The large differential diagnosis for chest pain plus the many end-organ ischemic manifestations associated with aortic dissections make the diagnosis challenging. The most important differential diagnoses are listed in Table 59-2.
TABLE 59-2Differential Diagnosis of Aortic Dissection ||Download (.pdf) TABLE 59-2 Differential Diagnosis of Aortic Dissection
Myocardial infarction or acute coronary syndromes
Musculoskeletal disease of the extremity
Spinal cord injuries and disorders
Pulmonary disorders, including pulmonary embolus, pneumonia, pleurisy, pneumothorax
Ischemic manifestations may change with time (as the dissection progresses), and this may distract the physician from making the correct diagnosis. Rupture of the dissection into the true aortic lumen may cause a cessation of symptoms, and the correct diagnosis may then be inappropriately dismissed. History, physical examination, and chest radiography can suggest the diagnosis, but only if one is alert to aortic dissection as one of the diagnostic possibilities in a patient with acute chest pain, syncope, or acute focal neurologic signs. Factors associated with misdiagnosis include walk-in mode of admission, normal mediastinal width/aortic contour on chest radiograph, absent extremity pulse amplitude differences, and nonspecific symptoms.17,18,19
It may be difficult to differentiate aortic dissection from acute coronary syndromes on ECG, because both conditions are associated with ECG changes; dissection may limit or obstruct coronary artery blood flow. Abnormal ECG findings include new Q waves or ST-segment elevation in 3% to 4%, ST depression in 15% to 22%, and nonspecific ST and T-wave changes in 41% to 62%.5,9,20 The ECG is normal in only 19% to 31% of patients.5,20
Several potential biomarkers have been investigated for their utility to identify or exclude aortic dissection.21 D-Dimer is the marker most thoroughly investigated. A meta-analysis of seven studies involving 298 subjects with acute aortic dissection and 436 without found a sensitivity of 97% (95% confidence interval, 94% to 99%) and negative predictive value of 96% (95% confidence interval, 93% to 98%) using a D-dimer cut point of 500 ng/mL (1620 nmol/L).22 The specificity was low at 56% (95% confidence interval, 51% to 60%). Guidelines do not endorse the use of D-dimer as the sole means of excluding aortic dissection,15 and several authors have cautioned against this practice.23,24,25,26 One report found that young adult patients with short dissection length and thrombosed false lumen were likely to have a false-negative D-dimer.24 The false-negative rate using D-dimer is as high at 18%.25
A plain chest radiograph may provide important clues for the diagnosis. However, from 12% to 37% of patients have no abnormality, and this study should not be used to exclude dissection.5,27 The most common radiographic abnormality is a widened mediastinum or abnormal aortic contour. Other possible findings include pleural effusion, displacement of aortic intimal calcification, and deviation of the trachea, mainstream bronchi, or esophagus (Figure 59-1).
Abnormal aortic contour on chest radiography. Frontal and lateral radiographs of the chest in a patient with type B aortic dissection reveal an abnormal aortic contour (arrow). A right pleural effusion is present, and multiple postoperative clips and wires are also seen.
CT (especially multidetector-row CT) is the imaging modality of choice for diagnosis of dissection.11,15,28 CT can reliably identify a false lumen (Figure 59-2) and can provide additional details such as the anatomy of the dissection, the location of the dissection flap, extension of the flap into great vessels (Figure 59-3), signs of aortic rupture, and signs of end-organ damage. CT protocols should be both with and without IV contrast. Invasive catheter angiography is rarely necessary.
CT image of a type A aortic dissection. True and false lumens are present in the ascending aorta and descending aorta (descending false lumen at arrow) on noncontrast (left) and contrast (right) images. AF = ascending false lumen; AT = ascending true lumen; DT = descending true lumen.
Type B dissection into the iliac arteries. Contrast CT image of dissection extending into the iliac arteries (anterior to vertebral body). True and false lumens are visible in both arteries (arrows).
CT may also diagnose intramural hematoma and penetrating atherosclerotic ulcer.28 Penetrating atherosclerotic ulcer can be difficult to distinguish from large atheromatous plaques (Figure 59-4). CT diagnosis of penetrating atherosclerotic ulcer depends on extension of the ulcer past the intima. Ulcers often have overhanging edges and focal outpouchings of the aorta itself. Intramural hematoma is often identified by a high-signal mass in the aorta on CT (Figure 59-5). This often appears as a crescent and is best seen on noncontrasted images.
Noncontrast CT image of a penetrating aortic ulcer in the descending aorta (arrows), demonstrating an outpouched, abnormal contour of the aorta in three sections. [Image used with permission of Dr. Ernest Scalzetti, MD.]
Contrast CT image of an intramural hematoma (arrows point to the crescent-shaped lesion along the posterior lateral aortic wall) in the descending aorta. [Image used with permission of Dr. Ernest Scalzetti, MD.]
In experienced hands, transesophageal echocardiography may be as sensitive and specific as CT. The procedure generally has to be performed under moderate sedation or even general anesthesia. Known esophageal disease is a relative contraindication. Sound transmission is disrupted by air in the trachea or left bronchia, which may make evaluation of the ascending aorta difficult. The accuracy and precision of transesophageal echocardiography are highly operator dependent. MRI has been used to evaluate stable patients with suspected aortic disease.28
Coronary CT angiography, or the "triple rule-out," can diagnose and differentiate coronary artery disease, pulmonary embolism, and acute aortic dissection.29,30 However, it requires a specialized contrast infusion protocol to image the three vascular beds of interest and an increased radiation dosage.31 Furthermore, the "triple rule-out" study has not been shown to improve diagnostic yield, reduce clinical events, or diminish downstream resource use.32