Effect of centralization and regionalization of pancreatic surgery on resection rates and survival.

BACKGROUND: Centralization of pancreatic surgery in the Netherlands has been ongoing since 2011. The aim of this study was to assess how centralization has affected the likelihood of resection and survival of patients with non-metastatic pancreatic head and periampullary cancer, diagnosed in hospitals with and without pancreatic surgery services. METHODS: An observational cohort study was performed on nationwide data from the Netherlands Cancer Registry (2009-2017), including patients diagnosed with non-metastatic pancreatic head or periampullary cancer. The period of diagnosis was divided into three time intervals: 2009-2011, 2012-2014 and 2015-2017. Hospital of diagnosis was classified as a pancreatic or non-pancreatic surgery centre. Analyses were performed using multivariable logistic and Cox regression models. RESULTS: In total, 10 079 patients were included, of whom 3114 (30.9 per cent) were diagnosed in pancreatic surgery centres. Between 2009-2011 and 2015-2017, the number of patients undergoing resection increased from 1267 of 3169 (40.0 per cent) to 1705 of 3566 (47.8 per cent) (P for trend < 0.001). In multivariable analysis, in 2015-2017, unlike the previous periods, patients diagnosed in pancreatic and non-pancreatic surgery centres had a similar likelihood of resection (odds ratio 1.08, 95 per cent c.i. 0.90 to 1.28; P = 0.422). In this period, however, overall survival was higher in patients diagnosed in pancreatic surgery than in those diagnosed in non-pancreatic surgery centres (hazard ratio 0.92, 95 per cent c.i. 0.85 to 0.99; P = 0.047). CONCLUSION: After centralization of pancreatic surgery, the resection rate for patients with pancreatic head and periampullary cancer diagnosed in non-pancreatic surgery centres increased and became similar to that in pancreatic surgery centres. Overall survival remained higher in patients diagnosed in pancreatic surgery centres.