We appreciate Drs. Lai and Yuen's comments. We agree that lifelong maintenance therapy would be preferred when safe, affordable, and effective treatment is available to prevent disease recurrence. However, the safety of lifetime use of nucleoside or nucleotide analogues for hepatitis B has not been established. Adefovir is associated with a small risk for nephrotoxicity. Entecavir at high doses in rodents was associated with a variety of tumors, prompting the manufacturer to pledge a 10-year follow-up study to establish its long-term safety in humans. Furthermore, currently available hepatitis B treatments are very expensive, and their efficacy in maintaining viral suppression is diminished by the selection of drug-resistant mutations during long-term therapy. Even in cancer that has the propensity for late recurrence, treatment is discontinued after disease remission or a finite duration of consolidation therapy (1).

We completely agree that cirrhosis and hepatocellular carcinoma are the key outcomes of chronic HBV infection. However, as Drs. Lai and Yuen indicated, a persistently high HBV DNA level is the most important predictor of these outcomes. Thus, the risk for adverse outcomes after treatment discontinuation is related to the risk for reactivation of viral replication.

Drs. Lai and Yuen cited a recent study (2) in Greek children showing that 21 of 34 Thracian Muslims (62%) had an HBV-infected mother. However, this study did not determine whether the mothers were the source of infection and whether the infection was acquired perinatally. In the same study, only 47 of 121 Thracian Christians (39%) had an HBV-infected mother. Other studies from Italy and Greece, where HBV genotype D is preponderant, reported increased prevalence of chronic HBV infection with age, with the highest prevalence in those older than 40 years of age. This supports the idea that HBV infection in these countries is mostly acquired during childhood and adult life (3, 4).

Although the strict definition of HBeAg seroconversion was not clearly described in the 2007 American Association for the Study of Liver Diseases Practice Guidelines (5), a recent analysis of 74 patients who lost HBeAg after 48 weeks of entecavir therapy found that 70 had HBeAg seroconversion, 71 had undetectable HBV DNA by polymerase chain reaction, and 63 had normalization of aminotransferases. These findings support the idea that most patients who achieve HBeAg seroconversion during nucleoside or nucleotide analogue therapy meet the strict definition of HBeAg seroconversion (6).