Hospital volume versus outcome following oesophagectomy for cancer in Australia and New Zealand

BACKGROUND: Volume-outcome relationships for mortality following oesophagectomy have been demonstrated in Europe and the USA, but not in Australia or New Zealand. We determined whether higher volume hospitals achieve better outcomes following oesophagectomy in Australia and New Zealand. METHODS: Administrative data for hospitals contributing data to the Health Roundtable were analysed. Hospitals performing oesophagectomy for cancer from July 2008 to June 2015 were grouped according to mean annual caseload: low (1-5), medium (6-11) and high (12+) volume. Univariate and multivariable analyses determined the impact of volume on 30-day and in-hospital mortalities, length of hospital stay and mechanical ventilation following surgery. RESULTS: A total of 2252 patients underwent oesophagectomy in 65 hospitals. Sixty-eight percent (n = 44) were low-, 26% (n = 17) were medium- and 6% (n = 4) were high-volume hospitals. Seven hundred and sixty-two (34%) procedures were performed in low-, 1042 (46%) in medium- and 448 (20%) in high-volume hospitals. Overall in-hospital mortality was 3.1% and 30-day mortality was 2.1%. In-hospital mortality was lowest in high-volume hospitals; 1.6% versus 2.6% and 4.1% for low- and medium-volume hospitals (P = 0.02). Surgery in high-volume hospitals was shorter (32 min, P = 0.001), and patients were less likely to require post-operative ventilation (16.7% versus 25.3% and 28.0%, P < 0.001), although patients requiring ventilation in high-volume hospitals were ventilated for longer. CONCLUSIONS: A volume-outcome relationship was demonstrated, with overall better performance in higher volume hospitals. Colocation of oesophagectomies to hospitals that can demonstrate appropriate caseload should be considered.