Hepatitis D-related cirrhosis and risk of hepatocellular carcinoma development.

Abstract

Editor: The article published by Dr Caturelli and colleagues in the March 2003 issue of Radiology (1) prompted me to write a few critical comments that I would like to share with the editorial board and the readers. It is known that a coarse nodular ultrasonographic (US) pattern represents a major added risk factor for hepatocellular carcinoma (HCC) development in patients with hepatitis C virus (HCV)–related cirrhosis. In their study, Dr Caturelli and colleagues (1) suggested that this ecographic pattern, even if encountered frequently in hepatitis D virus (HDV)–related cirrhosis (51% of patients), is not associated with an increased risk of HCC development as it is in cirrhotic patients with HCV and hepatitis B virus (HBV) and to a lesser extent in patients with alcohol-related cirrhosis (table 3 of the article). These conclusions are based on their retrospective analysis in which they first assessed the prevalence of the coarse nodular US pattern in cirrhotic patients grouped according to disease origin (table 2). Next, they analyzed the relationships between HCC development and US patterns (table 3), thus giving the prognostic implications of this US pattern in terms of HCC risk for each subgroup (HCV-, HBV-, HDV-, and alcohol-related cirrhosis and primary biliary cirrhosis). To explain the lower risk of HCC in patients with HDVrelated cirrhosis and coarse nodular US pattern, the authors suggested that the ability of HDV to suppress HBV replication could represent a protective mechanism to lower the risk of HCC development in patients with HBV and HDV superinfection. A careful revision of the data reported by Dr Caturelli and colleagues may suggest a different explanation of these findings. It is widely accepted that age represents a major risk factor for the development of HCC. Stroffolini et al (2) have demonstrated that, in Italy, the mean age at which HCC appears in patients with HCV is 65 years (male to female ratio, 2.8:1), while the mean age at which HCC appears in patients with HBV is 59 years (male to female ratio, 10.4:1). If we analyze the demographic data reported in table 1, the authors do not make any comments on patient age, which is presumably substantially lower in patients with HDV (mean age, 46 years 11 [standard deviation]) compared with those with HCV (mean age, 63 years 9) and HBV (mean age, 55 years 10) and those who abuse alcohol (mean age, 54 years 12). These respective ages agree with the expected lower incidence of HCC found in patients with HDV (4%) than that observed in patients with HBV or HCV (16%) and those who abuse alcohol (8%). In conclusion, the younger age, per se, possibly accounts for the lower tumor incidence observed in patients with HDV. In this context, statistical analysis performed after stratification of patients according to age would help to clarify the real effect of the coarse nodular US pattern as a risk factor for HCC in patients with HDV-related cirrhosis.