BACKGROUND: Platelet transfusion is universally employed in acute leukemia. Platelet concentrate supernatants contain high concentrations of biologic mediators that might impair immunity. We investigated whether washed platelet and red cell transfusions could improve clinical outcomes in adult patients with acute leukemia. METHODS: A pilot randomized trial of washed, leukoreduced ABO identical transfusions versus leukoreduced ABO identical transfusions was conducted in 43 adult patients with acute myeloid or lymphoid leukemia during 1991-94. Primary endpoints to be evaluated were platelet transfusion refractoriness, infectious and bleeding complications and overall survival. RESULTS: There were no significant differences in infectious or major bleeding complications and only one patient required HLA matched platelet transfusions. Minor bleeding was more frequent in the washed, leukoreduced arm of the study. Confirmed transfusion reactions were more frequent in the leukoreduced arm of the study. Overall survival was superior in the washed arm of the study (40% versus 22% at 5 years), but this difference was not statistically significant (p = 0.36). A planned subset analysis of those </=50 years of age found that those in the washed, leukoreduced arm (n = 12) had a 75% survival at five years compared with 30% in the leukoreduced arm (n = 10) (p = 0.037) CONCLUSION: This study provides the first evidence concerning the safety and efficacy of washed platelets, and also raises the possibility of improved survival. We speculate that transfusion of stored red cell and platelet supernatant may compromise treatment, particularly in younger patients with curable disease. Larger trials will be needed to assess this hypothesis.