We thank Fazili et al. for their interest in our article. The development of antibodies to hepatitis B e antigen (anti-HBe) and hepatitis B surface antigen (anti-HBs) were lower in the lamivudine-treated group relative to placebo, although these did not reach statistical significance. We concur with Fazili et al. that it is an important observation, and it seems plausible that interruption of the protective host immune response may be a consequence of pharmacologic therapy. Because virus-specific antibody-producing B cells are enriched early after acute viral infection, 1 the host immune system needs to be exposed to viral antigen during the early phase in order to induce production of neutralizing antibodies. Therefore, giving lamivudine early may inhibit the production of neutralizing antibodies to some extent. 2 A previous study also reported that after the use of lamivudine in treating acute viral hepatitis B, 18.18% of patients lost serum HBsAg but did not develop anti-HBs during treatment and follow-up. 3 Nevertheless, it has also been reported that anti-HBs did not develop in 10% of untreated patients with acute hepatitis B. 4 As rightly mentioned, only a long-term follow-up of such patients would help us know the risk of reactivation of hepatitis B virus (HBV) in patients treated with antiviral therapy.
We do believe however that treatment of patients with severe acute hepatitis B with lamivudine is not harmful. Lamivudine or other more potent antivirals may be useful in a small subset of patients with acute hepatitis B who have more severe disease at presentation and high HBV DNA. Although there are no studies defining the patients with more severe disease, we suggest that such patients might include individuals with fulminant acute hepatitis B, 5 or individuals who fulfill any 2 of the following criteria: 3,6 (1) hepatic encephalopathy; (2) serum bilirubin ี 10.0 mg/dL; and
(3) international normalized ratio (INR) ี 1.6, especially if it is increasing.
Because patients belonging to the above categories would often need liver transplantation, it would be best to achieve, through use of potent antiviral drugs, HBV DNA levels which are undetectable or as low as possible at the time of liver transplantation, as measured with current real-time polymerase chain reaction methods. Whether such a therapeutic approach may translate into spontaneous recovery in a proportion of patients awaiting transplantation can only be predicted by the use of antiviral drugs in proper RCTs including such a group of patients.