In their recent paper [1], Martinot and associates remark that hepatitis G virus (HGV) infection has a minor etiologic role in liver disease and little direct pathogenic effect on the liver. They found that HGV infection did not influence the severity of liver disease in patients with chronic hepatitis due to hepatitis C virus (HCV). The pathogenic role of HGV in isolated infection is actually unclear.

We had a chance to assess, over a long period, the course of isolated HGV infection in a 35-year-old health care worker and to evaluate the changes over time in the HGV RNA viral load and in the 5'-untranslated (UTR) HGV region. Aminotransferase levels in this patient were occasionally found to be twice the upper limit of normal in 1983 and have since remained abnormal. In 1991, the patient was first admitted for investigation of his disease. He had no history of alcohol or illicit drug abuse, blood transfusions, or familial liver disease. Alanine aminotransferase (ALT) and -glutamyltransferase (GGT) values were abnormal (Figure 1). At that time, and on subsequent studies (1992, 1993, 1994, and 1997), the patient was seronegative for hepatitis B surface antigen, antibodies to hepatitis B surface antigen, antibodies to hepatitis B core antigen IgG and IgM, and hepatitis B virus DNA (by core region polymerase chain reaction [PCR]); antibodies to HCV (by enzyme immunoassay); HCV RNA (by 5'-UTR nested PCR); and IgM, antibodies to hepatitis A virus, and antibodies to hepatitis E virus. Non-organ-specific autoantibodies were absent, and serum ceruloplasmin, ferritin, and ₁-antitrypsin levels were normal. On liver biopsy, liver structure was normal and no necroinflammation was apparent. Mild centrilobular cholestasis, with ballooning of hepatocytes but no bile duct cell damage, and focal microvesicular steatosis were also noted. No therapy was given. Because of abnormal liver enzyme levels, biopsy was repeated in 1993 and 1997. Histologic findings were changed, and no "hepatitic" features or any progression to fibrosis or cirrhosis and persistent mild steatosis was seen.

View larger version (19K): [in this window] [in a new window]   Figure 1. ALT = alanine aminotransferase; GGT = -glutamyltransferase; HGV = hepatitis G virus; UTR = untranslated region. The upper line represents the ALT level; the lower line represents the IgG level.

Stored serum samples were retrospectively tested for HGV RNA by PCR with primers for 5'-UTR, NS3, and NS5. Semiquantitation was performed by end-titer dilution on 5'-UTR amplicons; titers were expressed as reciprocals of the highest positive dilution. All available samples since 1991 tested positive for HGV RNA by all three sets of primers. Viral titers tended to decrease over the years and seemingly did not relate to ALT or GGT levels or to liver damage. A difference of 3.27% with respect to a previously reported sequence of HGV [2] was observed. An average sequence divergence of 0.33% per year was observed in a 214 base-pair fragment of the 5'-UTR region.

Thus, after at least 6 years (and probably more than 14 years) of persistent HGV infection, no significant necroinflammation or progressive liver damage developed in this patient. Even in the absence of viral interference, such as HCV co-infection, HGV replication declines over the years while the virus undergoes mutations in a region that should be highly conserved. This long, benign course would again argue against a major role for HGV as a cause of chronic liver disease [3].