Patients admitted for surgical removal of varicose veins were treated in a blinded manner for 48 hours prior to surgery with either placebo, low-dose aspirin (25 mg twice daily), dipyridamole (150 mg twice daily) or both. Segments of vein excised at surgery were incubated with or without sodium arachidonate and subsequent prostacyclin (PGI2) production was measured without knowledge of treatment given. During the first 5 minute period of incubation in the presence of arachidonate, veins from dipyridamole-treated patients demonstrated increased (by 75%) arachidonate-stimulated PGI2 production compared to placebo-treated patients. By contrast, PGI2 production was reduced by 64% by aspirin treatment and 67% by aspirin plus dipyridamole compared to placebotreated patients (p=<0.05). In unstimulated vein segments incubated in the absence of arachidonate, spontaneous PGI2 production during the first 5 minute incubation period was increased 32% following dipyridamole treatment but was unchanged following aspirin treatment. By contrast, unstimulated (spontaneous) PGI2 production in patients treated with aspirin plus dipyridamole was reduced by 57% (p=<0.05), compared to both placeboand aspirin-treated patients, and by 71% (p=<0.05) compared to dipyridamole-treated patients. With repeated change of incubation medium, the ability of vein walls to produce PGI2 declined. This exhaustion was not prevented by drug treatment. However, drug effects between patient treatment groups were consistent over successive incubation periods. These results suggest that certain therapeutic benefits that might be achieved by enhancement of PGI2 production from vascular endothelium following dipyridamole treatment may be reduced by simultaneous aspirin treatment. © 1990.