Early Prediction of Biliary Atresia Using Combi-Elastography in Infants ≤ 60 Days of Age

TL;DR

This study introduces a new model using combi-elastography and lab tests to predict biliary atresia in infants under 60 days old.

Contribution

The novel predictive model integrates combi-elastography with clinical indicators for early detection of biliary atresia.

Findings

A nomogram model using GGT, gallbladder morphology, and fibrosis index achieved an AUC of 0.887 in predicting biliary atresia.

The model showed high sensitivity (86.4%) and specificity (76.0%) in differentiating biliary atresia from other cholestatic diseases.

Abstract

Background: To improve the early recognition of biliary atresia (BA) and timely treatment, this study developed a predictive model integrating combi-elastography, a novel form of elastography, to distinguish biliary atresia from other cholestatic liver diseases (non-BA) in infants. Method: A total of 69 children aged < 60 days with cholestatic hepatitis were retrospectively enrolled. All patients underwent conventional ultrasonography, combi-elastography, and laboratory testing. The variables were selected using logistic regression to construct a nomogram model, and the performance of the model was evaluated. Results: Multifactorial logistic regression analysis indicated that GGT (p = 0.015), the gallbladder morphology (p = 0.017), and the fibrosis index of the combi-elastography (p = 0.017) could be used as independent predictors to differentiate BA from other causes of cholestasis. A nomogram model constructed with these three indexes showed better performance, with an area under the operating characteristic curve (AUC) of 0.887 (0.823, 0.952) (p < 0.001), sensitivity of 86.4%, and specificity of 76.0%. Using 1000 Bootstrap resamples for internal validation of the model, the predictive effect of the nomogram model to identify biliary atresia from other cholestatic liver diseases was in good agreement with the actual situation. Decision-curve analysis showed that the use of the nomogram model to predict biliary atresia gained more clinical value at a risk threshold of 0.10–0.80. Conclusion: The nomogram constructed integrating combi-elastography and liver function indices shows promising value for predicting the risk for developing biliary atresia.

Figures