Immunohematology JOURNAL OF BLOOD GROUP SEROLOGY AND EDUCATION

V OLUME 20, NUMBER 2, 2004 Immunohematology JOURNAL OF BLOOD GROUP SEROLOGY AND EDUCATION V OLUME 20, NUMBER 2, 2004 CONTENTS

79 Letter to the Readers Introduction to the review articles S. MURPHY,MD 80 Review: matching for alloimmunized patients—room for improvement S.T.NANCE,S.HSU, R.R.VASSALLO,AND S. MURPHY 89 Review: platelet alloantigens and and their clinical significance A. NORTON,D.L.ALLEN,AND M.F.MURPHY 103 Review: transfusion-related acute lung injury: pathophysiology, laboratory investigation, and donor management P.M. KOPKO 112 Review: immune thrombocytopenic purpura: an update for immunohematologists S.G. SANDLER 118 The sensitivity, specificity, and clinical relevance of gel versus tube DATs in the clinical laboratory N. PAZ,D.ITZHAKY,AND M.H. ELLIS 122 124 COMMUNICATIONS Letter to the Editor-in-Chief Letter to the Editors Looking back at the history of the journal An unusual reacting with pre-diluted 0.8% DENISE HARMENING,PHD, MT(ASCP) reagent RBCs and with 0.8% older (aged) RBCs prepared at the time of testing J.TRIMBLE 125 126 128 Letters From the Editor-in-Chief SPECIAL SECTION Book Review The second 20th anniversary issue! Excerpts from the Stem Cells Handbook My sincere apology Red Cell Free Press N. REBECCA HALEY,MD Memories from the past! What is next? 129 131 ANNOUNCEMENTS CLASSIFIED AD 132 134 ADVERTISEMENTS INSTRUCTIONS FOR AUTHORS EDITOR-IN-CHIEF MANAGING EDITOR Delores Mallory, MT(ASCP)SBB Mary H. McGinniss,AB, (ASCP)SBB Supply, North Carolina Bethesda, Maryland MEDICAL EDITOR SENIOR MEDICAL EDITOR S. Gerald Sandler, MD Scott Murphy, MD Washington, District of Columbia Philadelphia, Pennsylvania TECHNICAL EDITOR GUEST EDITOR Christine Lomas-Francis, MSc Scott Murphy, MD New York, New York Philadelphia, Pennsylvania ASSOCIATE MEDICAL EDITORS David Moolton, MD Geralyn Meny, MD Ralph Vassallo, MD Philadelphia, Pennsylvania Philadelphia, Pennsylvania Philadelphia, Pennsylvania EDITORIAL BOARD Patricia Arndt, MT(ASCP)SBB W. J ohn Judd, FIBMS, MIBiol Paul M. Ness, MD Los Angeles, California Ann Arbor, Michigan Baltimore, Maryland James P.AuBuchon, MD Christine Lomas-Francis, MSc Mark Popovsky, MD Lebanon, New Hampshire New York, New York Braintree, Massachusetts Geoffrey Daniels, PhD Gary Moroff, PhD Marion E. Reid, PhD, FIBMS Bristol, United Kingdom Rockville, Maryland New York, New York Richard Davey, MD Ruth Mougey, MT(ASCP)SBB Susan Rolih, MS, MT(ASCP)SBB New York, New York Carrollton, Kentucky Cincinnati, Ohio Sandra Ellisor, MS, MT(ASCP)SBB John J. Moulds, MT(ASCP)SBB David F.Stroncek, MD Anaheim, California Raritan, New Jersey Bethesda, Maryland George Garratty, PhD, FRCPath Marilyn K. Moulds, MT(ASCP)SBB Marilyn J.Telen, MD Los Angeles, California Houston, Texas Durham, North Carolina Brenda J. Grossman, MD Sandra Nance, MS, MT(ASCP)SBB St. Louis, Missouri Philadelphia, Pennsylvania

EDITORIAL ASSISTANT PRODUCTION ASSISTANT Linda Berenato Marge Manigly COPY EDITOR ELECTRONIC PUBLISHER PROOFREADER Lucy Oppenheim Paul Duquette George Aydinian

Immunohematology is published quarterly (March, June, September, and December) by the American Red Cross, National Headquarters,Washington, DC 20006. The contents are cited in the EBASE/Excerpta Medica and Elsevier BIOBASE/ Current Awareness in Biological Sciences (CABS) databases. The subscription price is $30.00 (U.S.) and $35.00 (foreign) per year. Subscriptions, Change of Address, and Extra Copies: Immunohematology, P.O. Box 40325 Philadelphia, PA 19106 Or call (215) 451-4902 Web site: www.redcross.org/pubs/immuno

Copyright 2004 by The American National Red Cross ISSN 0894-203X LETTER TO THE READERS

Introduction to the review articles P. Kopko then grapples with the syndrome of transfusion-related acute lung injury (TRALI). This may As part of the celebration of the 20th anniversary now be the second most common cause of death, after of Immunohematology,this issue contains four review transfusion, and there can be considerable morbidity articles that take the reader from the serology for those who survive. The etiology is hotly debated. laboratory to the bedside and back. This constant Most, but not all, cases develop after a transfusion interplay between the laboratory and the patient containing a relatively large volume of plasma, as in makes immunohematology an extraordinarily reward- fresh-frozen plasma or . Antibodies ing endeavor. The first three articles describe clinical to leukocytes and are often found in the mischief caused by antibodies to human platelet donor’s plasma. In a typical case, the to which (HPA), human leukocyte antigens (HLA), the antibody has formed is present on the patient’s human antigens (HGA), and human white blood cells. antigens. The fourth review describes a red Finally, G. Sandler discusses the clinical blood cell antibody that went from the bedside to the characteristics and pathophysiology of idiopathic laboratory and then returned to the bedside as therapy thrombocytopenic purpura (ITP). Recently, several for patients with low platelet counts. therapeutic modalities have been added to the old S. Nance et al. describe the exploits of a young standbys, corticosteroids and splenectomy. These physician, Dr. Peter Petz-Garratty, and then compare include intravenous gamma globulin (IVIg), WinRho and contrast the evolutions of red blood cell matching (sometimes referred to as anti-D), and Rituximab. For and platelet matching. Most patients with refracto- half a century, the ability to detect anti-D, using the riness to have HLA antibodies. The antiglobulin test, has been crucial to the management authors conclude that perhaps methods of platelet of many patients. Now it has returned to the bedside matching should be modeled after red blood cell as a therapeutic agent. matching. In fact, this possibility has been made more attractive by the development of new, more sensitive Scott Murphy, MD methods to determine the specificities of HLA Chief Medical Officer and antibodies. Guest Editor of this issue, A. Norton et al. provide a thorough review of the American Red Cross Blood Services HPAs that are present on platelets but not on other Penn-Jersey Region blood cells. During pregnancy, a woman may develop 700 Spring Garden Street HPA antibodies, and they, in turn, may cross the Philadelphia, PA 19123 placenta to produce thrombocytopenia in the fetus (NAIT). In other patients, HPA antibody may be produced in response to associated with the puzzling syndrome posttransfusion purpura (PTP).

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 79 Review: platelet matching for alloimmunized patients—room for improvement

S.T.NANCE,S.HSU, R.R.VASSALLO,AND S. MURPHY

A Parable The next day,a new resident,Dr.Peter Petz-Garratty, A 40-year-old woman was admitted at midnight to came on service. He reviewed the chart and said,“Why HMO General Hospital for evaluation of profound don’t you examine the patient’s serum for antibodies fatigue. Her hemoglobin was found to be 2 g/dL. Her and determine their antigenic specificities? Then, medical history included several pregnancies and crossmatch products without those antigens.” transfusions. The intern had attended a lecture on The medical director said,“This man has a future.” that day. He ordered a “six-pack,” six units of random-donor RBCs. The technologist on Evolution of Strategies for RBC Matching call for the was a bacteriologist, cross- Practice was quite primitive when physicians first covering the transfusion service. The technologist considered transfusing blood from one individual to removed six units of RBCs from the refrigerator,pooled another. They began transfusing between them, and sent a liter of the mixture to the floor for experimental animals, without any consideration transfusion, which was quickly followed by back pain whatsoever for RBC antigen-antibody compatibility. and black urine for the patient. Interestingly, this lack of compatibility testing persists The intern called the on-call transfusion service even today for the majority of platelet transfusions. medical director, who was cross-covering from In an excellent historical review,1 Myhre described cardiology. The intern asked, “Can’t you do immuno- a report by John Wilkins of a dog-to-dog transfusion of logic testing or provide a matched product?” about 2 ounces of blood via syringe. At that time,it was The director responded,“No way! What was the 1- not recognized that there was a possibility of naturally hour CCI? We insist upon two documented failures occurring alloantibodies (not present in dogs) or the before we do an immunology workup. I’m sending the relevance of a volume possibly too small to produce a patient another pool.” reaction. Other investigators in London and France “Oh no, not again!” said the patient. Shortly after also experimented with animal-to-animal transfusions. transfusion, back pain and black urine recurred. This was done,of course,without compatibility testing. The resident called the medical director and said, Since animal-to-animal transfusions went quite well, “That’s two, Doc!” soon the time came to try animal-to-human trans- “OK.We’ll phenotype the patient’s RBCs and search fusions. Jean Baptiste Denys performed the first animal- for a good BX match in inventory. You know,Jka and Jkb to-human transfusion in mid-1667.2 Nine ounces of are crossreactive.” lamb’s blood was transfused to a 16-year-old patient Once again the patient said, “Are you sure?” who had been bled at least 20 times over 2 months. However, the transfusion was given and soon she was During the transfusion, it was reported that the patient having back pain and black urine again. experienced “very great heat along his arm.” Again, no In desperation, the medical director said,“I’ll take testing of the blood or the recipient was performed. random units off the shelf and crossmatch until I find a James Blundell was a major contributor to the early compatible unit.” The unit was found and transfused years of human-to-human blood transfusion. He and there was no back pain or black urine until 5 days reported a transfusion to a patient with obstinate later. vomiting, in 1819. The donors were several physicians;

80 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: platelet matching for immunized patients

12 to 24 ounces of blood was given by syringe.2 This is bovine albumin11 and antiglobulin testing.12 Although reminiscent of the use of six packs for platelet other enhancement media and techniques (enzyme, transfusion today. As in the preceding two accounts,no LISS, PEG, and solid phase and column agglutination laboratory testing was carried out on the donor or technology) have been described, the basic premise of recipient blood. These early reports share relatively the major crossmatch remains remarkably the same,ie., small transfused volumes and a lack of pretesting. One with testing of the patient’s serum against donor RBCs. report remarked that animal blood was thought to be At first, the minor crossmatch was also felt to be more pure, as the animal did not participate in some of necessary, but a paradigm shift occurred. The minor the risky pursuits that humans did. crossmatch was eventually abandoned in favor of Fast forward to today,when involved parties’blood, routine donor antibody screening. The need for both donor and recipient, are subjected to a large complement-reactive antihuman globulin (AHG) sera battery of tests. But how did the testing algorithm has been eliminated because of the rare occurrence of develop, and why? As mentioned previously, the first clinically significant antibodies leaving only comple- transfusions were given without testing of the donor or ment on the cell surface. The requirement for serum as the recipient. Interest in transfusion waned in the late a source of a test sample has been changed for the 19th century due to a high rate of severe adverse same reason. reactions to transfusion.3 Landsteiner’s experiments Another change was the departure from room and direct transfusions by Carrel, described by Crile,4 temperature testing to 37°C for detection of potentially promoted a recurrence of interest. The subsequent significant antibodies. Most clinically significant determination of ABO groups gave rise to renewed antibodies are reactive at 37°C, and antibodies reactive interest in human-to-human blood transfusion. solely at room temperature are generally not thought to In a publication by Ludwig Hektoen,5 what would have clinical significance. become the major and minor crossmatches were More recently, AABB standards have allowed the described in 1907. In the early days, the methods use of immediate spin testing as long as the patient’s concentrated solely upon detection of room antibody screen is negative. In some laboratories, temperature IgM antibodies, as the “” had electronic crossmatches have taken the place of the not yet been described. Thus, these investigators were AHG crossmatch. Thus, few patients’ samples are studying ABO compatibility. Ottenberg6 and Epstein7 routinely subjected to an AHG crossmatch. When the studied a method for and agglutination tests. antibody screen is positive, RBC units are selected that Epstein promoted a pretransfusion crossmatch, using lack antigens to which the antibody is directed. This is tests both for hemolysis and for agglutination. It was quite similar to the evolution of antibody-based later recognized that when agglutination was present matching for platelets. hemolysis might also be present, but in vitro hemolysis The number of RBC transfusions requiring a full did not occur without agglutination.8 With this crossmatch is proportionate to the number of platelet information, agglutination tests could be used without transfusions currently guided by crossmatch testing, tests for hemolysis. At this time, all testing still but for vastly different reasons. In the case of potential concentrated upon detection of IgM antibodies. RBC recipients, an AHG crossmatch will be required by Changes like the recommendation for testing at 37°C the serologic determination of alloantibodies or and concentration upon the major crossmatch . In the case of potential platelet followed. During World War I,Lee proposed that if type recipients, however, crossmatching is ordered after O blood were used, preliminary testing could be several transfusions have not yielded expected avoided.9 increments. The present status of platelet cross- These red cell innovations in the early part of the matching, then, parallels the state of RBC testing in the 20th century were quite similar to the use of “blind” early 1900s. crossmatching to assess platelet compatibility today. The first practical method for platelet crossmatching Evaluation of Strategies for Platelet Matching was not described until 1988.10 This will be discussed In current practice,platelet transfusion begins with further in the next portion of this paper. randomly selected platelet products, either pooled Major changes to detect IgG antibodies followed whole-blood–derived platelets or platelets prepared by the reports of enhancement of agglutination with apheresis. This is analogous to the initial management

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 81 S.T.NANCE ET AL. of the anemic woman in the parable. This approach antigens within CREGs of their own HLA antigens.16 continues until the patient is refractory, i.e., fails to Therefore,Duquesnoy et al.16 defined matches between show adequate increments after transfusion. To donor and patient as A, HLA identical; BU, (partially identify a patient as refractory, most clinicians require homozygous) donor antigens all present in the two or three failed transfusions. Again, this was the recipient; BX, all donor HLA antigens either identical to way the patient in the parable was managed. or within recipient CREGs; C & D, one or more donor Commonly, increments are expressed as corrected antigens not present in recipient or in recipient CREGs. count increments (CCI) where This is analogous to the phenotypic matching of RBCs. The use of BX matches greatly expanded the CCI = Plt. Ct. Increment (/µL) × Body Surface Area (m2) number of potential donors. This was an advance Plts.Transfused (× 1011) because many BX matches provided good increments in the refractory patient. However, the predictive Refractoriness is commonly defined as a CCI less power of this approach was poor. Poor responses were than 4000 per µL per m2 per 1011 platelets.13 Thus, frequent with BX matches and a surprising number of patients have to “earn” immunologic intervention by good responses were seen with C and D matches. The experiencing several therapeutic failures, as the patient in the parable had a poor response to a woman in the parable did. This reactive policy “phenotypic match” at the Kidd locus. contrasts with RBC transfusion,where patients are now The study of Duquesnoy et al.16 can be criticized in screened for antibody before each transfusion, as retrospect because no serologic assessment was suggested by “Dr. Petz-Garratty.” Most immunologic platelet refractoriness results performed. They did not know if the patients had HLA from antibodies to HLA. It is important to emphasize antibodies and, if they did, what the specificities of the that clinical refractoriness is often due not to alloim- antibodies were. These determinations are always munization but, rather, to clinical factors such as performed prior to a RBC transfusion, as was suggested infection, splenomegaly, or disseminated intravascular by “Dr. Petz-Garratty.” coagulation.14 Thus, it would make sense to perform The poor predictive power of the CREG method one or more serologic tests to prove that alloimmu- can be understood to some extent by examining the nization was present. However, this approach was rare results of serologic studies in these patients. For many during the first 20 to 25 years of matched-platelet years,the standard serologic test for HLA antibodies has transfusions.15,16 been a lymphocytotoxicity (LCT) assay, the antihuman The classical HLA approach to platelet matching globulin–augmented complement-dependent cyto- began with the paper by Yankee et al.15 that compared toxicity test (AHG-CDC). In this assay,leukocytes from the use of platelets from HLA-identical siblings with 50 to 100 donors with an appropriate heterogeneity of randomly selected platelets in refractory patients. HLA types are incubated in wells with the patient’s There were no increments with random products but serum and complement. LCT is assessed micro- excellent responses to the perfectly HLA-matched scopically. The presence of HLA antibody correlates sibling donors. with poor response to randomly selected platelets17 It is generally not realistic to support refractory and with improved response to HLA-selected patients with sibling donors. Duquesnoy et al.16 patients.18 On the other hand, if such antibodies are introduced the concept of supporting patients with not present, poor responses are infrequently improved HLA-matched platelets from the general population. by HLA matching.18–20 However, they noted that a pool of 50,000 donors In our laboratory, one half of patients referred for would be required to provide a 50 percent probability matched platelets have no antibody.21 We discourage of finding ten perfectly matched donors for each of 100 matching in such patients except in the rare potential recipients. This was also unrealistic. They circumstance that immunologic refractoriness is proposed a method that we will call the CREG method. mediated solely by antibodies to human platelet Cross-reactive groups (CREGs) of HLA antigens have antigens (HPAs). The results of the AHG-CDC assay been defined by serologic testing. Cross-reactivity determine not only the presence or absence of among antigens within a CREG results from the sharing antibody, but also the breadth of of one or more public . Patients do not make expressed as the percent reactive antibody (PRA), the antibody to their own HLA antigens nor commonly to percentage of wells in which lymphocytotoxicity is

82 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: platelet matching for immunized patients seen. Furthermore, by analyzing the positive wells, one The crossmatch method has the advantage of being frequently can determine the specificities of the fast, once the patient’s serum is in hand. In addition, it antibodies. When we analyze results with the LCT is a form of screening for the presence of antibody. In assay in refractory patients, the PRAs can range from 1 our experience21 comparing the SPRCA to the AHG- percent to 100 percent, so one can classify patients as CDC assay, both tests detect antibody in 50 percent of mildly (1–40% PRA), moderately (40–70% PRA), and referred patients, while all crossmatches are compat- severely (PRA > 70%) immunized. ible in 30 percent. It is very rare for the SPRCA to be Furthermore, when the specificities are entirely negative (compatible) when the AHG-CDC is determined, frequently one can identify intra-CREG positive. Thus, if all crossmatches are compatible, it is antibodies: antibodies against antigens within the very unlikely that the patient is refractory for patient’s CREGs. With these facts in mind, one can immunologic reasons. On the other hand, 20 percent understand why the CREG method has a relatively poor of these cases are SPRCA positive and AHG-CDC predictive capacity. When a BX match fails, there may negative. The majority of such cases are seen in group be intra-CREG antibody. When a C or D match O patients who have positive reactions to donors who succeeds, the patient, although immunized, may have are group A and/or B. In this situation, the positive no antibody against the antigens present on the crossmatch would be considered falsely positive since transfused platelets. There is a computer program the platelets may be an excellent HLA match. available to better predict successful BX and C and D Recent results from Gelb and Leavitt24 in 76 matches.22,23 It is known as “HLA Matchmaker.” potentially immunized patients indicated a mean CCI of Furthermore, donor selection may be improved by 1800 for randomly selected products and 9800 for “HLA Matchmaker.” It is based upon the concept that crossmatch-compatible products. However, there were HLA molecules express their antigenic determinants as no compatible products for ten patients. Another mode amino acid triplets. These triplets are shared among of support based on HLA matching might have helped otherwise unrelated HLA molecules. Antigens whose these ten patients. Furthermore, in the other 66 triplet complement is identical to ones present in the patients, the mean percentage of screened units that patient are highly likely to provide compatible were compatible was 69 percent (range: 24–100%), products for an immunized patient. BX matches with suggesting that these patients were only mildly or fewer triplet mismatches may be more likely to moderately immunized. Overall, 59 percent of succeed, as would triplet-identical C and D matches. crossmatch-compatible products produced CCIs greater This concept awaits validation in a clinical study. than 7500. On the other hand, 41 percent did not. A quite different approach, crossmatching of donor Crossmatching is an excellent way to begin support platelets with a patient’s serum, was introduced in when a patient needs a product urgently, if appropriate 1988 and, later, in 1997.10,24 The most commonly used technical staff are available (i.e., perhaps not on nights method, at least in the United States, has been the solid or weekends). It provides a quick answer to the phase red cell adherence (SPRCA) assay, which is question of whether or not the patient is alloimmunized. available commercially as Capture-P from Immucor, In our experience,it is generally satisfactory for mild and Inc. (Norcross, GA). The platelets of the potential moderate degrees of alloimmunization but less helpful donor are layered in microtiter wells and the patient’s for the severely immunized. Furthermore, it may find a serum is added. The platelets are then washed and anti- very good HLA match incompatible because of ABO IgG–coated indicator RBCs are added. If there is antibodies in the patient. In our experience21 with antibody to the platelets, the RBCs form a thin film in highly immunized patients, the majority of crossmatch- the well. If there is no antibody, the RBCs puddle in a compatible products are not successful unless the button on the bottom of the well. degree of HLA compatibility is high. Many centers simply take platelets from inventory A final approach was suggested by Dr. Peter Petz- and perform “blind” crossmatching, with no pre- Garratty in the parable. Why not examine the patient’s selection of the donors by identifying the antigens to serum for antibodies and determine their antigenic which the patient has formed antibodies. This stands in specificities? Then, provide antigen-negative products, contrast to RBC matching, in which cells are perhaps with a final check by crossmatching. In fact, deliberately chosen for crossmatching that lack the this approach was evaluated by Petz and Garratty et antigens to which the patient has been sensitized. al.25 in a large number of alloimmunized patients. They

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 83 S.T.NANCE ET AL. called it the antibody specificity prediction (ASP) not tolerate this with RBCs, since morbidity from an method. The specificities of the patient’s antibodies incompatible RBC transfusion can be severe. However, were determined with the AHG-CDC method and there is morbidity from a failed platelet transfusion: a antigen-negative platelets from an HLA-typed donor continuation of the patient’s bleeding risk. pool were selected for transfusion. This approach was In addition, a new generation of technologies compared to “blind” crossmatching. The increments developed and used for organ transplantation can also were better with the ASP method, although not be used for the platelet-refractory patient. These are statistically so since there was a large range around the discussed in detail in the next section. Briefly, new means for both methods. The authors also found that a techniques for determining the specificity of perfect A or BU match produced the best result. antibodies take advantage of the ability to use isolated, We studied the PRA of patients over time, up to a single HLA molecules in ELISA or flow cytometric– maximum of 16 months. Our results were similar to based assays. This is in contrast to the AHG-CDC assay, those of Lee and Schiffer.26 PRAs tend to remain in which two to four antigens are expressed on each constant over time, even as the patients continue to be target . transfused. On the other hand, approximately 25 In addition, there are simple ELISA methods in percent of immunized patients may lose their which the target is a wide variety of HLA antigens. One antibodies over a period of months, a phenomenon could use these methods to screen patients for the that we do not understand. In practice, this means that presence of HLA antibodies, perhaps every 2 to 4 one can perform an AHG-CDC assay once monthly to weeks. With this approach, one can be proactive and monitor the patient’s status without concern for major begin matching earlier. We could abandon the practice changes during the period. of forcing patients to “earn” their matched products by A great advantage of the ASP method is that it experiencing several therapeutic failures. increases the number of potential donors. Further, if one has identified the specificities of the patient’s Platelet Matching: New Technology for the antibodies, one can distribute compatible products Future from inventory at any time without the need for For solid organ transplantation, the detection of testing. Our approach is outlined in Table 1. circulating anti-HLA antibodies in allograft recipients 25 It is clear that what Petz and Garratty et al. are correlates not only with hyperacute, acute, accelerated, suggesting brings us to a method very similar to what and chronic graft rejection,27–29 but also with reduced we do with RBCs. We are using more complex methods graft function and graft failure. Over the past 40 years, but their cost is less than that of a failed matched- a plethora of papers has been published, discussing the platelet product. With current methods, frequent failed merits of various antibody detection methodologies in platelet transfusions are accepted routinely. We would relation to the clinical outcomes of solid organ transplants. In contrast, very little can be found in the Table 1. Suggested patient management approach literature relating these methods to platelet transfusion A. Patient Referral: 1. Begin support from inventory by: outcome. This may partly be due to the difficulty of a. classic CREG method16 using patient’s HLA type such research due to the confounding effect of variables b. crossmatching of random units by SPRCA24 such as storage-related changes in platelet viability, 2. Particularly if using (a), strongly suggest HLA antibody screen 3. Urge clinicians to provide CCIs concurrent recipient nonimmunologic plate-let B. If: destruction, and the difficulties of demonstrating 1. No HLA antibody is demonstrated, return to randomly selected clinically significant changes in outcome in the context products; consider testing for HPA antibody, especially if concomitant nonimmunologic platelet destruction is unlikely of prophylactic platelet use. This meager literature is 2. Only ABO antibody is demonstrated, use platelets from ABO-identical further confused by the use of less sensitive “gold or group O donors standard”assays (unenhanced lymphocytotoxicity tests) 3. HLA antibody is demonstrated but PRA < 70%, support from 30 inventory using crossmatching24 or antigen-negative platelets selected for antibody screening in many comparative studies. by antibody specificity prediction method25 The objective of this portion of the review is to C. Monitor CCIs to judge response discuss different testing methodologies currently in use D. If PRA > 70% and/or CCIs are poor: 1. Crossmatch antigen-negative platelets24,25 for the detection of anti-HLA antibodies as they relate 2. Establish transfusion schedule and selectively recruit A, BU, or other to platelet transfusion. We will also outline what we very close matches believe to be the optimal approach to characterizing

