulocytes) that was neither CMV-seronegative nor filtered. However, in at least one instance unequivocal molecular evidence for transfusiontransmitted CMV was provided by demonstrating identical restriction endonuclease digestion patterns of CMV isolates from a seropositive blood donor and from the neonates who were transfused with his blood component and subsequently developed CMV infection.148 Transfusion can lead to active CMV infection in the recipient by three mechanisms. The term transfusion-transmitted CMV is used to describe a primary CMV infection occurring in a seronegative recipient transfused with an infectious blood component. In contrast, reactivated CMV infection can occur when a seropositive transfusion recipient experiences reactivation of their latent CMV infection after blood transfusion from a seronegative donor. The mechanism underlying reactivated CMV infection likely involves immunomodulatory interactions between MHC mismatched leukocytes of the donor and recipient. Consistent with this mechanism, studies indicate that the incidence of CMV reactivation is independent of donor serostatus and component storage time, but increases with the volume of blood transfused.149,150 Finally, CMV superinfection (second-strain infection) occurs when a seropositive recipient contracts a new strain of CMV from an infectious blood component. The diagnosis of both reactivation and superinfection is based on a fourfold or greater rise in the titer of anti-CMV antibodies and/or renewed viral shedding in secretions of seropositive transfusion recipients.70 Although reactivation and superinfection can be distinguished from one another by restriction endonuclease genotyping of CMV strains,151

this analysis has no significant clinical implications. These three mechanisms of transfusion-associated CMV infection appear to occur with similar frequencies. A review of five early studies of transfused CMV-seropositive patients calculated a 26% cumulative incidence of CMV reactivation or superinfection (66 of 252 patients), compared to a 31% incidence (99 of 323) of transfusion-transmitted CMV in seronegative recipients reported in seven contemporary studies (reviewed in Adler70). Nonetheless, the clinical significance of transfusion-transmitted CMV overshadows that of CMV reactivation and superinfection because transfusion-transmitted CMV results in a primary infection against which the recipient has no preexisting immunologic memory. In contrast, CMV reactivation and superinfection are unlikely to cause morbidity in the transfusion recipient.152 Finally, it should be noted that although most cases of suspected transfusion-transmitted CMV result from the transfused component,153,154 a minority of cases may result from community-acquired CMV infection occurring in temporal proximity to the transfusion. Patients at Risk and Incidence Although transfusion-transmitted CMV produces a primary CMV infection, in the immunocompetent transfusion recipient it is of no more clinical significance than community-acquired CMV infection. Furthermore, the risk of transfusion-transmitted CMV is very low in these patients. Early studies showed that 1.2% or fewer of immunocompetent patients experienced transfusion-transmitted CMV.155,156 In a more recent study of 76 seronegative children with malignancies, there were no cases of transfusion-transmitted CMV in patients who received either seronegative or unscreened units, resulting in a calculated risk of less than 1 in 698 donor exposures in this population.157 It should be noted that for reasons unrelated to study design, some of the units transfused in this study were washed or filtered. Follow-up of these children revealed that 2 of 76 subsequently developed community-acquired CMV infections (1.7% per patient-year), demonstrating the relatively greater incidence of community-acquired CMV infections than transfusion-transmitted CMV in this immunocompetent population. Furthermore, even in one early study where 32% of nonimmunosuppressed patients undergoing tumor resection developed transfusion-transmitted CMV, there was no evidence of CMV disease.149 Thus, at present there are no compelling reasons to provide nonimmunosuppressed seronegative transfusion recipients with special components for the purposes of preventing transfusion-transmitted CMV. In contrast, transfusion-transmitted CMV can be an important cause of morbidity and mortality in immunocompromised patients (Table 46–3). Most studies suggest that 13% to 37% of these patients will contract CMV from transfusion of unscreened and unfiltered cellular blood components.97,158–161 The most well-established patient groups at risk for transfusion-transmitted CMV include premature low-birth-weight infants ( 1500 grams, born to seropositive mothers • All other transfusion recipients not listed above Modified with permission from Preiksaitis JK. The cytomegalovirus-”safe” blood product: Is leukoreduction equivalent to antibody screening? Transfus Med Rev 2000;14(2):112–136.

in liver function tests. The illness can progress to disseminated tissue-invasive CMV disease, including CMV hepatitis, retinitis, interstitial pneumonitis, encephalitis, and gastroenteritis, including esophagitis.163 Progression to disease is more likely in patients with elevated viral loads. CMV infections are also associated with, and may predispose to, other complications in immunocompromised patients, including graftversus-host disease (GVHD) in allogeneic marrow transplant recipients,164–166 accelerated solid organ graft rejection,167,168 and other opportunistic infections, including invasive fungal disease.163,169 Seronegative infants transfused with unscreened blood products had a 13.5% incidence of transfusion-transmitted CMV.97 When greater than 50 mL of packed red cells were transfused, the incidence of transfusion-transmitted CMV increased to 24%. Of the infants who acquired transfusiontransmitted CMV, five (50%) developed serious symptoms or fatal disease, all of them weighing less than 1200 grams.97 In other studies, seronegative neonates weighing less than 1250 to 1500 grams also experienced a high incidence of transfusion-transmitted CMV (reviewed in Preiksaitis170), likely due to their immature immune systems. It should be noted, however, that low-birth-weight infants born to seropositive mothers can also be at risk for lethal CMV infection, despite the transfer of humoral immunity.171 Marrow transplant recipients are at significant risk of morbidity and mortality from CMV infections. Up to one third of those patients who contract CMV infection can develop CMV pneumonitis, a frequently fatal complication.172 In seropositive marrow recipients, CMV infection is usually due to viral reactivation, making CMV seropositivity the most important risk factor for CMV infection and disease.172–175 CMV infection also occurs frequently in seronegative recipients of seropositive marrow.158–160 However, in

seronegative recipients of seronegative marrow or autologous transplants, transfusion is the primary mechanism for CMV infection.158–160 Solid-organ transplant recipients are also susceptible to CMV infection and disease. In contrast to marrow transplantation, the most important source of CMV infection is the donor organ, with transfusion-transmitted CMV being less significant.144,176–179 In seronegative recipients of seronegative organs, transfusion of unscreened blood products has been associated with an incidence of CMV ranging from 0 to 33%, with a cumulative incidence of approximately 9% (reviewed in Roback147 and Preiksaitis170). Among organ transplant recipients, those receiving heart, heart-lung, liver, and pancreas transplants usually require numerous transfusions and thus have an increased risk of transfusion-transmitted CMV. Early studies showed that even in heavily transfused organ transplant recipients, the use of seronegative blood products could effectively prevent transfusion-transmitted CMV.180–182 In addition to these patient populations with well-defined susceptibility to transfusion-transmitted CMV, there are other groups that may be susceptible and may also benefit from transfusion of CMV-safe blood components (see Table 46–3). For example, it is well-documented that primary CMV infection during pregnancy carries high risks of congenital fetal infection. Although there is no direct evidence that primary maternal infection resulting from transfusiontransmitted CMV can in turn lead to fetal infection,68,156 it is prudent to provide CMV-safe blood components to pregnant women who are seronegative. Because of the high incidence of CMV reactivation and infection in seropositive marrow recipients, 69% in one study,172 transfusion-transmitted CMV is not a significant concern in these patients. However, special components may be considered for seronegative patients who are candidates for BMT, including

46 625

TRANSFUSION MEDICINE

III 626

immunosuppressed oncology patients, to prevent infection before transplant. Blood Components Implicated in Transfusion-Transmitted CMV Most evidence suggests that the primary vector for transfusion-transmitted CMV is the CMV-infected leukocyte (Table 46–4). Transfusion-transmitted CMV has not been observed in patients receiving blood components that are free of WBCs, arguing that plasma free virus is not significantly involved in the pathogenesis of transfusion-transmitted CMV.183 For example, there was no evidence of transfusion-transmitted CMV in a group of 21 immunosuppressed seronegative recipients of seronegative BMT undergoing total plasma exchange, although they were exposed to an average of 47.6 +/− 19.5 units of unscreened fresh frozen plasma (FFP).184 The absence of CMV after transfusion of FFP may be due to the scarcity of plasma free virus in healthy seropositive donors, as well as neutralization of virus by anti-CMV antibodies. In contrast, there is an abundance of evidence that transfusion-transmitted CMV can be mediated by WBCs in blood components and that the incidence of transfusion-transmitted CMV correlates with the WBC load. For example, multiple studies have demonstrated transfusion-transmitted CMV after granulocyte transfusions, most often by seropositive granulocyte preparations transfused to seronegative recipients.154,172,185 Red blood cell and platelet transfusions are also known to transmit CMV infections97 (see Table 46–4). Comparison of data over the past 4 decades reveals a correlation between decreased usage of fresh blood since the 1960s and a decline in the incidence of transfusion-transmitted CMV produced by unscreened units over the same period.186 Most early evidence suggested that fresh blood (donated within 24 hours) from seropositive donors was more infectious than stored blood.145,155 For example, in the initial descriptions, transfusion-transmitted CMV typically occurred following open-heart surgery in which the patient was exposed to fresh whole blood.138–140 In nonimmunosuppressed seronegative transfusion recipients, 6 of 7 patients who seroconverted were transfused with fresh whole blood. In contrast, among 585 patients who did not seroconvert, only 53 had received fresh blood (p < 0.001).155 Similar findings were made in a pediatric study where transfusiontransmitted CMV occurred in 13 of 15 children (87%) who received fresh blood (6 months), they are likely to have peripheral blood viral loads near or below the limits of detection of even sensitive NAT assays. Assays sufficiently sensitive to detect these low viral loads may be subject to problems, including nonspecific amplification of background DNA. For this reason, it is not surprising that several studies investigating the presence of CMV DNA in healthy blood donors have yielded conflicting results. Some investigators have identified CMV DNA in WBCs or isolated monocytes from both seropositive and seronegative donors.224–226 In a study of 270 healthy blood donors in Sweden, the use of nested PCR demonstrated CMV DNA in peripheral blood mononuclear cells from 14% of seronegative donors using both UL123 and UL32 PCR assays, although the CMV DNA detection rate increased to 55% if monocyte-enriched samples from seronegative donors were assayed.227 These investigators also identified CMV DNA in 100% of seropositive donors. In contrast, other investiga-

tors have been unable or only rarely able to identify CMV by PCR in peripheral blood from seropositive or seronegative donors.83,94,126,127,130,228–230 These latter findings appear more consistent with the clinical experience that seronegative blood components only rarely transmit CMV infections to transfusion recipients, and that only 0.4% to 12% of seropositive blood units appear capable of CMV transmission.155,156,180 The variability of the results in these prior studies demonstrate that technical hurdles remain prior to implementation of CMV NAT for testing of the blood supply. To identify CMV PCR assays with appropriate performance characteristics for screening healthy blood donors, a multicenter trial was undertaken to determine the sensitivity, specificity, and reproducibility of seven previously described assays.231 In practice, the DNA yield from 250,000 WBCs (approximately 2 μg of DNA) is the maximum tolerated input for most PCR assays. Based on estimates that 0.004% to 0.12% of PBMCs are latently infected,35,38 it was predicted that an average of 4 to 120 latently infected cells would be present per 250,000 WBCs (PBMCs typically comprise 40% of WBCs). Given the reported presence of 2 to 13 viral genomes per latently infected cell,38 it was hypothesized that PCR assays used in donor screening should be sufficiently sensitive to detect 8 to 1560 CMV genome equivalents (GE) in 250,000 WBCs. Five of the examined assays displayed sufficient sensitivity for donor screening based on consistent detection of a minimum of 25 CMV GE in analytical controls constructed to contain from 1 to 100 CMV GE in background DNA from 250,000 cells.231 Of these five assays, two detected CMV DNA in a subset of 20 pedigreed CMV seronegative samples. However, these results were found to be inconsistent when the seronegative samples were reanalyzed. The other three sensitive assays did not detect CMV DNA in the seronegative samples, and two of these were selected for further study. These assays, a nested PCR directed at the CMV UL93 ORF and the Roche Monitor assay, were used to screen 1000 blood samples from healthy donors.69 Of the 416 seropositive donor samples, 2 (0.5%) tested positive for CMV DNA with both assays, and 7 other samples tested positive with only one assay and did not confirm on retesting. When the two positive samples were subjected to limiting dilution analysis they were found to have viral loads of 1 to 10 copies per 250,000 WBCs, demonstrating that CMV DNA loads in healthy blood donors are at or below the limits of detection of even sensitive PCR assays. Six of the seronegative samples tested positive with only one assay and were not confirmed on repeat testing. These results suggest that only a small percentage of seropositive samples from healthy blood donors have viral loads that are reproducibly detectable by extremely sensitive PCR assays, potentially limiting the use of PCR to prevent transfusion-transmitted CMV. It should be noted that the minimal viral load required for transfusion-transmitted CMV has not been determined, and it must be assumed at present that any CMV-seropositive or CMV DNA-positive unit is potentially infectious. VIRAL INACTIVATION, CMV VACCINATION, AND ADOPTIVE IMMUNOTHERAPY TO DECREASE THE INCIDENCE OF TT-CMV

As opposed to current approaches such as the use of seronegative and filtered components, the application of pathogen inactivation technology to cellular blood components carries the potential for completely preventing transfusion-transmitted CMV, as well as eliminating

HUMAN HERPESVIRUS INFECTIONS

they experienced increased mortality due to other infectious complications, including Aspergillus fumigatus, Streptococcus pneumoniae, and Pneumocystis carinii pneumonia.220 In a randomized, placebo-controlled trial of prophylactic ganciclovir in marrow transplant recipients, ganciclovir usage lead to delayed recovery of anti-CMV CTL activity, which may predispose patients to late-onset CMV disease.106 Up to 15% of BMT recipients develop late CMV disease after discontinuation of ganciclovir prophylaxis,215 and the mortality of late-onset disease approaches 70%.106 Prolonged ganciclovir therapy can also lead to selection of drug-resistant CMV mutants. In one study, 38% (5 of 13) of AIDS patients treated with ganciclovir longer than 3 months shed resistant mutants.221 Children receiving T cell–depleted marrow transplants also appear to be at increased risk for developing ganciclovir-resistant mutants, as well as strains resistant to the antiviral compounds foscarnet and cidofovir.222 These concerns underscore the importance of preventing CMV transmission, rather than treating resulting CMV infections and disease.223 Furthermore, when compared to the proven efficacy of seronegative or filtered units to decrease the incidence of transfusion-transmitted CMV, the use of IVIg and/ or antiviral agents would be more expensive and possibly not as effective, and thus these interventions cannot be recommended for abrogation of transfusion-transmitted CMV.

46 629

TRANSFUSION MEDICINE

III 630

the transmission of other infectious agents. The psoralen derivative 8-methoxypsoralen (8-MOP), when exposed to ultraviolet light, can inactivate a spectrum of both grampositive and gram-negative organisms that can contaminate platelet components, including Staphylococcus aureus, Staphylococcus epidermidis, Escherichia coli, Yersinia enterocolitica, Salmonella choleraesuis, Serratia marcescens, and Pseudomonas aeruginosa.232–234 Bacterial concentrations of 104 to 107 CFU/mL can be inactivated, which compares favorably with estimated bacterial concentrations of 10 to 103 CFU/mL in naturally infected units.235 In vitro platelet function was not adversely affected by 8-MOP treatment. Similar results were obtained with another psoralen derivative, S-59, which could efficiently inactivate more than 106 plaque-forming units per milliliter of HIV, duck hepatitis B virus, and bovine diarrhea virus (a model for HCV) in experimentally infected platelet components.236–238 S-59 also effectively prevented transfusion-transmitted CMV in a murine transfusion model.195 Viral inactivation technology has entered clinical trials and is the subject of Chapter 27 in this volume. Given the limited risk of transfusion-associated CMV disease in seropositive recipients, vaccination of seronegative patients prior to immunosuppression may mitigate the risks of CMV in this population. Use of an attenuated Towne CMV strain as a live virus vaccine in 237 renal transplant recipients demonstrated that vaccinated seronegative recipients of seropositive grafts had similar rates of infection to unvaccinated controls but experienced less severe disease. Furthermore, the survival of cadaveric renal grafts at 36 months was improved by vaccination compared with the control group (p = 0.04).239 Although there were no episodes of Towne CMV reactivation following latency in these immunosuppressed patients, this possibility must be considered when evaluating the safety of this approach. The use of recombinant CMV proteins as subunit vaccines represents a potentially safer method of vaccination. In experimental settings, recombinant gB protein can induce protective immunity,101 and anti-gB antibodies in naturally infected donors comprise a large component of the CMV-neutralizing activity in seropositive donor serum.99,101 Recovery of CMV-specific CTLs following marrow transplantation provides protection from CMV infection and disease.102–104,106 However, at 40 days post-transplantation, 65% of marrow transplant recipients had deficient CD8+ CTL responses in one study.104 In immunocompromised patients who cannot mount an effective CTL response, passive transfer of CTLs is an attractive approach to augment cellular immunity. However, early attempts to restore cellular immunity through administration of unselected donor Tcells to marrow recipients was associated with GVHD.240–242 In an alternative approach, CD3+,CD8+,CD4− CMV-specific CTLs were cloned from CMV-seropositive marrow donors, expanded in culture, and then infused into marrow recipients.243 Each recipient received four infusions, 1 week apart, in escalating doses. Importantly, the CMV antigen-specific CTLs did not produce GVHD or other morbidity in the recipient.243 Further studies demonstrated that transferred clones could persist and retain anti-CMV cytotoxic activity for at least 12 weeks following infusion.244 However, CTL activity declined if recipient anti-CMV CD4+ T cells were not generated. There was no evidence of CMV viremia or disease in any patient receiving CTL therapy.244 Although adoptive cellular immunotherapy is an attractive approach

for prophylaxis of immunocompromised patients, the currently used methods for deriving and expanding donor CTL clones is lengthy, labor-intensive, and costly. However, as this technology improves and becomes more commonplace, transfusion medicine physicians will likely play a significant role in the procurement, processing, and administration of CTL components. For example, studies in a murine model have demonstrated that a rapid and simple method using psoralen-treated MHC-mismatched donor lymphocytes is highly efficacious at clearing CMV infections without causing GVHD following bone marrow transplantation.245 RECOMMENDATIONS

The exclusive use of CMV-seronegative blood components for the transfusion of seronegative immunocompromised patients susceptible to transfusion-transmitted CMV remains the standard of care (see Table 46–4). If seronegative units are not available, the evidence suggests that filtered, unscreened components are an acceptable substitute. Although the AABB guidelines state that seronegative and leukoreduced units are equivalent for prevention of transfusion-transmitted CMV,246 other panels that have reviewed this issue disagree. For example, Swiss clinical practice recommendations consider seronegative units to be the gold standard for prevention of transfusion-transmitted CMV and conclude that leukoreduced units have not yet been proven equivalent for this purpose.247 A Canadian consensus conference reached similar conclusions when the majority of the panel agreed that seronegative blood components should continue to be provided to at-risk patients despite implementation of universal leukoreduction in Canada.248 The use of frozen-deglycerolized red cells is also an acceptable alternative to seronegative units, but their preparation is expensive and labor intensive. In contrast, washed red cell and platelet components cannot be considered CMV-safe. The use of IVIg and ganciclovir or other antiviral drugs, are unlikely to be clinically efficacious or cost-effective approaches to preventing transfusion-transmitted CMV. At the present time, there is no compelling scientific argument for the implementation of CMV NAT screening to prevent transfusion-transmitted CMV. Using the most sensitive and reproducible PCR tests currently available, only about 0.2% of blood donations are found to contain CMV DNA. Furthermore, there are no studies to correlate the risk of transfusion-transmitted CMV with the presence of CMV DNA, although such an association is theoretically attractive. Nonetheless, the potential of CMV NAT to detect seronegative window phase donations, as well as components that retain a residual CMV load postfiltration, remains an interesting possibility to be evaluated. However, implementation of CMV vaccination or virus-inactivated cellular blood components in the future is likely to render the above issues moot.

EPSTEIN-BARR VIRUS (EBV, HHV-4) The genome of EBV, a gammaherpesvirus, is 172 kbp in length. Two genetically distinguishable types of EBV exist, EBV-1 and EBV-2, as well a number of variants resulting from genomic recombination. In immunocompetent patients primary infection with EBV results in a spectrum of clinical sequelae ranging from an asymptomatic infection to heterophile-positive infectious mononucleosis.39,249 EBV can

Table 46–5

B-cell viral loads.256 Although the use of EBV-seronegative blood components may reduce the incidence of transfusiontransmitted EBV, they are difficult to obtain given the high seroprevalence of EBV (Table 46–5). Because EBV is highly B-cell associated and B cells are efficiently removed by blood filtration,199 exclusive use of leukoreduced cellular blood components for transfusion is likely to decrease the probability of transfusion-transmitted EBV infections in at-risk seronegative patients. This possibility has, however, not been subject to controlled clinical trials.

HUMAN HERPESVIRUS INFECTIONS

also cause lymphomagenesis, as can the other human gammaherpesvirus HHV-8. EBV is an etiologic agent in endemic Burkitt’s lymphoma, AIDS-related lymphoma, post-transplant lymphoproliferative disease (PTLD) and nasopharyngeal carcinoma.249 After immune-mediated resolution of acute infection, EBV is not completely eliminated but rather achieves a lifelong latent state. The EBV genome is episomal in latently infected B cells, which probably represent the true EBV reservoir in the latently infected host. One in 105 to 106 B cells is estimated to be latently infected after primary infection. Latent EBV in B cells may undergo sporadic reactivation with subsequent release of infectious progeny virus. Eleven viral gene products are known to be expressed during latent infection, including nuclear antigens (EBNA 1, 2, 3A, 3B, 3C, LP), membrane proteins (LMP 1, 2A, 2B), and noncoding nuclear RNAs (EBER 1, 2).39 These latent gene products disrupt B-cell regulatory mechanisms, leading to the characteristic polyclonal B-cell lymphoproliferation seen in infectious mononucleosis. Epstein-Barr virus is transmitted by close contact, and the majority of the adult population (>95%) has been infected with EBV, based on serologic investigations. Seropositive individuals are at risk for reactivation of latent EBV infection if they become immunocompromised (e.g., in the setting of pharmacologic immunosuppression for organ or marrow transplantation).249 However, the most significant EBV-associated risk is that of primary EBV infection occurring during immunosuppression for transplantation. Primary infection can be community acquired or can result from blood transfusion or transplantation from an EBV-seropositive donor.249 An important clinical example is pediatric orthotopic liver transplantation, where EBV-associated PTLD has been identified in 10% to 20% of patients immunosuppressed with tacrolimus.250–252 PTLD encompasses a range of disorders from benign polyclonal B-cell lymphoid hyperplasia to malignant high-grade non-Hodgkin’s lymphoma. In pediatric orthotopic liver transplantation, PTLD is a cause of significant morbidity, including graft loss,253 as well as mortality rates of up to 20%.254 Transmission of EBV by blood transfusion can present in a similar manner to classic infectious mononucleosis. As with transfusion-transmitted CMV, viral genotyping has been used to document transfusion-transmitted EBV from a blood donor to a recipient, who subsequently developed EBV-driven PTLD.255 In most cases of transfusion-transmitted EBV the blood donor has been found to be in the incubation period for infectious mononucleosis with high

HUMAN HERPESVIRUS 6 (HHV-6) HHV-6 was first identified in lymphocytes from AIDS patients with lymphoproliferative diseases and was originally named human B-cell lymphotropic virus.257 However, the primary tropism of HHV-6 is now recognized to be CD4+ T lymphocytes, although it can also infect CD8+ T cells, NK cells, monocytes, macrophages, and megakaryocytes. The linear double-stranded DNA genome of HHV-6 consists of a 143 kb unique segment flanked by 8 to 13 kb terminal direct repeats. Overall, there is 66% sequence similarity between HHV-6 and CMV, the prototypical betaherpesvirus,258 and greater than 90% sequence similarity between the two HHV6 variants, HHV-6A and HHV-6B. Human herpesvirus-6B is the principal etiologic agent of the childhood illness roseola infantum (sixth disease), which presents with a fever of 2 to 5 days’ duration followed by a maculopapular skin rash (exanthem subitum) and rapid defervescence.259 Serologic investigations have demonstrated that HHV-6 is endemic, with 70% or more of children infected by age 1,260 and over 90% of the population displaying evidence of community-acquired infection during infancy. HHV-6, like other herpesviruses, can persist in a latent state, possibly within PBMCs. In addition to roseola infantum, in immunocompetent individuals there is a weak association between HHV-6 infection and diseases that include heterophile-negative mononucleosis, hepatitis, multiple sclerosis, chronic fatigue syndrome, hemophagocytic syndrome, encephalitis, Rosai-Dorfman disease, Kawasaki disease, Kikuchi lymphadenitis, sarcoidosis, a variety of lymphoproliferative disorders, and rare cases of anemia and granulocytopenia.261 However, significant problems are rare in immunocompetent individuals. In contrast, immunocompromised patients can experience HHV-6–related complications, including fever, leukopenia, graft rejection,

Transfusion Transmission of WBC-Associated Human Herpesviruses Special Components for Susceptible Recipients

Virus

Seronegative Recipients

Morbidity following Transfusion Transmission

Seronegative

Filtered

CMV EBV HHV-6 HHV-7 HHV-8

20%–60% 90% UND >90%

1–2

Minutes

>107

Low

na

Seconds

107–108

Low

Results of Routine Platelet Screening with the BacT/ALERT 3D System No. (rate) of Results

Center/ Country American Red Cross, Washington, DC/US Puget Sound Blood Center, Seattle, WA/US New York Blood Center, NY/US American Red Cross, Dedham, MA/US Community Blood Center, Kansas City, MO/US Red-Cross Flanders/Belgium Copenhagen Blood Transfusion Service Center/ Denmark Sanquin Blood Bank / The Netherlands Department of Immunology and Transfusion Medicine/Norway Héma Québec/ Canada Canadian Blood Services (CBS)/Canada

Plts*

Time of Testing (hr)†

Sample Volume (mL)

A

>24

A

>24

4

A

24

4

A

>24

A

24

4

A

22

7

B B

22 §

7 8–10

B

16–24

7.5

B

24

5–10

88 (1/419)

12 (1/3074)

A

>24

4–6

3 (1/4450)

A

>24

4–6

11 (1/1213) 28 (1/843)

Initial Positive

True Positive

False Positive‡

False Negative

Units Tested

Collection Period Reference

50 (1/1440)

13 (1/5538)

23 (1/3130)

unknown

72,000

Mar-May 2004

5 (1/2000)

11 (1/909)

unknown

10,000

June 2003– 95

5 (1/4981) 1 (1/6255)

10 (1/2490) 5 (1/1245)

unknown

24,909

unknown

6,225

Oct 2003– Apr 2004 Mar and Apr 2004

2 (1/3657)

4 (1/1828)

unknown

7,315

Sep 2003– 98 Apr 2004

181 (1/176) 622 (1/121) 34 (1/651)

41 1 (1/780) (1/31,998) 140 (1/541) 16 || (1/1385)

31,998

1999–2002 3

75,829 22,165

1999–2002 3 Jan 2002 3 –2004

2 (1/19,332)

38,664

14 (1/2635)

#

36,896

Oct 2001– Apr 3 2004 May 1998– 100 May 2004

8 (1/1668) 28 (1/843)

1 (1/13,351) 1 (1/23,617)

13,351

15

237 (1/135) 793 (1/95) 50 (1/443)

295 (1/131)

23,617

94

96 97

2002–Aug 99 2004 Mar 2004– CBS May 2005 database

*

Plt, platelets; A, apheresis platelets; B, buffy coat-derived platelets. Time of testing after blood collection. ‡ False-positive cultures include cultures that were not confirmed on reculture and false signals of the BacT/ALERT system. § Immediately after platelet production. || Screening of 2472 buffy-coat platelets on day 3 of storage revealed 6 false negatives (1/412).3 # Screening of 1061 outdated platelets yielded 2 false negatives (1/531).100 †

cantly reduces the platelet unit content. The Hong Kong Red Cross developed a pooled culture method that involves stripping of five platelet segments, which are disinfected before collecting a sample of pooled platelets that is inoculated into a BacT/ALERT culture bottle.101,102 This method was modified, validated, and adopted by one Canadian Blood Services Center.103 An adaptation to the Hong Kong method using sterile docking has also been reported.104 Similarly, the BacT/ALERT system has been validated for the detection of bacteria in WBDPs in a pooled format by Brecher and colleagues.92 Because the initial inoculum of bacteria is probably very small (21 >21 to 300 days

DAT may persist as positive; eluates may demonstrate alloantibody specificity or panagglutination

Accelerated destruction of transfused donor RBC Antibody titer increases; sensitizes donor red cells In vivo removal of antibody-sensitized donor red cells from the circulation Alloantibody binding nonspecifically to autologous RBC, or development of warm autoantibodies

DAT, direct antiglobulin test; DHTR, Delayed hemolytic transfusion reaction; RBC, red blood cell.

HEMOLYTIC TRANSFUSION REACTIONS

results due to clinically insignificant antibodies or other contaminating substances. In the case of AHTR, in which the culprit is most commonly a sampling problem en route to the laboratory, increased sensitivity in testing is unlikely to offer any real progress. Another common teaching by transfusion professionals is to avoid unnecessary transfusions, in part to reduce the fear that they may cause adverse consequences such as AHTR. Most hospitals have reduced their transfusion trigger for RBCs22 largely to handle concerns about infectious complications, so transfusion avoidance is not likely to reduce the rate of AHTR significantly. Because safe blood transfusion continues to depend on welltrained practitioners to obtain samples and administer blood products, it seems unlikely that enhanced training programs will ever totally eliminate this significant problem that is largely caused by human error. Technologic developments that might better handle this problem should be encouraged. The evolving capability to create universal group O blood by enzymatically removing A and B blood group antigens from RBC might be a technologic solution to the serious problem of AHTR.23

49 671

TRANSFUSION MEDICINE

III 672

trend of increased frequency, but decreased clinical significance, appears to be related, in part, to the manner in which DHTRs were defined. If the frequency of DHTR is determined on the basis of clinical reporting alone, then the incidence would be far lower than the true rate, because the signs and symptoms of hemolysis are nonspecific and may be difficult to distinguish from the complicated medical course of multitransfused patients.26,27 If DHTR is defined by post-transfusion serologic testing, then the frequency is likely to increase, and the clinical findings may diminish, because correlation between serologic results and clinical hemolysis is often poor.30 The definition of DHTR was clarified by Ness and colleagues30 in 1990, when the authors introduced the term delayed serologic transfusion reaction (DSTR) to describe cases with serologic evidence of DHTR (i.e., the development of a positive post-transfusion DAT result and a newly identified alloantibody in eluate studies or plasma studies or both), but no clinical evidence of hemolysis. The authors strictly defined DHTR as a subset of DSTR in which clinical evidence of hemolysis was attributable to a transfusion reaction. In their series at the Johns Hopkins Hospital, only 6 (18%) of 34 consecutive patients identified with serologic findings consistent with DHTR had clinical evidence of hemolysis associated with transfusion and could be defined as DHTR by strict definition. Twenty-eight (82%) of the 34 DSTR cases had no clinical evidence of hemolysis, as determined by retrospective chart review. The ratio of DHTR to DSTR can be expressed as 18:82 (Table 49–5). The combined frequency of DSTR and DHTR was calculated as 1:1605 (0.06%) RBC units transfused. The frequency of true DHTR was only 1 case per 9094 units (0.01%) transfused. By using the same definitions as prescribed by Ness and colleagues,30 the Mayo Clinic reported a combined frequency of DHTR/DSTR as 1 per 1899 allogeneic RBC units transfused with a DHTR/DSTR ratio of 36:64 when patients were evaluated for clinical hemolysis concurrent with the detection of positive post-transfusion serology.43 The most recent study from the Mayo Clinic spanning the years 1993 to 1998 indicates a decline in DHTR and a concomitant increase in DSTR, which the authors attributed to the implementation of a moresensitive antibody screening method and a decrease in the average length of hospital stay for inpatients.44 Interestingly, their current rates of observed reactions and the DHTR/DSTR ratio are very similar to those reported by Ness and colleagues30 in 1990 (see Table 49–5).

