Selection of platelets for refractory patients by HLA matching and prospective crossmatching

A multi-site clinical study compared platelets chosen for refractory patients by prospective platelet crossmatching using stored donor platelets and HLA-based selection. Seventy-three patients who were refractory to random-donor platelets received two plateletpheresis components, one chosen by HLA-based criteria and the other by crossmatching. Patients were carefully evaluated to exclude nonimmune factors that could adversely affect transfusion results. Each of the five study sites used a crossmatch procedure with which it had experience. Results from this study indicate the following: 1) The overall rate of successful transfusion was similar when an HLA-based method of donor selection that includes all grades of matching and mismatching was compared to a crossmatch-based method of donor selection. 2) HLA-based selection that restricts recipients to grade A and BU matches was superior to a selection method based upon crossmatching alone. Donor selection based on HLA matching (grades A or BU) was also superior to selection based on any degree of HLA mismatching (grades BX, C, or D). 3) Selection of donors based on HLA-cross-reactive groups (defined by in vitro serologic crossreactivity) was no more successful than that based on grade C and D mismatches and was no more successful than selection by crossmatching alone. 4) Lymphocytotoxic and platelet antibodies were not detected in many of the enrolled patients, even though patients demonstrating nonimmune factors were eliminated from the study. It can be concluded that HLA-compatible (grades A and BU) platelets provide optimal support for refractory patients, but that crossmatch-selected platelets are acceptable as an alternative component.