PTU-076 Factors at first ERCP which influence the decision to proceed to subsequent surgery or repeat ERCP for complex biliary problems

Introduction ERCP is a safe and highly effective solution to many pancreaticobiliary problems. However, surgical options also exist. After a challenging first ERCP, it can be unclear whether surgery or repeat ERCP is preferred. The aim of this study was to identify predictive factors at first ERCP which inform this decision. Methods All ERCPs performed at one hospital (April 2008–March 2011) were analysed. Patients having more than one ERCP were evaluated in detail. Demographics, disease-specific and procedure-specific variables relating to ERCPs and any subsequent surgery were extracted. The primary outcome measure was a requirement for surgery after two or more ERCPs. Descriptive statistics and logistic regression were performed. Results 1729 ERCPs were done in 1270 patients, of which 317 patients had more than one ERCP. Of these, 140 patients were randomly sampled and analysed in detail. These form the denominator for this analysis. The primary diagnosis was gallstones in 62.8%, malignancy in 16.9% and stricture in 10.2%. Combinations of these or other diagnoses occurred in 17.6%. 74.5% of first ERCPs were urgent or emergent. Cannulation was attempted in 96.3% and successful in 81.5% of patients. The operator deemed the first ERCP to be successful in 40.6%. Multiple stones requiring a stent and planned revisit occurred in 15.2% and a large stone requiring lithotripsy in 9.8%. Repeat ERCP was deemed successful by the operator in 65.2% of cases. 40.2% went on to subsequent ERCP attempts. 31.1% of patients having a second or subsequent ERCP ended up having surgery (open biliary exploration, biliary bypass and other operations). On logistic regression, a primary diagnosis of gallstones was associated with likelihood of endoscopic success (OR (95% CI): 3.8 (1.2 to 12.3, p=0.027). In those patients with a primary diagnosis of gallstones, younger patients (OR 1.07 (1.01 to 1.12, p=0.012)) and those with sepsis at presentation (OR 5.3 (1.1 to 25.2, p=0.038)) were significantly more likely to require surgery. No other pattern was predictive of subsequent ERCP success after a first attempt. Conclusion From this analysis, there are no unequivocal clinical or technical factors which make either ERCP or surgery preferable following an incomplete first ERCP. Repeat ERCP should be considered in gallstone disease. In gallstone disease, younger or septic patients should be considered for early surgery if a first ERCP is not successful. This decision is not straightforward; multidisciplinary teamwork and communication between surgeon and endoscopist are essential. Competing interests None declared.