BACKGROUND: To validate the hypothesis that cryptogenic stroke with multiple infarcts included embolic stroke due to left atrial appendage (LAA) dysfunction, the present retrospective observational study was aimed to clarify the association between LAA flow velocity (LAA-FV) and multiple infarcts in patients with cryptogenic stroke. METHODS: From consecutive patients with cryptogenic stroke admitted to our hospital within 7 days after onset, patients without brain magnetic resonance imaging (MRI) on admission or without transesophageal echocardiography (TEE) during acute hospitalization were excluded, and the remaining patients were enrolled. Multiplicity of fresh infarcts was assessed using diffusion-weighted images from brain MRI. LAA-FV was defined as LAA peak emptying flow velocity on TEE. RESULTS: Of 786 enrolled patients, 522 patients (66%) had a single infarct, and the remaining 264 patients (34%) had multiple infarcts. The percentage of multiple infarcts decreased with increasing quartiles of LAA-FV (p for trend <0.001). The adjusted odds ratio for multiple infarcts decreased with increasing quartiles of LAA-FV (adjusted odds ratio in the fourth quartile, 0.39; 95% confidence interval, 0.25-0.60; compared with the first quartile). LAA-FV as a continuous variable was negatively associated with multiple infarcts (adjusted odds ratio per 10 cm/s, 0.87; 95% confidence interval, 0.81-0.92). CONCLUSIONS: Reduced LAA-FV on TEE was associated with multiple infarcts in patients with cryptogenic stroke. The present findings indicate that cryptogenic stroke with multiple infarcts includes embolic stroke due to LAA dysfunction.
BACKGROUND: To validate the hypothesis that cryptogenic stroke with multiple infarcts included embolic stroke due to left atrial appendage (LAA) dysfunction, the present retrospective observational study was aimed to clarify the association between LAA flow velocity (LAA-FV) and multiple infarcts in patients with cryptogenic stroke. METHODS: From consecutive patients with cryptogenic stroke admitted to our hospital within 7 days after onset, patients without brain magnetic resonance imaging (MRI) on admission or without transesophageal echocardiography (TEE) during acute hospitalization were excluded, and the remaining patients were enrolled. Multiplicity of fresh infarcts was assessed using diffusion-weighted images from brain MRI. LAA-FV was defined as LAA peak emptying flow velocity on TEE. RESULTS: Of 786 enrolled patients, 522 patients (66%) had a single infarct, and the remaining 264 patients (34%) had multiple infarcts. The percentage of multiple infarcts decreased with increasing quartiles of LAA-FV (p for trend <0.001). The adjusted odds ratio for multiple infarcts decreased with increasing quartiles of LAA-FV (adjusted odds ratio in the fourth quartile, 0.39; 95% confidence interval, 0.25-0.60; compared with the first quartile). LAA-FV as a continuous variable was negatively associated with multiple infarcts (adjusted odds ratio per 10 cm/s, 0.87; 95% confidence interval, 0.81-0.92). CONCLUSIONS: Reduced LAA-FV on TEE was associated with multiple infarcts in patients with cryptogenic stroke. The present findings indicate that cryptogenic stroke with multiple infarcts includes embolic stroke due to LAA dysfunction.