Epidemiology Of Invasive Bacterial Infections In Pneumococcal Conjugate Vaccine-Vaccinated And -Unvaccinated Children Under 5 Years Of Age In Soweto, South Africa: A Cohort Study From A High-Hiv Burden Setting

Background: There are limited data on paediatric invasive bacterial infections (IBI) and the impact of pneumococcal conjugate vaccine (PCV) on the spectrum of IBI pathogens, specifically in African countries with a high prevalence of HIV infection. Aim: To describe the epidemiology of IBI in a cohort of children <5 years of age in Soweto, South Africa. Methods: A cohort of children enrolled into a PCV9 efficacy trial conducted from 1998 until 2005 was used for secondary data analysis. Surveillance data were collected from admission wards at Chris Hani Baragwanath Academic Hospital. The incidence of IBI was calculated using person-time, stratified by age group, gender, PCV9 vaccination status and HIV infection status. Risk factors for IBI were investigated using binomial logistic regression. Results: A total of 395 cases of laboratory-confirmed IBI were identified. HIV infection and not receiving PCV9 vaccination were risk factors for IBI hospitalisation. PCV9 vaccination was associated with reductions in IBI hospitalisation (IRR 0.76, p = 0.006) solely owing to reductions in the incidence of Streptococcus pneumoniae (IRR 0.56, p < 0.001). PCV9 vaccination had no effect on the incidence of Haemophilus influenza type b or Salmonella species IBI. There was an increase in Klebsiella species IBI (IRR 3.50, p = 0.019) and a trend towards a higher incidence of Staphylococcus aureus IBI (IRR 1.90, p = 0.099) in PCV9-vaccinated children. Conclusions: PCV9 vaccination was effective in reducing the incidence of IBI hospitalisation in children through reductions in the incidence of S. pneumoniae. The results show that trends in other IBI causative pathogens (specifically S. aureus and Klebsiella species) should be monitored in the era of PCV vaccination.