BACKGROUND/AIMS: Liver biopsy is considered a gold standard for fibrosis staging, but it has a high risk of morbidity. Therefore, there is an interest in developing noninvasive markers for the prediction of liver fibrosis stages. METHODS: Hyaluronic acid, ferritin, N-acetyl-β-D-glucosaminidase, β-glucuronidase, glucosamine, aspartate transaminase, and alanine transaminase were assayed in 210 individuals with chronic hepatitis C infection. Statistical analysis was carried out by logistic regression and receiver-operating characteristic curves. RESULTS: The best linear combination of only significant blood markers was used for the determination of the fibrosis discriminant score; score=[1.64 (numerical constant)-0.002×hyaluronic acid (pg/l)-2.68×β-glucuronidase (µmol/ml/min)-0.026×glucosamine (µg/dl)-0.001×ferritin-0.033 (ng/ml)×aspartate transaminase/alanine transaminase]. The selected fibrosis discriminant score function correctly classified 81% of patients with severe liver fibrosis at a discriminant cut-off score=0.55 (i.e. less than 0.55 indicated mild liver fibrosis and greater than 0.55 indicated severe liver fibrosis), with a sensitivity of 100% and a specificity of 73%. CONCLUSION: A simple fibrosis index can be useful to select hepatitis C virus-infected patients with a very low risk of significant fibrosis in whom the protocol of liver biopsies may be avoided.
BACKGROUND/AIMS: Liver biopsy is considered a gold standard for fibrosis staging, but it has a high risk of morbidity. Therefore, there is an interest in developing noninvasive markers for the prediction of liver fibrosis stages. METHODS:Hyaluronic acid, ferritin, N-acetyl-β-D-glucosaminidase, β-glucuronidase, glucosamine, aspartate transaminase, and alanine transaminase were assayed in 210 individuals with chronic hepatitis C infection. Statistical analysis was carried out by logistic regression and receiver-operating characteristic curves. RESULTS: The best linear combination of only significant blood markers was used for the determination of the fibrosis discriminant score; score=[1.64 (numerical constant)-0.002×hyaluronic acid (pg/l)-2.68×β-glucuronidase (µmol/ml/min)-0.026×glucosamine (µg/dl)-0.001×ferritin-0.033 (ng/ml)×aspartate transaminase/alanine transaminase]. The selected fibrosis discriminant score function correctly classified 81% of patients with severe liver fibrosis at a discriminant cut-off score=0.55 (i.e. less than 0.55 indicated mild liver fibrosis and greater than 0.55 indicated severe liver fibrosis), with a sensitivity of 100% and a specificity of 73%. CONCLUSION: A simple fibrosis index can be useful to select hepatitis C virus-infectedpatients with a very low risk of significant fibrosis in whom the protocol of liver biopsies may be avoided.