84 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: platelet matching for immunized patients the HLA antibody profile of highly sensitized platelet- The original NIH two-stage complement- refractory patients. dependent cytotoxicity assay (CDC)36 used for HLA HLA Class I antigens are expressed on all nucleated phenotyping is illustrated in Figure 1,30 which depicts cells, including platelets. HLA Class I antibodies are the two steps of the basic CDC method with relevant not only to solid organ and tissue trans- alternative steps (indicated by asterisks) included to plantation, but also to WBC and platelet transfusion. illustrate (1) extended incubation to allow low Class II antigens are expressed on B cells, monocytes, avidity/affinity antibodies to bind cellular antigens; (2) ,dendritic cells,and activated T cells. HLA a wash step to remove unbound serum and minimize Class II antibodies are therefore unimportant in the “anticomplement factors,” which may inhibit or context of platelet transfusion. weaken complement activity (Amos modified Within the relevant Class I HLA-A and HLA-B technique)37; and (3) goat AHG-CDC, which overrides families (HLA-C being present on the platelet surface in inefficient C1q binding by single IgG molecules.38,39 a density too low to be clinically important), there are Many of the anti-public or “CREG” antibodies are not several CREGs, which express multiple public detectable by the standard CDC method due to epitopes. These public epitopes are common antigenic inefficient complement activation by HLA antibodies. determinants shared by different HLA antigens. HLA-B This is termed the cytotoxic-negative, adsorption- 40 locus antigens have two mutually exclusive public positive phenomenon (CYNAP). The AHG-CDC epitopes, Bw4 and Bw6, as well as a number of others method is the most sensitive of the four complement- defining the CREGs, such as 5C, 7C, 22C, 27C, 8C, 12C, dependent cytotoxicity tests. In some laboratories, the and 21C. Private epitopes refer to antigenic sensitivity of the AHG-CDC is comparable to the determinants specific for an individual antigen. In multiple-antigen-panel ELISA-based technique. addition to “foreign” CREGs, patients can occasionally The major advantage of complement-dependent make antibodies to antigens within their own CREGs, assays is the ability to detect both IgG and IgM either to private epitopes or to different public antibodies. While IgM anti-HLA antibodies do not play a epitopes within a CREG.31 The CREG antibody profile role in solid organ transplantation, this may not be the remains remarkably constant over time in patients.31,32 case in platelet transfusions. The disadvantages of Monthly fluctuations in PRA, if not due to new complement-dependent assays include the detection of not only HLA-specific antibodies but also nonspecific transfusion events or changes in assay technique or specificities or autoantibodies. The latter can be panel composition, usually reflect the waxing and eliminated by heating the serum at 63°C for 10 minutes, waning of detectable specificities within the same CREG, or they can reflect the increasing strength or number of reactivities within the same CREG cluster.30 Stage One: Incubate Cells In the transfusion setting, we have observed that Antigen-Antibody with Antibody greater than 50 percent of sera screened by flow Interaction × cytometric methods contained both IgM and IgG Wash 3 or 4* antibodies.33 These are mostly directed against public 30–60* min. epitopes or high-frequency private epitopes present in Add AHG* transfused units. Additionally, some antibodies arise from an anamnestic response from prior sensitization Stage Two: Add Rabbit Serum 2 min. to paternal antigens during pregnancy.34,35 Complement- (Complement Source) Antibody screening provides information regarding Mediated Cell Injury the presence or absence of HLA antibodies, the PRA against HLA alloantigens, antibody specificities, the 60–120* min. immunoglobulin isotypes produced, and even antibody titers. This information may in turn be used to predict products likely to provide successful patient platelet Add Vital or Supravital Dyes concentration increments.21,25 The ideal screening Visualize Cell Membrane Injury method should be sensitive enough to detect the lowest level of clinically relevant IgM and/or IgG Fig. 1. The original NIH two-stage complement-dependent cytotoxicity antibodies. assay (CDC) and its modifications *(see text).

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 85 S.T.NANCE ET AL. treating the serum with DTT, or performing an technology. This technology combines two of the most autocrossmatch in parallel. CDC-determined PRAs may powerful detection and identification tools, the fluctuate due to changes in panel cell composition or exquisite binding specificity between biotin and poor viability of frozen cells. Without prior adsorption, streptavidin and the versatility of the flow cytometer. sera obtained from patients receiving OKT3 or anti- By the precise blending of different fluorescent immunoglobulins cannot be tested by CDC methods. intensities of the two dyes, 100 unique color codes are There are two forms of solid phase–based methods: created. When the colored beads are coated with the ELISA41,42 and flow cytometric assays.43–45 These purified Class I antigens,they can be used for screening methods employ purified soluble HLA antigens affixed and identification of HLA antibody specificities. to solid phase matrices (plastic trays for ELISA or beads Although a number of large parallel studies between for flow cytometry) as targets for the binding of HLA standard and Luminex flow methods are still in antibodies instead of the live cells employed in CDC progress, data published thus far indicate that they are assays. likely to be comparable to each other.47 The advantages of solid phase–based methods are It is well established that flow cytometry methods the ability to distinguish between IgG and IgM are more sensitive than AHG-CDC and ELISA. Better isotypes, better detection of HLA-specific antibodies, clinical correlations are obtained using flow methods objective end-result determination by instrument not only for antibody screening but also for optical density readings, the lack of assay interference transplantation crossmatches. Flow methods can by OKT3 treatment, and faster turnaround time. detect extremely low-level antibodies undetectable by Currently,ELISA assays are available in two or three either AHG-CDC or ELISA methods.48–50 formats. The first is the screening test for the presence Considering reported sensitivities and specificities or absence of HLA antibodies. A pool of HLA antigens together with our own comparison studies of AHG- is affixed to a single well for this purpose. Positive sera CDC versus ELISA or flow methods,33 our ranking of are further defined using a panel of HLA antigens preferred antibody elucidation methods for platelet- purified from EBV cell lines or platelets to allow the refractory patients is: flow cytometry (HD panel) > determination of PRA and antibody specificity. The ELISA (HD panel) > AHG-CDC. sensitivity and specificity of ELISA using heterozygous The optimal approach to characterizing the HLA antigen panels (i.e., each well coated with multiple antibody profiles of highly sensitized patients, in our antigens) is comparable to that of the AHG-CDC opinion, is to HLA type potential platelet recipients and method. A third type of ELISA tray is the high-definition to perform a thorough determination of the antibody (HD) or single-antigen tray. Each well in the tray is specificities using either ELISA or flow methods on coated with a single HLA Class I–soluble antigen high-definition antigen or bead panels. Knowledge of derived from recombinant DNA technology.46 This the recipient’s HLA type assists in analyzing the high-definition antigen panel provides the ideal tool to antibody profiles. Patients can only produce HLA characterize antibody specificities contained in high- antibodies against mismatched public and private PRA sera. Both the sensitivity and the specificity of epitopes. Patients are less likely to be sensitized by screening sera using high-definition ELISA trays are antigens belonging to their own CREG groups,although increased when compared with the AHG-CDC intra-CREG antibody formation is not uncommon. method.47 In general,the ELISA method is not sensitive Screening high-PRA individuals using high for the detection of IgM HLA antibodies. definition trays sometimes results in near-100 percent Similar to the variety of ELISA methods, there are panel reactivities. This is mostly due to the presence of also three kinds of flow beads available: polyspecific high-titer antibodies. We now use high initial optical beads, useful as a screening tool; beads coated with density readings (O.D. > 2.0) from the screening assay heterozygous soluble Class I antigens to determine for the presence of HLA antibodies as a guide to dilute PRA; and antibody specificity or high-definition flow these high-titer PRA sera 1:10 or 1:20 for subsequent beads. Currently the high definition beads consist of screening on high-definition antigen or bead panels. 34 individual recombinant A locus antigens, 57 B locus The few antigens to which these individuals are not antigens, and 19 C locus antigens. sensitized may be confirmed with compatible Recently, another flow cytometric platform was lymphocyte crossmatches from products negative for developed, based upon Luminex Microsphere (bead) the corresponding antigens.

86 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: platelet matching for immunized patients

Our ultimate goal is to rapidly identify individuals 15. Yankee RA, Grumet FC, Rogentine GN. Platelet with HLA Class I antibodies and attempt to provide so- transfusion. The selection of compatible platelet called antigen-negative platelet products, even to highly donors for refractory patients by lymphocyte HL-A sensitized patients, akin to the traditional approach typing. N Engl J Med 1969;281:1208-12. outlined above for RBC transfusions. While this approach 16. Duquesnoy RJ, Filip DJ, Rodey GE, et al. Successful may be somewhat expensive compared with the current transfusion of platelets “mismatched” for HLA practice of therapeutic trial and error, the savings antigens to alloimmunized thrombocytopenic engendered by early identification of alloimmunization patients.Am J Hematol 1977;2:219-26. and consequent avoidance of failed transfusion, 17. Hogge DE, Dutcher JP, Aisner J, Schiffer CA. prolonged thrombocytopenia, and associated morbidity Lymphocytotoxic antibody is a predictor of and costs may be justified in the long run. response to random donor platelet transfusion.Am J Hematol 1983;14:363-9. References 18. McFarland JG, Anderson AJ, Slichter SJ. Factors 1. Myhre, BA. The first recorded blood transfusions: influencing the transfusion response to HLA- 1656–1688. Transfusion 1990;30:358-62. selected apheresis donor platelets in patients 2. Myhre BA. James Blundell—pioneer transfusionist. refractory to random platelet concentrates. Br J Transfusion 1995;35:74-8. Haematol 1989;73:380-6. 3. Oberman HA. The crossmatch. A brief historical 19. Phekoo KJ, Hambley H, Schey SA, et al. Audit of perspective.Transfusion 1981;21:645-51. practice in platelet refractoriness. Vox Sang 4. Crile GW.Hemorrhage and Transfusion. New York: 1997;73:81-6. D. Appleton & Co., 1909:318. 20. Engelfriet CP, Reesink HW, Aster RH, et al. 5. Hektoen L. Isoagglutination of human corpuscles. Management of alloimmunized, refractory patients JAMA 1907;48:1739. in need of platelet transfusions. Vox Sang 6. Ottenberg R. Transfusion and arterial anastamosis. 1997;73:191-8. Am J Surg 1908;47:486. 21. Murphy S, Varma M. Selecting platelets for 7. Epstein AA, Ottenberg R. A method for hemolysis transfusion of the alloimmunized patient:A review. and agglutinin tests. Arch Intern Med 1909;3:286. Immunohematology 1998;14:117-23. 8. Moss, WL. Studies on isoagglutination and 22. Duquesnoy RJ. HLA Matchmaker: A molecularly isohemolysis. Bull J Hopk Hosp 1910;21:63. based algorithm for determin- 9. Lee RI.A simple and rapid method for the selection ation. I. Description of the algorithm. Hum of suitable donors for transfusion by the determin- Immunol 2002;63:339-52. ation of blood groups. Br Med J 1917; 2:684. 23. Duquesnoy RJ, Marrari M. HLA Matchmaker: A 10. Rachel JM, Summers TC, Sinor LT,Plapp FV.Use of a molecularly based algorithm for histocompatibility solid phase red blood cell adherence method for determination. II.Verification of the algorithm and pretransfusion platelet compatibility testing, Am J determination of the relative immunogenicity of Clin Pathol 1988;90:63-8. amino acid triplet-defined epitopes. Hum Immunol 11. Diamond LK, Denton RL. Rh agglutination in 2002;63:353-63. various media with particular reference to the 24. Gelb AB, Leavitt AD. Crossmatch-compatible value of albumin. J Lab Clin Med 1945;30:821. platelets improve corrected count increments in 12. Coombs RR, Mourant AE, Race R.A new test for the patients who are refractory to randomly selected detection of weak and “incomplete”Rh agglutinins. platelets.Transfusion 1997;37:624-30. Br J Exp Pathol 1945;26:255-66. 25. Petz LD, Garratty G, Calhoun L, et al. Selecting 13. Bishop JF,Matthews JP,Yuen K, et al.The definition donors of platelets for refractory patients on the of refractoriness to platelet transfusions. Transfus basis of HLA antibody specificity. Transfusion Med 1992;2:35-41. 2000;40:1446-56. 14. Bishop JF,McGrath K,Wolf MM,et al.Clinical factors 26. Lee EJ, Schiffer CA. Serial measurement of influencing the efficacy of pooled platelet trans- lymphocytotoxic antibody and response to fusions. Blood 1988;71:383-7. nonmatched platelet transfusions in alloimmunized patients. Blood 1987;70:1727-9.

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 87 S.T.NANCE ET AL.

27. Terasaki PI, Marchioro TL, Starzl TE. Sero-typing of 40. Yunis EJ, Ward FE, Amos DB. Observations on the human lymphocyte antigens: preliminary trials on CYNAP phenomenon. In: Terasaki PI, ed. Histo- long-term kidney homograft survivors. In: compatibility 1970. Copenhagen: Munksgaard, Histocompatibility Testing. Report of a Conference 1970:351-6. and Workshop. Washington DC: National Academy 41. Kao KJ, Scornik JC, Small SJ. Enzyme-linked of Sciences—National Research Council 1965: immunoassay for anti-HLA antibodies—an 83-96. alternative to panel studies by lymphocytotoxicity. 28. Kissmeyer-Nielsen F,Olsen S, Peterson VP,Fjeldborg Transplantation 1993;55:192-6. O. Hyperacute rejection of kidney allografts, 42. Wang G, Tarsitani C, Denham S, Chen M, Lee J-H. A associated with pre-existing humoral antibodies new micro-ELISA for anti-Class I and Class II anti- against donor cells. Lancet 1966;2(7465):662-5. body detection. Human Immunol 1998:59(1): 133. 29. Terasaki PI. Humoral theory of transplantation. Am 43. Garovoy MR, Rheinschmidt MA, Bigos M, et al. Flow J Transplant 2003;3:665-73. cytometry analysis: A high technology crossmatch 30. Rodey G. HLA beyond tears: introduction to human technique facilitating transplantation. Transplant histocompatibility. 2nd ed. Durango, CO: De Novo, Proc 1983;15:1939-44. Inc., 2000: Chapters 4, 8, 9 and 10. 44. Pei R,Wang G, Tarsitani C, et al. Simultaneous HLA 31. Rodey GE,Revels K,Fuller TC. specificity of Class I and Class II antibodies screening with flow HLA Class I alloantibodies. II. Stability of cross- cytometry. Human Immunol 1998;59:313-22. reactive group antibody patterns over extended 45. Pei R, Lee J, Chen T,et al. Flow cytometric detection time periods.Transplantation 1997;63:885-93. of HLA antibodies using a spectrum of microbeads. 32. Rodey GE, Neylan JF, Whelchel JD, et al. Epitope Hum Immunol 1999;60:1293-1302. specificity of HLA Class I alloantibodies. I. 46. Pei R, Lee JH, Shih NJ, et al. Single human leukocyte Frequency analysis of antibodies to private versus antigen flow cytometry beads for accurate public specificities in potential transplant identification of antibody recipients. Hum Immunol 1994;39:272-80. specificities.Transplantation 2003;75:43-9. 33. Meny GM, Hsu S, Murphy S, Pei R. ELISA and flow 47. Visuals of the Clinical Histocompatibility cytometry provide advantages over lympho- Workshop, Cancun, Mexico 2002, Hawaii 2003, cytotoxicity in detecting HLA antibodies. (In sponsored by One Lambda, Inc. Paul Terasaki, Ed. preparation) 2004. 48. Tambur AR, Bray RA, Takeomoto SK, et al. Flow 34. Scornik JC, Ireland JE, Howard RJ, Pfaff WW. cytometric detection of HLA-specific antibodies Assessment of the risk for broad sensitization by as a predictor of heart allograft rejection. blood transfusions. Transplantation 1984;37: Transplantation 2000;70:1055-9. 249-53. 49. Gebel HM, Bray RA. Sensitization and sensitivity: 35. Braun, W.E. Donor-specific antibodies. Clinical defining the unsensitized patient. Transplantation relevance of antibodies detected in lymphocyte 2000;69:1370-4. crossmatches. Clin Lab Med 1991:11:571-602. 50. Gebel HM, Bray RA, Nickerson P. Pre-transplant 36. Terasaki PI, McClelland JD. Microdroplet assay of assessment of donor-reactive, HLA-specific anti- human serum cytotoxins. Nature 1964;204: bodies in renal transplantation: contraindication vs. 998-1000. risk.Am J Transplant 2003;3: 1488-1500. 37. Amos DB, Bashir H, Boyle W, et al. A simple micro cytotoxicity test.Transplantation 1969;7:220-3. Sandra Taddie Nance, MS, MT(ASCP)SBB, Director, 38. Johnson AH, Rossen RD, Butler WT. Detection of Technical Services; Susan Hsu, PhD, Director, alloantibodies using a sensitive antiglobulin Histocompatibility/Molecular Genetics; Ralph R. microcytotoxicity test: identification of low levels Vassallo, Jr., MD, Senior Medical Officer; Scott Murphy, of pre-formed antibodies in accelerated allograft MD, Chief Medical Officer, American Red Cross Blood rejection.Tissue Antigens 1972;2:215-26. Services, Penn-Jersey Region, 700 Spring Garden 39. Fuller TC, Fuller AA, Golden M, Rodey GE. HLA Street, Philadelphia, PA 19123. alloantibodies and the mechanism of the antiglobulin-augmented lymphocytotoxicity proce- dure. Hum Immunol 1997;56:94-105.

88 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: platelet alloantigens and antibodies and their clinical significance

A. NORTON,D.L.ALLEN, AND M.F.MURPHY

Antigens on human platelets are categorized deficiency of these GPs results in bleeding disorders, according to their biochemical nature into carbohy- e.g., lack of GPIIb/IIIa causes Glanzmanns’ drate antigens on glycolipids and glycoproteins (GPs) thrombasthenia, and absence of GPIb/IX/V results in (A, B, O, P, Le antigens), protein antigens (human Bernard-Soulier syndrome. The expression of platelet leukocyte antigens [HLA] Class-I A, B, and C, GPIIb/IIIa, alloantigens located on these GPs may be altered in GPIb/IX/V), and (quinine, quinidine, heparin, these disorders, and HPA typing performed by and some antibiotics, e.g., penicillins and serologic assays (phenotyping) may give discrepant cephalosporins). results when compared with results obtained by Many platelet antigens are shared with other blood molecular typing (genotyping). cells,e.g.,ABO and HLA class I antigens,but some of the glycoprotein antigens are expressed predominantly on Inheritance and nomenclature platelets. These antigens are commonly referred to as Most of the platelet-specific alloantigen systems platelet-specific alloantigens or human platelet reported to date have been shown to be biallelic, with alloantigens (HPAs), although some of these are also each allele being codominant. Historically, systems present to a lesser extent on other blood cells, e.g., were named by the authors first reporting the system, HPA-5 on activated T . usually using an abbreviation of the name of the patient in whom the antibody was detected. Some systems Human Platelet Alloantigens were published simultaneously by different labora- There are a number of well-characterized biallelic tories and with different names, e.g., Zw and PlA,or platelet alloantigen systems, and a number of rare, Zav/Br/Hc, and only later were they found to be the private, or low-frequency antigens have also been same polymorphism. In 1990, a working party for described (Table 1). Most of these antigens were first platelet immunology of the ISBT agreed on a new discovered during the investigation of cases of neonatal nomenclature for platelet polymorphisms, the Human alloimmune thrombocytopenia (NAIT). Platelet Antigen (HPA) nomenclature. Any new platelet Platelet-specific alloantigens are located on platelet GP alloantigens are now accepted and named according membrane GPs involved in hemostasis through to guidelines established by the recently founded interactions with extracellular matrix proteins in the International Platelet Nomenclature Committee.1 vascular endothelium and plasma coagulation proteins. In the HPA nomenclature,each system is numbered The majority of these antigens are on the GPIIb/IIIa consecutively (HPA-1, -2, -3, and so on) according to its complex, which plays a central role in platelet date of discovery. The high-frequency allele in each aggregation as a receptor for fibrinogen, fibronectin, system is designated “a” and the low-frequency allele vitronectin, and von Willebrand factor. Other “b.” Newly discovered systems are only officially important GPs are GPIb/IX/V,the main receptor for von included when confirmed by a second party and Willebrand factor involved in platelet adhesion to approved by the nomenclature committee. If an damaged vascular endothelium; GPIa/IIa, which is antibody against only one allele has been reported, a involved in adhesion to collagen; and CD109, which “w” (for workshop) is added after the antigen name, also appears to be a collagen receptor. Congenital e.g., HPA-10bw. One possible reason why antibodies

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 89 A. NORTON ET AL.

Table 1. Platelet-specific alloantigen systems

Platelet Alternative Phenotype membrane Nucleotide Amino acid System Antigen names frequency* (%) glycoprotein change change

HPA-1 HPA-1a Zwa,PlA1 97.9 GPIIIa T196 Leucine33 HPA-1b Zwb,PlA2 28.8 C196 Proline33

HPA-2 HPA-2a Kob > 99.9 GPIbα C524 Threonine145 HPA-2b Koa, Siba 13.2 T524 Methionine145

HPA-3 HPA-3a Baka, Leka 80.95 GPIIb T2622 Isoleucine843 HPA-3b Bakb 69.8 G2622 Serine843

HPA-4 HPA-4a Yukb,Pena > 99.9 GPIIIa G526 Arginine143 HPA-4b Yuka,Penb < 0.1 A526 Glutamine143

HPA-5 HPA-5a Brb,Zavb 99.0 GPIa G1648 Glutamic acid505 HPA-5b Bra,Zava,Hca 19.7 A1648 Lysine505

HPA-6 GPIIIa A1564 Arginine489 HPA-6bw Caa,Tua 0.7 G1564 Glutamine489

HPA-7 GPIIIa C1267 Proline407 HPA-7bw Mo 0.2 G1267 Alanine407

HPA-8 GPIIIa T2004 Arginine636 HPA-8bw Sra < 0.01 C2004 Cysteine636

HPA-9 GPIIb G2603 Valine837 HPA-9bw Maxa 0.6 A2603 Methionine837

HPA-10 GPIIIa G281 Arginine62 HPA-10bw Laa < 1.6 A281 Glutamine62

HPA-11 GPIIIa G1996 Arginine633 HPA-11bw Groa < 0.25 A1996 Histidine633

HPA-12 GPIbβ G141 Glycine15 HPA-12bw Iya 0.4 A141 Glutamic acid15

HPA-13 GPIa C2531 Threonine799 HPA-13bw Sita 0.25 T2531 Methionine799

HPA-14 HPA-14bw Oea < 0.17 GPIIIa ∆ AAG1929–1931 ∆ Lysine611

HPA-15 HPA-15a Govb 74 CD109 C2108 Serine703 HPA-15b Gova 81 A2108 Tyrosine703

HPA-16 C517 Threonine140 HPA-16bw Duva < 1 GPIIIa T517 Isoleucine140

Va a < 0.4 GPIIb/IIIa

PlT > 99.9 GPV

Vis GPIV

Pea GPIbα

Dya 38 kD GP

Moua 26 unknown

* Frequencies based on studies in Caucasians

90 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: platelet antigens and antibodies against the “a” antigen have not yet been reported for typing. One such assay is the PCR using sequence- many of the recently discovered systems is that the “b” specific primers (PCR-SSP). This is a fast and reliable allele is of such low frequency that “bb” homozygous molecular typing technique with minimal post-PCR individuals either do not exist or are extremely rare. handling. It has become one of the cornerstone In Caucasian populations, the allele frequency for techniques in HLA typing and is widely used for HPA the majority of HPA systems is skewed toward the “a” genotyping (Fig. 2).3,4 Novel, high-throughput, DNA- allele and homozygosity for the “b” allele is below 3 based typing techniques with automated readout are percent. This places significant pressures on the blood under development and will be used in platelet services in the management of alloimmunized patients, immunology reference laboratories in the near future. as compatible RBCs and platelets are difficult to obtain Knowledge of the genetic basis of platelet-specific for patients with antibodies against the “a”alloantigen. antigens makes it possible to carry out molecular Allele frequencies vary between populations, e.g., HPA- genotyping on whatever DNA-containing material is 1b is extremely rare or absent in the Far East, while available, e.g., platelet typing using fetal DNA from HPA-4b does not occur in Caucasians (Table 2). These amniocytes or from chorion villous biopsy samples. differences are important when investigating cases of However, in the setting of first trimester fetal HPA suspected platelet alloimmunity in different ethnic typing, extreme caution is required to exclude possible groups. contamination with maternal cells and consequent Until the early 1990s, platelet typing was erroneous typing. performed by serologic assays. These assays required the use of monospecific antisera, which were relatively Platelet Antibodies uncommon, as the majority of immunized individuals Platelet antigens can be targeted by different types produced HLA Class-I antibodies in addition to the of antibodies: autoantibodies, alloantibodies, isoanti- platelet-specific antibodies. Therefore, typing that bodies, and drug-dependent antibodies. could be performed was limited,and many laboratories were only able to phenotype for HPA-1a. The Platelet Autoantibodies publication of more advanced assays, such as Platelet autoantibodies cause the persistent -specific immobilization of thrombocytopenia (peripheral platelet count < 150 × platelet antigens (MAIPA) (Fig. 1), permitted more 109/L) seen in idiopathic thrombocytopenic purpura extensive phenotyping, but some antisera were simply (ITP). Autoantibodies bind to platelet antigens and not available.2 cause the premature destruction of platelets in the With the advent of techniques such as immuno- reticuloendothelial system.5 The main targets appear to precipitation of radioactive labeled platelet-membrane be the platelet membrane GP GPIIb/IIIa and GPIb/IX/V. proteins, and the PCR, the molecular basis of the Such autoantibodies bind to the platelets of all majority of clinically relevant platelet-specific allo- individuals, regardless of their HPA types. antigen systems was elucidated. The molecular basis for ITP may be seen both in children and in adults, all of the HPA alloantigen systems has been determined, where it can follow an acute or chronic course. In and in all but one (Oea) the difference between the two adults, the condition typically has an insidious onset alleles is based on a single nucleotide difference that and runs a chronic course. Symptoms and signs are results in a single amino acid substitution. Based on this variable and include bruising, mucocutaneous molecular knowledge, a plethora of molecular typing bleeding, and frank hemorrhage. In general, serious techniques has been developed over the last decade, bleeding symptoms are rare unless the ITP is severe and these have largely overcome problems in platelet (platelet count < 30 × 109/L).6 Adult chronic ITP has an

Table 2. Platelet antigen frequencies in different populations HPA- 1a 1b 2a 2b 3a 3b 4a 4b 5a 5b Naka Caucasians 97.5 30.8 99.8 11.8 86.1 62.9 100 0 98.8 20.7 100 African blacks 100 16 97 33 85 60 100 0 96 38 97.6 Japanese 99.9 3.7 ? 25.4 78.9 70.7 99.9 1.7 99.8 8.7 97 Chinese 99.9 0.15 ? 9 78.5 71.9 99.9 0.17 99.9 17.7 85.7

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 91 A. NORTON ET AL.

