Introduction: African children hospitalized with symptoms of severe acute pneumonia, which resolves following empiric antibiotic therapy, are sometimes identified to have underlying culture-confirmed pulmonary tuberculosis (PTB). Experimental studies suggest Mycobacterium tuberculosis infection predisposes to Streptococcus pneumoniae infection; however, diagnostic limitations make it difficult to quantify this association in children. We aimed to probe the extent of pneumococcal coinfection in children with PTB, using a vaccine-probe design study.
Materials and methods: A post hoc analysis of PTB cases occurring among 39,836 participants in a phase III randomized, double-blind placebo-controlled 9-valent pneumococcal polysaccharide-protein conjugate vaccine (PCV9) trial in South Africa was undertaken. Hospitalization for PTB occurring during the 5.3 years of follow-up were identified and categorized as culture-confirmed PTB or probable/possible-PTB. The incidence rates of hospitalized PTB were compared between PCV9 vaccinees and placebo recipients.
Results: Hospitalization for culture-confirmed PTB was 43.4% (95% CI, 9.7%–65.1%) less likely among vaccinees (n = 30) compared with placebo recipients (n = 53), incidence, 20 versus 35 per 100,000 child-years of follow-up (P = 0.0117). In HIV-infected children, culture-confirmed PTB was 47.3% (95% CI, 8.6%–69.6%) less likely among vaccinees (n = 19) compared with placebo recipients (n = 36), P = 0.0203. The incidence of possible/probable PTB did not differ by vaccination status.
Conclusions: This vaccine-probe design study suggests that in a setting with high HIV and TB prevalence, culture-confirmed PTB in African children, which frequently presents with symptoms of acute pneumonia, is probably associated with superimposed pneumococcal pneumonia. Children admitted with pneumonia in these settings should be investigated for underlying PTB.