Vascular imaging in the initial evaluation of embolic stroke source.

PURPOSE: During acute ischemic stroke evaluations, neurovascular imaging is commonly performed to localize the source of a thromboembolus and to identify vascular stenoses. In this study, we aimed to analyze 1) the usefulness of intracranial and/or cervical CTA and MRA and carotid doppler ultrasound (DUS) for identifying the stroke source and 2) the incidence of vascular stenoses across stroke etiologies.

METHODS: We retrospectively reviewed intracranial and/or cervical CTA, DUS and MRA studies to identify the source of the acute stroke by Trial of ORG 10172 in Acute Stroke Treatment (TOAST) criteria for 200 consecutive patients admitted with a stroke to our tertiary hospital. Stroke etiologies were ascertained and the rates of intracranial and cervical vascular stenoses were stratified by stroke type.

RESULTS: Of 200 patients, the most probable sources of stroke were small vessel disease (29%), cardioembolic (26.0%) and atheroembolic (23.5%). Across all groups, 27.5% of patients had ≥70% stenosis on neurovascular imaging. The rate of ≥70% vascular stenosis in the neck was 6.9% and 5.8% in the small vessel and cardioembolic categories, respectively.

CONCLUSIONS: The TOAST etiologies of strokes were nearly equally distributed. Neurovascular imaging was of high utility for identifying large vessel intracranial stenoses in patients presenting with acute stroke across all etiologies, although neck CTA/MRA had a lower rate of positive studies with cardiogenic and small vessel strokes. These findings have implications on the use of CTA/MRA in acute stroke work-up.