P-HPB-12: The utility of endoscopic ultrasonography in common bile duct dilatation of unknown etiology detected from cross-sectional imaging

BACKGROUND: Endoscopic ultrasonography (EUS) is frequently being performed when common bile duct (CBD) dilatation is detected on cross-sectional imaging without identifiable causes. OBJECTIVES: To assess the diagnostic yield of EUS in patients with inconclusive cause of CBD dilatation detected by cross-sectional imaging. METHODS: This was a retrospective review of patients with dilated CBD on either computed tomography (CT) or magnetic resonance imaging/magnetic resonance cholangiopancreatography (MRI/MRCP) without definite cause undergoing EUS during 2008–2015. RESULTS: A total of 131 patients were recruited (mean age 63, 47% males). The mean diameter of CBD on CT or MRI/MRCP was 12 mm. The etiology of CBD dilatation was identified in 88/131 (67%) patients. Among these, 47% had malignant obstruction with distal cholangiocarcinoma being the most common cancer (51%), 27% had bile duct stone, and 26% had benign stricture. EUS offers an accuracy of 98% in the detection of pathological CBD dilatation. In addition, it is 100% accurate for CBD stone detection, 93% accurate for both malignant and benign obstruction, and 100% specific in cases of nonpathological CBD dilatation. Multivariate analysis indicated that male gender, serum glutamic-pyruvic transaminase = 3 × upper limit of normal (ULN), alkaline phosphatase (ALP) = 3 × ULN, and intrahepatic duct (IHD) dilatation were predictors of pathological obstruction with the odds ratio of 5.46, 5.02, 4.63, and 4.03, respectively. In contrast, ALP = 2 × ULN, female gender, and no IHD dilatation were predictive of nonpathological CBD dilatation with the odds ratio of 7.15, 5.18, and 4.12, respectively. CONCLUSION: EUS is essential in the evaluation of CBD dilatation in inconclusive CT or MRI/MRCP and thus should be performed routinely when clinically or biochemistry indicative of pancreatobiliary diseases.