Platelet count, platelet aggregation responses, and plasma beta-thromb
oglobulin were assessed before and at intervals during each study dial
ysis. As expected, modest thrombocytopenia was observed following trea
tment with all dialysers, but Filtral showed the least effect. Concord
antly, modest increases in plasma beta-thromboglobulin were observed w
ith all devices except Filtral (caveat: the device also seems to adsor
b beta-thromboglobulin). Interpatient variability was more striking fo
r platelet aggregation responses, and made it difficult to discern pat
terns; however, aggregation in response to ADP was augmented during ha
emodialysis with the Filtral device, and blunted when the G120 M was e
mployed. Differences between hollow-fibre and parallel plate devices w
ith the same membrane material suggest that membrane geometry, manufac
turing residues, membrane support, or potting materials may also contr
ibute to platelet activation.