Fig. 1. Monoclonal antibody-specific immobilization of platelet antigens. (1) A cocktail of target platelets;murine monoclonal antibody (MoMab) directed against the glycoprotein being studied, e.g., GPIIb/IIIa; and human serum is prepared in (a) the test serum containing anti-HPA-1a and (b) no anti- platelet antibodies. (2) After incubation, a trimeric (a) or dimeric (b) complex is formed. Excess serum and MoMab is removed by washing. (3) The platelet membrane is solubilized in a non-ionic detergent,which releases the complexes into the fluid phase,and particulate matter is removed by centrifugation. (4) The lysates containing the glycoprotein/antibody complexes are added to the wells of a microtiter plate previously coated with goat anti-mouse antibody. (5) Unbound lysate is removed by washing and an enzyme-conjugated goat anti-human antibody added. (6) Excess conjugate is removed by washing and a substrate solution is added. Cleavage of the substrate, i.e., a color reaction indicates binding of human antibody to the target platelets.

92 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: platelet antigens and antibodies

HPA-1 HPA-2 HPA-3 HPA-4 HPA-5 HPA-15 Testing for platelet autoantibodies may be of value abababababab when there is a combination of bone marrow failure Control Donor 1 band associated with immune-mediated thrombocytopenia, Allele- in ITP patients refractory to first and second line specific bands treatments, in drug-dependent immune thrombocy- topenias, and in rare disorders such as acquired antibody-mediated thrombasthenia.13 Donor 2 Secondary immune thrombocytopenias can occur in association with underlying autoimmune disorders, e.g., systemic lupus erythematosus. Platelet is also frequently associated with B-cell malignancies, and formation is not infrequent in the Donor 3 posthematopoietic stem cell transplantation period, during immune cell re-engraftment. In these situations, the presence of autoantibodies might contribute to Fig. 2. Simultaneous genotyping for HPA-1, -2, -3, -4, -5 and -15 by PCR- SSP. The upper band present in all lanes is the 429-bp product of refractoriness to donor platelets. human growth hormone. The lower bands are the products of sequence-specific primers. Results are read from left to right,i.e., lane 1 is HPA-1a, lane 2 is HPA-1b, etc. The genotypes of the 3 Platelet Alloantibodies donors shown are: Donor 1: HPA-1b1b; 2a2b; 3b3b; 4a4a; 5a5a; 15a15a Detection of platelet alloantibodies Donor 2: HPA-1b1b; 2a2a; 3a3a; 4a4a; 5a5a; 15a15b Tests for the detection of platelet-specific antibodies Donor 3: HPA-1b1b; 2a2a; 3a3a; 4a4a; 5a5b; 15a15a have evolved over the last 4 decades. Currently the most Courtesy of Dr Paul Metcalfe. widely used assay is the indirect PIFT,but it is unable to distinguish between platelet-specific and HLA Class I incidence of 58 to 66 new cases per million population antibodies. The MAIPA assay has become the gold 7 per year in the United States and affects predominantly standard for the identification of platelet-specific 8 women of childbearing age. In children,the incidence antibodies. This capture assay uses murine monoclonal of ITP appears to be lower than in adults, with antibodies specific for platelet GPs and can analyze published figures of between 4.0 and 5.3 per complex mixtures of antibodies in patient sera. Third 9,10 100,000. The disorder also tends to run a different generation antibody detection assays, using purified or clinical course, with an acute and abrupt onset, often recombinant platelet GP, have also been developed; following a viral illness or immunization. The majority however,the sensitivity of these assays is not satisfactory of children with the disorder require no treatment, and for all alloantigen systems and some systems (e.g., HPA- it resolves within 6 months. However, 15 percent of 15) are not included.14 affected children develop a chronic form of ITP similar to that seen in adults. Clinical significance of HPA alloantibodies The diagnosis of ITP usually relies on a typical HPA alloantibodies are responsible for the history,blood count,and blood film. Although there are following clinical conditions: assays available for the detection of platelet auto- • Neonatal alloimmune thrombocytopenia (NAIT), antibodies,these are not robust enough to establish the sometimes called fetomaternal alloimmune diagnosis of ITP alone. The direct platelet immuno- thrombocytopenia (FMAIT) fluorescence test (PIFT) detects platelet-associated •Posttransfusion purpura (PTP) immunoglobulin (PAIg) levels, which are found to be • Refractoriness to platelet transfusions increased in most patients with ITP. Unfortunately, the assay is not specific, since positive results can also be seen in nonimmune thrombocytopenias, for example NAIT secondary to septicemia.11 Assays for antibodies to Definition specific platelet membrane GPs IIb/IIIa and Ib/IX are NAIT is defined as thrombocytopenia in the specific (90%) in ITP but less sensitive (50–65%), and neonate (platelet count < 150 × 109/L) due to trans- currently their routine use in the diagnosis of ITP is not placental transfer of maternal platelet alloantibodies. considered to be justified.12,13 However, thrombocytopenia can also occur in utero,

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 93 A. NORTON ET AL. resulting in intrauterine death or intracranial Clinical features hemorrhage (ICH). The typical infant with NAIT is full term and otherwise healthy, with a normal coagulation screen Incidence and isolated thrombocytopenia. The thrombocyto- The estimated incidence of severe thrombocyto- penia may be severe and unexpected, and it may occur penia due to maternal alloantibodies is 1 per 1200 in the firstborn child. Clinically, the neonate may pregnancies in a Caucasian population, of which the present with petechiae, purpura, and/or ecchymoses.17 majority are due to fetomaternal incompatibility for The most feared complication is ICH, which occurs in HPA-1a.15 NAIT is probably under-recognized in 14 to 26 percent of cases,17,20,21 resulting in death in 7 clinical practice.16 percent of affected cases20,21 and long-term morbidity with neurologic sequelae in 21 percent.20 Eighty Pathophysiology percent of ICHs occur in utero, with 14 percent NAIT is the platelet equivalent of the anemia seen occurring before 20 weeks’ and a further 28 percent 21 in HDN. Maternal IgG alloantibodies against a fetal before 30 weeks’ gestation. This is consistent with platelet alloantigen cross the placenta and bind to fetal the expression of platelet antigens from 16 weeks’ platelets, resulting in their increased destruction in the gestation and placental transfer of IgG antibodies that reticuloendothelial system. Unlike HDN, up to 50 can occur from 14 weeks. There are also less common percent of cases of NAIT occur in the first pregnancy. presentations of NAIT, including fetal hydrocephalus, 22 In Caucasian women, the antibody most commonly hydrops fetalis, or recurrent miscarriages. implicated in NAIT (78–89% of cases) is anti-HPA-1a.17 The differential diagnoses of thrombocytopenia in the newborn include infection, prematurity, Anti-HPA-5b occurs more frequently in pregnancy, but intrauterine growth retardation, maternal platelet it tends to cause mild thrombocytopenia,18 and autoimmunity, and inadequate megakaryocytopoiesis clinically significant NAIT is less common (6–15% of (such as in the thrombocytopenia absent radii [TAR] cases). The HPA-15 system,described a decade ago,has syndrome). recently been found to have clinical significance, with HPA-15 alloantibodies being the third most frequently Laboratory diagnosis encountered antibodies. Rarely,NAIT can occur due to The diagnosis of NAIT requires the demonstration incompatibility for other HPA, HLA, or blood group of maternal platelet alloantibodies that react against ABO antigens. In non-Caucasian populations, other platelet-specific antigens present in the father and HPA alloantigens are more commonly implicated, for infant but not in the mother. Detection of maternal example HPA-4b in Orientals. platelet-specific antibodies is usually carried out by Approximately 2.5 percent of pregnant Caucasian two techniques, the indirect PIFT and the MAIPA assay women are HPA-1a negative and most of these women using a panel of HPA-typed platelets. The serum of the will carry HPA-1a positive fetuses, but only about 12 mother is also tested against paternal platelets by both percent of these women will become HPA-1a tests so that alloantibodies against low-frequency 15 alloimmunized. The ability to develop anti-HPA-1a is alloantigens and private antigens are not missed. The HLA class II restricted, with a strong association seen parents are also genotyped for the HPA-1,-2,-3, -5, and 19 with HLA-DRB3*0101 positivity. As a negative -15 alloantigens (the frequency of other alloantigens predictive factor, the absence of HLA-DRB3*0101 for being comparatively low in the Caucasian population). HPA-1a alloimmunization in HPA-1a negative women is It should also be noted that in up to 30 percent of cases > 90 percent, but its positive predictive value as a of NAIT associated with fetomaternal incompatibility single marker is only 35 percent.15 for the HPA-1a antigen,maternal HPA-1a antibodies may The pathogenicity of antibodies in an unselected not be detectable.20 However, some authors have population is highly variable. In a study including 26 shown that the choice of MoAb used to capture cases of NAIT due to anti-HPA-1a,9 of the 26 (34%) had GPIIb/IIIa in the MAIPA assay is critical and falsely severe thrombocytopenia; 35 percent (7 of 26 cases) negative results may be obtained if the epitope had a normal cord platelet count, with a further 38 recognized by the MoAb is blocked by the human percent (10/26) having a “safe”platelet count of > 50 × antibody.23 Others have reported that an increased 109/L.15 serum:cell ratio is required to detect low levels of

94 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: platelet antigens and antibodies antibody. Sometimes the antibodies that were not hours, and thrombocytopenia may persist for up to 6 detectable at the time of delivery become detectable a weeks postnatally (although 1–2 weeks is more usual). few weeks or months after delivery. All neonates with NAIT should also undergo some form of cerebral imaging to exclude ICH. Postnatal management The postnatal management of NAIT is influenced Antenatal management by the degree of thrombocytopenia and clinical The realization that spontaneous ICH may occur in symptoms present. Any neonate with suspected NAIT utero has led many to a search for methods of who is bleeding or has a platelet count of < 30 × 109/L preventing serious antenatal bleeding. Unfortunately, should be transfused with compatible platelets to 30 percent of cases of NAIT occur in the first minimize the risk of ICH, without waiting for pregnancy, and at present there are no reliable tests confirmatory laboratory tests. available to predict which women will become Anti-HPA-1a and anti-HPA-5b are responsible for alloimmunized and which of those will have severely approximately 75 percent and 20 percent of cases of affected babies. What is well recognized is that NAIT, respectively.17,24 Approximately 2 percent of mothers with HPA-1a alloimmunization and previously blood donors are negative for both these antigens and affected infants (particularly those with ICH) have a can be easily identified using established platelet high risk of recurrence and poor outcome in typing methods.25,26 It is possible to establish in blood subsequent pregnancies.30 These are the women centers a stock of HPA-1a-negative and 5b-negative targeted for antenatal treatment. platelet concentrates from donors who donate In women with a known history of alloimmu- regularly,and these can be issued without delay for the nization and a previously affected pregnancy, HPA treatment of suspected cases of NAIT before laboratory typing of the partner is important. This allows the risk confirmation of the diagnosis.27 An alternative for of the fetus being HPA-1a positive to be assessed (i.e., platelets is random donor platelets, which, while more all infants are affected when the father is homozygous readily available, have been shown to have poorer for the pathogenic platelet antigen as opposed to 50 responses, with one study of 36 cases of NAIT percent of infants affected in heterozygous cases). reporting a median increase in platelet count 24 hours Where the father is heterozygous it may be useful to posttransfusion of only 3.5 × 109/L.21 Washed maternal HPA type the fetus by chorionic villous or amniotic platelets have been found to be successful in the fluid sampling. treatment of NAIT28 and may also be used. However, There is considerable experience in the antenatal maternal platelets need to be collected by apheresis management of FMAIT where there has been a machines, washed to remove the maternal HPA previously affected pregnancy,but which is the optimal antibodies, which may prolong the thrombocytopenia, management remains controversial.31 The therapeutic and gamma-irradiated to prevent transfusion-associated options that have been explored are maternal graft-versus-host disease; all of which is time administration of IVIG and/or steroids and fetal platelet consuming and requires specialist equipment and transfusions. Early cesarean section alone is not facilities. considered to be effective in preventing antenatal or In the stable neonate, an alternative approach is to perinatal hemorrhage. For both approaches to use IVIG at a dose of 1g/kg/day for 2 days. In a study antenatal management, fetal blood sampling (FBS) has of 12 cases of NAIT, a 75 percent response rate was been used for initial assessment of the fetal platelet reported. However, the increase in platelet count was count, usually at 20 to 22 weeks’ gestation, and for the delayed by 24 to 48 hours.29 In urgent situations, monitoring of the effectiveness of treatment in the where antigen-negative platelets are not available and early studies. the neonate is bleeding, a combination of random Bussel et al.30 found maternal administration of platelets and IVIG should be given until compatible IVIG to be successful, with no instances of ICH and donor platelets become available. most, but not all, infants, achieving a platelet count of Monitoring of neonates with NAIT is important in greater than 30 × 109/L at the end of pregnancy. The the postnatal period,irrespective of initial platelet level addition of steroids did not add to the effect of IVIG. and clinical symptoms. The platelet count may However, ICH has been found during maternal continue to fall after birth, particularly in the first 48 treatment with IVIG,32 and a group of European centers

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 95 A. NORTON ET AL. treating 37 pregnancies only found success with the Routine antenatal screening use of maternal IVIG in 7 of 27 cases (26%), and Advances in laboratory diagnosis and antenatal steroids in 1 of 10 cases (10%).33 It is difficult to under- management have drawn attention to the fact that the stand why there was such a difference in the success of first affected fetus/neonate in a family is only maternal administration of IVIG between these two recognized after bleeding has occurred, and this has studies. Relevant factors may include the methods raised the question of whether routine screening for used for assessing the success of treatment and the NAIT would be advantageous. dose, timing, and type of IVIG used. The selection of An important initial question is whether to carry cases may also be important. out screening antenatally or postnatally. The advan- tages of antenatal screening are that alloimmunized A number of studies have shown the value of women can be identified during pregnancy, allowing platelet transfusions given by cordocentesis in raising time for antenatal intervention if it is agreed that this is the platelet count, but the platelet count is raised for appropriate. Even if no antenatal intervention is under- only a few days. A single pre-delivery transfusion may taken, the mode and timing of delivery can be planned protect against bleeding at the time of delivery,but the to ensure minimal trauma to the baby’s head and that fetus remains at risk of spontaneous ICH earlier in compatible platelets are available, if needed. Postnatal pregnancy. Weekly in utero platelet transfusions have screening can be achieved by simply carrying out a been shown to be effective in preventing ICH in severe platelet count on a cord blood sample, but the major cases of FMAIT,34 but this approach is invasive. drawback of this approach is that the key objective of In a recent study from the European Study Group screening, to prevent morbidity and mortality from for NAIT,the outcome of 56 fetuses receiving antenatal ICH, is unlikely to be achieved in the index pregnancy. treatment for NAIT due to HPA-1a alloimmunization Although it is recognized that antenatal hemor- compared favorably with the outcome in previous rhage due to NAIT can produce devastating clinical pregnancies.35 Cases with a sibling history of antenatal effects, significant shortcomings exist in the know- ICH or severe thrombocytopenia (platelet counts of ledge about NAIT necessary for the introduction of an < 20 × 109/L) had significantly lower pretreatment antenatal screening program. Further research is platelet counts than cases whose siblings had less required on a number of issues, including the range of severe thrombocytopenia or postnatal ICH. Maternal clinical outcomes in affected cases,the identification of therapy resulted in a platelet count exceeding 50 × factors useful for predicting severe disease, and the preferred option for antenatal management in women 109/L in 67 percent of cases. None of the fetuses with anti-HPA-1a but no previous history of affected managed by serial platelet intrauterine transfusions pregnancies.38 (IUTs) suffered ICH after treatment started. However, the most serious complications encountered by the study cases were associated with fetal blood sampling Posttransfusion Purpura (FBS). The results of this study support the use of A clinical case of thrombocytopenia developing 7 maternal therapy as first line treatment for the antenatal days after elective surgery, then spontaneously management of NAIT. The observations of this study resolving 3 weeks later, was first described in 1959 by van Loghem and colleagues.39 The 51-year-old woman suggest that the commencement of maternal therapy involved was found to have a strong platelet can be stratified on the basis of the sibling history of alloantibody,subsequently described as the first human NAIT. In two recent studies, concern regarding the platelet alloantibody,Zw. Two years later, a similar case safety of FBS led to the use of a less invasive treatment was described by Shulman and colleagues, who coined strategy involving maternal administration of IVIG the term posttransfusion purpura (PTP).40 In each case without FBS for monitoring of the fetal platelet count, the same human platelet alloantibody was implicated 36,37 without an increased incidence of ICH. (named Zw in the first case and anti-PlA1 in the case by Antenatal treatment appears to have the potential Shulman, now known as HPA-1a). to improve the outcome of severely affected cases of NAIT, but there is little information on the long-term Definition development of children who have been treated in PTP describes an acute episode of severe throm- utero. bocytopenia occurring 5 to 12 days after a blood

96 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: platelet antigens and antibodies transfusion. It is usually seen in HPA-1a negative antigen-negative individuals. The majority of cases of women previously alloimmunized by pregnancy or PTP (80 to 90%) are associated with the development transfusion. The implicated transfusion is thought to of HPA-1a antibodies, but other HPA antibodies have induce a secondary immune response, boosting the also been implicated, including HPA-1b, HPA-3a, HPA- production of HPA-1a antibodies and destruction of 3b, HPA-4a, HPA-5a, HPA-5b, HPA-15b, and Naka, and transfused donor platelets. At the same time, the occasionally multiple antibodies are present. patient’s own platelets are also destroyed. Platelet-specific antibodies are detected using the MAIPA assay, which is able to resolve mixtures of Incidence antibodies, including HLA antibodies, that are often PTP is considered to be a rare complication of present in patients with PTP but do not appear to have transfusion,although the true incidence is unknown. A a pathological role. UK-based voluntary and confidential scheme for Other causes of a rapid onset of severe thrombo- reporting serious hazards of transfusion (SHOT), set up cytopenia should also be excluded, such as dis- in 1996, reported 43 cases in the first 6 years of the seminated intravascular coagulation (the coagulation scheme, during which approximately 20 million blood screen is normal in uncomplicated cases of PTP), components were transfused, giving an approximate autoimmune thrombocytopenia, and drug-induced incidence of 1 case in 465,000 transfusions. However, thrombocytopenia, e.g., heparin-induced thrombo- since the introduction of universal leukoreduction of cytopenia (HIT). Rarely, thrombocytopenia may occur blood components in 1999, there has been a reduction within 48 hours of a transfusion secondary to passively in the annual number of cases reported to SHOT.41 transfused platelet-specific alloantibodies from an As in NAIT, the susceptibility of HPA-1a negative immunized blood donor. individuals to PTP appears to reflect their ability to make HPA-1a antibodies, which is strongly associated Pathophysiology with HLA Class II –DRB3*0101. A rapid secondary antibody response (usually against HPA-1a) is stimulated by the RBC transfusion, Clinical features with the result of acute thrombocytopenia about 1 The typical patient is a middle-aged or elderly week later. This time course of events is well woman who has had a previous exposure to platelet established. However, it is still unclear why the antigens through pregnancy, transfusion, or both. The patient’s own antigen-negative platelets are destroyed. occurrence of PTP has also been reported in a small A number of hypotheses have been suggested: number of male patients. The time interval between •Transfused HPA-1a-positive platelets release HPA- the initial sensitizing event and the subsequent 1a antigen, which is adsorbed onto the patient’s transfusion stimulating PTP is variable, with previous HPA-1a-negative platelets, making them a target reports ranging from 3 to 52 years. for anti-HPA-1a. Support for this hypothesis Clinically, the patient presents 5 to 12 days after comes from observations such as the elution of transfusion with an acute severe thrombocytopenia anti-HPA-1a from HPA-1a-negative platelets in (platelet count < 10 × 109/L) that has fallen from some cases of PTP and the demonstration of the normal within 12 to 24 hours. Associated hemorrhage adsorption of HPA-1a antigen onto HPA-1a- is very common and sometimes severe, with wide- negative platelets after incubation with plasma spread purpura,and bleeding from mucous membranes from HPA-1a-positive stored blood. and the gastrointestinal and urinary tracts. If untreated, • The released HPA-1a antigen forms immune the condition usually lasts between 7 and 28 days, complexes with anti-HPA-1a in the plasma and although occasionally it may persist longer. the immune complexes become bound to the Implicated blood products include whole blood, patient’s platelets, causing their destruction. packed RBCs,and RBC concentrates. PTP has also been • The transfusion stimulates the production of reported following transfusion of plasma. platelet autoantibodies as well as anti-HPA-1a. Evidence in favor of this mechanism is the Laboratory investigations detection of positive reactions of some PTP A clinical diagnosis of PTP needs to be confirmed patients’ sera from the acute thrombocytopenic by the finding of platelet-specific alloantibodies in phase with autologous platelets and the isolation

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 97 A. NORTON ET AL.

of autoantibodies that recognize calcium- penia. They may be given therapeutically to patients dependent epitopes on GPIIb/IIIa during the with active bleeding or prophylactically to patients acute phase.42 with thrombocytopenia secondary to bone marrow • In the early phase of the secondary antibody failure. Serious spontaneous hemorrhage is unlikely to response, anti-HPA-1a may be produced, which occur at platelet counts above 10 × 109 /L,44 and this has the ability to crossreact with autologous as level has been widely adopted as a threshold above well as allogeneic platelets. which prophylactic platelet transfusions are not required in an otherwise stable, nonbleeding patient.45 Management Although prophylactic platelet transfusions are The main aim of treatment is to prevent morbidity standard practice for patients with bone marrow and mortality associated with severe thrombocyto- failure, no recent randomized controlled trials have penia by shortening the duration of thrombocytopenia. compared therapeutic versus prophylactic transfusions A particular risk is ICH, which may cause early in terms of incidence of hemorrhage and associated fatalities. Prompt treatment is therefore essential, to morbidity. Indeed, concerns about platelet refracto- prevent this. riness secondary to alloimmunization have led some to There have been no randomized controlled trials suggest lowering the platelet transfusion threshold to of treatment for PTP, and comparison of various 5 × 109 /L.45 therapeutic measures is complicated because of the occurrence of spontaneous remissions. Currently the Definition treatment of choice is high-dose intravenous immuno- Platelet refractoriness is defined as the repeated globulin (IVIG) (2g/kg given over 2 or 5 days), with failure to achieve satisfactory responses to platelet rapid responses seen in about 85 percent of cases.43 transfusions. This can be assessed clinically in a Steroids and plasma exchange were the preferred bleeding patient. However, where platelet transfusions treatments before the availability of IVIG, and plasma have been given prophylactically, response is assessed exchange, in particular, appeared to be effective in by measuring the posttransfusion platelet count some, but not all, cases. increment. Platelet transfusions are usually ineffective in raising Various formulae have been devised to assess the the platelet count, but they may be needed in large platelet increment, including the percent platelet doses to control severe bleeding in the acute phase, recovery and CCI. In practice, a 24-hour increment of particularly in patients who have recently undergone < 5 × 109/L on two or more occasions is a good surgery before their response to high-dose IVIG. There indicator of refractoriness to random donor platelets. is no evidence that platelet concentrates from HPA-1a- negative platelets are more effective than those from Causes of platelet refractoriness random donors; the dose probably is more important. Refractoriness is due to the shortened survival of the transfused platelets in the recipient’s circulation, Prevention historically described in 20 to 60 percent of patients PTP may recur, although this is unpredictable and receiving multiple transfusions.46 The causes may be occurs many years later. Patients with a previous classified as immune or nonimmune, but in any episode should be issued a card to indicate that they individual patient there may be a multifactorial need special blood products, and ideally future blood etiology. or platelet transfusions should be either autologous or Nonimmune clinical factors include disseminated from HPA-compatible donors. intravascular coagulation (DIC), splenomegaly, and The reduction in cases of PTP reported to the intravenous antibiotics (especially antifungal drugs SHOT scheme since the introduction of universal such as amphotericin B).47,48 Fever has also been leukoreduction in the UK suggests that leukoreduced implicated in causing poor responses to platelet blood products may be safe. transfusions, although whether this is a reflection of sepsis, associated DIC, or antibiotic therapy rather than Refractoriness to Platelet Transfusions the temperature itself is unclear.49 Another unknown Platelet transfusions are effective in decreasing factor is whether platelet refractoriness may be due in hemorrhagic complications of severe thrombocyto- part to inadequate dosage of platelets. Further studies

