Liver stiffness measurement (LSM) may effectively correlate to the presence of liver fibrosis, but it is controversial to use for the prediction of clinical outcomes. Therefore, we aimed to evaluate the predictive value of liver stiffness for the regression of liver fibrosis.

In this study, we collected data from a clinical cohort of patients who are received anti-virus therapies for 48 weeks. 180 naive chronic hepatitis B (CHB) patients, who received paired LSM and liver biopsy with pre- and post-treatments were analyzed. Two methods (FibroScan and iLivTouch) test LSM.

The area under the receiver operating characteristics curve (AUROC) of changing LSM for fibrosis regression is higher in advanced fibrosis patients (F5/6) than in moderate fibrosis patients (F3/4) in both FibroScan (0.719, 95%CI, 0.590–0.848; P=0.003; vs 0.617, 95%CI, 0.379–0.856, P=0.282) and iLivTouch (0.707, 95%CI, 0.567–0.847; P=0.011; vs 0.583, 95%CI, 0.422–0.744; P=0.377). A higher kappa value was received in advanced stage than in moderate stage both in FibroScan (0.392, P=0.001 vs 0.265, P=0.053) and iLivTouch (0.326, P=0.019 vs 0.030, P=0.833). Cut-off set as 4.10 kPa (sen, 69.4%; spe, 73.9%) in FibroScan, as 4.25 kPa (sen, 56.8%; spe, 72.2%) in iLivTouch.

The changing LSM can be used for predicting the liver fibrosis regression in advanced stage of CHB patients.