Pancreatic resection in Veterans Affairs and selected university medical centers: Results of the Patient Safety in Surgery Study

Background: Pancreatectomy is a high-risk, technically demanding operation associated with substantial perioperative morbidity and mortality. This study aims to describe the 30-day morbidity and mortality for pancreatectomy and to compare outcomes between private-sector and Veterans Affairs hospitals using multiinstitutional data. Study Design: This is a retrospective review of patients who underwent pancreatic resection for neoplasia at private-sector (PS) and Veterans Affairs (VA) hospitals participating in the National Surgical Quality Improvement Program Patient Safety in Surgery Study in fiscal years 2002 to 2004. The variables reviewed were demographics, preoperative medical conditions, intraoperative variables, and outcomes. Using logistic regression to control for differences in patient comorbidities, 30-day mortality and morbidity rates between PS and VA hospitals were compared. Results: A total of 1,069 patients underwent pancreatectomy for neoplasia at 97 participating hospitals. Six hundred ninety-two patients were treated at PS hospitals and 377 at VA hospitals. The average number of patients treated at each hospital was 11.0, with a range of 1 to 83 during the 3-year study period. There were 842 patients who underwent pancreaticoduodenectomy (CPT 4815x) and 227 who underwent distal/subtotal pancreatectomy (CPT 4814x). Significant differences were observed between PS patients and VA patients with regard to comorbidities and patient demographics. The 30-day unadjusted morbidity rate was 33.8% overall, 42.2% at VA hospitals versus 29.1% at PS hospitals (p<0.0001). Unadjusted and adjusted odds ratio (OR) for postoperative morbidity comparing VA with PS hospitals was 1.781 (95% CI, 1.369-2.318) and 1.581 (95% CI, 1.064-2.307). The 30-day unadjusted operative mortality rate was 3.8% overall, 6.4% at VA hospitals and 2.5% at PS hospitals (p=0.0015). Unadjusted and adjusted OR for postoperative mortality was 2.909 (95% CI, 1.525-5.549) and 2.533 (95% CI, 1.020-6.290), respectively. Similar outcomes were observed when looking at pancreaticoduodenectomy (CPT 4815x) when analyzed independent of other types of pancreatic resections. Conclusion: Pancreatectomies are high-risk operations with substantial perioperative morbidity and mortality. Risk-adjusted outcomes for patients treated at PS hospitals were found to be superior to those for patients treated at VA hospitals in the study.