Validation of FibroTest to Diagnose Liver Fibrosis in Patients with Chronic Hepatitis B

赵连荣,许德军,陆忠华,赵鸿,窦晓光,郎振为,王贵强
DOI: https://doi.org/10.3969/j.issn.1005-2194.2007.16.015
2007-01-01
Abstract:Objective To testify the diagnostic value of FibroTest in patients with CHB by comparing their results with histological features.Methods During Aug 2002 to Dec 2005 cases of chronic hepatitis B were colllected from Peking University First Hospital,No.5 People's Hospital of Anyang,Henan and Wuxi Hospital of Infectious Disease.All of them underwent liver biopsy with a blood sample taken simultaneously.The serum level of α2-macroglobulin,haptoglobin,apolipoprotein AⅠ,total bilirubin and gamma-glutamyl transpeptidase were tested.These results together with age and sex of the patients were put into the algorithm and final results of FibroTest were computed.Three different endpoints were studied according to liver fibrosis stage,namely significant fibrosis(S2 to S4),extensive fibrosis(S3 to S4)and cirrhosis(S4).With liver biopsy as the gold standard,ROC curves were delineated for different endpoints.The area under the ROC curves reflected its diagnostic values.The diagnostic value of FibroTest was compared with AST to platelet ratio index(APRI).Results One hundred and twenty-three patients were included.The distribution of their fibrosis stage was as follows,S0:25(20.3%);S1:27(22.0%);S2:31(25.2%);S3:29(23.6%);S4:11(8.9%).That means seventy-one patients(57.7%)had significant fibrosis(S2 to S4),forty(32.5%)had extensive fibrosis(S3 to S4)and eleven(8.9%)had cirrhosis(S4).The AUCs of FibroTest for significant fibrosis,extensive fibrosis and cirrhosis were 0.814(95%CI:0.740~0.888,P0.01),0.824(95%CI: 0.749~0.898,P0.01),0.723(95%CI:0.575~0.870,P=0.015)respectively.Comparatively,the AUCs of APRI were 0.715(95%CI:0.625~0.805,P=0.001),0.725(95%CI:0.631~0.818,P=0.002)and 0.646(95%CI:0.497~0.795,P0.05),respectively.Conclusion FibroTest identifies CHB patients with and without significant fibrosis with high accuracy.FibroTest may prevent liver biopsies in 45.5% of patients while maintaining 87.5% accuracy by restricting biopsy to patients with intermediate scores(0.31 and 0.72).
What problem does this paper attempt to address?