Estimates of possible severe bacterial infection in neonates in sub-Saharan Africa, south Asia, and Latin America for 2012: a systematic review and meta-analysis

Background Bacterial infections are a leading cause of the 2.9 million annual neonatal deaths. Treatment is usually based on clinical diagnosis of possible severe bacterial infection (pSBI). To guide programme planning, we have undertaken the first estimates of neonatal pSBI, by sex and by region, for sub-Saharan Africa, south Asia, and Latin America. Methods We induded data for pSBI incidence in neonates of 32 weeks' gestation or more (or birthweight >= 1500 g) with livebirth denominator data, undertaking a systematic review and forming an investigator group to obtain unpublished data. We calculated pooled risk estimates for neonatal pSBI and case fatality risk, by sex and by region. We then applied these risk estimates to estimates of livebirths in sub-Saharan Africa, south Asia, and Latin America to estimate cases and associated deaths in 2012. Findings We induded data from 22 studies, for 259 944 neonates and 20 196 pSBI cases, with most of the data (18 of the 22 studies) coming from the investigator group. The pooled estimate of pSBI incidence risk was 7.6% (95% CI 6.1-9.2%) and the case-fatality risk associated with pSBI was 9.8% (7.4-12.2). We estimated that in 2012 there were 6.9 million cases (uncertainty range 5.5 million-8.3 million) of pSBI in neonates needing treatment: 3.5 million (2.8 million-4.2 million) in south Asia, 2.6 million (2.1 million-3.1 million) in sub-Saharan Africa, and 0.8 million (0.7 million-1.0 million) in Latin America. The risk of pSBI was greater in boys (risk ratio 1.12, 95% CI 1.06-1.18) than girls. We estimated that there were 0.68 million (0.46 million-0.92 million) neonatal deaths associated with pSBI in 2012. Interpretation The need-to-treat population for pSBI in these three regions is high, with ten cases of pSBI diagnosed for each associated neonatal death. Deaths and disability can be reduced through improved prevention, detection, and case management.