We hypothesized that paradoxical embolism may be one of the main mechanisms of silent cerebral infarction (SCI) located outside the perforating artery territory. In the present study, we evaluated the association between SCI and paradoxical embolism detected by agitated saline transcranial Doppler (TCD) monitoring in healthy subjects without history of stroke or transient ischemic attack (TIA). We also compared the frequency and amount of paradoxical embolism between healthy subjects with SCI and acute stroke patients with presumed cryptogenic embolism or conventional stroke mechanisms.

Inclusion Criteria:

Healthy subjects

• Subjects with age over 20 years
• Subjects who consecutively visited a health promotion center for a health screening examination that included routine laboratory test, cardiac work-ups (12-lead electrocardiography and/or echocardiography), and brain magnetic resonance image (MRI) and MR angiography (MRA)
• Subjects who showed silent brain infarction (SCI) which was defined as a focal hyperintensity lesion 3 mm or large in diameter in the fluid-attenuation inversion recovery (FLAIR) images without a history of a corresponding stroke or transient ischemic attack (TIA)
• Subjects who performed agitated saline transcranial Doppler (TCD) monitoring

Patients with acute stroke (both cryptogenic embolic stroke and conventional stroke mechanisms)

• Patients with age over 20 year
• Patients with focal neurologic deficits presented within 7 days of symptom onset
• Patients with acute ischemic lesions on diffusion-weighted image (DWI)
• Patients who performed agitated saline TCD monitoring

Exclusion Criteria:

Healthy subjects

• Subjects with the presence of previous history of stroke or TIA
• Subjects with the presence of dementia on history taking or mini-mental status examination
• Subjects with the presence of proximal source of embolism, including atrial fibrillation or significant (≥50%) stenosis on cervicocerebral artery on MRA or carotid Duplex
• Subjects who had SCI restricted within perforator territory (basal ganglia or pons) suggesting that microangiopathy rather than embolism could be the main mechanism of SCI
• Subjects in whom agitated saline TCD monitoring was not performed or those who had poor temporal windows

Patients with acute stroke (both cryptogenic embolic stroke and conventional stroke mechanisms)

• Patients in whom agitated saline TCD monitoring was not performed or those who had poor temporal windows

Tertiary Hospital