Reply

Abstract

In the study, ultrasonography of the abdomen was performed every 1-2 years for surveillance of hepatocellular carcinoma (HCC) or, more frequently, if the alpha-fetoprotein level rose above 20 lg/ L. This appeared not to be in accord with the recommendations of screening intervals of every 6 months by the Asian Pacific Association for the Study of the Liver or American Association for the Study of liver Diseases. 2,3 A high proportion of hepatic events were ascribed to variceal bleeding. If complied with the recommendations to prevent first variceal bleeding, these serious events could be effectively reduced. 4,5 Most important, baseline clinical characteristics varied significantly between both cohorts. Though baseline data appeared to be milder disease in the entire untreated control cohort than entecavir-treated cohort, patients with cirrhosis in the untreated control cohort were more severe than the entecavir-treated cohort, as expressed by Child-Pugh's score. There was no difference in hepatic events between both cohorts, except among patients with cirrhosis. Based on sensitivity analyses among patients with cirrhosis, the risk of hepatic events is greatly reduced by entecavir among patients with advanced liver disease, as reflected by low platelet counts, low serum albumin, high total bilirubin, higher Child-Pugh score, and Model for End-Stage Liver Disease score. These baseline clinical parameters were all worse in the untreated control cohort than the entecavirtreated cohort. Duration of follow-up was significantly longer in the untreated control cohort. All these factors may elicit significant bias between groups. Use of entecavir is proven to be beneficial for hepatitis B virus patients to reverse fibrosis and cirrhosis 6 ; however, the evidence of reduction in hepatic events and incidence of HCC, as revealed by the current study and previous work, 7 was still flawed with significant bias. Long-term follow-up of controlled studies would be more convincing. Reply:We thank Dr. Lo for his interest in our study. 1 His first point related to the frequency of abdominal ultrasonography for surveillance of hepatocellular carcinoma every 1-2 years, instead of every 6 months, according to the international guidelines. [2][3][4] This was the result of limited resources of radiology services in our public sector, which is often an issue in some regions. 5 Instead, alphafetoprotein (AFP) was tested during every visit. In our recent analysis, the specificity of AFP was as high as 99% at a cut-off value of 20 mg/L in patients receiving entecavir. 6