Upper airway viruses and bacteria detection in clinical pneumonia in a population with high nasal colonisation do not relate to clinical signs

Indigenous Australian children have high (up to 90%) rates of nasopharyngeal microbial colonisation and of hospitalisation for pneumonia. In Indigenous children hospitalised with pneumonia in Central Australia, we describe the nasopharyngeal detection of viruses and bacteria and assessed whether their presence related to signs of pneumonia (tachypnoea and/or chest in-drawing) on hospital admission and during subsequent days. Nasopharyngeal swabs (NPS) and data were prospectively collected from 145 children (median age = 23.5 months, interquartile range [IQR] 8.7–50) hospitalised with pneumonia at Alice Springs Hospital, Australia, between April 2001 and July 2002. The cohort was enrolled in a randomised controlled study using zinc and/or vitamin A supplementation. NPS were taken within 24 hours of hospitalisation and kept frozen at-80°C until analysed in 2014. Polymerase chain reaction (PCR) was used to detect Moraxella catarrhalis, Haemophilus influenzae, Streptococcus pneumoniae, Staphylococcus aureus, Chlamydophila pneumoniae, Mycoplasma pneumoniae, and 16 respiratory viruses. Uni- and multi-variate analyses were used to examine the relationships. One or more organisms were present in 137 (94.5%) NPS; 133 (91.7%) detected ≥ 1 bacterium, 34 (37.2%) for ≥ 1 virus and 50 (34.5%) were positive for both viruses and bacteria. C. pneumoniae (n = 3) and M. pneumoniae (n = 2) were rare. In multi-variate analyses, age < 12 months (odds ratio [OR] 6.6 [95% confidence interval {CI} 1.7–25.4]) and fever (OR 4.1 [95% CI 1.7–10.4]) were associated with tachypnoea and chest in-drawing. However the presence of bacteria and/or virus type was not associated with tachypnoea and/or chest in-drawing on admission or during recovery. In children with high nasopharyngeal microbial colonisation rates, the utility of NPS in determining the diagnosis of clinical pneumonia or duration of tachypnoea or in-drawing is likely limited. Larger cohort and case-control studies are required to confirm our findings.