Objective: Patients with marginal venous conduit, poor arterial runoff, and prior failed bypass grafts are at high risk for infrainguinal graft occlusion and limb loss. We sought to evaluate the effects of anticoagulation therapy after autogenous vein infrainguinal revascularization on duration of patency, limb salvage rates, and complication rates in this subset of patients. Methods:This randomized prospective trial was performed in a university tertiary care hospital and in a Veterans Affairs Hospital. Fifty-six patients who were at high risk for graft failure were randomized to receive aspirin (24 patients, 27 bypass grafts) or aspirin and warfarin (WAR; 32 patients, 37 bypass grafts). All patients received 325 mg of aspirin each day, and the patients who were randomized to warfarin underwent anticoagulation therapy with heparin immediately after surgery and then were started on warfarin therapy to maintain an international normalized ratio between 2 and 3. Perioperative blood transfusions and complications were compared with the Student t test or with the χ² test. Graft patency rates, limb salvage rates, and survival rates were compared with the Kaplan-Meier method and the log-rank test. Results: Sixty-one of the 64 bypass grafts were performed for rest pain or tissue loss, and 3 were performed for short-distance claudication. There were no differences between the groups in ages, indications, bypass graft types, risk classifications (ie, conduit, runoff, or graft failure), or comorbid conditions (except diabetes mellitus). The cumulative 5-year survival rate was similar between the groups. The incidence rate of postoperative hematoma (32% vs 3.7%; P = .004) was greater in the WAR group, but no differences were seen between the WAR group and the aspirin group in the number of packed red blood cells transfused, in the incidence rate of overall nonhemorrhagic wound complications, or in the overall complication rate (62% vs 52%). The immediate postoperative primary graft patency rates (97.3% vs 85.2%) and limb salvage rates (100% vs 88.9%) were higher in the WAR group as compared with the aspirin group. Furthermore, the cumulative 3-year primary, primary assisted, and secondary patency rates were significantly greater in the WAR group versus the aspirin group (74% vs 51%, P = .04; 77% vs 56%, P = .05; 81% vs 56%, P = .02) and cumulative limb salvage rates were higher in the WAR group (81% vs 31%, P = .01). Conclusions: Perioperative anticoagulation therapy with heparin increases the incidence rate of wound hematomas, but long-term anticoagulation therapy with warfarin improves the patency rate of autogenous vein infrainguinal bypass grafts and the limb salvage rate for patients at high risk for graft failure. (J Vasc Surg 1998;28:446-57.)