Predictors of outcome after obesity surgery in New York state from 1991 to 2003

Abstract

This study aimed to examine hospital volume in New York state as a predictor of complications over 13 years. Data from the New York state Statewide Planning and Research Cooperative Systems database were analyzed by diagnosis-related group, patient demographics, and outcomes. The institution data were analyzed on the basis of volume and logistic regression using chi-square analysis for predicting morbidity and mortality. From 1991 to 2003, 24,534 patients underwent bariatric surgery at 79 institutions. Case volume increased 36-fold in high-volume centers (HVCs), whereas the overall number of institutions doubled. The average mortality rate was 1.1% in low-volume centers (LVCs), 0.4% in medium-volume centers (MVCs), and 0.2% in HVCs. Whereas 12.4% of the LVC patients had postoperative anastomotic complications (PACs), only 9.7% had PACs in MVCs and only 7.3% had PACs in HVCs (p < 0.05). Furthermore, only 0.7% of the patients required postoperative reintubation (PRI) after surgery at HVCs, as compared with 1.5% at MVCs and 1.8% at LVCs (p < 0.05). Finally, 1.8% of the patients required blood transfusion of packed red blood cells at HVCs, whereas 2.4% required transfusion at MVCs and LVCs. Experience correlates with fewer postoperative complications, including death, in bariatric centers in New York state.