Lung Ultrasound Findings and Bronchiolitis Ultrasound Score for Predicting Hospital Admission in Children With Acute Bronchiolitis

Abstract

Objectives The purposes of this study were to determine the benefit of the bronchiolitis ultrasound score (BUS) in predicting hospital admission in children with acute bronchiolitis and to characterize lung sonography findings. Methods This prospective observational study was performed in an academic pediatric emergency department. Children younger than 24 months presenting to the emergency department, diagnosed with acute bronchiolitis by 2 independent pediatricians were included in the study. Lung ultrasound was performed by a single sonographer, who was blinded to as much clinical information as possible. In addition, the treating physicians were blinded to the lung ultrasound findings. Logistic regression analysis models were used to identify admission predictors. Receiver operating characteristic analysis was used to evaluate the predictive value for effects of the BUS and the modified Bronchiolitis Severity Score on admission. Results The median age of the 76 patients diagnosed with acute bronchiolitis was 6 months (interquartile range, 3.6–10 months). Forty-two (55.3%) of the 76 patients enrolled were admitted. Lung ultrasound was compatible with acute bronchiolitis in 74 patients (97%). A significant correlation was determined between modified Bronchiolitis Severity Score and BUS in children with acute bronchiolitis (r = 0.698, P < 0.001). The most effective parameter in determining admission on logistic regression analysis, independently of other variables, was BUS (P = 0.044; adjusted odds ratio, 1.859; 95% confidence interval, 1.016–3.404). Bronchiolitis ultrasound score values of 3 or greater exhibited 73.81% sensitivity and 73.53% specificity, whereas BUS values of 4 or greater exhibited 50% sensitivity and 91.18% specificity. Conclusions Point-of-care lung ultrasound can accurately detect pulmonary anomalies in children with acute bronchiolitis, has a close correlation with clinical findings, and is a useful tool in predicting hospital admission.