Volume and outcome for major upper GI surgery in England

Abstract

Background: The correlation between hospital or surgeon volume and outcome for complex surgical procedures has been the subject of several studies in recent years. In the UK, such studies have been used to strengthen the case for centralization of such procedures. The recent availability of easily accessible and fully independent data on hospital outcomes for surgical services in the UK has provided the opportunity to review any potential associations between volume and outcome in the UK.

Methods: Hospital Episode Statistic (HES) data were collected through Dr Foster for four different upper GI procedures (gastrectomy, esophagectomy, pancreaticoduodenectomy, and liver resection) for a 6-year period from 1999 to 2005. Data for each procedure were divided into volume-dependant quartiles to assess any differences in mortality outcome.

Results: Generally, mortality rates for all four procedures are lower than previously studies have suggested. A significant trend favoring high volume providers was noted for esophagectomy, with mortality rates varying from 7.8% to 4.0% for lowest to highest volume providers (p < 0.001). A similar but less clear-cut trend was noted for pancreaticoduodenectomy. There was no significant difference for gastric and liver resection between low- and high-volume providers. There was a 20% decrease in centers performing esophagectomy and 28% for centers performing pancreaticoduodenectomy.

Conclusion: There is a volume outcome association for esophagectomy and pancreaticoduodenectomy. There is no association for gastrectomy or hepatectomy.