Diagnostic value of HR‐MRI and DCE‐MRI in unilateral middle cerebral artery inflammatory stenosis

Abstract Purpose High‐resolution magnetic resonance imaging (HR‐MRI) has high spatial resolution and can simultaneously perform wall and lumen imaging. Dynamic contrast‐enhanced magnetic resonance imaging (DCE‐MRI) can evaluate the integrity of the blood–brain barrier. In this paper, the result of 3.0T HR‐MRI and 3.0T DCE‐MRI has been evaluated to explore the application value of unilateral middle cerebral artery inflammatory stenosis and changes in vascular permeability parameters of stroke events. Methods Thirty‐six cases of neurological suspicion of central nervous system vasculitis of our hospital were selected from 20 January 2018 to 1 January 2019, who were diagnosed as unilateral middle cerebral artery M1 stenosis/occlusion by 3D TOF MRA. 3.0T HR‐MRI and 3.0T DCE‐MRI has been applied. Results Among the 36 patients who met the inclusion criteria, 23 patients with central nervous system vasculitis were diagnosed. The 23 patients with HR‐MRI showed diffuse thickening and enhanced stenosis. The K trans value of 10/23 patients with acute–subacute cerebral infarction and 3/23 patients in chronic phase were significantly higher than that of the mirror side, and the K trans value of these patients remeasured in the same region of interest is lower than before after 6 months treatment. The K trans value in the target area of 10 patients without cerebrovascular events was not statistically significant compared with the mirror side. The K trans value of patients with acute–subacute cerebral infarction was significantly higher than that without cerebrovascular events (0.098 ± 0.038 vs. 0.007 ± 0.001, p = .000), and there was no significant difference between K trans in the chronic infarction group and the other two groups (0.098 ± 0.038 vs. 0.044 ± 0.012, p = .058; 0.044 ± 0.012 vs. 0.007 ± 0.001, p = .057). Conclusion HR‐MRI is an accurate direct imaging method and has a high value for the etiological diagnosis of central nervous system vasculitis. DCE‐MRI could be an effective way to evaluate and monitor blood–brain barrier to prevent clinical ischemic stroke.

important in differential diagnosis and pathophysiology of cerebrovascular diseases.
The integrity of the collateral circulation is closely related to the occurrence of ischemic stroke, hemorrhage after thrombolysis, early neurological recovery, and long-term prognosis. The blood-brain barrier (BBB) function can be measured to reflect the establishment of the collateral circulation. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) can evaluate the integrity of the BBB by quantitatively calculating vascular permeability parameters which could effectively help the clinical evaluation of the ischemic stroke event, determine the treatment plan, and understand the treatment effect. The core parameter of DCE-MRI is the transfer constant of the contrast agent from the intravascular to the extracellular space (Volume Transfer Constant, K trans ), and the unit is min-1, which can quantitatively reflect the permeability of tiny blood vessels. By tracking the vascular damage dynamic change, DCE-MRI obtains image information, and processes, and analyzes it through corresponding software to obtain various parameters reflecting the microcirculation of the tissue (Heye, Culling, Valdés Hernández, Thrippleton, & Wardlaw, 2014).
To sum up, the purpose of this study is to apply HR-MRI to make accurate etiological diagnosis of intracranial arterial stenosis caused by central nervous system vasculitis, so as to guide clinical treatment. At the same time, DCE-MRI was used to quantitatively analyze the cerebrovascular diseases caused by vasculitis, so as to provide more information for the evaluation of the curative effect and prognosis of patients.

