Risk of renal failure and death when renal arteries are involved in endovascular aortic aneurysm repair

Abstract

ObjectiveEndovascular abdominal aortic repair can involve the incorporation of renal arteries. Revascularization after intentional or unintentional renal artery (RA) coverage is not always technically successful, and the loss of a single RA may result in the need for postoperative dialysis. Thus, we compared the outcomes after endovascular aneurysm repair (EVAR) stratified by RA involvement (RAI). MethodsPatient data from the Vascular Quality Initiative from 2009 to 2018 registry were analyzed. The exclusion criteria were preoperative dialysis, missing RAI data, and repair above the superior mesenteric artery. The repair type cohorts were defined as (1) no RAI (NRAI), (2) RAI with revascularization (RAI-R), and (3) RAI with no revascularization (RAI-NR). A sensitivity analysis was performed by excluding ruptured presentations. The primary outcome was the need for postoperative dialysis. The secondary outcomes were 30-day mortality, dialysis at follow-up, postoperative renal function, and 2-year survival. Multivariate analysis was used to determine the independent predictors for postoperative dialysis. The 2-year survival analysis was performed using Kaplan-Meier log-rank test. ResultsOf 54,020 patients in the EVAR and TEVAR (thoracic EVAR)/complex EVAR modules in the Vascular Quality Initiative, 25,724 met the criteria for inclusion (NRAI, n = 24,879; RAI-R, n = 733; RAI-NR, n = 112). The demographics and comorbidities were similar among the three groups. The RAI-NR group had more frequently had ruptured or symptomatic aneurysms. The postoperative dialysis requirement was higher in the RAI-NR group (NRAI, 0.7%; RAI-R, 2.2%; RAI-NR, 17%; P < .0001), as were the 30-day mortality and dialysis requirement at follow-up. On multivariate analysis, RAI-R (odds ratio [OR], 2.2; P = .03) and RAI-NR (OR, 5.9; P < .0001) were independent predictors of postoperative dialysis and remained so after excluding ruptured presentations (RAI-R: OR, 3; P = .003; RAI-NR: OR, 22.3; P < .0001). Other independent predictors of the need for postoperative dialysis were worse preoperative renal function, a symptomatic presentation, any preoperative or intraoperative blood transfusion, and larger blood loss (≥200 mL). Excluding those with rupture, the overall survival at 2 years on Kaplan-Meier analysis was lower for the RAI-NR group (NRAI, 92%; RAI-R, 89%; RAI-NR, 80%; P = .004). ConclusionsRAI is highly predictive of the need for postoperative and permanent dialysis after EVAR. RAI-NR was associated with lower overall survival. These risks should be considered when planning and performing EVAR and should be weighed against the risks of open repair when considering the treatment options.