+++
Make the Diagnosis: Thoracic Aortic Dissection
++
Although no current studies address the prior probability of an acute aortic dissection, a recent population-based epidemiologic study allows us to infer a 2% thoracic aortic dissection prevalence among patients with chest pain.5
+++
Population for Whom a Thoracic Aortic Dissection Might Be Considered
++
+++
Detecting the Likelihood of a Thoracic Aortic Dissection
++
Although clinical history, physical examination, and chest radiography can be suggestive of aortic dissection, none of these elements alone is sufficiently sensitive or specific to independently rule in or rule out this high-stakes diagnosis. Nonetheless, certain findings on the clinical evaluation can be helpful in suggesting the diagnosis and the need to perform a reference standard investigation such as CT angiography or TEE (Table 50-9). Almost all patients have severe pain (pooled sensitivity, 90%) of sudden onset (pooled sensitivity, 84%). The presence of a pulse or blood pressure differential from one side of the body to the other in a patient with severe chest pain is not often found in patients with dissection (sensitivity, 31%), but the finding increases the likelihood of aortic dissection when discovered (positive likelihood ratio [LR], 5.7). Similarly, a new focal neurologic deficit occurs infrequently (sensitivity, 17%) but also increases the likelihood of an aortic dissection when it is present (positive LR, 6.6-33.0). A widened mediastinum on chest radiograph is neither reliably present (pooled sensitivity, 64%) nor diagnostic of aortic dissection (positive LR, 2.0). However, almost all chest radiographs from patients with dissection will have some abnormality (pooled sensitivity, 90%), so a completely normal chest radiograph result decreases the probability of dissection being present (LR, 0.3).
++