(665) Azithromycin for Critically Ill Children With Bronchiolitis: A Multicenter Observational Study

Introduction: Macrolides (e.g., azithromycin [AZI]) are used as immunomodulation for pediatric conditions including acute viral bronchiolitis. We aimed to estimate prescribing trends for AZI among critically ill children with bronchiolitis and identify prescribing variation (regional and chronologic) using an administrative registry.

Methods: We performed a multicenter, retrospective cohort study using the Pediatric Health Information Systems registry of children 0-3 years of age admitted to 47 United States (US) pediatric intensive care units (PICUs) for bronchiolitis from Jan 2013 - Dec 2023. We excluded for alternative indications for AZI exposure (e.g., pertussis & atypical pneumonia). The primary outcome was AZI prescribing rates by center (i.e., regional variation) and calendar year (i.e., chronologic variation). Cohorts defined by AZI exposure were characterized by demographics, inpatient therapies, and clinical outcomes using comparative (i.e., Chi-square and Wilcoxon rank-sum) statistics. Trends were assessed via Joinpoint regression and Pearson’s correlation.

Results: Of the 46,596 children studied, 1,757 (3.8%) were prescribed AZI. A linear decline in AZI prescribing was noted from 8.3% in 2013 to 1.5% in 2022 (-0.7% annual decline, R²=0.91). Regional AZI prescribing rates by hospital center varied widely with a median of 3.1% (interquartile range [IQR]:1.5-4.8%, total range: 0.6-39.5%). Children prescribed AZI more commonly had a history of prematurity (8.9 vs 6.9%, P=0.002) and were prescribed treatment for bronchospasm with systemic glucocorticoids (50.8 vs 35%, P< 0.001), continuous albuterol (31.5% vs 17.8%, P< 0.001), and heliox (12.4 vs 2.5%, P< 0.001). Children prescribed AZI experienced a more severe clinical trajectory with a greater rate of invasive ventilation (35.3 vs 23.7%, P< 0.001) and a longer median length of stay (6 [IQR:4,9] vs 4 [IQR:3,6], P< 0.001). No adverse events were observed including QT prolongation.

Conclusions: Among 47 US PICUs, we observed a downward trend (-0.7% annually) in AZI prescribing with wide institutional variation. Those given AZI were more commonly treated for presumed bronchospasm and had a more severe clinical trajectory. Prospective study is needed to validate these findings, delineate safety, and assess efficacy for AZI use in this population.