98 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: platelet antigens and antibodies are required to assess the optimal dosage of platelets, refractoriness, seen usually when there are high-titer which may need to be adjusted according to a patient’s ABO antibodies in the recipient. ABO-identical platelet blood volume. concentrates should be given to exclude this The most common cause of alloimmune platelet possibility. refractoriness is HLA alloimmunization. Other immune causes include HPA alloimmunization, high titer ABO Prevention antibodies in the recipient, platelet autoantibodies, and Leukoreduction of blood components has been drug-related platelet antibodies. shown to reduce the incidence of HLA alloimmu- HLA antibodies predominantly occur in women nization and platelet refractoriness.50 with a history of pregnancies and/or a history of multiple transfusions. If they are suspected as a cause Drug-Dependent Antibodies for refractoriness, a combination of screening tests for Drug-dependent antibodies are an important cause both cytotoxic and noncytotoxic HLA antibodies of shortened platelet survival that can often be should include the lymphocytotoxicity test (LCT) with overlooked, especially in hemato-oncology patients either the lymphocyte or PIFT, or an ELISA-based who may have thrombocytopenia for a variety of method. reasons. In some cases, drugs too small to elicit an The role of platelet-specific antigens in platelet immune response by themselves may bind as a refractoriness is unclear. HPA antibodies occur at a to platelet GPs in vivo. This haptenized platelet GP can frequency of 8 percent50 to 20–25 percent51 in various trigger the formation of antibodies that only bind to the studies and are usually found in combination with HLA GP in the presence of the hapten. A classic example is antibodies,52 although rarely they may occur in quinine and its optical stereo-isomer, quinidine. isolation. Most commonly, HPA alloimmunization is Typically, quinine-dependent antibodies are against directed toward antigens with phenotypic frequencies GPIIb/IIIa, GPIb/IX/V, or a combination of both. below 30 percent. Some studies have suggested that Vancomycin has also been shown to induce the there is no clear correlation between HPA antibodies formation of IgG antibodies that bind specifically to GP and poor responses to platelet transfusions,53,54 but IIb, IIIa, or both in the presence of the antibiotic.56 others have found that matching for platelet-specific Similarly,the interaction of heparin with platelet factor antigens in those refractory to HLA-matched platelets 4 can cause antibody formation and lead to a drug- may be beneficial.55 dependent thrombocytopenia.57,58 Evidence is also emerging implicating teicoplanin as a cause of Management immune-mediated thrombocytopenia.59 The management of platelet refractoriness first requires an assessment of possible nonimmune clinical causes. These should be corrected if possible and Acknowledgments prophylactic platelet transfusions from random donors We are indebted to Dr Paul Metcalfe, Division of continued in the usual way. If poor responses to Haematology,National Institute for Biological Standards platelet transfusions persist, HLA antibodies should be and Control, Potters Bar Herts, UK, for the photograph sought in the patient’s serum and, if they are present, of results of simultaneous HPA-1 to -5 and -15 platelet transfusions matched for the HLA-A and -B genotyping by PCR-SSP (Fig.2). The schematic diagram antigens of the patient should be used. for the MAIPA assay is reproduced with permission If poor responses continue, consideration should from Allen DL, Lucas GF, Ouwehand WH, Murphy MF. be given to ABO and/or HPA incompatibility. Testing Platelet and antigens. In: Practical for platelet-specific antibodies may identify HPA transfusion medicine. 2nd ed. Murphy MF, Pamphilon antibodies, and future platelet transfusions lacking the DH, eds. Blackwell Science Ltd, Oxford, UK (in press relevant antigen may be indicated. Platelet 2004). crossmatching may be helpful in cases where the platelet-specific antibodies have no obvious HPA specificity and the only way of identifying compatible donors is to find those who are crossmatch negative.45 ABO incompatibility is an unusual cause of platelet

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 99 A. NORTON ET AL.

References the detection of antibodies against human platelet 1. Metcalfe P, Watkins NA, Ouwehand WH et al. antigens. Transfus Med 1999;9:383-6. Nomenclature of human platelet antigens. Vox 15. Williamson LM, Hackett G, Rennie J, et al. The Sang 2003;85:240-5. natural history of fetomaternal alloimmunization to 2. Kiefel V,Santoso S,Weisheit M, Mueller-Eckhardt C. the platelet-specific antigen HPA-1a (P1A1,Zwa) as Monoclonal antibody-specific immobilization of determined by antenatal screening. Blood platelet antigens (MAIPA): a new tool for the 1998;92(7):2280-7. identification of platelet-reactive antibodies. Blood 16. Murphy MF,Verjee S,Greaves M.Inadequacies in the 1987;70:1722-6. postnatal management of fetomaternal alloimmune 3. Metcalfe P, Waters AH. HPA-1 typing by PCR thrombocytopenia. Br J Haematol 1999;105:123-6. amplification with sequence-specific primers (PCR- 17. Mueller-Eckhardt C, Kiefel V, Grubert A, et al. 348 SSP): a rapid and simple technique. Br J Haematol cases of suspected neonatal alloimmune neonatal 1993;85:227-9. thrombocytopenia. Lancet 1989;1:363-6. 4. Jones DC, Bunce M, Fuggle SV,Young NT, Marshall 18. Kaplan C, Morel-Kopp MC, Kroll H, et al. HPA-5b SE. Human platelet alloantigens (HPAs): PCR-SSP (Bra) neonatal alloimmune thrombocytopenia: genotyping of a UK population for 15 HPA alleles. clinical and immunological analysis of 39 cases. Br Eur J Immunogenet 2003;30:415-9. J Haematol 1991;78:425-9. 5. Woods Jr VL, Kurata Y, Montgomery RR, et al. 19. Décary F, L’Abbé D, Tremblay L, Chartrand P. The Autoantibodies against platelet glycoprotein Ib in immune response to the HPA-1a antigen: associa- patients with chronic immune thrombocytopenic tion with HLA-DRw52a.Transfus Med 1991;1:55. purpura. Blood 1984a;64:156-60. 20. Kaplan C,Daffos F,Forestier F,Morel MC,Chesnel N, 6. George JN, Raskob GE. Idiopathic thrombocy- Tchernia G. Current trends in neonatal alloimmune topenic purpura: a concise summary of the thrombocytopenia: diagnosis and therapy. In: pathophysiology and diagnosis in children and Kaplan-Gouet C,Schlegel N,Salmon Ch,Mcgregor J, adults. Semin Hematol 1998;35:5-8. eds. Platelet immunology: fundamental and 7. McMillan R. Therapy for adults with refractory clinical aspects. Paris: Colloque INSERM/John chronic immune thrombocytopenic purpura. Ann Libbey Eurotext, 1991;267-78. Intern Med 1997;126:307-14. 21. Spencer JA, Burrows RF.Feto-maternal alloimmune 8. Waters AH. Autoimmune thrombocytopenia: thrombocytopenia: a literature review and clinical aspects. Semin Hematol 1992;2:18-25. statistical analysis. Aust N Z J Obstet Gynaecol 9. Lilleyman JS. Management of childhood idiopathic 2001;41:45-55. thrombocytopenic purpura. Br J Haematol 22. Murphy MF, Hambley H, Nicolaides K, Waters AH. 1999;105:871-5. Severe fetomaternal alloimmune thrombocy- 10. Zeller B, Helgestad J, Hellebostad M, et al. Immune topenia presenting with fetal hydrocephalus. thrombocytopenic purpura in childhood in Prenat Diagn 1996;16:1152-5. Norway: a prospective, population-based regis- 23. Morel-Kopp MC, Daviet L, McGregor J, Kaplan C. tration. Pediatr Hematol Oncol 2000;17:551-8. Drawbacks of the MAIPA technique in 11. Kelton JG.The serological investigation of patients characterising human antiplatelet antibodies. with autoimmune thrombocytopenia. Thromb Blood Coagul Fibrinolysis, 1996;7:144-6. Haemost 1995;74:228-33. 24. Kroll H, Kiefel V, Santoso S. Clinical aspects and 12. Warner MN, Moore JC, Warkentin TE, Santos AV, typing of platelet alloantigens. Vox Sang Kelton JG. A prospective study of protein-specific 1998;74(suppl 2):345-54. assays used to investigate idiopathic thrombocy- 25. Bessos H,Mirza S,McGill A,Williamson LM,Hadfield topenic purpura. Br J Haematol 1999;104:442-7. R, Murphy WG. A whole blood assay for platelet 13. Provan D, Newland A, Norfolk D, et al. Guidelines HPA1 (PlA1) phenotyping applicable to large-scale for the investigation and management of idiopathic screening. Br J Haematol 1996;92:221-5. thrombocytopenic purpura in adults, children and 26. Garner SF, Smethurst PA, Merieux Y, et al. A rapid in pregnancy. Br J Haematol 2003;120:574-96. one-stage whole-blood HPA-1a phenotyping assay 14. Allen DL, Murphy MF. Evaluation of an enzyme- using a recombinant monoclonal IgG1 anti-HPA-1a. linked immunosorbent assay kit (GTI PakPlus) for Br J Haematol 2000;108:440-7.

100 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: platelet antigens and antibodies

27. Ranasinghe E,Walton JD, Hurd CM, et al. Provision 39. van Loghem JJ, Dorfmeijer H, van der Hart M, of platelet support for fetuses and neonates Schreuder F.Serological and genetical studies on a affected by severe fetomaternal alloimmune platelet antigen (Zw).Vox Sang 1959;4:161–9. thrombocytopenia. Br J Haematol 2001;113:40-2. 40. Shulman NR, Aster RH, Leitner A, Hiller MC. 28. Adner MM, Fisch GR, Starobin SG, Aster RH. Use of Immunoreactions involving platelets. V. Post- compatible platelet transfusions in treatment of transfusion purpura due to a complement-fixing congenital isoimmune thrombocytopenic purpura. antibody against a genetically controlled platelet New Engl J Med 1969;280:244-7. antigen. A proposed mechanism for thrombocy- 29. Mueller-Eckhardt C, Kiefel V, Grubert A. High dose topenia and its relevance in ‘autoimmunity’. J Clin IgG treatment for neonatal alloimmune thrombo- Invest 1961;40:1597–620. cytopenia. Blut 1989;59:145-6. 41. Serious Hazards of Transfusion. Annual report 30. Bussel JB,Zabusky MR,Berkowitz RL,McFarland JG. 2001–2002. Serious hazards of transfusion scheme, Fetal alloimmune thrombocytopenia. New Engl J Manchester, UK, 2003. (www.shotuk.org). Med 1997;337:22-6. 42. Watkins NA, Smethurst PA, Allen D, Smith GA, 31. Engelfriet CP, Reesink HW, Kroll H, et al. Ouwehand WH. Platelet alpha II beta 3 International forum: Prenatal management of recombinant autoantibodies from the B-cell alloimmune thrombocytopenia.Vox Sang 2003;84: repertoire of a post-transfusion purpura patient. Br 142-9. J Haematol 2002;116:677-85. 32. Kroll H, Muntean W, Kiefel V, et al. Anti-Koa als 43. Becker T, Panzer S, Maas D, et al. High-dose ursache der neonatalen alloimmunthrombo- intravenous immunoglobulin for post-transfusion purpura. Br J Haematol 1985;61:149-55. zytopenie. Beitr Infusionsther Transfusionsmed 44. Slichter SJ. Controversies in platelet transfusion 1994;32:244-6. therapy. Annu Rev Med 1980;31:509-40. 33. Kaplan C, Murphy MF, Kroll H, Waters AH. 45. British Committee for Standards in Haematology Fetomaternal alloimmune thrombocytopenia: (2003). Guidelines for the use of platelet antenatal therapy with IVIgG and steroids—more transfusions. Br J Haematol 2003;122:10-23. questions than answers. Br J Haematol 1998;100: 46. Slichter SJ. Mechanisms and management of 62-5. platelet refractoriness. In:Nance SJ,ed. Transfusion 34. Murphy MF,Waters AH, Doughty A, et al. Antenatal medicine in the 1990’s. Arlington, VA: American management of fetomaternal alloimmune thrombo- Association of Blood Banks, 1990:95-179. cytopenia—report of 15 affected pregnancies. 47. Bishop JF,McGrath K,Wolf MM,et al.Clinical factors Transfus Med 1994;4:281-92. influencing the efficacy of pooled platelet 35. Birchall J, Murphy MF,Kaplan C, Kroll H. European transfusions. Blood 1998;71:383-7. collaborative study of the antenatal management of 48. Doughty HA, Murphy MF,Metcalfe P,Rohatiner AZS, feto-maternal alloimmune thrombocytopenia. Br J Lister TA, Waters AH. Relative importance of Haematol 2003;122:275-88. immune and non-immune causes of platelet 36. Mackenzie F, Brennand J, Peterkin M, Cameron A. refractoriness.Vox Sang 1994;66:200-5. Management of fetal alloimmune thrombocyto- 49. Delaflor-Weiss E, Mintz PD. The evaluation and penia—a less invasive option? J Obstet Gynaecol management of platelet refractoriness and 1999;19:119-21. alloimmunisation. Transfus Med Rev 2000;14: 37. Radder CM, Brand A, Kanhai HHH. A less invasive 180-96. treatment strategy to prevent intracranial 50. The Trial to Reduce Alloimmunisation to Platelets hemorrhage in fetal and neonatal alloimmune Study Group. Leucocyte reduction and ultraviolet B thrombocytopenia. Am J Obstet Gynecol irradiation of platelets to prevent alloimmunisation 2001;185:683-8. and refractoriness to platelet transfusions.New Eng 38. Murphy MF,Williamson LM, Urbaniak SJ. Antenatal J Med 1997;337:1861-9. screening for fetomaternal alloimmune thrombocy- 51. Murphy MF,Waters AH. Immunological aspects of topenia: should we be doing it? Vox Sang platelet transfusions. Br J Haematol 1985;60: 2002;83(Suppl.1):409-16. 409-14.

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 101 A. NORTON ET AL.

52. Kurz M, Greinix H, Hocker P, et al. Specificities of 57. Warentin TE. Heparin-induced thrombocytopenia: anti-platelets antibodies in multitransfused patients pathogenesis and management. Br J Haematol with hemato-oncological disorders. Br J Haematol 2003;121:535-55. 1996;95:564-9. 58. Warkentin TE. Heparin-induced thrombocytopenia 53. Godeau B, Fromont P,Seror T,Duedari N, Bierling P. and thrombosis. Hematology 2003. American Platelet alloimmunisation after multiple Society of Hematology Education Program Book, transfusion: A prospective study of 50 patients. Br J 2003;503-9. Haematol 1992;81:395-400. 59. Campbell K, Garner SF, Hurd C, et al. Teicoplanin- 54. Meenaghan M, Judson PA, Yousaf K, Lewis L, dependent platelet antibodies as a cause of severe Pamphilon DH. Antibodies to platelet glycoprotein thrombocytopenia (abstract). J Thromb Haemost V in polytransfused patients with hematological 2003;1(suppl 1)OC024. disease.Vox Sang 1993;64:167-70. 55. Kekomaki S, Volin L, Koistinen P, et al. Successful Alice Norton, BSc, MBChB, MRCPCH, Department of treatment of platelet transfusion refractoriness:The Hematology; David L. Allen, FIBMS, National Blood use of platelet transfusions matched for both Service, and Michael F. Murphy, MD, FRCP, FRCPath, human leukocyte antigens (HLA) and human Department of Hematology and National Blood platelet alloantigens (HPA) in alloimmunised Service, Oxford, UK, OX3 9BQ. patients with leukaemia. Eur J Haematol 1998;60:112-8. 56. Christie D J, Van Buren N, Lennon SS, Putman JL. Vancomycin-dependent antibodies associated with thrombocytopenia and refractoriness to platelet transfusion in patients with leukemia. Blood 1990;75:518-23.

Phone, Fax, and Internet Information: If you have any questions concerning Immunohematology, Journal of Blood Group Serology and Education, or the Immunohematology Methods and Procedures manual,contact us by e-mail at [email protected] information concerning the National Reference Laboratory for Blood Group Serology, including the American Rare Donor Program, please contact Sandra Nance, by phone at (215) 451-4362, by fax at (215) 451-2538, or by e-mail at [email protected]

Free Classified Ads and Announcements Immunohematology will publish classified ads and announcements (SBB schools, meetings, symposia, etc.) without charge. Deadlines for receipt of these items are as follows:

Deadlines 1st week in January for the March issue 1st week in April for the June issue 1st week in July for the September issue 1st week in October for the December issue

Fax these items to Mary H. McGinniss, Managing Editor, (301) 299-7443.

102 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: transfusion-related acute lung injury: pathophysiology, laboratory investigation, and donor management

P.M. KOPKO

Transfusion-related acute lung injury (TRALI) is a serious clinical beginning within 6 hours of transfusion with rapid syndrome that is temporally associated with the transfusion of resolution of symptoms and reappearance of more plasma-containing blood components. The syndrome typically 15 occurs within 6 hours of transfusion. Approximately 80 percent of severe symptoms about 24 hours after transfusion. cases will resolve within 96 hours with supportive care. The Signs and symptoms include fever, dyspnea, syndrome has been associated with antibodies to WBC antigens and hypotension, hypertension followed by hypotension, generation of biologically active mediators in stored cellular blood components. Appropriate laboratory investigation of TRALI can be and noncardiogenic pulmonary edema. Although any crucial in confirmation of the clinical diagnosis, as well as in of these signs and symptoms can be seen in TRALI, not decisions regarding donor management. Immunohematology all of them will be present in every case. 2004;20:103–111. Patients who are intubated, at the time of the Key Words: TRALI, TRALI pathophysiology and reaction or shortly after the reaction begins, are laboratory investigation, donor management—TRALI described as producing copious quantities of frothy fluid from the endotracheal tube.16–19 The frothy quality Transfusion-related acute lung injury (TRALI) is a of this fluid is secondary to its high protein content. syndrome that is clinically indistinguishable from acute The ratio of protein concentration in edema fluid to respiratory distress syndrome (ARDS).1 TRALI can protein concentration in the blood in patients with occur within 6 hours of transfusion, but usually occurs noncardiogenic pulmonary edema is typically > 0.7, within 1 to 2 hours.2 Symptoms and signs can include while the ratio is usually < 0.5 in cardiogenic shortness of breath, difficulty breathing, hypoxia, fever, pulmonary edema.20 Yost et al.21 analyzed the protein hypotension, and hypertension preceding hypo- content of pulmonary edema fluid in seven cases of tension.3 All plasma-containing blood components, TRALI that occurred after liver transplantation. In all including whole blood, packed RBCs, fresh frozen seven cases, the ratio of protein in the pulmonary 2 4,5 6,7 plasma, platelet concentrates, apheresis platelets, edema fluid to protein in the blood was > 0.7. 8 9 , , allogeneic bone The chest X-ray becomes abnormal at the time of 10 11 marrow, and IVIG, have been associated with TRALI. the reaction. In recumbent patients, a patchy infiltrate Although the pathogenesis of TRALI is not completely may first appear in the dependent areas of the lungs. understood, it has been associated with antibodies to The infiltrate rapidly progresses to bilateral “white out.” 2,7,12 WBCs and with the infusion of biologically active This pattern is caused by pulmonary edema and is 13,14 mediators in stored cellular blood components. indistinguishable from the pattern seen in ARDS. In approximately 80 percent of cases, the Clinical Syndrome symptoms completely resolve within 96 hours of In most cases of TRALI, clinical symptoms begin transfusion.2 Once the symptoms resolve, there are no within 1 to 2 hours of completion of the transfusion. chronic manifestations of TRALI. Fatal reactions occur Symptoms can begin up to 6 hours after transfusion.2 in 5 percent to 10 percent of cases. Fatalities can occur There are rare case reports of mild symptoms acutely at the time of the reaction or after a protracted

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 103 P.M. KOPKO course of mechanical ventilation. In the remaining the granulocyte 5b antigen (HNA-3a) and had donated cases, patients fully recover after intensive support and apheresis plasma (600 mL) in a frequent plasma mechanical ventilation. donation program. The medical records of 36 recipients of this donor’s plasma were reviewed by transfusion Differential diagnosis service medical directors for indications of TRALI The most important factor in the differential associated with transfusion of the blood components. diagnosis of TRALI is to determine the nature of the Reactions were classified as mild to moderate or severe. pulmonary edema. The edema in TRALI is noncardio- Severe reactions were defined as acute onset of genic in nature. Other forms of pulmonary edema pulmonary edema, or need for mechanical ventilation, include cardiogenic and volume overload. TRALI, or associated with transfusion. Mild to moderate reactions noncardiogenic pulmonary edema, can easily be were defined as dyspnea, and/or oxygen desaturation confused with these other two types of pulmonary without overt clinical evidence of pulmonary edema, edema, particularly in critically ill patients and those associated with transfusion. Eight severe reactions were who have received large volumes of intravenous fluids, identified in eight recipients, while seven mild to including blood components. moderate reactions were identified in six recipients. Volume overload pulmonary edema is associated Only two of the eight severe reactions were reported to with other signs of fluid overload, including jugular the transfusion service, while five of the seven mild to venous distension and increased blood pressure. moderate reactions were reported to the transfusion Cardiogenic pulmonary edema is associated with signs service. Only two of the reactions (one mild to and symptoms of heart failure, including gallops, moderate and one severe) were reported to the blood murmurs, or possibly evidence of myocardial collection facility. The two reactions reported to the ischemia/infarction on electrocardiogram. Additionally, blood collection facility were the fatal reaction that laboratory tests for myocardial infarction and heart prompted the look-back and a mild to moderate failure, including creatine kinase, troponin, and B-type reaction that occurred at the same institution during natriuretic peptide,can aid in the differential diagnosis. the initial investigation of the fatality. The observation All of these tests should be within normal limits in that mild to moderate TRALI reactions were reported to TRALI. the transfusion service more frequently than severe reactions may be explained by the fact that the mild to Incidence moderate manifestations of TRALI can clinically TRALI is one of the most common causes of resemble febrile nonhemolytic transfusion reactions transfusion-related mortality. It has been reported as (FNHTR) that are routinely reported to the transfusion the third leading cause of transfusion-related mortality service. in the United States22 and the most common cause in This “look-back” study emphasizes the importance the United Kingdom.23 An accurate estimate of the of clinical recognition of TRALI. Pulmonary symptoms incidence of TRALI has been difficult to establish. Most that occur within 6 hours of transfusion should be of the published estimates of the incidence of TRALI reported to the transfusion service for further are based upon calculating the number of cases of investigation, particularly when there is no other TRALI that occur at a specific institution over a period clinical explanation for the symptoms. After of time, compared to the number of components or investigation by the transfusion service, probable cases patients transfused during the same period. These of TRALI should be referred to the blood collection studies have led to estimates of one case of TRALI per facility for donor testing and possible deferral of 5000 components transfused2 and one case per 2400 implicated donors, depending upon the results of the patients transfused.24 laboratory investigation. Although these studies give an estimate of the incidence of TRALI,there is a growing appreciation that The Antibody Hypothesis for TRALI TRALI may be under-diagnosed and underreported.25 A The first published case report of what was most recent retrospective study examined the medical likely TRALI was reported in 1951.26 During the next records of the recipients of blood components donated 30 years there were several case reports of by a female donor who was implicated in a fatal case of noncardiogenic pulmonary edema associated with TRALI.25 The implicated donor had a strong antibody to transfusion.8,16,27–34 These case reports attribute the

