Feasibility study of rapid three dimensional MR angiography in visualization of carotid atherosclerotic plaque
10.3760/cma.j.issn.1005?1201.2017.04.013
- VernacularTitle:采用快速三维MR血管壁成像显示颈动脉粥样硬化斑块的可行性
- Author:
Li DONG
;
Bo LI
;
Zhanhong WANG
;
Dongxu LU
;
Yanyang WANG
;
Zhaoqi ZHANG
- Keywords:
Atherosclerosis;
Magnetic resonance imaging;
Vessel wall
- From:
Chinese Journal of Radiology
2017;51(4):299-303
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the feasibility of delineation of the atherosclerotic plaque in carotid artery using compressed sensing three dimensional motion-sensitized driven equilibrium prepared rapid gradient echo (CS-3D MERGE) technique. Methods Twenty-three patients who underwent carotid endarterectomy (CEA) were enrolled prospectively. In all patients, bilateral carotid arteries were scanned by CS-3D MERGE and contrast-enhanced MR angiography (MRA) simultaneously. Image quality of CS-3D MERGE images was scored. Images from the CS-3D MERGE sequence and contrast-enhanced MRA were used to measure the carotid stenosis, which were divided into four groups:normal group, mild stenosis group, moderate stenosis group, and severe stenosis group. The results were compared between the two methods. And images from the CS-3D MERGE sequence were compared with corresponding histology in identifying major plaque components including lipid rich necrotic core (LRNC), intraplaque hemorrhage (IPH), and calcification (CA). Spearman rank correlation coefficient was used to compare the correlation between the CS-3D MERGE sequence and contrast-enhanced MRA in measuring the carotid stenosis. Agreement was tested comparing images from the CS-3D MERGE sequence with corresponding histology in identifying major plaque components including lipid-rich necrotic core, intraplaque hemorrhage, and calcification. Results Twenty-three CEA patients finished bilateral CS-3D MERGE scanning successfully. Image quality was 3.16±0.25. There was an excellent correlation between CS-3D MERGE and MRA in measuring stenosis (r=0.95, P<0.01). The agreement between CS-3D MERGE and histological results for LRNC detection was 76.2% (16/21). It was less sensitive for IPH detection (71.4%, 15/21). CS-3D MERGE identified all CA accurately (100.0%, 21/21). Sensitivity and specificity were 86.6% (13/15) and 50.0% (3/6) for LRNC, 73.3% (11/15) and 66.6% (4/6) for IPH, 100.0% for CA respectively(16/16, 5/5). Conclusion CS-3D MERGE, a single sequence, can be used to quantify carotid stenosis and plaque components conveniently.
