Limiting chest computed tomography in the evaluation of pediatric thoracic trauma

J Trauma Acute Care Surg. 2016 Aug;81(2):271-7. doi: 10.1097/TA.0000000000001110.

Abstract

Background: Computed tomography (CT) of the chest (chest CT) is overused in blunt pediatric thoracic trauma. Chest CT adds to the diagnosis of thoracic injury but rarely changes patient management. We sought to identify a subset of blunt pediatric trauma patients who would benefit from a screening chest CT based on their admission chest x-ray (CXR) findings. We hypothesize that limiting chest CT to patients with an abnormal mediastinal silhouette identifies intrathoracic vascular injuries not otherwise seen on CXR.

Methods: All blunt trauma activations that underwent an admission CXR at our Level 1 pediatric trauma center from 2005 to 2013 were retrospectively reviewed. Patients who had a chest CT were evaluated for added diagnoses and change in management after CT.

Results: An admission CXR was performed in 1,035 patients. One hundred thirty-nine patients had a CT, and the diagnosis of intra-thoracic injury was added in 42% of patients. Chest CT significantly increased the diagnosis of contusion or atelectasis (30.3% vs 60.4%; p < 0.05), pneumothorax (7.2% vs 18.7%; p < 0.05), and other fractures (4.3% vs 10.8%; p < 0.05) on CXR compared to chest CT. Chest CT changed the management of only 4 patients (2.9%). Two patients underwent further radiologic evaluation that was negative for injury, one had a chest tube placed for an occult pneumothorax before exploratory laparotomy, and one patient had a thoracotomy for repair of aortic injury. Chest CT for select patients with an abnormal mediastinal silhouette on CXR would have decreased CT scans by 80% yet still identified patients with an intrathoracic vascular injury.

Conclusions: The use of chest CT should be limited to the identification of intrathoracic vascular injuries in the setting of an abnormal mediastinal silhouette on CXR.

Level of evidence: Therapeutic study, level IV; diagnostic study, level III.

MeSH terms

  • Adolescent
  • Child
  • Child, Preschool
  • Female
  • Humans
  • Infant
  • Infant, Newborn
  • Injury Severity Score
  • Male
  • Radiography, Thoracic / statistics & numerical data
  • Retrospective Studies
  • Thoracic Injuries / diagnostic imaging*
  • Tomography, X-Ray Computed / statistics & numerical data*
  • Trauma Centers
  • Wounds, Nonpenetrating / diagnostic imaging*