Objectives: Infectious pneumonia is the most common cause of acute respiratory distress syndrome, with viruses frequently implicated as causative. However, the significance of viruses in pediatric acute respiratory distress syndrome is unknown. We aimed to characterize the epidemiology of viral pneumonia in pediatric acute respiratory distress syndrome and compare characteristics and outcomes between pneumonia subjects with and without viruses. Secondarily, we examined the association between specific viruses and outcomes.
Design: We performed a secondary analysis of a prospectively enrolled pediatric acute respiratory distress syndrome cohort. Subjects with pneumonia acute respiratory distress syndrome underwent testing of respiratory secretions for viruses and culture for bacteria and fungi and were stratified according to presence or absence of a virus.
Setting: Tertiary care children's hospital.
Patients: Children with acute respiratory distress syndrome.
Interventions: None.
Measurements and main results: Of 544 children with acute respiratory distress syndrome, 282 (52%) had pneumonia as their inciting etiology, of whom 212 were virus-positive. In 141 of 282 (50%) pneumonia acute respiratory distress syndrome cases, a virus was the sole pathogen identified. Virus-positive pneumonia had fewer organ failures but worse oxygenation, relative to virus-negative pneumonia, with no differences in antibiotic use, ventilator duration, or mortality. Subjects with respiratory syncytial virus-associated acute respiratory distress syndrome had lower mortality (0%), and subjects with influenza-associated acute respiratory distress syndrome had shorter ventilator duration, relative to other viral acute respiratory distress syndrome. Nonadeno herpesviruses, tested for exclusively in immunocompromised subjects, had greater than 80% mortality.
Conclusions: Pneumonia was the most common cause of pediatric acute respiratory distress syndrome, and viruses were commonly isolated as the sole pathogen. Respiratory syncytial virus and influenza were associated with better outcomes relative to other viral etiologies. Viral pneumonias in immunocompromised subjects, particularly nonadeno herpesviruses, drove the mortality rate for pneumonia acute respiratory distress syndrome. Specific viral etiologies are associated with differential outcomes in pediatric acute respiratory distress syndrome and should be accounted for in future studies.