Preoperative Functional Status Predicts 2-year Mortality in Patients Undergoing Fenestrated/branched Endovascular Aneurysm Repair

Abstract

Fenestrated/branched endovascular aneurysm repair (F/BEVAR) is a minimally invasive alternative for patients at high risk for open repair of complex aortic aneurysms. Nearly all investigative study protocols evaluating F/BEVAR require a predicted life expectancy >2 years for study inclusion. However, accurate risk models for predicting 2-year survival in this patient population are lacking. We sought to identify preoperative predictors of 2-year survival in patients undergoing F/BEVAR. Prospectively collected data on all consecutive F/BEVAR procedures, performed in an institutional review board-approved registry and/or physician-sponsored investigational device exemption trial (IDE# G130210), were reviewed (November 2010 to February 2019). Forty-four preoperative patient characteristics, including comorbidities, preoperative functional status, and aneurysm/repair characteristics, were assessed. Preoperative functional status was defined as totally dependent (any impairment in activities of daily living (ADLs) or residence in a skilled nursing facility), partially dependent (any impairment in instrumental ADLs), or independent (no impairment of ADLs/instrumental ADLs). Based on a univariate screen (P < .2), a Cox proportional hazards model was constructed to identify independent predictors of 2-year all-cause mortality. Among 256 consecutive F/BEVARs (six [2.3%] common iliac, 94 [41%] juxtarenal, 35 [14%] pararenal, 119 [47%] thoracoabdominal, and 2 [0.8%] arch aneurysms), the 2-year mortality was 18%. On Cox modeling, the only independent preoperative predictor of 2-year mortality was functional status (totally dependent, hazard ratio [HR], 5.4 [95% CI, 1.8-16; P < .01]; partially dependent, HR, 4.7 [95% CI, 2.5-8.7; P < .01]). A history of coronary revascularization was protective (HR, 0.5, 95% CI, 0.2-0.9; P = .03). Factors such as age, congestive heart failure, chronic kidney disease, diabetes, chronic obstructive pulmonary disease, aneurysm extent, and prior aortic surgery were not significant. The 2-year mortality for the independent (n = 176 [69%]), partially dependent (n = 69 [27%]), and totally dependent groups (n = 10 [3.9%]), was 11%, 33%, and 40%, respectively (Fig). In patients undergoing F/BEVAR, decreased preoperative functional status is the strongest predictor of 2-year mortality, with totally dependent patients experiencing poor survival. Traditional risk factors were not independently significant, perhaps reflecting a high prevalence of severe chronic illness in these high-risk patients participating in an investigational device exemption trial. Given that 2-year survival in the independent group of this high-risk cohort is 89%, which mirrors that of infrarenal EVAR, expanding the indication for F/BEVAR to patients who are independent is reasonable for low-risk patients as well.