Diagnosis The blood bank plays an important role in recognizing and reporting potential cases of DHTR. The clinician may submit a post-transfusion sample with a request for RBC transfusions, not suspecting that the patient’s decreased hemoglobin

and hematocrit are due to DHTR. Generally, positive antibody screening tests on the post-transfusion sample trigger antibodyidentification studies and investigation of DHTR. The laboratory investigation of DHTR is outlined in Table 49–6. The serologic findings in classic cases of DHTR include the development of a post-transfusion positive DAT that may have a mixed-field appearance because of the presence of IgG antibody–sensitized donor RBCs interspersed with DAT-negative autologous RBCs and the presence of a newly identified alloantibody in the red cell eluate or plasma studies or both approximately 3 to 10 days (or usually by 20 days) after the index transfusion (see Table 49–4).26 If the patient’s transfusion history and post-transfusion serologic findings are consistent with DHTR, then the patient should be evaluated for clinical and laboratory evidence of hemolysis associated with RBC transfusions (see Table 49–6). The patient’s physician should be notified, and a transfusion reaction report should be generated so that the findings become a part of the patient’s medical records.

Serologic Complexities of DHTR Diagnostic Difficulties For many years, it was generally accepted that in DHTR, the sensitized donor red cells would be removed from the circulation within approximately 14 days (or no later than 21 days) after the putatively responsible transfusion (see Table 49–4).26 Therefore, DAT and eluate studies should become negative coinciding with the in vivo destruction of transfused donor RBCs that provoked the immune response. It is now clear from long-term studies of DHTR that it is not uncommon for DAT to persist as positive because of IgG antibody or complement sensitization or both of RBCs for many weeks or months after the transfusion reaction, long after the transfused, antigen-positive donor RBCs would be expected to be removed from the circulation.30,45 In addition, these long-term studies showed that eluates prepared from samples drawn more than 6 months after DHTR, when all the transfused donor RBCs would have been removed from circulation, may still demonstrate the alloantibody responsible for the reaction. Several theories are proposed to explain these remarkable phenomena.26,45 Salama and Mueller-Eckhardt45 suggested that alloantibody binds nonspecifically in vivo to antigen-negative autologous red cells and activates complement via the classic pathway in all DHTRs. Thus DAT may remain positive long after the removal of transfused donor RBCs because of nonspecific sensitization of autologous RBCs with alloantibody or complement or both during DHTR.45–48 Although the mechanisms remain unclear and a subject of contention, the persistence of a positive DAT and reactive eluates after DHTR does not generally appear to correlate with in vivo hemolysis.26,30 The practical application of these long-term

Table 49–5 Incidence of Delayed Hemolytic Transfusion Reaction and Delayed Serologic Transfusion Reaction Series

Incidence*

DHTR/DSTR

Ness et al30 Johns Hopkins Hospital (Jan 1986–Aug 1987) Vamvakas et al43 Mayo Clinic (1980–1992) Pineda et al44 Mayo Clinic (1993–1998)

1:1605 1:1899 1:1300

18:82 36:64 19:81

*

Incidence expressed as DHTR/DSTR per number of red cell units transfused.

Outline for Laboratory Investigation of Suspected DHTR

1. Initial Serologic Investigation Post-transfusion sample tests Antibody identification studies: Initiated as a result of positive antibody screening tests (and/or positive DAT results) on the posttransfusion sample DAT profile: (DAT with polyspecific and monospecific anti-IgG and anti-C3 antiglobulin reagents) Eluate studies: Performed if DAT positive due to IgG sensitization of RBC and history of recent RBC transfusions 2. Supplemental Serologic Tests Retrieve pretransfusion sample (if available): Perform DAT and repeat antibody screening tests Retrieve retained segments from transfused RBC donor units: Phenotype for antigen corresponding to identified antibody 3. Review Serologic Findings and Transfusion History Results evaluated by transfusion medicine attending physician 4. Generate Suspected Transfusion Reaction Report The transfusion medicine attending physician assesses the patient for clinical signs and symptoms of DHTR and notifies the patient’s physician

studies of DHTR is the recognition that serology consistent with DHTR may persist long after the reaction has occurred and the patient has recovered. Consequently, serologic findings alone can be misleading and difficult to interpret. The investigation of DHTR must consider not only the serologic findings, but also the temporal relation of antigen-positive RBC transfusions to the serologic test results, coupled with a clinical evaluation and laboratory assessment of the patient for evidence of hemolysis associated with RBC transfusions. DHTR and Autoantibody Many anecdotal cases of allo- and autoimmunization occurring concurrently have appeared in the literature over the past 40 years.49–59 For example, Lalezari and colleagues55 described a patient who had DHTR due to allo-anti-D and subsequently developed autoantibodies. The post-transfusion DAT remained positive for 6 months after DHTR, and eluates demonstrated panagglutination consistent with the presence of warm autoantibodies having a broad specificity. Worlledge54 noted as early as 1978 that alloimmunization can lead to autoimmunization and that even immunized recipients who are given compatible blood may subsequently develop a positive DAT because of autoantibodies. It is not surprising that long-term studies by Ness and associates30 found that the development of warm autoantibodies after DHTR was relatively common. Approximately one third of patients with DHTR who were followed for more than 25 days after DHTR had persistence of a positive DAT due to warm autoantibodies with broad specificity (i.e., eluate panagglutinin) (see Table 49–4). The frequency of autoantibody development after DHTR may be even higher, because autoantibodies can mimic alloantibody specificities.58,59 It is possible that DHTR cases with apparent alloantibody recovered in long-term eluate studies may actually represent cases of warm autoantibodies mimicking alloantibody specificities.30,45 Although the postreaction autoantibodies appeared to be clinically benign in the cases followed by Ness and colleagues,30 some reports indicate that DHTRs have evolved into the production of pathologic warm autoantibodies leading to severe autoimmune hemolytic anemia.49–53

Severe DHTR Fortunately, the clinical symptoms in most DHTRs generally resolve within 2 to 3 weeks of the index transfusion without medical intervention other than transfusion support with

appropriate antigen-negative blood transfusions. However, severe DHTR with life-threatening anemia has been reported and occurs most often in alloimmunized patients with sickle cell disease (SCD).60,61 Because patients with SCD have a high alloimmunization rate ranging from 17.6% to 36%, they are at greater risk for DHTR.62–67 Frequencies of DHTR in SCD range from 4% to 22%63,66,69,70 compared with the frequency of true, clinical DHTR in all transfused patients of only 0.04% (one DHTR per 2537 transfused patients)71 to 0.1% (one DHTR per 854 transfused patients).30 The diagnosis of DHTR in patients with SCD may be difficult. Patients may have symptoms that are misdiagnosed as sickle cell pain crisis, and delays in initiating appropriate treatment could lead to significant morbidity or even death.72–77 Confusion can be avoided by making certain that an accurate transfusion history is obtained on admission and that a sample is sent to the transfusion service for processing in the event that blood transfusions are required. The causes of severe DHTR in SCD are unclear and controversial. Explanations for the profound anemia observed in severe DHTR include bystander hemolysis, sickle cell hemolytic transfusion reaction syndrome, and hyperhemolysis.60,61,78,79 Bystander hemolysis has been defined as the immune destruction of autologous RBCs that may occur during DHTR. The mechanism(s) of bystander hemolysis is unclear. One theory has proposed that activation of complement could occur when alloantibodies react with transfused antigen-positive donor RBCs, leading to destruction of allogeneic and “innocent bystander” autologous red cells.80–82 Bystander hemolysis is difficult to document but may be suspected when post-transfusion reaction hemoglobin and hematocrit are less than the pretransfusion values, suggesting that simultaneous destruction of transfused donor RBCs and autologous RBCs may be present. The phenomenon of bystander hemolysis in DHTRs has been reported in patients without SCD.80–82 However, bystander hemolysis may actually be more prevalent in patients with SCD, because sickle cells have a regulatory defect in the formation of the complement membrane attack complex, causing the cells to be more susceptible to complement-mediated hemolysis.83 Garratty71 suggested that complement activation in these cases may be triggered not only by antibody reacting with transfused RBC antigens, but also by antibody reacting with other foreign antigens (e.g., human leukocyte antigen [HLA], plasma proteins). Thus the mechanism of bystander hemolysis during DHTRs in patients with

HEMOLYTIC TRANSFUSION REACTIONS

Table 49–6

49 673

TRANSFUSION MEDICINE

SCD may be similar to that of paroxysmal nocturnal hemoglobinuria (PNH), in which immune complex formation may result in hemolysis of “innocent” autologous RBCs.82 One well-documented case of bystander hemolysis was reported by King and coworkers60 in a study in which they monitored five patients who had DHTR after preoperative exchange transfusions. By monitoring the hemoglobin A (transfused allogeneic donor RBCs) and hemoglobin S (autologous sickle cells) levels, the authors showed that in one of the cases, a substantial loss of autologous RBCs occurred (i.e., bystander hemolysis), as well as loss of allogeneic RBCs during DHTR. Petz and colleagues61 reported five cases of hemolytic transfusion reactions in SCD in which the severe anemia appeared to be due to the destruction of transfused donor RBCs coincident with suppression of erythropoiesis. The authors defined these cases as “the sickle cell hemolytic transfusion reaction syndrome,” in which reticulocytopenia coupled with the hemolysis of transfused donor RBCs caused life-threatening anemia. In their series, none of the patients appeared to have accelerated autologous RBC destruction (i.e., bystander hemolysis). Hyperhemolysis is a term that has been used to describe severe DHTR in SCD in which the ongoing hemolysis of autologous cells seems to be accelerated during the course of a hemolytic reaction.78,79 Win and associates78 suggested that the accelerated destruction of autologous cells that may occur in DHTR could be due to hyperactive macrophages that readily sequester both mature RBCs and reticulocytes. Interestingly, Darabi and Dzik84 recently reported hyperhemolysis after DHTR due to anti-K1 in a patient without SCD.

III

Treatment of Severe DHTR

674

Regardless of the mechanisms that may be involved, it is important to recognize that severe DHTR in patients with SCD is not an infrequent finding and may mimic a painful crisis.72–77 Effective treatment requires prompt diagnosis and conservative transfusion support with appropriate antigennegative RBCs. Steroid therapy and intravenous immune globulin infusions have been successfully used for the treatment of the most severe cases.78,79 Some have suggested administration of recombinant erythropoietin, depending on the patient’s reticulocyte count.85

Prevention of DHTRs The frequency of DHTRs has diminished with implementation of improved pretransfusion antibody-screening methods.44 The current challenge is to find a rapid, automated antibody-screening method that is sufficiently sensitive to detect most clinically important antibodies, but not so highly sensitive that clinically benign, unwanted antibodies are detected.86 Because antibody titers may decline below detectable levels, it is important that clinically significant antibodies be well documented in the transfusion service and medical records, and it may be helpful to provide patients with personal identification cards listing the antibodies and transfusion requirements.87,88 It must be emphasized that patients diagnosed with DSTRs are at risk for severe DHTR with future transfusions and must receive donor blood negative for the antigen(s) corresponding to the implicated alloantibodies.26,30

The high rate of alloimmunization and the high risk of DHTRs in SCD appears to be due, at least in part, to the disparity in antigen frequencies between the African American recipient and white donor population.63 For example, 73% of African Americans are C negative and may develop anti-C antibodies when transfused with blood from white donors, because 70% of whites are C positive. Therefore, increasing African American donations, particularly if the donations could be linked to patients with SCD requiring transfusion support, might be of benefit. It is generally accepted that patients with SCD should be extensively phenotyped (e.g., Rh antigens [CEce], K1, MNSs, Fya, Fyb, Jka, Jkb) so that antibodies the patients may potentially produce can be easily discerned.89–91 Having the extensive or complete phenotype on file is particularly advantageous in resolving complex antibody problems that are frequently presented by multitransfused patients with SCD. DNA typing for RBC antigens is now available and can be used in cases in which phenotyping is precluded by recent RBC transfusions or by scarcity of rare antisera.92,93 Prophylactic antigen matching of donor RBCs with the recipient’s complete phenotype has been advocated for patients with SCD in an effort to reduce the rate of alloimmunization and severe DHTR.89,90,94 Various transfusion protocols that differ in the number of antigens that are considered for prophylactic matching have been examined by Castro and colleagues94 Ness91 has proposed that prophylactic antigen matching should be considered only when the patient has developed one or more RBC alloantibodies, thereby indicating that the patient is a “responder” and may be at greater risk for further alloimmunization and DHTRs with subsequent RBC transfusions. This protocol is particularly appealing because it avoids unnecessary prophylactic antigen matching of donor blood for the approximately 70% of patients with SCD who do not become alloimmunized even after multiple transfusions. Prophylactic antigen matching may prevent some cases of DHTR, but even protocols that call for extensive prophylactic antigen matching cannot prevent severe hemolytic transfusion reactions due to unusual alloantibody specificities or the development of pathologic warm autoantibodies.60,95–97

FUTURE DIRECTIONS The development of blood substitutes for transfusion and/or exchange transfusion in SCD seems to offer the most hope for preventing severe DHTR and for circumventing difficulties in providing compatible blood for patients with complex antibody problems.98,99 With the rapidly growing sophistication of DNA technology, we may be able to distinguish “responders” from “nonresponders” by genotyping, so that special transfusion protocols could be applied more appropriately. Automated technology for rapid genotyping of blood donors is already under study and may permit the provision of genotypematched donor blood to recipients in the near future.100 REFERENCES 1. Davenport RD. Hemolytic transfusion reactions. In Popovsky MA (ed). Transfusion Reactions, 2nd ed. Bethesda, Md, American Association of Blood Banks, 2001, p 1. 2. Josephson CD, Mullis NC, Van Demark C, et al. Significant numbers of apheresis-derived group O platelet units have “high-titer” anti-A/A,B: Implications for transfusion policy. Transfusion 2004;44:805–808.

32. Rauner RA, Tanaka KR. Hemolytic transfusion reactions associated with the Kidd system antibody (Jka). N Engl J Med 1967;276:1486. 33. Kurtides ES, Salkin MS, Widen AL. Hemolytic reaction due to anti-Jkb. JAMA 1966;197:816. 34. Fudenberg H, Allen FJ Jr. Transfusion reactions in the absence of demonstrable incompatibility. N Engl J Med 1957;256:1180–1184. 35. Walker PC, Jennings ER, Monroe C. Hemolytic transfusion reactions after the administration of apparently compatible blood. Am J Clin Pathol 1965;44:193–197. 36. Moore SB, Taswell HF, Pineda AA, et al. Delayed hemolytic transfusion reactions: Evidence of the need for an improved pretransfusion compatibility test. Am J Clin Pathol 1980;74:94–97. 37. Taswell HF, Pineda AA, Moore SB: Hemolytic transfusion reactions: Frequency and clinical and laboratory aspects. In Bell CA (ed). A Seminar on Immune-mediated Red Cell Destruction. Washington, D.C., American Association of Blood Banks, 1981, pp 71–92. 38. Taddie SJ, Barrasso C, Ness PM. A delayed transfusion reaction caused by anti-K6. Transfusion 1982;22:68–69. 39. Patten E, Reddi CR, Riglin H, et al. Delayed hemolytic transfusion reaction caused by a primary immune response. Transfusion 1982;22: 248–250. 40. Davey RJ, Gustafson M, Holland PV. Accelerated immune red cell destruction in the absence of serologically detectable alloantibodies. Transfusion 1983;23:40–44. 41. Baldwin ML, Barrasso C, Ness PM, et al. A clinically significant erythrocyte antibody detectable only by 51R survival studies. Transfusion 1983;23:40–44. 42. Kim HH, Park TS, Oh SH, et al. Delayed hemolytic transfusion reactions due to anti-Fyb caused by a primary immune response: a case study and review of the literature. Immunohematology 2004;20: 384–386. 43. Vamvakas EC, Pineda AA, Lreisner R, et al. The differentiation of delayed hemolytic and delayed serologic transfusion reactions: incidence and predictors of hemolysis. Transfusion 1995;35:16–32. 44. Pineda AA, Vamvakas EC, Gorden LD, et al. Trends in the incidence of delayed hemolytic and delayed serologic transfusion reactions. Transfusion 1999;10:1097–1103. 45. Salama A, Mueller-Eckhardt C. Delayed hemolytic transfusion reactions: evidence for complement activation involving allogeneic and autologous red cells. Transfusion 1984;24:188–193. 46. Chaplin H Jr. The implication of red cell-bound complement and delayed hemolytic transfusion reactions [Editorial]. Transfusion 1984; 24:185–187. 47. Salama H, Mueller-Eckhardt C. Binding of fluid phase C3b to nonsensitized bystander human red cells: A model for in vivo effects of complement activation on blood cells. Transfusion 1985;25:528–534. 48. Devine DV. Complement. In Anderson KC, Ness PM (eds). Scientific Basis of Transfusion Medicine: Implications for Clinical Practice, 2nd ed. Philadelphia, WB Saunders, 2000, pp 107–119. 49. Sosler SD, Perkins JT, Saporito C, et al. Severe autoimmune hemolytic anemia induced by transfusion in two alloimmunized patients with sickle cell disease [Abstract]. Transfusion 1989;29(Suppl):495. 50. Chaplin H Jr, Zarkowsky HS. Combined sickle cell disease and autoimmune hemolytic anemia [Abstract]. Arch Intern Med 1981;141:1091. 51. Reed W, Walker P, Haddix T, et al. Acute anemic events in sickle cell disease. Transfusion 2000;40:267–273. 52. Castellino SM, Combs MR, Zimmerman MA, et al. Erythrocyte autoantibodies in pediatric patients with sickle cell disease receiving transfusion therapy: frequency, characteristics and significance. Br J Haematol 1999;104:189–194. 53. Chan D, Poole GD, Binney M, et al. Severe intravascular hemolysis due to autoantibodies stimulated by blood transfusion. Immunohematology 1996;12:80–83. 54. Worlledge SM. The interpretation of a positive direct antiglobulin test. Br J Haematol 1978;39:157–162. 55. Lalezari P, Talleyrand NP, Wenz B, et al. Development of direct antiglobulin reaction accompanying alloimmunization in a patient with Rh(d) (D, category III) phenotype. Vox Sang 1975;28:19–24. 56. Cook IA. Primary rhesus immunization of male volunteers. Br J Haematol 1971;20:369–375. 57. Beard MEJ, Pemberton J, Blagdon J, et al. Rh immunization following incompatible blood transfusion and a possible long-term complication of anti-D immunoglobulin therapy. J Med Genet 1971;8:317–320. 58. Issitt PD, Zellner DC, Rolih SD, et al. Autoantibodies mimicking alloantibodies. Transfusion 1977;17:531–538. 59. Issitt PD, Pavone BG. Critical re-examination of the specificity of autoanti-Rh antibodies in patients with a positive direct antiglobulin test. Br J Haematol 1978;38:63–74.

HEMOLYTIC TRANSFUSION REACTIONS

3. Pineda AA, Brzica SM Jr, Taswell HF. Hemolytic transfusion reactions: recent experience in a large blood bank. Mayo Clin Proc 1978;53:378. 4. Lichtiger B, Perry-Thorton E. Hemolytic transfusion reactions in oncology patients: experience in a large cancer center. J Clin Oncol 1984;25:438. 5. Honig CL, Bove JR. Transfusion associated fatalities: review of Bureau of Biologics reports 1976–1978. Transfusion 1980;20:653. 6. Sazama K. Reports of 355 transfusion-associated deaths, 1976 through 1985. Transfusion 1990;30:583. 7. Robillard P, Itaj NK, Corriveau P. ABO incompatible transfusions, acute and delayed hemolytic transfusion reactions in the Quebec hemovigilance system-year 2000. Transfusion 2002;42:(Suppl): 25S. 8. Bluemle LW Jr. Hemolytic transfusion reactions causing acute renal failure: serologic and clinical considerations. Postgrad Med 1965; 38:484. 9. Davenport RD, Strieter RM, Kunkel SL. Red cell ABO incompatibility and production of tumour necrosis factor-alpha. Br J Haematol 1991;79:525. 10. Morat HZ, DiLorenzo NL. Activation of plasma kinin system by antigen-antibody aggregates. I: Generation of permeability factor in guinea pig serum. Lab Invest 1968;19:187. 11. Davenport RD, Strieter RM, Standiford TJ, et al. Interleukin-8 production in red blood cell incompatibility. Blood 1990;76:2439. 12. Davenport RD, Burdick MD, Streiter RM, et al. Monocyte chemoattractant protein production in red cell incompatibility. Transfusion 1994;34:16. 13. Pfueller SL, Luscher EF. Studies of the mechanisms of human platelet release reaction induced by immunologic stimuli. I: Complement-dependent and complement-independent reactions. J Immunol 1974;112:1201. 14. Davenport RD, Polak TJ, Schmouder RE, et al. Endothelial cell procoagulant activity inducted by red cell incompatibility. Transfusion 1994;34:943. 15. Goldfinger D. Acute hemolytic transfusion reactions: a fresh look at pathogenesis and considerations regarding therapy. Transfusion 1974;27:171. 16. Capon SM, Goldfinger D. Acute hemolytic transfusion reaction, a paradigm of the systemic inflammatory response: new insights into pathophysiology and treatment. Transfusion 1995;35:513. 17. Capon SM, Sacher RA. Hemolytic transfusion reactions: a review of mechanisms, sequelae, and management. J Intensive Care Med 1989;4:100–111. 18. Rock RC, Bove JR, Nemerson Y. Heparin treatment of intravascular coagulation accompanying hemolytic transfusion reactions. Transfusion 1969;9:57. 19. Rivard GE, David M, Farrell C, et al. Treatment of purpura fulminans in meningococcemia with protein C concentrate. J Pediatr 1995; 126:646. 20. Mercuriali F, Inghilleri F, Colotti MT, et al. One year use of the Bloodloc system in an orthopedic institute. Transfus Clin Biol 1994;1:227. 21. Dzik W. Emily Cooley Lecture 2002: Transfusion safety in the hospital. Transfusion 1003;43:11290. 22. Carson JL, Hill S, Carless P, et al. Transfusion triggers: a systematic review of the literature. Transfus Med Rev 2002;16:187. 23. Lenny LL, Hurst R, Goldstein J, et al. Transfusions to group O subjects of 2 units of red cells enzymatically converted from group B to group O. Transfusion 1999;34:209. 24. Mollison PL. Antibody-mediated destruction of red cells. Clin Immunol Newsl 1985;6:700–703. 25. Mollison PL. Blood Transfusion in Clinical Medicine, 9th ed. Oxford, Blackwell Scientific, 1993, pp 434–542. 26. Shirey RS, Ness PM. New concepts of delayed hemolytic transfusion reactions. In Nance SJ (ed). Clinical and Basic Science Aspects of Immunohematology. Arlington, Va., American Association of Blood Banks, 1991, pp 179–197. 27. Solanki D, McCurdy PR. Delayed hemolytic transfusion reactions: an often missed entity. JAMA 1978;239:729–731. 28. Holland PV, Wallerstein RO. Delayed hemolytic transfusion reaction with acute renal failure. JAMA 1968;204:1007–1008. 29. Pineda AA, Taswell HF, Brzica SM Jr. Delayed hemolytic transfusion reaction: An immunologic hazard of blood transfusion. Transfusion 1978;18:1–7. 30. Ness PM, Shirey RS, Thoman SK, et al. The differentiation of delayed serologic and delayed hemolytic transfusion reactions: incidence, long-term serologic findings and clinical significance. Transfusion 1990;30:688–693. 31. Boorman KE, Dodd BE, Loutit JF, et al. Some results of transfusion of blood to recipients with “cold” agglutinins. Br Med J 1946;i:751.

49 675

TRANSFUSION MEDICINE

III 676

60. King KE, Shirey RS, Lankiewicz MW, et al. Delayed hemolytic transfusion reactions in sickle cell disease. Transfusion 1997;37:376–381. 61. Petz LD, Calhoun L, Shulman IA, et al. The sickle cell hemolytic transfusion reaction syndrome. Transfusion 1997;37:382–392. 62. Shirey RS, King KE. Alloimmunization to blood group antigens. In Anderson KC, Ness PM (eds). Scientific Basis of Transfusion Medicine: Implications for Clinical Practice, 2nd ed. Philadelphia, WB Saunders, 2000, pp 393–400. 63. Vichinsky EP, Earles PNP, Johnson RA, et al. Alloimmunization in sickle cell anemia and transfusion of racially unmatched blood. N Engl J Med 1990;322:1617–1621. 64. Rosse WF, Gallagher D, Kinney TR, et al. Cooperative study of sickle cell disease: transfusion and alloimmunization in sickle cell disease. Blood 1990;76:1431–1437. 65. Orlina AR, Unger PJ, Koshy M. Post-transfusion alloimmunization in patients with sickle cell disease. Am J Hematol 1978;5:101–106. 66. Cox JV, Steane E, Cunningham G, et al. Risk of alloimmunization and delayed hemolytic transfusion reactions in patients with sickle cell disease. Arch Intern Med 1988;148:2485–2489. 67. Ambruso DR, Githens JH, Alcorn R, et al. Experience with donors matched for minor blood group antigens in patients with sickle cell anemia who are receiving chronic transfusion therapy. Transfusion 1987;27:94–98. 68. Rosse WF, Telen M, Ware RE. Transfusion Support for Patients with Sickle Cell Disease. Bethesda, Md., AABB Press, 1998, pp 73–92. 69. Koshy M, Burd L, Wallace D, et al. Prophylactic red cell transfusions in pregnant patients with sickle cell disease: a randomized cooperative study. N Engl J Med 1988;319:1447–1452. 70. Diamond WJ, Brown Fl Jr, Bitterman P, et al. Delayed hemolytic transfusion reaction presenting as sickle-cell crisis. Ann Intern Med 1980;93:231–234. 71. Garratty G. Severe reactions associated with transfusion of patients with sickle cell disease [Editorial]. Transfusion 1997;37:357–361. 72. Heddle NM, Sortar RL, O’Hoski P, et al. A prospective study to determine the frequency and clinical significance of alloimmunization post-transfusion. Br J Haematol 1995;91:1000–1005. 73. Rao KR, Patel AR. Delayed hemolytic transfusion reactions in sickle cell anemia. South Med J 1989;9:1034–1036. 74. Kalyanaraman M, Heidemann SM, Sarnaik AP, et al. Anti-s antibodyassociated delayed hemolytic transfusion reaction in patients with sickle cell anemia. J Pediatr Hematol Oncol 1999;21:10–13. 75. Fabron JA, Moreira JG, Bordin JO. Delayed hemolytic transfusion reaction presenting as a painful crisis in a patient with sickle cell anemia. Rev Paul Med 1999;117:385–389. 76. Milner PF, Squires JE, Larison PJ, et al. Posttransfusion crises in sickle cell anemia: role of delayed hemolytic reactions to transfusion. South Med J 1985;78:1462–1469. 77. Petz LD, Garratty G. Acquired Immune Hemolytic Anemia, 2nd ed. New York, Churchill Livingstone, 2004, pp 547–549. 78. Win N, Doughty H, Telfer P, et al. Hyperhemolytic transfusion reaction in sickle cell disease. Transfusion 2001;41:323–328. 79. Cullis JO, Win N, Dudley JM, et al. Post-transfusion hyperhaemolysis in a patient with sickle cell disease: use of steroids and intravenous immunoglobulin to prevent further red cell destruction. Vox Sang 1995;69:355–357. 80. Polesky HF, Bove JR. A fatal hemolytic transfusion reaction with acute autohemolysis. Transfusion 1964;4:285–292. 81. Wiener AS. Hemolytic reactions following transfusions of blood of the homologous group. II: Further observations on the role of property

82. 83. 84. 85. 86.

87.

88. 89. 90. 91. 92. 93. 94.

95. 96. 97. 98. 99. 100.

