Surveillance for Invasive Streptococcus pneumoniae Disease among Hospitalized Children in Bangladesh: Antimicrobial Susceptibility and Serotype Distribution

Background. Vaccines offer the prospect of primary disease prevention of pneumococcal disease in childhood. For introduction of such vaccines in developing countries, information about disease epidemiology is necessary. Methods. We evaluated antimicrobial susceptibility and serotype distribution of invasive Streptococcus pneumoniae disease in children aged < 5 years in a network of 7 hospitals in Bangladesh from May 2004 through May 2007. Results. Of 17,969 blood cultures and 3765 cerebrospinal fluid cultures, 139 yielded S. pneumoniae isolates; 94 were from meningitis cases, 13 were from pneumonia cases, and 32 were from sepsis cases. Among the children with positive culture results, 73% were aged < 12 months and 90% were aged < 24 months. Complete resistance against penicillin, chloramphenicol, and cotrimoxazole was found in 0%, 6%, and 32% of isolates, respectively. Of the 37 serotypes observed, the predominant serotypes were 2 (17%), 1 (12%), 14 (7%), 5 ( 6%), 7F (6%), 45 (7%), and 12A (4%). Serotypes differed between meningitis cases and nonmeningitis cases, especially for serotype 2 (25% of meningitis cases vs. 0% of pneumonia cases; P <.001). The 7-, 10-, and 13- valent vaccines would cover 20%(95% confidence interval [CI], 13%-27%), 43% (95% CI, 35%-51%), and 50% (95% CI, 42%-58%) of these cases of invasive pneumococcal disease overall, with higher coverage of nonmeningitis cases, compared with meningitis cases (7-valent coverage, 23% vs. 18%; 10-valent coverage, 55% vs. 38%; 13-valent coverage, 66% vs. 42%). Conclusions. High levels of nonsusceptibility to cotrimoxazole and susceptibility to penicillin suggest that penicillin may be a drug of choice for treatment of invasive pneumococcal disease. Although serotype distribution is diverse, with changes over time and differences between syndromes observed, implementation of use of the currently available 10- or 13- valent vaccines would have a substantial impact on pneumococcal disease in Bangladesh.