Antibiotic Use in Hospitalized Children With Respiratory Viruses Detected by Multiplex Polymerase Chain Reaction

Background: Multiplex polymerase chain reaction-based methods are increasingly used to detect respiratory pathogens in children. While rapid identification of viruses has been shown to reduce antibiotic use, the impact of detecting specific viruses on antibiotic utilization has not been ascertained. This study compared antibiotic utilization among hospitalized children who tested positive for different respiratory viruses at admission. Methods: A single-center study of hospitalized children under 21 years of age who tested positive at admission for at least 1 respiratory virus by multiplex polymerase chain reaction from October 1, 2012 to October 1, 2015 was performed. Multivariable logistic regression was used to determine the association of testing positive for specific viruses with the use of antibiotics for >= 2 days, adjusted for demographic and clinical characteristics. Results: The study included 1416 patients with a median age of 2.1 years (interquartile range: 0.6-6.2 years). Patients positive for influenza (odds ratio: 2.0, 95% confidence interval: 1.1-3.4) and human metapneumovirus (odds ratio: 2.0, 95% confidence interval: 1.1-3.7) were more likely to receive >= 2 days of treatment compared with patients positive for respiratory syncytial virus (RSV). Other variables affecting prolonged use of antibiotics included respiratory support, primary nonrespiratory diagnosis, complex comorbid conditions and admission to the intensive care unit. Conclusions: Providers are more likely to use antibiotics in non-RSV-infected patients compared with RSV. These trends likely represent concern about bacterial superinfection and may reflect lack of familiarity with these pathogens.