Bacterial and viral etiology of acute respiratory infection among the Forcibly Displaced Myanmar Nationals (FDMNs) in fragile settings in Cox's Bazar- a prospective case-control study.

ABSTRACT

The leading infectious cause of death in children worldwide is lower acute respiratory infection (LARI), particularly pneumonia. We enrolled a total of 538 acute respiratory infection (ARI) cases according to WHO criteria and age-sex matched 514 controls in the Forcibly Displaced Myanmar National (FDMN) refugee camps in Cox's Bazar, Bangladesh, between June 2018 and March 2020 to investigate the role of bacteria, viruses, and their co-infection patterns and observe Streptococcus pneumoniae (S. pneumoniae) serotype distribution. According to the etiological findings, children ≤5 years of age have a higher bacterial positivity (90%) and viral positivity (34%) in nasopharyngeal samples (NPS) compared to those >5 years of age, in both ARI cases as well as for the control group. Among the bacteria, S. pneumoniae was predominant in both cases and controls (85% and 88%). Adenovirus (ADV)(34), influenza virus A and B (IFV-A, B)(32,23), and respiratory syncytial virus (RSV)(26) were detected as the highest number among the viruses tested for the ARI cases. The total number of viruses was also found higher in ≤5 years of age group. Within this group, positive correlation was observed between bacteria and viruses but negative correlation was observed between bacteria. Both single and co-infection for viruses were found higher in the case group than the control group. However, co-infection was significantly high for Streptococcus aureus (S. aureus) and Haemophilus influenzae b (H. influenza b) (p<0.05). Additionally, semi-quantitative bacterial and viral load was found higher for the ARI cases over control considering Cycle threshold (Ct)≤30. Pathogen identification from blood specimens was higher by qRT-PCR than blood culture (16% vs 5%, p<0.05). In the S. pneumoniae serotype distribution, the predominant serotypes in ARI cases were 23F, 19A, 16F, 35B, 15A, 20 and 10F, while 11A, 10A, 34, 35A and 13 serotypes were predominant in the control group. Pathogen correlation analysis showed RSV positively correlated with human metapneumovirus (HMPV), S. aureus and H. influenza b while S. pneumoniae was negatively correlated with other pathogens in ≤5 years age group of ARI cases. However, in >5 years age group, S. aureus and H. influenza b were positively correlated with IFVs, and S. pneumoniae was positively correlated with HMPV and ADV. Logistic regression data for viruses suggested among the respondents in cases were about 4 times more likely to be RSV positive than the control. Serotype distribution showed 30% for PCV10 serotypes, 41% for PCV13 and 59% for other serotypes. Also, among the 40 serotypes of S. pneumoniae tested, the serotypes 22F, Sg24, 9V, 38, 8, and 1 showed strong positive correlation with viruses in the case group whereas in the control group, it was predominant for serotypes 14, 38, 17F and 39 ARI cases were prevalent mostly in monsoon, post-monsoon, and winter periods, and peaked in September and October. Overall these region-specific etiological data and findings, particularly for crisis settings representing the FDMNs in Cox's Bazar, Bangladesh, is crucial for disease management and disease prevention control as well as immunization strategies more generally in humanitarian crisis settings.