Hospital and Surgeon Annual Volume Thresholds Associated With Improved 30-Day Mortality After Fenestrated and Branched Aortic Repair.

Zettervall, Sara L; Dun, Chen; Columbo, Jesse A; Mendes, Bernardo C; Schanzer, Andres; Oderich, Gustavo; Schermerhorn, Marc L; Black, James H et al. · Ann Surg · 2026

retrospective_cohort · Level III

Where this comes from

Abstract

This study aims to assess the impact of surgeon and hospital volume on mortality following fenestrated and branched repair of aortic aneurysms (F/BEVAR) and identify minimum volume thresholds for repair. While hospital volume predicts outcomes after standard open and endovascular aortic repair, the volume-outcome relationship for F/BEVAR remains poorly defined. We performed a retrospective cohort study using 100% Medicare fee-for-service claims (2016-2023) to identify adults undergoing F/BEVAR. We used locally estimated scatterplot smoothing (LOESS) and derivative-based methods to identify hospital- and surgeon-level annual volume thresholds associated with 30-day mortality. Multivariable logistic regression were used to assess the association between hospital and surgeon volume (separately) and 30-day mortality. There were 8015 patients who received F/BEVAR (mean age 76.1±6.8 y, 27.7% female, 90.9% White race). Median annual hospital and surgeon volume was 2 cases (range 1-70) and 1 case (range 1-51), respectively. Overall, 350 (4.4%) of patients experienced mortality within 30 days. The LOESS plots identified inflection points at 9 cases/year for hospitals and 7 cases/year for surgeons. After adjusting for patient risk factors, there was a significant association of both low-hospital volume (≤9 cases/year) (aOR 1.43, 95% CI 1.02, 1.99) and surgeon volume (≤7 cases/year) (aOR 1.51, 95% CI 1.04, 2.18) with 30-day mortality after F/BEVAR. There is a significant volume-outcome relationship after F/BEVAR. As commercially available devices become available, volume thresholds may provide an opportunity to guide dissemination of the technology while ensuring acceptable patient outcomes.