Abstract
OBJECTIVE: To evaluate the diagnostic accuracy of different imaging modalities in patients with partial biliary obstruction with no obvious aetiology on initial imaging. METHODS: This is a prospective single-centre cohort study carried out at Pak Emirates Military Hospital, Rawalpindi from June 2019 to June 2021 with non-probability consecutive sampling. Patients with ages 16 to 75 years, presenting with partial biliary obstruction and undetermined aetiology on initial imaging (TUS and MRCP) were enrolled. EUS was performed for each of these patients and the case was regarded as "true positive" or "true negative" if the findings of imaging modality correlated to those of ERCP. ROC curve, sensitivity, specificity, PPV, NPV and AUC (with 95% confidence interval) were drawn for all the diagnostic tools using SPSS V. 21. RESULTS: A total of 65 patients were enrolled over a period of two years with male to female ratio of 1.4:1. Forty-four patients had an intermediate risk of choledocholithiasis upon preliminary evaluation whereas, 48(74%) of the participants had CBD calculi or sludge confirmed upon subsequent ERCP. Trans-abdominal ultrasound showed the lowest sensitivity (29.2%), specificity (85%), NPV 12% and PPV 93% for diagnosing CBD calculi. This was followed by MRCP with a sensitivity of 37.5%, specificity of 100%, NPV of 36.2% and PPV of 100%. EUS showed the maximum diagnostic accuracy with AUC of 1.0 and a 100% sensitivity and specificity when compared with ERCP as gold standard. CONCLUSION: EUS is superior to MRCP in terms of diagnostic accuracy as minimally invasive diagnostic tool and EUS superiority is particularly relevant in patients with intermediate risk of choledocholithiasis.