Traumatic Aortic Injury (TAI) Imaging and Diagnosis

Traumatic aortic disruption is a time-sensitive injury requiring rapid and accurate diagnosis to prevent death. Although the clinical, or mechanism, score is of primary importance in the prompt diagnosis of patients with traumatic aortic injury (TAI), the radiologic findings play a vital supportive role. TAI is initially characterized by contained rupture (pseudoaneurysm), which is relatively clinically silent. After a variable period of time, contained rupture is followed by a rapid transition to free, uncontained pseudoaneurysm rupture; exsanguination; and death. The clinical challenges are to rapidly stabilize the patient and then transport the patient to a level I trauma center for evaluation, diagnosis, and definitive treatment before free rupture occurs. ^{[1, 2]}

Approximately 80% of patients with acute TAI will die before reaching the hospital. In a Japanese study by Katayama et al, later timing of the first CT scanning was associated with a higher incidence of death. Patients were divided into 3 groups: early (≤26 min), middle (27-40 min), and late (≥41 min) phases, based on the time interval from hospital arrival to start of first CT scanning. A total of 421 patients were assessed. Mortality was 7.7% (11/142) for the early group, 11.1% (15/135) for the middle group, and 17.6% (25/144) for the late group. ^{[1, 3, 2, 4]}

Blunt TAI results most frequently from high impact deceleration (eg, high-velocity motor vehicle accidents) and falls from a height. It is important that radiologists are familiar with both the direct (intimal flap, pseudoaneurysm, aortic contour irregularity and contrast extravasation) and indirect (periaortic haematoma) imaging findings. ^[5]

Signs of an underlying aortic injury include widening of the mediastinum, apical pleural capping, an abnormal aortic contour, and deviation of an endotracheal or nasogastric tube to the right. Chest radiographs can be helpful but should not delay cross-sectional imaging if warranted. Chest radiography is better used to assess for life-threatening injuries, such as tension pneumothorax and a large hemothorax requiring immediate drainage. ^[5]

(TAI images are shown below.)

Blunt traumatic aortic injury is usually located distal to the left subclavian artery, but injuries have also been reported in other locations in the descending thoracic aorta. ^[6]

Preferred examination

Plain radiography is usually the first test to be performed. The optimal upright posteroanterior (PA) chest evaluation is often deferred for a portable examination with the patient still on the backboard.

Computed tomography (CT) scanning and angiography are the prime imaging modalities in the planning of treatment for blunt trauma. The choice between CT scanning and angiography may depend on institutional preferences, the patient's condition, and the likelihood of other injuries. For example, a patient with hemorrhage from a crushed pelvis would undergo angiography first, whereas a hemodynamically stable patient with a suspected renal or splenic injury would undergo CT scanning first. ^{[7, 8, 9, 10, 11, 12, 3, 13]}

Diagnostic imaging should be deferred in patients presenting in hypovolemic shock or cardiac arrest. Advanced (64-slice and greater) spiral CT can produce near-angiographic-quality images, and when available, it should be considered the diagnostic procedure of choice for TAI. Angiography is appropriate in cases where endovascular intervention (eg, stent-graft) is contemplated.

CT angiography has a sensitivity of 98% and a specificity of 100% in the assessment for TAI and  is indicated in all cases of high-velocity trauma. Patel et al recommended CT angiography to evaluate the aorta using multiplanar reformatting (aortic injury is most conspicuous in the sagittal-oblique plane [left anterior oblique; 20-90 degrees]) and identify the following ^[5] :

• Lower curvature contour of the aortic arch
• Intimal tear/flap
• Pseudoaneurysm formation
• Mediastinal hematoma
• Periaortic hematoma
• Presence of hemothorax
• Additional injuries, including  fractures (vertebral, scapula, first/second ribs)

Except in cases involving exsanguinating hemorrhage from a pelvic fracture, angiography has no role in the early treatment of a patient with polytrauma who is in unstable condition. Magnetic resonance imaging (MRI) generally has no role in the acute evaluation of polytrauma. Some centers advocate the early use of transesophageal echocardiography. ^{[14, 15]}