Rh, particularly in cases without demonstrable iso-antibodies. Arch Pathol 1941;32:227–250. Petz LD. The expanding boundaries of transfusion medicine. In Nance LST (ed). Clinical and Basic Science Aspects of Immunohematology. Arlington, American Association of Blood Banks, 1991, pp 73–113. Test ST, Woolworth VS. Defective regulation of complement by the sickle erythrocyte: evidence for a defect in control of membrane attack complex formation. Blood 1994;83:842–852. Darabi K, Dzik S. Hyperhemolysis syndrome in anemia of chronic disease. Transfusion 2005;45:1930–1933. Telen MJ, Coombs M. Management of massive delayed hemolytic transfusion reactions in patients with sickle cell disease [Abstract]. Transfusion 1999;36(Suppl):97S. Bunker ML, Thomas Cl, Geyer SJ. Optimizing pretransfusion antibody detection and identification: a parallel, blinded comparison of tube PEG, solid-phase, and automated methods. Transfusion 2001;41: 621–626. Taswell HF, Pineda AA, Moore SB. Hemolytic transfusion reactions: Frequency and clinical and laboratory aspects. In Bell CA (ed). A Seminar on Immune-mediated Red Cell Destruction. Washington, D.C., American Association of Blood Banks, 1981, pp 71–92. Schonewille H, Haak HL, van Zijl AM. RBC antibody persistence. Transfusion 2000;40:1127–1131. Tahhan HR, Holbrook CT, Braddy LR, et al. Antigen matched donor blood in the transfusion management of patients with sickle cell disease. Transfusion 1994;34:562–569. Vichinsky EP, Luban NLC, Wright E, et al. Prospective RBC phenotype matching in a stroke prevention trial in sickle cell anemia: a multicenter transfusion trial. Transfusion 2001;41:1086–1092. Ness PM. To match or not to match: The question for chronically transfused patients with sickle cell anemia [Editorial]. Transfusion 1994;34: 558–560. Storry JR, Westhoff CM, Charles-Pierre D, et al. DNA analysis for donor screening of Dombrock blood group antigens. Immunohematology 2003;19:33–36. Hult A, Hellberg A, Wester ES, et al. Blood group genotype analysis for the quality improvement of reagent test cells. Vox Sang 2005;88: 465–470. Castro O, Sandler SG, Houston-Yu P, et al. Predicting the effect of transfusing only phenotype-matched RBCs to patients with sickle cell disease: Theoretical and practical considerations. Transfusion 2002;42:684–690. Strupp A, Cash K, Uehlinger J. Difficulties in identifying antibodies in the Dombrock blood group system in multiply alloimmunized patients. Transfusion 1998;38:1022–1025. Campbell SA, Shirey RS, King KE, et al. An acute hemolytic transfusion reaction due to anti-IH in a patient with sickle cell disease. Transfusion 2000;40:828–831. Callahan DL, Kennedy MS, Ranalli MA, et al. Delayed hemolytic transfusion reaction caused by Jk(b) antibody detected by only solid phase technique [Abstract]. Transfusion 2000;40(Suppl):113S. Lanskron S, Moliterno AR, Norris EJ, et al. Polymerized human Hb use in acute chest syndrome: a case report. Transfusion 2002;42:1422–1427. Winslow RM. Red cell substitutes. In Anderson KC, Ness PM (eds). Scientific Basis of Transfusion Medicine: Implications for Clinical Practice, 2nd ed. Philadelphia, WB Saunders, 2000, pp 588–598. Beiboer SHW, Wieringa-Jelsma T, VanWijk M. Rapid genotyping of blood group antigens by multiplex polymerase chain reaction and DNA microarray hybridation. Transfusion 2005;45:667–679.

Chapter 50

Febrile, Allergic, and Other Noninfectious Transfusion Reactions Nancy Heddle

●

Kathryn E. Webert

The transfusion of blood and blood products can be associated with a number of different noninfectious complications. The frequency of these events is variable, depending on the type of blood product being transfused and the patient population requiring the product. It is useful to categorize these types of reactions as acute (occurring during or within 8 hours of the transfusion) or delayed (presenting a week or two after transfusion), as this will facilitate the differential diagnosis and identification of these adverse events. The challenge for physicians and health care providers when these reactions occur is, first, to recognize that the reaction may be transfusion associated, and second, to use the clinical signs, symptoms, and situation to identify systematically the differential diagnosis, the reaction type, and the cause so that appropriate treatment and prevention strategies can be used. These challenges are complicated by the fact that many of the clinical signs and symptoms are common to several different types of reactions and that comorbidities may complicate the recognition of these events. In this chapter, the immediate and delayed noninfectious complications to transfusion are summarized in terms of their definition, manifestation, prevalence, pathophysiology (when known), diagnosis, treatment, and prevention. An approach is presented to illustrate

how this information can be used to work systematically through the complex clinical presentation, generate a differential diagnosis, and identify the cause of the event.

ACUTE NONINFECTIOUS TRANSFUSION REACTIONS Most of the acute noninfectious reactions occur during a transfusion or within a few hours of the transfusion; however, in some situations, signs and symptoms may not present until 6 to 8 hours after transfusion.1 As the interval between the transfusion and presentation of symptoms increases, it becomes less likely that the reaction will be recognized as transfusion related; hence it is important for physicians to be aware not only of reactions with early presentation but also of those reactions with delayed onset. The noninfectious reactions in the acute category include febrile nonhemolytic, allergic, anaphylactic, transfusion-related acute lung injury (TRALI), hypotensive, transfusion-associated circulatory overload (TACO), hemolytic reactions (immediate and delayed), and a variety of metabolic complications (Tables 50–1 and 50–2). Hemolytic reactions, TRALI, and bacterial

Table 50–1 Summary of the Signs and Symptoms Typically Observed with Different Types of Acute Reactions to Blood and Blood Products Types of Symptoms Cutaneous

Reaction Type Hemolytic Allergic (mild) Allergic (anaphylactic) FNHTR TRALI Bacterial sepsis Hypotensive Circulatory overload (TACO)

Pruritis Urticaria Erythema Flushing

Inflammatory

Pain

Respiratory

Cardiovascular

GI

Fever Chills Shakes Cold Feeling Rigors

Headache Chest Back Abdominal Site of Infusion

Hoarseness Stridor Lump in Throat Wheezing Chest Tightness Substernal Pain Dyspnea Cyanosis

Hypotension Loss of Consciousness Shock Tachycardia Cardiac Arrhythmias Cardiac Arrest

Nausea Vomiting Abdominal Cramps Diarrhea

X

X

X

X

X

X

X

X X X

x

X x

X X x X

X X X X (hypertension)

X X X

x x

X x

X, typical symptoms; x, symptoms sometimes observed; GI, gastrointestinal; FHNTR, febrile nonhemolytic transfusion reaction; TRALI, transfusion-related acute lung injury.

50 677

*

Delayed serologic transfusion reaction (DSTR)

Mild Allergic

Red cells: 0.02–1.07 Platelets: 0.06–7.6 All components: 3.0

Febrile nonhemolytic Red cells: reactions (FNHTRs) 0.07–6.8 Platelets: 0.13–37.5

*

Delayed hemolytic transfusion reaction (DHTR)

Red cells: 28.9–68.4

Red cells: 0.7–24.9

Red cells: 1.1–9.0

*

Incidence (per 100 transfusions)6

Acute hemolytic reactions

Type of Acute Reaction

Incidence (per 100,000 transfusions*)6 or incidence (per 100,000 patients**) 1. Pre-existing antibody in the patient’s plasma binds to an incompatible antigen on the transfused red cells causing destruction. 2. Antibody in donor plasma binds to the patient’s red cells causing hemolysis Patient is alloimmunized (sometimes primary but usually secondary) by the transfusion and the antibody binds to the transfused red cells causing destruction. Can be caused by active or passive antibody binding to the red cells and causing a positive DAT. No clinical signs of hemolysis. Red cells: Most reactions are due to a white cell antibody in the patient’s plasma that interacts with leukocytes in the red cell product. Platelets: Storage generated proinflammatory cytokines cause of reactions; white cell antigen/antibody interactions cause about 9% of reactions. Soluble antigens in the donor plasma react with IgE bound to mast cells causing histamine release.

Etiology

Temporarily stop the infusion keeping line open with saline. Administer antihistamines. If urticaria abates, transfusion may be restarted.

Discontinue transfusion. Antipyretics. Supportive care.

No treatment required. Observe patient to ensure that delayed hemolysis does not occur.

Stop the transfusion. Notify transfusion service. Send appropriate samples to investigate. Provide supportive care including intravenous fluids. Maintain urine output. Treat DIC and shock as required. Supportive care, if required. No specific treatment usually indicated.

Treatment

Summary of the Incidence, Etiology, Treatment, and Prevention of Noninfectious Transfusion Reactions

678

Table 50–2

III Premedication with antipyretics. Leukoreduced blood products (must be prestorage LR for preventing platelet reactions. Other options include fresher products, plasma reduced, or washed cellular products. Pretransfusion medication with antihistamine.

Future transfusions should be with antigen-negative blood

Follow SOPs (clinical areas and laboratory) to ensure appropriate sample, patient and product identification. Ensure all staff are educated to the importance of these procedures. Blood for future transfusions should be antigen negative.

Prevention

TRANSFUSION MEDICINE

Unknown

Post-transfusion purpura

Discontinue transfusion; antihistamine administration. Subcutaneous, intramuscular, or intravenous epinephrine if needed. Supportive care.

Mediated by kinins generated by activation of the contact system. Patients on ACE inhibitors appear to be more susceptible. Recipient has alloantibody directed against an antigen on the donor’s platelets. Patient’s own platelets are also destroyed: mechanism not clearly understood.

IVIG infusions are the first-line therapy. Plasmapheresis may also be effective. Patients may respond to antigen-negative platelets if bleeding.

Stop the transfusion.

Stop the transfusion. Provide aggressive supportive therapy including broad-spectrum antibiotics.

Stop the transfusion. Diuretics and supplemental oxygen therapy.

Multiple causes: passive transfusion on Supportive care only. No specific leukocyte antibody in the donor plasma. therapy indicated. Storage generated lipids in the blood product (two-hit model). Other mechanisms not yet understood.

Patient is IgA deficient and has IgG antibodies to IgA. Other possible mechanisms: antibodies to other serum proteins absent in the patient; transfused allergens; passive transfer of IgE; mast cell activation by anaphylactoxins.

Infusion of too much fluid into the circulation in a short time. Patients with cardiovascular disease at greater risk. * Platelet pools: 1.1–47.5† Bacteria in the blood product release * Apheresis: 1.0–7.5 endotoxin or once in the patient’s * Red cells: 0.02–1.5 circulation cause infection. * Platelet pools: 0.2–6.3 * Apheresis: 0.2–1.5 * Red cells: 0–1.0

All products: 16.8–8000*

**

*

Red cells: 0.3–4.3 Apheresis platelets: 7.7 * Platelet pools: 62.5 * All components: 2.1

*

Future transfusions should consist of platelets negative for the antigen to which the antibody is directed (i.e., HPA-1a negative).

Depends on the reaction severity and etiology. Anaphylactoid reactions may be prevented by antihistamine use. Anaphylactic reactions may require avoidance of all plasma containing products if reaction is due to anti-IgA or stringent washed cellular products. Removal of donors with potent leukocyte antibodies from the donor pool. Transfusion of fresher blood products or washed blood products may have a lower risk of causing these reactions. Recognize patients at high risk for TACO. Administer transfusion slowly (split units if necessary). Use of diuretics. Inspect all blood products for visual evidence of contamination before infusion; employ measures at the blood center and transfusion service to decrease risk. Avoid the use of negatively charged bedside leukoreduction filters.

†

The higher rate was observed in elderly patients (mean age, 67 years) undergoing hip or knee surgery. One study from Hong Kong reported the risk of TTBI with platelet pools to be 280/100,000 transfusions; however, all other reports have ranged between 1.1 and 47.5.

*

Unknown

Red cells: 1–11 Platelets: 1–46 Not specified: 7–36

Hypotensive reactions

Transfusiontransmitted bacterial infection (TTBI)† Deaths from TTBI

Transfusion-associated circulatory overload (TACO)

TRALI

Anaphylactic

FEBRILE, ALLERGIC, AND OTHER NONINFECTIOUS TRANSFUSION REACTIONS

50

679

TRANSFUSION MEDICINE

contamination are summarized in Chapters 41, 43, and 50, respectively; further clinical details on these complications are not presented. However, these two complications are included in the practical approach to classify acute transfusion reactions described at the end of this chapter.

FEBRILE NONHEMOLYTIC TRANSFUSION REACTIONS Definition/Manifestations/Prevalence

III 680

Febrile nonhemolytic transfusion reactions (FNHTRs) have traditionally been defined as a temperature increase of at least 1° C with or without chills, a cold sensation, or discomfort, occurring during or within a few hours of the transfusion.1 However, increasing recognition exists that this traditional definition should be modified. Although the name implies that fever is always present, this may not be the case. A recent study has shown that, frequently, the symptoms of chills, cold sensation, and discomfort may occur in the absence of fever.2 It is not uncommon that the temperature increase is suppressed by the use of antipyretic premedication. Studies performed at our center have shown that FNHTR characterized by chills, cold, and/or discomfort can occur without fever in about 85% of reactions. Severe reactions can be characterized by rigors and, rarely, headache, nausea, and vomiting.2 The frequency of FNHTRs varies depending on the type and age of the blood product being transfused, the patient population, and differences in the recording of symptoms by nursing staff and whether premedications are given.3–5 The reported incidence of FNHTR with red cell products ranges from 0.07 to 6.8 reactions per 100 red cell units transfused.6 The use of prestorage leukoreduced red cells appears to result in an absolute risk reduction ranging from 0.14% to 0.22%.7–9 FNHTRs appear to be more common when platelets are transfused; however, the incidence reported in the literature is extremely variable, ranging from 0.19 to 37.5 per 100 platelet transfusions.6 Historically, FNHTRs have been the most commonly encountered reaction, accounting for approximately 43% to 75% of all transfusion reactions reported at some centers.10,11 However, centers that use prestorage leukoreduced blood products have seen a dramatic decrease in the frequency of these events, especially those reactions associated with platelet transfusions.2,7,12

Pathophysiology At least two mechanisms seem to result in the clinical manifestations of FNHTR. The first mechanism involves the presence of a white cell alloantibody in the patient’s plasma that interacts with white cells in the blood product.13–15 Both granulocyte14,16–20 and human leukocyte antigen (HLA) antibodies18,20 have been implicated in these reactions. This antigen-antibody interaction causes release of endotoxins that act on the hypothalamus, resulting in fever and the other symptoms typical of FNHTR. This mechanism appears to be the primary cause of FNHTR after the transfusion of red cells but accounts for only approximately 9% of platelet reactions.21 The second mechanism involves the generation of leukocyte-derived cytokines during storage of the product. A variety of cytokines have been detected in stored blood products; however, the proinflammatory cytokines (interleukin [IL]-1, IL-6, IL-8, and tumor necro-

sis factor [TNF]-α) are likely to be the predominant cause of FNHTR.21–29 The release and/or production of cytokines during storage occurs mainly at warmer temperatures; hence this mechanism accounts for approximately 90% of the reactions to platelets that are not prestorage leukoreduced, as this product is stored at room temperature but is unlikely to be the cause of red cell reactions, as cytokine production/release is inhibited by the 4° C storage.30 Therefore, poststorage leukoreduction will be effective at preventing FNHTR to red cells, which are mainly due to white cells in the product, but will have minimal impact on platelet reactions. In contrast, prestorage leukoreduction of both of these products is an effective method for prevention of FNHTR due to both red cells and platelets.2,7–9 Although this approach has been shown to be effective for preventing most reactions, residual reactions still occur (0.2% for red cells and 1.0% with platelets), suggesting that other minor mechanisms may also play a role.2 The exact cause of these residual reactions is unknown; however, possible mechanisms that have been suggested include platelet-derived substances such as CD154 (CD40 ligand).31,32 It has also been postulated that some febrile reactions may represent immune responses to infused substances, such as IL-6, which has been found to be elevated in patients who experience an FNHTR.33,34

Diagnosis and Treatment The diagnosis of FNHTR is one of exclusion. Other causes of fever must be excluded, including other acute transfusion reactions that manifest with fever and clinical comorbidities and/or medications (Fig. 50–1). Sometimes the timing of the reaction may be helpful in the differential diagnosis. FNHTRs tend to be dependent on the dose or concentration of leukocytes and/or cytokines infused and therefore appear toward the end of the transfusion.1,2,12 It is atypical for this type of reaction to occur during the initial stage of the transfusion. Clinical experience has also suggested that the rate of infusion is another important variable that may contribute to these events.35 Treatment of these reactions consists of the administration of an antipyretic such as acetaminophen. Severe shaking chills (rigors) may necessitate meperidine administration. These symptoms resolve with or without treatment and do not typically lead to additional or long-term adverse sequelae. The decision to stop the transfusion requires clinical judgment based on the severity of the symptoms, the type of product being transfused, the timing of the symptoms in relation to the amount of product infused, and the history and clinical state of the patient.5

Prevention The approach for preventing FNHTR differs slightly depending on whether the product is red cells or platelets. For red cell reactions, the first line of prevention would include either poststorage or prestorage leukoreduction. If reactions still occur to leukoreduced products, other approaches that may be effective include the transfusion of fresher blood products and or washed blood products. To prevent FNHTR to platelet transfusions, prestorage leukoreduction has been shown to be the most effective method. For patients who still react to the prestorage leukoreduced product, other options include removing the residual plasma from the platelet product, washing the product, or selecting fresher products

Yes

No

Differential diagnosis • Acute hemolytic • FNHTR • TRALI • Bacterial contamination (Sepsis)

Consider: • Allergic • Anaphylactic • Hypotensive Can’t rule out: • Hemolytic, FNHTR or TRALI

How high did the temperature rise?

IgG/IgA

Alloantibodies to Blood Cells Alloantibodies to red blood cells and platelets may contribute to a disease process in a number of situations. Antibody removal has been used as a treatment in several of these pathologic states. Hemolytic disease of the newborn was one of the first conditions to be approached with automated apheresis instruments.8–10,202 Efforts were made to lower anti-D titers by TPE in sensitized D-negative mothers carrying a D-positive fetus in the hope of ameliorating fetal red blood cell destruction. The incidence of this condition has decreased dramatically as prophylaxis with Rh immune globulin has become more widespread, and intrauterine transfusion with D-negative red blood cells has proved to be a more effective treatment for an affected fetus.203 TPE is seldom used in this circumstance now, but it may still be helpful in pregnancies with early evidence of fetal involvement because intrauterine transfusion is not feasible before about 18 weeks of gestation.204,205 Early institution of TPE was associated with successful delivery after multiple prior abortions in compelling case reports involving anti-M206 and anti-P.207 TPE has also been used to remove isoagglutinins in the context of transplantation. Allogeneic stem cell transplantation across a major ABO barrier (e.g., group A donor into group O recipient) is feasible if a hemolytic transfusion reaction to red blood cells in the transplant can be avoided. For bone marrow transplants, this was first accomplished by exhaustive pretransplantation TPE of the recipient to reduce the relevant isoagglutinin titer.208 Immunoadsorption columns containing A or B substance have subsequently been used.209 Ultimately, however, it has proved preferable simply to remove most of the red blood cells from the bone marrow graft before transplantation.210 Peripheral blood stem cell grafts contain smaller quantities of red blood cells and can usually be infused safely without any manipulation. Red blood cell engraftment is sometimes delayed after ABOincompatible transplantation, however, and TPE has been tried, both before transplantation to avoid delayed engraftment and after transplantation to correct it, with uncertain results in both cases.211–213 Resolution of delayed engraftment after donor leukocyte infusions has also been reported.214 Solid-organ transplantation across a major ABO barrier can result in hyperacute rejection and is usually avoided for this reason. In desperate circumstances, however,

considerations of organ availability have sometimes dictated use of an ABO-incompatible liver after pretransplantation TPE of the recipient, often with a satisfactory outcome.215–217 More recently, the prolonged shortage of donor kidneys has been addressed by elective transplantation of kidneys from informed living donors across major ABO barriers.218–220 Both A2 and non-A2 organs have been successfully engrafted. Pretransplant immunosuppressive regimens include TPE to lower isoagglutinin titers.221 Organ transplantation across a minor ABO barrier (e.g., group O donor into group A recipient) does not increase the risk of rejection; however, it is sometimes followed, within 1 to 3 weeks after transplantation, by a hemolytic anemia related to isoagglutinin secretion by B lymphocytes carried in the transplanted organ. Hemolysis mediated by these “passenger lymphocytes” is most often seen in heart-lung or liver transplantation, where the volume of lymphoid tissue transplanted is relatively high.222 It may, however, be seen after renal transplantation and may involve anti-D.223 Severe hemolysis in this setting may be partially ameliorated by TPE in combination with compatible red blood cell transfusion.224 Alloimmunization to platelets can cause clinical refractoriness to platelet transfusions.225 Removal of antibody has been attempted, both with TPE226 and with a protein A-silica column.227,228 IVIG has also been tried.229–231 However, the results have been inconclusive and transfusion of compatible platelets is still the best option.

THERAPEUTIC PLASMA EXCHANGE

with macroglobulinemia, 40% of patients with IgA myeloma, and 15% of patients with IgG myeloma.194 Should they prove clinically significant, TPE therapy can help to restore adequate hemostasis. Renal failure occurs in 3% to 15% of myeloma patients and carries a poor prognosis. In many such cases, renal biopsy demonstrates accumulation of free light chains in renal tubules. With normal renal function, urinary excretion of light chains far exceeds the amounts that could be removed by TPE, but in renal failure, the reverse may be true.195–198 Two controlled studies have shown higher rates of recovery of renal function in patients who received TPE therapy. In one study, three of seven dialysis-dependent patients who received TPE recovered renal function whereas none of five control patients did.199 In the other, 13 of 15 patients in the treatment group recovered renal function compared with only 2 of 14 control subjects.200 Thus TPE is recommended for such patients.201

Thrombotic Thrombocytopenic Purpura and Hemolytic-Uremic Syndrome Thrombotic Thrombocytopenic Purpura Thrombotic thrombocytopenic purpura (TTP) is characterized by microangiopathic hemolytic anemia and thrombocytopenia, often severe. Fever, CNS changes, and renal abnormalities may be seen in advanced cases, although frank renal failure is uncommon.232 A rare relapsing form usually begins in childhood, but most cases are sporadic and occur in adults.233 Women are predominantly affected, accounting for 70% of cases.234 TTP is usually idiopathic, but the syndrome may be seen in association with other illnesses, such as SLE235 and HIV infection.236 Idiopathic TTP was formerly associated with a mortality rate of 95%, but empirical studies in the 1970s and 1980s demonstrated much better survival in patients who received TPE with plasma replacement. Some patients, including children with the relapsing form of the disease, respond to simple plasma infusion.233 A convincing account of the pathogenesis of TTP has been offered. It involves a plasma enzyme (a metalloproteinase designated ADAMTS 13) that cleaves the ultralarge von Willebrand factor (ULvWF) multimers secreted by endothelial cells, yielding the smaller polymers found in normal plasma. Children with relapsing TTP have an inherited deficiency of this enzyme,237 whereas an acquired deficiency, related to the presence of an autoantibody that inhibits the enzyme, has been found in idiopathic adult cases238,239 and some SLE240 and HIV-associated241 cases. In either instance, persistence of ULvWF in the circulation or on the surface of endothelial cells or both apparently promotes inappropriate adherence of platelets to endothelial cells and to each other, leading to consumptive thrombocytopenia and microvascular obstructions. The latter cause mechanical trauma to red

55 747

TRANSFUSION MEDICINE

blood cells and varying degrees of end-organ ischemia.233 Periodic plasma infusion may abort or prevent attacks in congenitally deficient patients.237 Idiopathic cases respond better to TPE (78% response rate vs 63% for plasma infusion in a Canadian study),242 presumably because exchange removes inhibitory antibody as well as supplying the deficient enzyme. Exchanges are usually carried out daily until the platelet count and lactate dehydrogenase (as a marker of hemolysis) have normalized. Various immunosuppressive measures, such as glucocorticoid therapy,243 vincristine,244 rituximab,245 and splenectomy,246 have been advocated as adjunctive treatments. Discovery of an autoimmune etiology provides support for their use in idiopathic but not in congenital cases. A syndrome similar to TTP has been identified in some patients taking the antiplatelet drug ticlopidine. Autoantibody-induced deficiency of the metalloproteinase has been demonstrated in patients with this syndrome,247 and TPE has been reported to improve survival (76% survival vs 50% for patients without exchange in a retrospective study).248 Hemolytic-Uremic Syndrome

III 748

The hemolytic-uremic syndrome (HUS) is characterized by microangiopathic hemolysis, renal insufficiency, and mild to moderate thrombocytopenia.249 In children, it may occur in a self-limited form that follows infection with a verotoxinproducing Escherichia coli (diarrhea + or D+); however, some pediatric cases lack this association (D–).250,251 Familial and nonfamilial forms affect adults.238 Nonfamilial cases may be idiopathic or may be associated with prior chemotherapy or stem cell transplantation.252 Because of overlapping manifestations that sometimes preclude distinction on clinical grounds alone, it has long been supposed that the pathogenesis of HUS was similar to that of TTP. Largely on this basis, TPE and protein A column therapy have been recommended for D-childhood HUS and for adult HUS. Although this is a prevalent standard of care, responses have generally been less favorable than in TTP, particularly in thrombotic microangiopathy associated with chemotherapy and transplantation.233,253–255 Levels of the vWF-cleaving metalloproteinase have now been studied in adult patients with both familial and nonfamilial forms of HUS. Neither severe deficiency nor inhibitory activity has been found.238 Another study in patients with clinically uncategorized thrombotic microangiopathy compared 50 patients with severe ADAMTS 13 deficiency to 57 patients who were not severely deficient. The severely deficient patients had a lower median platelet count (13,000/μL vs 44,000/μL; P < 0.001), a lower median creatinine (1.3 mg/ dL vs 2.7 mg/dL; P < 0.001), and a higher rate of relapse (16 of 46 vs 4 of 47 evaluable patients; P < 0.01).256 Thus the prior supposition of a shared pathogenic mechanism for TTP and HUS would seem to be in error. Deficiency of complement factor H has been identified in some patients with familial HUS.257–259 This might provide some rationale for plasma therapy; however, for most cases, the role of TPE is unclear at this time. Similar comments apply to the HELLP syndrome of hemolysis, elevated liver enzymes, and low platelets in pregnant women,260 which shares clinical features with TTP, HUS, and preeclampsia. ADAMTS 13 Assays Several techniques have been devised to measure ADAMTS 13 activity in patient plasma and to detect inhibition of

exogenous enzyme by patient plasma. End points after incubation of vWF with a putative enzyme or inhibitor source include electrophoretic assessment of vWF multimer size, extent of collagen binding by residual vWF, residual ristocetin cofactor activity, and immunoradiometric quantitation of cleavage sites.261 Currently none of these assays is widely available to clinicians with a rapid turnaround time. This inaccessibility alone limits the value of such assays in clinical decision making. Furthermore, in multilaboratory comparative studies, the intra- and interlaboratory precision and accuracy of results obtained on replicate samples have been far from ideal (with coefficients of variation as high as 83%, even in the most respected laboratories), although most laboratories seem capable of recognizing severely deficient samples, variously defined as containing less than 5% or less than 10% of the ADAMTS 13 activity level in pooled normal plasma.262,263 Interlaboratory variability in detection and quantitation of ADAMTS 13 inhibitors has been even more disappointing.262 Several retrospective surveys have questioned the value of ADAMTS 13 inhibitor assays in recognizing TTP, in differentiating TTP from HUS, and/or in predicting responsiveness to TPE.264,265 Given the limitations of current assays and the disagreement about what degree of deficiency is “clinically significant,” it is perhaps not surprising that their predictive value is suboptimal. Other confounding factors may also contribute to apparent discrepancies. These might include (1) alternative pathogenic mechanisms in some patients (for example, an antibody that blocks access of ADAMTS 13 to endothelial cell membranes could theoretically produce a TPE-responsive TTP syndrome with normal plasma enzyme levels233); (2) inability to eliminate very high titer inhibitors even with aggressive TPE, as is seen with some factor VIII inhibitors; and (3) spontaneous resolution of illness or response to concurrent therapies such as corticosteroids in patients with disease processes unrelated to ADAMTS 13 who nevertheless undergo TPE. Despite these reservations, it seems certain that the elucidation of ADAMTS 13 deficiency was a major advance in understanding the pathogeneses of TTP. It is likely that accurate assays for ADAMTS 13 activity and inhibitors that are widely available and have short turnaround times would be useful in evaluation of patients with thrombocytopenia and microangiopathic hemolysis.240,265,266 However, prospective controlled studies and/or additional clinical experience will be necessary to assess fully the value of such assays.

Post-transfusion Purpura Post-transfusion purpura (PTP) is a rare syndrome in which the platelet count decreases abruptly to dangerously low levels about 1 week after an allogeneic transfusion. Patients who develop the syndrome lack the common allele for one of the platelet-specific glycoprotein antigens, most often the human platelet-specific antigen 1a (HPA-1a) on glycoprotein IIIa. Most affected patients have had multiple pregnancies or prior transfusions and have been immunized thereby to the platelet-specific antigen coded by the prevalent allele. The transfusion that precipitates the illness appears to stimulate an anamnestic response with an increasing IgG antibody titer, most often anti–HPA-1a. PTP is self-limited and usually resolves without treatment in several weeks; however, bleeding complications, including fatal CNS hemorrhage, may occur.267–269

Idiopathic Thrombocytopenic Purpura Idiopathic thrombocytopenic purpura (ITP) is an autoimmune illness affecting platelets. Most patients have an IgG autoantibody directed against a platelet membrane glycoprotein antigen. ITP sometimes accompanies warm autoimmune hemolytic anemia (Evans syndrome). A pediatric form of ITP is acute and self-limited; recovery is the rule with or without treatment. Adult cases, most of which occur in women, seldom remit without therapy, and most progress to the chronic form of the disease. An ITP-like syndrome also occurs in association with SLE and HIV.273 The goal of treatment in ITP is prevention of bleeding. Because most platelets circulating in ITP patients are relatively young and more hemostatically active, avoidance of hemorrhage can be accomplished without normalization of the platelet count. Corticosteroids, splenectomy, and IVIG are the mainstays of therapy.273 Some favorable anecdotal reports of TPE in ITP appeared in the 1970s.274,275 One small controlled trial suggested a lower rate of splenectomy in patients receiving exchange,276 whereas a retrospective survey found little difference.277 In the absence of confirmatory trials, however, enthusiasm for this approach has waned, and TPE is seldom used in ITP. Favorable responses to protein A–silica column treatment have been reported, first in ITP associated with HIV infection278 and later in patients with chronic ITP without HIV.279 The mechanism of action of protein A in this disease is uncertain. A subtractive mechanism involving antiplatelet antibody is unlikely because responses have been reported with treatment of as little as 250 mL of plasma per week. Bearing this uncertainty in mind, the protein A–silica column remains an option for chronic ITP refractory to more standard therapies.

Autoimmune Hemolytic Anemia Autoimmune hemolytic anemia (AHA) is caused by autoantibodies to red blood cells. Such antibodies are either cold or warm agglutinins. Cold agglutinins are usually IgM

antibodies directed against the I/i antigens; they bind preferentially at low temperatures and produce the syndrome of complement-mediated intravascular hemolysis known as cold agglutinin disease (CAD). Warm agglutinins are usually IgG and are often directed against an antigen that is not expressed on Rhnull cells. They bind better at body temperature and produce a predominantly extravascular hemolytic syndrome (warm autoimmune hemolytic anemia [WAHA]). AHA can be idiopathic but may also be associated with an infection, a lymphoproliferative disorder, or another autoimmune disease.280,281 Most cases require treatment. Standard therapy is aimed at inhibiting antibody production or reducing destruction of sensitized cells, or both. Corticosteroids, IVIG, and splenectomy are often effective in WAHA, and other immunosuppressive drugs may be tried if these measures fail.280 All approaches are less successful in CAD.281 TPE to deplete circulating antibody has been tried in both WAHA and CAD when conventional treatments have failed.282–288 IgM antibodies in CAD are predominantly intravascular and only loosely bound to cells; thus their removal by TPE should be relatively efficient. Such therapy, when added to conventional drug treatment, has been reported to transiently reduce antibody titers and transfusion requirements in CAD. In WAHA much of the circulating antibody is bound to red blood cells; TPE has also been tried in this disorder, but it is less likely to be helpful.

