Lung ultrasound for detecting pneumothorax in injured children: preliminary experience at a community-based Level II pediatric trauma center.

Abstract

Ultrasound (US) has been used in the adult trauma population with reported moderate to high sensitivities, but data are scarce in the pediatric trauma population. The purpose of this study was to specifically examine the sensitivity and specificity of one lung US methodology (single-point anterior exam) in the pediatric trauma population when compared to chest radiography or CT. We conducted a retrospective review of pediatric trauma patients who received lung US as an extension of the focused assessment with sonography for trauma (FAST) exam. We compared lung US findings with chest radiography and CT scans. Two hundred twenty-six pediatric trauma patients underwent lung US exam with confirmatory exams; 11 pneumothoraces (4.8%) were observed. Of those 11, 6 were evaluated as false negatives on the lung US. Analyses resulted in 45.5% sensitivity, 98.6% specificity and 96.0% accuracy. Pneumothoraces undetected by lung US were small and apical and were likely not observed because of their size and location. None of the false negatives required intervention. All true positives were associated with lung contusions. Pneumothorax is less common in the pediatric than the adult trauma population, and when encountered in children pneumothorax is often occult and might be associated with lung contusions. Existing evidence supports the usefulness of chest US in detecting pneumothorax in adults and suggests that it can be translated to injured children. However, our findings suggest that the sensitivity of lung US as a single-point anterior exam extension of the FAST exam might not be as reliable in the pediatric trauma population as in adults. Other methodologies using lung US might improve sensitivity.