Author + information
- Ajay Gupta, MD∗ (, )
- Gino Gialdini, MD,
- Ashley E. Giambrone, PhD,
- Michael P. Lerario, MD,
- Hediyeh Baradaran, MD,
- Babak B. Navi, MD, MS,
- Randolph S. Marshall, MD, MS,
- Costantino Iadecola, MD and
- Hooman Kamel, MD
- ↵∗Department of Radiology, Weill Cornell Medical College, 525 East 68th Street, Starr 8A, Box 141, New York, New York 10065
Most ischemic stroke classification schemes require an atherosclerotic plaque to cause ≥50% stenosis for the stroke to be ascribed to large-artery atherosclerosis (1). However, recent data (2–4) suggest that some of the 30% of ischemic strokes whose cause is currently classified as unknown may actually arise from nonstenosing large-artery atherosclerotic plaque. Improving our ability to recognize culprit nonstenosing plaque as the cause of stroke has important implications for precision medicine for secondary stroke prevention because such patients may benefit from dual antiplatelet drugs and emerging drugs for lowering serum cholesterol.
We examined the association between nonstenosing, vulnerable large-artery plaque and ischemic stroke using data from our prospective stroke registry. We reviewed all available medical records and assigned a stroke etiology using a standard ischemic stroke classification scheme (1). Because we were interested in stroke risk from atherosclerotic plaques that are currently not recognized as a cause of stroke, we excluded patients with large-artery atherosclerosis based on these standardized criteria, and instead focused on patients with stroke from cardioembolism, small-vessel occlusion, and undetermined cause. To reduce the risk of unmeasured confounding between patients, we performed a within-subjects comparison of the prevalence of vulnerable internal carotid artery (ICA) plaque on the side ipsilateral to an infarct versus the contralateral side. We therefore limited our study to patients with acute brain infarction(s) limited to the vascular territory of a single ICA as confirmed on magnetic resonance imaging (MRI).
We ascertained vulnerable plaque using sequences from routine magnetic resonance angiograms (MRAs). The details of our imaging technique are provided in a previous report (2). Briefly, all brain MRI and neck MRA studies were performed using standard MRI equipment without high-resolution surface carotid coils or gadolinium. We ascertained complicated carotid plaque based on the qualitative assessment of intraplaque high-intensity signal (IHIS) on axial 3-dimensional-time-of-flight images, a presumed marker for intraplaque hemorrhage (2). To be classified as IHIS, an atherosclerotic lesion’s signal had to meet a certain threshold (50% greater than the background signal intensity of the sternocleidomastoid muscle using region-of-interest analysis) and be distinguishable from the normal flow-related enhancement visible in the patent ICA lumen.
We identified 109 eligible patients (50.0% females, mean age 69.5 ± 14.7 years). Patients with cardioembolic stroke had significantly higher National Institutes of Health Stroke Scale scores and a higher proportion of atrial fibrillation. Patients with cryptogenic stroke had a slightly higher proportion of patients receiving echocardiography compared with other stroke subtypes. The stroke subtypes were otherwise comparable in demographics, vascular risk factors, and the extent of diagnostic evaluation. The median interval between MRI brain and MRA neck and between stroke onset and MRA neck was 0 and 1 day, respectively.
Overall, 22 of 109 patients (20.2%) had <50% ICA plaque with IHIS ipsilateral to the side of infarction, compared with 9 of 109 (8.3%) patients who had IHIS in <50% ICA plaque contralateral to the side of infarction (p = 0.01) (Table 1). The median degree of vessel luminal narrowing was not significantly different on the ipsilateral versus contralateral side of the infarction (p = 0.67). Subgroup analysis revealed a significantly higher proportion of IHIS in ICA plaques ipsilateral to the side of infarction in cryptogenic stroke patients (p <0.001), but not in patients with strokes from cardioembolism (p = 0.76) or small-vessel occlusion (p = 0.49).
Our results suggest that some strokes from large-artery atherosclerosis are currently not being recognized as such because the plaque causes <50% stenosis. Our findings also suggest that useful plaque composition data can be extracted using only a standard luminal imaging technique (time-of-flight MRA) that can be integrated into a rapid acute stroke imaging protocol. Because our IHIS-positive subgroup was small (n = 31), larger confirmatory prospective studies are now warranted which should also investigate stroke recurrence rates in such patients. Such studies are important because if patients who are currently labeled as cryptogenic stroke are more correctly identified as harboring a culprit large-artery atherosclerotic lesion, they may benefit from intensified and targeted therapy aimed at reducing their risk of recurrent stroke and other major adverse cardiovascular events.
Please note: Dr. Gupta was supported in part by the Foundation of the American Society of Neuroradiology Scholar Award. Dr. Kamel was supported by grant K23NS082367 from the National Institute of Neurological Disorders and Stroke. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Adams H.P. Jr..,
- Bendixen B.H.,
- Kappelle L.J.,
- et al.
- Gupta A.,
- Gialdini G.,
- Lerario M.P.,
- et al.