ABO antibody titers are not predictive of hemolytic reactions due to plasma-incompatible platelet transfusions


William J. Savage, MD, Transfusion Medicine Division, Department of Pathology, Johns Hopkins Medical Institutions, 600 N. Wolfe Street, Carnegie 667, Baltimore, MD 21287; e-mail: wsavage1@jhmi.edu.


BACKGROUND: The overall risk of hemolytic transfusion reactions (HTRs) from plasma (minor)-incompatible platelet (PLT) transfusions and the role of a critical anti-A or anti-B titer in predicting and preventing these reactions has not been clearly established.

STUDY DESIGN AND METHODS: We evaluated all apheresis PLT (AP) transfusions for 3 months. Using the gel titer method, we determined the anti-A and/or the anti-B immunoglobulin (Ig)G titer for all incompatible APs. Reported febrile transfusion reactions and HTRs were recorded; transfusions were not prospectively evaluated by the study team. A posttransfusion direct antiglobulin test (DAT) and eluate were performed after a reported febrile or hemolytic reaction for patients who received plasma-incompatible APs.

RESULTS: A total of 647 of 4288 AP transfusions (15.1%) were plasma incompatible. Group O APs (n = 278) had significantly higher anti-A and anti-B titers than group A or B APs (p < 0.0001). No group A or B APs had a titer of more than 128 (0/342). For group O APs, 73 had titers of 256 or greater (26.3%), and 27 had titers of 512 or greater (9.7%). No HTRs were reported to any plasma-incompatible AP transfusion during the study period. Two plasma-incompatible AP transfusions were associated with fever and chills and positive DATs, of which one had a positive eluate. The incidence of a DAT and eluate-positive febrile transfusion reaction in the plasma-incompatible AP population is 0.15% (95% confidence interval, 0.0%-0.86%).

CONCLUSION: A critical anti-A or -B titer is not sufficient to predict the risk of hemolysis in patients receiving plasma-incompatible APs, although underreporting of reactions to the blood bank may limit the generalizability of this study.