Noninvasive models for the prediction of liver fibrosis in patients with chronic hepatitis B

Objective To evaluate the diagnostic accuracy of aspartate aminotransferase(AST)/ alanine transaminase (ALT), AST to platelet ratio index (APRI), fibrosis-4 score (FIB-4) and gamma-glutamyl transpeptidase to platelet count ratio (GPR) for hepatic fibrosis in patients with chronic hepatitis B (CHB).Methods A total of 1210 CHB patients who underwent liver biopsy were divided into two groups: patients with no significant fibrosis (control group) and patients with significant fibrosis, and routine laboratory tests were retrospectively included. Logistic regression models were used for the prediction, and the area under the receiver operating characteristic (AUROC) was used to assess the diagnostic accuracy.Results A total of 631 (52.1%) and 275 (22.7%) patients had significant fibrosis (>= S2) and advanced fibrosis (>= S3), respectively. The GPR showed significantly higher diagnostic accuracy than that of APRI, FiB-4, and AST/ALT to predict >= S2(significant fibrosis) and >= S3 fibrosis(advanced fibrosis), with an AUROC was 0.69 (95%CI: 0.66-0.71) and 0.72 (0.69-0.75), respectively. After stratified by the status of HBeAg ( positive or negative), GPR, APRI, and FiB-4 showed improved predicting performance for significant fibrosis and advanced fibrosis in HBeAg positive patients, with the most significant improvement was shown for GPR in predicting significant fibrosis (AUROC = 0.74, 95%CI: 0.70-0.78).Conclusions Among the four noninvasive models, GPR has the best performance in the diagnosis of hepatic fibrosis in CHB patients and is more valuable in HBeAg-positive patients.