Chest CT Can Be Used to Exclude Aortic Injury

Abstract & Commentary

Synopsis: Physicians in two busy trauma centers examined the usefulness of computed tomography of the chest as a screening examination for trauma patients with suspected aortic injury. More than 800 patients were examined with CT and approximately half had follow-up aortography, which is considered the gold standard for the diagnosis of aortic injuries. CT was positive in all cases of aortic injury and a negative CT accurately ruled out aortic injuries. It was concluded that chest CT is an excellent screening tool for traumatic aortic injuries.

Source: Dyer DS, et al. Can chest CT be used to exclude aortic injury? Radiology 1999;213:195-202.

The majority of patients with blunt chest trauma and aortic rupture do not survive the initial injury. If the patient survives and arrives in the emergency department, the diagnosis of rupture must be made rapidly so that treatment may be initiated. Aortography is considered the reference-standard procedure for the diagnosis of aortic injury; however, it is invasive, expensive, and resource intensive. This study was conducted over a period of 4.5 years in two busy trauma centers and its purpose was to check the reliability of computed tomography (CT) in excluding aortic injury in blunt chest trauma. Eight hundred two patients with blunt chest trauma had chest CT as the initial exam, 382 of whom underwent follow-up aortography. All CT examinations, except seven, were contrast enhanced and approximately half were performed with a helical CT scanner. Aortography was performed using routine intra-arterial digital subtraction technique in two projections.

In the 382 patients who had both CT scan and aortography, when a positive scan was defined as that showing any mediastinal hematoma and/or direct signs of aortic injury, 10 CT scans were true positives, 230 were true negatives, 142 were false positives, and none was a false negative. Thus, the sensitivity of CT in the diagnosis of aortic injury was 100%, negative predictive value 100%, specificity 62%, and positive predictive value 7%. When only direct signs of aortic injury were considered positive on CT, the specificity increased to 96% and positive predictive value increased to 40%. Sensitivity and negative predictive value remained 100%. It was concluded that chest CT is an excellent screening test for the diagnosis of thoracic aortic injury in patients with blunt trauma.

Comment by Moni Stein, MD

Over the past 10 years, there has been a change in the approach to the diagnosis of aortic injury in trauma. Earlier, CT was not considered a reliable test to exclude aortic injury and the algorithm was to perform a plain chest film and, if necessary, perform aortography. Primarily trauma surgeons who wanted to expedite the diagnosis and send the appropriate patients as quickly as possible to the operating room promoted this algorithm.

In an article published in 1992, Raptopoulos and colleagues showed that chest CT was a reliable screening test to exclude thoracic aortic injuries.1 Dynamic chest CT was positive (mediastinal blood) in all injuries and there were no false negatives. With CT there was a significant improvement over plain chest radiography in specificity, accuracy, and predictive value of positive results. With the addition of chest CT in the screening for traumatic aortic tear, the need for aortography decreased by 56%.

More recently, a wide variety of traumatic and nontraumatic emergency conditions are quickly and accurately diagnosed with helical CT.2 The speed of helical technology allows expeditious CT examination of seriously ill patients. In addition, helical technology allows multiple, sequential CT scans to be rapidly obtained in the same patient, a great advantage for the multiple-trauma patient. Better quality CT examinations result from decreased misregistration, better intravenous contrast material opacification of vascular structures and parenchymal organs, and more flexibility in image reconstruction. Helical chest CT can also show a number of conditions that may be overlooked otherwise. Gavant et al reported a series of 1518 helical CT scans obtained for blunt trauma.3 One hundred twenty-seven patients with abnormal CT scans underwent aortography. Helical CT was shown to be more sensitive (100% vs 94.4%) but less specific (81.7% vs 96.3%) than aortography in the detection of aortic injuries. The earlier CT studies focused primarily on the presence of mediastinal blood to trigger performance of aortography. More recently, with more advanced scanners and cumulative diagnostic experience, even in the presence of mediastinal hematoma, if there are no direct signs of an aortic injury on the CT scan, then the study is considered to be negative for aortic injury.2


1. Raptopoulos V, et al. Traumatic aortic tear: Screening with chest CT. Radiology 1992;182:667-673.

2. Novelline RA, et al. Helical CT in emergency radiology. Radiology 1999;213:321-339.

3. Gavant ML, et al. Blunt traumatic aortic rupture: Detection with helical CT of the chest. Radiology 1995; 197:125-133.