Retrospective study reveals a novel nomogram predicts advanced liver fibrosis in autoimmune hepatitis, indicating a potential improvement in assessment methods.
Key Points
The aim is to develop and validate a non-invasive nomogram for predicting advanced liver fibrosis in autoimmune hepatitis patients.
Conducted a retrospective study with patients who had liver biopsy.
Used LASSO regression for identifying predictors.
Employed multivariable logistic regression to establish independent predictors.
Evaluated performance using AUC, calibration curves, and decision curve analysis.
The nomogram was based on liver stiffness measurement, platelet count, and prothrombin time.
AUC of training set is 0.851 and validation set is 0.922, indicating excellent discriminatory ability.
Calibration was strong, and the nomogram demonstrated significant clinical net benefit.