The decision of two surgical teams to repair an infrarenal aneurysm with a straight or bifurcation graft in 94 patients was investigated. Nine patients had an emergency operation and the two who died within 30 days of surgery (mortality rate 2%) had a ruptured aneurysm. Of 88 patients with follow-up data of at least 1 year (mean 3.5 years), 46 received bifurcation grafts and 42 tube grafts. The mean transverse diameter of the aneurysm on ultrasonography was 6.5 cm for straight and 6.1 cm for bifurcation grafts. The mean duration of operation was longer for bifurcation than straight grafts (4.2 versus 3.3h, P< 0.05) and intraoperative blood loss greater for the former (2292 versus 1350 ml, P < 0.05). The incidences of late complications for the two grafts were not different. The only secondary aneurysm repair was a staged procedure on a descending thoracic aortic aneurysm. No postoperative iliac aneurysms were detected by routine physical examination and pelvic ultrasonography in patients receiving straight grafts. Even though there was no preselection of patients by disease status, one surgical team relying on preoperative arteriography successfully replaced 28 of 39 aneurysms (72%) with tube grafts while the other team using primarily intraoperative assessment placed straight grafts in only 14 of 49 patients (29%). In conclusion, straight grafts can be implanted in approximately two-thirds of patients undergoing aortic aneurysm repair with significantly less operating time and blood loss and without increase in the incidence of late iliac occlusion or dilatation. Preoperative arteriography and computed tomography facilitate this decision by identifying occlusive disease or iliac dilatation.