The influence of a composite hospital volume on outcomes for gastric cancer surgery: A Dutch population-based study.

Abstract

Volume-outcome associations for complex surgical procedures have motivated centralization of care worldwide. The aim of this study was to investigate the association between overall hospital experience with complex upper gastrointestinal (GI) cancer resections and outcomes after gastric cancer surgery. Data on all patients (n = 4837) who underwent a resection for non metastatic invasive gastric cancer between 2005 and 2014 were obtained from the Netherlands Cancer Registry (NCR). Annual hospital volume categories were based on the combined volume of gastrectomies, esophagectomies, and pancreatectomies (composite hospital volume). Volume-outcome analyses were performed for lymph node yield, 30-day mortality, and overall survival. The proportion of gastric cancer resections performed in hospitals with an annual composite hospital volume of ≥40 upper GI cancer resections increased from 6% in 2005 to 80% in 2014. A higher composite hospital volume was univariably associated with a higher lymph node yield, lower 30-day mortality, and increased overall survival. Statistical significance was lost after adjusting for case mix. But, sub group analysis including only elderly patients (≥75 years) showed a significant association between composite hospital volume and 30-day mortality. In the Netherlands, an increasing proportion of gastric cancer resections is performed in hospitals with a high composite hospital volume of gastric, esophageal, and pancreatic cancer resections. Special attention is warranted to referral of elderly patients, as these patients might specifically benefit from this centralization.