Early surgical bypass versus endoscopic stent placement in pancreatic cancer.

Abstract

391 Background: Biliary obstruction frequently occurs in locally advanced or metastatic pancreatic cancer and is often managed by surgical biliary bypass or endoscopic stenting. We compared readmissions/reinterventions among pancreatic cancer patients undergoing bypass vs. stenting. Methods: Retrospective analysis of unresected pancreatic cancer patients in the Healthcare Cost and Utilization Project (HCUP) Florida State Inpatient Database and Florida State Ambulatory Surgery Database 2007-2011 using revisit variables. Patients with early surgical or endoscopic approach analyzed. Subsequent admissions and surgical, endoscopic or percutaneous interventions identified. Propensity score matching by approach. Univariate analysis of patient characteristics and outcomes before and after matching. Multivariate analysis of readmission and reintervention performed by logistic regression. Results: 1,823 and 342 underwent endoscopic treatment vs. early surgical bypass, respectively. After propensity score matching, 684 patients analyzed (table). 64.0% (219) of endoscopic and 70.5% (241) of surgical patients readmitted (p=0.07) and 15.2% (57) and 9.1% (31) underwent reintervention (p=0.01). Endoscopic patients had lower index median length of stay (6 vs 11 days, p<0.01) and admission costs ($11,549 vs $23,215, p<0.01). In multivariate analysis, surgical biliary bypass was predictive of readmission (OR 1.50; 95% CI 1.03-2.18), but initial procedure was not predictive of reintervention (p=0.20). Conclusions: Surgical biliary bypass is less commonly performed than endoscopic stenting. Among propensity score-matched patients, readmission rates are similar, though endoscopic patients require more subsequent interventions. Candidates for both techniques may experience fewer invasive procedures if offered initial surgical biliary bypass. [Table: see text]