Aortic atheromas and acute ischemic stroke: a transesophageal echocardiographic study in an ethnically mixed population.

PURPOSE: Proximal aortic atheromas have been suggested as a potential ischemic stroke determinant in the elderly, especially in cases of unexplained (cryptogenic) stroke. Our aim was to assess the potential role of proximal aortic atheromas as an independent risk factor for stroke by comparing their frequency in patients with acute ischemic stroke and in stroke-free control subjects. The frequency of atheromas was also compared among different ethnic groups. PATIENTS AND METHODS: A case-control study was conducted in 106 patients with acute ischemic stroke and 114 stroke-free control subjects. The presence of atheromas of the proximal portion of the aorta was assessed by biplane transesophageal echocardiography. Atheromas were categorized on the basis of their thickness (0.2 to 0.4 cm, small; > or = 0.5 cm, large) and complexity (i.e., ulceration or mobility). The association between aortic atheromas and ischemic stroke was tested, controlling for patients' demographic variables and stroke risk factors. In stroke patients, subgroup analyses were performed to test the associations between aortic atheromas and stroke diagnostic subtypes (determined cause versus cryptogenic) and presence and degree of carotid stenoses by duplex Doppler examination.
RESULTS: The frequency of large aortic atheromas was greater in stroke patients than in controls (26% versus 13%; crude odds ratio [OR] 2.4, 95% CI 1.2 to 4.7); ulcerated or mobile atheromas also tended to be more frequent in stroke patients (12% versus 5%; OR 2.5, 95% CI 1.0 to 6.8). Differences were entirely attributable to the subgroup of patients aged 60 years or older, in whom the frequency of ulcerated or mobile atheromas was particularly high among cryptogenic stroke patients (22% versus 8% in control subjects; OR 3.4, 95% CI 1.1 to 11.2). Multivariate analysis showed the presence of large atheromas to be independently associated with stroke in the entire study group (adjusted OR 2.6, 95% CI 1.1 to 5.9) and in the older subgroup (OR 2.4, 95% CI 1.1 to 5.7). Carotid stenosis > or = 60% was more frequent with increasing size and complexity of aortic atheromas but had low predictive value (16%) for presence of large atheromas; moreover, 36% of patients with mild or no carotid stenosis had large or complex aortic atheromas. No significant differences were found in the frequency of atheromas by ethnic group.
CONCLUSIONS: Proximal aortic atheromas > or = 0.5 cm in size are a risk factor for ischemic stroke in patients aged 60 years or older. Ulcerated or mobile atheromas may play a role in explaining some cryptogenic strokes in the elderly. The risk for stroke of patients with aortic atheromas may be similar across different ethnic groups. The absence of carotid stenosis does not exclude aortic atheromas as a potential cause for ischemic stroke.