3.0Tesla High-Resolution Magnetic Resonance Imaging-Based Research On Middle Cerebral Arterial Atherosclerotic Disease

Part I Efficiency of3.0Tesla High Resolution MRI vs Digital Subtraction Angiography in Measuring Stenosis Degree of Atherosclerotic Middle Cerebral Artery[Objective]To investigate the diagnostic efficiency of3.0tesla high resolution MRI (HR MRI) vessel wall imaging in measuring stenosis degree of atherosclerotic middle cerebral artery while taking digital subtraction angiography (DSA) findings as the gold standard.[Methods]Totally89MCA segments from78patients were included in this study. All patients underwent DSA and HR MRI vessel wall imaging for examining the degrees of MCA-M1stenosis; the accuracy of HR MRI was evaluated while the DSA measurements were taken as the gold standard, and we also analyzed the specificity and sensitivity of HR MRI in detecting moderate-to-severe MCA-M1stenosis (50%-99%).[Results]1. Of the89MCA segments, six were normal,10had mild stenosis (1%-49%),32had moderate stenosis,36had severe stenosis, and5had complete occlusion.2. The stenosis degrees of MCA-M1measured by HR MRI were significantly higher than those measured by DSA (P<0.001). But HR MRI-derived stenosis degree correlated well with that detected by DSA (Pearson r=0.731, P<0.001; Spearman r=0.663, P<0.001), and the results of two methods were consistent to some extent (κ=0.471,P<0.001).3. The sensitivity, specificity, and area under ROC curve of HR MRI for detecting MCA-M1moderate-to-severe stenosis (50%-99%) were98.63%,81.25%,0.985; for detecting severe stenosis (70%-99%) were80.49%,60.42%,0.747; and for differentiating moderate from severe stenosis (50%-69%,70%-99%) were78.26%,35.56%, and0.586, respectively.[Conclusion]1.3.0T high-resolution MRI wall imaging can accurately measure the degree of of MCA-M1stenosis, and the results have good consistency with DSA findings.2.3.0T high-resolution MRI has satisfactory diagnostic efficiency in detecting moderate-to-severe stenosis (50%-99%) within MCA-M1, and it may be widely used for non-invasive screening of middle cerebral arterial atherosclerotic disease. Part II3.0T HR MRI-based Stratification of Stroke Risk Associated with MCA Atherosclerotic Disease[Objective]To assess the feasibility of HR-MRI imaging for stratifying stroke risk in patients with MCA atherosclerotic disease by comparing the vessel wall and plaque characteristics between patients with symptomatic and asymptomatic MCA diseases.[Methods]Sixty-six patients with MCA-M1segment moderate-to-severe atherosclerotic stenosis (50%-99%), which were confirmed by DSA, were divided into symptomatic group (n=27) and asymptomatic group (n=39),3.0T HR-MRI wall imaging examination was done to all patients. The vessel area (VA) and lumen area (LA) at the maximal lumen narrowing (MLN) site and reference site were measured on cross-sectional images of MCA wall and plaque on HR-MRI. The quantitative and qualitative parameters derived from cross-sectional images were compared between the two groups. The quantitative parameters included VAnarrow, LAnarrow, VAreference, LAreference, wall area (WA=VA-LA), plaque size (PS=WAnarrow-WAreference), percent plaque burden (PB=PS/VAnarrow×100%), and remodeling index (RI=VAnarrow/VAreference).The qualitative parameters included plaque distribution of MCA (superior, inferior, ventral, dorsal, or annular) and plaque stability (enhanced or non-enhanced). RI≥1.05was defined as positive remodeling (PR) and RI<1.05as non-PR.[Results]1. There were no significant differences regarding the demographic data or degree of MCA stenosis between the symptomatic and asymptomatic groups.2. PS and PB of the symptomatic group were significantly higher than those of asymptomatic group (6.30±3.41vs4.26±2.90mm2, P=0.011;0.42±0.16vs0.31±0.16, P=0.008). The VAnarrow, LAnarrow, and WAnarrow of the symptomatic group were higher than those of asymptomatic group, but showing no significant difference. The VAreference, LAreference, and WAreference were similar between the two groups.3. Compared with the asymptomatic group, the symptomatic group had significantly greater RI (1.17±0.30vs0.96±0.20, P=0.001), higher prevalence of PR (66.67%[18/27] vs43.59%[17/39]) and lower prevalence of non-PR (33.33%[9/27] vs56.41%[22/39]).4. Atherosclerotic plaques of MCA-M1stenosis were most frequently located at the ventral wall (47.26%), followed by annular area (25.90%) and inferior wall(19.09%), and least located at the superior (5.49%) and dorsal wall (2.56%), with significant difference found between the ventral wall and all the other parts(P<0.001), also between the annular, inferior and superior, dorsal walls(P<0.001). Further analysis showed that the plaque distribution was similar beween the two groups, with the locations mainly in the ventral, inferior and the annular walls, and less in the posterior and superior walls.5. Of all the plaques,73.77%were reinforced on contrast-enhanced T1WI and26.23%were not enhanced, with no significant difference found between asymptomatic and symptomatic groups (P=0.310).