THERAPEUTIC PLASMA EXCHANGE

Although PTP is linked to a platelet-specific alloantibody response, the mechanism of extensive destruction of antigennegative autologous platelets remains obscure. The following four possibilities have been proposed: (1) platelet antigenantiplatelet antibody immune complexes bind to autologous platelets and mediate their destruction; (2) autologous platelets adsorb soluble alloantigen derived from the transfusion; (3) a simultaneous platelet autoantibody response occurs; and (4) a broad alloimmune response produces antibodies that crossreact with autologous platelets.270 Treatment of PTP is advisable because many patients have bleeding complications. High-dose corticosteroids are usually given empirically, and pulse methylprednisolone at 1 g/day has been reported effective. Platelet transfusions, even those from antigen-negative donors, do not increase the platelet count and may cause severe reactions. Daily TPE usually promotes an increase in platelet count within a few days and is considered an effective treatment for this reason, even though controlled trials have not been done.271 Exchanges usually include FFP replacement to avoid a superimposed humoral coagulopathy. IVIG leads to similarly prompt increases in platelet count and has become the favored treatment modality for this group of patients.272

Aplastic Anemia and Pure Red Blood Cell Aplasia Aplastic anemia and pure red blood cell aplasia are bone marrow disorders. The former leads to pancytopenia, whereas in the latter, only reticulocytopenic anemia results. At least some cases of both conditions are likely to be immunologically mediated. Allogeneic bone marrow transplantation is the preferred treatment for aplastic anemia if a suitable donor is available, but immunosuppressive therapies, such as corticosteroids, cytotoxic drugs, cyclosporin, and antithymocyte globulin, are effective in some cases.289–291 In a minority of patients, it is possible to demonstrate a serum factor, probably antibody, that is inhibitory to the growth of relevant marrow-derived precursor cells in culture.292 These observations provided an explicit rationale for TPE, and this therapy has been reported in a handful of cases in both disorders. Results in aplastic anemia have been mixed, with responses more likely in patients demonstrating serum inhibitory activity.292–294 In pure red blood cell aplasia, all reported instances of TPE treatment have led to improvement, which is sometimes quite dramatic in patients with serum inhibitory activity.295–298 Thus although TPE is not a primary therapy for either disorder, it is an option for patients who have not responded to more conventional treatment, especially those who have serum inhibitory factors.

Coagulation Factor Inhibitors Coagulation factor inhibitors are IgG antibodies directed against components of the humoral clotting cascade. They inactivate the targeted factor and interfere with clotting. They may be autoantibodies that arise in individuals with no prior bleeding disorder. Alternatively, they may be alloantibodies arising in genetically deficient patients as a result of exposure to a “foreign protein” in the course of factor-replacement

55 749

TRANSFUSION MEDICINE

III 750

therapy.299,300 Factor VIII is the clotting protein most often affected by antibodies of either type. The goals of treatment for patients with inhibitors are, first, control of individual bleeding episodes and, eventually, suppression of inhibitor synthesis.301,302 Depending on the inhibitor titer, the first goal can usually be achieved by infusion of high doses of human factor VIII, porcine factor VIII (which is usually only partially crossreactive with anti–human factor VIII antibodies),303 or, for patients with the highest titers, factor VIII–bypassing products, including recombinant factor VIIa.304 TPE with plasma replacement may be used during a bleeding episode to reduce inhibitor titers enough to allow infused human or porcine factor VIII to circulate and bring about hemostasis.305–307 Immunoadsorption with protein A–Sepharose may also be effective for this purpose.308 Suppression of inhibitor synthesis is approached with immunosuppressive measures, including high-dose corticosteroids, cytotoxic agents,309 cyclosporin,310 IVIG,311 and rituximab.312,313 Tolerance-inducing protocols have been devised for congenitally deficient patients with alloimmune inhibitors.314,315 These involve regular infusion of exogenous factor VIII. In the so-called Malmö protocol, extensive TPE or IgG depletion with protein A–Sepharose column procedures is used to reduce the inhibitor titer at the onset of treatment so that infused factor can circulate.316 Patients with inhibitors usually need frequent, large (two or three plasma volumes) exchanges with FFP replacement. Central venous access is frequently required; catheter placement in the context of a refractory bleeding disorder is challenging to all concerned and often mandates use of a factor VIII–bypassing product for wound hemostasis.317 TPE has also been reported for treatment of patients with antiphospholipid antibodies.318 These may interfere with in vitro assays of coagulation, such as the partial thromboplastin time. However, unlike the inhibitory antibodies described earlier, they usually promote inappropriate coagulation in vivo and cause thrombotic events.

THERAPEUTIC PLASMA EXCHANGE IN OTHER IMMUNOLOGIC DISORDERS TPE has been tried in a number of rheumatic diseases and other diseases that are considered to have an immune or autoimmune etiology. These are listed in Table 55–6 along with the indication categories assigned by ASFA and AABB.40,41

Cryoglobulinemia Cryoglobulins are defined as serum proteins that precipitate reversibly at 4° C, although some precipitate at higher temperatures. The precipitates always contain immunoglobulins, and immunoelectrophoretic or immunofixation analysis allows distinction of three types.319 Type I cryoglobulins consist of a single species of monoclonal immunoglobulin. These are usually found in myeloma, Waldenström macroglobulinemia, or other lymphoproliferative disorders (see Table 55–5). Cryoglobulin levels are often quite high (>500 mg/dL) and may cause Raynaud phenomenon or acral necrosis related to microvascular obstruction, as well as other symptoms. Type II cryoglobulins have both a monoclonal and a polyclonal component. The former is usually an IgMk with anti-IgG activity that binds polyclonal IgG in an immune complex. These typically produce

Table 55–6 ASFA/AABB Indication Categories for Therapeutic Plasma Exchange in Rheumatic and Other Immune Disorders Disorder Cryoglobulinemia Rheumatoid arthritis (IA) Systemic lupus erythematosus Systemic vasculitis Polymyositis and dermatomyositis Goodpasture syndrome Rapidly progressive glomerulonephritis Renal transplantation Rejection Presensitization Recurrent focal glomerulosclerosis Heart transplant rejection

Indication Category* II II IV III IV I II IV III III III

* I, standard first-line therapy; II, second-line therapy; III, controversial; IV, no efficacy. IA, protein A-silica immunoadsorption.

a lower-extremity cutaneous vasculitis and may cause other, more serious visceral manifestations of immune complex disease. Many cases occur in association with chronic hepatitis C infection.320 Type III cryoglobulins are mixed polyclonal, often with IgM anti-IgG that binds IgG in immune complexes. These may arise in acute infections, such as hepatitis B, and in severe rheumatoid arthritis or other chronic inflammatory states. Clinical manifestations resemble those of serum sickness.319,321 If an associated condition is present, cryoglobulin levels and related symptoms may decrease with treatment of the primary disorder (for example, chemotherapy for myeloma or interferon for hepatitis C virus infection). For idiopathic and secondary cases of mixed cryoglobulinemia, prednisone therapy may be effective in relieving symptoms, and alkylating agents may be useful for patients with severe symptoms resistant to prednisone.321 TPE reduces cryoglobulin levels and controls symptoms.322 It can do so even in the absence of other treatments,35 but expense and inconvenience usually preclude such use. Plasma exchange therapy should be started promptly for patients with severe acral ischemia or visceral manifestations of vasculitis.323 In such patients, it can help gain and maintain control of symptoms until aggressive drug therapy takes hold. Patients with chronic ulcers in the setting of cutaneous vasculitis may also benefit.324 Care should be taken to warm replacement fluids to body temperature before reinfusion.

Rheumatoid Arthritis Rheumatoid arthritis (RA) is a disease of unknown etiology that is more prevalent in women. It is the most common chronic inflammatory joint disease and a leading cause of disability. Most patients have rheumatoid factor, an IgM autoantibody to IgG; however, because this finding is absent in many otherwise typical cases and because it occurs in other patients who do not have arthritis, it is not thought to be directly involved in pathogenesis.325 Conservative treatment of RA includes nonsteroidal anti-inflammatory agents, low-dose oral corticosteroids, and intra-articular steroids. More severely affected patients

Systemic Lupus Erythematosus SLE has been viewed as the prototypical autoimmune disease. The occurrence of antibodies to DNA, especially those specific for double-stranded DNA (anti-ds-DNA), identifies a group of patients who may have autoantibodies to a variety of other determinants. Their clinical syndromes compose a disparate mixture in which skin disease, joint disease, cytopenias, or nephritis may be the sole or dominant manifestation.333,334 Immunosuppressive measures are the cornerstone of therapy for SLE. Most patients receive prednisone in varying doses, and those with severe disease may also be given azathioprine or cyclophosphamide.335,336 The plethora of autoantibodies that seem relevant to clinical signs made SLE an obvious target for plasma-exchange therapy, and it was one of the first illnesses to be treated with automated TPE in the early 1970s. A number of case reports and uncontrolled series suggested a favorable effect.11,12,337,338 Lupus nephritis is a particularly devastating expression of SLE in which glomerular deposition of anti-ds-DNA and immune complexes is believed to play a prominent pathogenic role. It thus seemed an attractive context for randomized trials. A controlled trial with only eight patients suggested benefit.339 However, a multi-institutional randomized controlled trial comparing oral cyclophosphamide plus TPE with oral cyclophosphamide alone failed to show any advantage for the patients receiving TPE.26 A later international trial included patients with a variety of severe manifestations. It was structured to take advantage of an enhanced sensitivity to a properly timed pulse dose of intravenous cyclophosphamide that was believed to follow removal of pathogenic antibody by TPE.340 As mentioned earlier, subsequent work with knockout mice deficient for the FcRn receptor suggested that enhanced susceptibility would not occur.24 In any case, this trial also failed to show any benefit associated with TPE, either in all patients341 or in a subgroup with nephritis.342 One group has also reported a higher incidence of infection in SLE patients treated with TPE.343 Thus several large controlled studies have failed to reveal any worthwhile effect of TPE in SLE.

Systemic Vasculitis Systemic vasculitis is an inclusive term applied to a group of disorders that cause inflammation in blood vessel walls with resultant ischemic tissue damage. Vasculitis syndromes are conveniently classified on the basis of the size of vessels typically involved, as shown in Table 55–7. Most are of unknown etiology. The presence of immune complexes in some syndromes and of autoantibodies in others, such as antineutrophil cytoplasmic antibodies (ANCAs) found in Wegener granulomatosis (c-ANCA) and polyarteritis (p-ANCA), has lent credence to the notion that humoral immune factors are somehow involved.344 Prednisone is the first-line therapy for most vasculitic syndromes. Cyclophosphamide or methotrexate is often added in more severe cases.345 Randomized controlled trials in renal vasculitis,346 as well as in a group of patients with polyarteritis or Churg-Strauss angiitis,347 have shown little evidence that long-term benefit is conferred by addition of TPE to immunosuppressive drug therapy. Nevertheless, it may be requested for patients who are not responding to maximal drug therapy. A recent observational trial revealed good outcomes in patients with polyarteritis nodosa associated with hepatitis B who received TPE in conjunction with an antiviral drug.348

THERAPEUTIC PLASMA EXCHANGE

eventually receive slow-acting antirheumatic agents that are probably immunomodulatory, such as antimalarials, gold compounds, and methotrexate.326 Tumor necrosis factor inhibitors have been approved for use in RA, and other cytokine antagonists may be beneficial as well.326,327 TPE was tried in RA in the 1970s and 1980s, but controlled trials did not show benefit.328,329 Subsequent reports of lymphapheresis were published, with or without TPE in addition. Some controlled trials showed significant but short-lived benefit,330 whereas others did not.331 In practice, the prospect of no more than a modest chance of only modest benefit from a costly and inconvenient therapy has discouraged treatment with therapeutic apheresis. A sham-controlled trial of 12 weekly protein A–silica column treatments produced improvement in 33% of 48 treated patients versus only 9% of 43 control subjects. Benefit persisted for about 8 months on average, and a subsequent course was again beneficial in seven of nine initial responders.332 Although its mechanism of action is unclear, this device has gained Food and Drug Administration (FDA) approval for use in RA, and postmarketing clinical experience is beginning to accumulate.

Polymyositis and Dermatomyositis Polymyositis and dermatomyositis are inflammatory diseases of skeletal muscle. A characteristic dermatitis, typically involving the eyelids, knuckles, neck, and shoulders, is also found in the latter condition. Proximal weakness with biochemical evidence of muscle cell enzyme leakage is the usual presentation. The diagnosis is confirmed by characteristic inflammatory histology and immunopathology in a biopsy specimen from an affected muscle. The natural history is progressive fiber loss, eventually leading to profound, irreversible weakness. An autoimmune etiology is suspected, but an antibody specific for skeletal muscle has not yet been implicated in pathogenesis.349 Initial treatment is high-dose prednisone, which can often be tapered to levels that are tolerable in the long term. Resistant disease is treated with azathioprine, methotrexate, cyclosporine, or an alkylating agent, or a combination thereof.349 Controlled trials have also shown that IVIG infusion reduces muscle enzyme levels and improves strength temporarily.350 Several uncontrolled series have been interpreted to show that TPE was beneficial, but these were confounded by concurrent

Table 55–7 Classification of Systemic Vasculitis Large-vessel vasculitis Giant cell (temporal) arteritis Takayasu arteritis Medium-sized vessel vasculitis Polyarteritis nodosa Kawasaki syndrome Small-vessel vasculitis Wegener granulomatosis Churg-Strauss syndrome Microscopic polyangiitis Henoch-Schönlein purpura Essential cryoglobulinemic vasculitis Cutaneous leukocytoclastic angiitis

55 751

TRANSFUSION MEDICINE

escalations in immunosuppressive drug therapy.351–353 A randomized, controlled trial was reported in which 12 patients received TPE, 12 received lymphapheresis, and 12 received sham apheresis, with no changes in drug therapy.354 No difference in the response rate was found among the three groups. Thus despite the successes with IVIG, no role for TPE appears obvious in the treatment of polymyositis.

Table 55–8 Subtypes of Rapidly Progressive Glomerulonephritis (RPGN) Nomenclature

Quality of Subendothelial Immune Deposits

Goodpasture syndrome Immune complex RPGN Pauci-immune RPGN

Linear Granular Scant or none

Goodpasture Syndrome

III 752

Goodpasture syndrome (GPS) is an illness characterized clinically by rapidly progressive glomerulonephritis and in some cases by pulmonary hemorrhage. Renal biopsy shows crescent formation in many glomeruli on light microscopy and linear subendothelial immune deposits on immunofluorescence and electron microscopy. Lung biopsy may show similar subendothelial deposits. In 95% of cases, the syndrome is associated with circulating antibody that binds to glomerular basement membrane (anti-GBM). These antibodies are directed against a noncollagenous sequence (NC1) of the α-3 chain of a type of collagen (type IV) that is found in appreciable quantities only in renal and pulmonary basement membranes. Untreated GPS progresses quickly and relentlessly, and most patients die of uremia or of complications of lung hemorrhage.355,356 Since the classic report by Lockwood and colleagues in 1975,357 the recommended treatment for GPS has been aggressive TPE combined with high-dose prednisone and cyclophosphamide.358,359 TPE reduces anti-GBM levels quickly to minimize progression of tissue damage.360 Exchanges are usually carried out daily and may be continued for 2 weeks. It is therefore prudent to give some FFP replacement in the latter part of each exchange to replete coagulation factors and avoid exacerbations of lung bleeding related to dilutional coagulopathy.355,356 The only controlled trial on record failed to show an advantage for GPS patients who received TPE; however, this study has been largely discounted because the TPE schedule (every 3 days) was not sufficiently aggressive and because, despite randomization, the extent of renal damage at entry was not comparable in the TPE and control groups.361 Early treatment is preferred because it has been noted that patients with dialysis-dependent renal failure at the onset of treatment are not likely to recover renal function.362,363 A corollary is that patients whose renal biopsies show an irreversible lesion and who do not have pulmonary hemorrhage are unlikely to benefit from TPE.

Other Rapidly Progressive Glomerulonephritis In addition to GPS, two other categories of rapidly progressive glomerulonephritis (RPGN) exist, as shown in Table 55–8. In both of these, the light microscopic findings are similar to those in GPS with severe glomerular inflammation and crescent formation. Some cases may even have associated lung hemorrhage. The distinction among RPGN subcategories rests on immunofluorescence microscopy and electron microscopy, which show granular (vs linear in GPS) subendothelial immune deposits, usually taken to be immune complexes, in one group of patients and absence or relative scarcity of immune deposits in the other (pauci-immune RPGN). Patients in each group may have isolated, idiopathic renal disease or may have accompanying features that

suggest a diagnosis of systemic vasculitis (for example, mixed cryoglobulinemia or Henoch-Schönlein purpura for granular-immune complex RPGN and microscopic polyangiitis or Wegener granulomatosis for patients with pauci-immune RPGN, many of whom test positive for ANCA).364 Because treatments for the two categories are similar, some trials and series have included patients with both types of RPGN.365,366 Virtually all patients receive prednisone, and most receive cyclophosphamide, either orally or intravenously. TPE has been used extensively in patients with both types of disease, and early uncontrolled series tended to credit it with a beneficial effect. Two controlled trials, one published in 1988367 and the other in 1992,368 showed no advantage overall for patients who received TPE in addition to immunosuppressive drugs. However, a separate analysis of the small subgroup in the second study who had dialysis-dependent renal failure suggested that such patients have a better chance of recovering renal function if they receive TPE.368 A comparable trend was not noticed in the trial conducted by Guillevin and colleagues.369 A prospective, randomized trial by Stegmayr and coworkers362 compared TPE with immunoadsorption. Among 38 patients with non-GPS RPGN, 87% of whom had ANCA, 70% avoided long-term dialysis. TPE and immunoadsorption were equally effective. In a more recent study of 22 patients with ANCA-positive RPGN, 12 patients who received prednisone, cyclophosphamide, and plasmapheresis had better outcomes than did 10 patients who received only drug therapy.370 Criteria used for treatment-arm assignment were not specified. Thus TPE continues to be controversial in non-GPS RPGN. It is probably not justified as a first-line therapy, except (paradoxically) in patients who initially have dialysis dependence,371 but it may be offered to patients who progress despite immunosuppressive drug therapy.

Solid Organ Transplantation TPE has been used in the organ transplant setting both to treat and to prevent rejection as well as to treat recurrence of certain diseases in a transplanted organ. Photopheresis has also been tried for the same purposes, mostly in heart transplantation. Rejection Cellular immunity is thought to mediate rejection of most organ allografts; however, rapid antibody-mediated rejection may occur in patients who have preexisting antibodies to ABO or human leukocyte antigen (HLA) expressed by the graft.372 Histologically, such hyperacute rejection is characterized by fibrin deposition, endothelial damage, and neutrophil infiltration in small blood vessels; damage to the graft parenchyma is mainly ischemic.373 All treatments, including TPE, have been futile in hyperacute rejection. Vascular

been suggested that this diagnosis be made, and TPE used, in patients who have relatively normal biopsies in the context of deteriorating cardiac function.399 Controlled data to support this assertion are lacking, as are any published data correlating this clinical syndrome with circulating DSAs. TPE has been used before transplantation in patients presensitized to HLA antigens. Patients whose sera react with lymphocytes from a large proportion of the population have a diminished chance of a compatible crossmatch with an available donor and hence a low likelihood of receiving a transplant. Prospective immunosuppression, combined with removal of antibody by TPE or protein A–Sepharose immunoadsorption, has been tried as a means to obtain a compatible crossmatch and prevent hyperacute rejection. Several groups have reported that patients prepared in this way who receive cadaveric kidney transplants have quite respectable graft survival rates.400–407 Similar protocols have facilitated successful transplantation of ABO-incompatible kidneys and livers.217,408–412 In more recent studies, pretransplant TPE has facilitated elective living-donor transplantation of HLA-crossmatch positive as well as ABO-incompatible organs.220,395,413,414 In summary, the preponderance of evidence from controlled trials shows that TPE is not effective in reversing established rejection of renal allografts, except possibly in patients with circulating DSAs. Pretransplantation antibody removal can allow transplantation for otherwise ineligible candidates, especially those with high-titer antibodies to one or two HLA antigens whose titer-sensitive crossreactions can be suppressed. It can also facilitate transplantation of kidneys from living donors across ABO and donor-specific crossmatch barriers. The role of TPE in cardiac transplantation, if any, remains unclear. Recurrence of Disease Focal glomerulosclerosis (FGS) is a disease that causes nephrosis and renal failure, predominantly in children. It recurs after transplantation in at least 30% of patients who receive an allograft for FGS, suggesting that a humoral factor may play a role in pathogenesis.415 Some evidence implicates a plasma factor having a molecular mass of 30 to 50 kDa that binds to protein A, is heat sensitive, and is soluble in 70% ammonium sulfate.416,417 Other factors having estimated masses of 12 to 100 kDa have also been described in some patients.418 All remain poorly characterized as to structure and half-life. Several reports describe reductions in proteinuria and improved renal function when recurrence of FGS is treated with stepped-up immunosuppression and TPE,415,419–423 but no prospective randomized controlled studies have been reported. GPS may occasionally recur in an allograft,424 although this can usually be avoided by delaying renal transplantation until anti-GBM titers have decreased spontaneously. If it recurs, it should be treated promptly with TPE and cyclophosphamide. A diffuse coronary artery disease, termed allograft vasculopathy, sometimes develops in transplanted hearts. It is a leading cause of morbidity and mortality in allograft recipients who survive beyond 1 year. It may be related to continuing hyperlipidemia as well as chronic immune injury.425 Selective depletion of LDLs, discussed in more detail in the following section on hypercholesterolemia, has been thought to be helpful in a few transplant patients with persistent lipoprotein abnormalities.426 Photopheresis has also been used

THERAPEUTIC PLASMA EXCHANGE

changes reminiscent of hyperacute rejection may be seen in later rejection episodes. These have sometimes been taken to indicate antibody-mediated rejection, even when immunofluorescence microscopy for immunoglobulin and tests for circulating antibody are negative.373 Prophylactic immunosuppression with corticosteroids and either cyclosporine or tacrolimus is standard posttransplantation management for kidney and liver transplantation. Heart recipients may also receive azathioprine or mycophenolate mofetil. Rejection episodes occurring in spite of these measures are treated with pulse steroids and/or therapeutic antibodies that deplete T cells, B cells, or both.374,375 TPE was used extensively in the late 1970s and early 1980s to treat renal transplant rejection after a number of case reports and uncontrolled series suggested it was beneficial.376–381 In the middle and late 1980s, however, five controlled trials were reported. Four showed no significant benefit for patients receiving TPE in addition to standard drug therapy, even in the subgroups with vascular histology in transplant biopsies.382–385 In the single study suggesting benefit, the mean time of treatment was 10 to 11 months after transplantation, when antibody-mediated rejection has been considered less likely.386 The last and largest study was reported by Blake and colleagues,385 who concluded that TPE therapy could no longer be recommended for this purpose. All of the aforementioned studies enrolled patients based on the timing of rejection (i.e., early rejection) and/or routine histologic criteria (i.e., acute rejection; vascular rejection). In none of them were patients assessed for the presence of donor-specific antibodies (DSAs). Despite this discouraging experience, use of TPE for renal transplant rejection continued to be reported.387–389 In the recent past, a new wave of enthusiasm for TPE has focused on patients with clinical and histologic evidence of acute rejection who also have DSAs and/or deposition of complement component C4d in transplant biopsy tissue.390–394 DSAs can be shown by a positive crossmatch with donor cells or by flow-cytometric reactions with donor antigens. C4d deposition is demonstrated by immunofluorescence microscopy.393 Uncontrolled studies have been encouraging. Observations in the pretransplant period have shown that titers of donor specific HLA and/or ABO antibodies can be reduced by TPE.219,395 Furthermore, in open trials, patients with refractory acute rejection who demonstrate DSAs and/or C4d have had relatively good long-term graft survival (>80%) after treatment regimens that include TPE and IVIG.392,393 A recent conference on antibody-mediated rejection recommended that demonstration of circulating DSA be considered an essential criterion for diagnosis of humoral rejection and that the effectiveness of TPE and IVIG in this entity be studied in “rigorous prospective, multicenter” trials.396 Notwithstanding its ineffectiveness in controlled trials in renal transplantation, preliminary experience with TPE has also been reported in the context of cardiac allograft rejection. Favorable courses have been seen in individual patients who were also receiving other therapies, but no controlled trials have been reported.397,398 The criteria for a diagnosis of humoral rejection in this setting remain controversial.372 Vascular histology is more difficult to detect or exclude in the endomyocardial biopsies done to monitor cardiac allografts because few blood vessels are found in this part of the heart muscle. Immunofluorescence microscopy positive for IgG is the usual criterion for humoral rejection; however, it has

55 753

TRANSFUSION MEDICINE

as an antirejection measure in this setting (see chapter 56 on cytapheresis).427

THERAPEUTIC PLASMA EXCHANGE IN TOXIC AND METABOLIC DISORDERS This section covers several conditions in which removal of nonantibody substances present in plasma has been deemed therapeutically attractive or advisable. These are listed in Table 55–9, along with the indication categories assigned to them by ASFA and AABB.40,41

Hypercholesterolemia

III 754

Familial hypercholesterolemia (FH) is an inherited disorder characterized by highly elevated levels of circulating LDL, cholesterol (650 to 1000 mg/dL), and lipoprotein (a) (Lpa). In homozygotes, cholesterol deposits in the form of skin xanthomas and coronary atheromas develop in the first decade of life, and death from myocardial infarction before age 20 years is common. Heterozygotes also have elevated LDL, cholesterol (350 to 500 mg/dL), and Lpa levels; xanthomas may develop by age 20 years, and coronary atherosclerosis, by age 30 years. A genetically determined deficiency of cell surface LDL receptors in these patients interferes with cholesterol off-loading from LDL into cells and with the normal negative-feedback regulation of LDL synthesis, leading to the elevations in LDL and cholesterol.428,429 Less severe forms of hypercholesterolemia can be influenced by dietary modifications and are amenable to therapy with several categories of drugs, including 3-hydroxy-3methylglutaryl coenzyme A reductase inhibitors, bile acid–binding resins, nicotinic acid, and ezetimibe, a new agent that inhibits gut absorption of both dietary and biliary cholesterol.429 However, FH homozygotes and some FH heterozygotes respond poorly to these measures and remain at high risk for premature death. For such patients, repeated physical removal of LDL and its associated cholesterol can be accomplished by various modalities of therapeutic apheresis. A 1-plasma-volume TPE lowers LDL and cholesterol levels by 50% or more, and long-term treatment every 1 to 2 weeks can bring about shrinkage of cutaneous xanthomas and regression of coronary deposits.430–433 TPE removes both LDL and Lpa, but it also depletes high-density lipoproteins (HDLs), which are thought to have a salutary antiatherogenic action. This apparent disadvantage has engendered efforts to

Table 55–9 ASFA/AABB Indication Categories for Therapeutic Plasma Exchange in Metabolic Disorders Disease

Indication Category*

Homozygous familial hypercholesterolemia Refsum disease Overdose or poisoning Acute hepatic failure

II (I for LDL-P) I III III

* I, standard first-line therapy; II, second-line therapy; III, controversial; IV, no efficacy. LDL-P, low-density lipoprotein apheresis (i.e., selective depletion of LDL).

remove LDL semiselectively and on-line from patient plasma separated by an apheresis device and then return the LDLdepleted plasma to the patient. Several systems that have been designed to accomplish this goal are listed in Table 55–10. Secondary filtration systems use a plasma filter with a pore size that retains the very large LDL molecule while sieving smaller ones such as albumin434; however, these systems typically remove about half of plasma HDL.435 In a heparin extracorporeal LDL precipitation (HELP) system, LDL precipitated by heparin at acid pH is removed by filtration from patient plasma that is then dialyzed on-line to restore a physiologic pH.39,436 Lpa is also depleted, and HDL levels decline by 15%.435 LDL immunoadsorption columns contain an LDL-specific antibody linked to Sepharose particles.437,438 The need to reuse columns to achieve cost effectiveness is an awkward feature of this technology.435 A final system, the Kaneka Liposorber, uses a pair of regenerable dextran sulfate columns that absorb LDL but not HDL or Lpa.38,439 All these systems remove LDL effectively, but only the dextran sulfate and HELP systems are FDA approved and commercially available in the United States. FDA eligibility guidelines for these devices specify the following minimum LDL cholesterol levels on maximal drug therapy: more than 500 mg/dL for homozygous FH; more than 300 mg/dL for heterozygous FH; and more than 200 mg/dL for heterozygous FH with documented coronary artery disease.440 Although these systems reduce HDL considerably less than TPE for a given decrement in LDL, they have no application other than the treatment of hypercholesterolemia. Consequently, a center acquiring one of them must use it regularly on multiple patients to make it economically feasible.

Refsum Disease Refsum disease is an inherited disorder caused by deficiency of the peroxisomal enzyme phytanoyl-CoA hydroxylase, which participates in the normal degradation of phytanic acid by α-oxidation. Symptoms usually begin in the third decade of life as a result of accumulation of diet-derived phytanic acid in plasma lipoproteins and in tissue lipid stores. Peripheral neuropathy, cerebellar ataxia, and retinitis pigmentosa are found in almost all cases. Anosmia, deafness, ichthyosis, renal failure, and arrhythmias may also occur. Slow progression is the usual course, but abrupt deterioration, including sudden death, may follow a marked increase in plasma phytanic acid levels.441 Early diagnosis is important so that dietary intake of phytanic acid through dairy products, meats, and ruminant fats can be curtailed. Diet is the mainstay of treatment, leading to gradual clearing of phytanate stores by slow ω-oxidation and gradual symptomatic improvement in most patients. Adequate nutrition must be maintained, however, because overly rapid

Table 55–10 Selective Extraction Methods for Lipoproteins Immunoadsorption with anti-LDL antibody Secondary membrane filtration Heparin precipitation (HELP system) Chemical adsorption to dextran sulfate HELP, heparin extracorporeal low-density lipoprotein precipitation; LDL, low-density lipoprotein.

