Serologic examination of hepatitis B infection and immunization in HIV-positive youth and associated risks. The Pediatric AIDS Clinical Trials Group Protocol 220 Team.

This seroprevalence report examines serologic evidence of hepatitis B immunization or infection and associated demographic/behavioral factors in adolescent (aged 12-20) subjects enrolled in a nontherapeutic clinical trial at 43 Pediatric AIDS Clinical Trials Group (PACTG) clinical centers. Subjects (n = 94) infected with the human immunodeficiency virus (HIV) through sexual activity were categorized as hepatitis B virus (HBV)-immunized, HBV-infected, or nonimmune by hepatitis B serology performed on specimens collected within the subject's first 48 weeks on study (1993-1995). Sixteen percent of the 94 serologically classified subjects were immunized; 19% HBV-infected; 65% nonimmune. Of the three risk factor scores examined (sociodemographic, sexual, and substance abuse), substance use alone demonstrated a significant difference among groups (despite virtually no reported injecting drug behavior), with the sexual risk score exhibiting marginally significant differences. Logistic regression analysis (restricted to nonimmunized subjects) showed that male-male sexual activity raised the odds of HBV infection by a factor of 5.14 (95% confidence interval [CI]: 1.45-18. 23) relative to heterosexual activity; and that for every one point increase on the substance abuse risk scale the odds of infection increased 5% (95% CI: 0.99-1.10). The HBV infection rate in PACTG 220 HIV-positive females is twice United States population-based rates; the rate in PACTG 220 HIV-positive males is nearly seven times higher. Past immunization efforts in this population appear to have been based on sexual activity volume without regard to injecting-drug use in sex partners.