Background: In Western countries, the prevalence of HHV8 is high in homosexual men (20%–30%). Among homosexual men, transmission is thought to occur via salivary or sexual exposures. In non-homosexual men and women, prevalence is low (2%–5%). Although salivary and sexual exposures are also frequent, the low prevalence makes transmission difficult to evaluate. However, in Africa, where KS is common, HHV8 prevalence is high and transmission is common, even in children. We examined correlates of infection in different populations of East and West Africa.

Methods: We studied 798 subjects with family data in rural Tanzania. Samples were collected in 1985, when HIV was uncommon (2 positive). Separately, we also studied 2,155 Nigerian adults with diverse sexual lifestyles when samples were collected in 1992–1994. Infection status was determined by EIA-detected antibody to K8.1, a HHV8 structural glycoprotein.

Results: By age 3–4 yrs old, 58% of Tanzanian children were already HHV8 seropositive. Infection in childhood was associated with maternal HHV8 status (OR: 7.4; 95% CI, 3.2–16.8) and, less strongly, with the antibody status of other household members, including older siblings. Maternal and familial child-caring patterns may result in frequent and extensive saliva exposures to these children. In adults (≥18 yrs old), prevalence was high (88% in men; 79% in women). After adjusting for age, women with seropositive husbands were 6.9-fold more like to be seropositive than those with seronegative husbands, indicating additional transmission occurs, probably via sexual intercourse. Among adults in Lagos, Nigeria, we found a lower prevalence in the background population (22% of men and 14% of women). However, female commercial sex workers (31%) and both female (20%) and male (35%) attendees of clinics for sexually transmitted diseases had a higher prevalence than the referent population. Furthermore, within each group, persons who had laboratory evidence of sexually transmitted diseases (chancroid, syphilis) or infections (HIV⁺, HTLV-I⁺) were significantly more likely to be HHV8-infected.

Conclusions: Our data from Africa are consistent with intra-familial transmission, probably through saliva in childhood, with some additional sexual transmission among adults. While HHV8 prevalence was lower in Nigerian than Tanzanian adults, the populations differed in several ways (urban/rural, socio-economic) that make a direct comparison difficult.