To the Editor:
    I read the paper by Chen et al^[1] with great interest. The authors performed a retrospective study to evaluate the short-term efficacy of antiviral therapy with entecavir for patients with severe chronic hepatitis B (CHB), and found that the survival at 3 months, the improvement of the model for end-stage liver disease (MELD) score and liver function tests in the entecavir group were not significantly different from those of the control group without antiviral therapy, although entecavir therapy could inhibit hepatitis B virus (HBV) replication rapidly.
      Included in this study were the patients with severe CHB whose serum bilirubin levels were more than ten times the upper limit of normal value (>171 µmol/L). The mean baseline bilirubin levels in the entecavir group and control group were 418.83±174.51 and 316.91±154.43 µmol/L, respectively and MELD scores were 21.76±4.68 and 17.33±6.61, respectively. It seemed that the bilirubin levels and MELD scores in the entecavir group were much higher although the differences were not statistically significant (P=0.50 and 0.13, respectively). Univariate analysis showed that the patients who survived less than 3 months had significantly higher baseline bilirubin levels, prothrombin time and MELD scores. The paper by Tseng et al^[2] also reported the value of these factors to predict early mortality in patients with decompensated HBV cirrhosis. Baseline HBV DNA level wasn't a predictive factor of early mortality in both studies. However, another study found that elevated bilirubin and creatinine levels as well as the presence of detectable HBV DNA (by the bDNA assay) pretreatment were significantly associated with the 6-month mortality in patients with decompensated CHB treated with lamivudine.^[3]
    Several studies have shown that inhibition of viral replication with nucleos(t)ide analogues such as lamivudine could potentially decrease hepatic necroinflammation and result in a significant improvement of liver function in patients with decompensated HBV cirrhosis.^[3-5] However, clinical efficacy was not obvious until 3-6 months later. So, antiviral therapy should be started promptly without the need of emphasis on high HBV DNA or elevated ALT levels.[3] The results of this study didn't show the efficacy of entecavir on the survival at 3 months. We couldn't deny the efficacy of entecavir by this short-term follow-up study and further longer studies are needed to evaluate whether antiviral therapy is efficient to improve the long-term survivals.