Are minimum volume standards appropriate for lung and esophageal surgery?

Background: Several medical systems have adopted minimum volume standards for surgical procedures, including lung and esophageal resection. We sought to determine whether these proposed hospital cutoffs are associated with differences in outcomes. Methods: Analyzing the State Inpatient Databases and Healthcare Cost and Utilization Project of the Agency for Healthcare Research and Quality, we evaluated all patients (aged ≥ 18 years) who underwent lobectomy/pneumonectomy or esophagectomy for cancer in California, Florida, and New York (2009-2011). Hospitals were defined as low volume for each procedure per proposed minimum volume standards by year: <40 lung resections and <20 esophagectomies. We compared demographic data and determined the incidence of complications and mortality between patients operated on at low- versus high-volume hospitals. Propensity matching (of demographic characteristics, income, payer, and comorbidities) was performed to balance the cohorts for analysis. Results: During the time period, 20,138 patients underwent lobectomy/pneumonectomy of which 12,432 operations (61.7%) were performed at low-volume hospitals (n = 456) and 7706 operations were performed at high-volume hospitals (n = 48). Of 1324 patients undergoing esophagectomy, 1087 operations (82.1%) were performed at low-volume hospitals (n = 184), whereas only 237 operations were at high-volume hospitals (n = 6). After propensity matching (lung 1:1 and esophagus 2:1), no major differences were apparent for in-hospital mortality nor major complications for either lung or esophageal resection. Length of stay was longer in low-volume hospitals after lung resection (median 6 vs 5 days; P <.001), but not after esophageal resection. Discussion: Although several groups have publicly called for minimum volume requirements for surgical procedures, the majority of patients undergo lung and esophageal resection at hospitals below the proposed cutoffs. The proposed standards for lung and esophageal resection are not associated with a difference in outcomes in this large administrative database. Efforts should be made to determine more meaningful minimum volume requirements and to determine whether such standards are appropriate.