Detection and Quantification of Symptomatic Atherosclerotic Plaques With High-Resolution Imaging in Cryptogenic Stroke

BACKGROUND AND PURPOSE:High-resolution vessel wall imaging (HR-VWI) is a powerful tool in diagnosing intracranial vasculopathies not detected on routine imaging. We hypothesized that 7T HR-VWI may detect the presence of atherosclerotic plaques in patients with intracranial atherosclerosis disease initially misdiagnosed as cryptogenic strokes. METHODS:Patients diagnosed as cryptogenic stroke but suspected of having an intracranial arteriopathy by routine imaging were prospectively imaged with HR-VWI. If intracranial atherosclerotic plaques were identified, they were classified as culprit or nonculprit based on the likelihood of causing the index stroke. Plaque characteristics, such as contrast enhancement, degree of stenosis, and morphology, were analyzed. Contrast enhancement was determined objectively after normalization with the pituitary stalk. A cutoff value for plaque-to-pituitary stalk contrast enhancement ratio (CR) was determined for optimal prediction of the presence of a culprit plaque. A revised stroke cause was adjudicated based on clinical and HR-VWI findings. RESULTS:A total of 344 cryptogenic strokes were analyzed, and 38 eligible patients were imaged with 7T HR-VWI. Intracranial atherosclerosis disease was adjudicated as the final stroke cause in 25 patients. A total of 153 intracranial plaques in 374 arterial segments were identified. Culprit plaques (n=36) had higher CR and had concentric morphology when compared with nonculprit plaques (P≤0.001). CR ≥53 had 78% sensitivity for detecting culprit plaques and a 90% negative predictive value. CR ≥53 (P=0.008), stenosis ≥50% (P