ABSTRACT
AIM: Assessment of liver fibrosis in chronic hepatitis C (CHC) patients is necessary before antiviral treatment. This study aimed to evaluate the effectiveness of eight non-invasive models (aspartate aminotransferase [AST]/alanine transaminase ratio [AAR], AST/platelet ratio index [APRI], fibrosis-cirrhosis index [FCI], fibrosis index [FI], fibrosis-4 [FIB-4] score, fibrosis quotient [FibroQ], King, and von Willebrand factor antigen (vWF-Ag)/thrombocyte ratio [VITRO] scores) for predicting fibrosis compared with liver biopsy and to create a new score for predicting different fibrosis stages with increased accuracy. METHODS: We prospectively studied 127 treatment-naive CHC patients who underwent liver biopsy. The AAR, APRI, FCI, FI, FIB-4, FibroQ, King and VITRO scores were calculated and correlated with fibrosis stages. A new score (VAP) was derived from vWF-Ag, AST, and platelets: [VAP = (AST (U/L) × vWF-Ag)/platelets (109 /L)]. RESULTS: Apart from AAR, readily available scores were correlated with liver fibrosis stages. VITRO (r = 0.62) and APRI (r = 0.46) showed the closest correlation. Our new (VAP) score significantly correlated with fibrosis stages (r = 0.702, P < 0.001). Compared to other scores, VAP had the highest area under the receiver operating characteristic curve, with 0.854, 0.921, 0.849, and 0.861 for mild (F1), significant (≥F2), advanced (≥F3) fibrosis, and cirrhosis (F4) respectively. At a cut-off value >1, VAP had 75.2% sensitivity and 100% positive predictive value for predicting mild fibrosis. At a cut-off value >2.3 for predicting cirrhosis, VAP had 73% sensitivity and 81.7% positive predictive value. CONCLUSIONS: The VAP score is a novel model that had higher diagnostic performance to predict different fibrosis stages and subclinical cirrhosis among CHC patients compared to the other studied scores and hence may offer a useful strategy to stratify patients who would benefit from direct-acting antivirals.
