Reduction of liver stiffness by antiviral therapy in chronic hepatitis B.

BACKGROUND: Liver stiffness (LS) has been reported to correlate with fibrosis stage (F). The correlation between LS and fibrosis stage and the reduction of LS by antiviral therapy were examined in patients with hepatitis B infection. METHODS: LS was measured by FibroScan in 212 patients infected with hepatitis B virus. Liver biopsies were done in 51 patients. Changes of LS were assessed in 29 patients treated with nucleotide or nucleoside analogs and 52 patients without antiviral therapy. RESULTS: LS was significantly correlated with fibrosis stage (ρ = 0.686, P < 0.0001). The optimal cut-off values of LS were 7.1 kPa for F ≥ 2, 10.7 kPa for F ≥ 3, and 16.0 kPa for F4. LS was significantly reduced by antiviral therapy, from 12.9 (range 6.2-17.9) kPa to 6.6 (4.4-10.3) kPa measured at an interval of 512 (range 366-728) days (P < 0.0001). Eleven of 19 (58%) patients with baseline fibrosis stages of F3-4 deduced from LS had 2-point or greater reductions of deduced stage at the last LS measurement. The change ratio of hyaluronic acid (P = 0.0390) was associated with a 2-point or greater reduction of deduced fibrosis stage. Without antiviral therapy, LS tended to increase, increasing from 6.1 (range 3.9-8.5) kPa to 6.3 (range 4.4-9.7) kPa at an interval of 422 (range 358-709) days (P = 0.0682). CONCLUSIONS: LS was significantly correlated with fibrosis stage in patients with chronic hepatitis B. The reduction of LS by antiviral therapy was significantly correlated with the reduction of hyaluronic acid. Thus, we conclude that LS can be useful to assess the progression and regression of liver fibrosis stage noninvasively.

A new index for non-invasive assessment of liver fibrosis.

AIM: To construct and evaluate a new non-invasive fibrosis index for assessment of the stage of liver fibrosis. METHODS: A new fibrosis index (Fibro-Stiffness index) was developed in 165 of 285 patients with chronic hepatitis C, and was validated in the other 120 patients where liver biopsy was performed. Its usefulness was compared with liver stiffness (LS) measured by FibroScan, the aminotransferase-to-platelet ratio index, the Forns index and the FibroIndex. RESULTS: The Fibro-Stiffness index consists of LS, platelet count and prothrombin time. The values of the Fibro-Stiffness index differed significantly between neighboring fibrosis stages except F0-F1. The area under the receiver operating characteristics curves of the Fibro-Stiffness index for prediction of F ≥ 2 (0.90), F ≥ 3 (0.90) and F = 4 (0.92) in the estimation group and those for F ≥ 3 (0.93) and F = 4 (0.97) in the validation group were the highest among the 5 methods examined. The accuracy of the Fibro-Stiffness index had highest values for F ≥ 2, F ≥ 3 and F = 4 in both the estimation and validation groups. The diagnostic performance for F = 4 was improved by a combination of the Fibro-Stiffness index with serum hyaluronic acid level. CONCLUSION: The Fibro-Stiffness index was constructed and validated. It showed superior diagnostic performance to other indices for F ≥ 2, 3 and 4.