Guidelines

According to the American College of Radiology (ACR) Appropriateness Criteria, chest CT with IV contrast, whole-body CT with IV contrast, chest CT angiography with IV contrast, or a radiography trauma series is usually appropriate for the initial imaging of major blunt chest trauma in hemodynamically stable patients. These modalities are considered equivalent to provide the clinical information needed for effective management. ^[16]  

The Eastern Association for the Surgery of Trauma (EAST) published a practice management guideline on evaluation and management of blunt traumatic aortic injury (BTAI), which included the following recommendations ^[17] :

• Patients with suspected BTAI - Strong recommendation in favor of computed tomography (CT) of the chest with IV contrast for diagnosis of clinically significant BTAI.
• Patients diagnosed with BTAI - Strong recommendation in favor of using endovascular repair in patients without contraindications for such repair.
• Patients diagnosed with BTAI - Suggestion in favor of delayed repair as opposed to immediate repair.

After a clinical evaluation, most patients are best evaluated with chest radiography, followed by CT scanning, angiography, or immediate surgery, depending on the specific features of the case and institutional preferences. In the typical emergency department evaluation of a patient with blunt trauma, an initial screening radiograph of the chest and cervical spine is obtained. The patient's clinical condition, the severity and mechanism of the trauma, and the initial radiographic findings are used to determine whether angiography or CT scanning is indicated. ^{[5, 18, 19, 20, 21, 22, 23, 24]}

Chest radiographs can be helpful but should not delay cross-sectional imaging if warranted. Chest radiography is better used to assess for life-threatening injuries, such as tension pneumothorax and a large hemothorax requiring immediate drainage. ^[5]

Initial chest radiography has a significant false-negative rate in patients with TAI. Ekeh et al found that it failed to identify the possibility of blunt thoracic aortic injury in 11% of cases. Consequently, researchers recommend follow-up CT scanning or CT angiography in all patients with suspected TAI. ^{[25, 26, 27]}

(Radiographs of aortic trauma are shown below.)

Blunt aortic trauma (BAT) has several classic radiographic signs. These signs are more a result of multiple injuries or the radiographic technique used and less a specific sign of BAT. Regarding the well-known sign of mediastinal widening, some trauma radiologists believe that the mediastinal contour is a better indication of BAT than is the transverse diameter.

Diagnoses such as masses, infection, and chronic injury may simulate acute traumatic aortic injury.

Stark et al reviewed the chest radiographic results in 49 cases of aortic rupture and found that a widened mediastinum, partial blurring of the aortic shadow, a left apical cap, and right tracheal deviation were the most common findings. ^[28] No patient in the study had normal results. Unfortunately, those findings are often nonspecific, and they can be present in patients without significant aortic injury. Therefore, clinical evaluation is essential in determining which patients require further studies.

The incidences of radiographic signs of TAI, as reported by Stark et al, are as follows ^[28] :

• Wide mediastinum - 70% (See the images below.)

• Partial obliteration of the descending aorta - 67%

• A left apical cap - 65%

• Downward displacement of the left bronchus - 65%

• Tracheal deviation to the right - 63%

• Obscuration of the aortic arch - 55%

• A right paratracheal stripe thickening - 53%

• Deviation of the nasogastric (NG) tube to the right - 50%

• Enlarged abnormal aortic contour - 39%

• Left hemothorax - 35%

• A displaced left paraspinal stripe - 35%

• A displaced right paraspinal stripe - 33%

• A fracture of the first rib - 16%

  Aorta, trauma. Chest radiograph shows widening of the mediastinal contour and deformity and blurred margins of the superior mediastinum.

  View Media Gallery

  Aorta, trauma. Chest radiograph shows a circumferential isthmus rupture. The aortic knob is not depicted, and the mediastinum is wide and deformed.