104 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: transfusion-related acute lung injury pulmonary edema to several different causes, including Recipients were typed for HLA Class I and II antigens incompatibility of an undetermined nature,28 human in all cases. If the donor(s) did not possess HLA or leukocyte antigen (HLA) incompatibility,32 non-HLA granulocyte antibodies, the recipient was evaluated for leukoagglutinins,30 undefined granulocyte leukoag- antibodies to HLA Class I and II antigens. In 14 of the glutinins,34 and severe allergic pulmonary edema.8,16,31 16 cases (87.5%), a correlation was identified between Antibodies to HLA Class I and granulocyte antigens antibody present in the donor or the recipient and were associated with TRALI by Popovsky et al.2,17 In a antigen in the other member of the donor-recipient series of 36 patients, granulocyte antibodies were pair. The identity of the antibody involved in these identified in the serum of at least one implicated donor cases was HLA Class I (4), HLA Class II (5), HLA Class I in 89 percent of cases. The presence of lymphocyte and II (2), granulocyte (1), and monocyte (2). In the antibodies was determined, by reverse lymphocyto- two cases where antibody was not identified, recipient toxic crossmatch, in the serum of at least one samples were not available for antibody testing. Several implicated donor, in 26 of the 36 cases (72%). recent case reports support the association of HLA Specificity of the HLA antibodies was further Class II antibodies with TRALI.12,37–40 characterized by lymphocyte panel testing in 17 of the Six of the 16 TRALI cases described above were 26 cases with HLA antibody. A clear HLA specificity further investigated for the ability of the sera was determined in 11 of the 17 cases (65%). The containing antibody from one member of the donor- specificity of the antibody was determined to recipient pair to activate monocytes in the other correspond to at least one of the recipient’s HLA member of the pair. The fraction of cells expressing the antigens in 10 of the 17 cases. interleukin 1β (IL-1β) and tumor necrosis Many of the early case reports of TRALI report the factor α (TNF-α) or tissue factor were measured as an presence of granulocyte or HLA antibodies based upon indicator of monocyte activation. Expression of the a positive reverse granulocyte or lymphocyte cross- two cytokines and tissue factor were measured after match. The results of these tests can be misleading exposure to autologous serum, control serum (serum because of the presence of other antigen systems on from the donor of another blood component the these cells. This has the potential to result in incorrect patient received or ABO-matched control serum), or interpretation of crossmatch results. An example of serum involved in the TRALI reaction. There was this is the granulocyte 5b antigen. Since this antigen is significantly increased expression of all three factors in present on lymphocytes in addition to granulocytes,35,36 the monocytes exposed to TRALI serum compared to a positive reverse lymphocyte crossmatch may lead to autologous serum and control serum (p < 0.05). the misinterpretation that lymphocyte or HLA TRALI has been reproduced in an ex vivo animal antibodies are responsible for the TRALI reaction. lung model.41 In this model, pulmonary edema is The technologies available to identify HLA produced after a latent period of 3 to 6 hours when antibodies have dramatically evolved in the last 10 years rabbit lungs are perfused with 5b-positive with the availability of solid phase assays such as flow granulocytes, antibody to granulocyte 5b, and rabbit cytometry and ELISA in most histocompatibility plasma as a source of complement. If any of these laboratories. Because these technologies are based three components is absent, pulmonary edema does upon isolation and adherence of HLA antigens to a latex not occur. The pulmonary edema produced in these bead (flow cytometry) or solid surface (ELISA), they are experiments was noncardiogenic in nature. This was more specific for detection of HLA antibodies than tests demonstrated by monitoring of pulmonary artery based upon lymphocytotoxicity. Additionally, these pressure, which displayed only a transient and tests do not depend upon the ability of the antibody to moderate increase. Increased pulmonary vascular activate complement and cause cellular lysis. permeability was the cause of the edema. A recent study by Kopko et al.7 investigated 16 In the antibody-mediated model of TRALI,antibody- cases of TRALI. This study examined donor-recipient coated leukocytes localize to the pulmonary micro- pairs involved in TRALI reactions. Donors were tested vasculature. release by these leukocytes is for HLA and monocyte antibodies by flow cytometry, thought to damage the vascular endothelium. Damage HLA antibodies via a lymphocytotoxic panel, and to the endothelium of the pulmonary microvasculature granulocyte antibodies by granulocyte agglutination is thought to cause increased vascular permeability. (GA) and granulocyte immunofluorescence (GIF). This results in leakage of protein-rich fluid into the

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 105 P.M. KOPKO pulmonary alveoli and results in pulmonary edema. presence of a “first event”that would have predisposed This theory is supported by the findings of McCullough them to develop TRALI. A control group of ten patients et al.,42 who have demonstrated that 111Indium-labeled who experienced a febrile or urticarial reaction was granulocytes localize to the pulmonary microvascula- also assessed for the presence of a “first event” prior to ture when transfused to a recipient with antibodies transfusion. All of the patients in the TRALI group had directed against the granulocytes. an underlying clinical factor that could have A recent case report suggests that interaction of predisposed them to develop TRALI, while only two antibody with leukocytes may not be necessary for patients in the control group had such underlying TRALI to occur.43 In this case, a 34-year-old woman clinical factors. Clinical factors thought to predispose who received two units of packed RBCs approximately a patient to developing TRALI included infection, 10 weeks after undergoing a lung transplant cytokine administration, recent surgery, and massive experienced a TRALI reaction. Pulmonary edema was transfusion. present in the transplanted, but not in the native, lung. This mechanism of TRALI has also been reproduced Antibodies to HLA-B44 were identified in the donor of in an ex vivo animal lung model.45 In this study,rat lungs the second unit of packed RBCs. No leukocyte were first pretreated with lipopolysaccharide (LPS) to antibodies were identified in the donor of the first unit simulate sepsis. The lungs were perfused with one of of packed RBCs. The transplanted lung possessed the the following solutions:buffered salt solution,5% human HLA-B antigen, while the patient did not. Since very plasma in saline, 5% supernatant from day 0 or day 5 few WBCs of donor origin would be expected to be platelets (both whole blood–derived and apheresis present in the transplant recipient 10 weeks after platelets), or lipid extracts from day 0 or day 5 platelets. transplant, and the pulmonary endothelial cells would Pulmonary edema (lung weight) and leukotriene B4 be expected to be of donor origin, this case suggests levels (a measure of lung injury) were measured. Plasma that TRALI may be initiated directly through antibody from day 5 apheresis or whole blood–derived platelets, interaction with endothelial cells. as well as lipid extracts from these components, caused increased pulmonary edema (p < 0.05) when the lungs The Biologically Active Mediator Model for were first treated with LPS compared to saline TRALI pretreated and saline perfused lungs. Pulmonary edema Biologically active mediators in cellular blood did not develop if the lungs were pretreated with saline components, at or near the time of outdate, have been instead of LPS. Lungs perfused with supernatant or lipid associated with TRALI.13 In this theory of TRALI, extract from day 0 whole blood–derived or apheresis biologically active mediators or lipids accumulate platelets did not develop pulmonary edema. during storage of cellular blood products, including Leukotriene B4 levels were increased in the lung packed RBCs and platelets. These agents enhance or perfusate (p < 0.05) only after pretreatment with LPS prime neutrophil NADPH Oxidase. In this model, two and perfusion of the lungs with 5% supernatant from day 44 events are required for a reaction to occur. In the first 5 platelets. Leukotriene B4 levels were not measured event, biologically active mediators are generated after treatment with lipid extracts. Prestorage during physiologic stress. These stressors can include leukoreduction of platelets did not alter the pulmonary events such as trauma, infection, recent surgery, and edema that developed in these experiments. Pulmonary massive transfusion. Generation of biologically active artery pressure was monitored in these experiments to compounds, related to stress, is hypothesized to ensure the edema was noncardiogenic in nature. activate the pulmonary vascular endothelium and Significant, sustained increases in pulmonary artery prime . These actions are thought to cause pressure were not observed. sequestration of neutrophils within the pulmonary Silliman et al.46 recently published a study of 90 microvasculature. The second event consists of the TRALI reactions occurring during a 4-year period at a infusion of biologically active mediators in a cellular single institution. The blood components implicated in blood component. these reactions were platelet concentrates (72), This model of TRALI is supported by a apheresis platelets (2), packed RBCs (15), and plasma retrospective study performed by Silliman and (1). The first 46 reactions were analyzed in a nested colleagues.14 The records of ten transfusion recipients case-control study. Patient and blood component data who experienced TRALI were assessed for the were compared in these cases to a control group of

106 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: transfusion-related acute lung injury

225 recipients of platelet concentrates. TRALI was not have been prestorage leukoreduced. Therefore, if associated with patient age or gender, incompatibility biologically active mediators are associated with of patient-donor blood groups, number of previous nonleukoreduced platelet and RBC concentrates,46 it transfusions, or number and type of previous would be unlikely that a case of TRALI due to these transfusion reactions. TRALI was associated with a substances would be found at the author’s center. diagnosis of hematologic malignancy (p < 0.0004) and cardiac disease (p < 0.0006). TRALI was also associated Clinical Investigation with increasing age of platelet concentrate transfused The laboratory investigation of TRALI can be a (p = 0.014). Pretransfusion and posttransfusion frustrating and costly task. Cases of TRALI are often samples from the last 51 patients who experienced referred to the clinical laboratory with a large number TRALI were analyzed for the accumulation of of implicated blood components. Additionally, the polymorphonuclear neutrophil leukocyte priming laboratory is seldom provided data regarding when the activity. There was significantly more priming activity implicated blood components were transfused relative (p < 0.05) in the posttransfusion samples compared to to the reaction. Since most cases of TRALI occur within pretransfusion samples and controls. 1 to 2 hours of transfusion, this information is essential to the laboratory investigation of TRALI. Therefore, it is Pathophysiology of TRALI important to obtain this detail prior to initiating Although two seemingly dissimilar mechanisms testing. This allows the laboratory to prioritize testing have been associated with TRALI, it is possible that of implicated donors. The use of a testing algorithm in both mechanisms may be involved in the reaction. the laboratory investigation of TRALI can significantly Activation of leukocytes with endothelial damage, reduce the cost of working up a case. increased vascular permeability, and resultant It is strongly recommended that donor-recipient pulmonary edema are common to the antibody- pairs be investigated in TRALI cases. One of the most mediated model of TRALI and the biologically active important steps needed to accomplish a complete mediator model of TRALI. Additionally, look-back laboratory investigation is to obtain the appropriate studies have shown that the presence of antibody in a samples from the recipient as soon as TRALI is donor is not sufficient to cause TRALI in all recipients, considered a possibility. If recipient samples for HLA even if the recipient possesses an antigen typing and antibody studies are not drawn early in the corresponding to the infused antibody.25,47 These investigation, they may not be available if they are studies suggest that two events may be needed for needed to correlate results in the recipient with results TRALI to occur. Biologically active mediators have from the donor(s). been demonstrated in cellular blood components, A suggested testing algorithm is presented in while they have not been routinely demonstrated in Table 1. Testing should begin with female donors of FFP. Since FFP has frequently been implicated in blood components transfused within 2 hours of the TRALI,7,18,31,48 biologically active mediators alone can’t reaction. Since testing for HLA antibodies is more explain all cases of TRALI. readily available than testing for granulocyte An important point to remember when antibodies, and HLA antibodies have recently been considering the two hypotheses for TRALI is that the associated with TRALI more often than granulocyte differences in transfusion practice around the country antibodies, testing can begin with HLA Class I and II and around the world may affect the findings in TRALI antibodies. If no HLA antibodies are detected, testing investigations. Although the author’s institution has for granulocyte antibodies should be performed. The reported identifying antibody in one member of the recipient should be HLA typed (Class I and II) to donor-recipient pair in the majority of cases, those data determine whether antibodies identified in a donor are possibly biased by our area’s transfusion practices. correspond to recipient antigens. Hospitals supplied by the author’s blood center have If female donors of blood components transfused exclusively received apheresis platelets for more than a within 2 hours of the reaction are negative for HLA and decade. For the last 7 years, the apheresis platelets granulocyte antibodies, female donors of blood have been leukocyte reduced during the collection components transfused within 6 hours of the reaction process. Additionally, the overwhelming majority (> should be tested for HLA antibodies, followed by 98%) of RBCs collected within the last few years testing for granulocyte antibodies. If all female donors

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 107 P.M. KOPKO

Table 1. Laboratory investigation of TRALI particularly if the patient received a nonleukocyte Recipient samples and testing reduced cellular blood component (i.e., packed RBCs 1. When TRALI is first suspected, obtain serum/plasma and cellular or platelets). samples (citrate anticoagulant) If the recipient and the female donors of blood 2. Obtain HLA Class I and II typing on cellular samples or freeze cells for components transfused within 6 hours of the reaction DNA for future testing are negative for HLA and granulocyte antibodies, male 3. Freeze recipient serum/plasma sample for future testing (prereaction donors of blood components transfused within 2 hours and postreaction samples, if available) of the reaction can be tested next. If the testing is still 4. Separate and freeze liquid portion of remaining contents of bag, if available, for future testing negative, male donors of blood components transfused within 6 hours of the reaction can be tested. At any Donor testing—HLA and granulocyte point in the TRALI investigation, if antibody is 1. Obtain transfusion history including: identified in one member of the donor-recipient pair a. Unit number of transfused component(s) and it corresponds to antigen in the other member of b. Start/stop times of all components transfused c. Time of first symptoms of reaction the donor-recipient pair,further workup can be halted. d. Gender of donor of each of the components transfused within 6 Currently, testing for monocyte antibodies and hours of reaction symptoms biologically active mediators is not commercially 2. Test female donors of components transfused within 2 hours of available. If testing for HLA and granulocyte antibodies reaction for HLA Class I and II antibodies a. If positive—correlate with recipient HLA typing (donor in step 4) is negative in the donor(s) and recipient, testing for i. If correlation exists—stop these other factors associated with TRALI can be b. If negative—test donors (recipient in step 4) for granulocyte obtained through research laboratories. antibodies i. If positive—correlate with antigens in recipient (donor in step 4) 1. If correlation exists—stop Treatment ii. If negative—proceed to next step Because TRALI occurs infrequently, there are no 3. Test female donors of components transfused within 6 hours of clinical studies of appropriate treatment after a reaction for HLA Class I and II antibodies (go to step 2a) reaction has occurred. The only treatments that can be 4. If recipient is at increased risk of having antibodies (multiparous or recommended are supportive care for the patient’s recipient of an organ allograft or multiple transfusions) and received a symptoms, including oxygen support with or without cellular blood component (go to step 2a) mechanical ventilation, and fluid administration. The 5. Test male donor of components transfused within 2 hours of reaction for HLA Class I and II antibodies (go to step 2a) use of steroids for treatment of TRALI is anecdotal. Therefore, no recommendation regarding steroid use 6. Test male donors of components transfused within 6 hours of reaction for HLA Class I and II antibodies (go to step 2a) can be made. 7. Consider obtaining monocyte antibody testing (available through One caution in the treatment of TRALI is the research laboratories only) avoidance of diuretics. There are several case reports that describe worsening of the patient’s clinical Testing for biologically active mediators condition and even fatality following administration of Testing should be considered (available through research laboratories 8,18 only) if: diuretics after a reaction. Since the pulmonary • Recipient received a cellular blood component with little plasma edema associated with TRALI is noncardiogenic, content patients are not typically volume overloaded. The • Reaction occurred after transfusion of an autologous blood component patient may be hypovolemic due to the underlying condition or secondary to the extravasation of fluid of blood components transfused within 6 hours of the into the lungs. Therefore, administration of diuretics can lead to hypotension, decreased cardiac output, and reaction are negative for HLA and granulocyte 8,18 antibodies, the recipient can be tested for these decreased pulmonary capillary wedge pressure. antibodies. This step will allow diagnosis of the approximately 10 to 15 percent of TRALI cases that are Donor Management Issues associated with antibodies in the recipient. If the One of the most important considerations transfusion recipient is at high risk of having leukocyte regarding TRALI is donor management. Several antibodies, due to a history of pregnancy or different suggestions regarding donor management transfusion, it may be advisable to test the recipient for have been published. These include deferring all antibodies earlier in the course of the workup, females from donating plasma for transfusion,49

108 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: transfusion-related acute lung injury deferring multiparous females from donating plasma 2. Popovsky MA, Moore SB. Diagnostic and for transfusion,49 deferring multiparous females from pathogenetic considerations in transfusion-related plasma donation unless their serum has been tested for acute lung injury.Transfusion 1985;25:573-7. antibodies to leukocytes,50 and deferring donors 3. Popovsky MA, Haley NR. Further characterization implicated in a TRALI case, if they have been found to of transfusion-related acute lung injury demo- have antibody in their serum that corresponds to graphics, clinical and laboratory features, and antigen on the TRALI patient’s leukocytes.25,50 Each of morbidity. Immunohematology 2000;16:157-9. these donor management strategies would reduce the 4. Ramanathan RK, Triulzi DF, Logan TF. Transfusion- risk of TRALI in transfusion recipients. However, none related acute lung injury following random donor of them would completely eliminate TRALI. platelet transfusion: a report of two cases.Vox Sang Deferring high-risk donors from donating plasma 1997;73:43-5. components would not eliminate the risks associated 5. Virchis AE, Patel RK, Contreras M, et al. Acute non- with the plasma in other blood components. A recent cardiogenic lung oedema after platelet transfusion. retrospective study correlated the volume of plasma in Br Med J 1997;314:880-2. a blood component with the likelihood that a blood 6. Eastlund T, McGrath PC, Burkart P. Platelet component would be implicated in a TRALI case.51 If transfusion reaction associated with interdonor the volume of plasma in a blood component correlates HLA incompatibility.Vox Sang 1988;55:157-60. to the risk of TRALI from a blood component, then 7. Kopko PM, Paglieroni TG, Popovsky MA, et al. apheresis platelets would have a slightly greater risk of TRALI: correlation of antigen-antibody and causing TRALI than plasma derived from a monocyte activation in donor-recipient pairs. whole–. None of the deferral strategies Transfusion 2003;43:177-84. that exclude donors based upon gender or parity 8. O’Connor JC, Strauss RG, Goeken NE, Knox LB. A include deferral from apheresis platelet donation. near-fatal reaction during granulocyte transfusion Additionally, very small volumes of plasma have been of a neonate.Transfusion 1988;28;173-6. shown to cause TRALI.27 These deferral strategies 9. Reese EP,McCullough JJ, Craddock PR. An adverse would not prevent cases of TRALI secondary to the pulmonary reaction to cryoprecipitate in a small volumes of plasma present in cryoprecipitate or hemophiliac.Transfusion 1975;15:583-8. RBC and platelet concentrates. 10. Urahama N, Tanosaki R, Masahiro K, et al. TRALI None of the donor management strategies would after the infusion of marrow cells in a patient with prevent cases of TRALI secondary to antibody in the acute lymphoblastic leukemia. Transfusion recipient. Finally, none of the strategies would address 2003;43:1553-7. the association of TRALI with biologically active lipids. 11. Rizk A, Gorson KC, Kenney L, Weinstein R. There are also obstacles to screening female or Transfusion-related acute lung injury after the multiparous donors for antibodies to leukocytes. In infusion of IVIG.Transfusion 2001;41:264-8. order to screen for leukocyte antibodies, a test that 12. Kopko PM, Popovsky MA, MacKenzie MR, et al. HLA could be performed on a large number of donors in a Class II antibodies in transfusion-related acute lung short period of time would be needed. Although injury.Transfusion 2001;41:1244-8. screening tests for HLA Class I and II antibodies using 13. Silliman CC, Thurman GW, Ambruso DR. Stored flow cytometry or ELISA methodologies are currently blood components contain agents that prime the available commercially,there are no such tests available neutrophil NADPH oxidase through the platelet- for screening for antibodies to the granulocyte 5b activating-factor receptor.Vox Sang 1992;63:133-6. antigen. Since this antibody has recently been reported 14. Silliman CC, Paterson AJ, Dickey WO, et al. The in association with three fatal cases of TRALI,25,52 association of biologically active lipids with the inclusion of a test for this antibody in a screening test development of transfusion-related acute lung would be ideal. injury: a retrospective study. Transfusion 1997;37: 719-26. References 15. Levy GJ, Shabot MM, Hart ME, et al. Transfusion- 1. Kopko PM, Holland PV. Transfusion-related acute associated noncardiogenic pulmonary edema. lung injury. Br J Haematol 1999;105:322-9. Transfusion 1986;26:278-81.

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 109 P.M. KOPKO

16. Culliford AT, Thomas S, Spencer FC. Fulminating 30. Thompson JS, Severson CD, Parmely MJ, et al. noncardiogenic pulmonary edema: a newly Pulmonary “” reactions induced by recognized hazard during cardiac operations. J transfusion of non-HLA leukoagglutinins. N Engl J Thorac Cardiovasc Surg 1980;80:868-75. Med 1971;20:1120-5. 17. Popovsky MA, Abel MD, Moore SB. Transfusion- 31. Kernoff PBA,Durrant IJ,Rizza CR,Wright FW.Severe related acute lung injury associated with passive allergic pulmonary oedema after plasma trans- transfer of antileukocyte antibodies. Am Rev Respir fusion. Br J Haematol 1972;23:777-81. Dis 1983;128:185-9. 32. Wolf CFW, Canale VD. Fatal pulmonary 18. Hashim SW, Kay HR, Hammand GL, et al. hypersensitivity reaction to HL-A incompatible Noncardiogenic pulmonary edema after cardio- blood transfusion.Transfusion 1976;16:135-40. pulmonary bypass: an anaphylactic reaction to 33. Carilli AD, Ramanamurty MV, Chang YS, et al. .Am J Surg 1984;147:560-4. Noncardiogenic pulmonary edema following blood 19. Kawamata M, Miyabe M, Omote K, et al. Acute transfusion. Chest 1978;74:310-2. pulmonary edema associated with transfusion of 34. Dubois M, Lotze MT, Diamond WJ, et al. Pulmonary . Intensive Care Med shunting during leukoagglutinin-induced 1995;21:443-6. noncardiac pulmonary edema. JAMA 1980;244: 20. Sprung CL, Rackow EC, Fein IA, et al.The spectrum 2186-9. of pulmonary edema:differentiation of cardiogenic, 35. Van Leeuwen A, Eernise JG, Van Rood JJ. A new intermediate, and noncardiogenic forms of leukocyte group with two alleles: leukocyte group pulmonary edema. Am Rev Respir Dis 1981;13: five.Vox Sang 1964;9:431-7. 718-22. 36. Stroncek D. Neutrophil alloantigens. Transfus Med 21. Yost CS,Matthay MA,Gropper MA.Etiology of acute Rev 2002;16:67-75. pulmonary edema during liver transplantation. 37. Flesch BK,Neppert J.Transfusion-related acute lung Chest 2001;119:219-23. injury caused by human leucocyte antigen Class II 22. Sazama K. Reports of 355 transfusion-associated antibody. Br J Haematol 2002;116:673-6. deaths; 1976 through 1985. Transfusion 1990;30: 38. Varela M, Mas A, Nogues N, et al. TRALI associated 583-90. with HLA Class II antibodies. Transfusion 2002;42: 23. SHOT Annual report 2000/2001. Available at 1102. http//www.shotuk.org. 39. Win N, Brown C, Navarrete C. TRALI associated 24. Weber JG, Warner MA, Moore SB. What is the with HLA Class II antibodies. Transfusion 2003;43: incidence of perioperative transfusion-related 545-6. acute lung injury? Anesthesiology 1995;82:789. 40. Wallis JP, Lubenko A, Wells AW, Chapman C. Single 25. Kopko PM, Marshall CS, MacKenzie MR, et al. hospital experience of TRALI.Transfusion 2003;43: Transfusion-related acute lung injury: report of a 1053-9. clinical look-back investigation. JAMA 2002;287: 41. Seeger W, Schneider U, Kreusler B, et al. 1968-71. Reproduction of transfusion-related acute lung 26. Barnard RD.Indiscriminate transfusion:a critique of injury in an ex vivo lung model. Blood 1990;76: case reports illustrating hypersensitivity reactions. 1438-44. N Y State J Med 1951;51:2399-402. 42. McCullough J, Clay M, Hurd D, et al. Effect of 27. Brittingham TE. Immunologic studies on leuko- leukocyte antibodies and HLA matching on the cytes.Vox Sang 1957;2:242-8. intravascular recovery, survival, and tissue 28. Phillips E, Fleischner FG. Pulmonary edema in the localization of 111Indium granulocytes. Blood course of a blood transfusion without overloading 1986;67:522-8. the circulation. Dis Chest 1966;50:619-23. 43. Dykes A, Smallwood D, Kotsimbos T, Street A. 29. Ward HN. Pulmonary infiltrates associated with Transfusion-related acute lung injury (TRALI) in a leukoagglutinin transfusion reactions. Ann Intern patient with a single lung transplant. Br J Haematol Med 1970;73:689-94. 2000;109:674-6. 44. Silliman CC. Transfusion-related acute lung injury. Transfus Med Rev 1999;13:177-86.

