Variability of intensive care management for children with bronchiolitis

Abstract

Objective: To determine the extent of variability in testing and treatment of children with bronchiolitis requiring intensive care.

Methods: This prospective, multicenter observational study included 16 academic children's hospitals across the United States during the 2007 to 2010 fall and winter seasons. The study included children<2 years old hospitalized with bronchiolitis who required admission to the ICU and/or continuous positive airway pressure (CPAP) within 24 hours of admission. Among the 2207 enrolled patients with bronchiolitis, 342 children met inclusion criteria. Clinical data and nasopharyngeal aspirates were collected.

Results: Respiratory distress severity scores and intraclass correlation coefficients were calculated. The study patients' median age was 2.6 months, and 59% were male. Across the 16 sites, the median respiratory distress severity score was 5.1 (interquartile range: 4.5-5.4; P<.001). The median value of the percentages for all sites using CPAP was 15% (range: 3%-100%), intubation was 26% (range: 0%-100%), and high-flow nasal cannula (HFNC) was 24% (range: 0%-94%). Adjusting for site-specific random effects (as well as children's demographic characteristics and severity of bronchiolitis), the intraclass correlation coefficient for CPAP and/or intubation was 21% (95% confidence interval: 8-44); for HFNC, it was 44.7% (95% confidence interval: 24-67).

Conclusions: In this multicenter study of children requiring intensive care for bronchiolitis, we identified substantial institutional variability in testing and treatment, including use of CPAP, intubation, and HFNC. These differences were not explained by between-site differences in patient characteristics, including severity of illness. Further research is needed to identify best practices for intensive care interventions for this major cause of pediatric hospitalization.

Keywords: continuous positive airway pressure; high-flow nasal cannula; institutional variability; intraclass correlation coefficient; intubation; respiratory distress.