Epidemiology and natural history of hepatitis G virus infection in chronic hemodialysis patients.

Unlabelled: Patients on chronic hemodialysis (HD) have recently been identified as having a high prevalence of hepatitis G virus (HGV) infection. The clinical significance of HGV in this population remains unclear, with no data available as to the acquisition and natural history of HGV infection in this group.

Aims: To assess the prevalence and risk factors of HGV in a large cohort of chronic HD patients, and to evaluate the incidence and clinical consequences of HGV over time in this population.

Methods: Paired sera from 292 patients undergoing chronic HD treatment in four units in the Los Angeles area were tested for HGV RNA before and after they had been on HD for a mean period of 9.7 +/- 1.9 months. HGV was tested by a single-step RT-PCR using two couples of primers located in two different portions (5'UTR, NS5a) of the genome. The amplified products were detected by hybridization with 5' biotin-labeled probes specific for each region.

Results: At study entry there were 50 HGV RNA-positive patients, thus the HGV prevalence was 17% (50/292). The multivariate analysis by ordinal logistic regression model showed association (p = 0.0013) between HGV RNA and the location of patients among the HD units. No other significant associations were observed. Three (3/50 = 6%) HGV RNA-positive patients at study entry and 3 (3/41 = 7%) at the end of the follow-up showed a mild increase of alanine aminotransferase (ALT) activity in absence of other apparent causes of liver damage. 35 (70%) out of 50 HGV viremic patients had persistently detectable viremia during the study period; 15 (30%) had non-persistently detectable HGV RNA in the second serum specimen. There was no significant difference between the patients with persistently detectable HGV RNA and those who showed non-persistently detectable HGV viremia with regard to demographic, clinical or virological features. Six patients without detectable HGV viremia at the start of the study showed de novo HGV infection during the follow-up, thus the HGV incidence was 3.07% per year. These individuals did not simultaneously acquire HBV or HCV markers; de novo HGV infection was not associated with other demographic, clinical or virological features. One (16.7%) out of 6 individuals with HGV acquisition had persistently raised ALT levels and chronic HBsAg positivity. The prevalence of HGV was 14% (41/292) at the end of the observation period.

Conclusions: The prevalence of HGV in our HD population was high; HGV positivity was strongly associated with the location of HD patients among the units; some HD individuals with current HGV infection showed biochemical signs of liver disease without other apparent causes. De novo acquisition of HGV occurred within HD units in the absence of evident parenteral risk factors for HGV other than their presence in the HD environment. A large portion of HGV viremic patients showed non-persistently detectable HGV viremia during the study. Acquisition of HGV was not associated with a rise in ALT activity unlike prior experience with de novo HCV in HD patients. Further investigations are warranted to explain the modes of HGV acquisition and the clinical significance of HGV in th HD population.