Defining Facility Volume Threshold for Optimization of Short- and Long-Term Outcomes in Patients Undergoing Resection of Perihilar Cholangiocarcinoma

Background We determined the minimum threshold (Tmin) of annual facility case volume to optimize outcomes for patients with resected perihilar cholangiocarcinoma. Methods We identified patients with localized perihilar cholangiocarcinoma who underwent resection within the National Cancer Database (2010–2017). We used marginal structural logistic regression models to estimate the average treatment effect of receiving care in facilities meeting/exceeding Tmin on 90-day mortality and other postoperative outcomes. Results A total of 2471 patients underwent resection for perihilar cholangiocarcinoma at 471 facilities. There was no effect of total hepatopancreatobiliary, surgical hepatopancreatobiliary, total hepatobiliary, surgical hepatobiliary, or total perihilar cholangiocarcinoma case volume on 90-day mortality. A Tmin of seven perihilar cholangiocarcinoma resections/year resulted in lower odds of 90-day mortality (IP-weighted OR = 0.49, 95% CI: 0.66–0.87). A total of two facilities met the Tmin. Patients receiving treatment at Tmin facilities had lower odds of length of stay ≥ 7 days (IP-weighted OR = 0.85, 95% CI: 0.75–0.97) and positive surgical resection margins (IP-weighted OR = 0.40, 95% CI: 0.47–0.55). Additionally, undergoing surgery at Tmin facilities resulted in higher (≥ 4 nodes) lymph node yields (IP-weighted OR = 1.94, 95% CI: 1.21–3.11) but no change in the odds of nodal positivity. There was no effect of undergoing surgery at Tmin facilities on 30-day mortality or re-admission. Conclusions Resection of perihilar cholangiocarcinoma is infrequently performed at a high number of facilities. A Tmin of ≥ 7 resections/year resulted in lower 90-day mortality and improved postoperative outcomes. Our data suggest that regionalization of care for patients with perihilar cholangiocarcinoma could potentially improve outcomes in the USA.