Misdiagnosis of bronchiolitis and misuse of antibiotics - experience from the largest paediatric hospital of Bangladesh.

Background: Bronchiolitis remains the commonest acute lower respiratory tract infection in younger children globally. Bronchiolitis is often misdiagnosed as World Health Organization (WHO)-classified pneumonia due to closely related clinical manifestations. Our primary objective was to assess the prevalence of viral bronchiolitis among WHO-classified pneumonia, and the secondary objective was to evaluate their management. Methods: This was a prospective observational study. Infants (aged <2 years) suffering from WHO-classified pneumonia attending the country's largest academic paediatric hospital 'Bangladesh Shishu Hospital and Institute', Dhaka, during July 2020 through June 2021 were enrolled (n = 60) following systematic sampling (every second child). We revalidated all infants meeting the inclusion criteria (cough, fast breathing and/or chest in-drawing) for bronchiolitis assessment using clinical, radiological, haematological, biochemical, and microbiological analyses. Nasopharyngeal aspirate (NPA) was subjected to polymerase chain reaction (PCR) assay for respiratory syncytial virus (RSV), influenza A, and influenza B viruses. Clinical features, treatment modalities, mode of discharge, and length of hospital stay (LOS) were subjected to differentiate bronchiolitis from WHO-classified pneumonia. Results: PCR of NPA of 60 WHO-classified pneumonia cases yielded RSV in 41 (68%) who had pertinent clinical symptoms and signs suggestive of bronchiolitis, attested by benign chest radiography (85%). Further, their mean (x̄) and standard deviation (SD) complete blood count (x̄ = 15,488; SD = 1874/cm) with x̄ = 36% (SD = 11) neutrophil) and C-reactive protein (x̄ = 4.49 (5.88) mg/L cut off reference value >5 mg/L) were indicative more as non-bacterial infections favouring RSV bronchiolitis. In contrast, all 19 non-RSV cases did not differ in clinical findings, as attested by haematological, microbiological and radiographical features, suggesting bronchiolitis-like illness, further supported by disease outcome, discharge pattern and LOS (x̄ = 4.7; SD = 2.29) days). Nevertheless, antibiotics were given in 95% of all 60 cases. Conclusions: There was a high proportion of bronchiolitis among WHO-classified pneumonia, and antimicrobials were prescribed to treat those cases. The findings need the attention of policymakers to address the overlap between WHO-classified pneumonia and bronchiolitis and minimise the potential overuse of antibiotics.