Results are presented from the first completed multicenter trial directed at gaining approval from the US Food and Drug Administration of endovascular versus open surgical repair of descending thoracic aortic aneurysms. Between September 1999 and May 2001, 140 patients with descending thoracic aneurysms were enrolled at 17 sites and evaluated for a Gore TAG Thoracic Endograft. An open surgical control cohort of 94 patients was identified by enrolling historical and concurrent subjects. Patients were assessed before treatment, at treatment, and at hospital discharge and returned for follow-up visits at 1 month, 6 months, and annually thereafter. One hundred thirty-seven of 140 patients had successful implantation of the endograft. Perioperative mortality in the endograft versus open surgical control cohort was 2.1% (n = 3) versus 11.7% (n = 11, P < .001). Thirty-day analysis revealed a statistically significant lower incidence of the following complications in the endovascular cohort versus the surgical cohort: spinal cord ischemia (3% vs 14%), respiratory failure (4% vs 20%), and renal insufficiency (1% vs 13%). The endovascular group had a higher incidence of peripheral vascular complications (14% vs 4%). The mean lengths of intensive care unit stay (2.6 +/- 14.6 vs 5.2 +/- 7.2 days) and hospital stay (7.4 +/- 17.7 vs 14.4 +/- 12.8 days) were significantly shorter in the endovascular cohort. At 1 and 2 years' follow-up, the incidence of endoleaks was 6% and 9%, respectively. Through 2 years of follow-up, there were 3 reinterventions in the endograft cohort and none in the open surgical control cohort. Kaplan-Meier analysis revealed no difference in overall mortality at 2 years. In this multicenter study early outcomes with descending aortic endovascular stent grafting were very encouraging when compared with those of a well-matched surgical cohort. However, at 2 years' follow-up, there is an incidence of endoleaks and reinterventions associated with endovascular versus open surgical repair. Continued vigilant surveillance of patients treated with an endograft is important.
