Bordetella pertussis Infection in South African HIV-Infected and HIV-Uninfected Mother–Infant Dyads: A Longitudinal Cohort Study

Background. There is a paucity of data regarding the burden of Bordetella pertussis in African women and young infants, and particularly the impact of maternal human immunodeficiency virus (HIV) infection thereon. We performed a retrospective analysis of respiratory illness samples from longitudinal cohorts of HIV-uninfected and HIV-infected women and their infants to evaluate the burden of pertussis illness in a black-African community. Methods. The women were followed up for respiratory illness from midpregnancy and together with their infants until 24 weeks postpartum. Respiratory samples obtained at the time of illness visits were tested for B. pertussis by polymerase chain reaction (PCR). Results. The study included 194 HIV-infected and 1060 HIV-uninfected women, and 188 and 1028 infant offspring, respectively. There were 7 PCR-confirmed pertussis cases in the HIV-exposed infants and 30 in HIV-unexposed infants (7.4 vs 5.5 episodes per 1000 infant-months; P = .47), at a mean age of 70.9 days. All infant pertussis cases had a history of cough (mean duration, 6.3 days). Six of 17 (35.3%) pertussis-confirmed cases in infants <2 months of age were admitted to hospital within 21 days of B. pertussis detection, whereas none of the 20 cases ≥2 months of age required hospitalization. Ten PCR-positive pertussis-associated illnesses were detected in HIV-infected women compared with 32 in the HIV-uninfected women (6.8 vs 3.9 episodes per 1000 person-months; P = .12). Conclusions. Bordetella pertussis identification was common among young infants with respiratory illness, most of whom were too young to be fully protected through direct vaccination. Vaccination of pregnant women might be a valuable strategy in a setting such us ours to prevent B. pertussis–associated illness in women and their young infants.