PTH-114 Real-world performance of non-invasive imaging modalities in the detection of common bile duct stones (CBDS). (8th June 2018)

• Record Type: Journal Article
• Authors: Iqbal, Javed; Giles, Benjamin; Palmer-Jones, Chris; Crinnion, William; Syed, Mohammed; Phillpotts, Simon; Tanwar, Sudeep
• Is Part Of: Gut. Volume 67(2018)Supplement 1
• Page Start: A135
• Page End: A136
• Publication Date: 2018-06-08
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gutjnl-2018-BSGAbstracts.270
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Introduction: Cholangiography at endoscopic retrograde cholangio-pancreatography (ERCP), the reference standard for detecting CBDS, is not practicable as the primary Method for identifying CBDS as it is invasive and has an associated morbidity. With regard to the non-invasive identification of CBDS, the current BGS guidance outlines pooled sensitivities of 73%, 69%–87% and 93% for ultrasound (US), computed tomography (CT) and magnetic resonance cholangio-pancreatography (MRCP) respectively. In this study we report the real-world sensitivity of these imaging modalities to detect CBDS. Method: All cases of CBDS confirmed at ERCP over 12 months were identified prospectively. The imaging modalities employed prior to ERCP were identified and evaluated for the presence or absence of duct dilatation (>6 mm) or intra-ductal contents (stone or sludge) thereby defining the sensitivity of these parameters for detecting CBDS. 95% confidence interval (CI) of the sensitivities was calculated. Results: In total, 102 patients had ERCP for CBDS (female 57%) after presenting with pain (83%) and jaundice (45%. Liver function tests were abnormal in 82%. 48 patients had CBDS ≥1 cm (maximum size 35 mm, mean 9.6 mm). All US scans in this study were performed by sonographers. Overall US performed poorly for the detection of CBDS even when accounting for stone size. As the presence of overlying bowel gas may have contributed to these Results, a further analysis was performed after excluding the 22 cases (25%) in which the bile duct could not be visualised at US. In this sub-group analysis, the sensitivity of US for detecting CBDS remained similar (dilated CBD sensitivity 64% and intra-ductal content sensitivity 18%). Conclusion: Our study demonstrates that US performs poorly in the detection of CBDS in a real-world setting even when accounting for confounding factors such as overlying bowel gas or stone size. In our study, approximately one third of patients with CBDS did not have a dilated CBD on US (diameter ≤6 mm). Moreover, due to spectrum bias it is likely that the true sensitivity of US in the wider population is much lower. By contrast, the performance of both CT and MRCP was more comparable with published data. Whereas our findings may relate specifically to the performance of US in our unit, a larger study continues to determine if the sample size has contributed to our findings.