Apheresis platelet transfusions: does ABO matter?

More than 100 years ago, landmark reports by Landsteiner and others related the serologic causes of acute hemolytic transfusion reactions to antigens and antibodies we now know as the ABO system. These discoveries opened the door to safe blood transfusion. Naturally occurring ABO antibodies of the immunoglobulin (Ig) M class form in antigen-negative individuals, making it imperative to identify the ABO type of the transfusion recipient. Transfusion of relatively small amounts of incompatible red blood cells (RBCs) can lead to severe intravascular hemolysis, kidney failure, vascular collapse, and even death. Concern over these drastic consequences has led our discipline to adopt stringent policies to assure ABO compatibility for transfusion. This concern extends to the transfusion of plasma products, although most adult patients can tolerate transfusion of hundreds of milliliters of incompatible plasma. Mismatched ABO antibodies become diluted throughout the blood volume and are inhibited by soluble ABO antigenic determinants and wide tissue distribution of ABO glycoconjugates.1 Owing to their smaller blood volumes, this is not considered a safe practice for pediatric transfusion recipients. The neutralizing ability in adults is saturable and can be overcome by large amounts of mismatched plasma. Donors can also become hyperimmune and deliver a much larger amount of mismatched ABO antibodies of both IgM and IgG classes, which can overcome this neutralizing ability. Good transfusion practice mandates that transfusion of fresh frozen plasma (FFP) be ABO compatible. Platelet (PLT) concentrates are suspended in plasma, and it would appear that these concerns about ABO compatibility should also apply. Nevertheless, the short storage life of PLTs and the often urgent nature of the need for PLT transfusion has led to common practices that override ABO compatibility in PLT transfusion for a significant number of patients. When PLT availability is curtailed through inventory shortages or through economic pressure to avoid outdating older products, the majority of PLT transfusions may be ABO mismatched. Adverse consequences of this practice have been said to be infrequent, attributed to dilution and the neutralizing capacity, as mentioned above. Although whole-blood-derived PLT concentrates are suspended in 50 to 60 mL of plasma and contribute only a fraction of the total plasma in a pooled PLT transfusion, apheresis PLTs provide 250 to 400 mL of plasma from a single donor. This can magnify the effect of a high-titer isoagglutinin in a donor's plasma. This issue of TRANSFUSION includes an article by Josephson and colleagues,2 which addresses one approach to the problem of hemolytic transfusion reactions because of high-titer isoagglutinins in incompatible apheresis PLT transfusions. The authors encountered two hemolytic transfusion reactions traced to apheresis donors with high-titer isoagglutinins during a period when half of the apheresis PLT transfusions were ABO mismatched in their hospital. They call attention to the numerous case reports of adult recipients of apheresis PLTs who experienced acute intravascular hemolytic reactions owing to mismatched ABO antibodies. In many of these reports, the hemolytic complications are attributed to donors with extraordinarily high-titer isoagglutinins, usually Group O donors transfused into Group A recipients.3 Using a gel technology method for rapid determination of isoagglutinin titer, they sought to identify products with what they consider “critical” titers of isoagglutinins with an eye toward using this technique to screen PLT apheresis products for inventory management before release for transfusion. Their efforts warrant several comments. First, the frequency of hemolytic complications related to mismatched PLT transfusions is poorly documented. Although the literature contains numerous case reports, there are no prospective studies assessing the frequency of the com-plication. Mair and Benson4 retrospectively calculated the risk of hemolysis at their institution based on one identified hemolytic transfusion reaction among 46,176 PLT transfusions of which 21 percent of the transfusions were plasma mismatched. They calculated a frequency of 1 hemolytic event in 9000 mismatched PLT trans-fusions. These data seem to support the common practice of largely ignoring ABO compatibility for PLT transfusions. Another issue that arises involves the lack of consensus regarding the serologic aspects of the problem. There are little data to define the level of a “critical” titer for either IgM or IgG ABO antibodies. Hemolytic reactions have been reported involving incompatible plasma with isoagglutinin titers ranging from 16 to 16,384.3 Further complicating the situation is the fact that many studies report ABO antibody titers as a single entity, making no distinction between IgM or IgG reactivity. Consequently, it is difficult to determine the significance of each immunoglobulin class in causing intravascular hemolysis. An analogous situation is seen in the use of intravenous immunoglobulin (IVIG), which has only trace amounts of IgM. Although positive direct antiglobulin tests (DAT) owing