Annual hospital volume of surgery for gastrointestinal cancer in relation to prognosis.

ABSTRACT

BACKGROUND: Studies examining hospital volume for surgery for various gastrointestinal (GI) cancer types have shown conflicting results regarding the influence on long-term prognosis. The aim of this study was to examine annual hospital volume in relation to long-term survival after elective surgery for all GI cancers (esophagus, stomach, liver, pancreas, bile ducts, small bowel, colon, and rectum). METHODS: Population-based cohort study including all 45,908 patients who underwent elective surgery for GI cancers in Sweden in 2005-2013. Follow-up was until 2016 for disease-specific 5-year mortality (main outcome) and 2018 for all-cause 5-year mortality (secondary outcome). Hospitals were divided into quartiles for each GI cancer according to a 4-year average annual volume of the year of surgery and three years earlier. Multivariable Cox regression provided hazard ratios (HRs) with 95% confidence intervals (CIs), adjusted for relevant confounders. RESULTS: Higher hospital volume was associated with a survival benefit in the large group of patients (n = 26,688) who underwent colon cancer resection, with HR 0.89 (95% CI 0.84-0.96) for disease-specific 5-year mortality comparing the highest with the lowest quartile. Higher hospital volume improved 5-year mortality in sub-groups of patients who underwent surgery for cancer of the esophagus, pancreas, and rectum. No such improvements were found for cancer of the stomach, liver, bile ducts, or small bowel. CONCLUSION: Long-term survival was improved at higher volume hospitals for some GI cancers (colon, esophagus, pancreas, rectum), but not for others (stomach, liver, bile ducts, small bowel).