Objective: Compared with open repair of abdominal aortic aneurysms (AAA), endovascular repair (EVAR) is associated with decreased perioperative morbidity and mortality in a standard patient population. This study sought to determine if the advantage of EVAR extends to patients aged <90 years. Methods: This was a retrospective review from a prospectively maintained computerized database. Of the 322 patients aged <80 treated with EVAR from January 1997 to November 2007, 24 (1.9%) were aged <90. Mean age was 91.5 ± 1.5 years (range, 90-95 years), and 83.3% were men. Mean aneurysm size was 6.8 cm (range, 5.2-8.7 cm). Results: Mean procedural blood loss was 490 mL (range, 100-4150 mL), and 20.8% required an intraoperative transfusion. Mean postoperative length of stay was 6.0 days, (median, 4 days; mode, 1 day; range, 1-42 days), with 33.3% of patients discharged on the first postoperative day. Amongst the 24 patients, there were 6 (25.0%) perioperative major adverse events, and 2 patients died, for a perioperative mortality rate of 8.3%. Mean follow-up was 20.5 months (range, 1-49 months). Overall, three patients (12.5%) required a secondary intervention, comprising thrombectomy, angioplasty, and proximal cuff extension. No patients required conversion to open repair. Two patients (8.3%) died of AAA rupture at 507 and 1254 days. Freedom from all-cause mortality was 83.3% at 1 year and 19.3% at 5 years. Freedom from aneurysm-related mortality was 87.5% at 1 year and 73.2% at 5 years. Endoleak occurred in five patients (20.8%), with three type I and two of indeterminate type; of these, two patients with type I endoleak underwent secondary intervention at 153 and 489 days after EVAR, of which one case was successful. Conclusion: Our study supports that EVAR in nonagenarians is associated with acceptable procedural success and perioperative morbidity and mortality. The medium-term results suggest that EVAR may be of limited benefit in very carefully selected patients who are aged <90 years.