Drug Overdose and Poisoning Toxic effects may arise from intentional or inadvertent exposure to excessive doses of pharmacologic agents as well as from harmful agents encountered in the environment. Management techniques are similar for both types of events and include attempts to remove toxin still in the gastrointestinal tract, efforts to promote or enhance renal elimination, and direct removal from the blood by hemodialysis, hemoperfusion (e.g., over charcoal columns), or TPE.444 If available, specific antidotes may also be given.445 Most patients are treated with more than one modality. TPE has been tried in a number of patients with drug overdose or poisoning. It has been reported to be beneficial, when used along with other therapies, in cases involving substances that bind tightly to plasma proteins.446,447 Examples include methylparathion,448 vincristine,449 vinblastine,450 cisplatin,451 carbamazipine,452 and digoxin bound to antidigoxin Fab fragments.453 It has also been reported for severe hyperthyroidism, either endogenous or exogenous, although its effectiveness is limited by extensive binding of l-thyroxine to tissue proteins.454 TPE has been used in several cases of poisoning caused by ingestion of the Amanita phalloides mushroom455,456; however, diuresis has been shown to clear far more Amanita toxin.457 Most of the literature on this topic is older and all of it is anecdotal. Furthermore, TPE has always been used in combination with other, presumably effective therapies, making it difficult to formulate firm, rational guidelines. Nevertheless, it seems reasonable to offer TPE to severely affected patients with poisoning or overdose who have high blood levels of an agent known to bind to plasma proteins. Conversely, TPE has shown minimal or no beneficial effect in overdose of drugs known to bind to tissue proteins and lipids, including barbiturates,458 chlordecone,459 aluminum,460 tricyclic antidepressauts,461 benzodiazepines,458 quinine,462 phenytoin,463 digoxin, digitoxin,464 prednisone, prednisolone,465 tobramycin,466 and propranolol.467

Acute Hepatic Failure Acute hepatic failure (AHF) is an uncommon condition that arises from a severe liver insult. Hepatitis B and acetaminophen overdosage are important causes. Additional cases are due to drug reactions, Wilson disease, vascular anomalies, acute fatty liver of pregnancy, and a variety of toxins. AHF results in myriad metabolic imbalances and synthetic defects. Clinical symptoms include jaundice, coagulopathy, renal failure, and encephalopathy. Liver transplantation is the treatment of choice, leading to 60% to 80% long-term survival versus greater than 60% mortality for patients without transplantation. Many fatal outcomes are due to complications of cerebral edema.468–470

Conservative therapy is essentially supportive. Fluid, electrolyte, and nutritional supplements are adjusted to correct metabolic abnormalities. Bowel sterilization with enteral antibiotics is recommended to minimize ammonia production by intestinal flora. Pressors are given as needed for hemodynamic support, and plasma products are infused to combat coagulopathy. Osmotic diuretics, sedatives, hyperventilation, and proper positioning are all used to reduce intracranial pressure.468–470 TPE with plasma replacement has inherent appeal as a strategy to restore metabolic homeostasis, remove toxic metabolites that may cause cerebral edema, and supply deficient plasma proteins, such as coagulation factors, in quantity without inducing volume overload. However, evaluations of the effectiveness of this approach have been mixed.469 Some investigators have thought that TPE is helpful in stabilizing patients and maintaining them until an organ became available for transplantation.471 Improvements in neurologic status, blood pressure, and cerebral blood flow were attributed to exchanges in one study472; however, intracranial pressure, which is a key prognostic indicator, did not decrease. Hemoperfusion over activated charcoal, which reduces plasma ammonia levels, has also shown no advantage over intensive supportive care alone.470 Potential problems with extensive TPE arise from the diminished ability of patients with AHF to metabolize the citrate in infused plasma. Citrate accumulation leads to ionized hypocalcemia and to alterations in arterial ketone body ratios that may interfere with regeneration of hepatocytes.473 Thus although TPE can partially reverse coagulopathy and other synthetic deficits in these patients, a favorable net impact on outcome has been difficult to demonstrate, and it is seldom used for this indication in the United States.474 Other methods for extracorporeal support of patients with liver failure have been investigated. Several devices place patient plasma in “metabolic contact” with hepatocyte suspensions of either human or porcine origin. It is proposed that such devices could provide both detoxification and synthetic functions; however, evidence for the latter has been disappointing. Other devices aim to remove toxins via binding to extracorporeal albumin molecules. None of these approaches has yet been shown to improve outcomes for acute liver failure patients in a large controlled trial, but research is ongoing.474,475

CONCLUSION It can be seen from the foregoing that TPE is an effective therapy for a number of disorders, especially those mediated by autoreactive antibodies or paraproteins. Refinements in instrumentation that are outside the scope of this chapter have made it a safe therapy as well; thus it should play an important role in the treatment of selected diseases for the foreseeable future. REFERENCES 1. Abel JJ, Rowntree LG, Turner BB. Plasma removal with return of corpuscles (plasmapheresis). J Pharmacol Exp Ther 1914;5:625. 2. Skoog WA, Adams WS. Plasmapheresis in a case of Waldenström’s macroglobulinemia. Clin Res 1959;7:96. 3. Schwab PJ, Fahey JL. Treatment of Waldenström’s macroglobulinemia by plasmapheresis. N Engl J Med 1960;263:574. 4. Cohn EJ, Tullis JL, Surgenor DM, et al. Biochemistry and biomechanics of blood collection, processing and analysis: abstract of paper

THERAPEUTIC PLASMA EXCHANGE

catabolism of endogenous fat can increase plasma phytanic acid levels acutely and cause clinical exacerbations.441 TPE has been reported in a number of cases of Refsum disease and removes large quantities of phytanic acid incorporated into plasma lipids.442 Selective lipoprotein depletion is also effective.443 Apheresis is probably most appropriate for patients who have very high plasma phytanate levels, particularly those with exacerbation of symptoms. Skin disease, neuropathic symptoms, and ataxia usually improve as plasma levels decrease. Cranial nerve defects usually do not.441

55 755

TRANSFUSION MEDICINE

5. 6. 7. 8. 9. 10. 11. 12. 13.

14. 15. 16. 17. 18. 19. 20.

III 756

21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.

presented at the National Academy of Science, New Haven, Conn., Nov 5–7, 1951. Science 1951;114:479. Latham A. Early developments in blood cell separation technology. Vox Sang 1986;51:249. De Laval CGP. U.S. Patent 247,804, 1881. Freireich EJ, Judson G, Levin RH. Separation and collection of leukocytes. Cancer Res 1965;25:1516. Powell L. Intense plasmapheresis in the pregnant Rh sensitized woman. Am J Obstet Gynecol 1968;101:153. Bowman JM, Peddle LJ, Anderson C. Plasmapheresis in severe Rh isoimmunization. Vox Sang 1968;15:272. Clarke CA, Elson CJ, Bradley J, et al. Intensive plasmapheresis as a therapeutic measure in Rhesus-immunised women. Lancet 1970;2:793. Verrier Jones J, Cumming RH, Bucknall RC, et al. Plasmapheresis in the management of acute systemic lupus erythematosus. Lancet 1976;1:709. Verrier Jones J, Cumming RH, Bacon PA, et al. Evidence for a therapeutic effect of plasmapheresis in patients with systemic lupus erythematosus. Q J Med 1979;448:555. Chopek M, McCullough J. Protein and biochemical changes during plasma exchange. In Berkman EM, Umlas J (eds). Therapeutic Hemapheresis: A Technical Workshop. Washington, D.C., American Association of Blood Banks, 1980, p 13. Kellogg RM, Hester JP. Kinetics modeling of plasma exchange: Intraand post-plasma exchange. J Clin Apheresis 1988;4:183. Orlin JB, Berkman EM. Partial plasma exchange using albumin replacement: removal and recovery of normal plasma constituents. Blood 1980;56:1055. McLeod BC, Sassetti RJ, Stefoski D, Davis FA. Partial plasma protein replacement in therapeutic plasma exchange. J Clin Apheresis 1983;1:115. Waldmann TA, Strober W. Metabolism of immunoglobulins. Prog Allergy 1969;13:1 Strober W, Wochner RD, Barlow MH, et al. Immunoglobulin metabolism in ataxia telangiectasia. J Clin Invest 1968;47:1905. Wells JV, Fudenberg HH. Metabolism of radio-iodinated IgG in patients with abnormal serum IgG levels. 1: Hypergamma-globulinaemia. Clin Exp Immunol 1971;9:761. Wells JV, Fudenberg HH. Metabolism of radio-iodinated IgG in patients with abnormal serum IgG levels. II. Hypogamma-globulinaemia. Clin Exp Immunol 1971;9:775. Brambell FWR, Hemmings WA, Morris IG. A theoretical model of γ-globulin catabolism. Nature 1964;203:1352. Junghans RP, Anderson CL. The protection receptor for IgG catabolism is the β2-microglobulin-containing neonatal intestinal transport receptor. Proc Natl Acad Sci U S A 1996;93:5512. Dau PC. Immunologic rebound. J Clin Apheresis 1995;10:210. Junghans RP. IgG biosynthesis: No “immunoregulatory feedback.” Blood 1997;90:3815. Yu Z, Lennon VA. Mechanism of intravenous immune globulin therapy in antibody-mediated autoimmune diseases. N Engl J Med 1999;340:227. Lewis EJ, Hunsicker LG, Lan SP, et al. A controlled trial of plasmapheresis therapy in severe lupus nephritis. N Engl J Med 1992;326:1371. Pool M, McLeod BC. Pyrogen reactions to human serum albumin in plasma exchange. J Clin Apheresis 1995;10:81. Keller AJ, Urbaniak SJ. Intensive plasma exchange on the cell separator: effects on serum immunoglobulins and complement components. Br J Haematol 1978;38:531. Cochrane Injuries Group Albumin Reviewers. Human albumin administration in critically ill patients: Systematic review of randomized controlled trials. 1998;317:235. Wilkes MM, Navickis RJ. Patient survival after human serum albumin administration: a meta-analysis of randomized, controlled trials. Ann Intern Med 2001;135:149–164. Finfer S, Bellomo R, Boyce N, et al. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med 2004;350:2247–2256. Rossi EC. Plasma exchange in the thrombotic microangiopathies. In Rossi EC, Simon TL, Moss GS, Gould SA (eds). Principles of Transfusion Medicine, 2nd ed. Baltimore, Williams & Wilkins, 1996, p 577. McLeod BC, Price TH, Owen H, et al. Frequency of immediate adverse effects associated with therapeutic apheresis. Transfusion 1999;39:282. Owen HG, Brecher ME. Partial colloid replacement for therapeutic plasma exchange. J Clin Apheresis 1997;12:146. McLeod B, Sassetti R. Plasmapheresis with return of cryoglobulin depleted autologous plasma (cryoglobulinpheresis) in cryoglobulinemia. Blood 1980;55:866.

36. Vamvakas EC, Pineda AA. Selective extraction of plasma constituents. In McLeod BC, Price TH, Drew MJ (eds). Apheresis: Principles and Practice. Bethesda, Md., AABB Press, 1997, p 378. 37. Gjöstrup P, Watt RM. Therapeutic protein A immunoadsorption: A review. Transfus Sci 1990;11:281. 38. Gordon BR, Kelsy SF, Dau PC, et al. Long-term effects of lowdensity lipoprotein apheresis using an automated dextran sulfate cellulose adsorption system: Liposorber Study Group. Am J Cardiol 1998;81:407. 39. Jaeger BR. The HELP system for treatment of atherothrombotic disorders. Ther Apheresis Dial 2003;7:391–396. 40. McLeod BC. Introduction to the third special issue on clinical applications of therapeutic apheresis. J Clin Apheresis 2000;15:1. 41. Smith JW, Weinstein R, Hillyer KL for the AABB Hemapheresis Committee. Therapeutic apheresis: a summary of current indication categories endorsed by the AABB and the American Society for Apheresis. Transfusion 2003;43:820–822. 42. Consensus Conference. The utility of therapeutic plasmapheresis for neurological disorders. JAMA 1986;256:1333. 43. Ropper AH. The Guillain-Barré syndrome. N Engl J Med 1992;326:1130. 44. van der Meché FGA. The Guillain-Barré syndrome. Baillieres Clin Neurol 1994;3:73. 45. Hughes RA, Rees JH. Guillain-Barré syndrome. Curr Opin Neurol 1994;7:386. 46. Cook SD, Dowling PC, Murray MR, Whitaker JN. Circulating demyelinating factors in acute idiopathic polyneuropathy. Arch Neurol 1971;24:136. 47. Feasby TE, Hahn AF, Gilbert JJ. Passive transfer studies in GuillainBarré polyneuropathy. Neurology 1982;32:1159. 48. Koski CI, Gratz E, Sutherland J, et al. Clinical correlation with antiperipheral myelin antibodies in Guillain-Barré syndrome. Ann Neurol 1986;19:573. 49. McFarlin DE. Immunological parameters in Guillain-Barré syndrome. Ann Neurol 1990;27(Suppl 1):525. 50. Vriesendorp FJ, Mayer RF, Koski CL. Kinetics of anti-peripheral nerve myelin antibody in patients with Guillain-Barré syndrome treated and not treated with plasmapheresis. Arch Neurol 1991;48:858. 51. Vriesendorp FJ, Mishu B, Blaser MJ, Koski CL. Serum antibodies to GM1, GD16, peripheral nerve myelin, and Campylobacter jejuni in patients with Guillain-Barré syndrome and controls. Ann Neurol 1993;34:130. 52. Rees JH, Gregson NA, Hughes RA. Anti-ganglioside GM1 antibodies in Guillain-Barré syndrome and their relationship to Campylobacter jejuni infection. Ann Neurol 1995;38:809. 53. Rees JH, Soudain SE, Gregson NA, Hughes RAC. Campylobacter jejuni infection and Guillain-Barré syndrome. N Engl J Med 1995;333:1374. 54. Willison HJ, Kennedy PGE. Gangliosides and bacterial toxins in Guillain-Barré syndrome. J Neuroimmunol 1993;46:105. 55. Yuki N, Taki T, Inagaki F, et al. A bacterium lipopolysaccharide that elicits Guillain-Barré syndrome has a GM1, ganglioside-like structure. J Exp Med 1993;178:1771. 56. McKhann GM, Cornblath DR, Griffen JW, et al. Acute motor axonal neuropathy: A frequent cause of flaccid paralysis in China. Ann Neurol 1993;33:333. 57. Jacobs BC, Rothbarth PH, van der Meche FG, et al. The spectrum of antecedent infections in Guillain-Barré syndrome: a case-control study. Neurology 1998;51:1110–1115. 58. Guillain-Barré Syndrome Steroid Trial Group. Double-blind trial of intravenous methylprednisone in Guillain-Barré syndrome. Lancet 1993;341:586. 59. Brettle RP, Gross MLP, Legg NJ, et al. Treatment of acute polyneuropathy by plasma exchange. Lancet 1978;2:1100. 60. Schooneman F, Janot C, Streiff F, et al. Plasma exchange in GuillainBarré syndrome: ten cases. Plasma Ther 1981;2:117. 61. Valbonesi M, Garelli S, Mosconi L, et al. Plasma exchange as a therapy for Guillain-Barré syndrome with immune complexes. Vox Sang 1981;41:74. 62. de Jager AEJ, The TH, Smit Sibinga CTS, Das PC. Plasma exchange in the Guillain Barré syndrome. Br Med J 1981;283:794. 63. Rumpl E, Mayr U, Gerstenbrand F, et al. Treatment of Guillain-Barré syndrome by plasma exchange. J Neurol 1981;225:207. 64. Greenwood RJ, Newsom-Davis J, Hughes RA, et al. Controlled trial of plasma exchange in acute inflammatory polyradiculoneuropathy. Lancet 1984;1:877. 65. Osterman PG, Lundemo G, Pirskanen R, et al. Beneficial effects of plasma exchange in acute inflammatory polyradiculoneuropathy. Lancet 1984;2:1296. 66. The Guillain-Barré Syndrome Study Group. Plasmapheresis and acute Guillain-Barré syndrome. Neurology 1985;35:1096.

93. Dyck PJ, Litchy WJ, Kratz KM, et al. A plasma exchange versus immune globulin infusion trial in chronic inflammatory demyelinating polyradiculoneuropathy. Ann Neurol 1994;36:838. 94. Kyle RA. Monoclonal proteins in neuropathy. Neurol Clin 1992; 10:713. 95. Bosch EP, Smith BE. Peripheral neuropathies associated with monoclonal proteins. Med Clin North Am 1993;77:125. 96. Baldini L, Nobile-Orazio E, Guffanti A, et al. Peripheral neuropathy in IgM monoclonal gammopathy and Waldenström’s macroglobulinemia: a frequent complication in elderly males with low MAG-reactive serum monoclonal component. Am J Hematol 1994;45:25. 97. Nemni R, Gerosa E, Piccolo G, Merlini G. Neuropathies associated with monoclonal gammopathies. Haematologica 1994;79:557. 98. Gabriel JM, Erne B, Miescher GC, et al. Expression patterns of human PNS myelin proteins in neuropathies associated with anti-myelin antibodies. Schweiz Arch Neurol Psychiatry 1994;145:22. 99. Tatum AH. Experimental paraprotein neuropathy: demyelination by passive transfer of human IgM anti-myelin-associated glycoprotein. Ann Neurol 1993;33:502. 100. Smith IS. The natural history of chronic demyelinating neuropathy associated with benign IgM paraproteinemia: a clinical and neurophysiological study. Brain 1994;117:949. 101. Simmons Z, Albers JW, Bromberg MB, Feldman EL. Long-term follow-up of patients with chronic inflammatory demyelinating polyradiculoneuropathy, without and with monoclonal gammopathy. Brain 1995;118:359. 102. Nobile-Orazio E, Barbieri S, Baldini L, et al. Peripheral neuropathy in monoclonal gammopathy of undetermined significance: prevalence and immunopathogenetic studies. Acta Neurol Scand 1992;85:383. 103. Miralles GD, O’Fallon JR, Talley NJ. Plasma-cell dyscrasia with polyneuropathy: The spectrum of POEMS syndrome. N Engl J Med 1992;327:1919. 104. Dyck PJ, Low PA, Windebank AJ, et al. Plasma exchange in polyneuropathy associated with monoclonal gammopathy of undetermined significance. N Engl J Med 1991;325:1482. 105. Siciliano G, Moriconi L, Gianni G, Richieri E. Selective techniques of apheresis in polyneuropathy associated with monoclonal gammopathy of undetermined significance. Acta Neurol Scand 1994;89:117. 106. Oksenhendler E, Chevret S, Leger JM, et al. Plasma exchange and chlorambucil in polyneuropathy associated with monoclonal IgM gammopathy. J Neurol Neurosurg Psychiatry 1995;59:243. 107. Bleasel AF, Hawke SH, Pollard JD, McLeod JG. IgG monoclonal paraproteinemia and peripheral neuropathy. J Neurol Neurosurg Psychiatry 1993;56:52. 108. Richman DP, Agius MA. Treatment of autoimmune myasthenia gravis. Neurology 2003;61:1652–1661. 109. Romi F, Gilhus NE, Aarli JA. Myasthenia gravis: clinical, immunological, and therapeutic advances. Acta Neurol Scand 2005;111:134–141. 110. Masselli, RA. Pathophysiology of myasthenia gravis and LambertEaton syndrome. Neurol Clin 1994;12:285. 111. Richman DP, Wollman RL, Maselli RA, et al. Effector mechanisms of myasthenic antibodies. Ann N Y Acad Sci 1993;681:264. 112. Dau PC, Lindstrom JM, Cassel CK, et al. Plasmapheresis and immunosuppressive drug therapy in myasthenia gravis. N Engl J Med 1977; 297:1134. 113. Gajdos P, Chevret S, Toyka K. Plasma exchange for myasthenia gravis. Cochrane Database Sys Rev 2002;4:CD002275. 114. d’Empaire G, Hoaglin DC, Perlo VP, Pontoppidan H. Effect of prethymectomy plasma exchange on postoperative respiratory function in myasthenia gravis. J Thorac Cardiovasc Surg 1985;89:592. 115. Rodnitzy RL, Bosch EP. Chronic long-interval plasma exchange in myasthenia gravis. Arch Neurol 1984;41:715. 116. Splendiani G, Passlacqua S, Barbera G, et al. Semi-selective immunoadsorption treatment in myasthenia gravis. Biomater Artif Cells Immobilization Biotechnol 1992;20:1145. 117. Ichikawa M, Koh CS, Hata Y, et al. Immunoadsorption plasmapheresis for severe generalized myasthenia gravis. Arch Dis Child 1993;69:236. 118. Sawada K, Malchesky PS, Koo AP. Myasthenia gravis therapy: Immunoadsorbent may eliminate need for plasma products. Cleve Clin J Med 1993;60:60. 119. Berta E, Confalonieri P, Simoncini O, et al. Removal of antiacetylcholine receptor antibodies by protein-A immunoadsorption in myasthenia gravis. Int J Artif Organs 1994;17:603. 120. Benny WB, Sutton DMC, Oger J, et al. Clinical evaluation of a staphylococcal protein A immunoadsorption system in the treatment of myasthenia gravis patients. Transfusion 1999;39:682. 121. Gajdos P, Chevret S, Toyka K. Intravenous immunoglobulin for myasthenia gravis. Cochrane Database Sys Rev 2002;4:CD002277.

THERAPEUTIC PLASMA EXCHANGE

67. French Cooperative Group on Plasma Exchange and Guillain-Barré Syndrome. Efficiency of plasma exchange in Guillain-Barré syndrome: role of replacement fluids. Ann Neurol 1987;22:753. 68. French Cooperative Group on Plasma Exchange in Guillain-Barré Syndrome. Plasma exchange in Guillain-Barré syndrome: one-year follow-up. Ann Neurol 1992;32:94. 69. Jansen PW, Perkin RM, Ashwal S. The Guillain-Barré syndrome in childhood: natural course and efficacy of plasmapheresis. Pediatr Neurol 1993;9:16. 70. Bouget J, Chevret S, Chastang C, Raphael J-C, et al. Plasma exchange in Guillain-Barré syndrome: results from the French prospective, doubleblind, randomized, multicenter study. Crit Care Med 1993;21:651. 71. McLeod BC. The technique of therapeutic apheresis. J Crit Illness 1991;6:487. 72. van der Meché FGA, Schmitz PIM, The Dutch Guillain-Barré Study Group. A randomized trial comparing intravenous immune globulin and plasma exchange in Guillain-Barré syndrome. N Engl J Med 1992;326:1123. 73. Plasma Exchange/Sandoglobulin Guillain-Barré Syndrome Trial Group. Randomised trial of plasma exchange, intravenous immunoglobulin, and combined treatments in Guillain-Barré syndrome. Lancet 1997;349:225. 74. Hughes RAC, Wijdicks EFM, Borohn R, et al. Practice parameter: immunotherapy for Guillain-Barré syndrome. Neurology 2003;61:736–740. 75. Mendell JR. Chronic inflammatory demyelinating polyradiculopathy. Annu Rev Med 1993;44:211. 76. Glass JD, Cornblath DR. Chronic inflammatory demyelinating polyneuropathy and paraproteinemic neuropathies. Curr Opin Neurol 1994;7:393. 77. Köller H, Kieseier BC, Jander S, Hartung H-P. Chronic inflammatory demyelinating polyneuropathy. N Engl J Med 2005;352:1343–1356. 78. Saida K, Sumner AJ, Saida T, et al. Antiserum-mediated demyelination: relationship between remyelination and functional recovery. Ann Neurol 1980;8:12. 79. Yan WX, Taylor J, Andrias-Kauba S, Pollard JD. Passive transfer of demyelination by serum or IgG from chronic inflammatory demyelinating polyneuropathy patients. Ann Neurol 2000;47:765–775. 80. Simone IL, Annunziata P, Maimone D, et al. Serum and CSF antiGM1 antibodies in patients with Guillain-Barré syndrome and chronic inflammatory demyelinating polyneuropathy. J Neurol Sci 1993;114:49. 81. Khalili-Shirazi A, Atkinson P, Gregson N, Hughes RAC. Antibody response to P0 and P2 myelin proteins in Guillain-Barré syndrome and chronic inflammatory demyelinating polyradiculoneuropathy. J Neuroimmunol 1993;46:245. 82. Connolly AM, Pestronk A, Trotter JL, et al. High-titer selective antibeta-tubulin antibodies in chronic demyelinating polyneuropathy. Neurology 1993;43:557. 83. Dyck PJ, Daube J, O’Brien P, et al. Plasma exchange in chronic inflammatory demyelinating polyradiculoneuropathy. N Engl J Med 1986;314:461. 84. Hahn AF, Bolton CF, Pillay N, et al. Plasma-exchange therapy in chronic inflammatory demyelinating polyneuropathy: a double-blind, sham-controlled, cross-over study. Brain 1996;119:1055. 85. Dalakas MC. Intravenous immunoglobulin in autoimmune neuromuscular diseases. JAMA 2004;291:2367–2375. 86. van Doom PA, Vermeulen M, Brand A, et al. Intravenous immunoglobulin treatment in patients with chronic inflammatory demyelinating polyneuropathy. Arch Neurol 1991;48:217. 87. Faed JM, Day B, Pollack M, et al. High-dose intravenous human immunoglobulin in chronic inflammatory demyelinating polyneuropathy. Neurology 1989;39:422. 88. van Doom PA, Brand A, Strengers PFW, et al. High-dose intravenous immunoglobulin treatment in chronic inflammatory demyelinating polyneuropathy. Neurology 1990;40:209. 89. Vermeulen M, van Doom PA, Brand A, et al. Intravenous immunoglobulin treatment in patients with chronic inflammatory demyelinating polyneuropathy: A double blind, placebo controlled study. J Neurol Neurosurg Psychiatry 1993;56:36. 90. Hahn AF, Bolton CF, Zochodne D, Feasby TE. Intravenous immunoglobulin treatment in chronic inflammatory demyelinating polyneuropathy: a double-blind, placebo-controlled, cross-over study. Brain 1996;119:1067. 91. Hughes R, Bensa S, Willison H, et al. Randomized controlled trial of intravenous immunoglobulin versus oral prednisone in chronic inflammatory demyelinating polyneuropathy. Ann Neurol 2001;50:195–201. 92. Choudhary PP, Hughes RAC. Long-term treatment of chronic inflammatory demyelinating polyradiculoneuropathy with plasma exchange or intravenous immunoglobulin. Q J Med 1995;88:493.

55 757

TRANSFUSION MEDICINE

III 758

122. Gajdos P, Chevret S, Clair B, et al. Clinical trial of plasma exchange and high-dose intravenous immunoglobulin in myasthenia gravis. Ann Neurol 1997;41:789. 123. Al-Qureshi AI, Choudry MA, Akber MS, et al. Plasma exchange vs. intravenous immunoglobulin treatment in myasthenic crisis. Neurology 1999;52:629. 124. Newsom-Davis J. Lambert-Eaton myasthenic syndrome. Springer Semin Immunopathol 1985;8:129. 125. Jablecki C. Lambert-Eaton myasthenic syndrome. Muscle Nerve 1984;7:250. 126. O’Neill JH, Murray NM, Newsom-Davis J. The Lambert-Eaton myasthenic syndrome: A review of 50 cases. Brain 1988;11:577. 127.Takamori M. Lambert-Eaton myasthenic syndrome as an autoimmune calcium channelopathy. Biochem Biophys Res Commun 2004; 322:1347–1351. 128. Rosenfield MR, Wong E, Dalmace J, et al. Cloning and characterization of a Lambert-Eaton myasthenic syndrome antigen. Ann Neurol 1993;33:113. 129. Kokontis L, Gutmann L. Current treatment of neuromuscular diseases. Arch Neurol 2000;57:939–940. 130. Maddison P, Newsome-Davis J. Treatment for Lambert-Eaton myasthenic syndrome. Cochrane Database Syst Rev 2005;3:CD003279. 131. Chalk CH, Murray NM, Newsom-Davis J, et al. Response of the Lambert-Eaton myasthenic syndrome to treatment of associated small-cell lung carcinoma. Neurology 1990;40:1552. 132. Dau PC, Denys EH. Plasmapheresis and immunosuppressive drug therapy in the Eaton-Lambert syndrome. Ann Neurol 1982;11:570. 133. Moll JWB, Vecht CJ. Immune diagnoses of paraneoplastic neurological disease. Clin Neurol Neurosurg 1995;97:71. 134. Graus F, Rene R. Clinical and pathological advances on central nervous system paraneoplastic syndromes. Rev Neurol (Paris) 1992;148:496. 135. Sutton I, Winer JB. The immunopathogeneses of paraneoplastic neurological syndrome. Clin Sci 2002;102:475–486. 136. Lorish TR, Thorsteinsson G, Howard FM Jr. Stiff-man syndrome updated. Mayo Clin Proc 1986;64:629–636. 137. Levy LM, Dalakas MC, Floeter MK. The stiff-person syndrome: An autoimmune disorder affecting neurotransmission of γ-aminobutyric acid. Ann Intern Med 1999;131:522–530. 138. Helfgott SM. Stiff-man syndrome. Arthritis Rheum 1999;42: 1312–1320. 139. Dalakas MC, Fujii M, Li M, McElroy B. The clinical spectrum of antiGAD antibody-positive patients with stiff-person syndrome. Neurology 2000;55:1531–1535. 140. Moll JWB, Vecht CJ. Immune diagnoses of paraneoplastic neurological disease. Clin Neurol Neurosurg 1995;97:71–81. 141. Graus F, Rene R. Clinical and pathological advances in central nervous system paraneoplastic syndromes. Rev Neurol (Paris) 1992;148: 496–501. 142. Butler MH, Hayashi A, Ohkoshi N, et al. Autoimmunity to gephyrin in stiff-man syndrome. Neuron 2000;26:307–312. 143. Hao W, Davis C, Hirsch IB, et al. Plasmapheresis and immunosuppression in stiff-man syndrome with type 1 diabetes: a 2-year study. J Neurol 1999;246:731–735. 144. Oguni H, Andermann F, Rasmussen TB. The natural history of the syndrome of chronic encephalitis and epilepsy: a study of the MNI series of forty-eight cases. In Andermann F (ed). Chronic Encephalitis and Epilepsy: Rasmussen’s Syndrome. Boston, Butterworth-Heinemann, 1991, p 7. 145. Bien CG, Widman G, Urbach H, et al. The natural history of Rasmussen’s encephalitis. Brain 2002;125:1751–1759. 146. Rodgers SW, Andrews PI, Gahring LC, et al. Autoantibodies to glutamate receptor Glu R3 in Rasmussen’s encephalitis. Science 1994; 265:648. 147. Bien CG, Bauer J, Dechwerth TL, et al. Destruction of neurons by cytotoxic T cells: a new pathogenic mechanism in Rasmussen’s encephalitis. Ann Neurol 2002;51:311–318. 148. Andrews PI, McNamara JO. Rasmussen’s encephalitis: An autoimmune disorder? Curr Opin Neurobiol 1996;6:673. 149. Andrews PI, Dichter MA, Berkovic SF, et al. Plasmapheresis in Rasmussen’s encephalitis. Neurology 1996;46:242. 150. Hart YM, Cortez M, Andermann F, et al. Medical treatment of Rasmussen’s syndrome (chronic encephalitis and epilepsy): effect of high dose steroids or immunoglobulins in 19 patients. Neurology 1994;44:1030. 151. Gronata T, Fusco L, Gobbi G, et al. Experience with immunomodulatory treatments in Rasmussen’s encephalitis. Neurology 2003;61:1807–1810. 152. Swedo SE, Leonard HL, Schapiro MB, et al. Sydenham’s chorea: physical and psychological symptoms of St. Vitus’ dance. Pediatrics 1993;91:706.

