Pediatr Infect Dis J 2017 Sep;36(9):e211-e218
From the *Botswana-UPenn Partnership, Gaborone, Botswana; †Division of Pediatric Infectious Diseases, Duke University Medical Center, Durham, North Carolina; ‡Department of Medicine, McMaster University, §Department of Pathology and Molecular Medicine, McMaster University, ¶St. Joseph's Healthcare, and ‖Department of Pediatrics, McMaster University, Hamilton, Ontario, Canada; **Global Health Center, and ††Division of Pediatric Infectious Diseases, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania; ‡‡Department of Pathology and Laboratory Medicine, BC Children's Hospital, Vancouver, British Columbia, Canada; §§University of Botswana School of Medicine, Gaborone, Botswana; ¶¶Department of Molecular Genetics and Microbiology, Center for the Genomics of Microbial Systems, Duke University Medical Center, Durham, North Carolina; and ‖‖Divisions of Hospital Medicine and Infectious Diseases, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio.
Background: Nearly half of child pneumonia deaths occur in sub-Saharan Africa. Microbial communities in the nasopharynx are a reservoir for pneumonia pathogens and remain poorly described in African children.
Methods: Nasopharyngeal swabs were collected from children with pneumonia (N = 204), children with upper respiratory infection symptoms (N = 55) and healthy children (N = 60) in Botswana between April 2012 and April 2014. We sequenced the V3 region of the bacterial 16S ribosomal RNA gene and used partitioning around medoids to cluster samples into microbiota biotypes. We then used multivariable logistic regression to examine whether microbiota biotypes were associated with pneumonia and upper respiratory infection symptoms.
Results: Mean ages of children with pneumonia, children with upper respiratory infection symptoms and healthy children were 8.2, 11.4 and 8.0 months, respectively. Clustering of nasopharyngeal microbiota identified 5 distinct biotypes: Corynebacterium/Dolosigranulum-dominant (23%), Haemophilus-dominant (11%), Moraxella-dominant (24%), Staphylococcus-dominant (13%) and Streptococcus-dominant (28%). The Haemophilus-dominant [odds ratio (OR): 13.55; 95% confidence interval (CI): 2.10-87.26], the Staphylococcus-dominant (OR: 8.27; 95% CI: 2.13-32.14) and the Streptococcus-dominant (OR: 39.97; 95% CI: 6.63-241.00) biotypes were associated with pneumonia. The Moraxella-dominant (OR: 3.71; 95% CI: 1.09-12.64) and Streptococcus-dominant (OR: 12.26; 95% CI: 1.81-83.06) biotypes were associated with upper respiratory infection symptoms. In children with pneumonia, HIV infection was associated with a lower relative abundance of Dolosigranulum (P = 0.03).
Conclusions: Pneumonia and upper respiratory infection symptoms are associated with distinct nasopharyngeal microbiota biotypes in African children. A lower abundance of the commensal genus Dolosigranulum may contribute to the higher pneumonia risk of HIV-infected children.