ASSOCIATION BETWEEN ANNUAL FACILITY VOLUME WITH OVERALL SURVIVAL AND POSTOPERATIVE MORTALITY FOR RESECTION OF PROXIMAL EXTRAHEPATIC CHOLANGIOCARCINOMA
Mohamedraed Elshami*, Jonathan J. Hue, Richard S. Hoehn, Luke Rothermel, Jeffrey Hardacre, John Ammori, Jordan M. Winter, Lee M. Ocuin
Department of Surgery, UH Cleveland Medical Center, Cleveland, OH
The association between annual facility volume of resection for proximal extrahepatic cholangiocarcinoma (pCCA) and survival or postoperative mortality is poorly defined. We aimed to determine the minimum threshold (Tmin) for annual facility volume for resections of pCCA at which patient outcomes are optimized.
We identified patients with localized pCCA (with cM0) who underwent resection in the National Cancer Database (2010-2017). We used multivariable Cox regression to analyze the association between annual volume of resections and overall survival (OS) to determine the Tmin. We performed multivariable logistic regression analyses adjusting for sociodemographic factors to examine the association between Tmin and postoperative outcomes (length of stay ?7d, 30d readmission, 30/90d mortality, number of lymph nodes examined ?4, surgical margins). We used marginal structural logistic regression models to estimate the average treatment effect of receiving care in facilities exceeding Tmin.
In total, 1473 patients underwent resection for pCCA at 379 facilities. A Tmin of 1 resection/year was associated with an improved OS (median OS: 36.0 vs 26.9mos, HR=0.83, 95% CI: 0.75-0.93). A total of 47 facilities (12.4%) met the Tmin. Patients treated at facilities that met the Tmin were less likely to receive adjuvant chemotherapy (42.2 vs 48.4%) and radiotherapy (24.9 vs 32.9%) but more likely to receive neoadjuvant chemotherapy (10.0 vs 6.2%) and/or radiotherapy (6.1 vs 2.1%)
On multivariable analysis, patients treated at Tmin facilities had a higher likelihood of length of stay >7d (72.4 vs 66.7%, OR=1.40, 95% CI 1.06-1.85) and having ? 4 lymph nodes examined (63.1 vs 49.2%, OR=1.72, 95% CI 1.30-2.27). Patients treated at Tmin facilities had lower likelihood of positive margins (31.1 vs 42.2%, OR=0.62, 95% CI 0.49-0.78) and 90d postoperative mortality (8.0 vs 11.2%, OR=0.65, 95% CI 0.44-0.97). There was no association between Tmin and 30d readmission or mortality.
There was 1 facility performing ?4 resections/year. On the marginal model, patients receiving treatment at facilities with ?4 resections/year had lower length of stay >7d, 30d readmission, 90d mortality, and positive margins but higher rates of ?4 lymph nodes examined. There was no effect of receiving treatment at facilities with ?4 resections/year on 30d mortality.
A large number of facilities performed a relatively low number of pCCA resections. A Tmin of ?1 resection/year was associated with higher OS and lower 90d mortality. Higher thresholds resulted in even further reductions of adverse outcomes. These data suggest that defining minimal volume standards of an infrequently performed procedure may help improve short- and long-term outcomes, and consideration should be given to regionalizing care for patients with localized disease.
Back to 2022 Abstracts