Abstract

INTRODUCTION: All-cause mortality in patients after repair of aortic aneurysms of the descending thoracic aorta thoracic endovascular aortic repair (TEVAR) is relatively high at mid-term follow-up. The aim of this study was to derive and validate a system that could predict all-cause mortality after TEVAR to aid with patient selection. METHODS: The MOTHER database contained 625 patients that underwent elective surgery for descending thoracic aortic aneurysms. Univariate analysis identified preoperative factors associated with mid-term all-cause mortality, and a Cox proportional hazards model was developed. The model was internally validated using Kaplan-Meier comparison of observed vs predicted mortality. External validation was performed using a data set from the University of Florida College of Medicine. RESULTS: There were 625 patients that underwent TEVAR for descending thoracic aortic aneurysm in the MOTHER database and 231 in the University of Florida College of Medicine validation set. The mid-term mortality rate at 6 years of follow-up was 34.4% and 34%, respectively. The all-cause mortality risk score was calculated using 0.0398 × (age) + 0.516 × (renal insufficiency) + 0.46 × (previous cerebrovascular disease) + 0.352 × (prior tobacco use) + 0.376 × (number of devices >2) + 0.016 × (maximum aneurysm diameter). Using this score, low-, medium-, and high-risk groups were defined, with predicted survival at 5 years of 80%, 60%, and 40%. Patients at high risk of mid-term all-cause death were identified in the validation cohort using the prediction rule. CONCLUSIONS: Identifying patients with a limited life expectancy after TEVAR is possible using a preoperative risk-stratification system. This information can be used to inform decision making regarding when and whether to proceed with TEVAR.