Regression of Plaque Enhancement Within Symptomatic Middle Cerebral Artery Atherosclerosis: A High-Resolution MRI Study

Objective: Contrast enhancement is a vital feature of the intracranial atherosclerotic plaque on high-resolution magnetic resonance imaging (HRMRI), but its clinical significance is still unclear. We aimed to quantitatively assess plaque enhancement patterns in the middle cerebral artery (MCA) atherosclerotic plaque. Methods: We conducted a cross-sectional study by prospectively recruiting stroke or transient ischemic attack patients with >30% of MCA stenosis of either side. All patients underwent contrast-enhanced HRMRI scans. Enrolled patients were classified into acute phase (<4 weeks), subacute phase (4–12 weeks) and chronic phase (>12 weeks) groups based on the time interval from stroke onset to imaging scan. Plaque enhancement index was calculated for each MCA lesion at the maximal narrowing site. Results: We identified a total of 89 MCA plaques [53 (60%) symptomatic and 36 (40%) asymptomatic; 57 (64%) acute, 18 (20%) subacute and 14 (16%) chronic] in 58 patients on HRMRI. Among the acute lesions, symptomatic plaques had a significantly stronger plaque enhancement than asymptomatic plaques (symptomatic vs. asymptomatic: 38.9 ± 18.2 vs. 18.2 ± 16.2, p < 0.001). Among the symptomatic lesions, plaque enhancement diminished with increasing time after stroke onset (38.9 ± 18.2, 22.0 ± 22.8, and 5.0 ± 10.1 for acute, subacute, and chronic phase, respectively; p = 0.001). Conclusion: Plaque enhancement in the acute atherosclerotic plaque is closely related to recent ischemic events. In symptomatic atherosclerosis, plaque enhancement regresses over time after ischemic stroke, which may offer the potential to monitor the plaque activity in intracranial atherosclerosis using HRMRI.


INTRODUCTION
Intracranial atherosclerotic disease (ICAD) is a leading cause of ischemic stroke, especially among Asian stroke patients (1,2). The challenge for current imaging modalities is not only to detect the presence of intracranial atherosclerotic plaque but also to clarify the high-risk lesions that are vulnerable to thrombosis. In addition to lumen stenosis that has long been used as an imaging standard for risk-stratifying patients, pathological studies reveal that some other plaque features, such as large lipid core, increased neovascularization, and plaque inflammation, also contribute to plaque instability and thromboembolic events (3,4), highlighting the importance of visualizing the plaque components to predict future vascular events.
High-resolution magnetic resonance imaging (HRMRI) has gained prominence as an imaging approach to reliably visualize the vessel wall pathology and thus is an ideal imaging tool to identify plaque characteristics that are indicative of plaque instability (5,6). Several clinical studies have been done to validate the utility of HRMRI in characterizing imaging features of high-risk intracranial atherosclerotic lesions, including plaque distribution, intraplaque hemorrhage (IPH), patterns of arterial remodeling (7)(8)(9)(10)(11). Contrast enhancement after intravenous gadolinium-based contrast administration can improve the accuracy of HRMRI in identifying the activity of atherosclerotic lesions by detecting inflammation and neovascularization (12,13). Prior studies of extracranial atherosclerotic plaque have proposed plaque enhancement as an imaging indicator of plaque vulnerability and high stroke risk (14)(15)(16). Plaque enhancement has less been studied in ICAD. Some recent studies revealed the clinical significance of gadolinium enhancement of intracranial lesions by showing the independent association between plaque enhancement and symptomatic intracranial plaques (17)(18)(19)(20)(21). However, the time-course changes of contrast enhancement in intracranial atherosclerosis are still controversial. A few studies demonstrated that the strength of contrast enhancement in symptomatic plaques decreased with increasing time after ischemic stroke (22,23), but the sample size is relatively small to draw a firm conclusion. In the present study, we aimed to investigate whether the strength of plaque enhancement in middle cerebral artery (MCA) atherosclerosis changes after the onset of ischemic stroke. Patients were classified as acute patients if scans were performed within 4 weeks post the onset of symptoms, subacute if between 4 and 12 weeks from stroke onset, and chronic if it was performed beyond 12 weeks from the presentation. Acute and subacute patients were given mono antiplatelet treatment (aspirin 80 mg/day). Chronic patients were extracted from a previous study (24), in which patients were given dual antiplatelet (aspirin 80 mg/day and clopidogrel 75 mg/day) for a total of 4 weeks from the onset of stroke, followed by aspirin 80 mg/day alone. All patients received statin treatment (Atorvastatin, Simvastatin, or Rosuvastatin) with the targeting low-density lipoprotein of < 70 mg/dL (1.8 mmol/L). Other drugs were used for risk factor management following the upto-date guidelines (25).