  View Media Gallery

Radiographic signs of TAI, as reported by Kirsh et al in a study of 43 patients, were as follows: widened mediastinum, 42 patients; abnormal aortic contour, 41 patients; deviation of the trachea to the right, 14 patients; depression of the left mainstem bronchus, 13 patients; left pleural effusion, 15 patients; and negative signs, 1 patient. ^[29]

The table below shows the poor predictive value of the classic chest radiographic signs of TAI.

Table 1. Frequency of Abnormal Radiographic Signs in Patients with Suspected TAI ^[30] (Open Table in a new window)

 

Traumatic aortic injury (TAI) may be diagnosed from CT scans on the basis of direct or indirect signs. Direct signs (eg, aortic intimal flap, contour abnormality) are more accurate than indirect signs (eg, mediastinal, periaortic hematoma). CT scanning is an excellent screening modality for patients who are not undergoing aortography. It is particularly convenient in patients with polytrauma who are undergoing other CT scan studies at the same time. Acquisition parameters and protocols should be developed locally, based on equipment, physician preference, and patient condition. If CT angiographic reconstruction is used, the axial source images should be reviewed. ^{[7, 8, 11, 3, 13]}

(CT scans of aortic trauma are shown below.)

The advantages of CT scanning are that it is quick, noninvasive, useful in evaluating multiple traumas at the same time, and capable of providing a larger field of interest, particularly with fast spiral technology. ^{[31, 32, 33]} The historic tradeoff between CT scanning and angiography was image resolution versus cross-sectional imaging. The submillimeter resolution available with angiography (cut-film and then digital subtraction angiography [DSA]) was invaluable for delineating the sometimes subtle findings of traumatic aortic disruption. CT scanning times were long, and sections were thick. Moreover, the intravascular density of contrast material on CT scans was far less than that which is currently achievable. False-negative results could occur with subtle injuries. False-positive CT scan findings were possible with mediastinal hemorrhage from venous bleeding and irregularities of the vessel wall due to atherosclerosis.

Multidetector spiral technology has allowed CT scanning to catch up to angiography in terms of image quality; 64- or 256-slice scanners can image with submillimeter-section thicknesses and a 512 × 512 image matrix rivaling DSA 1024 examinations. CT scanning is ideal for evaluating the nonarterial injuries in patients with polytrauma, such as patients with brain, spinal, pelvic, spleen, liver, and/or kidney injuries. A single intravenous administration of contrast material can be used for a combined vascular and nonvascular evaluation. Multidetector-row units may allow CT scanning to eventually become the new criterion standard. ^{[24, 34, 35, 36]}

Technological improvements, however, are a double-edged sword. With decreasing section thickness comes a commensurate increase in the number of images to be interpreted. Examinations involving 600-800 individual sections are possible. High-speed diagnostic workstations can help manage the data load. With these workstations, reviewers can rapidly scroll through high-resolution datasets and combine the data into fewer, thicker sections.

Another data-management tool brings diagnosis full circle to the angiographic interpretation of BAT images. Cross-sectional imaging may be processed with 3-dimensional (3D) and 2-dimensional (2D) surface reconstruction techniques. These reconstruction techniques produce images that simulate anatomic dissections and angiograms. The images allow the clinician to obtain an examination overview and to focus on particular areas of pathology or surgical interest. As technology improves, reconstructed images will eventually supplant source images in routine image interpretation.

The imaging endpoints of reconstructed CT scans and conventional angiograms are essentially similar. Similar interpretation approaches are likely to apply. Angiographic signs and interpretation skills should have significant overlap with the evaluation of CT angiograms.

The disadvantages of CT scanning are as follows: (1) CT scanning may be limited by partial-volume effects (eg, those due to small disruption or subtle intimal injury), (2) cardiac and respiratory motion (eg, artifact in the region of the aortic root) can affect the image quality, and (3) CT scanning can expose the patient to contrast material, particularly if angiography and/or embolization are required. These disadvantages are minimized with modern scanner technology.

