Regional Differences in the Delivery and Outcomes of Oesophageal Cancer Surgery Across Australia

BACKGROUND: In Europe, the lowest perioperative mortality is seen when oesophagectomies are performed in high-volume hospitals. In Australia, the optimum annual volume for the lowest mortality has been shown to be 21+. The objective was to evaluate regional variation in hospital mortality following oesophagectomy from a prospective database using univariate and multivariable logistic regression models. METHODS: The Australia and New Zealand Intensive Care Society Adult Patient Database was used to identify patients undergoing oesophagectomy for cancer in Australia from 2012 to 2021. The hospital mortality for each jurisdiction per year was determined, and a three-level mixed-effects logistic model was used to assess hospital volume versus hospital mortality. RESULTS: 4189 patients underwent oesophagectomy in 124 hospitals. Regions with the highest number of hospitals performing oesophagectomies were NSW/ACT (44) and VIC/TAS (38). The jurisdiction with the highest proportion of operations in high-volume hospitals (21+ operations per year) was Queensland (39.7%), followed by Western Australia (39.6%) and South Australia (31.8%). Mortality rates were 1.4% in Queensland, 2.5% in South Australia, 3.3% in Western Australia, 3.8% in VIC/TAS and 5.4% in NSW/ACT. The differences in mortality between jurisdictions were primarily driven by differences in surgical volume across hospitals within each jurisdiction, accounting for 73.9% of the between-hospital variance, with higher volume associated with lower mortality. CONCLUSIONS: Differences exist between jurisdictions in Australia for outcomes following oesophagectomy. Perioperative mortality was lower in regions where surgery was more commonly undertaken in a higher-volume hospital. Hospital mortality in Australia might be reduced if oesophagectomies are uniformly centralised to high-volume hospitals.