Is it safe to omit the 37 degrees C reading from pretransfusion red blood cell antibody detection testing?

The routine pretransfusion red blood cell antibody detection test (PADT), performed at the authors' institution, consists of a three-phase, saline-antiglobulin technique (immediate spin, 30-minute incubation at 37 degrees C, IgG indirect antihuman globulin test [IAT]), and each phase is examined for hemolysis and agglutination. To determine if it would be safe to omit reading the 37 degrees C phase of the PADT, a 6-year retrospective review of records (February 1986 to February 1992) was undertaken. Of approximately 280,000 sera tested for unexpected red cell alloantibodies, 1480 (.53%) were reactive at only 37 degrees C. Of 1480 sera, 1313 contained alloantibodies of no or questionable significance (eg, anti-Le(a), anti-Leb, and anti-P1), 71 sera contained antibodies of undetermined specificity, and 10 sera were reactive because of rouleaux. Eighty-six serum samples from 53 different patients contained alloantibodies of potential significance (anti-K, anti-E, etc). These 86 sera represented approximately 2.2% of all reactive sera that contained potentially significant alloantibodies. Although most antibodies (approximately 94%) detected only at 37 degrees C were of no or questionable significance, the other 6% could have increased the risk of missing a potentially significant antibody from approximately 1 in every 4700 sera tested to 1 in every 1875 sera tested, had they not been detected. The authors suggest that the increased risk of an acute or delayed hemolytic transfusion reaction would be too high to justify a procedural change. Based on these data, the authors continue to read the 37 degrees C phase of the PADT for hemolysis and agglutination.