Pancreatic necrosectomy: North American mortality is much lower than expected.

Abstract

The aim of this analysis was to explore the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database to determine outcomes of patients undergoing debridement for pancreatic and peripancreatic necrosis. Single-institution series suggest that the mortality of patients undergoing pancreatic necrosectomy has improved but remains at 15% to 20%. But no national data have been available for patients with necrotizing pancreatitis. In 2007, a CPT code specific for debridement of pancreatic necrosis became available. The ACS-NSQIP Participant Use File was queried for all patients who had debridement of pancreatic and peripancreatic necrosis (CPT code 48105) from January 1, 2007, through December 31, 2007. Patient demographics, observed (O) and expected (E) morbidity and mortality, and indices (O/E) were evaluated. A multivariate stepwise logistic regression was performed to determine predictors of mortality. During this 12-month period, data were accumulated on 161 patients. The mean age was 54 years; 71% were male; and 75% were Caucasian. The mean body mass index was 30.3 kg/m(2); 29% had diabetes; and 11% abused alcohol. Forty-two percent were transferred to NSQIP hospitals from other facilities. Overall morbidity was 62%, and 30-day mortality was 6.8%, but morbidity and mortality indices were 0.86 and 0.33, respectively. Increased age and blood urea nitrogen were independent predictors of mortality. These data suggest that patients undergoing debridement for pancreatic and peripancreatic necrosis at ACS-NSQIP hospitals provide a new North American sample and have better than predicted outcomes. We concluded that ACS-NSQIP is a powerful tool to assess contemporary outcomes of uncommon, high-risk procedures.