110 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: transfusion-related acute lung injury

45. Silliman CC, Bjornsen AJ, Wyman TH, et al. Plasma 50. Popovsky MA, Chaplin HC, Moore SB. Transfusion- and lipids from stored platelets cause acute lung related acute lung injury: a neglected, serious injury in an animal model. Transfusion 2003;43: complication of hemotherapy.Transfusion 1992;32: 633-40. 589-92. 46. Silliman CC, Boshkov LK, Mehdizadehkashi Z, et al. 51. Kopko PM, MacKenzie M, Holland PV,Popovsky M. Transfusion-related acute lung injury: epidemiology Transfusion-related acute lung injury: dispropor- and a prospective analysis of etiologic factors. tionate involvement of plasma components. Blood 2003;101:454-62. Transfusion 2002b;42:29S. 47. Cooling L. Transfusion-related acute lung injury. 52. Davoren A, Curtis BR, Shulman IA, et al. TRALI due JAMA 2002;288:315-6. to granulocyte-agglutinating human neutrophil 48. Eastlund T, McGrath PC, Britten A, Propp R. Fatal antigen-3a (5b) alloantibodies in donor plasma: a pulmonary transfusion reaction to plasma report of 2 fatalities.Transfusion 2003;43:641-5. containing donor HLA antibody.Vox Sang 1989;57: 63-6. Patricia M. Kopko, MD, Associate Medical Director, 49. Boulton-Jones R, Norris A, O’Sullivan A, et al. The Blood Source, 1625 Stockton Blvd., Sacramento, CA impact of screening a platelet donor panel for 95816-7089. human leucocyte antigen antibodies to reduce the risk of transfusion-related acute lung injury. Transfus Med 2003;13:169-70.

Manuscripts: The editorial staff of Immunohematology welcomes manuscripts pertaining to blood group serology and education for consideration for publication.We are especially interested in case reports, papers on platelet and white cell serology,scientific articles covering original investigations, and papers on the use of computers in the blood bank. Deadlines for receipt of manuscripts for consideration for the March, June, September, and December issues are the first weeks in November, February, May, and August, respectively. Instructions for scientific articles and case reports can be obtained by phoning or faxing a request to Mary H. McGinnniss, Managing Editor, Immunohematology, at (301) 299-7443, or see “Instructions for Authors” in every issue of Immunohematology or on the Web. Include fax and phone numbers and e-mail address with your manuscript.

REMEMBER: THE PASSWORD IS “2000” For www.redcross.org/pubs/immuno Now, as a subscriber, you can enter the password, 2000, and access the back issues. That means cover to cover! You will receive every article in total, every letter to the editor, every review, every ad, every notice, and every literature review! All of the other services will continue to be available on the Web site, including sending a letter to the editor, subscribing with a credit card on the secure order site, performing a literature search, reviewing instructions for authors, and linking to other important sites. Log on now to see this great service!

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 111 Review: immune thrombocytopenic purpura: an update for immunohematologists

S.G. SANDLER

Immune thrombocytopenic purpura (ITP) is an acquired disease in highly variable disease into categories without a which autoantibodies to platelets cause their sequestration and significant number of exceptions. The goal of destruction by mononuclear macrophages, principally in the spleen. If increased production of platelets by megakaryocytes classifying patients is to be able to match a newly does not compensate for platelet destruction, the number of diagnosed patient with a group of other patients with circulating platelets decreases (thrombocytopenia), resulting in a similar findings for purposes of selecting an characteristic bleeding tendency (purpura). While most children with the disease experience a relatively short and benign clinical appropriate treatment and providing prognostic course, ITP in adults often lasts more than 6 months (chronic ITP) information. and is resistant to conventional treatment (corticosteroids, intravenous immune globulin, or splenectomy). The goal of medical management is to increase the platelet count to a safe level, Primary versus secondary immune without the risks of bacterial infections associated with thrombocytopenia splenectomy or toxicity from prolonged corticosteroid therapy. Conventionally, thrombocytopenias are divided Splenectomy increases platelet counts in hours to days in most patients with acute ITP, but nearly 50 percent experience into three major categories: those resulting from (1) recurrent thrombocytopenia by 5 years postsplenectomy. abnormally decreased platelet production (e.g.,aplastic Immunohematology 2004;20:112–117. anemia), (2) platelet sequestration (e.g., hypersplenism), or (3) abnormally increased platelet Key Words: platelets, thrombocytopenia, platelet destruction (e.g., immune destruction). Immune antibodies, reticuloendothelial system, Fc (antibody-mediated) thrombocytopenia may occur as a receptors, splenectomy, intravenous immune globulin, complication of another disease (secondary IV Rh immune globulin, rituximab thrombocytopenia) (e.g., HIV/AIDS, systemic lupus erythematosis, or a primary Introduction syndrome) or as the only abnormal finding (primary Immune thrombocytopenic purpura (ITP, immune thrombocytopenia). In the following review, idiopathic or autoimmune thrombocytopenic purpura) “ITP” is intended to refer to primary immune is an acquired bleeding disorder of children and adults thrombocytopenia. that presents with thrombocytopenia as the only significant abnormal laboratory finding. There is ITP in children versus in adults neither a specific clinical nor a laboratory finding that Typically, children with ITP present with an acute reliably establishes a diagnosis of ITP. A diagnosis of onset of petechiae and diffuse bruising approximately ITP is established by excluding other potential causes 4 to 8 weeks after a viral illness. As many as 85 percent of thrombocytopenia. of affected children experience an uneventful course and spontaneous remission within 6 months.1 In 2003, Classification the International Childhood ITP Study Group reported ITP is a heterogeneous disease with a wide range of the results of a prospective study of 2540 infants and clinical and laboratory presentations. Typically,patients children with newly diagnosed ITP, providing are classified according to age at the time of diagnosis voluminous data on the clinical spectrum of childhood (ITP in children versus adults),but no one classification ITP.2 In this study, the incidence of intracranial system is capable of segregating patients with this hemorrhage, the most feared complication of ITP in

112 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: thrombocytopenic purpura children,was only 0.17% during the first 6 months after might expect that the severity of bleeding in patients diagnosis. In contrast to ITP in children, ITP in adults with ITP would correlate with the presence of one or typically presents with an insidious onset; affects a more of these antibodies, since their glycoprotein preponderance of females, as seen in other auto- targets are integral to normal platelet function and immune diseases in adults; and progresses to chronic hemostasis. However, current laboratory assays for ITP (persistent thrombocytopenia or a requirement for platelet antibodies do not demonstrate a convincing treatment 6 months after onset). This traditional profile correlation between specific glycoprotein targets and of ITP in adults has been challenged by the results of a disease severity.6 Neither research-level nor prospective study reported in 2003 for a population commercially marketed assays for detecting platelet cohort of 245 adults with ITP in the UK.3 In this study, antibodies have sufficient specificity to be useful as there were 134 females/111 males (1.2:1.0) and the diagnostic assays for ITP in clinical practice.8 An majority of patients (63.3%) achieved remission evaluation of the performance of Capture-P (Immucor, (platelet count > 100,000/µL) and an additional 24 Inc., Norcross, GA), a commercially marketed kit for percent achieved a partial remission. indirect platelet antibody tests, concluded that the method was not useful for establishing or excluding Pathophysiology the diagnosis of ITP.8 This finding is consistent with the American Society of Hematology’s practice Harrington’s “thrombocytopenic factor” guideline, which concluded that current tests for While splenectomy has been recognized as an platelet antibodies are neither necessary nor effective treatment for patients with ITP since it was appropriate in the evaluation of ITP in children or first described in 1913,4 the modern concept of the adults.8 Recently, serologic reactivity of ITP pathophysiology of ITP began in 1951 when Harrington et al.5 reported “a thrombocytopenic autoantibodies has been further localized to epitopes on α 9 or the amino-terminal portions of both GPIIbα factor” in the blood of patients with ITP. Harrington IIb α 10 11 infused himself and seven other nonthrombocytopenic and GPIII . McMillan et al. developed an subjects with whole blood or plasma from patients immunobead assay to detect ITP antibodies reactive to with “idiopathic” thrombocytopenic purpura and GPIIb/IIIa or GPIb/IX that has a minimum specificity of observed a rapid decrease in most recipients’ platelet 84.4 percent, raising the possibility that the long- counts. One patient’s plasma was fractioned by Cohn’s awaited laboratory diagnostic assay for ITP may be method and the thrombocytopenic effect was localized forthcoming. to the globulin fraction. This observation might have γ justified changing the name from “idiopathic” to Polymorphic Fc R macrophage receptors “immune” thrombocytopenic purpura more than 50 In ITP, autoantibody-coated platelets are removed years ago, but the designation “idiopathic” persists in from the circulation by Fcγ receptors on mononuclear common usage, even today. macrophages, principally in the spleen. Among the three families of Fcγ receptors that have been Platelet autoantibodies recognized (FcγRI, FcγRII, and FcγRIII), there is Autoantibodies that react with platelet membrane considerable diversity of structure and function.12 glycoproteins have been demonstrated in as many as FcγRI has a strong affinity for monomeric IgG. FcγRII 80 percent of patients with ITP.6 These antibodies and FcγRIII bind IgG only when it presents in an α β γ react with glycoproteins IIb/IIIa ( IIb 3 integrin or . Fc RII is polymorphic and two CD41/CD61), GPIb/IX, and others. As a consequence, codominant alleles have been described, FcγRIIA-H131 autoantibody-coated platelets are phagocytized in the and FcγRIIA-R131.12 In one study, the distribution of reticuloendothelial system (RES) via Fcγ receptors on FcγRII receptors was skewed toward the FcγRIIA-R131 mononuclear macrophages. Technically, the name , suggesting that polymorphisms “reticuloendothelial system” is a misnomer, with may have a role in the pathophysiology of ITP or may recognition that RES mononuclear are be responsible for modulating the immune response.12 neither reticular nor endothelial cells.7 However, the In another study, there was no significant difference in traditional concept that the spleen, liver, and bone the distribution of FcγRIIA genotypes between ITP marrow are functional sites for RES phagocytic activity patients and controls, but there was a correlation is useful and is preserved in this review.Intuitively,one between FcγRIIIA genotypes and the response to

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 113 S.G. SANDLER treatment.13 ITP autoantibodies have been identified in beneficial effect in ITP is a consequence of all IgG subclasses. The most prevalent IgG subclass “overloading and blocking of the reticuloendothelial detected among ITP antibodies is IgG1 (77%),but IgG2, system by IgG catabolism.”18 Fehr et al.19 showed that IgG3, and IgG4 ITP antibodies have been identified.14 increased platelet counts following infusions of IVIG in Further complicating the effort to identify a correlation four patients with ITP correlated with decreased between diagnosis or prognosis for ITP and a specific clearance of IgG-coated RBCs, providing experimental laboratory marker (i.e., antibody specificity, FcγR evidence supporting Imbach’s hypothesis. genotype, or the IgG subclass) is the observation that A single infusion of IVIG (1 gm/kg) is likely to be some patients with ITP appear to have oligoclonal highly effective (> 85%) for increasing the platelet platelet antibodies, whereas other patients have count by at least 30,000/µL in a previously untreated .14 child or adult with ITP. Typically, the beneficial effect will last for 2 to 4 weeks, when a repeat infusion is Selected Treatments for ITP needed to sustain the beneficial effect. However, if In many persons with ITP, particularly children, thrombocytopenia persists, the inconvenience of long thrombocytopenia resolves without specific treatment infusions (3 to 5 hours),higher cost compared to IV Rh after a benign course lasting only a few days to a few immune globulin (see below), and acute side effects weeks. Other persons may have sustained thrombocy- (headache, flulike symptoms) often cause patients and topenia with life-threatening hemorrhages and a physicians to consider alternative treatment programs. treatment-resistant clinical course lasting several years. For readers of Immunohematology with an interest in Anti-D (IV Rh immune globulin) reading more than selected aspects of the treatment for In 1983, Salama et al.20 reported that microgram ITP, I recommend the March 2002 issue of Blood doses of anti-D (IV Rh immune globulin, IV RhIG) in D+ Reviews, which contains 21 articles focused primarily patients with ITP increased platelet counts in ITP to on clinical management.6 The following brief levels comparable to those reported after infusing gram comments are intended to address selected aspects of doses of IVIG. Whereas IVIG-induced Fcγ receptor the treatment of ITP that have special relevance to block results from random IgG molecules binding to immunohematologists and blood bank serologists. mononuclear macrophages’ Fcγ receptors, IV RhIG- The goal of current treatments for ITP is to induced Fcγ receptor block in D+ patients leverages the decrease the synthesis of the platelet autoantibodies larger and more numerous anti-D-coated D+ RBCs to and/or decrease the rate of splenic destruction of compete with smaller and fewer autoantibody-coated autoantibody-coated platelets. platelets for Fcγ receptor binding. Assuming a representative platelet count of 10,000/µL and a normal Splenectomy RBC count of 5,000,000/µL in a patient with ITP, the The rationale for splenectomy in ITP is supported competition of approximately 500 anti-D-coated D+ by studies that show elimination of both the white RBCs versus one autoantibody-coated platelet (500:1) pulp (immune function) and red pulp (phagocytic favors of the RBCs and “RES blockade” of function) are beneficial.4 Following splenectomy in platelet sequestration and destruction.21 The efficacy of ITP, antiplatelet antibodies decrease.15 This finding is IV RhIG for increasing platelet counts in ITP has been not unexpected, since cultured splenic cells isolated confirmed by several studies (summarized in reference from ITP patients synthesize IgG platelet antibodies.16 22). In North America, treatment of patients with ITP Also, splenectomy removes those mononuclear with IV RhIG is increasingly popular compared to IVIG, macrophages most likely to be responsible for because of the availability of an FDA-approved IV Rh destroying ITP autoantibody-coated platelets. immune globulin (WinRho SDF, Nabi, Boca Raton, FL), and its lower cost, ease of administration, and fewer IVIG acute side effects.23, 24 WinRho is neither effective nor A single infusion of a standard dose of IVIG (1 FDA-approved for treatment of ITP in D– patients. D– gram/kg) causes broad perturbations of both humoral patients with ITP (approximately 15% of Caucasians are and cellular immune function.17 However, the D–) may be treated with IVIG. Rarely, recipients of a preponderance of evidence supports Imbach’s original standard dose have experienced acute intravascular explanation that the principal mechanism of IVIG’s hemolysis,with hemoglobinemia and hemoglobinuria.25

114 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: thrombocytopenic purpura

No specific factor has been identified that predicts an a consequence, rituximab is used increasingly “off acute hemolytic reaction,nor is there an explanation for label” to treat adults28 and children29 with chronic or the variable clinical responses observed after treatment refractory ITP. Reports of sustained remissions in many with IV RhIG, raising the possibility that WinRho may patients, together with the relatively selective B-cell have a variable avidity for different Rh phenotypes, immunosuppression, suggest that rituximab is likely to reflecting the number of D antigen sites on the RBC have an increasingly important role in the future membrane (dosage effect).26 This hypothesis was management of ITP. investigated using WinRho-coated RBCs of selected Rh phenotypes in direct hemagglutination and monocyte Treating Helicobacter pylori infection monolayer assays (MMAs), but neither serologic dosage In 1998, Gasbarrini et al.30 reported that eradication effect nor differentiating reactivity of MMAs for RBCs of Helicobacter pylori (H. pylori) in eight Italian adults with a double dose of D antigen was demonstrated.26 with ITP and H. pylori gastric infections resulted in Readers are reminded that while WinRho is commonly significantly increased platelet counts. This finding was called “anti-D,”it is manufactured from pools of plasma supported subsequently by studies of H.pylori–infected collected from alloimmunized D– persons and different persons with ITP in Italy and Japan.31–34 In contrast, in a production lots may contain, in addition to anti-D, carefully controlled study of H. pylori–infected ITP variable concentrations of anti-E, -C, -G and/or other patients in New York, only one of 15 patients whose H. alloantibodes.27 When WinRho is infused intravenously pylori infection had been eradicated experienced an to treat ITP in a D+ patient, the IV bolus (50–75 µg/kg) increased platelet count, and that response was is approximately 10 times the conventional transient.35 Similarly, in a study in Spain, no significant intramuscular dose used for Rh immunoprophylaxis in improvements in platelet counts were observed in 56 D– women (300 µg, total dose). Therefore, serologic patients with chronic ITP after eradication of H. pylori testing shortly after an infusion may detect multiple infection.36 The discrepancy in treatment outcomes passively infused blood group alloantibodies, as well as remains unexplained and has resulted in divergent the expected positive DAT.27 The presence of anti-D in recommendations concerning the role of screening for the plasma of D+ ITP patients after an infusion of H. pylori infection in patients with ITP. As one might WinRho often raises the question, “Should D+ or D– anticipate, the authors of the study in Spain concluded RBCs be selected for transfusion, if required?” In this that “there is insufficient evidence to include a search author’s opinion, the decision should be based on the for H. pylori in the initial work-up of ITP patients.”36 immediate clinical priority. For example, if RBCs are Responding to this opinion, the authors of one of the required to treat anemia secondary to metrorrhagia in a studies in Italy wrote that “the data acquired to date still woman with newly diagnosed ITP who remains prompt us to consider the investigation and eradication thrombocytopenic, I favor transfusing D+ RBCs to react of H. pylori in ITP patients as a simple, inexpensive tool with any circulating WinRho and contribute to the in early management of the chronic disease. Even immediate goal of increasing the platelet count. In though the percentage of patients who are responsive contrast, if a patient is anemic, but no longer to eradication treatment will be small, it should be thrombocytopenic after an infusion of WinRho, I favor mandatory [italics added] for the physicians to avoid transfusing D– RBCs to support the immediate goal of the toxicity and discomfort that accompany long-term correcting the anemia. prednisone and other immunosuppressive treatments, also spenectomy.”37 In an editorial reviewing these Rituximab diverse findings, McCrae concluded, “Should patients Rituximab (Rituxan,Genentech,Inc.,San Francisco, with ITP be routinely screened for H. pylori? At this CA; and Biogen Idec, Inc., Cambridge, MA) is a point, probably not.”38 humanized IgG1/κ monoclonal anti-CD20 that is For immunohematologists, a more intriguing indicated for the treatment of patients with relapsed or question may be whether antibodies to H. pylori refractory, low-grade or follicular, CD20-positive B-cell crossreact with antigens on platelets, causing ITP-like non-Hodgkin lymphoma. Typically, an infusion of immune destruction. As one hypothesis to explain the rituximab is followed by a rapid decrease in B- inconsistent association of ITP and H. pylori infection, lymphocytes with less myelosuppression than usually Michel et al.35 suggested that “the expression of various is observed with other immunosuppressive agents. As Lewis (Le) antigens by H. pylori isolates, and the

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 115 S.G. SANDLER subsequent production of anti-Le antibodies, could play thrombocytopenic purpura bind primarily to α a role in ITP pathogenesis since platelets may adsorb epitopes on IIb. Blood 2001;97:2171-2. Lewis antigens from the plasma.” In fact, there are 10. Kuwana M, Kaburaki J, Kitasato H, et al. significant data to support this hypothesis, since Leb Immunodominant epitopes on glycoproteins IIb- functions as a receptor for H. pylori in the gastric IIIa recognized by autoreactive T cells in patients mucosa39–41 and Leb may be adsorbed from plasma by with immune thrombocytopenic purpura. Blood platelet membranes.42 At the 1982 annual meeting of the 2001;98:130-9. American Association of Blood Banks, McGinniss 11. McMillan R,Wang L,Tani P. Prospective evaluation reviewed the “ubiquitous nature of human blood group of the immunobead assay for the diagnosis of adult antigens” that are shared between RBCs, bacteria, chronic immune thrombocytopenic purpura (ITP). viruses, and parasites.43 The relationship between Lewis J Thromb Haemost 2003;1:485-91. antigens, RBCs, gastric mucosa, platelets, and H. pylori 12. Williams Y,Lynch S, McCann S, et al. Correlation of was not recognized at that time. However, the H. platelet FcγRIIA polymorphism in refractory pylori–ITP issue would appear to present yet one more idiopathic (immune) thrombocytopenic purpura. example of antigens that are shared among human blood Br J Haematol 1998;101:779-82. cells and the diverse microbial agents that infect us. 13. Fujimoto TT, Inoue M, Shimomura T, Fujimura K. Involvement of Fc gamma receptor polymorphism References in the therapeutic response of idiopathic 1. Nugent DJ. Childhood immune thrombocytopenic thrombocytopenic purpura. Br J Haematol 2001; purpura. Blood Rev 2002;16:27-9. 115:125-30. 2. Kuhne T, Buchanan GR, Zimmerman S, et al. A 14. Chan H, Moore JC, Finch CN, et al. The IgG prospective comparative study of 2540 infants and subclasses of platelet-associated autoantibodies children with newly diagnosed idiopathic detected against platelet glycoproteins IIb/IIIa in thrombocytopenic purpura (ITP) from the patients with idiopathic thrombocytopenic International Childhood ITP Study Group. J Pediatr purpura. Br J Haematol 2003;122:818-24. 2003;143:605-8. 15. Dixon R, Rosse W,Ebbert L. Quantitative detection 3. Neylon AJ, Saunders PW, Howard MR, et al. of antibody in idiopathic thrombocytopenic Clinically significant newly presenting autoimmune purpura. N Engl J Med 1975;292:230-6. thrombocytopenic purpura in adults: a prospective 16. McMillan R, Longmire R, Yelenosky R, et al. study of a population-based cohort of 245 patients. Immunoglobulin synthesis in vitro by splenic tissue Br J Haematol 2003;122:966-74. in idiopathic thrombocytopenic purpura. N Engl J 4. Sandler SG. The spleen and splenectomy in Med 1972;286:681-4. immune (idiopathic) thrombocytopenic purpura. 17. Zimmerman SA, Malinoski FJ, Ware RE. Semin Hematol 2000;37(suppl 1):10-2. Immunologic effects of anti-D (WinRho-SD) in 5. Harrington WJ, Minnich V,Hollingsworth JW,et al. children with immune thrombocytopenic purpura. Demonstration of a thrombocytopenic factor in the Am J Hematol 1998;57:131-8. blood of patients with thrombocytopenic purpura. 18. Imbach P,Barandum E, d’Apuzzo V,et al. High-dose J Lab Clin Med 1951;115:636-45. intravenous gammaglobulin for idiopathic 6. McFarland J. Pathophysiology of platelet thrombocytopenic purpura in childhood. Lancet destruction in immune (idiopathic) thrombocyto- 1981;1:1228-30. penic purpura. Blood Rev 2002;16:1-2. 19. Fehr J,Hofmann V,Kappeler U. Transient reversal of 7. Horsewood P,Keton JG. Macrophage-mediated cell thrombocytopenia in idiopathic thrombocytopenic destruction. In: Garratty G, ed. Immunobiology of purpura by high-dose intravenous gammaglobulin. transfusion medicine. New York: Marcel Dekker, N Engl J Med 1982;306:1254-8. 1994: 435-64. 20. Salama A, Mueller-Eckhardt C, Kiefel V. Effect of 8. Raife TJ, Olson JD, Lentz SR. Platelet antibody intravenous immunoglobulin in immune thrombo- testing in idiopathic thrombocytopenic purpura cytopenia. Competitive inhibition of reticulo- (letter). Blood 1997;89:112-4. endothelial system function by sequestration of 9. McMillan R, Lopez-Dee J, Loftus JC. Autoantibodies autologous red blood cells? Lancet 1983;2:193-5. to α β in patients with chronic immune 21. Savasman CM, Sandler SG. Serologic aspects of IIb 3 treating immune thrombocytopenic purpura using

