Limited survival in dialysis patients undergoing intact abdominal aortic aneurysm repair.

OBJECTIVE: Elective abdominal aortic aneurysm (AAA) repair in suitable candidates is a standard modality. The outcomes of AAA repair in patients with end-stage renal disease on dialysis are not well characterized, and there is questionable survival advantage in such patients with limited life expectancy. We sought to describe outcomes after AAA repair in U.S. dialysis patients.

METHODS: The United States Renal Data System was used to collect data on intact asymptomatic AAA repair procedures in dialysis patients in the United States between 2005 and 2008. Endovascular AAA repair (EVAR) and open aortic repair (OAR) were identified by Current Procedural Terminology codes. Primary outcomes were perioperative (30-day) mortality and long-term survival. Predictors of mortality were identified by multivariate regression models.

RESULTS: A total of 1557 patients were identified who had undergone elective AAA repair: 261 OAR and 1296 EVAR. The 30-day mortality was 11.3% (EVAR, 10.3%; OAR, 16.1%; P = .010), with increased age associated with increased mortality (odds ratio, 1.04; 95% confidence interval [CI], 1.02-1.07; P = .001). Kaplan-Meier survival estimates were 66.5% at 1 year (EVAR, 66.2%; OAR, 68%) and 37.4% at 3 years (EVAR, 36.8%; OAR, 40%; P = .33). Median survival was 25.3 months after EVAR and 27.4 months after OAR. Women had a higher mortality rate at 1 year (38.7%) compared with men (32.0%) (P = .015). There was no significant mortality difference at 1 year in comparing type of procedure in both men (EVAR, 31.6%; OAR, 34%; P = .55) and women (EVAR, 39.3%; OAR, 36%; P = .60). A Cox proportional hazards model demonstrated that male gender (hazard ratio [HR], 0.75; 95% CI, 0.62-0.92; P = .005), increased time on dialysis (HR for each year on dialysis, 0.79; 95% CI, 0.75-0.83; P < .001), kidney transplantation history (HR, 0.62; 95% CI, 0.43-0.88; P = .008), and diagnosis of hypertension (HR, 0.60; 95% CI, 0.48-0.75; P < .001) were protective against mortality. Increased age (HR, 1.02; 95% CI, 1.01-1.03; P < .001) and diabetes diagnosis (HR, 1.39; 95% CI, 1.13-1.71; P = .002) predicted increased mortality.

CONCLUSIONS: AAA patients on dialysis have high perioperative and 1-year mortality rates after EVAR or OAR, particularly diabetics, women, and the elderly. This raises questions about the indications for intact AAA repair in dialysis patients, in whom the size threshold may need to be raised. Dialysis patients may be best served by deferring repair of AAA until AAAs reach large size or become symptomatic, especially if OAR is required, given the higher perioperative mortality compared with EVAR.