Background and Purpose Transesophageal echocardiography has a high yield for detecting potential cardiac sources of embolism in patients with clinical risk factors for cardioembolism or unexplained stroke. The yield in other stroke subtypes is unknown.Methods We classified 145 consecutively admitted patients into stroke subtypes based on clinical findings, brain imaging, and carotid ultrasound. Both transesophageal and transthoracic echocardiography were performed to detect left atrial thrombi, spontaneous echo contrast, atrial septal aneurysm, interatrial shunts, ventricular thrombus or aneurysm, and myxomatous mitral valve.Results Transesophageal echocardiography documented at least one of these findings in 46% of the patients compared

Transcranial Doppler can detect cerebral microemboli. These emboli may be a risk factor for embolic stroke. We studied the prevalence of microemboli in patients referred for echocardiography. Forty-two patients were evaluated. Patients were studied with continuous monitoring over one middle cerebral artery for 30 minutes, and the number of microemboli was recorded. Patients were divided into three groups, those with prosthetic heart valves (group A, n = 15), atrial fibrillation (group B, n = 14), and no major cardiac risk factor (group C, n = 14). Seventeen percent (7 of 42) of all patients had microemboli. In group A, 5 of 15 (33%) had microemboli. In group B, 2 of 13 (15%) patients had microemboli. Twenty-five percent (7 of 28) of patients in groups A and B combined (A+B) had microemboli. No patients (0 of 14) in group C had microemboli. Groups A and A+B had significantly more emboli than group C (P < .05). Prosthetic heart valve patients with emboli more commonly had a history of prior stroke than valve patients without emboli (3 of 5 versus 2 of 10). The number of emboli seen per 30-minute monitoring session was greater in patients with a prior history of stroke than in patients without (10 microemboli versus 3). Microemboli can be found in a significant percentage of selected patients referred for echocardiography. The prevalence of microembolism is greater in patients with a known high risk of embolization (eg, prosthetic valves) and less in patients with a lower risk of embolization (eg, atrial fibrillation). These microemboli may be associated with an increased prevalence of previous stroke in patients with prosthetic valves.

No abstract