153. Swedo SE, Rapoport JL, Cheslow DL, et al. High prevalence of obsessive-compulsive symptoms in patients with Sydenham’s chorea. Am J Psychiatry 1989;146:246. 154. Swedo SE, Leonard HL, Garvey M, et al. Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections: Clinical description of the first 50 cases. Am J Psychiatry 1998;155:264. 155. Husby G, van de Rijn I, Zabriskie JB, et al. Antibodies reacting with cytoplasm of subthalamic and caudate nuclei neurons in chorea and acute rheumatic fever. J Exp Med 1976;144:1094. 156. Zabriskie JB. Rheumatic fever: a model for the pathological consequences of microbial-host mimicry. Clin Exp Rheumatol 1986;4:65. 157. Snider LA, Swedo SE. Post-streptococcal autoimmune disorders of the central nervous system. Curr Opin Neurol 2003;16:359–365. 158. Singer HS. PANDAS and immunomodulatory therapy. Lancet 1999; 354:1137. 159. Garvey MA, Swedo SW, Shapiro MB, et al. Intravenous immunoglobulin and plasmapheresis as effective treatments in Sydenham’s chorea. Neurology 1996;46:A147. 160. Perlmutter SJ, Leitman SF, Garvey MA, et al. Therapeutic plasma exchange and intravenous immunoglobulin for obsessive-compulsive disorder and tic disorders in childhood. Lancet 1999;354:1153. 161. Martino D, Giovannoni G. Antibasal ganglia antibodies and their relevance to movement disorders. Curr Opin Neurol 2004;17:425–432. 162. Weinshenker BG, Sibley WA. Natural history and treatment of multiple sclerosis. Curr Opin Neurol Neurosurg 1992;5:203. 163. Owens T, Sriram S. The immunology of multiple sclerosis and its animal model, experimental allergic encephalomyelitis. Neurol Clin 1995;13:51. 164. Hafler DA, Weiner HL. Immunologic mechanisms and therapy in multiple sclerosis. Immunol Rev 1995;144:75. 165. Weiner HL. Multiple sclerosis is on inflammatory T-cell-mediated autoimmune disease. Arch Neurol 2004;61:1613–1615. 166. Genain CP, Cannella B, Hauser SL, Raine CD. Identification of autoantibodies associated with myelin damage in multiple sclerosis. Nat Med 1999;5:170–175. 167. Egg R, Reindl M, Deisenhammer F, et al. Anti-MOG and anti-MBP antibody subclasses in multiple sclerosis. Mult Scler 2001;7:285–289. 168. Berger T, Rubner P, Schautzer F, et al. Antimyelin antibodies as a predictor of clinically definite multiple sclerosis after a first demyelinating event. N Engl J Med 2003;349:139–145. 169. Noseworthy JH, Ebers GC, Vandervoort MK, et al. The impact of blinding on the results of a randomized, placebo-controlled multiple sclerosis clinical trial. Neurology 1994;44:16. 170. Jacobs L, Goodkin DE, Rudick RA, Herndon R. Advances in specific therapy for multiple sclerosis. Curr Opin Neurol 1994;7:250. 171. Rizvi SA, Bashir K. Other therapy options and future strategies for treating patients with multiple sclerosis. Neurology 2004;63(Suppl 6): S47–S54. 172. Rizvi Sa, Bashir K. Current approved options for treating patients with multiple sclerosis. Neurology 2004;63(Suppl 6):S8–S14. 173. Goodin DS, Frohman EM, Garmany GP, et al. Disease modifying therapies in multiple sclerosis: report of the therapeutics and technology assessment subcommittee of the American Academy of Neurology and the MS Council for Clinical Practice Guidelines. Neurology 2002;58:169–178. 174. Fazekas F, Deisenhammer F, Strasser-Fuchs S, et al. Randomised placebo-controlled trial of monthly intravenous immunoglobulin therapy in relapsing-remitting multiple sclerosis. Austrian Immunoglobulin in Multiple Sclerosis Study Group. Lancet 1997;349:589. 175. Sorensen PS, Wanscher B, Jensen CV, et al. Intravenous immunoglobulin G reduces MRI activity in relapsing multiple sclerosis. Neurology 1998;50:1273. 176. Achiron A, Gabbay U, Gilad R, et al. Intravenous immunoglobulin treatment in multiple sclerosis: Effect on relapses. Neurology 1998;50:398. 177. Dau PC, Petajan JM, Johnson KP, et al. Plasmapheresis and immunosuppressive drug therapy in multiple sclerosis. Neurology 1980;30:1023. 178. Weiner HL, Dawson DM. Plasmapheresis in multiple sclerosis: preliminary study. Neurology 1980;30:1029. 179. Khatri BO, McQuillen MP, Hoffman RG, et al. Plasma exchange in chronic progressive multiple sclerosis: a long term study. Neurology 1991;41:409. 180. Hauser SL, Dawson DM, Lehrich JR, et al. Intensive immunosuppression in progressive multiple sclerosis: a randomized three-arm study of high dose intravenous cyclophosphamide, plasma exchange, and ACTH. N Engl J Med 1983;308:173. 181. Tindall RSA, Walker JE, Ehle AL, et al. Plasmapheresis in multiple sclerosis: prospective trial of pheresis and immunosuppression versus immunosuppression alone. Neurology 1982;32:739.

212. Benjamin RJ, Connors JM, McGurk S, et al. Prolonged erythroid aplasia after major ABO-mismatched transplantation for chronic myelogenous leukemia. Biol Blood Marrow Transplant 1998;4:151–156. 213. Üstün C, Celebi H, Arat M, et al. Treatment of aregeneratoric anemia following an ABO-incompatible allogeneic peripheral blood stem cell transplantation: a case report. Ther Apheresis 1999;3:275–277. 214. Verholen F, Stalder M, Helg C, Chalandon Y. Resistant pure red cell aplasia after allogeneic stem cell transplantation with major ABO mismatch treated by escalating dose donor leukocyte infusion. Eur J Hematol 2004;73:441–446. 215. Fischel RJ, Ascher NL, Payne WD, et al. Pediatric liver transplantation across ABO blood group barriers. Transplant Proc 1989;21:2221. 216. Renard TH, Shimaoka S, Le Bherz D, et al. ABO incompatible liver transplantation in children: a prospective approach. Transplant Proc 1993;25:1953. 217. Mor E, Skerrett D, Manzarbeitia C, et al. Successful use of an enhanced immunosuppressive protocol with plasmapheresis for ABO-incompatible mismatched grafts in liver transplant recipients. Transplantation 1995;59:986. 218. Gloor JM, Lager DJ, Moore B. ABO-incompatible kidney transplantation using both A2 and non-A2 living donors. Transplantation 2003; 75:971–977. 219. Warren DS, Zachary AA, Sonnenday CJ, et al. Successful renal transplantation across simultaneous ABO incompatible and positive crossmatch barriers. Am J Transplant 2004;4:561–568. 220. Sonnenday CJ, Warren DS, Cooper M, et al. Plasmapheresis, CMV hyperimmune globulin, and anti-CD20 allow ABO-incompatible renal transplantation without splenectomy. Am J Transplantation 2004; 4:1315–1322. 221. Winters JL, Gloor JM, Pineda AA, et al. Plasma exchange conditioning for ABO-incompatible renal transplantation. J Clin Apheresis 2004;19:79–85. 222. Ramsey G. Red cell antibodies arising from solid organ transplants. Transfusion 1991;31:76. 223. Ramsey G, Israel L, Lindsay GD, et al. Anti-rho(D) in two Rh-positive patients receiving kidney grafts from an Rh-immunized donor. Transplantation 1986;41:67–69. 224. Lundgren G, Asaba H, Bergstrom J, et al. Fulminating anti-A autoimmune hemolysis with anuria in a renal transplant recipient: a therapeutic role of plasma exchange. Clin Nephrol 1981;16:211. 225. Kickler TS. The challenge of platelet alloimmunization: Management and prevention. Transfusion 1990;30:8. 226. Bensinger WI, Buckner CD, Clift RA, et al. Plasma exchange for platelet alloimmunization. Transplantation 1986;41:602. 227. Christie DJ, Howe RB, Lennon SS, Sauro SC. Treatment of refractoriness to platelet transfusion by protein A column therapy. Transfusion 1993;33:234. 228. Lopez-Plaza I, Miller K, Leitman SF. Ineffectiveness of protein A adsorption in the treatment of platelet refractoriness. J Clin Apheresis 1992;7:33. 229. Lee EJ, Norris D, Schiffer CA. Intravenous immune globulin for patients alloimmunized to random donor platelet transfusions. Transfusion 1987;27:245. 230. Kickler T, Braine HG, Piantadosi S, et al. A randomized, placebocontrolled trial of intravenous gammaglobulin in alloimmunized thrombocytopenic patients. Blood 1987;70:313. 231. Ziegler ZR, Shadduck RK, Rosenfeld CS, et al. High-dose intravenous gamma globulin improves responses to single-donor platelets in patients refractory to platelet transfusion. Blood 1987;70:1433. 232. Amorosi EL, Ultmann JE. Thrombotic thrombocytopenic purpura: report of 16 cases and review of the literature. Medicine (Baltimore) 1966;45:139. 233. Moake JL. Thrombotic microangiopathies. N Engl J Med 2002;347: 589–600. 234. Torok TJ, Holman RC, Chorba TL. Increasing mortality from thrombotic thrombocytopenic purpura in the United States: analysis of national mortality data, 1968–1991. Am J Hematol 1995;50:84. 235. Porta C, Caporali R, Montecucco C. Thrombotic thrombocytopenic purpura and autoimmunity: a tale of shadows and suspects. Haematologia 1999;84:260. 236. Rarick MU, Espina B, Mocharnuk R, et al. Thrombotic thrombocytopenic purpura in patients with human immunodeficiency virus infection: a report of three cases and review of the literature. Am J Hematol 1992;40:103. 237. Furlan M, Robles R, Solenthaler M, et al. Deficient activity of von Willebrand factor-cleaving protease in chronic relapsing thrombotic thrombocytopenic purpura. Blood 1997;89:3097. 238. Furlan M, Robles R, Galbusera M, et al. Von Willebrand factor-cleaving protease in thrombotic thrombocytopenic purpura and the hemolytic-uremic syndrome. N Engl J Med 1998;339:1578.

THERAPEUTIC PLASMA EXCHANGE

182. Khatri BO, McQuillen MP, Harrington GJ, et al. Chronic progressive multiple sclerosis: double-blind controlled study of plasmapheresis in patients taking immunosuppressive drugs. Neurology 1985;35:312. 183. Weiner HL. An assessment of plasma exchange in progressive multiple sclerosis. Neurology 1985;35:320. 184. Goodin DS. The use of immunosuppressive agents in the treatment of multiple sclerosis: a critical review. Neurology 1991;41:980. 185. Goodkin DE, Ransohoff RM, Rudick RA. Experimental therapies for multiple sclerosis: current status. Cleve Clin J Med 1992;59:63. 186. Weiner HL, Dau P, Khatri BO, et al. Double-blind study of true versus sham plasma exchange in patients being treated with immunosuppression for acute attacks of multiple sclerosis. Neurology 1989;39:1143. 187. The Canadian Cooperative Multiple Sclerosis Study Group. The Canadian cooperative trial of cyclophosphamide and plasma exchange in progressive multiple sclerosis. Lancet 1991;337:441. 188. Weinshenker B, O’Brien PC, Petterson TM, et al. A randomized trial of plasma exchange in acute central nervous system inflammatory demyelinating disease. Ann Neurol 1999;46:878–886. 189. Keegan M, Pineda AA, McClelland RL, et al. Plasma exchange for severe attacks of CNS demyelination: predictors of response. Neurology 2002;58:143–146. 190. McGrath MA, Penny R. Paraproteinemias: blood hyperviscosity and clinical manifestations. J Clin Invest 1996;58:1155. 191. Salmon SE, Cassady RJ. Plasma cell neoplasms. In DeVita VT, Hellman S, Rosenberg SA (eds). Cancer: Principles and Practice of Oncology. New York: Lippincott-Raven, 1997, p 2344. 192. Kipps TJ. Macroglobulinemia. In Lichtman MA, Beutler E, Kipps TJ, Seligsohn U, et al (eds). Williams Hematology, 7th ed. New York, McGraw-Hill, 2006, p 1549. 193. Gertz MA. Waldenström macroglobulinemia: a review of therapy. Am J Hematol 2005;79:147–157. 194. Glaspy JA. Hemostatic abnormalities in multiple myeloma and related disorders. Hematol Oncol Clin North Am 1992;6:1301. 195. Bear RA, Cole EH, Slang A, et al. Treatment of acute renal failure due to myeloma kidney. Can Med Assoc J 1980;123:750. 196. Knudsen LM, Hjorth M, Hippe E, for the Nordic Myeloma Study Group. Renal failure in multiple myeloma: Reversibility and impact on the prognosis. Eur J Hematol 2000;65:175–181. 197. Misiani R, Tiraboschi G, Mingardi G, Mecca G. Management of myeloma kidney: an anti-light chain approach. Am J Kidney Dis 1987; 10:28. 198. Solling K, Solling J. Clearance of Bence-Jones proteins during peritoneal dialysis or plasmapheresis in myelomatosis associated with renal failure. Contrib Nephrol 1988;68:259. 199. Johnson WJ, Kyle RA, Pineda AA, et al. Treatment of renal failure associated with multiple myeloma: plasmapheresis, hemodialysis, and chemotherapy. Arch Intern Med 1990;150:863. 200. Zucchelli P, Pasquali S, Cagnoli L, Ferrari G. Controlled plasma exchange trial in acute renal failure due to multiple myeloma. Kidney Int 1988;33:1175. 201. Kaplan AA. Therapeutic apheresis for the renal complications of multiple myeloma and the dysglobulinurias. Ther Apheresis 2001;5:171–175. 202. Fraser ID, Bothamley JE, Bennett MO, et al. Intensive antenatal plasmapheresis in severe Rhesus isoimmunisation. Lancet 1976;1:6. 203. Mollison PL, Engelfriet CP, Contreras M. Haemolytic disease of the fetus and newborn. In Blood Transfusion in Clinical Medicine. London: Blackwell Science, 1997, p 390. 204. Filbey D, Berseus O, Lindeberg S, Wesstrom G. A management programme for Rh alloimmunization during pregnancy. Early Hum Dev 1987;15:11. 205. Watson WJ, Katz VL, Bowes WA. Plasmapheresis during pregnancy. Obstet Gynecol 1990;76:451. 206. Furukawa K, Nakajima T, Kogure T, et al. Example of a woman with multiple intrauterine deaths due to anti-M who delivered a live child after plasmapheresis. Exp Clin Immunogenet 1993;10:161. 207. Shirey R, Ness P, Kickler T, et al. The association of anti-P and early abortion. Transfusion 1987;27:189. 208. Gale RP, Feig S, Ho W, et al. ABO blood group system and bone marrow transplantation. Blood 1977;50:185. 209. Bensinger WI, Baker DA, Buckner DD, et al. Immunoadsorption for removal of A and B blood group antibodies. N Engl J Med 1981; 301:160. 210. Braine HG, Sensenbrenner LL, Wright SK, et al. Bone marrow transplantation with major ABO incompatibility using erythrocyte depletion of marrow prior to infusion. Blood 1982;60:420. 211. Gmur JP, Burger J, Schaffner A, et al. Pure red cell aplasia of long duration complicating major ABO incompatible bone marrow transplantation. Blood 1990;75:290.

55 759

TRANSFUSION MEDICINE

III 760

239. Tsai H-M, Lian EC-Y. Antibodies to von Willebrand factor-cleaving protease in acute thrombotic thrombocytopenic purpura. N Engl J Med 1998;339:1585. 240. Rick ME, Austin H, Leitman SF, et al. Clinical usefulness of a functional assay for the von Willebrand factor cleaving protease (ADAMTS 13) and its inhibitor in a patient with thrombotic thrombocytopenic purpura. Am J Hematol 2004;75:96–100. 241. Sahud MA, Claster S, Liu L, et al. Von Willebrand factor-cleaving protease inhibitor in a patient with human-immunodeficiency syndromeassociated thrombotic thrombocytopenic purpura. Br J Haematol 2002;116:909–911. 242. Rock GA, Shumak KH, Buskard NA, et al. The Canadian Apheresis Study Group: comparison of plasma exchange with plasma infusion in the treatment of thrombotic thrombocytopenic purpura. N Engl J Med 1991;325:393. 243. Bell WR, Braine HG, Ness PM, Kickler TS. Improved survival in thrombotic thrombocytopenic purpura-hemolytic uremic syndrome: Clinical experience in 108 patients. N Engl J Med 1991;325:398. 244. O’Connor NTJ, Bruce-Jones P, Hill LF. Vincristine therapy for thrombotic thrombocytopenic purpura. Am J Hematol 1992;39:234. 245. Reddy PS, Deanna-Limayo D, Cook JD, et al. Rituximab in the treatment of relapsed thrombotic thrombocytopenic purpura. Ann Hematol 2005;84:232–235. 246. Onundarson PT, Rowe JM, Heal JM, Francis CW. Response to plasma exchange and splenectomy in thrombotic thrombocytopenic purpura. Arch Intern Med 1992;152:791. 247. Tsai H-M, Rue L, Sarode R, et al. Antibody inhibitors to von Willebrand factor metalloproteinase and increased binding of von Willebrand factor to platelets in ticlopidine associated thrombotic thrombocytopenic purpura. Ann Intern Med 2000;132:794. 248. Bennett CL, Weinberg PD, Rozenberg-Ben-Dor K, et al. Thrombotic thrombocytopenic purpura associated with ticlopidine: a review of 60 cases. Ann Intern Med 1998;128:541. 249. Gasser C, Gautier E, Steck A, et al. Hämolytisch-urämische Syndrome: Bilaterale Nierenrinden-nekrosen bei akuten erworbenen hämolytischen Anämien. Schweitz Med Wochenschr 1955;85:905. 250. Robson WL, Leung AKC, Kaplan BS. Hemolytic-uremic syndrome. Curr Probl Pediatr 1993;23:16. 251. Loirat C, Baudouin V, Sonsino E, et al. Hemolytic-uremic syndrome in the child. Adv Nephrol 1993;22:141. 252. George JN, Gilcher RO, Smith JW, et al. Thrombotic thrombocytopenic purpura-hemolytic uremic syndrome: diagnosis and management. J Clin Apheresis 1998;13:120. 253. Von Baeyer H. Plasmapheresis in thrombotic microangiopathy associated syndromes: review of outcome data derived from clinical trials and open studies. Ther Apheresis 2002;4:320–328. 254. George JN, Li X, McMinn JR. Thrombotic thrombocytopenic purpura-hemolytic uremic syndrome following allogeneic HPC transplantation: a diagnostic dilemma. Transfusion 2004;44:294–304. 255. Elliot MA, Nichols WL, Plumhoff EA, et al. Post transplantation thrombotic thrombocytopenic purpura: a single-center experience and a contemporary review. Mayo Clin Proc 2003;78:421–430. 256. Raife T, Atkinson B, Montgomery L, et al. Severe deficiency of VWFcleaving protease (ADAMTS 13) activity defines a distinct population of thrombotic microangiopathy patients. Transfusion 2004;44:146–150. 257. Rougier N, Kazatchkine MD, Rougier J-P, et al. Human complement factor H deficiency associated with hemolytic uremic syndrome. J Am Soc Nephrol 1998;9:2318. 258. Warwicker P, Donne RL, Goalship JA, et al. Familial relapsing haemolytic uraemic syndrome and complement factor H deficiency. Nephrol Dial Transplant 1999;14:1229. 259. Zipfel PF, Hellwage J, Friese MA, et al. Factor H and disease: a complement regulator affects vital bodily functions. Mol Immunol 1999;36:241. 260. Martin JN, Files JC, Blake PG, et al. Plasma exchange for preeclampsia. I: Postpartum use for persistently severe preeclampsia-eclampsia with HELLP syndrome. Am J Obstet Gynecol 1990;162:126. 261. Veyradier A, Girma J-P. Assays of ADAMTS-13 activity. Semin Hematol 2004;41:41–47. 262. Studt JD, Böhm M, Budde U, et al. Measurement of von Willebrand-cleaving protease (ADAMTS-13) activity in plasma: a multicenter comparison of different assay methods. J Thromb Haemost 2003;1:1882–1887. 263. Tripodi A, Chantarangkul V, Böhm M, et al. Measurement of von Willebrand factor cleaving protease (ADAMTS-13): results of an international collaborative study involving 11 methods testing the same set of coded plasmas. J Thromb Haemost 2004;2:1601–1609.

264. Vesely SK, George JN, Lämmle B, et al. ADAMTS-13 activity in thrombotic thrombocytopenic purpura-hemolytic uremic syndrome: relation of presenting features and clinical outcomes in a prospective cohort of 142 patients. Blood 2003;102:60–68. 265. Zheng X, Majerus EM, Sadler JE. ADAMTS 13 and TTP. Curr Opin Hematol 2002;9:389–394. 266. Zheng X, Kaufman RM, Goodnough LT, Sadler JE: Effect of plasma exchange on plasma ADAMTS 13 metalloprotease activity, inhibitor level, and clinical outcome in patients with idiopathic and nonidiopathic thrombotic thrombocytopenic purpura. Blood 2004;103:4043–4049. 267. Mueller-Eckhardt C. Post-transfusion purpura. Br J Haematol 1986; 64:419. 268. Shulman NR, Reid DM: Platelet immunology. In Colman RW, Hirsh J, Marder VJ, Salzman EW (eds). Hemostasis and Thrombosis. Philadelphia: JB Lippincott, 1994, p 414. 269. McCrae KR, Herman JH. Posttransfusion purpura: two unusual cases and a literature review. Am J Hematol 1996;52:205. 270. Newman PJ, McFarland JG, Aster RH. Alloimmune thrombocytopenias. In Loscalzo J, Schafer AI (eds). Thrombosis and Hemorrhage. Boston: Blackwell Scientific, 1994, p 429. 271. Laursen B, Morling N, Rosenkvist J, et al. Post-transfusion purpura treated with plasma exchange by Haemonetics cell separator. Acta Med Scand 1978;203:539. 272. Mueller-Eckhardt C, Kiefel V. High dose IgG for post-transfusion purpura-revisited. Blut 1988;57:163. 273. Diz-Küçcükkaya R, Gushiken FC, Lopez JA. Thrombocytopenia. In Lichtman MA, Beutler E, Kipps TJ, Seligsohn U, et al (eds). Williams Hematology, 7th ed. New York, McGraw-Hill, 2006, p 1749. 274. Branda RF, Tate DY, McCullough JJ, et al. Plasma exchange in the treatment of fulminant idiopathic (autoimmune) thrombocytopenic purpura. Lancet 1978;1:688. 275. Isbister JP, Biggs JC, Penny R. Experience with large volume plasmapheresis in malignant paraproteinemia and immune disorders. Aust N Z J Med 1978;8:154. 276. Marder VJ, Nusbacher J, Anderson FW. One-year follow-up of plasma exchange therapy in 14 patients with idiopathic thrombocytopenic purpura. Transfusion 1981;21:291. 277. Buskard N, Rock G, Nair R. The Canadian experience using plasma exchange for immune thrombocytopenic purpura. Transfusion Sci 1998;19:295–300. 278. Mittelman A, Bertram J, Henry DH, et al. Treatment of patients with HIV thrombocytopenia and hemolytic uremic syndrome with protein A (proSorba) immunoadsorption. Semin Hematol 1989;26(Suppl 11):15. 279. Snyder HW Jr, Cochran SK, Balint JP, et al. Experience with protein Aimmunoadsorption in treatment resistant immune thrombocytopenic purpura. Blood 1992;79:2237. 280. Packman CH. Hemolytic anemia resulting from immune injury. In Lichtman MA, Beutler E, Kipps TJ, Seligsohn U, et al (eds). Williams Hematology, 7th ed. New York, McGraw-Hill, 2006, p 729. 281. Packman CH. Cryopathic hemolytic syndromes. In Beutler E, Lichtman MA, Coller BS, Kipps TJ, Seligsohn U (eds). Williams Hematology, 6th ed. New York, McGraw-Hill, 2001, p 649. 282. Silberstein LE, Berkman EM. Plasma exchange in autoimmune hemolytic anemia (AIHA). J Clin Apheresis 1983;1:238. 283. Valbonesi M, Guzzini D, Zerbi D, et al. Successful plasma exchange for a patient with chronic demyelinating polyneuropathy and cold agglutinin disease due to anti-Pra. J Clin Apheresis 1986;3:109. 284. Kutti J, Wadenvik H, Safai-Kutti S, et al. Successful treatment of refractory autoimmune hemolytic anemia by plasmapheresis. Scand J Hematol 1984;32:149. 285. Andersen O, Taaning E, Rosenkvist J, et al. Autoimmune hemolytic anemia treated with multiple transfusions, immunosuppressive therapy, plasma exchange and desferrioxamine. Acta Paediatr Scand 1984;73:145. 286. McConnell ME, Atchison JA, Kohaut E, Castleberry RP. Successful use of plasma exchange in a child with refractory autoimmune hemolytic anemia. Am J Pediatr Hematol Oncol 1987;9:158. 287. von Keyserlingk H, Meyer-Sabellek W, Arntz R, Hiller H. Plasma exchange treatment in autoimmune hemolytic anemia of the wane antibody type with renal failure. Vox Sang 1982;52:298. 288. Geurs F, Ritter K, Mast A, Van Maele V. Successful plasmapheresis in corticosteroid-resistant hemolysis in infectious mononucleosis: Role of autoantibodies against triosephosphate isomerase. Acta Haematol 1992;88:142. 289. Segel GB, Lichtman MA. Aplastic anemia. In Lichtman MA, Beutler E, Kipps TJ, Seligsohn U, et al (eds). Williams Hematology, 7th ed. New York, McGraw-Hill, 2006, p 419.

318. Nakamura Y, Yoshida K, Itoh S, et al. Immunoadsorption plasmapheresis as a treatment for pregnancy complicated by systemic lupus erythematosus with positive antiphospholipid antibodies. Am J Reprod Immunol 1999;41:307. 319. Brouet JC, Clauvel JP, Danon F, et al. Biologic and clinical significance of cryoglobulins: a report of 86 cases. Am J Med 1974;57:775. 320. Bloch KJ. Cryoglobulinemia and hepatic C Virus. N Engl J Med 1992; 327:1521. 321. Ferri C, Zignego AL, Pileri SA. Cryoglobulins. J Clin Pathol 2002; 55:4–13. 322. Berkman EM, Orlin JB. Use of plasmapheresis and partial plasma exchange in the management of patients with cryoglobulinemia. Transfusion 1980;20:171. 323. Bombardieri S, Maggiore Q, L Abbate A, et al. Plasma exchange in essential mixed cryoglobulinemia. Plasma Ther Transfus Technol 1981; 2:101. 324. McGovern TW, Enzenauer RJ, Fitzpatrick JE. Treatment of recalcitrant leg ulcers in cryoglobulinemia types I and II with plasmapheresis. Arch Dermatol 1996;132:498. 325. Goronzy J, Weyland C. Rheumatoid arthritis. In Klippel R (ed). Primer on the Rheumatic Diseases. Atlanta, Arthritis Foundation, 1997, p 155. 326. O’Dell JR. Therapeutic strategies for rheumatoid arthritis. N Engl J Med 2004;350:2591–2602. 327. Weaver AL. The impact of new biologicals in the treatment of rheumatoid arthritis. Rheumatology 2004;43(Suppl 3):iii17–iii23. 328. Rothwell R, Davis P, Gordon P, et al. A controlled study of plasma exchange in the treatment of severe rheumatoid arthritis. Arthritis Rheum 1980;23:785. 329. Dwosh I, Giles A, Ford P. Plasmapheresis therapy in rheumatoid arthritis: A controlled, double-blind, crossover trial. N Engl J Med 1983; 308:1124. 330. Wallace D, Goldfinger D, Lowe C, et al. A double-blind, controlled study of lymphoplasmapheresis versus sham apheresis in rheumatoid arthritis. N Engl J Med 1982;306:1406. 331. Karch J, Klippel J, Plotz P, et al. Lymphapheresis in rheumatoid arthritis: A randomized trial. Arthritis Rheum 1981;24:867. 332. Hester J, Felson D, Gendreau M. Phase III trial of ProSorba column for severe rheumatoid arthritis. J Clin Apheresis 1998;13:90. 333. Pisesky D. Systemic lupus erythematosus. In Klippel J (ed). Primer on the Rheumatic Diseases. Atlanta, Arthritis Foundation, 1997, p 246. 334. Moc CC, Lau CS. Pathogenesis of systemic lupus erythematosus. J Clin Pathol 2003;56:481–490. 335. Klippel J. Systemic lupus erythematosus: treatment. In Klippel J (ed). Primer on the Rheumatic Diseases. Atlanta, Arthritis Foundation, 1997, p 258. 336. Ad Hoc Working Group. Criteria for steroid-sparing ability of interventions in systemic lupus erythematosus: report of a consensus meeting. Arthritis Rheum 2004;50:3427–3431. 337. Parry HF, Moran CJ, Snaith ML, et al. Plasma exchange in systemic lupus erythematosus. Ann Rheum Dis 1981;40:224. 338. Jones JV. Plasmapheresis in SLE. Clin Rheum Dis 1982;8:243. 339. Huston DP, White MJ, Maltiolo C, et al. A controlled trial of plasmapheresis and cyclophosphamide therapy of lupus nephritis. Arthritis Rheum 1983;26(Suppl):S33. 340. Euler HH, Schwab UM, Schroeder JO, Hasford J. The Lupus Plasmapheresis Study Group: Rationale and updated interim report. Artif Organs 1996;20:356. 341. Schroeder JO, Schwab U, Zennet R, et al. Plasmapheresis and subsequent pulse cyclophosphamide in severe systemic lupus erythematosus: preliminary results of the LPSG Trial. Arthritis Rheum 1997; 40:S325. 342. Wallace DJ, Goldfinger D, Pepkowitz S, et al. Randomized control of pulse/synchronization cyclophosphamide/apheresis for proliferative lupus nephritis. J Clin Apheresis 1998;13:163. 343. Aringer M, Smolen JS, Graninger WB. Severe infections in plasmapheresis-treated systemic lupus erythematosus. Arthritis Rheum 1998; 41:414–420. 344. Langford CA. Vasculitis. J Allergy Clin Immunol 2003;111:S602–S612. 345. Jayne D. Evidence-based treatment of systemic vasculitis. Rheumatol 2000;39:589–595. 346. Pusey CD, Rees AJ, Evans DJ, et al. Plasma exchange in focal necrotizing glomerulonephritis without anti-GBM antibodies. Kidney Int 1991;40:757. 347. Guillevin L, Fain O, Lhote F, et al. Lack of superiority of steroids plus plasma exchanges to steroids alone in the treatment of polyarteritis nodosa and Churg-Strauss syndrome: a prospective, randomized trial in 78 patients. Arthritis Rheum 1992;35:208.