Degree of confidence

Studies have shown that CT scanning is sensitive for TAI (83-100%) and that it has a negative predictive value (NPV) of 99-100%. Its specificity of 54-99.8% and its related positive predictive value (PPV) of 9-89% are generally lower than those of angiography.

Mirvis et al explained the reason for the lower specificity of CT scanning. The authors stratified the sensitivity and specificity results based on direct and indirect signs. Although direct signs were 99% specific, they reported that the specificity of indirect signs was only 87%. ^[37]

Further study is needed before the field of CT scanning in traumatology is considered mature. Most of the studies that show a 100% NPV for CT scanning is limited to clinical follow-up. Data about follow-up CT or angiography in patients with negative initial CT scan findings are limited.

However, in a study of blunt trauma patients that included 72 patients who underwent CT angiography (CTA) followed by catheter angiography, Sammer et al found that when CTA findings are indeterminate (ie, showing mediastinal hematoma without direct evidence of aortic or intrathoracic great vessel injury), conventional angiography is unlikely to show an aortic or intrathoracic great vessel injury and may be unnecessary. ^[38] In this study, isolated mediastinal hematoma on CTA had an NPV of 100% for aortic or intrathoracic great vessel injury.

In a study of patients with acute traumatic aortic injury (TAI), Steenburg and Ravenel found that catheter angiograms after contrast-enhanced 64-channel multidetector CT (64-MDCT) scans were of limited value. Of the 10 such cases in their study, direct signs on 64-MDCT scans were confirmed in 3 cases; indirect signs were found to be normal findings in 5 cases; and equivocal findings remained equivocal in 2 cases. No patient with equivocal or indirect findings on 64-MDCT needed surgical repair. These researchers concluded that when 64-MDCT scans show direct signs of TAI, standard angiography is unnecessary. ^[39]

A number of investigators recommend CT scanning as the screening examination of choice for TAI as well as for other injuries (eg, pulmonary laceration, pneumothorax, tracheobronchial injury, spinal injury). Unless the CT scan results are diagnostic, confirmatory angiography is generally required.

Pate et al reported that CT scanning led to a 50% reduction in the use of angiography; this is a benefit in institutions with limited access to angiography. ^[40]

Because of the nearly 100% sensitivity and specificity of helical CT scanning, this modality has largely replaced aortography in the diagnosis of TAI in adults and children.

Atelectasis, the thymus, or the pericardial recess can mimic mediastinal hematoma. Ductus diverticulum or other variants may be confused with traumatic aortic injury (TAI). Mediastinal hematoma may be due to venous or arterial injury. Subtle ruptures, particularly intimal injuries, may not be clearly delineated. Nevertheless, authors of published reports speak highly of CT evaluation.

Transesophageal echocardiography (TEE) for the diagnosis of traumatic aortic injury has its proponents, who support the modality mostly because of its portability and rapid availability in a trauma center setting. TEE has significant proponents in the pediatric trauma literature.

Intravascular ultrasonography is very helpful in identifying traumatic intimal flaps when CTA and aortographic findings are inconclusive in the diagnosis of aortic injuries.

Limitations of the modality include blind spots where the aorta is suboptimally imaged with ultrasonography. Like angiography, TEE is operator and reader dependent and must be performed by well-trained personnel.

Thoracic aortography may be performed with a standard 100-cm-long 5F pigtail catheter, although a 110-cm-long 6F pigtail is preferred. The longer length is helpful for reaching the aortic root in tall patients.

With the 6F catheter, the 27-mL/s contrast-agent injection rate typical of 100-cm-long 5F catheters can be exceeded. The 6F pigtail may be more rigid than 5F catheters. Predilation of the tract with a 5F or 6F dilator or a 6F sheath is recommended.

The imaging rate of thoracic aortography is 4-6 frames/sec on deep inspiration. ^[41]

(Angiograms of aortic trauma in the region of the isthmus are shown below.)

