Graft inclusion and vessel reattachment to openings made in the graft were employed in the treatment of 605 patients with thoracoabdominal aortic aneurysms. These patients were divided into four groups on the basis of the extent of aneurysm. Group I consisted of those patients with involvement of most of the descending thoracic and upper abdominal aorta; group II involved most of the descending thoracic aorta and most or all of the abdominal aorta; group III involved the distal descending thoracic aorta and varying segments of abdominal aorta; and group IV involved most or all of the abdominal aorta including the segment from which the visceral vessels arose. The cause of aneurysm formation was medial degenerative disease in 80%, and dissection in 17%; other causes were responsible in the remaining 3%. The median age was 65 years and associated diseases including aneurysms involving other segments, atherosclerotic occlusive disease, heart disease, chronic obstructive pulmonary disease (COPD), hypertension, and renal insufficiency were frequent. The aneurysm was symptomatic in 70% of cases and rupture had occurred in 4% of cases. There were 54 (8.9%) early (30-day) deaths and 151 late deaths; 400 (66%) patients were still alive 3 months to 20 years after operation, including 60% at 5 years. Statistically significant pre- and intraoperative variables by univariate analysis that were predictive of increased risk of early death were advancing age, associated diseases that included COPD, renal artery occlusive disease, atherosclerotic heart disease, renal insufficiency, and long aortic clamp time. Three of these (age, clamp time, and the presence of COPD) retained significance by multivariate analysis. Variables predictive of risk of late death were age, dissection, extent of aneurysm, rupture, heart disease, cerebrovascular disease, COPD, hypertension, and poor renal function. Age, rupture, renal dysfunction, extent of aneurysm, and dissection retained their significance by multivariate analysis. Variables predictive of neurologic disturbances of the lower extremities included rupture, reattachment of intercostal and lumbar arteries, clamp time, dissection, extent and age. Rupture, reattachment of vessels, dissection, and extent of aneurysm retained significance by multivariate analysis. Thus, the risk of this complication was greatest in patients with extensive lesions (group II) with aortic dissection. The greatest risk of renal failure after operation that required dialysis was in patients who had impaired renal function before operation. Methods employed did not prevent these complications.