THERAPEUTIC PLASMA EXCHANGE

290. Young NS. Pure red cell aplasia. In Lichtman MA, Beutler E, Kipps TJ, Seligsohn U, et al (eds). Williams Hematology, 7th ed. New York, McGraw-Hill, 2006, p 437. 291. Young NS, Barrett AJ. The treatment of severe acquired aplastic anemia. Blood 1995;85:3367. 292. Fitchen JJ, Cline MJ, Saxon A, et al. Serum inhibitors of hematopoiesis in a patient with aplastic anemia and systemic lupus erythematosus. Am J Med 1979;6:537. 293. Abdou NI. Plasma exchange in the treatment of aplastic anemia. In Tindall RSA (ed). Therapeutic Apheresis and Plasma Perfusion. New York: Alan R. Liss, 1982, p 337. 294. Young NS, Klein HG, Griffith P, et al. Therapeutic plasma exchange and lymphocyte depletion in aplastic anemia and pure red cell aplasia. In Nose Y, Malchesky PS, Smith JW, et al (eds). Plasmapheresis. New York, Raven Press, 1983, p 339. 295. Messner HA, Fause AA, Curtis JE, et al. Control of antibody-mediated pure red cell aplasia by plasmapheresis. N Engl J Med 1981;304:1334. 296. Freund LG, Hippe E, Strandgaard S, et al. Complete remission in pure red cell aplasia after plasmapheresis. Scand J Hematol 1985;35:351. 297. Khelif A, Van HV, Tremisi JP, et al. Remission of acquired pure red cell aplasia following plasma exchanges. Scand J Hematol 1985;35:13. 298. Berlin G, Lieden G. Long-term remission of pure red cell aplasia after plasma exchange and lymphocytapheresis. Scand J Haematol 1986;36:121. 299. Delgado J, Jimenez-Yuste V, Hernandez-Navarro F, Villar A. Acquired haemophilia: review and meta-analysis focused on therapy and prognostic factors. Br J Haematol 2003;121:21–35. 300. Ewenstein BM. Factor VIII and other coagulation factor inhibitors. In Loscalzo J, Schafer AI (eds). Thrombosis and Hemorrhage. Boston: Blackwell Scientific, 1994, p 729. 301. Garvey MB. Incidence and management of patients with acquired factor VIII inhibitors: the practical experience of a tertiary care hospital. In Kessler CM (ed). Acquired Hemophilia. Princeton, NJ, Excerpta Medica, 1995, p 91. 302. Kessler CM. Factor VIII inhibitors: an algorithmic approach to treatment. Semin Hematol 1994;31(Suppl 4):33. 303. Gribble J, Garvey MD. Porcine factor VIII provides clinical benefit to patients with high levels of inhibitors to human and porcine factor VIII. Haemophilia 2000;6:482–486. 304. Lusher J, Ingerslev S, Roberts H, Hedner U. Clinical experience with recombinant factor VIIa. Blood Coagul Fibrinol 1998;9:119. 305. Wensley RT, Stevens RF, Burn AM, et al. The role of intensive plasma exchange in the prevention and management of haemorrhage in patients with inhibitors to factor VIII. Br Med J 1980;281:1388. 306. Slocombe GW, Newland AC, Colvin MP, et al. The role of intensive plasma exchange in the management of hemophilia patients with inhibitors. Br J Haematol 1981;47:577. 307. Bona RD, Pasquale DN, Kalish RI, et al. Porcine factor VIII and plasmapheresis in the management of hemophilia patients with inhibitors. Am J Hematol 1986;21:201. 308. Freedman J, Garvey MB. Immunoadsorption of factor VIII inhibitors. Curr Opin Hematol 2004;11:327–333. 309. Lusher JM. Management of patients with factor VIII inhibitors. Transfus Med Rev 1987;1:123. 310. Pflieger G, Boda Z, H’arsfalvi J, et al. Cyclosporin treatment of a woman with acquired hemophilia due to factor VIII:C inhibitor. Postgrad Med J 1989;65:400. 311. Schwartz RS, Gabriel DA, Aledort LM, et al. A prospective study of treatment of acquired (autoimmune) factor VIII inhibitors with highdose intravenous gammaglobulin. Blood 1995;86:797. 312. Mathias M, Khair K, Hann I, Liesner R. Rituximab in the treatment of alloimmune factor VIII and IX antibodies in two children with severe haemophilia. Br J Haematol 2004;125:366–368. 313. Stasi R, Brunetti M, Stipa E, Amadori S. Selective B-cell depletion with rituximab for the treatment of patients with acquired hemophilia. Blood 2004;103:4424–4428. 314. Ewing NP, Sanders NL, Dietrich SL, Kaspar CK. Induction of immune tolerance to factor VIII in hemophiliacs with inhibitors. JAMA 1988;259:65. 315. Gruppo RA, Valdez LP, Stout RD. Induction of immune tolerance in patients with hemophilia A and inhibitors. Am J Pediatr Hematol Oncol 1992;4:82. 316. Nilsson IM, Berntorp E, Zettervoll O. Induction of immune tolerance in patients with hemophilia and antibodies to factor VIII by combined treatment with intravenous IgG, cyclophosphamide, and factor VIII. N Engl J Med 1988;318:947. 317. Smith OP, Hann IM. Factor VIIa therapy to secure haemostasis during central line insertion in children with high-responding FVIII inhibitors. Br J Haematol 1992;92:1002.

55 761

TRANSFUSION MEDICINE

III 762

348. Guillevin L, Mahr A, Cohen P, et al. Short-term corticosteroids then lamivudine and plasma exchanges to treat Hepatitis B virus-related polyarteritis nodosa. Arthritis Rheum 2004;51:482–487. 349. Dalakas MC. Polymyositis and dermatomyositis. Lancet 2003;362: 971–982. 350. Dalakas MC. Controlled studies with high-dose intravenous immunoglobulin in the treatment of dermatomyositis, inclusion body myositis and polymyositis. Neurology 1998;51:537. 351. Dau PC. Plasmapheresis in idiopathic inflammatory myopathy. Arch Neurol 1981;38:544. 352. Khatri BO, Luprecht G, Weiss SA. Plasmapheresis and immunosuppressive drug therapy in polymyositis. Muscle Nerve 1982;5:568. 353. Cecere FA, Spiva DA. Combination plasmapheresis/leukocytapheresis for the treatment of dermatomyositis/polymyositis. Plasma Ther Transfus Technol 1982;3:401. 354. Miller FW, Leitman SF, Cronin ME, et al. Controlled trial of plasma exchange and leukapheresis in polymyositis and dermatomyositis. N Engl J Med 1992;326:1380. 355. Salama AD, Levy JB, Lightstone L, Pusey CD. Goodpasture’s disease. Lancet 2001;358:917–920. 356. Hudson BG, Tryggvason K, Sundaramoorthy M, Nelson EG. Alport’s syndrome, Goodpasture’s syndrome and type IV collagen. N Engl J Med 2003;348:2543–2556. 357. Lockwood CM, Boulton-Jones JM, Lowenthal RM, et al. Recovery from Goodpasture’s syndrome after immunosuppressive treatment and plasmapheresis. Br Med J 1975;2:252. 358. Pusey CD, Lockwood CM, Peters DK. Plasma exchange and immunosuppressive drugs in the treatment of glomerulonephritis due to antibodies to the glomerular basement membrane. Int J Artif Organs 1983;6:15. 359. Glassock RJ. Intensive plasma exchange in crescentic glomerulonephritis: help or no help? Am J Kidney Dis 1992;20:270. 360. Madore F, Lazarus JM, Brady HR. Therapeutic plasma exchange in renal diseases. J Am Soc Nephrol 1997;7:367. 361. Johnson JP, Moore J, Austin HA, et al. Therapy of anti-glomerular basement membrane antibody disease: Analysis on the prognostic significance of clinical, pathologic, and treatment factors. Medicine (Baltimore) 1985;64:219. 362. Stegmayr BG, Almroth G, Berlin G, et al. Plasma exchange or immunoadsorption in patients with rapidly progressive glomerulonephritis: a Swedish multicenter study. Int J Artif Organs 1999;22:81. 363. Levy JB, Turner AN, Rees AJ, Pusey CD. Long-term outcome of anti-glomerular basement membrane antibody disease treated with plasma exchange and immunosuppression. Ann Intern Med 2001;134: 1033–1042. 364. Jennette JC. Rapidly progressive glomerulonephritis. Kidney Int 2003;63:1164–1177. 365. Burran WP, Avasthi P, Smith KJ, et al. Efficacy of plasma exchange in severe idiopathic rapidly progressive glomerulonephritis. Transfusion 1986;26:382. 366. Sakellariou G. Plasmapheresis as a therapy in specific forms of acute renal failure. Nephrol Dial Transplant 1994;9:210. 367. Glöckner WM, Sieberth HG, Wichmann HE, et al. Plasma exchange and immunosuppression in rapidly progressive glomerulonephritis: a controlled, multi-center study. Clin Nephrol 1988;29:1. 368. Cole E, Cattran D, Magil A, et al. A prospective randomized trial of plasma exchange as additive therapy in idiopathic crescentic glomerulonephritis. Am J Kidney Dis 1992;20:261. 369. Guillevin L, Fain O, Lhote F, et al. Lack of superiority of steroids plus plasma exchange to steroids alone in the treatment of polyarteritis nodosa and Churg-Strauss syndrome: a prospective, randomized trial in 78 patients. Arthritis Rheum 1992;35:208. 370. Nakamura T, Matsuda T, Kawagoe Y, et al. Plasmapheresis with immunosuppressive therapy vs. immunosuppressive therapy alone for rapidly progressive anti-neutrophil cytoplasmic autoantibody-associated glomerulonephritis. Nephrol Dial Transplant 2004;19:1935–1937. 371. Levy JB, Pusey CD. Still a role for plasma exchange in rapidly progressive glomerulonephritis? J Nephrol 1997;10:7. 372. Michaels PJ, Fishbein MC, Colvin RB. Humoral rejection of human organ transplants. Springer Semin Immunopathol 2003;25:119–40. 373. Croker BP, Ramos EL. Pathology of the renal allograft. In Tishler CC, Brenner BM (eds). Renal Pathology with Clinical and Functional Correlations. Philadelphia, JB Lippincott, 1994, p 1591. 374. Halloran PF. Immunosuppressive drugs for kidney transplantation. N Engl Med 2004;351:2115–2129. 375. Lindenfeld J, Miller GG, Shakar SF, et al. Drug therapy in the heart transplant recipient. Part I: Cardiac rejection and immunosuppressive drugs. Circulation 2004;110:3734–3740.

376. Cardella CJ, Sutton D, Uldall PR, de Veber GA. Intensive plasma exchange and renal transplant rejection. Lancet 1977;1:264. 377. Rifle G, Chalopin JM, Ture JM, et al. Plasmapheresis in the treatment of renal allograft rejections. Transplant Proc 1979;11:20. 378. Adams MB, Kauffman HM, Hussey CV, et al. Plasmapheresis in the treatment of refractory renal allograft rejection. Transplant Proc 1981; 13:491. 379. Vangelista A, Frasca GM, Nanni Costa A, et al. Value of plasma exchange in renal transplant rejection induced by specific HLA antibodies. Trans Am Soc Artif Intern Organs 1982;28:599. 380. Cardella CJ, Sutton D, Falk JA, et al. Effect of intense plasma exchange on renal transplant rejection and serum cytotoxic antibody. Transplant Proc 1978;10:617. 381. Bankowski LHW, Corteste J, Lutton JL, Saunders PH. Plasmapheresis: adjunctive treatment for steroid-resistant rejection in renal transplantation. J Urol 1984;131:14. 382. Soulillou JP, Guyot C, Guimbretiere J, et al. Plasma exchange in early kidney graft rejection associated with anti-donor antibodies. Nephron 1983;35:158. 383. Allen NH, Ayer P, Geoghegan T, et al. Plasma exchange in acute renal allograft rejection: a controlled trial. Transplantation 1983;35:425. 384. Kirubakaran MG, Disney APS, Norman J, et al. A controlled trial of plasmapheresis in the treatment of renal allograft rejection. Transplantation 1981;32:164. 385. Blake P, Sutton D, Cardella C. Plasma exchange in acute renal transplant rejection. Prog Clin Biol Res 1990;337:249. 386. Bonomini V, Vangelista A, Frasca GM, et al. Effects of plasmapheresis in renal transplant rejection: a controlled study. Trans Am Soc Artif Intern Organs 1985;31:698. 387. Grandtnerova B, Javorsky P, Kolacny J, et al. Treatment of acute humoral rejection in kidney transplantation with plasmapheresis. Transplant Proc 1995;27:934. 388. Loss GE Jr, Grewal HP, Siegel CT, et al. Reversal of delayed hyperacute renal allograft rejection with a tacrolimus-based therapeutic regimen. Transplant Proc 1998;30:1249. 389. Aichberger C, Nussbaumer W, Rosmanith P, et al. Plasmapheresis for the treatment of acute vascular reaction in renal transplantation. Transplant Proc 1997;29:169. 390. Gannedahl G, Ohlman S, Persson U, et al. Rejection associated with early appearance of donor-reactive antibodies after kidney transplantation treated with plasmapheresis and administration of 15-deoxyspergualin: a report of two cases. Transplant Int 1992;5:189. 391. Abe M, Sannomiya A, Koike T, et al. Removal of anti-donor antibody by double-filtration plasmapheresis to prevent chronic rejection in kidney transplantation. Transplant Proc 1998;30:3108. 392. Montgomery RA, Zachary AA, Racusen LC, et al. Plasmapheresis and intravenous immune globulin provides effective rescue therapy for refractory humoral rejection and allows kidneys to be successfully transplanted into cross-match-positive recipients. Transplantation 2000;70:887–895. 393. Crespo M, Pascual M, Tolkff-Rubin M, et al. Acute humoral rejection in renal allograft recipients. I: Incidence, serology and clinical characteristics. Transplantation 2001;71:652–658. 394. Jordan SC, Vo AA, Tyan D, et al. Current approaches to treatment of antibody-mediated rejection. Pediatr Transplant 2005;9:408–415. 395. Zachary AA, Montgomery RA, Ratner LE, et al. Specific and durable elimination of antibody to donor HLA antigens in renal-transplant patients. Transplantation 2003;76:1519–1525. 396. Takemoto SK, Zeevi A, Feng S, et al. National conference to assess antibody-mediated rejection in solid organ transplantation. Am J Transplantation 2004;4:1033–1041. 397. Partanen J, Nieminen MS, Krogerus L, et al. Heart transplant rejection treated with plasmapheresis. J Heart Lung Transplant 1992;11:301. 398. Malafa M, Mancini MC, Myles JL, et al. Successful treatment of acute humoral rejection in a heart transplant patient. J Heart Lung Transplant 1992;11:486. 399. Fishbein MC, Kobashigawa J. Biopsy-negative cardiac transplant rejection: etiology, diagnosis, and therapy. Curr Opin Cardiol 2004;19:166–169. 400. Taube D, Palmer A, Welsh K, et al. Removal of anti-HLA antibodies prior to transplantation: an effective and successful strategy for highly sensitized renal allograft recipients. Transplant Proc 1989;21:694. 401. Backman U, Fellstrom B, Frodin L, et al. Successful transplantation in highly sensitized patients. Transplant Proc 1989;21:762. 402. Alarabi A, Backman U, Wikstrom B, et al. Plasmapheresis in HLAimmunosensitized patients prior to kidney transplantation. Int J Artif Organs 1997;20:51. 403. Hodge EE, Klingman LL, Koo AP, et al. Pretransplant removal of antiHLA antibodies by plasmapheresis and continued suppression on

405. 406. 407. 408. 409. 410. 411. 412. 413. 414.

415. 416. 417. 418. 419. 420.

421.

422. 423.

424. 425. 426.

427. 428.

429.

430. Thompson GR, Miller JP, Breslow JL. Improved survival of patients with homozygous familial hypercholesterolaemia treated with plasma exchange. Br Med J 1985;291:1671. 431. Kamanabroo D, Ulrich K, Grobe H, Assman G. Plasma exchange in type II hypercholesterolemia. Prog Clin Biol Res 1988;255:347. 432. Leren TP, Fagerhol MK, Leren P. Sixteen years of plasma exchange in a homozygote for familial hypercholesterolemia. J Intern Med 1993; 233:195. 433. Beigel Y, Bar J, Cohen M, Hod M. Pregnancy outcome in familial homozygous hypercholesterolemic females treated with long-term plasma exchange. Acta Obstet Gynecol Scand 1998;77:603. 434. Leitman SF, Smith JW, Gregg RE. Homozygous familial hypercholesterolemia: selective removal of low-density lipoproteins by secondary membrane filtration. Transfusion 1989;9:341. 435. Matsuda Y, Malchesky PS, Nose Y. Assessment of currently available low-density lipoprotein apheresis systems. Artif Organs 1994;18:93. 436. Armstrong VW, Eisenhauer T, Noll D, et al, Extracorporeal plasma therapy: the HELP system for the treatment of hyperlipoproteinemia. In Widhalm K, Maito HK (eds). Recent Aspects of Diagnosis and Treatment of Lipoprotein Disorders: Impact on Prevention of Atherosclerotic Diseases. New York, Alan R. Liss, 1988, p 327. 437. Stoffel W, Borberg H, Greve V. Application of specific extracorporeal removal of low density lipoprotein in familial hypercholesterolaemia. Lancet 1981;2:1005. 438. Saal SD, Parker TS, Gordon BR, et al. Removal of low-density lipoproteins in patients by extracorporeal immunoadsorption. Am J Med 1986;80:583. 439. Yamamoto A, Kojima S, Shiba-Harada M, et al. Assessment of the biocompatibility and long-term effect of LDL-apheresis by dextran sulfate-cellulose column. Artif Organs 1992;16:177. 440. Vella A, Pineda AA, O’Brien T. Low-density lipoprotein apheresis for the treatment of refractory hyperlipidemia. Mayo Clin Proc 2003;76:1039–1046. 441. Wills AJ, Manning NJ, Reilly MM. Refsum’s disease. Q J Med 2001; 94:403–406 442. Gibberd FB. Plasma exchange for Refsum’s disease. Transfus Sci 1993;14:23. 443. Gutsche H-U, Siegmund JB, Hoppmann I. Lipapheresis: an immunoglobulin-sparing treatment for Refsum’s disease. Acta Neurol Scand 1996;94:190. 444. Giorgi DF, Jagoda A. Poisoning and overdose. Mt Sinai J Med 1997; 64:283. 445. Trujillo MH, Guerrero J, Fragachan C, Fernandez MA. Pharmacologic antidotes in critical care medicine: a practical guide for drug administration. Crit Care Med 1998;26:377. 446. Jones JS, Dougherty J. Current status of plasmapheresis in toxicology. Ann Emerg Med 1986;15:474 447. Kale-Pradhan PB, Woo MH. A review of the effects of plasmapheresis on drug clearance. Pharmacology 1997;17:684–695. 448. Luzhnikov EA, Yaroslavsky AA, Molodenkov MN, et al. Plasma perfusion through charcoal in methylparathion poisoning. Lancet 1977;1:38. 449. Pierga JY, Beuzeboc P, Dorval T, et al. Favorable outcome after plasmapheresis for vincristine overdose. Lancet 1992;1:185. 450. Spiller M, Marson P, Perilongo G, et al. A case of vinblastine overdosage managed with plasma exchange. Pediatr Blood Cancer 2005;45:344–346. 451. Chu G, Mantin R, Shen YM, et al. Massive cisplatin overdose by accidental substitution for carboplatin: toxicity and management. Cancer 1993;72:3707. 452. Duzova A, Baskin E, Usta Y, Ozen S. Carbamazepine poisoning: treatment with plasma exchange. Hum Exp Toxicol 2001;20:175–177. 453. Chillet P, Korach JM, Petitpas JM, et al. Digoxin poisoning and acute anuric renal failure: efficiency of treatment associating digoxin-specific antibodies (Fab) and plasma exchanges. Int J Artif Organs 2002;25:538–541. 454. Ligtenberg J, Tulleken J, Zijlstra J. Plasmapheresis in thyrotoxicosis. Ann Intern Med 1999;131:71. 455. Mercuriali F, Sichia G. Plasma exchange for mushroom poisoning. Transfusion 1977;17:644. 456. Ponikvar R, Drinovec J, Kandus A, et al. Plasma exchange in management of severe acute poisoning with Amanita phalloides. In Rock G (ed). Apheresis. New York, Wiley-Liss, 1990, p 327. 457. Piqueras J, Duran-Suarez JR, Massuet L, Hernandez-Sanchez JM. Mushroom poisoning: therapeutic apheresis or forced diuresis. Transfusion 1987;27:116. 458. Seyffart G. Plasmapheresis in treatment of acute intoxications. Trans Am Soc Artif Organs 1982;28:673. 459. Guzelian PS. New approaches for treatment of humans exposed to a slowly excreted environmental chemical (chlordecone). Z Gastroenterol 1984;22:16.

THERAPEUTIC PLASMA EXCHANGE

404.

cyclosporine-based therapy after heart-kidney transplant. Transplant Proc 1994;26:2750. Hakim R, Milford E, Himmelfarb J, et al. Extracorporeal removal of antiHLA antibodies in transplant candidates. Am J Kidney Dis 1990;16:423. Kupin WL, Venkat KK, Hayashi H, et al. Removal of lymphocytotoxic antibodies by pretransplant immunoadsorption therapy in highly sensitized renal transplant recipients. Transplantation 1991;51:324. Miura S, Okazaki H, Sato T, et al. Beneficial effects of double-filtration plasmapheresis on living related donor renal transplantation in presensitized recipients. Transplant Proc 1995;27:1040. Miura S, Okazaki H, Sato T, et al. Successful renal transplantation in presensitized recipients with double-filtration plasmapheresis and 15-deoxyspergualin. Transplant Proc 1997;29:350. Ishikawa A, Itoh M, Ushlyama T, et al. Experience of ABO-incompatible living kidney transplantation after double filtration plasmapheresis. Clin Transplant 1998;12:80. Renard TH, Andrews WS. An approach to ABO-incompatible liver transplantation in children. Transplantation 1992;53:116. Takahashi K, Yogisawa T, Sonda K, et al. ABO-incompatible kidney transplantation in a single center trial. Transplant Proc 1993;25:271. Boudreaux JP, Hayes DH, Mizrahi S, et al. Successful liver/kidney transplantation across ABO incompatibility. Transplant Proc 1993;25:1874. Aswad S, Mendez R, Mendez RG, et al. Crossing the ABO blood barrier in renal transplantation. Transplant Proc 1993;25:267. Montgomery RA, Zachary AA. Transplanting patients with a positive donor-specific crossmatch: a single center’s perspective. Pediatr Transplant 2004;8:535–542. Sonnenday CJ, Ratner LE, Zachary AA, et al. Preemptive therapy with plasmapheresis/intravenous immunoglobulin allows successful live donor renal transplantation in patients with a positive cross-match. Transplant Proc 2002;34:1614–1616. Vincenti F, Ghigger GM. New insights into the pathogenesis and the therapy of recurrent focal glomerulosclerosis. Transplantation 2005;5:1179–1185. Artero ML, Sharma R, Savin VJ, Vincenti F. Plasmapheresis reduces proteinuria and serum capacity to injure glomeruli in patients with recurrent focal glomerulosclerosis. Am J Kidney Dis 1994;23:574. Savin VJ, McCarthy ET, Sharma M. Permeability factors in focal segmental glomerulosclerosis. Semin Nephrol 2003;23:147–160. Benchimal C. Focal segmental glomerulosclerosis: pathogenesis and treatment. Curr Opin Pediatr 2003;15:171–180. Cochat P, Kassir A, Colon S, et al. Recurrent nephrotic syndrome after transplantation: early treatment with plasmapheresis and cyclophosphamide. Pediatr Nephrol 1993;7:50. Oetliker OH, Zimmerman A, Bianchetti MG. Treatment of recurrent idiopathic nephrotic syndrome after transplantation using plasmapheresis and intensified immunosuppression over 2 months [Letter]. Pediatr Nephrol 1993;7:508. Delucchi A, Cano F, Rodriguez E, Wolff E. Focal segmental glomerulosclerosis relapse after transplantation: treatment with high cyclosporine doses and a short plasmapheresis course [Letter]. Pediatr Nephrol 1994;8:786. Davenport RD. Apheresis treatment of recurrent focal segmental glomerulosclerosis after kidney transplantation: reanalysis of published case-reports and case-series. J Clin Apheresis 2001;16:175–178. Deegens JKJ, Andresdottir MB, Crookewit S, Wetzels JFM. Plasma exchange improves graft survival in patients with recurrent focal glomerulosclerosis after renal transplantation. Transplant Int 2004; 17:1511–1517. Dixon FJ, McPhaul JJ, Lemer RA. The contribution of transplantation to the study of glomerulonephritis: the recurrence of glomerulonephritis in renal transplants. Transplant Proc 1969;1:194. Ramzy D, Rao V, Brahm J, et al. Cardiac allograft vasculopathy. Can J Surg 2005;48:319–327. Thiery J, Meiser B, Wenke K, et al. Heparin-induced extracorporeal low-density-lipoprotein plasmapheresis (HELP) and its use in heart transplant patients with severe hypercholesterolemia. Transplant Proc 1995;27:1950. Barr ML, McLaughlin SN, Murphy MP, et al. Prophylactic photopheresis and effect on graft atherosclerosis in cardiac transplantation. Transplant Proc 1995;27:1993. Goldstein JL, Hobbs HH, Brown MS. Familial hypercholesterolemia. In Scriver CR, Beuadet AL, Sly WS, Valle D (eds). The Metabolic and Molecular Basis of Inherited Disease. New York, McGraw-Hill, 2001, pp 1863–1913. Rader DJ, Cohen J, Hobbs HH. Monogenic hypercholesterolemia: new insights in pathogenesis and treatment. J Clin Invest 2003;11:1795–1803.

55 763

TRANSFUSION MEDICINE

III 764

460. Elliott HL, MacDougall AI, Haase G, et al. Plasmapheresis in the treatment of dialysis encephalopathy. Lancet 1978;2:940. 461. Tilz GP, Teubl I, Kopplhuber C, et al. Therapeutic plasmapheresis: A new form of adjuvant treatment. Med Klin 1976;71:1952. 462. Sabato JK, Pierce RM, West RH, Gurr FW. Hemodialysis, peritoneal dialysis, plasmapheresis and forced diuresis for the treatment of quinine overdose. Clin Nephrol 1981;16:264. 463. Larsen LS, Sterrett JR, Whitehead B, Marcus SM. Adjunctive therapy of phenytoin overdose: a case report using plasmapheresis. J Toxicol Clin Toxicol 1986;24:37. 464. Keller F, Hauff A, Schultze G, et al. Effect of repeated plasma exchange on steady state kinetics of digoxin and digitoxin. Arzneimittelforschung 1984;34:83. 465. Stigelman WH, Henry DH, Talbert RL, Townsend RJ. Removal of prednisone and prednisolone by plasma exchange. Clin Pharm 1984;3:402. 466. Appelgate R, Schwartz D, Bennett WM. Removal of tobramycin during plasma exchange therapy. Ann Intern Med 1981;94:820. 467. Talbert RL, Wong YY, Duncan DB. Propranolol plasma concentrations and plasmapheresis. Drug Intell Clin Pharm 1981;15:993.

468. 469. 470. 471. 472.

473. 474. 475.

Lee WM. Acute liver failure. Am J Med 1994;96(Suppl 1A):3S. Lee WM. Acute liver failure. N Engl J Med 1993;329:1862. Caraceni P, van Thiel DH. Acute liver failure. Lancet 1995;345:163. Kondrup J, Almdal T, Vilstrup H, et al. High volume plasma exchange in fulminant hepatic failure. Int J Artif Organs 1992;15:669. Larsen FS, Hansen BA, Ejlersen L, et al. Cerebral blood flow, oxygen metabolism and transcranial Doppler sonography during high-volume plasmapheresis in fulminant hepatic failure. Eur J Gastroenterol Hepatol 1995;8:261. Saibara T, Maeda T, Onishi S, Yamamoto Y. Plasma exchange and the arterial blood ketone body ratio in patients with acute hepatic failure. J Hepatol 1994;20:617. Barshes NR, Gay AN, Williams B, et al. Support for the acutely failing liver: a comprehensive review of historic and contemporary strategies. J Am Col Surgeons 2005;201:458–476. Jalan R, Sen S, Williams R. Prospects for extracorporeal liver support. Gut 2004;53:890–898.

Chapter 56

Therapeutic Cytapheresis Bruce C. McLeod

INTRODUCTION This chapter will cover apheresis procedures performed on patients to modify or remove excess or abnormal formed elements. Because leukocytes and platelets occupy only a small fraction of total blood volume, they can generally be collected or depleted without a need for volume replacement. Red cells, by contrast, occupy a major fraction of total blood volume. In addition, their presence in the blood in adequate quantities is essential for respiratory gas transport. For these reasons, therapeutic manipulation of the red cell content of blood most often takes the form of an exchange procedure in which patient red cells are removed and simultaneously replaced with normal donor cells.

The extent of reduction in the concentration of a targeted cell in blood by therapeutic cytapheresis is not easily predicted. Formulas based on blood volume, initial cell concentration, and volume of blood processed may give unreliable estimates for several reasons, including (1) the blood volume of patients with elevated cell counts is sometimes underestimated by standard nomograms, (2) more cells can be released into the circulation during a procedure from bone marrow or an enlarged spleen, and (3) the behavior of abnormal cells in a centrifugal field may differ from expectations2 (Table 56–2) Therefore, rather than prescribe in advance a specific volume to be processed for a therapeutic cell-depletion procedure, it is more reasonable to monitor the cell count of interest during a procedure and, if possible, continue until a meaningful reduction (e.g., 30% to 50%) has been achieved.