116 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Review: thrombocytopenic purpura

intravenous Rh immune globulin. secondary autoimmune thrombocytopenic Immunohematology 2001;17:106-10. purpura. Br J Haematol 2002;118:584-8. 22. Sandler SG. Intravenous Rh immune globulin for 34. Hashino S, Mori A, Suzuki S, et al. Platelet recovery treating immune thrombocytopenic purpura. Curr in patients with idiopathic thrombocytopenic Opin Hematol 2001;8:417-20. purpura after eradication of Helicobacter pylori. 23. Sandler SG. Treating immune thrombocytopenic Int J Hematol 2003;77:188-91. purpura and preventing Rh alloimmunization using 35. Michel M, Cooper N, Jean C, et al. Does intravenous Rho (D) immune globulin. Trans Med Helicobacter pylori initiate or perpetuate immune Rev 2001;15:67-76. thrombocytopenic purpura? Blood 2004;103: 24. Sandler SG,Novak SC,Roland B.The cost of treating 890-6. immune thrombocytopenic purpura using intra- 36. Jarque I,Andreu R,Llopis I,et al.Absence of platelet venous Rh immune globulin versus intravenous response after eradication of Helicobacter pylori immune globulin.Am J Hematol 2000;63:245-7. infection in patients with chronic idiopathic 25. Gaines AR. Acute onset hemoglobinemia and/or thrombocytopenic purpura. Br J Haematol 2001; hemoglobinuria and sequelae following Rho(D) 115:1002-3. immune globulin intravenous administration in 37. Emilia G, Luppi M, Morselli M, et al. Helicobacter immune thrombocytopenic purpura patients. pylori infection and idiopathic thrombocytopenic Blood 2000;95:2523-9. purpura (correspondence). Br J Haematol 118: 26. Sandler SG, Mallory D, Trimble J, Nance ST. 1198-9. Intravenous anti-D treatment of immune thrombo- 38. McCrae KR. Helicobacter pylori and ITP: many cytopenic purpura (letter). Blood 1998;91:2624-5. questions, few answers. Blood 2004;103:752-3. 27. Rushin J, Rumsey DH, Ewing CA, Sandler SG. 39. Ilver D, Arnqvist A, Ögren J, et al. Helicobacter Detection of multiple passively acquired alloanti- pylori adhesin binding fucosylated histo-blood bodies following infusions of IV Rh immune group antigens revealed by retagging. Science globulin.Transfusion 2000;40:551-4. 1998;270: 373-7. 28. Stasi R, Pagano A, Stipa E, Amadori S. Rituximab 40. Heneghen MA, Moran AP,Feeley KM, et al. Effect of chimeric anti-CD20 monoclonal antibody host Lewis and ABO blood group antigen treatment for adults with chronic idiopathic expression on Helicobacter pylori colonisation thrombocytopenic purpura. Blood 2001;98:952-7. density and the consequent inflammatory 29. Bengtson KL, Skinner MA,Ware RE. Successful use response. FEMS Immunol Med Microbiol 1998;20: of anti-CD20 (rituximab) in severe, life-threatening 257-66. childhood immune thrombocytopenia. J Pediatr 41. Gerhard M, Lehn N, Neumayer N, et al. Clinical 2003;143:670-3. relevance of the Helicobacter pylori gene for 30. Gasbarrini A, Franceschi F, Tatagilone R, et al. blood-group antigen-binding adhesin. PNAS 96: Regression of autoimmune thrombocytopenia after 12778-83. eradication of Helicobacter pylori. Lancet 1998; 42. Dunstan RA, Simpson MB. Heterogeneous 352-78. distribution of antigens on human platelets 31. Emilia G, Longo G, Luppi M, et al. Helicobacter demonstrated by fluorescence flow cytometry.Br J pylori eradication can induce platelet recovery in Haematol 1985;61:603-9. idiopathic thrombocytopenic purpura. Blood 43. McGinniss M. The ubiquitous nature of human 2001;97:812-4. blood group antigens as evidenced by bacterial, 32. Veneri D, Franchini M, Gottardi M, et al. Efficacy of viral and parasitic infections. In: Garratty G, ed. Helicobacter pylori eradication in raising platelet Blood group antigens and disease. Arlington, VA: count in adult patients with idiopathic thrombo- American Association of Blood Banks, 1983: 25-43. cytopenic purpura. Br J Haematol 2002;118: 1198-9. S. Gerald Sandler, MD, Professor, Medicine and 33. Kohda K, Kuga T, Kogawa K, et al. Effect of Pathology, and Director, Transfusion Medicine, Helicobacter pylori eradication on platelet Department of Laboratory Medicine, Georgetown recovery in Japanese patients with chronic University Hospital, 3800 Reservoir Road, NW, idiopathic thrombocytopenic purpura and Washington, DC 20007

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 117 The sensitivity, specificity, and clinical relevance of gel versus tube DATs in the clinical immunology laboratory

N. PAZ,D.ITZHAKY, AND M.H. ELLIS

The DAT is a test used to demonstrate in vivo antibody and/or DAT is an essential test in the evaluation of autoimmune complement coating of RBCs. Typically, the DAT is performed in hemolytic anemia (AIHA). The DAT also has importance test tubes; however, recently a number of commercially available tests using gel-filled microtubes have become available. Few data in the blood bank, where it is performed as part of the comparing the sensitivity of these test media are available. To evaluation of hemolytic transfusion reactions. compare the rate of detection of a positive DAT performed in test Typically, the DAT is performed in test tubes; tubes versus in gel-filled microtubes and to assess the clinical however, recently a number of commercially available significance of the results in patients undergoing evaluation of anemia, we tested 310 consecutive EDTA-anticoagulated blood tests using gel-filled microtubes have become samples from adult patients. The samples were analyzed using both available.2 Only limited data comparing the sensitivity the conventional tube technique and a gel-based assay (DiaMed®; of these test media and their clinical relevance are Cressier sur Morat, Switzerland). Test results were expressed as 3,4 either positive or negative. When a positive result by either available. In this study, we compare the rate of technique was encountered,the treating physician was interviewed detection of a positive DAT performed in test tubes to determine whether the result warranted further patient versus in gel-filled microtubes and assess the clinical investigation or treatment. In 268 out of 310 cases the DAT was significance of the results in patients undergoing negative by both methods. Of the 42 patients with a positive DAT, the test was positive by both methods in 18 patients. In the evaluation of anemia. remaining 24 cases the DAT was positive by the gel test only. In all cases positive by both techniques the test result affected patient management. Of the 24 cases that were positive only by gel test, 3 Materials and Methods were judged to be clinically significant. In this study, the gel test Patients and samples was more sensitive than the tube technique for performance of the DAT. However, the clinical significance of a DAT positive only by a We studied 310 consecutive blood samples sent to gel test is doubtful. We believe that use of the gel-based DAT should our hospital’s Clinical Immunology laboratory for the be more extensively evaluated before it is adopted as a standard DAT. The samples were collected in EDTA-containing technique in general clinical laboratory practice. tubes (Becton Dickinson and Co., UK) and were Immunohematology 2004;20:118–121. analyzed, using both the conventional tube technique Key Words: gel-based DAT test, tube DAT technique, and a gel-based assay, on the day of collection. clinical significance of a positive DAT DAT—tube technique The DAT is used to demonstrate in vivo antibody DATs were performed by the classic tube and/or complement coating of RBCs. The principle of technique using polyspecific antihuman globulin this test was first demonstrated in 1908 when rabbit (AHG) serum (anti-IgG + anti-C3d; Gamma Biologicals, erythrocytes, sensitized by goat anti-rabbit serum, were Houston,Texas). In this technique, RBCs were washed stongly agglutinated by the subsequent addition of × 3 using NaCl 0.9% and resuspended to a 3–5% saline rabbit anti-goat serum. However, it was not until 1945, suspension. Two drops of polyspecific AHG were when Coombs showed that blood group antibodies added to 1 drop of the washed cells, centrifuged for 1 could be demonstrated in serum or on sensitized RBCs, minute at 900 × g, and examined macroscopically for that the test was adopted by clinical laboratories.1 The agglutination. Positive samples were further tested

118 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Gel versus tube DATs with monospecific anti-IgG (Gamma Biologicals) and Table 1. Number of positive and negative DATs (using polyspecific AHG) anti-IgA, -IgM, -C3c, and -C3d (DiaMed AG, Cressier sur tests by tube and gel testing Morat, Switzerland). Agglutination reactions were Tube tests graded strongly positive (4+ and 3+), moderately Positive Negative Total positive (2+ and 1+), and weakly positive (W+), Positive 18 21 39 according to the manufacturer’s directions. Gel tests Negative 3 268 271 Total 21 289 310 DAT—gel technique Gel tests were performed by adding 50 µL of a 0.8% Of the 18 samples that were positive using both RBC suspension in LISS (ID-Diluent 2) to the top of techniques, the IgG test only was positive in 11, while each microtube in a LISS/Coombs ID card (DiaMed in 7 samples both IgG and C3 were found on the RBCs. AG). The cards were centrifuged at 910 rpm for 10 No samples were positive for C3 only. Identical results minutes, using the ID-Centrifuge 24S. All positive were obtained in tube and gel tests. Of the 21 samples results (presence of agglutinated RBCs in the gel positive using gel testing only,20 were positive for IgG matrix) were re-examined using rabbit monospecific and 1 was positive for C3 only. anti-IgG, -IgM, and -C3d (DC-screening monospecific The sensitivity of the gel test compared to the tube Coombs sera, DiaMed AG). Negative reactions test using a polyspecific AHG reagent was 85 percent appeared as a discrete cell button at the base of the and its specificity was 93 percent. When a monoclonal column. IgG reagent was used, the sensitivity and specificity were 95 percent and 93 percent, respectively. The Further investigation of positive DAT tests positive predictive value (PPV) of a positive gel test When a positive DAT result was obtained by either using the polyspecific reagent was only 46 percent the tube or the gel technique, the treating physician while the negative predictive value (NPV) was 99 was interviewed by one of the investigators to percent. For the monoclonal IgG reagent, the PPV and determine the patient’s diagnosis and whether the NPV were 47 percent and 99 percent, respectively. result warranted further patient investigation or treatment. In addition, laboratory data pertinent to Clinical features of patients with a positive DAT possible hemolysis were abstracted from the patients’ The clinical diagnoses of the patients with a charts (serum bilirubin, lactate dehydrogenase [LDH], positive DAT by both gel and tube tests are listed in and, when available, haptoglobin levels and Table 2. Of note is that in patients with a positive DAT reticulocyte count). An isolated elevation of LDH was by tube and gel testing,the clinical diagnosis of most of not considered to be indicative of hemolysis because of the patients is one known to be associated with a its lack of specificity for this diagnosis. positive DAT. Table 2. Clinical diagnoses of patients with positive DATs by both tube Statistical analysis and gel techniques The sensitivity and specificity of the gel technique Number of Patients Diagnosis was calculated, using a result of the tube test, to be 7 Chronic lymphocytic leukemia either truly positive or negative. From these 3 Systemic lupus erythematosus calculations,the positive and negative predictive values 3 Immune thrombocytopenic purpura were determined. 1 Non-Hodgkins lymphoma 1 Monoclonal gammopathy Results 1 Hypersplenism, Hepatitis C positive Three hundred and ten samples were analyzed. Of 2 Unknown these, 268 (86%) were negative by both tube and gel testing and 42 (14%) were postitive by at least one test Laboratory parameters of patients with a positive technique. Of the 42 positive samples, 21 (50%) were DAT positive by gel and negative by tube testing, 3 samples Of the 18 patients with a positive DAT by tube and (7%) were positive by tube testing and negative by gel gel testing, 8 (44%) had at least 1 laboratory marker of testing, and 18 (43%) were positive by both tube and hemolysis. However, of the 21 patients with a positive gel testing (Table 1). gel test only, 2 patients (9%) had positive hemolytic

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 119 N. PAZ ET AL. parameters while none of the patients with a tube-test- therapy based on the test result. Our results show that only DAT had abnormal markers of hemolysis. of 42 samples positive by either technique using polyspecific antihuman globulin, 18 (43%) were Effect of a positive DAT on further patient positive by tube and gel testing, while 21 (50%) were investigation or treatment positive by gel testing only. Three samples (17%) were In 94 percent of the 18 patients with a positive positive by tube testing and negative by the gel test. DAT by tube and gel testing, the result elicited further Thus the sensitivity of the gel test was 86 percent. diagnostic or therapeutic action, while an isolated When monospecific reagents (anti-IgG and anti-C3) positive gel test led to further diagnostic or therapeutic were used, the sensitivity increased to 95 percent for activity in only 9 percent of cases. None of the patients IgG and decreased to 78 percent for C3. These results with a tube-test-only DAT had further tests performed are similar to those of Tissot et al.,12 except that in their nor received any treatment based on the result of the study the sensitivity of the gel test using anti-C3 was test. only 16 percent. This may be significant because it is known that in patients with AIHA, hemolysis is more Discussion severe when both C3 and IgG are present on the RBCs. The use of gel-based microcolumn tests was Our study has demonstrated that the gel test is introduced into the blood bank laboratory at the end of highly specific compared to the tube test: 93 percent the 1980s for blood typing, antibody detection, and using polyspecific and anti-IgG reagents, and 99 DATs.5,6 Gel tests have grown in popularity because of percent using anti-C3. The results are concordant with 12 their increased accuracy and ease of use compared to those of Tissot et al., but differ from those of 3 classic tube tests. Furthermore, gel tests require Nathanlang et al. The latter group compared gel and smaller samples of RBCs and serum for testing, a tube DAT tests in newborns with suspected fetal- distinct advantage when testing newborn and maternal ABO incompatibility and in adults with premature infants. Gel test systems also decrease the known AIHA and found a sensitivity of 94 percent and exposure of laboratory staff to potentially hazardous a specificity of 85 percent. When only the adults in the blood samples and breakable glassware. Finally,the test study are considered, the sensitivity and specificity result obtained using gel test cards remains stable for were 100 percent and 80 percent, respectively. These up to 48 hours and can thus be saved for comparison differences may be accounted for by the fact that or consultation after the test has been performed. Nathanlang et al. studied a different patient population A number of studies have been published than ours: all of their adult patients had known AIHA demonstrating the increased sensitivity of the gel test and thus would be more likely to have a positive DAT compared to tube testing in detecting RBC by any technique, compared to our patients, who were alloantibodies in patients’ sera.7,8 This increased undergoing an evaluation for anemia whose cause was sensitivity permits the detection of alloantibodies that undetermined at the time of DAT testing. This would otherwise have remained undiagnosed, and has difference in patient population also explains the been shown to have definite clinical value in lower PPV of the gel test in our study (46%) compared preventing potential hemolytic transfusion reactions in to 94 percent among the adult patients in their study. a number of cases.9,10 When we analyzed the clinical relevance of the Gel testing for the DAT compared to tube testing positive DAT, we noted that in 17 of the 18 patients has been studied to a lesser extent, and the clinical (94%) with a positive DAT by both techniques, the relevance of a positive DAT by gel test is unknown.3,4,11 result impacted clinical decision making. By contrast, We performed the current study to determine the in only 2 of the 21 patients (9%) with a positive DAT by sensitivity, specificity, and positive and negative gel testing only did the result lead to further relevant predictive values of one gel test (DiaMed) compared to clinical activity. A similar trend was found when traditional tube testing, which until now has been laboratory markers of hemolysis were examined in considered the standard test system. We also sought to patients with a positive DAT. When the DAT was assess the immediate clinical relevance of a DAT positive by both techniques, hemolytic parameters positive by either technique by determining the were observed in 74 percent of patients, while only 9 treating physicians’ response to the test result in terms percent of patients with a gel-only positive DAT had of ordering further diagnostic tests or instituting other laboratory evidence of hemolysis.

120 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Gel versus tube DATs

These results suggest that the gel test is more 6. Reis KJ, Chachowski R, Cupido A. Column sensitive than the conventional tube test for agglutination technology: the antiglobulin test. performance of the DAT. The relatively low number of Transfusion 1993;33:639-43. patients with a gel-only positive test having clinical or 7. Dittmar K, Procter JL, Cipolone K, Njoroge JM, laboratory evidence for hemolysis suggests a high Miller J, Stroncek DF. Comparison of DATs using falsely positive rate for the gel test. However, we traditional tube agglutination to gel column and cannot exclude the possibility that some patients in affinity column procedures. Transfusion 2001;41: our study who had a positive gel test may develop 1258-62. hemolytic anemia or another disease related to a 8. Judd WJ, Steiner EA, Knafl PC, Masters C. The gel positive DAT in the future. Furthermore, in this study test: use in the identification of unexpected we did not take into account the strength of antibodies to blood group antigens. agglutination obtained in the gel test. This was Immunohematology 1998;14:59-62. because, at the time that this study was performed, our 9. Weisbach V, Ziener R, Zimmerman A, Glaser J, hospital’s immunology laboratory reported DAT test Zingsam R, Eckstein R. Comparison of the results as “positive”or “negative”and we postulated that performance of four microtube column treating physicians were likely to make clinical agglutination systems in the detection of red cell judgements on this basis without considering the alloantibodies. Transfusion 1999;39:1045-50. strength of a positive test result. 10. Champagne K,Spruell P,Chen L,Voll A,Schlanser G. In conclusion, we believe that aspects of the gel- Moulds M. Comparison of affinity column tech- based DAT, such as the correlation of strength of nology and LISS tube tests. Immunohematology agglutination with evidence for hemolysis, should be 1999;15:149-51. more extensively evaluated before it is adopted as a 11. Lai M, Rumi C, D’Onofrio G, Voso MT, Leone G. standard technique in place of conventional tube Clinically significant autoimmune hemolytic testing in general clinical laboratory practice. anemia with a negative direct antiglobulin test by routine tube test and positive by column aggluti- References nation method. Immunohematology 2002;18: 1. Coombs RR. Historical note: past, present and 109-13. future of the antiglobulin test. Vox Sang 12. Tissot JD, Kiener C, Burand B, Schneider P. The 1998;74:67-73. direct antiglobulin test: still a place for the tube 2. Lapierre Y,Rigal D,Adam J, et al. The gel test: a new technique? Vox Sang 1999;77:223-6. way to detect red cell antigen-antibody reactions. Na’ama Paz, MD, Blood Bank, Meir Hospital, Israel; Transfusion 1990;30:109-13. Dganit Itzhaky, PhD, Immunohematology 3. Nathanlang O, Chuansumrit A, Prayoonwitat W, Laboratory, Meir Hospital; and Martin H. Ellis, MD, Siripoonya P,Sriphasial T. Comparison between the Blood Bank, Meir Hospital, Kjar Saba 44281, Israel, conventional tube technique and the gel technique and Sackler School of Medicine, Tel Aviv University, in direct antiglobulin tests. Vox Sang 1997;72:169- Tel Aviv, Israel (corresponding author). 71. 4. Moulds JM, Diekman L, Wells TD. Evaluation of column technology for direct antiglobulin testing. Immunohematology 1998;14:146-8. 5. Cate JC and Reilly N. Evaluation and implementation of the gel test for indirect antiglobulin testing in a community hospital laboratory. Arch Pathol Lab Med 1999;123:693-7.

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 121 COMMUNICATIONS

Letter to the Editors

An unusual antibody reacting with pre-diluted 0.8% reagent RBCs and with 0.8% older (aged) RBCs prepared at the time of testing

A 58-year-old, nontransfused, group O male Since this case was referred to us as a possible presented at hospital for pretransfusion testing prior to antibody to a high-frequency antigen, we prepared a surgery. The preoperative screens and initial panel selected panel of RBCs negative for high-frequency were performed using a gel test and reagent RBCs pre- antigens from our liquid inventory. We prepared the diluted with LISS to 0.8% by the manufacturer (Ortho- RBCs as 0.8% suspensions in Diluent 2 and tested the Clinical Diagnostics, Inc., Raritan, NJ). All cells tested, RBCs with the patient’s serum immediately by the gel except the autocontrol, were 2+ positive. The patient’s test. To this point, the only reactive RBCs had been DAT was negative. The hospital had also performed those that had been pre-diluted by the manufacturer. tube LISS crossmatches and screens but observed no We now had some 2+ reactions and some negative reactivity. results. We then observed that the only reactive RBCs The sample was referred to our red cell reference were greater than 3 months old at the time of testing. laboratory for investigation of a potential antibody to a We then tested older (aged) examples of previously high-frequency antigen reacting only by gel test. nonreactive RBCs, side by side, in the gel cards and As is our standard testing protocol, the patient’s found that the older examples now gave us up to 2+ serum sample was tested with a three-cell screen plus reactions. The DAT on these older RBCs was negative an autocontrol using the following: LISS-antiglobulin using gel-IgG cards. (ImmuAdd; Immucor Inc., Norcross, GA), ficin We then decided to “age” some reagent RBCs at antiglobulin technique (GammaZyme-F; Gamma 37°C overnight in a water bath and test them the next Biologicals Inc., Houston,TX), and a gel test using RBCs day. The so-called “aged” cells gave 2+ reactions and diluted at the time of testing with Diluent 2 their “fresh” counterparts were nonreactive. (MicroTyping Systems Inc., Pompano Beach, FL). The patient’s serum and ten random male donor No reactivity was observed with any of the test sera were tested against two 0.8% pre-diluted, DAT- methods. This excluded clinically significant negative screening cells by the gel technique. The alloantibodies to common RBC antigens and antibodies patient’s serum reacted 2+ with both reagent RBCs and to high-frequency antigens reacting by routine all ten of the donor sera were nonreactive. methods currently used by our laboratory. These same donor sera were then tested in parallel The hospital had tested the sample using a gel with the patient’s serum against DAT-negative RBCs method and pre-diluted RBC suspensions. Since our from one donor. The single-source RBCs were tested center was also using Ortho’s pooled pre-diluted 0.8% “fresh,” i.e., 2 weeks from expiration, 3 months RBCs for high-volume donor screening, we tested this postexpiration, and 6 months postexpiration. The patient’s serum against these pooled RBCs in gel and three RBC samples were diluted to 0.8% suspensions also observed a 2+ reaction. Another lot number of the and tested immediately using the gel test. The ten pooled pre-diluted cells was available and the 2+ donor sera were nonreactive with all cells tested but reaction was reproduced. This latter example was the patient’s serum reacted as expected. No reactivity washed (× 3) free of diluent with 0.9% normal saline was observed with the “fresh” RBCs, a 1+ reaction was and retested. There was no change in reactivity. This observed with the 3-month aged RBCs, and a 2+ demonstrated that the reactivity was neither diluent- reaction with the 6-month aged RBCs. dependent nor caused by something that could be The conclusion was that a constituent of the washed from the RBC surface. diluent used by Ortho to predilute their RBCs was

122 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 COMMUNICATIONS CONT’D producing an immediate effect on the RBC surface that understand the advantages and disadvantages of was detected by this patient’s serum, an effect not different antibody screening techniques. apparent with other commercially prepared RBCs until they were aged during storage. This effect was Jeff Trimble, BSc, ART(CSMLS), CLS (NCA) demonstrable by gel test. Reference Laboratory Supervisor Although the unusual antibody observed in this Michigan Community Blood Centers case had no relevance to treatment of the patient, it is 1036 Fuller Ave., NE prudent to share abnormal findings with our PO Box 1704 colleagues. Sharing such results enables us to better Grand Rapids, MI 49501-1704

Attention: State Blood Bank Meeting Organizers If you are planning a state meeting and would like copies of Immunohematology for distribution, please contact Mary McGinniss, Managing Editor, 4 months in advance, by phone or fax at (301) 299-7443.

Attention SBB and BB Students: You are eligible for a free 1-year subscription to Immunohematology. Ask your education supervisor to submit the name and complete address for each student and the inclusive dates of the training period to Immunohematology, P. O. Box 40325, Philadelphia, PA 19106.

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 123 COMMUNICATIONS CONT’D

Letter to the Editor-in-Chief

Looking back at the history of the journal

It’s hard to believe that 20 years have passed since The straightforward presentations of serologic case you asked me to write the first article for the very first studies make it an invaluable resource! The actual issue of your new journal, Vol. 1, No. 1, 1984, serologic characteristics of certain antibodies reported Immunohematology. Each year, you and your staff in your journal has made it a convenient resource for have managed to produce a prodigious series of blood bank technologists to recognize unusual journals at a price which was, and still is, affordable to antibodies. In addition, your willingness to guide and all! The journal has covered a wide range of topics, all nurture first-time authors has made the journal a great practical, timely, and important to the field of success! immunohematology. In addition, the journal has Even after 20 years,Immunohematology remains a attracted an editorial board composed of individuals, well-balanced, timeless journal, designed for easy well known in the field, who have donated their time reading,and infused with a richness of tradition! Thank and expertise in making it a quality peer-reviewed you for a wonderful journal under your leadership! publication. As the needs of blood bank technologists, The commitment and dedication of all those who have physicians, and educators have changed to reflect the contributed to this journal have made this publication advances and redesign of transfusion medicine, so has well acknowledged and admired by everyone in the the focus of the journal kept pace over the past 20 field. I would like to personally thank you and applaud years. the journal for its many years of service to the Immunohematology has come a long way and has profession. become a favorite journal of all blood bankers! It was the first of its kind to provide current and timely Denise Harmening, PhD, MT(ASCP) articles on practical issues for blood bank serologists Department of Medical & Research Technology and educators. In the beginning, and now, the journal University of Maryland School of Medicine continues to provide wonderful review articles, Baltimore, MD 21201 original scientific studies, unusual serologic cases, and literature reviews.

IMPORTANT NOTICE ABOUT MANUSCRIPTS FOR IMMUNOHEMATOLOGY Mail all manuscripts (original and 2 copies) to Mary H. McGinniss, Managing Editor, 10262 Arizona Circle, Bethesda, MD 20817. In addition, please e-mail a copy to Marge Manigly at [email protected]

124 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 COMMUNICATIONS CONT’D

Letters From the Editor-in-Chief

The second 20th anniversary issue!

This second issue of Immunohematology truly recognized experts in the field who have written celebrates 20 years of publishing by bringing you four reviews for the reader that will be of great value for a great reviews, one original paper, and one scientific long time to come. letter to the editor. The guest editor of this Platelet In this issue you will also find a letter from Dr. Review Issue is Dr. Scott Murphy, medical director of Denise Harmening who looks back at the history of the the Penn-Jersey Red Cross in Philadelphia, journal. She was the first author in the first issue of the Pennsylvania, and senior sedical editor of journal, published 20 years ago. Immunohematology. Dr. Murphy has invited

My sincere apology

In the last issue I neglected to explain that tology at the New York Blood Center and Technical Christine Lomas-Francis, MSc, guest editor of the first Editor of Immunohematology. I hope the readers and 20th anniversary issue of Immunohematology, is the Christine will accept my sincere apology. technical director of the laboratory of immunohema-

Memories from the past!

In this issue and for the rest of the issues in 2004, answer to the crossword puzzle at the end of this issue. we are publishing excerpts from early issues of the Red The Red Cell Free Press was the precursor of Cell Free Press. See Special Section. They represent Immunohematology, started in 1984 by Sandy Ellisor, some of the interesting case studies, methods, and fun Marion Reid, and Helen Glidden at the American Red things, such as a crossword puzzle, that were published Cross. It was designed to communicate and interact in the newsletter in the late 1970s. You will find the with the ARC Reference Laboratory staff.

What is next?

