Outcomes of esophagectomy at academic centers: an association between volume and outcome

Studies have shown that esophagectomies performed at high-volume centers have lower in-hospital mortality. However, the volume-outcome relationship for esophagectomy performed at academic centers is unknown. Using the University HealthSystem Consortium national database, we examined the influence of the hospital's volume of esophagectomy on outcome at academic centers between January 2003 and October 2007. Outcomes including length of stay, 30-day readmission, morbidity, and observed and expected mortality were compared between high (> 12), medium (6-12), and low-volume centers' (< or = 5) annual cases. There were 30 high (n = 3984), 23 medium (n = 822), and 54 low-volume (n = 430) hospitals. Compared with low-volume counterparts, high-volume hospitals had shorter lengths of stay (14.1 vs 17.2 days, P < 0.01), fewer overall complications (51.1% vs 56.5%, P = 0.03), fewer cardiac complications (1.1% vs 2.5%, P = 0.01), fewer pulmonary complications (18.5% vs 29.8%, P < 0.01), fewer hemorrhagic complications (3.2% vs 6.7%, P < 0.01), fewer patients requiring skilled nursing facility care (9.5% vs 19.7% P < 0.01), and lower in-hospital mortality (2.5% vs 5.6%, P < 0.01). The observed-to-expected mortality ratio was 0.6 for high-volume and 1.0 for low-volume centers. Within the context of academic centers, there is a threshold of > 12 esophagectomies annually whereby there is a lower mortality and improved outcome.