When performing an aortobifemoral bypass, the surgeon may choose between an end-to-end or end-to-side aortic anastomosis. Although several authors have claimed the superiority of the former, controversy still exists. The aim of this study was to compare the early and late results of these two techniques in patients for which both procedures were possible. Of 158 patients, aortic anastomosis was performed end-to-end in 47 (group I) and end-to-side in 111 (group II). Both groups were similar in all other aspects. The type of proximal anastomosis did not affect the rate of perioperative mortality or early thrombosis. There were no secondary aortic aneurysms or aortoenteric fistulae in either group. Actuarial primary (without reoperation) and secondary (after thrombectomy) five year patency rates were 90.2 and 98.9 per cent in group I, and 90.8 and 98.5 per cent in group II, respectively. Five-year limb survival was 95.3 and 95.7 per cent, respectively. As we could not find any difference between the results in the two groups, we suggest choosing the simplest procedure which maintains adequate pelvic and colonic blood supply, according to angiographic findings.