Serum fibrosis markers, such as the enhanced liver fibrosis (ELF) test, have been suggested as alternatives for liver biopsy (LB) in assessing liver fibrosis. We investigated the efficacy of the ELF test in predicting development of liver-related events (LREs) in patients with chronic hepatitis B (CHB). A total of 170 patients (103 men; 60.6%) with CHB who underwent LB and serological tests for determining ELFs were enrolled. All patients were followed up to monitor LRE development, defined as hepatic decompensation, hepatocellular carcinoma, and/or liver-related death. The mean age was 45.3 years. During the follow-up period (median, 41 months), 39 (22.9%) patients experienced LREs. In patients with LREs, age, proportion of male gender, ELF test results, age-spleen-platelet ratio (ASPRI), liver stiffness (LS) value, and proportion of histological cirrhosis were significantly higher than those in patients without LREs (all P < 0.05). Areas under the receiver operating characteristic curves to predict LRE development were 0.808 for the ELF test, 0.732 for LS value, 0.713 for histological fibrosis stages using Batts and Ludwig's scoring system, and 0.687 for ASPRI. On multivariate analysis, along with age, the ELF test was an independent predictor of LRE development (adjusted hazard ratio [HR], 1.438; P < 0.001). When we applied a three-tier stratification of our study population using cut-off ELF values of 8.10 and 10.40, patients with low (P = 0.002; adjusted HR: 0.045; 95% confidence interval [CI]: 0.006-0.330) and intermediate (P < 0.001; adjusted HR: 0.239; 95% CI: 0.122-0.469) ELF range were found less likely to develop LREs, compared to those with high ELF range.
Conclusion: ELF is useful in a noninvasive prediction of LRE development. Transient elastography showed a statistically similar prognostic performance for LREs as the ELF, but other noninvasive tests were inferior.
© 2014 by the American Association for the Study of Liver Diseases.