The Etiology of Childhood Pneumonia in Mali: Findings From the Pneumonia Etiology Research for Child Health (PERCH) Study

Background We present findings from the Pneumonia Etiology Research for Child Health (PERCH) site in Bamako, Mali. Methods Cases were patients 28 days to 59 months of age, admitted to hospital with severe or very severe pneumonia (2005 World Health Organization definition). Community controls were frequency matched by age. Both provided nasopharyngeal and oropharyngeal swabs for multiplex polymerase chain reaction and Streptococcus pneumoniae culture. Cases underwent blood culture and induced sputum culture for Mycobacterium tuberculosis. A subset had pleural fluid and lung aspirates collected for culture and polymerase chain reaction. Primary analyses included participants with negative or unknown HIV status (HIV-) and cases with abnormal chest radiographs (CXR+). Cases and controls were compared using logistic regression adjusting for age. Etiologic fractions were calculated by a Bayesian nested partially latent class analysis, the PERCH integrated analysis. Results Between January 1, 2012, and January 14, 2014, we enrolled 241 CXR+/HIV- cases and 725 HIV- controls. Compared with controls, cases were more likely to have moderate-to-severe wasting (43.1% vs. 14.1%, P Conclusions PERCH uncovered high case fatality among children with severe pneumonia in Mali, highlighting a role for new interventions (eg, respiratory syncytial virus vaccines) and a need to improve vaccine coverage and strengthen healthcare delivery.