BACKGROUND Buffy coats (BCs) are used as an alternative to platelet-rich plasma in the preparation of platelet concentrates (PCs). For this purpose the BCs have to be stored for same time at 20-24 degrees C which implies cellular metabolic activity. However, little information is available concerning the effects of a number of factors which may influence the suitability of the preparation as the source of PC. METHODS We studied the effects on BCs of a high and low gas permeability of the wall of the plastic containers, PL2209 and PL146, respectively, mixing versus non-mixing during storage for 48 h at 22 degrees C, and two types of anticoagulant solutions, CPD and half strength citrate CPD (0.5CPD). The buffy coats were prepared by the bottom and top technique. The median values of volume and haematrocrit were 58-64 ml and 39-45%, respectively. A total of 48 BCs were tested. Blood gases, pH, bicarbonate concentration and haemolysis were determined in the blood mixtures and beta-thromboglobulin (beta-TG), lactate dehydrogenase (LDH), complement factor 3a, and elastase in the extracellular fluid. RESULTS The pH decreased in all units but to a lesser extent in PL2209 containers than in PL146 units. In the former the pCO2 decreased slowly in contrast to the latter where it increased by about 50%. Mixing during storage increased the pH and decreased the pCO2 in 0.5CPD-PL146 and CPD-PL2209 units, as compared to resting, while no effects of mixing were observed in the other groups. The pO2 decreased to low levels in PL146 units. The haemolysis and LDH release were higher in mixed than in unmixed units. The initial beta-TG levels were lowest in 0.5CPD-PL146 units which also had the lowest 24-hour levels. The release of beta-TG during storage was smallest in CPD resting units. The elastase release was significantly higher in 0.5CPD than in CPD units already from the beginning of storage and increased during storage at about the same rate irrespective of mixing. The C3a levels were higher in 0.5CPD-PL2209 units than in the other units at 2 h. Storage for 24 h caused an increase by 2-3 times of the original level without any clear relation to storage conditions. CONCLUSIONS In BC units accumulation of CO2 occurs in containers with low gas permeability. These also show the most rapid pH decrease during storage. Prolonged holding of BCs puts extra emphasis on the need of satisfactory gas permeability of the container for platelet storage in BC-derived PCs. Continuous mixing causes red cell damage and does not seem to have any clear benefit. The release of granulocyte elastase was higher in 0.5CPD than in CPD units but there was no indication of an associated increase in platelet activation. CONCLUSIONS Study of buffy coats stored in various media and containers at 22 degrees C suggests that it is better to restrict storage to 24 h or less to avoid activation or other deleterious effects on the platelets.