The purpose of this single-center study was to compare findings at presentation and surgical outcome in patients in whom abdominal aortic aneurysms (AAAs) ruptured after endovascular repair and patients in whom AAAs ruptured before any treatment, over a defined period. From May 1992 to September 2003, 1043 patients underwent elective repair of intact infrarenal AAAs. Endovascular repair was performed in 609 patients, and open repair in 434 patients. Eighteen of 609 patients (3%) who underwent endovascular AAA repair required treatment because of rupture of the aneurysm after a mean of 29 months (group 1). During the same 11-year period, another 91 patients without previous treatment required urgent repair of a ruptured AAA (group 2). Rupture was diagnosed at contrast material-enhanced computed tomography or by presence of extramural extravasation of blood at open repair. Except for a higher incidence of women in group 2, patients in both groups were similar with regard to demographics and clinical characteristics but differed in findings at presentation. Eight patients in group 1 had a known endoleak before AAA rupture, whereas contrast-enhanced computed tomography, performed in 15 patients at presentation, demonstrated an endoleak in all. Hypotension (systolic blood pressure <100 mm Hg) was noted at presentation in 4 of 18 patients (22%) in group 1 and 76 of 91 patients (84%) in group 2. All patients underwent open repair via a transperitoneal approach, except for 4 patients in group 1 and 3 patients in group 2 who underwent endovascular repair of ruptured AAAs. The proportion of patients with hypotension at presentation in group 1 (4 of 18) was significantly less than in group 2 (76 of 91; P <. 01). The difference in perioperative (30 day) mortality rate in group 1 (3 of 18; 16.6%) compared with group 2 (49 of 91; 53.8%) was also significant (P <. 01). The outcome in group 1 was therefore superior to that in group 2. This study confirms that endovascular AAA repair complicated by endoleak does not prevent rupture. The data suggest, however, that rupture, when it occurs in these circumstances, may not be accompanied by such major hemodynamic changes and high mortality as rupture of an untreated AAA. Further long-term follow-up and analysis in a larger group of patients are required to confirm the apparent intermediate level of protection afforded by failed endovascular repair, which does not prevent rupture but enhances survival after operation to treat rupture, possibly by ameliorating the hemodynamic changes associated with the rupture process.