The diagnostic accuracy of the Forns index, platelet count and AST to Platelet Ratio Index derived fibrosis index for the prediction of Hepatitis C virus-related significant liver fibrosis and cirrhosis

Abstract

Aim: To provide a simple fibrosis index combining the routine laboratory markers for predicting significant fibrosis (SF) and cirrhosis in patients with chronic HCV.

Methods: Platelet count, ALT, AST, AST to ALT Ratio, AST to Platelet Ratio Index (APRI), Forns index, FIB-4 and Age Platelet Index of 202 liver biopsy performed HCV-infected patients were reviewed. METAVIR classification was used to determine the stage of liver fibrosis. The predictive fibrosis index was constructed by multiple linear regression analysis (- 2.948 + 0.562 × Forns index + 0.288 × APRI + 0.006 × platelet count [10(9)/L]).

Results: Median (25th-75th interquartile range) age was 52 (42-59) years, and 61% were male. 65.8% (n = 133) had SF (F2-F4) and 23.3% (n = 47) had cirrhosis (F4). For discrimination of SF, AUROCs were: Fibrosis index = 0.869, Forns index = 0.837, APRI = 0.814, platelet count = 0.764. For cirrhosis, AUROCs were: Fibrosis index = 0.911, Forns index = 0.883, APRI = 0.847, platelet count = 0.827. A cut-off point of ≤ 1.2 for fibrosis index excluded SF in 89% of patients with sensitivity of 96%, while > 2.0 predicted SF in 88% of patients with specificity of 86%. Threshold of ≤ 1.9 excluded cirrhosis in 95% of patients with sensitivity of 94%, while > 2.7 showed cirrhosis in 88% of patients with specificity of 95%. In multivariate logistic regression analysis, OR (95% CI) of fibrosis index was 7.825 (3.682-16.629) for SF (p < 0.001) and was 8.672 (4.179-17.996) for cirrhosis (p < 0.001).

Conclusion: SF and cirrhosis were predicted with accuracy of 82% and 89% and were excluded with accuracy of 74% and 82% using this fibrosis index which may potentially decrease the need for liver biopsy in 76% and 83% of patients, respectively.