to incompatible isoagglutinins are seen in well over half of all recipients of IVIG, episodes of hemolysis are only rarely reported. Nevertheless, whether these positive DATs cause shortened RBC survival through extravascular hemolysis is largely unexplored. Acknowledging these problems, Josephson and coworkers2 attempted to define critical titers for IgM isoagglutinins as greater than or equal to 64 and for IgG isoagglutinins as greater than or equal to 256, based on the few case reports available. In addition to titer, the volume of the product should be considered. Hemolytic reactions may be caused not just by high-titer antibody, but by higher volumes of lower-titer mismatched ABO antibodies. Larger amounts of mismatched plasma with lower titers may also overwhelm the protective factors of dilution and inhibition by delivering a higher total dose of mismatched ABO antibodies. In a similar manner, there may be a cumulative effect of repeated out-of-group PLT transfusions over a short time, even though each product had a low titer. Many transfusion authorities recommend using ABO-mismatched pooled PLTs rather than mismatched apheresis PLTs or tracking the amount of out-of-group plasma received each day, for just this reason. These are attempts at risk stratification, because apheresis PLT products have some advantages over pooled concentrates. Transfusion services are faced with declining resources to manage the increasingly complex task of improving the safety margin of transfusion. Even if there were a true critical titer that could be defined by clinical study, hospital blood banks would need effective and validated techniques to identify those donors. Although Josephson and associates2 showed good correlation between tube technology and the gel method they used in their limited study, others have found less correlation between the techniques. Alternatives to screening products at the hospital before release would be to have blood collection centers perform this before distribution or even to screen and defer high-titer donors from apheresis. These latter options are unattractive; they would impair the availability of apheresis PLTs and would certainly contribute to higher product costs because they would be applied to all products, not just those being considered for out-of-group transfusion. Perhaps the most important consideration we face is the balance between the risk of hemolysis from ABO-mismatched plateletpheresis transfusion against the risk of not correcting thrombocytopenic bleeding through cumbersome inventory management based on this risk. Blood bank practices must take clinical reality into account when policies affecting PLT product availability are instituted. Deciding on PLT transfusion is not a black or white issue, but requires numerous risk-benefit analyses, both overt and implicit. There is tremendous difference between a stable, thrombocytopenic chemotherapy recipient who is scheduled for a prophylactic PLT transfusion and a patient with a subarachnoid hemorrhage and neurologic decompensation who was taking daily aspirin and clopidogrel bisulfate and needs urgent improvement in PLT function. Most of us would not mind waiting for ABO-matched PLTs for the former, but would demand even out-of-group PLTs for the latter if that was what was available that night. In summary, the risk of hemolysis from ABO-mismatched apheresis PLT transfusion is an uncommon but real concern. As we strive for a zero-risk blood supply, transfusion complications that were previously deemed to be relatively uncommon and consequently inconsequential should be reassessed, quantified, and addressed. At present, there are no standards or guidelines for transfusion of mismatched PLTs, and there are no requirements for assessing ABO antibody titer before transfusion of mismatched apheresis PLTs. Many institutions have volume limitations, such that there is a maximum allowable volume of mismatched plasma transfused within a 24-hour period. Obviously, this approach may be protective at some titers, but may be futile at extraordinarily high titers of ABO antibodies. Some institutions are assessing isoagglutinin titers before transfusion and only transfusing products with a titer below a specified level; this is the approach that is proposed by Josephson et al.2 Others advocate the use of plasma removal or concentration, especially if the volume or isoagglutinin titers are excessive. Few blood centers are screening apheresis donors for ABO antibody titers. The true answer to this quandary must lie elsewhere, because none of these practices address all of the causes and risks attendant to this problem. An attractive solution would be to eliminate plasma from all PLT products through the use of artificial PLT storage media. There is hope that one of the candidates recently reviewed5 may some day obviate the need to be concerned about the isoagglutinin burden in any PLT product. This would be the most efficient way of assuring maximal PLT availability by taking plasma compatibility out of the picture. It would still leave the issue of those hyperimmune patients who are refractory to PLT transfusion because of their own isoagglutinins, but that is a matter to discuss another day.