Midterm results of endovascular repair of descending aneurysms with first-generation stent grafts

Objective: Five years after reporting our initial stent-graft repair of descending thoracic aortic aneurysms experience, we determined the 5- to 10-year results of stent-graft treatment and identified risk factors for adverse late outcomes. Methods: Between 1992 and 1997, 103 patients (mean age 69 ± 12 years) underwent repair using first-generation (custom-fabricated) stent grafts. Sixty-two patients (60%) were unsuitable candidates for conventional open surgical repair ("inoperable"). Follow-up was 100% complete (mean 4.5 ± 2.5 years; maximum 10 years). Outcome variables included death and treatment failure (endoleak, aortic rupture, reintervention, and/or aortic-related or sudden death). Results: Overall actuarial survival was 82% ± 4%, 49% ± 5%, and 27% ± 6% at 1, 5, and 8 years. Survival in open surgical candidates was 93% ± 4% and 78% ± 6% and at 1 and 5 years compared with 74% ± 6% and 31% ± 6% in those deemed inoperable (P < .001). Independent risk factors for death were older age (hazard ratio = 1.1; P = .008), previous stroke (hazard ratio = 2.8; P = .003), and being designated an inoperable candidate (hazard ratio = 1.9; P = .04). Actuarial freedom from aortic reintervention and treatment failure at 8 years was 70% ± 6% and 39% ± 8%, respectively. Earlier operative year (hazard ratio = 1.2; P = .07), larger distal landing zone diameter (hazard ratio = 1.1; P = .001), and transposition of the left subclavian artery (hazard ratio = 3.3; P = .008) were determinants of treatment failure. Conclusions: Survival after aneurysm repair using crude, first-generation stent grafts was satisfactory in good operative candidates but bleak in the inoperable cohort, raising the question of whether asymptomatic patients should have even been treated. Late aortic complications were detected in many patients, reemphasizing the importance of serial imaging surveillance.