THERAPEUTIC CYTAPHERESIS Cytapheresis procedures deplete or collect a component of the buffy coat having a density intermediate between red cells and plasma (Table 56–1). These procedures were the original goal for which centrifugal apheresis instruments were designed, which is why these instruments are sometimes called “blood cell separators.” It is possible for procedures to emphasize collection of either platelets or mononuclear white blood cells (MNCs) depending on centrifugal force, separation-chamber geometry, and, for intermittent collection techniques, the timing of collection events. Addition of an agent such as hydroxyethyl starch to blood entering the centrifuge will promote rouleaux formation, which will accelerate red cell sedimentation in the centrifuge. This enhances the separation between red cells and granulocytes (polymorphonuclear leukocytes; PMNs) (whose density is higher than that of MNCs and overlaps that of the youngest red cells) and thereby enables efficient collection of PMN by an apheresis instrument.1,2 Strictly speaking, the term therapeutic cytapheresis might be reserved for procedures designed to deplete an abnormal and/ or overabundant buffy-coat component. This chapter will have a somewhat larger scope and will include other cytapheresis procedures performed on patients, even though some of them could equally well be considered autologous donations because the expected benefit is not realized until the collected cells are reinfused at a later time. Examples of the latter type of procedure would include collection and infusion of peripheral blood progenitor cells to restore hematopoiesis and collection and subsequent reinfusion of cells that have undergone ex vivo immunization or genetic alteration intended to enhance host immunity or correct a genetic defect.

Therapeutic Plateletpheresis An elevated platelet count may be found in three settings: (1) rare familial cases may be due to excessive production of thrombopoietin; (2) a reactive thrombocytosis can occur after several types of inciting events including splenectomy, iron deficiency, malignancy, and chronic inflammation; and (3) finally, a supernormal platelet count may be seen in clonal myeloproliferative disorders such as chronic myelogenous leukemia. When a high platelet count is the only sign of an abnormal clone, the disorder is called essential thrombocythemia. Symptoms attributable to thrombocytosis are seen only in patients who have a myeloproliferative disorder.3,4 Rationale Many patients with thrombocytosis remain asymptomatic indefinitely; however, both thrombosis and hemorrhage are noted with increased frequency in the context of a myeloproliferative disorder. Clotting may be venous (e.g., deep vein thrombosis, Budd-Chiari syndrome) or arterial (e.g., erythromelalgia, stroke). Hemorrhage may be caused by a platelet-function defect or by aspirin given to protect against thrombosis. An age exceeding 60 years, a preexisting cardiovascular disease, or a prior history of thrombosis (or a combination of these) all confer an increased risk for clinical events, the magnitude of which correlates roughly with platelet count (Table 56–3). In younger patients without cardiovascular disease, the risk of clinical events seems unrelated to platelet count (i.e., symptoms may develop with a platelet count of 500,000/μL in some patients, whereas others may remain problem free for years with platelet counts exceeding 1 million/μL).3,4

52 56 765

TRANSFUSION MEDICINE

Table 56–1 Elements

Density of Plasma and Formed

Component

Specific Gravity

Plasma Platelets Lymphocytes Monocytes Granulocytes Erythrocytes

1.025–1.029 1.040 1.050–1.061 1.065–1.070 1.087–1.092 1.093–1.096

Clinical Studies Little definitive evidence guides therapy for thrombocytosis. A single trial in high-risk patients revealed a reduced incidence of clinical events when the platelet count was kept below 600,000/μL,5 but no comparable data exist for lowerrisk patients. Drug therapy with hydroxyurea or anagrelide will reduce the platelet count of most patients. Therapeutic plateletpheresis is usually reserved for patients with acute, serious thrombotic or hemorrhagic events, or for higher-risk patients with very high platelet counts. When it occurs in such circumstances, thrombocytosis has been rated a category I (standard practice) indication for therapeutic plateletpheresis by the American Society for Apheresis (ASFA)6 and AABB7 (Table 56–4). An urgent plateletpheresis will reduce the platelet count promptly, and additional procedures can maintain a reduced level until drug therapy takes effect.2,3 Technique

III 766

Therapeutic plateletpheresis can be performed with any centrifugal apheresis instrument. Procedural variables such as anticoagulation, centrifuge speed, and separatory chamber configuration are usually the same as those used for a platelet donation; however, it may be desirable to maintain a higher flow rate in the component-removal line in an instrument so equipped.2 As mentioned, the platelet count decrement for a single procedure cannot be reliably predicted on the basis of the volume of blood processed. Monitoring intraprocedural platelet counts is the best way to ensure that the desired decrement (e.g., a 50% decline) is obtained before a procedure is discontinued.

Therapeutic Leukapheresis The great majority of therapeutic leukapheresis procedures are carried out to treat hyperleukocytosis occurring in association with leukemia, but a few other indications should be briefly mentioned. Depletion of nonmalignant lymphocytes by apheresis, with or without simultaneous therapeutic plasma exchange (TPE), has been reported in several autoimmune disorders. Rheumatoid arthritis,8 progressive multiple sclerosis,9 inclusion body myosi-

Table 56–2 Sources of Error in Predicting the Extent of Cell Depletion by Leukapheresis 1. Blood volume exceeds predictions 2. Extravascular cells mobilized during procedure 3. Collection efficiency differs from expectations

Table 56–3 Risk Factors for Clinical Events in Myeloproliferative Disease with Thrombocytosis 1. Age older than 60 years 2. Cardiovascular disease 3. History of thrombosis

tis,10 polymyositis, and dermatomyositis11 have all been assigned indication categories by ASFA6 and AABB7 (see Table 56–4); however, leukapheresis to deplete nonmalignant lymphocytes has not become an accepted treatment for any autoimmune disorder. Leukapheresis has been used to remove malignant lymphocytes from patients who have Sézary syndrome in the context of cutaneous T-cell lymphoma (CTCL)12,13; this is also a category III indication (see Table 56–4).6,7 However, it has been largely replaced by photopheresis, as discussed later. Rationale The clinical manifestations associated with a highly elevated white blood cell (WBC) count (hyperleukocytosis) in leukemia patients include leukostasis, tumor lysis syndrome, and early mortality. These are not mutually exclusive, and it is therefore useful to keep all of them in mind when considering the potential for benefit from therapeutic leukapheresis. Leukostasis is diagnosed when evidence is present of organ dysfunction due to microvascular obstruction and consequent patchy ischemia. Neurologic and pulmonary dysfunction are the most common examples in acute leukemia; priapism is another possible complication in patients with chronic myelogenous leukemia (CML).14,15 The pathologic basis for these symptoms is believed to be small-vessel occlusion by masses of leukemic cells, possibly with thrombosis, and sometimes with hemorrhage distal to the occlusion.16,17 Such obstructions have been attributed to rheologic consequences of an increased blood concentration of WBCs,

Table 56–4 ASFA/AABB Indication Categories for Therapeutic Cell Depletion Disease

Procedure

Category

Thrombocytosis Hyperleukocytosis Rheumatoid arthritis Progressive multiple sclerosis Inclusion body myositis Polymyositis Dermatomyositis Heart transplant rejection Cutaneous T-cell lymphoma Sickle cell disease Severe malaria/ babesiosis

Plateletpheresis Leukapheresis Lymphoplasmapheresis Leukapheresis

I I II III

Leukapheresis Leukapheresis Leukapheresis Photopheresis

IV IV IV III

Photopheresis Leukapheresis Red cell exchange Red cell exchange

I III I III

I, standard therapy; II, evidence suggests efficacy; III, inadequately tested; IV, ineffective in controlled trials.

Clinical Studies It is widely accepted, based on accumulated clinical experience rather than controlled studies, that reducing the WBC count by leukapheresis can reverse symptoms of leukostasis, although it may not do so if tissue infarction as well as ischemia has occurred. This is the basis for the ASFA/AABB ranking of hyperleukocytosis as a category I indication for leukapheresis (see Table 56–4).6,7 Leukapheresis should be carried out as an emergency in leukemia patients with symptomatic leukostasis, especially acute leukemia patients with blast counts exceeding 100,000/μL. Wide acceptance of this principle led to the question of whether urgent prophylactic leukapheresis might be prudent in acute leukemia patients with WBC or blast counts exceeding 100,000/μL, even if they do not have signs of leukostasis. Evidence for benefit from such prophylactic leukapheresis is limited, and this practice remains controversial; nevertheless, urgent leukapheresis may be requested routinely on this basis by individual physicians or as institutional policy, and requests for repeated leukapheresis may be made when acute leukemia patients’ WBC counts remain high for several days before chemotherapy is instituted.24,25 An extreme example of the latter was a pregnant patient who received only leukapheresis therapy for several weeks after a diagnosis of acute leukemia was made because she wished to delay chemotherapy until after delivery.29 It is still not clear that reducing the WBC count by leukapheresis before initiation of chemotherapy prevents or

meaningfully reduces the severity of tumor lysis syndrome. A counter-argument to the prediction of such benefit is that the leukemic cells accessible for removal from the circulation represent only a small fraction of the total tumor burden and that a meaningful reduction in tumor load is not likely to be achieved in this way. Controlled studies that address this point have not been reported. It is also unclear whether prophylactic leukapheresis improves survival in patients with hyperleukocytosis, either by attenuating tumor lysis syndrome or in some other way. Three observational studies of prophylactic leukapheresis have been published in the past decade. One group reported 48 consecutive AML patients who received leukapheresis if the presenting WBC count exceeded 100,000/μL. The extent of WBC count reduction was not significantly different between the 14 patients who died within 2 weeks and the remaining patients, suggesting that WBC count reduction was not an important predictor of mortality.30 Another nonrandomized, retrospective study of 146 patients with AML and a WBC count exceeding 50,000/μL showed a higher 2-week survival (87% vs 77%) in patients who had leukapheresis; this was statistically significant only in a logistic regression analysis. Follow-up observation, however, showed no advantage in 4-week, 6-week, or overall survival. Paradoxically, long-term survival was significantly lower in the group that had leukapheresis.31 A third study looked at outcomes in 53 patients with AML and WBC counts exceeding 100,000/μL, all of whom underwent daily leukapheresis (76 procedures total) until the WBC count dropped below 100,000/μL or the performance status improved. Only two patients died in the first week; however, despite the prophylactic WBC removal, 47% developed coagulopathy, 85% developed tumor lysis syndrome that was severe in 53%, and only 55% achieved a complete remission. Median survival among the responders was only 8 months.32 Taken as a whole, these reports provide little indication of benefit from routine prophylactic leukapheresis in hyperleukocytic AML patients. Indeed, the results might equally well imply that hyperleukocytosis is merely a marker for other risk factors, such as morphologic category, specific chromosomal abnormalities, and/or total tumor burden, that cannot be altered simply by leukapheresis. Prospective randomized controlled studies are needed before routine prophylactic leukapheresis can be recommended. Technique Therapeutic leukapheresis can be accomplished with most centrifugal apheresis instruments. Urgent treatment of patients with inadequate peripheral venous access may require equally urgent placement of a dual-lumen central venous catheter, which can be challenging in acutely ill patients with the low platelet counts that often occur in leukemia. When contemplating leukapheresis in a patient who does not have leukostasis, the unproven benefits of treatment should be balanced against the potential hazards of emergency central-line placement and delay in starting pharmacologic cytoreduction with hydroxyurea and definitive chemotherapy.24,25 Cell removal will be more efficient in most instances if hydroxyethyl starch or another sedimenting agent is added to patient blood before it enters the centrifuge. Also, removal of leukemic cells, even in a concentrate containing 500,000 to 750,000 leukocytes per microliter, may entail large plasma losses. The volume-expanding properties of a sedimenting

THERAPEUTIC CYTAPHERESIS

especially increased viscosity.18 However, concomitant anemia keeps the whole blood viscosity within the normal range in most patients with leukemia,19,20 unless ill-advised red cell transfusions are given before the WBC count has been lowered.21 More recent studies have identified reduced deformability, cytokine secretion, and altered adherence properties of blasts or other primitive cells as factors contributing to leukostasis.2,22,23 These factors could explain not only why leukostasis correlates more closely with the circulating blast count than with total WBC count, but also why the absolute blast count at which symptoms occur can vary widely from patient to patient.24,25 Leukostasis occurs in about 5% of patients with acute myelogenous leukemia (AML).14 Autopsy studies have shown microscopic evidence of leukostasis in most AML nonsurvivors whose WBC counts exceeded 200,000/μL.17 The blast count exceeds 100,000/μL in most clinically evident instances of leukostasis; however, the syndrome is occasionally suspected in patients when the blast count is only in the 50,000 to 100,000/μL range.26 Leukostasis is seldom seen in acute lymphocytic leukemia, and then only when the WBC count reaches the 250,000 to 300,000/μL range.24 Higher cell counts, usually in the 300,000 to 500,000/μL range, are required before the more mature cells that circulate in CML can cause signs of leukostasis,15 whereas counts above even this level may be tolerated without symptoms in chronic lymphocytic leukemia.2 Even in patients who do not have leukostasis, hyperleukocytosis is a marker for a poorer prognosis in both acute and chronic leukemias. Both short-term and long-term survival are inversely correlated with presenting WBC count, as are the complete remission rate and the mean duration of response to chemotherapy.27 Finally, hyperleukocytosis is associated with a higher incidence of hyperuricemia and other manifestations of tumor lysis syndrome, all of which may be exacerbated by cytotoxic chemotherapy.28

56 767

TRANSFUSION MEDICINE

agent can offset this to some extent, but additional colloid replacement is wise if the volume of leukocyte concentrate removed exceeds 1 L.2 The extent to which the WBC count must be reduced to prevent or reverse leukostasis cannot be known in advance for an individual patient with any degree of certainty. Nevertheless, it is reasonable to expect clinical improvement in a patient with leukostasis if the WBC count is promptly reduced by 30% to 50%, ideally to less than 100,000/μL in AML or less than 300,000/μL in CML. The extent of prophylactic leukocyte removal that might be needed to prevent tumor lysis syndrome or improve survival is even more uncertain. As mentioned previously, it is difficult to predict WBC count reduction merely on the basis of the volume of blood processed.2 It is therefore worthwhile to monitor intraprocedural WBC counts to assure that a targeted reduction is reached.

Photopheresis

III 768

In an extracorporeal photochemotherapy (ECP or photopheresis) treatment, patient MNCs separated by leukapheresis are exposed to a standardized dose of ultraviolet A radiation (UVA) in the presence of 8-methoxypsoralen (8-MOP) at a concentration of 60 to 200 ng/μL. 8-MOP is a photoactive compound that will crosslink nuclear DNA in the treated cells by means of diadducts between proximate thymidine residues in complementary strands. Protein changes may occur as well. After a short incubation, the irradiated cells are returned to the patient. Although conceptually independent, the modules performing the apheresis and irradiation steps can be housed in a single instrument such as the UVAR XTS system (Therakos, Exton, Pa.). 8-MOP was originally given orally shortly before apheresis but is now available in a preparation suitable for intravenous administration that can be added to the MNC concentrate after collection but before UVA irradiation. This provides much better control of 8-MOP concentration during irradiation; in addition, it reduces the dose of 8-MOP given to the patient by a hundredfold, thereby reducing side effects such as photosensitivity. In most of the applications to be discussed, ECP has usually been performed on one or two consecutive days at 2- to 4-week intervals.33 Cutaneous T-Cell Lymphoma ECP was devised in the 1980s as a therapy for CTCL, a disease in which malignant lymphocytes are typically abundant in the bloodstream as well as in skin infiltrates.34 Certain cutaneous manifestations of CTCL were known to improve after depletion of malignant cells by leukapheresis12,13 and also after UVA irradiation of skin after ingestion of 8-MOP (psoralen-UVA treatment; PUVA).35 Collection, UVA irradiation and reinfusion of 5% to 10% of circulating MNCs on 2 consecutive days monthly led, after a delay of some months, to dramatic improvement in some patients. Responses of the magnitude observed cannot be accounted for simply by destruction of UVA-irradiated tumor cells, and this quantitative discrepancy suggested the possibility that ECP may elicit, enhance, or otherwise modulate a host immune response to the tumor.23,24 This hypothesis provides a theoretical framework for trials of ECP in other diseases in which immunomodulation seems a reasonable approach to therapy. Nine North American studies have reported 282 CTCL patients who received ECP.34,36–43 Of these, 18% had a

complete response, and another 38% had a partial response.44 Long-term follow-up suggested better median survival than had been observed in historic controls. ECP seems most active in patients who have had relatively recent onset of diffuse skin disease (erythroderma) but still have relatively little of the immunosuppression that often accompanies advanced disease. Results are not as good in patients with localized skin plaques, long-standing disease, and/or involvement of lymph nodes and viscera.44,45 The FDA approved Therakos’s original UVAR system for treatment of CTCL in 1987, and this use has been rated a category I indication for ECP by ASFA6 and AABB7 (see Table 56-4). Unfortunately, although its ability to produce regression of erythroderma is well established, the optimal role of ECP in CTCL remains uncertain because, unlike almost all drug therapies for malignant disease, it has never been subjected to controlled trials with uniform enrollment and response criteria comparing it with no treatment, sham ECP, or alternative therapies.46 Graft versus Host Disease (GVHD) GVHD is most often encountered as a complication of allogeneic stem cell transplantation. It affects skin, liver, and gastrointestinal (GI) tract. Acute GVHD begins before day 100 after transplant and is mostly inflammatory, whereas chronic GVHD begins after day 100 and may include sclerodermalike skin thickening and other fibrotic changes. Both forms occur frequently in allograft recipients despite prophylaxis with corticosteroids and other immunosuppressive drugs. Each form has about a 30% incidence after HLA-matched sibling transplants; after partially mismatched related or matched unrelated transplants, the respective incidences are in the 50% to 80% range.33,47 Skin GVHD sometimes improves after PUVA therapy.48 ECP was tried in hopes of addressing visceral involvement via a similar mechanism and has been reported to be beneficial in both acute and chronic GVHD. In both settings, the best responses are noted in skin manifestations.33,49 Eleven reports on acute GVHD,47,50–59 including observations of 76 patients, have been analyzed.47 Of the patients with skin disease, 83% showed improvement after ECP, with a complete response recorded in 67%. Complete regression was also observed in 54% of cases with GI involvement and 38% of cases with liver disease. Twenty reports on chronic GVHD47,52–54,56,59,60–72 also have been analyzed.47 Of 160 patients with skin disease, 76% improved after ECP, with a complete response being noted in 35%. Improvement was also reported in 48% of 84 patients with liver involvement, 39% of 31 patients with lung disease, and 63% of 59 patients with oral manifestations. Responses in this context are said to be quicker than those seen in CTCL, with improvement evident in a matter of weeks. Survival is also thought to be improved.47 Again, however, the precise contribution of ECP to treatment of GVHD must be regarded as uncertain because ECP-treated patients generally receive other therapies concomitantly, and no prospective controlled trials have been reported. GVHD has not been rated as an indication for ECP by either ASFA or AABB. Solid Organ Transplant Rejection Immunomodulation by means of ECP has also been undertaken in patients who are rejecting a solid organ transplant. A number of case studies and small series have described recipients of kidney73–75 and lung76–80 allografts who improved

Autoimmune Diseases Immunomodulation with ECP has been tried in a number of autoimmune diseases, including scleroderma,85 systemic lupus erythematosus,86 pemphigus vulgaris,87 rheumatoid arthritis,88 and psoriatic arthritis.89 Favorable results have been reported in case studies and small series. More extensive experience in scleroderma has indicated limited benefit at most,90 whereas data from controlled trials of ECP in other autoimmune disorders are still lacking. Mechanism of Action The mechanism of action of ECP is not known with certainty for any of the applications mentioned earlier. No doubt exists, however, that it leads to apoptosis in a majority of treated lymphocytes while leaving treated monocytes relatively intact and viable. UVA-induced DNA crosslinking by 8-MOP is probably an important contributor to this effect, although other mechanisms, such as protein damage, may operate as well. In CTCL, it is supposed that apoptosis of malignant lymphocytes exposes or releases tumor-specific antigens in a more immunogenic fashion, thereby eliciting or enhancing host immune responsiveness to the tumor. Recent studies suggest that contact with plastic surfaces of the disposables in the apheresis instrument and the UVA irradiation apparatus induces collected monocytes to differentiate into dendritic

cells. It is theorized that activated dendritic cells then ingest tumor antigens being released from apoptotic cells and facilitate antigen presentation, thus leading to enhanced immunity.91,92 A similar mechanism could be envisioned for ECP in other applications; that is, a downregulating response to a pathogenic though nonmalignant lymphocyte clone is enhanced through release of clone-specific antigens by apoptosis, followed by vigorous presentation of these antigens by abundant activated dendritic cells.49,93 At present, however, this mechanism should be regarded as hypothetical. Circulating clonal lymphocytes have been found in scleroderma94 and GVHD,49 but thus far, neither pathogenicity nor downregulation by ECP has been shown for such clones. The putative effects of ECP in GVHD have also been attributed to more general immunomodulatory effects. One hypothesis involves a shift from an inflammatory Th1 cytokine expression profile to a more inhibitory Th2 profile.95 A second envisions a paradoxical decrease in dendritic cell function in this context.96

THERAPEUTIC CYTAPHERESIS

when ECP was added to a conventional antirejection regimen. The largest experience, however, has come from heart transplantation. In considering the use of ECP in heart transplantation, it is important to point out that the term rejection has a subtle additional connotation in that context that does not apply to other organ allografts. Transplanted kidneys and lungs are seldom biopsied unless signs of organ dysfunction are present; however, the low risk and relative simplicity of endomyocardial biopsy (EMB) permits a cardiac allograft to be sampled routinely, often in the absence of cardiac dysfunction. It is also customary for patients who have mild histologic changes in such surveillance biopsies to be treated preemptively and aggressively for rejection in the absence of any signs of cardiac dysfunction. The first reports of ECP in cardiac allograft recipients described cases with severe, hemodynamically significant rejection who improved after ECP was added to other antirejection therapies.33,81 Subsequent studies have tended to focus on the more common patients with normal hemodynamics and mild histologic changes on EMB. A controlled trial with eight patients per arm found ECP to be equivalent to high-dose corticosteroids in reversing mild histologic changes on EMB.82 The significance of this finding is open to question, however, because other studies have suggested that such changes usually resolve without adjustments to antirejection therapy.83 A controlled trial of prophylactic ECP in 60 patients with cardiac transplants showed a lower incidence of mild histologic changes in patients who received ECP prophylaxis, but no difference in the rate of severe rejection with hemodynamic compromise.84 Absent controlled trials in patients with severe rejection, the uncertainty about the prognostic significance of mild histologic changes contributes to uncertainty about the proper role for ECP in treatment of rejection of heart and other organ transplants. Heart transplant rejection has been rated as a category III indication for ECP by ASFA6 and AABB7 (see Table 56–4).

Peripheral Blood Progenitor Cell Collection Autologous MNCs, collected by leukapheresis and cryopreserved, have all but replaced bone marrow as the preferred source of stem and progenitor cells for autologous hematopoietic rescue after myeloablative antitumor therapy for lymphomas, leukemias, myeloma, and other malignancies.97 A similar approach is being tried in some autoimmune diseases.98 Stem and progenitor cells do not circulate in the basal state in quantities sufficient for this purpose; however, they can be stimulated to enter the bloodstream in any of several ways. One is to give a “conventional” dose of an appropriate chemotherapeutic agent as “mobilizing chemotherapy” roughly 10 to 14 days before the planned date of collection. When the WBC count begins to increase after the expected interval of leukopenia, the blood concentration of CD34-positive cells, which is a marker for peripheral blood progenitor cell (PBPC) concentration, increases by as much as 25-fold and remains elevated for up to a week.99,100 A second method is to give daily injections of a hematopoietic growth factor. Both granulocyte colony–stimulating factor (G-CSF; filgrastim; Neupogen)101,102 and granulocytemacrophage colony–stimulating factor (GM-CSF; sargramostim; Leukine)103 injections will induce worthwhile levels of circulating CD34-positive cells in most individuals in 4 to 5 days; these levels will persist for up to a week with continued daily injections. In the autologous setting, G-CSF or GM-CSF or both can be given during recovery from mobilizing chemotherapy for maximal mobilization.104–106 A number of other cytokines will also mobilize PBPCs, but none is commercially available at this time.107 Goals for total CD34-positive cell yield from autologous PBPC collections range from 2 × 106 to 5 × 106 CD34positive cells per kilogram per transplant at different centers. Most patients will respond to mobilizing stimuli well enough to reach their goal after one to three daily leukapheresis procedures. A smaller proportion with suboptimal mobilization may require four to five daily collections. Perhaps 10% of patients, most of whom are older or have had extensive prior chemotherapy or both, will fail to have a useful response to conventional mobilization strategies. Their blood CD34positive cell levels (expressed in cells per microliter) will not increase above the low single-digit range, levels at which

56 769

TRANSFUSION MEDICINE

III 770

daily leukapheresis procedures will seldom produce the target yield within the limited period of mobilization.107 PBPCs are preferred over cells from bone marrow harvests for two reasons. The first is that the collection procedure does not require hospitalization or general anesthesia. The second is that WBC and platelet counts recover about a week sooner after PBPC transplantation than after bone marrow transplantation. The resultant reduction in early morbidity and mortality from infection and bleeding more than offsets a somewhat higher incidence of GVHD with PBPC.97 Candidates for PBPC collection have unique vascularaccess needs. They require a multilumen, tunneled central vein catheter for diagnostic blood draws, transfusion support, fluid maintenance, and chemotherapy infusions during both the pre- and post-transplant periods; however, catheters designed solely for these purposes are too flexible and have a diameter too small to support the flow rates required for efficient leukapheresis. Fortunately, tunneled dual- or triplelumen apheresis catheters (e.g., Raaf catheter) that accommodate both PBPC collection and the full range of other intravenous-access requirements are now commercially available. Potential autologous PBPC transplant recipients will usually benefit from having such a catheter placed. A notable exception would be a patient having a precautionary collection that would only be used in the event of an unexpected or late relapse. Such “harvest and hold” patients should have leukapheresis via peripheral veins if possible. PBPC collections are quite prolonged apheresis procedures. It is commonplace for 15 to 30 L of patient blood to be processed in a 4- to 6-hour period. Such extensive procedures may entail enough incidental platelet loss to warrant platelet transfusions for patients who are already somewhat thrombocytopenic. They also entail prolonged infusion of citrate anticoagulant and therefore a large cumulative citrate dose. Conflicting reports concerned the frequency of adverse effects during PBPC collection. A multicenter survey that included both autologous and allogeneic donors but did not track either mild paresthesias or calcium-replacement practices revealed only a 1.33% incidence of spontaneously reported adverse effects in 664 PBPC collections.108 A later study at a single center found a 54% incidence of symptoms in carefully questioned donors during 71 allogeneic PBPC collections performed without intravenous calcium supplementation. Serum ionized calcium decreased by 20% to 35% during these procedures. Symptoms were especially prevalent in smaller female donors. The same investigators found a 20% incidence of symptoms among 244 donations performed with prophylactic intravenous calcium supplementation, with only a 10% to 15% decrease in ionized serum calcium.109 Although the latter studies monitored allogeneic donations, they suggest that intravenous calcium supplementation might prevent mild paresthesias during autologous PBPC collections as well, especially those done on smaller female patients.

Autologous Mononuclear White Blood Cells Altered Ex Vivo Autologous MNCs collected by leukapheresis provide the starting material for a number of established and experimental cellular therapy techniques. Purification and Expansion Purification of MNC subsets has been explored in autologous PBPC transplantation. This has been done primarily to

address potential tumor cell contamination. In positive selection approaches, stem and progenitor cells are purified from all other MNC, including any contaminating tumor cells. The most prevalent techniques are immunologic ones that target the CD34 antigen. Several of these are commercially available. Stem cells thus purified also provide an attractive starting material for in vitro expansion techniques that could enhance the outcome of transplantation in other ways. An alternative approach to purification is negative selection to deplete tumor cells actively from an MNC concentrate. Sequential positive and negative selection approaches also are possible.110 Gene-marking experiments identified tumor cells derived from autograft contamination in metastases developing after transplantation111; however, to date, no stem cell purification process has been shown to enhance survival after transplantation in any illness. Immunization A number of in vitro strategies have been proposed to induce and/or enhance antimicrobial or antitumor immunity expressed by autologous immunocytes present in MNC concentrates. Enhanced antigen presentation, release from inhibitory influences operative in vivo, and/or selective stimulation by supraphysiologic “cytokine cocktails” are potential mechanisms that could facilitate immune responses that could not be achieved in vivo and might result in desirable therapeutic effects after reinfusion of treated cells. Some approaches of this nature have shown promise in melanoma, renal cell carcinoma, and prostate cancer.112 Gene Insertion Autologous hematopoietic stem and progenitor cells collected by leukapheresis are promising targets for therapeutic gene insertion. They can be obtained in abundance with relative ease, and they or their progeny can potentially survive indefinitely after genetic alteration and reinfusion. Encouraging early successes were achieved in correcting immunodeficiency due to adenosine deaminase deficiency, but gene therapy has subsequently proven to be a very challenging endeavor.113

RED CELL EXCHANGE Description of Procedure The most common manipulation of red cells by apheresis is red cell exchange (RCE), in which patient red cells separated by the instrument are removed while compatible donor red cells are added to patient plasma and reinfused. Two other manipulations are possible. Red cells can be depleted rapidly and isovolemically from patients with polycythemia by removing red cells and adding 5% human serum albumin (HSA) or fresh frozen plasma (FFP) at the same rate to the patient’s plasma for reinfusion.114,115 Alternatively, red cells can be transfused rapidly and isovolemically to anemic patients by removing patient plasma while adding compatible donor cells to the patient’s red cells at the same rate to be reinfused.116 The kinetics of a “pure” RCE are similar to those of TPE, such that the equation and graph in the previous chapter (see Fig. 55–1) can be used to determine the fraction of patient red cells remaining (FCR) after exchange of a given number of patient red cell volumes. Some apheresis instruments include

Sickle Cell Disease The great majority of RCEs are done for complications of SCD in patients who are homozygous for hemoglobin S (HbS) or have another hemoglobinopathy. Briefly stated, the rationale for removing patient red cells and transfusing normal red cells containing hemoglobin A (HbA) in RCE is to create a hemoglobin mixture similar to that found in sickle trait cells, in which a majority of circulating hemoglobin is HbA. Because HbS-containing cells are less likely to sickle in such cell mixtures, RCE can interrupt the vicious cycle of sickling, stasis, and progressive hypoxia that occurs during sickle cell crisis.118 The exact proportions of HbA and

HbS needed to achieve this are not known with certainty; however, plans for RCE are usually formulated with a goal of increasing HbA above 70% of the total (i.e., FCR 24%), isovolemic removal of patient red cells with HSA replacement can be carried out at the beginning of the exchange to an extent that will reduce the hematocrit to the minimum safe level. Because no donor cells are infused, none can be subsequently withdrawn, and removal of patient cells during this preliminary phase is therefore 100% efficient. Similarly, if the hematocrit at the end of an exchange is not too high (i.e., 500/μL

PLT > 20,000/μL

BM

PBSC

BM

PBSC

BM

PBSC

P

BM

PBSC P

52 91 118 116

48 81 109 163

2.4 2.4 2.4 2.7

6.6 7.3 6.7 5.8

21 21 23 15

15 16 19 12

10−5