Abstract WP49: Distal Hyperintense Vessels Were not a Predictor of Large Vessel Occlusion in Patients With Transient Ischemic Attack

    loading  Checking for direct PDF access through Ovid

Abstract

Introduction: Distal hyperintense vessels (DHV) detected by FLAIR imaging are not uncommon in patients with acute ischemic stroke. The presence of DHV and its predictors has been scarcely evaluated in patients with transient ischemic attack (TIA), being associated to the presence of large vessel occlusion in such patients. We assessed the hypothesis that DHV are frequent in patients with TIA and do correlate with relevant clinical and neuroimaging characteristics.

Methods: We evaluated a database of consecutive patients admitted with TIA from February 2009 to June 2012 who had undergone magnetic resonance imaging within 30 h of symptoms onset and intracranial and extracranial vascular imaging. We analyzed the relationship between DHV, clinical presentation, risk factors, neuroimaging characteristics and large artery stenosis or occlusion. DHV signals were defined on FLAIR images as focal, linear or serpentine, hyperintense signals relative to gray matter. Two neuroradiologists blinded to clinical information reached consensus regarding the presence of DHV.

Results: Seventy-two TIA patients were enrolled. The median time from symptoms onset to MRI was 8:39 h [4:21, 14:13]. DHV signals on FLAIR images were present in 12 (16.7 %) patients. The overall agreement between examiners was good (k 0.67). Patients with DHV had more atrial fibrillation (AF) than those without (41.7% versus 21.7%, p=0.05) and a trend towards more congestive heart failure (CHF) (8.3% versus 1.7%, p=0.2) and diabetes (41.7% versus 21.7%, p=0.1). There were no differences in the frequency of intracranial or cervical arterial stenosis, cerebral microbleeds and white matter abnormalities in patients with and without DHV. In a multivariate logistic regression analysis, only AF had a trend to be a predictor of DHV (OR=4.24, p=0.1). The statistical model to predict DHV including AF, diabetes, and CHF had a moderate fit in terms of discrimination (c statistic=0.62)

Conclusion: DHV signals on FLAIR images occur in patients with TIA and might correlate with clinical variables like AF and not only with large vessel occlusion as previously described. The presence of DVH in patients with TIA and AF might be a surrogate marker for a previous large vessel occlusion spontaneously recanalized.

Related Topics

    loading  Loading Related Articles