OBJECTIVE Because of widespread use of leukocyte reduction in platelet concentrates (PCs) and the need to store such concentrates, we investigated the effects of leukocyte depletion on the quality of stored PCs. METHODS Ten double-sized PCs were divided into 2 equal units which were tested simultaneously. One half was stored for 5 days after filtration through a polyester filter, the other one was stored unfiltered. RESULTS The volume of the 10 "oversized' PCs was 483 +/- 40 ml (mean +/- standard deviation) and they contained 5.9 +/- 1.5 x 10(11) platelets and 80 +/- 23 x 10(6) leukocytes. Filtration significantly reduced the leukocyte concentration (168 +/- 56/microliter before, 6 +/- 4 /microliter after filtration) and leukocyte count (39.9 +/- 11.3 x 10(6) vs. 1.3 +/- 0.9 x 10(6); p < 0.0005). Filtration caused a platelet loss of 16%, the platelet count decreasing not significantly from 2.91 +/- 0.75 x 10(11) to 2.40 +/- 0.94 x 10(11) (p = 0.26). After 5 days of storage all parameters of platelet function (platelet aggregation to several stimuli, hypotonic shock reaction [HSR] and platelet retraction), mean platelet volume, and pH and pCO2 showed no advantage for PCs filtered prior to storage compared to PCs stored unfiltered. Moreover, platelet aggregation on day 5 using 4 agonists at 10 concentrations showed worse results in 4 assays in prestorage filtered PCs (collagen [4 micrograms/ml: p < 0.05, ADP [0.2 mM]: p < 0.05, ADP [0.3 mM]: p < 0.05, thrombin [0.6 E/ml]: p < 0.05). But there is no convincing trend in all aggregation tests, and HSR, presumably the most useful parameter, was not different or day 5. CONCLUSIONS There is no advantage in terms of improved quality for prestorage leuko-depletion of PCs. Taking into account the obvious disadvantages of filtration, such as platelet loss and increasing costs per transfusion, we conclude that pre- or post-storage filtration of single-donor PCs should be done only for patients who have a clear indication for the transfusion of leukocyte-poor blood products.