Image Analysis
Two reviewers who were blinded to the clinical data analyzed the matched pre-and post-contrast T1-weighted images of MCA. The images were graded on a 3-point scale: 1 = poor, 2 = adequate, and 3 = excellent. Images with a score of 2 or 3 were analyzed after being zoomed to 400% by using VesselMass software (Leiden University Medical Center, The Netherlands).
The cross-sectional area of MCA plaques was measured at the site of maximal lumen narrowing on T1-weighted images. Vessel-cerebrospinal fluid interface was used to trace vessel area (VA) manually, and blood-intima interface was used to determine lumen area (LA). The wall area (WA) was estimated by subtracting LA from VA. Plaque load was calculated as WA/VA * 100%. LA of the lesion-free site at the MCA distal portion (LA reference ) was measured as a reference to calculate the luminal stenosis as follows: luminal stenosis = (1 -LA stenosis /LA reference ) * 100%. IPH was defined as an area of high signal (>150% of the adjacent area of the vessel wall) in the plaque, as described previously (10).
Plaque enhancement was quantified by manually tracing the vessel lumen and the outer boundary of vessel wall at the most stenotic site of MCA and measuring the signal intensity of plaque (SI plaque ) on pre-and matched post-contrast T1weighted images. The signal intensity of gray matter (SI graymatter ) was measured by manually drawing a round region of 10-12 mm 2 at the adjacent normal gray matter on matched preand post-contrast T1-weighted images, respectively. The plaque enhancement index was calculated to characterize the extent of plaque contrast enhancement by using the following formula: Enhancement index = [(SI plaque /SI graymatter on post-contrast T1w images) -(SI plaque /SI graymatter on pre-contrast T1w images)] / (SI plaque /SI graymatter on pre-contrast T1w images) * 100% (Figures 1, 2), according to the methods described in a previous study (26).
Lesions were classified as symptomatic or asymptomatic according to the brain infarct on DWI and/or signs of neurological deficits during the ischemic events. A plaque was considered as symptomatic if: (1) it is the only lesions within the ipsilateral MCA distribution of stroke, or (2) it is the most stenotic plaque when tandem lesions were present within the same vascular distribution of stoke. A plaque was defined as asymptomatic if: (1) it is not within the vascular distribution of stoke, or (2) it is within the vascular distribution of stoke but not the most stenotic lesion.

Statistical Analysis
All data analyses were conducted using the SPSS 20.0 software package (SPSS, Inc., USA). Independent-samples T-test (for normally distributed variables) or Mann-Whitney U-test (for non-normally distributed variables) was used to compare means between symptomatic and asymptomatic groups. ANOVA and post-hoc pairwise comparisons (LSD) were used to compare means among acute, subacute and chronic groups. Chi-square or Fisher's exact test was used to compare categorical variables. Univariate and multivariate linear mixed regression analysis were performed to assess how the variables, such as stroke phases (acute, subacute, chronic), symptomatic status, plaque load, luminal stenosis, and IPH, were related to contrast enhancement indexes. A two-tailed P < 0.05 was considered indicative of a significant difference.

Univariate and Multivariate Analysis
The univariate analysis showed that stroke phases (B = 23.853,  Table 3).