[Conclusion]1. Based on the vessel wall images on HR-MRI between the symptomatic and asymptomatic patients with moderate-to-severe MCA stenosis, it is found that greater plaque burden and positive remodeling are high stroke risks. Further prospective studies are required to investigate whether HR-MRI can help to stratify stroke risk in patients with MCA atherosclerotic disease.2. Moderate-to-severe MCA plaques mainly locate opposite to the orifices of penetrating arterial branches (ventral, inferior), and the annular plaque is an important plaque subtype.3. The plaque stability estimated by plaque enhanced or not on contrast-enhanced T1WI has a limited value in predicting stroke risks. Part Ⅲ Preinterventional3.0T HR MRI-based analysis of In-Stent Restenosis factors Following Stent Implantation for Middle Cerebral Artery Atherosclerotic Stenosis[Objective]To compare the preinterventional atherosclerotic vessel wall properties and plaque characteristics of MCA on3.0T HR MRI between in-stent restenosis (ISR) and non-ISR patients, so as to explore the effect of preinterventional arterial wall and plaque characteristics for ISR following self-expanding stent implantation, hoping to lay a foundation for selecting proper MCA atherosclerotic patients with potential benefit from endovascular stent placement.[Methods]From February2011to June2013, a retrospective nested case-control study was carried out among31patients who had a recent transient ischemic attack or stroke attributed to atherosclerotic stenosis of50%to99%of the MCA Ml segments. All patients underwent aggressive medical management plus percutaneous transluminal angioplasty and stenting (PTAS) using the self-expanding stent (Wingspan, Solitaire, or Enterprise), and they were followed up for over6months at regular intervals using DSA, and the ISR patients were noted. ISR was defined as an angiographically verified50%stenosis within the stent or at the edge of the stent in the range of5mm, or with20%absolute luminal loss. All patients received3.0T HR MRI imaging pre-interventional PTAS for calculating VAnarrow, LAnarrow, VAreference, and LAreference.Five patients were diagnosed as ISR by DSA follow-up (ISR group), each of them were individually matched with two patients confirmed free from ISR (n=10, non-ISR group). All of them were matchable in age (<55yr,≥55yr), diabetic mellitus (yes/no), length of vascular lesion (≤5mm,5-10mm,≥10mm), and residual stenosis immediately after stenting (<30%,≥30%) as baseline. The pre-interventional quantitative and qualitative parameters of MCA wall and plaque measured on the cross-sectional images of HR-MRI were compared between the ISR and non-ISR groups. The quantitative parameters included VAnarrow, LAnarrow, VAreference, LAreference, WAnarrow, WAreference, PS, PB, and RI. The qualitative parameters included plaque distribution of MCA (superior, inferior, ventral, dorsal, or annular). RI≥1.05was defined as positive remodeling (PR) and RI≤0.95as negative remodeling (NR).[Results]1. The demographic data and degree of MCA stenosis were similar between ISR and non-ISR groups. 2. Compared with the non-ISR patients, ISR patients had smaller VAnarrow, WAnarrow, PS, and PB (8.70±4.47vs14.62±3.55mm2, P=0.015;7.37±4.43vs13.35±3.22mm2, P=0.010;2.29±1.67vs7.47±3.68mm2, P=0.011;0.23±0.13vs0.50±0.15, P=0.004), while LAnarrow, VAreference, LAreference, and WAreference were similar between the two gorups.3. The RI of ISR group was significantly smaller than that of non-ISR (0.78±0.17vs1.26±0.24, P=0.002). ISR patients had more frequent NR(5/5,100%) than non-ISR patients (0/10,0), and non-ISR patients had more frequent PR (9/10,90%) than ISR patients (0/5,0).4. Atherosclerotic plaques of MCA-M1stenosis were most frequently located at the ventral wall (62.00%, P<0.001), followed by annular (17.78%) and inferior (11.33%) wall, superior (4.44%), and dorsal wall (4.44%). ISR patients had more annular plaques than non-ISR patients (48.33%vs2.5%, P=0.028), while there were no significant difference for other plaque distribution subtypes (superior, inferior, ventral, or dorsal) between the two groups.[Conclusion]1. Based on3.0T HR MRI, we are the first to explore the factors (pre-interventional wall and plaque characteristics) influencing ISR after intracranial self-expanding stent implantation.2. Contrary to the coronary artery, intracranial atherosclerotic lesions with negative remodeling before intervention have higher incidence of ISR, indicating that negative remodeling may be the major mechanism for middle cerebral arterial ISR after self-expanding stent implantation.3. This study for the first time find that Intracranial atherosclerotic lesions with annular plaque before intervention have higher incidence of ISR. The mechanism of this phenomena is unclear.4. Preinterventional3.0T HR MRI-based examination of vessel wall properties and plaque characteristics of MCA can help to stratify lesions at high-risk of ISR.5. Large sample randomized controlled study is needed to confirm the above evidence.