Daniel J. Pallin, MD, MPH reviewing Langdorf MI et al. Ann Emerg Med 2015 Dec.
Trauma patients commonly undergo computed tomographic (CT) imaging of the chest, yet evidence to guide chest imaging is conflicting, with some studies suggesting that we should scan more and others suggesting that we should scan less. Investigators from the NEXUS group conducted a secondary analysis of 5912 patients with blunt trauma who had both a chest radiograph and a chest CT scan, based on clinician discretion, at 10 Level 1 trauma center emergency departments. These patients were drawn from a larger sample of 14,533 patients who underwent either CT or radiography.
Overall, 25% of the 5912 patients had injuries discovered on chest CT but missed on chest x-ray. Of these, 14% underwent major interventions (chest tube, mechanical ventilation, or surgery), 24% underwent minor interventions (observation or pain control for >24 hours), and 62% received no intervention.
This study does not prove that it is good to do chest CT scans on trauma patients, because it
provides no evidence that doing the scans was beneficial. It remains possible that deferring the scans would have led to injury detection later in the subset of patients requiring intervention. It also remains possible that some of the interventions were unnecessary (e.g., do we really know which hemopneumothoraces will not resolve without a chest tube?). I agree with the authors: we need a decision rule, and the decision rule should be tested in a prospective randomized fashion so we can really know whether patients benefit.
Langdorf MI et al. Prevalence and clinical import of thoracic injury identified by chest computed tomography but not chest radiography in blunt trauma: Multicenter prospective cohort study. Ann Emerg Med 2015 Dec; 66:589. (http://dx.doi.org/10.1016/j.annemergmed.2015.06.003) PubMed abstract (Free)