To determine the clinical utility of routine intraoperative completion arteriography, we prospectively evaluated 214 consecutive infrainguinal bypass grafts (209 reversed-vein and 5 polytetrafluoroethylene grafts) performed from July 1987 to August 1991. Visual inspection, pulse palpation, and continuous-wave Doppler examination were performed in all cases. At least 1 completion arteriogram was obtained in 213 cases (99%). The bypasses were to the popliteal artery in 130 cases and to the tibial or pedal arteries in 84 cases. Graft patency was confirmed at 30 days in all patients by ankle-brachial index determinations (greater than 0.2 increase) and duplex scan-derived peak-systolic flow velocities (greater than 45 cm/s). Significant technical problems requiring revision were identified in 18 grafts (8%), including 6% of popliteal grafts and 12% of tibial/pedal grafts. Only three of these problems were suspected by pulse palpation or continuous-wave Doppler examination. The intraoperative angiographic findings leading to revision included distal anastomotic stenoses (n = 6), distal arterial disease requiring sequential bypass (n = 4), mid-graft valvular or branch ligature stenoses (n = 4), distal arterial thrombosis (n = 2), and graft kink or twist (n = 2). Thirty-day primary patency was 99% (129 of 130) for femoropopliteal grafts and 93% (78 of 84) for femorodistal grafts. Secondary patency was 100% (130 of 130) and 96% (81 of 84), respectively. Primary patency was 89% (16 of 18) for those grafts that required intraoperative revision based on arteriographic findings. We conclude that routine completion arteriography is an excellent method of ensuring the intraoperative technical adequacy of infrainguinal bypass. The test is easy to perform, reproducible, and should be considered the "gold standard" for intraoperative bypass assessment. Prior to adopting angioscopy or duplex scanning for intraoperative surveillance, randomized, controlled validation studies against angiography should be performed.