Although most traumatic aortic injuries occur in the region of the isthmus, the entire thoracic aorta should be evaluated. To exclude aortic regurgitation and other aortic root injury, the catheter is first placed above the sinuses of Valsalva. A test injection is used to verify that the catheter tip does not interfere with the aortic root, to avoid spurious aortic regurgitation (see the image below). The contrast-agent injection rate is 20 to 30 mL/s for 50 mL, depending on the patient's cardiac output. Patients with blunt aortic trauma (BAT) are typically young and present in a state of high cardiac output.

The injection and imaging rate may be adjusted on the basis of the test injection results. For intubated patients, mechanical ventilation may be suspended during each angiographic run. Once the aortic root and the ascending aorta are cleared, the catheter may then be withdrawn to the upper ascending aorta to concentrate on the isthmus and distal aorta. The side holes of the catheter should be positioned just proximal to the area of interest for maximum opacification.

Aortographic signs of aortic rupture may be subtle. A minimum of 2 angiographic projections is required to exclude traumatic aortic injury (TAI). Typical projections are a 45° left anterior oblique (LAO) view followed by a steep LAO or lateral view (see the images below). When an isthmus abnormality is suspected, obtain a lateral or steep LAO view to differentiate between the contour abnormality of a pseudoaneurysm and a normal ductus. Other projections are obtained depending on clinical suspicion and the preliminary angiographic findings. In one case from the author's experience, a rupture was better seen on the PA image than on the shallow oblique image (no steep oblique or lateral image had been obtained).

Biplane imaging or rotational angiography can reduce the examination time and the contrast agent load. A biplane combination of 45° right anterior oblique (RAO) and/or LAO views may be used. The imaging frame rate, the source-to–image intensifier distance (SID), and the projection may be altered to accommodate tube loading, kilovolt-peak issues, or cardiac output. A typical frame rate is 4-6 per second. DSA magnification may be helpful in equivocal cases.

Tube-loading limits may become significant for steep oblique or lateral magnified studies. Rotational angiography typically increases tube loading. Whenever tube loading becomes an issue, the traditional steps may be taken—that is, a reduction in the frame rate and the duration of the examination, in the SID, in the magnification, or in the dose (increased DSA quantum mottle).

Degree of confidence

Angiography is the traditional modality for evaluating patients with suspected acute TAI. Approximately 10-20% of patients with blunt trauma who are referred for angiography have positive angiographic findings. In BAT, angiography is best used as a confirmatory tool for patients with positive CT scan findings or as a primary diagnostic tool in hemodynamically stable patients in whom the clinical suspicion for TAI is high.

Gavant reported that angiography is 94% sensitive and 96% specific, with an NPV of 99% and a PPV of 81%. ^[42] Fabian et al reported that angiography had a sensitivity of 100%, a specificity of 97%, an NPV of 97%, and a PPV of 97%. ^[43]

False positives/negatives

A high index of suspicion is required when a patient is evaluated for TAI. Mortality associated with a missed diagnosis (false-negative angiographic finding) is high. Equivocal cases should undergo close clinical and imaging follow-up or exploratory surgery. Surgery is a relatively benign procedure when the risk of morbidity and mortality associated with a missed TAI is considered. As the quality of CT and other noninvasive imaging modalities improve, the surgical exploration rate can be expected to decrease.

False diagnoses can be minimized by paying meticulous attention to the imaging technique and by following up any equivocal findings. False-negative angiographic results may occur when the injury is subtle, when it is not seen in profile (out of plane), or when the injured area is not delineated by the contrast agent column or when it is out of the field of view.

False-positive angiographic results (examples of which appear below) may occur as a result of inadequate angiographic technique (catheter position, injection rate, exposure and position), motion or inflow artifacts, comorbidities (eg, atherosclerosis), the presence of nontraumatic entities (eg, infection), and normal variants in the anatomy.

Common variants at the isthmus are a smooth, fusiform, circumferential widening at the isthmus; a ductus diverticulum; and an asymmetrical ductus with bronchointercostal artery. No other signs of TAI should be present if a normal variant is to be safely diagnosed.