Purpose: Few studies have evaluated the validity of noninvasive models for the diagnosis of liver fibrosis in chronic hepatitis B (CHB) patients, with controversial results. In this study, we evaluated the ability of seven noninvasive models in staging liver fibrosis in a large cohort of CHB.
Methods: This is a retrospective study. A total of 1168 severe CHB patients with a clear diagnosis of liver fibrosis were included in this study. Data from routine laboratory tests were collected to establish noninvasive models. The stage of fibrosis was defined by the Metavir scoring system. Fibro-quotient, AST/ALT ratio, AST to PLT ratio index (APRI), cirrhosis discriminant score, age-PLT index (API), fibrosis index based on the four factors (FIB-4), and Lok's model were adapted as noninvasive models in this study.
Results: FIB-4 (rs=0.542), API (rs=0.427), and Lok's model (rs=0.452) showed a higher positive correlation with liver fibrosis in CHB patients than the other models. APRI, FIB-4, and Lok's model were effective in distinguishing fibrotic stage. APRI, API, FIB-4, and Lok's model were the most effective models in distinguishing significant (S1, S2) and extensive (S3, S4) fibrosis, with area under receiver-operating characteristic values of 0.721, 0.727, 0.789, and 0.712, respectively. However, only FIB-4 and Lok's model showed higher sensitivity, specificity, positive predictive value, and negative predictive value at the cutoff value of 1.433-1.858 and 0.415-0.511, respectively.
Conclusion: FIB-4 and Lok's model are the most effective models for distinguishing significant and extensive fibrosis, whereas APRI, FIB-4, and Lok's model are suitable for staging fibrosis in CHB patients.