S0563 Complications of Endoscopic Retrograde Cholangiopancreatography in Patients With Previous Bariatric Surgery: A Population-Based Cohort Study

Abstract

INTRODUCTION: Bariatric surgery has been proven to be effective in the treatment of obesity and weight-related diseases. The anatomic changes after bariatric surgery make Endoscopic Retrograde Cholangiopancreatography (ERCP) technically challenging, associated with modest clinical success. This study aims to assess the safety and clinical outcomes of ERCP in patients with previous bariatric surgery (BS). METHODS: The National Inpatient Sample from 2007 to 2013 was queried for hospitalizations of adults over 18 years of age with procedure diagnoses of ERCP. Those with prior BS were selected as cases and those without BS as controls. Case-control matching at a ratio of 1 case to 2 controls was performed based on sex, age, race, comorbidity and obesity. The primary outcomes were in-patient mortality and ERCP-related complications. Multivariate regression analysis was used to identify independent risk factors associated with the outcomes. RESULTS: A total of 1,068,862 weighted hospitalizations with ERCP procedure codes were identified. 6,689 with BS were selected as cases, and 13,246 were matched as controls. Length of hospital stay was 7.0 vs 5.7 days for patients with and without BS, respectively, P < 0.05. Total hospital cost was $55,719 and $49,795, with and without BS, P < 0.05. The BS group had lower rates of acute pancreatitis (0.1% vs 1.3%), cholecystitis (0.1% vs 0.3%), and bleeding (1.0% vs 1.4%) but had higher rates of cholangitis (5.0% vs 3.7%) and systemic infections (7.4% vs 5.6%), all P < 0.05. In-patient mortality was lower in the BS group than in the control group (0.2% vs 0.5%, P < 0.05). CONCLUSION: Post-ERCP pancreatitis (PEP) and bleeding are the two most common complications after ERCP in the general population, but in the BS group, the most common complications were systemic infections, followed by cholangitis, and bleeding. Both groups had lower prevalence of PEP (0.1% to 1.3%) than what had been reported in the literature, and prior BS was associated to decreased risk of PEP in the regression analysis. Hospitalized cases may have a lower prevalence of PEP than what has been previously reported in the literature. BS patients had a lower rate of PEP compared to non-BS patients, possibly due to weight loss as obesity has been linked with increased rates of PEP. Prior BS was associated with lower occurrences of PEP, cholecystitis, bleeding, and in-patient mortality but was an independent risk predictor for cholangitis, and systemic infections after adjusted for age, sex, and comorbidities.Table 1Table 2Table 3