Risk factors for 1-year mortality after thoracic endovascular aortic repair.

Abstract

Thoracic endovascular aortic repair, although physiologically well tolerated, may fail to confer significant survival benefit in some high-risk patients. In an effort to identify patients most likely to benefit from intervention, the present study sought to determine the risk factors for 1-year mortality after thoracic endovascular aortic repair. A retrospective review was performed on prospectively collected data from all patients undergoing thoracic endovascular aortic repair from 2002 to 2010 at a single institution. Univariate analysis and multivariate Cox proportional hazards regression analysis were used to identify risk factors associated with mortality within 1 year after thoracic endovascular aortic repair. During the study period, 282 patients underwent at least 1 thoracic endovascular aortic repair; index procedures included descending aortic repair (n = 189), hybrid arch repair (n = 55), and hybrid thoracoabdominal repair (n = 38). The 30-day/in-hospital mortality was 7.4% (n = 21) and the overall 1-year mortality was 19% (n = 54). Cardiopulmonary pathologies were the most common cause of nonperioperative 1-year mortality (22%, n = 12). Multivariate modeling demonstrated 3 variables independently associated with 1-year mortality: age older than 75 years (hazard ratio, 2.26; P = .005), aortic diameter greater than 6.5 cm (hazard ratio, 2.20; P = .007), and American Society of Anesthesiologists class 4 (hazard ratio, 1.85; P = .049). A baseline creatinine greater than 1.5 mg/dL (hazard ratio, 1.79; P = .05) and congestive heart failure (hazard ratio, 1.87; P = .08) were also retained in the final model. These 5 variables explained a large proportion of the risk of 1-year mortality (C statistic = 0.74). Age older than 75 years, aortic diameter greater than 6.5 cm, and American Society of Anesthesiologists class 4 are independently associated with 1-year mortality after thoracic endovascular aortic repair. These clinical characteristics may help risk-stratify patients undergoing thoracic endovascular aortic repair and identify those unlikely to derive a long-term survival benefit from the procedure.