The September issue of Immunohematology will Arndt and Sandy Nance; and “Evaluating patients with be a very exciting issue and one to look forward to. immune hemolysis,”by Dr. Larry Petz. The September Sandra Nance will be the guest editor and the topics of issue will be outstanding! the invited reviews will be: “Drug-induced hemolytic Remember, we are always ready for more papers anemia: an update,” by Dr. George Garratty; “What do from original studies. you do when all units are incompatible: clinical aspects,”by Dr. Geralyn Meny;“What do you do when Delores Mallory all units are incompatible: serologic aspects,” by Pat Editor-in-Chief

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 125 SPECIAL SECTION

Lan2 and Gonsowski1 Obtaining Cells from a Clot for Absorption, Elution, etc. We were referred a blood sample on a patient who was admitted to a hospital for TUR surgery. The blood Although we request serum and an anticoagulated center suspected the patient’s serum contained an specimen, hospitals often send just a clotted specimen. antibody to a high frequency antigen. In our To obtain sufficient cells for absorption, elutions, etc., laboratory, the routine antibody workup was we have found the following procedure to be useful: performed and a battery of high incidence negative • Cut the top off the filter of a Recipient Set. We cells tested. Both Lan negative and Gn(a–) cells were use Fenwal HB-98D. Discard the drip chamber found non-reactive with this patient’s serum, whereas and attached tubing. all other cells were strongly incompatible in the antiglobulin phase of testing. After resolving that this •Pour the clot into the filter chamber, with the patient’s antibody was anti-Lan and that he was Lan spike sitting in a 15 cc tube. negative, we investigated the Gn(a–) red blood cells • Mash the clot around the filter with applicator and serum from Mrs. Gonsowski. sticks and squirt saline to wash as many cells as Red cells from Mrs. Gonsowski drawn on two possible into the tube. different occasions were tested with several examples •Wash the cells until free of hemolysis. of anti-Lan. No reactivity occurred. Mrs. Gonsowski’s serum, which was known to contain anti-D as well as We think this is great for salvaging a lot of cells with a high incidence antibody, did not agglutinate three minimum amount of trouble. Rh (D) negative Lan negative red cell samples. Mrs. o Beverly Pohl Gonsowski’s serum was then absorbed onto a random Reprinted as published in the Red Cell Free Press group O, Rh negative (rr) red cell sample and an eluate 1978;3(No.4):5. prepared from the absorbing cells. Lan specificity was once again obtained. A fresh blood sample on Mrs. Gonsowski was kindly sent to us for repeat testing. The above results were duplicated, and other investigators have also verified these results. We are therefore noting that anti- Gna, which was reported earlier as a new antibody independent of other blood group systems, is really anti-Lan and that Gn(a–) cells are Lan negative.

1. Fox, J.A.;Taswell, H.F. Anti-Gna,a new antibody reacting with a high-incidence erythrocytic antigen. Transfusion 9:265-269; 1969. 2. Grindon,A.J.; McGinniss, M.H.; Issitt, P.D.; Reihart, J.K.;Allen, F.H. A second example of anti-Lan. Vox Sang. 15:293-296; 1968. Rosia E. Nesbitt Reprinted as published in the Red Cell Free Press 1979;4(No.1):7.

126 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Crossword Puzzle Number 4

123456789 46. An animal used to produce antihuman antibodies. 47. An antigen which non-specifically stimulates 10 11 12 lymphocytes. 13 14 15 DOWN 16 17 18 19 20 2. Antibody reacting only with cells possessing I and 21 22 23 B antigens. 3. A procedure for demonstrating immunoglobulin 24 on the surface of B cells. 25 26 27 28 29 4. Symbol for the Duffy blood group system. 5. The site used to obtain blood samples from the 30 31 32 immediate neonate. 33 34 35 6. Another symbol for HD or Pr. 7. Antibody and complex antigen described by 36 37 38 39 40 Rosenfield et al.Vox Sang. 9:415 (1964). 41 42 43 44 8. A female gamete. 9. Of no commercial value. 45 11. A type of polyagglutination. 46 47 12. The process of separating antibody from its antigen. 13. Bilirubin is ______to albumin for transport ACROSS within the body. 1. Ouchterlony technique is an agar gel ______17. Low frequency antigen in the MN system. method. 18. First half of the name for the biphasic hemolysin 10. White cell responsible for humoral found in patients with PCH. (2 words). 20. The Langereis or Soa blood group system. 14. Initials for term synonomous to resumé. 26. This can affect antigen and antibody strength. 15. Antibody in serum of Claas and McLeod. 27. Rhw1 16. Antigen complex absent or weak on ABH ,A1B null 28. Rheumatoid arthritis. cells and Oi cells. 29. Spleen, liver, et al. 19. K10 antigen. 31. Isotonic, low ionic strength solution used in 21. An abnormal respiratory sound heard in immunohematology. auscultation. 32. A system of adjacent genes on a chromosome of 22. Complement component bound before C3 in which one gene, the operator gene, controls the classical pathway. activity of the structural genes. 23. AutoAnalyser. 34. Y and X genes control this (X-rated). 24. This form of polyagglutination is specifically 38. Low frequency antigen in the MN blood group agglutinated by S. Scleria. system described in Nature 172:688(1953). 25. Highfalutin Sd antigen. 39. A relative of gastropod mollusk that can be used 28. Unusual, uncommon or low incidence antigen. as a source of anti-A. 30. Serum and plasma mixed together will ______. 40. A major city in the USA. 33. Group A secretor gene. 42. A third generation method to test for HB Ag. 35. Gene controlling secretion of Lewis substance in s 43. Automated reagin test. body fluids. 44. Usual site of venipuncture. 36. Stuart Prower factor. 37. Sc:1 antigen. Marion Reid 39. Systemic lupus erythematosus. Reprinted as published in the 41. RES red cell destruction. Red Cell Free Press 1978;3(No. 2): 4. 45. Immune response genes. For solution, see this issue, page 133.

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 127 BOOK REVIEW

Stem Cells Handbook. Stewart Sell, MD, ed. Humana Chapter 6 covers the cloning of mammalian cells, Press, 2003. ISBN:1-58829-113-8. Hardcover, 499 pages. with a description of the difficulties of the processes as List price: $175. To order call: 973-256-1699, fax: 973- well as the observed outcomes of producing either 256-8341; e-mail: [email protected]. new cloned organisms or sources of cells for repair of the current organism. No scientific topic has been more in the press and In Chapter 8, repair and wound healing are in legislative hearing rooms than stem cells. Behind the described first in the amphibian model, where these news-catching events of the last 5 years,there has been mechanisms are extensive. This is followed by two an amazing amount of research into the mechanisms of examples of human repair systems in the next two action of stem cells: (1) from differentiated tissue, chapters: dermal repair and reconstruction of skin; and where the stem cells tend to repair or repopulate the bone marrow mesenchymal cells for reconstituting organ where they reside; (2) from primitive or mesodermal tissues such as bone, cartilage, muscle, embryonic tissue that has not yet made the journey of tendons,and ligaments,as well as bone marrow stromal differentiation;and (3) from cells that appear to be able cells. Normal and abnormal hematopoietic stem cell to transdifferentiate from the organ where they reside lineages are explored from the starting point of the to other tissues. This phenomenon of plasticity is hematopoietic stem cell. The authors also discuss use under intensive research by investigators studying of the hematopoietic stem cell as a means of rescue, readily available adult stem cells, for example, those in treatment, or both of leukemias and lymphomas. the blood and bone marrow, and their ability to Beginning with Chapter 16, the book takes us first become a stem cell for another organ of the body. through the development of organ systems, then This book begins with a historical perspective on through the development of disease, with suggestions stem cells and walks the reader straight to the for areas of research that may lead to treatment using knowledge revolution of the last few years. It stem cell–based or augmented strategies. The coverage accomplishes this with information instead of passion, of neural, cardiovascular, and hepatic organs is with clear studies and data instead of polarized points particularly informative and captures the extensive of view. This book helps place into context the much- research being done in those areas. publicized and controversial studies highlighted by the All in all, the book gives a very timely review of news media. these varied topics, with up-to-date references. The The first chapter, by the editor, gives an important illustrations provided by the chapter authors are top- overview of the path of differentiation and some notch and very helpful to the understanding of the possible side trips of nature. It also offers clear topics. Although not every reader may wish to go definitions of terms that make the information in the through every page of this reference text,this reviewer, rest of the book more understandable. Each chapter for one, will be going back section by section many author has a depth of knowledge in his or her field, yet times in the years to come. The concise setting of the they are all careful to include examples of practical use assembled knowledge and the insight that the authors of the information,either in further basic research or in give on each of these complex subjects are quite possible treatment applications. valuable in promoting understanding of the topics Methods of growing and isolating embryonic stem described and their potential medical importance. cells are covered in Chapters 2 and 3. Chapter 4 traces the journey from embryonic stem cell to functional, N. Rebecca Haley, MD differentiated tissue architecture. Chapter 5 explains Vice President, StemCo Biomedical, Inc. germ cell interactions and combinations as they move 2810 Meridian Parkway, Suite 148 on their way to becoming embryonic stem cells or Durham, NC 27707 supporting their growth.

128 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 ANNOUNCEMENTS

Monoclonal antibodies available at no cost. The HEMATOLOGÍA HABANA′ 2005–First Laboratory of Immunochemistry at the New York Announcement. The 5th National Congress and Blood Center has developed a wide range of the 7th Latin American Meeting in Hematology, monoclonal antibodies (both murine and humanized) Immunology, and Transfusion Medicine will present a that are useful for screening for antigen–negative scientific program at the International Conference donors and for typing patients’ RBCs with a positive Center, Havana, Cuba, May 16–20, 2005. A preliminary DAT. Monoclonal antibodies available include anti-M, program lists malignant hemopathies, disorders of -Fya, -Fyb, -K, -k, -Kpa, -Jsb, -Dob, -Wrb, and –Rh17. For a RBC membranes, immunotherapy, histocompatibility, complete list of available monoclonal antibodies,please immunohematology, hemolytic disease of the see our Web site at www.nybloodcenter.org/ newborn, and blood components as some of the framesets/FS-4C7.htm. Most of those antibodies are topics. For more information contact: Prof. José M. murine IgG and,thus,require the use of anti-mouse IgG Ballester, President, Organizing Committee, for detection, i.e, anti-K, -k, and -Kpa. Some are directly Hematology Habana′ 2005,Apartado 8070, Ciudad de agglutinating (anti-M, -Wrb, and -Rh17), and a few have la Habana, CP 10800, Cuba. been humanized into the IgM isoform and are directly b a agglutinating (anti-Js and -Fy ). The monoclonal Workshop on Blood Group Genotyping. The antibodies are available at no charge to anyone who ISBT/ICSH Expert Panel in Molecular Biology has requests them. Contact: Marion Reid (mreid@ recommended that a workshop be held on blood nybloodcenter.org) or Gregory Halverson (ghalverson@ group genotyping by molecular techniques. The results nybloodcenter.org), New York Blood Center, 310 East would culminate in a report at the ISBT Congress, July 67th Street, New York, NY 10021. 11–15, 2004, in Edinburgh, Scotland. It was decided that only laboratories that provide a reference service New York State Blood Bank Association annual in blood group genotyping would be included in the meeting. The Blood Banks Association of New York workshop. One of the aims of the workshop would be State, Inc., will hold its 53rd annual meeting June 9–11, to establish an external quality assurance plan. If you 2004, in Albany, NY. The preliminary program includes have any suggestions as to how the workshop should bacterial contamination, stem cells, quality be organized, we would be grateful for your opinions. improvement, cost containment in the transfusion If you are interested in taking part in such a workshop, service setting, technical workshops, an infectious please contact Geoff Daniels (geoff.daniels@nbs. disease update, and a nursing track. Advance registra- nhs.uk). Offer presented by Geoff Daniels, Martin L. tion is encouraged. For more information and Olsson, and Ellen van der Schoot. registration, contact: Kevin Pelletier, MT(ASCP), Blood Banks Association of New York State, Inc., PO Box 38002,Albany, NY 12203-8002; phone: (518) 485-5341; e-mail: [email protected].

Annual Symposium. The National Institutes of Health, Department of Transfusion Medicine, will hold their 24th annual symposium, Immunohematology and Blood Transfusion, on September 23, 2004. The symposium is co-hosted by the Greater Chesapeake and Potomac Region of the American Red Cross and is free of charge. Advance registration is encouraged. For more information and registration, Contact: Karen Byrne, NIH/CC/DTM, Bldg.10/Rm.1C711,10 Center Drive,MSC 1184,Bethesda,MD 20892-1184;e-mail:[email protected];or visit our Web site: www.cc.nih.gov/dtm>education.

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 129 Masters (MSc) in Transfusion and Transplantation Sciences At The University of Bristol, England

Applications are invited from medical or science graduates for the Master of Science (MSc) degree in Transfusion and Transplantation Sciences at the University of Bristol. The course starts in October 2004 and will last for one year. A part-time option lasting three years is also available. There may also be opportunities to continue studies for PhD or MD following MSc. The syllabus is organised jointly by The Bristol Institute for Transfusion Sciences and the University of Bristol, Department of Transplantation Sciences. It includes:

• Scientific principles underlying transfusion and transplantation • Clinical applications of these principles •Practical techniques in transfusion and transplantation •Principles of study design and biostatistics • An original research project

Applications can also be made for Diploma in Transfusion and Transplantation Science or a Certificate in Transfusion and Transplantation Science.

The course is accredited by the Institute of Biomedical Sciences.

Further information can be obtained from the Web site:

http://www.bloodnet.nbs.nhs.uk/ibgrl/MSc/MScHome.htm

For further details and application forms please contact:

Professor Ben Bradley

University of Bristol, Department of Transplantation Sciences

Southmead Hospital, Westbury-on-Trym, Bristol BS10 5NB, England

FAX +44 1179 595 342, TELEPHONE +44 1779 595 455, E-MAIL: [email protected]

130 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 CLASSIFIED AD

SBB Program. The Department of Transfusion Medicine, National Institutes of Health, is accepting applications for its 1-year Specialist in Blood Bank Technology Program. Students are federal employees who work 32 hours per week. This program introduces students to all areas of transfusion medicine, including reference serology,cell processing, and HLA and infectious disease testing. Students also design and conduct a research project. NIH is an Equal Opportunity Employer. Application deadline is June 30, 2004, for the January 2005 class. Contact: Karen M.Byrne,NIH/CC/DTM,Bldg.10/Rm.1C711,10 Center Drive,MSC 1184,Bethesda,MD 20892-1184;phone: (301) 496-8335; or e-mail: [email protected]

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 131 ADVERTISEMENTS

NATIONAL REFERENCE LABORATORY FOR National Platelet Serology Reference BLOOD GROUP SEROLOGY Laboratory

Immunohematology Reference Diagnostic testing for: Laboratory • Neonatal alloimmune thrombocytopenia (NAIT) AABB,ARC, New York State, and CLIA licensed •Posttransfusion purpura (PTP) (215) 451-4901—24-hr. phone number • Refractoriness to platelet transfusion (215) 451-2538—Fax • Heparin-induced thrombocytopenia (HIT) American Rare Donor Program • Alloimmune idiopathic thrombocytopenia (215) 451-4900—24-hr. phone number purpura (AITP) (215) 451-2538—Fax Medical consultation available [email protected] Test methods: Immunohematology • GTI systems tests (215) 451-4902—Phone, business hours —detection of glycoprotein-specific platelet (215) 451-2538—Fax antibodies [email protected] —detection of heparin-induced antibodies (PF4 Quality Control of Cryoprecipitated-AHF ELISA) (215) 451-4903—Phone, business hours • Platelet suspension immunofluorescence test (215) 451-2538—Fax (PSIFT) • Solid phase red cell adherence (SPRCA) assay • Monoclonal antibody immobilization of platelet Granulocyte Antibody Detection and Typing antigens (MAIPA)

• Specializing in granulocyte antibody detection For information, e-mail: [email protected] and granulocyte antigen typing or call: •Patients with granulocytopenia can be classified through the following tests for proper therapy Maryann Keashen-Schnell and monitoring: (215) 451-4041 office —Granulocyte agglutination (GA) (215) 451-4205 laboratory —Granulocyte immunofluorescence (GIF) —Monoclonal Antibody Immobilization of Sandra Nance Granulocyte Antigens (MAIGA) (215) 451-4362 For information regarding services, call Gail Eiber at: (651) 291-6797, e-mail: [email protected], Scott Murphy, MD or write to: (215) 451-4877 Neutrophil Serology Reference Laboratory American Red Cross Blood Services American Red Cross St. Paul Regional Blood Services Musser Blood Center 100 South Robert Street 700 Spring Garden Street St. Paul, MN 55107 Philadelphia, PA 19123-3594

CLIA LICENSED CLIA LICENSED

132 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 ADVERTISEMENTS CONT’D

IgA/Anti-IgA Testing Reference and Consultation Services

IgA and anti-IgA testing is available to do the Red cell antibody identification and problem following: resolution • Monitor known IgA-deficient patients HLA-A, B, C, and DR typing •Investigate anaphylactic reactions • Confirm IgA-deficient donors HLA-disease association typing Our ELISA assay for IgA detects antigen to Paternity testing/DNA 0.05 mg/dL. For information regarding our services, contact For information on charges and sample Zahra Mehdizadehkashi at (503) 280-0210, or requirements, call (215) 451-4351, e-mail: [email protected], write to: or write to: Pacific Northwest Regional Blood Services American Red Cross Blood Services ATTENTION: Tissue Typing Laboratory Musser Blood Center American Red Cross 700 Spring Garden Street 3131 North Vancouver Philadelphia, PA 19123-3594 ATTN: Ann Church Portland, OR 97227

CLIA LICENSED CLIA LICENSED, ASHI ACCREDITED

IMMUNOHEMATOLOGY IS ON THE WEB! Answers to Crossword Puzzle Number 4

www.redcross.org/pubs/immuno D IFFU S ION Password “2000” BLYMPHOCYTE C U B C VKL For more information or to send an e-mail message “To the editor” O O I H Y D UL N R A L E T WO AA [email protected] J E I E N TN U S C A D RARE Notice to Readers: All articles published, G C O A G UALTEO including communications and book reviews, A S E L E I H P reflect the opinions of the authors and do not necessarily reflect the official policy of the T E N SM SLE American Red Cross. E X T RAVASCULAR D I R R UO Notice to Readers: Immunohematology, Journal COAT M ITOGEN of Blood Group Serology and Education, is Reprinted as published in the Red Cell Free Press printed on acid-free paper. 1978;3(No. 3)1.

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 133 Immunohematology

JOURNAL OF BLOOD GROUP SEROLOGY AND EDUCATION Instructions for Authors

SCIENTIFIC ARTICLES, REVIEWS, AND CASE REPORTS 5. Acknowledgments Before submitting a manuscript, consult current issues of Acknowledge those who have made substantial contributions to Immunohematology for style. Type the manuscript on white bond the study, including secretarial assistance; list any grants. paper (8.5" × 11") and double-space throughout. Number the pages 6. References consecutively in the upper right-hand corner, beginning with the A.In text, use superscript, arabic numbers. title page. Each component of the manuscript must start on a new B. Number references consecutively in the order they occur in page in the following order: the text. 1. Title page C.Use inclusive pages of cited references, e.g., 1431–7. 2. Abstract D.Refer to current issues of Immunohematology for style. 3. Text 7. Tables 4. Acknowledgments A.Head each with a brief title, capitalize first letter of first word 5. References (e.g., Table 1. Results of ...), and use no punctuation at the end 6. Author information of the title. 7. Tables—see 6 under Preparation B. Use short headings for each column needed and capitalize first 8. Figures—see 7 under Preparation letter of first word. Omit vertical lines. C.Place explanations in footnotes (sequence: *, †, ‡, §, ¶, **, ††). Preparation of manuscripts 8. Figures 1. Title page A.Figures can be submitted either by e-mail or as photographs A.Full title of manuscript with only first letter of first word (5″×7″ glossy). capitalized (bold title) B. Place caption for a figure on a separate page (e.g.,Fig.1.Results B. Initials and last name of each author (no degrees; all CAPS), of ...), ending with a period. If figure is submitted as a glossy, e.g., M.T.JONES and J.H. BROWN place first author’s name and figure number on back of each C.Running title of ≤ 40 characters, including spaces glossy submitted. D.3 to 10 key words C.When plotting points on a figure, use the following symbols if 2. Abstract possible: ● ● ▲ ▲ ■ ■. A.One paragraph, no longer than 300 words 9. Author information B. Purpose, methods, findings, and conclusions of study A.List first name, middle initial, last name, highest academic 3. Key words—list under abstract degree, position held, institution and department, and 4. Text (serial pages) complete address (including zip code) for all authors. List Most manuscripts can usually,but not necessarily,be divided into country when applicable. sections (as described below). Results of surveys and review papers are examples that may need individualized sections. SCIENTIFIC ARTICLES AND CASE REPORTS SUBMITTED A.Introduction AS LETTERS TO THE EDITOR Purpose and rationale for study, including pertinent back- ground references. Preparation B. Case Report (if study calls for one) 1. Heading—To the Editor: Clinical and/or hematologic data and background serology. 2. Under heading—title with first letter capitalized. C.Materials and Methods 3. Text—write in letter format (paragraphs). Selection and number of subjects, samples, items, etc. studied 4. Author(s)—type flush right; for first author: name, degree, and description of appropriate controls, procedures, methods, institution, address (including city, state, ZIP code, and country); equipment, reagents, etc. Equipment and reagents should be for other authors: name, degree, institution, city, and state. identified in parentheses by model or lot and manufacturer’s 5. References—limited to ten. name, city, and state. Do not use patients’ names or hospital 6. One table and/or figure allowed. numbers. D.Results Presentation of concise and sequential results, referring to perti- nent tables and/or figures, if applicable. Send all submissions (original and two copies) to: E. Discussion Mary H. McGinniss, Managing Editor, Immunohematology, Implications and limitations of the study,links to other studies; 10262 Arizona Circle, Bethesda, MD 20817 and e-mail your if appropriate,link conclusions to purpose of study as stated in manuscript to Marge Manigly at [email protected] introduction.

134 IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 Becoming a Specialist in Blood Banking (SBB)

What is a certified Specialist in Blood Banking (SBB)? • Someone with educational and work experience qualifications who successfully passes the American Society for Clinical Pathology (ASCP) board of registry (BOR) examination for the Specialist in Blood Banking. •This person will have advanced knowledge, skills, and abilities in the field of transfusion medicine and blood banking.

Individuals who have an SBB certification serve in many areas of transfusion medicine: • Serve as regulatory, technical, procedural and research advisors •Perform and direct administrative functions •Develop, validate, implement, and perform laboratory procedures • Analyze quality issues preparing and implementing corrective actions to prevent and document issues • Design and present educational programs •Provide technical and scientific training in blood transfusion medicine • Conduct research in transfusion medicine Who are SBBs? Supervisors of Transfusion Services Managers of Blood Centers LIS Coordinators Educators Supervisors of Reference Laboratories Research Scientists Consumer Safety Officers Quality Assurance Officers Technical Representatives Reference Lab Specialist Why be an SBB? Professional growth Job placement Job satisfaction Career advancement How does one become an SBB? • Attend a CAAHEP-accredited Specialist in Blood Bank Technology Program OR • Sit for the examination based on criteria established by ASCP for education and experience Fact #1: In recent years, the average SBB exam pass rate is only 38%. Fact #2: In recent years, greater than 73% of people who graduate from CAAHEP-accredited programs pass the SBB exam. Conclusion: The BEST route for obtaining an SBB certification is to attend a CAAHEP-accredited Specialist in Blood Bank Technology Program

Contact the following programs for more information: PROGRAM CONTACT NAME CONTACT INFORMATION Walter Reed Army Medical Center William Turcan 202-782-6210; [email protected] Transfusion Medicine Center at Florida Blood Services Marjorie Doty 727-568-5433 x 1514; [email protected] Univ. of Illinois at Chicago Veronica Lewis 312-996-6721; [email protected] Medical Center of Louisiana Karen Kirkley 504-903-2466; [email protected] NIH Clinical Center Department of Transfusion Medicine Karen Byrne 301-496-8335; [email protected] Johns Hopkins Hospital Christine Beritela 410-955-6580; [email protected] ARC-Central OH Region, OSU Medical Center Joanne Kosanke 614-253-2740 x 2270; [email protected] Hoxworth Blood Center/Univ. of Cincinnati Medical Center Catherine Beiting 513-558-1275; [email protected] Gulf Coast School of Blood Bank Technology Clare Wong 713-791-6201; [email protected] Univ. of Texas SW Medical Center Barbara Laird-Fryer 214-648-1785; [email protected] Univ. of Texas Medical Branch at Galveston Janet Vincent 409-772-4866; [email protected] Univ. of Texas Health Science Center at San Antonio Bonnie Fodermaier SBB Program: 210-358-2807, Linda Smith [email protected] MS Program: 210-567-8869; [email protected] Blood Center of Southeastern Wisconsin Lynne LeMense 414-937-6403; [email protected] Additional information can be found by visiting the following Web sites: www.ascp.org, www.caahep.org, and www.aabb.org

IMMUNOHEMATOLOGY, VOLUME 20, NUMBER 2, 2004 135