Effect of operative volume on morbidity, mortality, and hospital use after esophagectomy for cancer

Objective: We sought to evaluate the effect of operative volume, hospital size, and cancer specialization on morbidity, mortality, and hospital use after esophagectomy for cancer. Methods: Data derived from the Health Care Utilization Project was used to evaluate all Medicare-reimbursed esophagectomies for treatment of cancer from 1994 to 1996 in 13 national cancer institutions and 88 community hospitals. The complications of care, length of stay, hospital charges, and mortality were assessed according to hospital size (greater than or equal to 600 beds vs <600 beds), cancer specialization (national cancer institution vs community hospital), and operative volume (esophageal [greater than or equal to 5 Medicare esophagectomies per year vs <5 Medicare esophagectomies per year] and nonesophageal operations [greater than or equal to 3333 cases per year vs <3333 cases per year]). Results: Mortality was lower in national cancer institution hospitals (4.2% [confidence interval, 2.0%-6.4%] vs 13.3% [confidence interval, 4.2%26.2%], P = .05) and in hospitals performing a large number of esophagectomies (3.0% [confidence interval, 0.09%-5.1%] vs 12.2% [confidence interval, 4.5%-19.8%], P < .05). Multivariate analysis revealed that the independent risk factor for operative mortality was the volume of esophagectomies performed (odds ratio, 3.97; P = .03) and not the number of nonesophageal operations, hospital size, or cancer specialization. Hospitals performing a large number of esophagectomies also showed a tendency toward decreased complications (55% vs 68%, P = .06), decreased length of stay (14.7 days vs 17.7 days, P = .006), and decreased charges ($39,867 vs $62,094, P < .005). Conclusions: These results demonstrate improved outcomes and decreased hospital use in hospitals that perform a large number of esophagectomies and support the concept of tertiary referral centers for such complex oncologic procedures as esophagectomies.