DISCUSSION
In this study, we quantitatively assessed the enhancement patterns of MCA by using contrast-enhanced HRMRI and revealed that both the symptomatic status and stroke phases were associated with the extent of plaque enhancement. Among acute plaques, plaque enhancement was strongly associated with the symptomatic ischemic presentation. Among symptomatic plaques, enhancement indexes decreased gradually after the onset of ischemic events. These findings may provide insight into stroke risk stratification.
Substantial contrast enhancement within the carotid plaque has been shown to be correlated with extensive neovascularization and inflammatory cell infiltration in histological studies (14,27), and be associated with elevated serum markers of inflammation, such as interleukin-6 (28). Given the fundamental role of inflammation in the progression and rupture of atherosclerotic plaque, plaque enhancement identified on HRMRI seems to be a vital indicator of plaque vulnerability. Several extracranial artery atherosclerosis studies demonstrated that plaque enhancement was closely related to recent cerebrovascular ischemic events (29,30), suggesting the potential of plaque enhancement as an MRI marker for identifying high-risk atherosclerotic plaque.
In the current study, plaque enhancement in the acute lesions was stronger in symptomatic MCA plaques than that in asymptomatic plaques, consistent with the recent findings showing that the intracranial atherosclerotic plaque tended to be more frequently enhanced in the arteries ipsilateral to the infarcts Frontiers in Neurology | www.frontiersin.org  ( [31][32][33][34]. This indicates that plaque enhancement in intracranial lesions may also serve as a marker of plaque instability, although the pathological validation of the intracranial plaques is generally unattainable. However, while the strong association between plaque enhancement and ischemic events was observed in acute lesions, the chronic plaques, either symptomatic or asymptomatic, rarely showed significant plaque enhancement. A previous study of coronary artery atherosclerosis has demonstrated a significant regression in plaque enhancement during the 3-month post-infarction period. The decrease was paralleled by declines in C-reactive protein, indicating that the contrast-enhanced cardiac magnetic resonance offers the potential to indirectly visualize the transient inflammatory process in coronary artery atherosclerosis by characterizing the changes in plaque enhancement (35). In this study, we also observed a trend toward a decrease in plaque enhancement strength within symptomatic MCA atherosclerosis after the onset of ischemic events, which may further support the potential link between altered plaque enhancement and dynamic inflammatory activity. Limited studies have investigated the changes of plaque enhancement after the onset of ischemic events. A prior MRI study evaluated the presence and strength of enhancement in symptomatic intracranial lesions, demonstrating that all acute plaques showed substantial enhancement while a majority of chronic plaques had mild or no enhancement (22). Another 7 Tesla MRI study followed up stroke or TIA patients in 1 month after symptom onset, in which only 17% of atherosclerotic lesions showed changes in the presence of plaque enhancement or enhancement patterns over time (36). However, 1 month is relatively short as compared to the long natural course of inflammatory activity. In a small study of 7 stroke or TIA patients with symptomatic >70% of stenosis in the M1 segment, Abe et al. observed an enhancement decrease over a 7-month follow-up period in 4 of 7 patients (23), implying that the enhancement patterns may change after a longer period of stroke onset. The mean period of chronic stroke patients in our study was 3 years and 3 months, which may explain why we observed a considerable low enhancement index in chronic lesions. Of note, in Abe's study, two of the three patients who did not show enhancement reduction had a recurrent stroke during the follow-up (23). Plaque enhancement is considered as an imaging marker of future stroke risk (37), we thus speculate that the less enhanced pattern of chronic lesions in our study may be attributable to no stroke recurrence in chronic stroke patients. Additionally, while acute and subacute patients received mono antiplatelet treatment, the chronic group was given dual antiplatelet treatment for 4 weeks before switching to monotherapy, which may also contribute to the less significant plaque enhancement pattern in the chronic group. The mechanisms underlying plaque enhancement is complex and unclear. Some other plaque features or components could also have confounding effects on the association of plaque enhancement with either stroke phase or symptomatic status. Consistent with a prior study (33), we found that degree of luminal stenosis was closely associated with plaque enhancement, implying that the decreased enhancement in chronic lesions may also be explained by plaque volume regression. However, no statistically significant difference in plaque load or area stenosis was found among the acute, subacute and chronic groups, suggesting that luminal stenosis may not account for the plaque regression from the acute to chronic phase in our study. IPH is attributed to inherently immature, fragile neovessels that surround and invade the plaque, and could also contribute to the entry of gadolinium into the plaque (38). However, no association between IPH and plaque enhancement was found in this study.
There are some limitations in our study. First, this is a crosssectional study comparing the strength of plaque enhancement in stroke patients of acute, subacute, and chronic phases, respectively, and follow-ups at various time points were not performed to evaluate the time-course changes in plaque enhancement. The enhancement differences in distinct groups could be due to the selection bias. However, the chronic patients in our study were consecutively recruited, and all the MCA plaque showed a considerably low enhancement index compared to acute patients, implying that plaque enhancement cannot persist for a long period after stroke onset. Second, although the relationship of contrast enhancement with inflammation has been well-established within the extracranial plaques, the exact mechanism accounting for intracranial plaque enhancement remains largely unexplored. Further studies by using molecular contrast agents to target specific inflammatory markers may assist in clarifying the basis of enhancement (39).
In summary, the quantitative assessment of MCA plaque shows an association between plaque enhancement and ischemic stroke, and also reveals temporal regression of plaque enhancement relative to the onset of the ischemic events. If the uptake of contrast in ICAD has the same mechanisms as that in extracranial and coronary vessels, intracranial plaque enhancement may act as a promising imaging biomarker for plaque activity.

DATA AVAILABILITY STATEMENT
The datasets generated for this study are available on request to the corresponding author.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Institutional Review Board of the Chinese University of Hong Kong. The patients/participants provided their written informed consent to participate in this study.

AUTHOR CONTRIBUTIONS
W-JY recruited patients, analyzed MRI data, and drafted the manuscript. JA acquired MRI data and helped to draft the manuscript. YS participated in study coordination and recruited patients. SW acquired the MRI data. K-SW and TL participated in the design and coordination of the study. WC and X-YC conceived the study, participated in its design and coordination, and revised the manuscript. All authors contributed to the article and approved the submitted version.