Abstract
Background: There is limited evidence comparing hepatitis A seroprevalence among HIV-exposed uninfected (HEU), HIV-infected (HIV), and unexposed uninfected (HUU) children. This compromises rational vaccine decision-making. Methods: This study comprised a retrospective health facility-based population of children aged 1 month-12 years. Archival sera were tested for markers of acute (anti-HAV IgM) or past (total anti-HAV) HAV infection. Subgroup analysis was conducted based on perinatal HIV exposure or infection status. Results: Among 513 children, the median age was 10 (IQR: 4-25) months. The median maternal age was 29 (IQR: 25-34) years. An anti-HAV seropositivity of 95.1% (117/122 [95% CI 90.2-98.4]) was found among those ≤6 months of age, indicative of the rate of transplacental antibody transfer. Among 1-12-year-olds, hepatitis A seroprevalence was 19.3% (37/192 [95% CI 14.1-25.7]), while 1.1% (2/188 [95% CI 0.12-2.76]) had evidence of acute infection. Compared to HIV-exposed subgroups (HIV = 60%, 6/10 [95% CI 27.4-86.3] and HEU = 45%, 9/20 [95% CI 23.8-68]), hepatitis A seroprevalence among HUU children was low (29.2%, 47/161 [95% CI 22.4-37.0]). Conclusions: Natural immunity among HIV-exposed and unexposed children in South Africa is insufficient to protect against severe liver complications associated with HAV infection later in adulthood.