OBJECTIVE The purpose of this study was to test a transfemoral system of bifurcated endovascular graft insertion for aortic aneurysm repair. METHODS Bifurcated endovascular grafts were inserted through bilateral femoral artery cutdowns in 41 patients. The results were assessed by completion angiography and follow-up computed tomography. RESULTS The second half of the study included more aneurysms 6 cm or larger (p < 0.05) and more instances of short proximal neck (p < 0.05), proximal neck angulation (p < 0.05), and iliac angulation (p < 0.05). Despite the increasingly challenging anatomy, the results were better in the second half of the study as illustrated by the lower overall combined morbidity/mortality rate (15% vs 50%) and higher overall success rate (85% versus 65%). The mortality rate for the series as a whole was 7.5%. Mean follow-up was 18.8 months for the first 20 patients and 10.9 months for the second 20. The commonest complication in the first half of the study was graft thrombosis (n = 5). This complication was absent from the second half of the study because of routine adjunctive stenting. Two patients died of complications of endovascular repair. In both cases aneurysm rupture on the third postoperative day was associated with coagulopathy and angiographic signs of perigraft leak. CONCLUSIONS Aneurysm exclusion with a bifurcated endovascular graft was feasible in a wide range of patients, but when the aneurysm was not entirely excluded from the circulation, the risk of rupture persisted.