Ticlopidine, a thienopyridine that prevents the progression of diabetic retinopathy in humans, was recently shown to increase nitric oxide (NO) production in human neutrophils. The thienopyridine clopidogrel has been found to be clinically useful in the secondary prevention of thrombotic events. The aim of the present study was to evaluate the effect of clopidogrel on ischemic retinopathy in streptozotocin-diabetic rats and its influence on prostanoids and NO production. We compared nondiabetic rats and rats after 3 months of diabetes that were given three doses (1, 10 or 20 mg/kg per day p.o.) of ticlopidine or clopidogrel from the first day of diabetes. The variables recorded after 3 months of diabetes were platelet aggregation, thromboxane B(2) (TxB(2)) production, 6-keto-prostaglandin F(1)(alpha) (stable metabolite of prostacyclin), aortic NO, plasma nitrites/nitrates, and the percentage of the retinal surface occupied by horseradish peroxidase (HRP)-permeable vessels. In diabetic rats, platelet aggregation and thromboxane concentration were increased, and prostacyclin, NO and area occupied by HRP-permeable vessels were decreased. Ticlopidine and clopidogrel reduced the maximum extent of platelet aggregation in a dose-dependent manner: maximal inhibition with respect to untreated diabetic rats was 48.6% with ticlopidine and 66.6% with clopidogrel. Ticlopidine reduced thromboxane B(2) only at a dose of 20 mg/kg per day p.o. (47.4% inhibition) and clopidogrel at doses of 10 mg/kg per day (51% inhibition) or 20 mg/kg per day (51.7% inhibition). Aortic prostacyclin production did not change after treatment with either thienopyridine. Treatment with ticlopidine reduced the inhibition of NO production in untreated rats (89.6% inhibition) to 0.9%, and clopidogrel reduced inhibition to 30%. Treatment with ticlopidine or clopidogrel reduced the retinal nonperfused area from 86.8% inhibition in untreated rats to 45.6% and 25.3%, respectively. In conclusion, the early administration of thienopyridines in streptozotocin-diabetic rats partly prevented the appearance of diabetic retinal ischemia.