Reintervention does not affect long-term survival after fenestrated endovascular aneurysm repair.
Academic Article
Overview
abstract
OBJECTIVE: Fenestrated endovascular aneurysm repair (FEVAR) is increasingly used in the treatment of juxtarenal aortic aneurysms and short-neck infrarenal aneurysms. Reinterventions (REIs) occur frequently, contributing to patient morbidity and resource utilization. We sought to determine whether REI affects long-term survival after FEVAR. METHODS: A single-institution retrospective review of all Cook Zenith fenestrated (ZFEN; Cook Medical, Inc, Bloomington, IN) repairs was performed. Patients with ≥6 months of follow-up and without adjunctive branch modifications were included. REI was defined as any aneurysm, device, target branch, or access-related intervention after the index procedure. REIs were categorized as early (<30 days) or late (≥30 days), by indication (ie, branch, endoleak, limb related, access related, other), and by target branch or device components. Patients were stratified into REI vs no REI groups and branch REI vs non-branch REI groups. RESULTS: Of 219 consecutive ZFEN repairs from 2012 to 2021, 158 patients met the inclusion criteria. Of these 158 patients, 41 (26%) required a total of 51 REIs (10 early and 41 late) during a mean follow-up of 33.9 months. The most common indication for REI was branch-related (31 of 51; 61%), with the renal arteries the most frequently affected (26 of 51; 51%). The only differences found in baseline, aneurysm, and device characteristics were a higher mean Society for Vascular Surgery comorbidity score (9.6 vs 7.9; P = .04) and larger suprarenal neck angle (23.3° vs 17.1°; P = .04) in the no REI group. In contrast, the REI group had a larger mean proximal seal zone diameter (26.3 mm vs 25.1 mm; P = .03) and device diameter (31.9 mm vs 30.0 mm; P = .002) compared with the no REI group. Technical success and operative characteristics were similar between the groups, except for a longer mean fluoroscopy time (74.9 minutes vs 60.8 minutes; P = .01) and longer median length of stay (2 vs 2 days; P = .006) for the REI group. Although the rate of early (<30 days) major adverse events was greater for the REI group (24.4% vs 6.0%; P = .001), the difference in 30-day mortality was not statistically significant (4.9% vs 0.9%; P = .10). On Kaplan-Meier analysis, freedom from REI at 1 and 5 years was 85.7% and 62.6%, respectively, for the overall cohort. No difference was found in the estimated 5-year survival between the REI and no REI groups (62.8% vs 63.5%; log-rank, P = .87) and branch REI and non-branch REI groups (71.8% vs 49.9%; log-rank, P = .16). On multivariate analysis, REI was not an independent predictor for mortality. However, age, Society for Vascular Surgery comorbidity score, and preoperative maximum aneurysm diameter each increased the hazard of death (hazard ratio [HR], 1.07; 95% confidence interval [CI], 1.02-1.12 [P = .007]; HR, 1.10; 95% CI, 1.01-1.18 [P = .02]; HR, 1.05; 95% CI, 1.02-1.08 [P = .003], respectively). CONCLUSIONS: After ZFEN, 41 patients (26%) had required a total of 51 REIs, with most occurring ≥30 days after the index procedure, and 61% were branch related, with no influence on 5-year survival. Age, comorbidity, and baseline aneurysm diameter independently predicted mortality. The use of FEVAR mandates lifelong surveillance and protocols to maintain branch patency. Despite their relative frequency, REIs did not influence 5-year postprocedural survival.
