Abstract WP213: Comparison of Clinical and Imaging Characteristic of Cryptogenic Stroke to Known Ischemic Subtypes

    loading  Checking for direct PDF access through Ovid


Introduction: Cryptogenic stroke is defined as not attributable to an identified source despite standard evaluation. The absence of small vessel or large artery disease in such evaluation suggests that cryptogenic stroke may be largely cardioembolic. We hypothesized that cryptogenic stroke would be similar to cardioembolic stroke in clinical and imaging characteristics.Methods: The Greater Cincinnati/Northern Kentucky Stroke Study (GCNKSS) is a population-based study that tracks the regional incidence of stroke. A convenient subsample from the 2010 GCNKSS ischemic stroke cohort (N= 368) was selected for detailed neuroimaging analysis. The study physician subtyped cases based on clinical, radiographic and laboratory findings (carotid ultrasound, echocardiography, vascular imaging). Subtypes included cryptogenic, cardioembolic, large-vessel, small-vessel, undetermined, and other. Three radiologists performed imaging analysis including number of acute infarcts, location and white matter hyperintensity (WMH). Infarct volume was segmented using manual tracing.Results: Of 368 ischemic stroke cases with imaging data, subtypes were 26.4% cryptogenic, 16.3% large vessel, 15.5% small vessel, 24.7% cardioembolic, 5.4% undetermined, and 11.7% other. Compared to cardioembolic, cryptogenic stroke patients were younger, had less hypertension, higher alcohol use, smaller infarct volume and differed in location of stroke. Cryptogenic stroke had more clinical and radiological features in common with large and small-vessel stroke (Table). Undetermined and other had no significant differences to cryptogenic.Conclusion: Contrary to our hypothesis, cryptogenic stroke was different from cardioembolic stroke and appeared more similar to large vessel stroke in clinical and radiological characteristics. Further testing on a larger sample size to evaluate the impact of cardiac event monitoring on subtype distribution is needed.

    loading  Loading Related Articles