Intermediate Likelihood of Choledocholithiasis: Do All Need EUS or MRCP?

Background Recently updated guidelines for choledocholithiasis stratify suspected patients into high, intermediate, and low likelihood, with the aim to reduce risk of diagnostic endoscopic retrograde cholangiopancreatography. This approach has increased proportion of patients in intermediate likelihood making it heterogenous. We aim to substratify intermediate group so that diagnostic tests (endoscopic ultrasound/magnetic resonance cholangiopancreatography) are judicially used. Methods This is a single-center retrospective analysis of prospectively maintained data. We used subset of patients who met intermediate likelihood of American Society of Gastrointestinal Endoscopy (ASGE) criteria from previously published data (PMID:32106321) as derivation cohort. Binominal logistic regression analysis was used to define independent predictors of choledocholithiasis. A composite score was derived by allotting 1 point for presence of each independent predictor. The diagnostic performance of a composite score of ≥ 1 was evaluated in validation cohort. Results A total of 678 (mean age [standard deviation]: 47.0 [15.9] years; 48.1% men) and 162 (mean age 47.8 [14.8] years; 47.4% men) patients in ASGE intermediate-likelihood group were included as derivation cohort and validation cohort, respectively. Binominal logistic regression analysis showed that male gender (p = 0.024; odds ratio [OR] = 1.92), raised bilirubin (p = 0.001; OR = 2.40), and acute calculus cholecystitis (p = 0.010; OR = 2.04) were independent predictors for choledocholithiasis. A composite score was derived by allotting 1 point for presence of independent predictors Using ≥ 1 as cutoff, sensitivity and specificity for detection of choledocholithiasis were 80% (95% confidence interval [CI]: 68.2–88.9) and 36.2% (95% CI: 32.2–40.0), respectively, in derivation cohort. Applying composite score in independent validation cohort showed sensitivity and specificity of 73.3% (95% CI: 44.9–92.2) and 40.1% (95% CI: 30.1–48.5), respectively. Conclusion Substratification of intermediate-likelihood group of ASGE criteria is feasible. It may be useful in deciding in whom confirmatory tests should be performed with priority and in whom watchful waiting may be sufficient.