The Impact of Surgical Volume on Outcomes in Newly Diagnosed Colorectal Cancer Patients Receiving Definitive Surgeries

Purpose: Colorectal cancer (CRC) patients who receive cancer surgeries by higher-volume providers may have better outcomes. Our previous studies support the notion that surgeon volume is more important to patient outcomes compared with hospital volume. However, the definitions of surgical volume may affect the results. We aim to analyze the effects of different definitions of surgical volume on patient outcome. Methods: We conducted a nationwide population-based study in Taiwan that enrolled all patients who underwent definitive surgery for newly diagnosed CRC between 2005 and 2016. We divided all patients into four quartiles according to different definitions of hospital and surgeon volumes, including total, cumulative, and annual volumes. The primary outcome was a five-year mortality rate, which was analyzed using a frailty model for Cox regression. We adjusted for patientand provider characteristics to control for potential confounding factors. The discrimination model of the different definitions of surgical volume was estimated by the Akaike information criterion (AIC) and the Bayesian information criterion (BIC). Results: : We included 100,009 newly diagnosed CRCpatients in this study, including 55,849 (55.8%) males and 44,160 (44.2%) females, of median age 66 years at diagnosis (range, 20–105 years). The five-year mortality rates were reversely associated with hospital and surgeon volume quartiles, respectively. After adjustment for the patient and provider characteristics, we found that surgeon volume, but not hospital volume, remained an independent predictor of death ( p < 0.001). We further compared three definitions of surgical volume and found cumulative volume provided the lowest AIC and BIC for the prediction of five-year mortality. After adjusting for basic characteristics of patient, hospital, and physician, CRC patients in the high hospital and high surgeon cumulative volume group had the significantly lowest risk of mortality (HR 0.72; 95% CI 0.68–0.75; p < 0.001), followed by the low hospital and high surgeon volume group (HR 0.80; 95% CI 0.77–0.83; p < 0.001) as compared with the low hospital and low surgeon volume group. Conclusion: CRC patients receiving definitive surgery by higher-volume providers had better outcomes, especially where surgeon volume may play a more important role than hospital volume. The cumulative volume could predict five-year mortality of the study cohort better than total and annual volume.