OBJECTIVE Repair of thoracovisceral aortic aneurysms (TVAA) after previous open repair of an infrarenal abdominal aortic aneurysm (AAA) poses significant challenges. We sought to better characterize such recurrent aneurysms and to evaluate their operative outcome. METHODS We reviewed the records and radiographs of 49 patients who underwent repair of TVAAs between 1988 and 2002 after previous repair of an AAA. Visceral artery reconstructions were completed with combinations of beveled anastomoses, inclusion patches, and side arm grafts. In 14 patients visceral endarterectomy was required to treat associated occlusive disease. Sixteen patients had cerebrospinal fluid drainage, and 10 patients had distal perfusion during cross-clamping. RESULTS Patient mean age was 72 years, and 80% were men. Fifty-one percent of patients had symptomatic disease, and average TVAA diameter was 6.2 cm. Mean time between AAA and TVAA repair was 77 months. Twenty-six percent of aneurysms were restricted to the lower visceral aortic segment, 35% extended to the diaphragm, another 35% extended to the distal or middle thoracic aorta, and 4% involved the entire remaining visceral and thoracic aorta. The 30-day operative mortality rate was 4.1% in patients with nonruptured aneurysms and 50% in patients with ruptured aneurysms, for an overall mortality rate of 8.2%. Fifteen patients (30.6%) had major morbidity, including paresis in two patients and dialysis-dependent renal failure in five patients. At late follow-up, three patients required further aortic operations to treat additional aneurysms, and four patients had fatal aortic ruptures. Two-year and 5-year cumulative survival rates were 61% (+/-7.5%) and 37% (+/-7.8%), respectively. At univariate analysis, operative blood loss was the sole significant predictor of major morbidity (P <.023), and rupture (P <.030, P <.0001) and aneurysm extent (P <.0007, P <.0001) correlated with both operative death and long-term survival. Only aneurysm extent (P <.010, relative risk 37.3) remained a significant predictor of long-term survival at multivariate analysis. CONCLUSIONS Elective repair of TVAAs after previous AAA repair can be performed with an acceptable level of operative mortality, though with considerable operative morbidity. Limited long-term survival mandates careful patient selection, and the high mortality associated with ruptured TVAA underscores the need for post-AAA surveillance.