| Summary: | Abstract Background Serum chitinase-3-like protein-1 (CHI3L1) level is significantly correlated with fibrosis in chronic liver diseases. Therefore, we aimed to develop a CHI3L1-based model to predict significant liver fibrosis in the general population. Methods This retrospective cross-sectional study enrolled 6361 participants (4452 training and 1909 validation). Least absolute shrinkage and selection operator (LASSO) with variance inflation factor (VIF) analysis was employed for variable selection. Multivariable logistic regression analyses were used to construct a nomogram model. The clinical utility of the nomogram model was compared with that of the fibrosis-4 index (FIB-4) and non-alcoholic fatty liver disease fibrosis score (NFS) using receiver operating characteristic curves with comparisons of areas under the curve (AUCs), decision curve analysis (DCA), clinical impact curves (CIC), and stratified subgroup analysis. Results The nomogram model was composed of five variables: waist circumference, body mass index, aspartate aminotransferase, homeostatic model assessment of insulin resistance, and CHI3L1. The model obtained AUCs of 0.795 and 0.802 in the training and validation cohorts, respectively, which were significantly greater than those of FIB-4, NFS, and CHI3L1 alone, ranging from 0.573 to 0.612. Furthermore, DCA demonstrated greater net benefit across clinically relevant threshold, while CIC confirmed robust risk stratification. Consistent performance was observed in the metabolic dysfunction-associated steatotic liver disease (MASLD) population. Conclusions This novel nomogram showed good performance and clinical utility in predicting significant liver fibrosis and could be used as an ideal screening tool for the general population.
|
|---|
