The Burden of Pertussis Hospitalization in HIV-Exposed and HIV-Unexposed South African Infants

BACKGROUND  There are limited data on pertussis in African children, including among human immunodeficiency virus (HIV)-exposed infants. We conducted population-based hospital surveillance to determine the incidence and clinical presentation of Bordetella pertussis-associated hospitalization in perinatal HIV-exposed and -unexposed infants. METHODS  Children <12 months of age hospitalized with any sign or symptom of respiratory illness (including suspected sepsis or apnea in neonates) were enrolled from 1 January 2015 to 31 December 2015. Detailed clinical and demographic information was recorded and respiratory samples were tested by polymerase chain reaction (PCR). RESULTS  The overall B. pertussis PCR positivity was 2.3% (42/1839), of which 86% (n = 36) occurred in infants <3 months of age. Bordetella pertussis was detected in 2.1% (n = 26/1257) of HIV-unexposed and 2.7% (n = 16/599) of HIV-exposed infants. The incidence (per 1000) of B. pertussis-associated hospitalization was 2.9 (95% confidence interval [CI], 1.8-4.5) and 1.9 (95% CI, 1.3-2.6) in HIV-exposed and HIV-unexposed infants, respectively (P = .09). The overall in-hospital case fatality ratio among the cases was 4.8% (2/42), both deaths of which occurred in HIV-exposed infants <3 months of age. Among cases, presence of cough ≥14 days (20.5%) and paroxysmal coughing spells (33.3%) at diagnosis were uncommon. Only 16 (38%) B. pertussis-associated hospitalizations fulfilled the Centers for Diseases Control and Prevention case definition of "definite" pertussis. CONCLUSIONS  Bordetella pertussis contributed to a modest proportion of all-cause respiratory illness hospitalization among black-African children, with a trend for higher incidence among HIV-exposed than HIV-unexposed infants. Maternal vaccination of pregnant women should be considered to reduce the burden of pertussis hospitalization in this population.