Blunt traumatic aortic rupture: detection with helical CT of the chest.

Abstract

To determine the effectiveness of helical computed tomography (CT) as a screening device to detect traumatic aortic rupture. Helical CT was used to examine 1,518 patients with nontrivial blunt trauma. Of these patients, 127 (8.37%) with abnormal CT scans of the mediastinum or aorta underwent thoracic aortography--89 patients solely for evaluation of mediastinal hematoma depicted at CT. Imaging abnormalities were correlated with surgical or clinical outcome. Twenty-one aortic injuries were identified that ranged from subtle intimal flaps to complete aortic disruption. Helical CT was more sensitive than aortography (100% versus 94.4%, respectively) but less specific (81.7% versus 96.3%, respectively) in detection of aortic injuries in patients who underwent both examinations. The association between CT findings and outcome was phi = 0.62 (chi2 = 49.1, 1 df, P < .01) and between aortography and outcome was phi = 0.85 (chi2 = 92.2, 1 df, P < .001). The P value of the difference between the phi coefficients was .10. Helical CT of the chest is effective for screening critically injured patients with possible blunt thoracic aortic injuries.