Performance of Diagnostic Guidelines in the Evaluation of Choledocholithiasis in Patients With Acute Biliary Presentation: A Systematic Review and Meta-Analysis.

BACKGROUND

Optimal management of acute biliary disease should include an assessment for possible choledocholithiasis (CBDS). Various diagnostic guidelines have been developed by expert bodies for this purpose, but uncertainties remain about their performance in wider practice.

METHODS

For this systematic review and meta-analysis, we searched MEDLINE, Embase, and Scopus for studies on adult populations published in English language between 2000 and August 2024. All studies pertaining to the performance of diagnostic guidelines developed or adopted by regional, national, or international professional bodies are considered, but we excluded diagnostic tools or scoring systems developed locally with limited scopes, such as those employed by a single institution or a group of related institutions. We extracted or derived performance measures in the forms of true positive, true negative, false positive, and false negative and performed meta-analysis using a multilevel random effects model to calculate pooled sensitivity and specificity for the reviewed guidelines and summarized their performance using summary ROC curves and AUCs. The quality of the evidence was assessed with the PROBAST risk of bias and applicability tool. This study is registered on PROSPERO (CRD42024581409).

FINDINGS

Of 1892 records identified, 31 studies were eligible with data available, all of which had a low to moderate risk of overall bias. All studies focused on one or more of three international guidelines, namely the ASGE guidelines in 2010, the revised ASGE guidelines in 2019, and the ESGE guidelines in 2019. For distinguishing patients at high risk for CBDS from those not at high risk, ASGE 2010, ASGE 2019, and ESGE guidelines have pooled sensitivities of 65% (CI: [57,73]), 63% (CI: [53,73]), and 62% (CI: [50,74]) and pooled specificities of 57% (CI: [48,66]), 75% (CI: [65,83]), and 82% (CI: [71,90]), respectively. For distinguishing patients at low risk for CBDS from those at greater than low risk, ASGE 2010, ASGE 2019, and ESGE guidelines have pooled sensitivities of 97% (CI: [92,99]), 95% (CI: [90,98]), and 84% (CI: [70,93]) and pooled specificities of 7% (CI: [3,18]), 11% (CI: [7,18]), and 15% (CI: [8,28]), respectively. Overall, the AUCs for ASGE 2010, ASGE 2019, and ESGE guidelines are 0.65, 0.74, and 0.73, respectively.

CONCLUSION

ASGE 2019 and ESGE guidelines have comparable performance, with their key strength being the ability to rule out CBDS in low-risk patients, allowing these patients to proceed with cholecystectomy without additional workup. All guidelines have limited specificity in identifying patients at high risk for CBDS and cannot reliably select patients for upfront ERCP.

PROSPERO REGISTRATION

CRD42024581409.