HLA-matched platelet transfusions are effective only in refractory patients with positive HLA antibody screening

BACKGROUND Recipients of platelet transfusions with 1-hour corrected count increments (1hCCIs) of 7.5 or less on two subsequent platelet transfusions with random platelets may benefit from human leukocyte antigen (HLA)-matched platelet concentrates. We aimed to quantify the efficacy of HLA-matched platelets concentrates expressed in 1hCCIs. METHODS We performed a cohort study among consecutive refractory patients who received HLA-matched platelet concentrates in the Netherlands between 1994 and 2017. We performed mixed-model linear regression comparing 1hCCIs after HLA split-antigen-matched transfusions with 1hCCIs after HLA-mismatched transfusions, adjusted for within-patient correlations. A donor-to-patient match was categorized as a split-match if all donor HLA-A and -B antigens were present in the patient as well; that is, donor and patient were HLA identical or compatible. Subgroup analyses were performed for patients with positive or negative HLA antibody screens. Finally, the additional effect of ABO mismatches on 1hCCIs was investigated. RESULTS The 1hCCI after an HLA-matched transfusion was 14.09 (95% reference interval, 1.13-29.89). This was 1.94 (95% confidence interval [CI], 0.74-3.15) higher than 1hCCI after HLA-mismatched transfusions. In patients with negative HLA antibody screening tests, HLA matching did not affect 1hCCIs. Conditional on HLA matching, 1hCCIs decreased by 3.70 (95% CI, -5.22 to -2.18) with major ABO mismatches. CONCLUSION Matched platelet concentrates yielded maximal 1hCCIs, whereas mismatched transfusions still resulted in adequate increments. There is no indication for HLA-matched platelets in patients with negative antibody screens.