Hospital volume effects on abdominal aortic aneurysm repair mortality–an international registry-based analysis

Abstract

Introduction: The influence of hospital volume on early mortality after abdominal aortic aneurysm (AAA) repair of intact and ruptured AAA internationally is poorly understood. The purpose of this study was to assess the impact of hospital volume on early mortality after endo- vascular aneurysm repair (EVAR) and open aneurysm repair (OAR) using worldwide data. Methods: Vascular registry data on AAA repair was ob- tained from 11 countries 2010-2016 in an international transatlantic registry collaboration involving data from Europe, North America and Australasia. Postoperative mortality was compared among four study cohorts [EVAR and OAR for intact AAA (iAAA) and ruptured AAA (rAAA)]. Hospitals from all countries were grouped into quartiles based upon aggregated annualized aortic case volume for OAR and EVAR. The effect of volume on outcome after adjustment for patient age, sex, and medical risk factors was assessed in a hierarchical logistic regression model. Results: Among 178,302 patients (mean age:738 years; fe- male:15%), 87% were treated for iAAA and 13% for rAAA. EVAR was used in 63% of patients [iAAA:68%; rAAA:28%]. Median individual hospital annual volumes varied between countries, EVAR 7.0-76.4, OAR 4.0-85.6 cases per year. Overall post- operative mortality was: EVAR, iAAA 1.0%, rAAA 23.0%; OAR, iAAA 4.7%, rAAA 37.2%. In crude and risk-adjusted analysis, no hospital volume effect on mortality was identified after EVAR for iAAA or rAAA. However, a significant mortality reduction was detected in the highest volume quartile for OAR (iAAA, odds ratio Q4 vs. Q1, 0.55;p1⁄4.007; rAAA, Q4 vs. Q1, 0.44;p< .001). During the study period, EVAR utilization increased for both iAAA (from 65 to 72%) and rAAA repair (from 24 to 34%). Mean hospital OAR volume decreased from 35.7 in 2010-2013 to 29.8 in 2014-2016(p< 0.001). Conclusion: Higher volume hospitals have significantly lower postoperative mortality after OAR for both iAAA and rAAA. Notably, OAR volumes are decreasing due to the increased utilization of EVAR. These results suggest that patients needing OAR should be referred to high volume centers if possible, while similar EVAR outcomes can be anticipated independent of hospital volume. These findings have significant implications on physician training, hospital accreditation, and care regionalization internationally to further improve AAA outcomes.