Evaluation Of The Diagnostic And Predictive Performance Of Non-Invasive Models For Assessing Liver Fibrosis In Patients With Chronic Hepatitis B Virus Infection

Abstract

The performance of non-invasive models in the diagnosis and prediction of liver fibrosis have not been evaluated in all populations. This study evaluated the accuracy of gamma-glutamyl transpeptidase-to-platelets ratio index (GPRI), and S-index in the diagnosis and prediction of liver fibrosis in patients with chronic hepatitis B virus infection (CHBV). Fifty patients with CHBV and 40 control were recruited into this case-control study. Albumin (ALB), alkaline phosphatase (ALP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), bilirubin, gamma-glutamyl transpeptidase (GGT) and platelet count (PLT) were determined by colorimetric methods and Sysmex XS-10000 haematology automated analyzer respectively. GPRI and S-index were computed. Data were analyzed using ANOVA, Pearson’s correlation and ROC curve at p<0.05. ALP, ALT, AST, GGT, GPRI and S-Index were higher (p<0.05), while ALB was lower (p<0.05) in patients with liver fibrosis (LF), compared to patients with CHBV. ALP, GGT, GPRI and S-index were higher (p<0.05), while ALB and PLT were lower (p<0.05) in patients with CHBV compared to the controls. ALP, ALT, AST, GGT, GPRI and S-index were higher (p<0.05), while ALB was lower (p=0.000) in patients with LF compared to the controls. The area under the curve (AUROC) for S-index and GPRI were (AUC=0.835, P=0.000), GPRI (AUC=0.778, P=0.003). This study shows that both S-index and GPRI are good test instruments for evaluating liver fibrosis in patients with chronic hepatitis B. S-index was a better marker than GPRI for predicting liver fibrosis.