On the basis of clinical empirical observations and classic theory of platelet aggregation, microvascular surgeons have held that platelet aggregation is more responsible for arterial thrombosis and fibrin clot formation is more responsible for venous thrombosis. This theory was tested in two models of thrombosis, one created in arteries and the other in veins, using systemic antithrombotic agents. The models were pair matched in rats, creating a crush-avulsion injury to the femoral artery on one leg and a crushing injury to the femoral vein on the contralateral leg. There were four treatment groups (n = 15 per group): systemic heparin, aspirin and dipyridamole, both sets of agents, and dual controls (vehicle treated). One-day patency rates in the controls were 13.4% and 0% for the arterial and venous models, respectively. Heparin treatment caused an increase in arterial patency to 73.3% but had no effect on venous patency (6.7%). Aspirin and dipyridamole had a modest effect on venous patency (40%) but had no effect on arterial patency (6.7%). Administration of both sets of agents led to 100% and 86.7% patency in the arterial and venous models, respectively. Patency rates were maintained at the 7-day evaluation in the arteries; in contrast, all thrombosed veins were found to be patent 7 days postoperatively (100% patency in all groups). These findings are in contrast to classic thinking about microvascular thrombosis, but this reverse effect may be unique to the rat. The high rate of recanalization in thrombosed veins indicates a need for caution in performing experimental studies of patency in rat veins.