Methodological considerations in studying natural history of HBeAg‐negative and ‐positive chronic hepatitis B: authors' reply

To the editor: We read with great interest the report of Croagh et al. (1), which confirmed a positive correlation between high HBV-DNA levels and the presence of significant liver fibrosis in HBeAg-negative, but not in HBeAg-positive disease. This study differed from REVEAL(2) in that all patients had histological quantification of fibrosis rather than ultrasound and over half the patients had HBeAg-positive disease. We offer the following comments. Firstly, it may not be suitable to identify the predictors of disease progression in a chronic progressive disease through a cross-sectional study. Many studies have shown that the natural history of chronic HBV infection varies significantly at an individual level (3). The risk of liver inflammation and fibrosis in chronic hepatitis B is the complex result of many interacting factors and HBV-DNA level is only one of them (1). As so many other variables affect the natural history, just examining viral load in a crosssectional study produces an oversimplified and potentially incorrect conclusion about disease progression. In addition, this study of an Australian clinical cohort was a heterogeneous one, with all major genotypes (A–D) being represented. It is well known(3) that Genotype D infection prevails in the Mediterranean region (>90%) and its associated HBeAg-negative precore mutations can lead to cirrhosis and hepatocellular carcinoma. Contrary to Southeast Asia, HBV vertical transmission is rare in the Mediterranean region where horizontal transmission in early childhood is the predominant mode of HBV transmission, accounting for the vast majority of chronic HBV infections (>90%). Moreover, the natural history of chronic HBV infection varies with age of infection (4). Although Croagh et al. pointed out that “the ethnic makeup of these groups did not differ significantly (P = 0.189), with the majority (>74%) of the patients being of Asian ethnicity in all the groups”, we think that the possible impact of the complex ethnic makeup of the patients on the findings cannot be ignored. In summary, we contend that the use of a cross-sectional study method and the ethnic heterogeneity of their cohort may reduce the validity of their conclusions.