|| Checking for direct PDF access through Ovid
To assess the circulation of human herpesvirus (HHV)-8 infection over the years, two seroprevalence surveys were conducted, which tested sera from HIV-infected individuals recruited 10 years apart (206 individuals from 1986 to 1988 and 177 individuals from 1997 to 1998). For all patients, antibodies to hepatitis C virus (HCV), hepatitis B virus (HBV), and HHV-8 lytic and latent antigens were evaluated.HHV-8 seroprevalence was higher among individuals recruited in the 1990s (31.6% for anti-lytic, 8.5% for anti-latent antibodies) compared with similar findings in those seen in the late 1980s (14.6% and 3.4% for anti-lytic and anti-latent antibodies, respectively), with a twofold increase of the risk of HHV-8 infection. However, the increase was observed only among injecting drug users, whereas seroprevalence tended to slightly increase among those infected by sexual contact. At univariate analysis, time of recruitment and being homosexual men were factors associated with HHV-8 infection, an association that remained after adjusting for age. HBV infection was significantly associated with HHV-8 infection (odds ratio [OR], 2.2; 95% confidence interval [CI], 1.3-3.6), whereas those infected with HCV had a lower probability of having HHV-8 antibodies (OR, 0.3; 95% CI, 0.20-0.6). After controlling for age and gender, time of recruitment remained independently associated with HHV-8 infection among injecting drug users.In conclusion, HHV-8 seroprevalence appears to be increased during 10 years among HIV-infected injection drug users but not among homosexual men, who remain those at the highest risk of infection.