| Patients
Thirty-six cases of neurological suspicion of central nervous system vasculitis were selected from 20 January 2018 to 1 January

| Data acquisitions
MRI was performed on a 3.0 Telsa magnetic resonance scanner (Philips Achieva) using an eight-channel phased array head coil. All patients underwent routine DWI and 3D-TOF MRA scans. According to the 3D-TOF MRA MIP image, HR-MRI T1WI plain sequences were performed parallelly and perpendicularly to the long axis of the stenotic/occlusion vessel; then, DCE-MRI T1-Mapping was performed.
After the end of the 45-stage dynamic contract enhanced scanning, the HR-MRI T1WI enhanced scanning was applied. The specific scan parameters of the above sequence are as follows:

| Data analyze
Apparent diffusion coefficient (ADC) image and magnetic resonance angiography-maximum intensity projection (MRA-MIP) image were obtain from Philips Extended MR WorkSpace 2.6 work station.
Measurement and calculation of vascular remodeling were performed on the HR-MRI T1WI axial image. To find the maximal lumen narrowing (MLN), the image has enlarged to 400%. The cross-sectional area of the middle cerebral artery at the target and the control was measured and compared, and the normal vascular lumen near the site was used as a control. The blood-outer membrane boundary in the middle cerebral artery was manually mapped as the vascular area (VA), and the blood-intimal boundary in the middle cerebral artery was used as the lumen area (LA). The value of the wall area of the middle cerebral artery (Wall Area, WA) is the difference between the area of the blood vessel and the lumen area. Lumen stenosis is the ratio of LA at the target to the LA at the control (Ma et al., 2010). While the ratio is less than 50% as mild stenosis; while the ratio is 50% to 70% as moderate stenosis; while the ratio is larger than 70% as severe stenosis and lager than 90% as occlusion. HR-MRI can be used to accurately assess the stenosis/occlusion range of MCA: focal stenosis/occlusion refers to lesion range ≤ 1/3 M1 segment, segmental stenosis/ occlusion refers to lesion range from 1/3 to 2/3 M1 segment, and fulllength stenosis/occlusion refers to a lesion range ≥2/3 M1 segment.
The image and data processing were using Omini Kinectics (GE Healthcare) software. To complete the conversion of time-luminance signal to the time-contrast concentration signal, the original data of 2°, 5°, 10°, 15°, 20°, 25°, and 30° flip angles are, respectively, loaded into the T1-Mapping calculation. The region of interest (ROI) is manually outlined in the superior sagittal sinus. The ROI is placed at the center of the superior sagittal sinus cross section and must not extend beyond the superior sagittal sinus. The time-concentration curve of the arterial input function (AIF) was obtained by calculation.
The extend Tofts linear model was selected as the kinetic model. The model is formalized as follows: where C t (t) represents the tissue concentration, V P represents the plasma volume fraction, C P (t) represents the plasma concentration, V e represents the extravascular extracellular volume fraction, and the core parameter is K trans (volume transfer constant). A region of interest (ROI) with the same size was selected as the research object in the middle cerebral artery blood supply area and the contralateral mirror position of the lesion side. The corresponding parameters were obtained by Omini Kinectics software.
Besides, we grouped the cases according to the different luminal stenosis degree and extent and made a statistical analysis of the relationship between cerebral infarction and the two factors, respectively. The comparisons of the different stages of cerebral infarction and the K trans value were also performed.

| Statistical analysis
Statistical analysis of the data was obtained by using the social science (SPSS) version 21.0 for Windows. When measurement data conform to the normal distribution, it was expressed by the mean plus or minus standard deviation and the t test was applied. When the measurement data do not satisfy the normal distribution, the median (interquartile range) was applied, and nonparametric two independent samples Wilcoxon's rank sum test and chi-square test were performed simultaneously. Levene variance analysis and two independent sample t test were performed on K trans value. Spearman correlation analysis was used for association between clinicopathological characteristics and factors. The data are expressed in terms of frequency and percentage. Fisher's exact test is used for comparison between the two groups since the total sample size is <40 cases.
If the p value is <.05, it is considered statistically significant.

| RE SULTS
In total, thirty-six patients were enrolled: twenty-three patients (16 male patients, mean age 38.7 years; seven female patients, mean age 35.3 years; mean age 37.7 years, age range 16-58 years, median age The distribution is shown in Table 2. The relationship between the difference of degree and extent of vascular stenosis and the occurrence of cerebral infarction was analyzed. The results were not statistically significant (p = .253).
For the 10 patients who were in the acute-subacute phase of cerebral infarction, the damage to the blood-brain barrier was significant. Also, the K trans value (0.098 ± 0.038) was significantly higher than the mirror side (0.003 ± 0.001) which considered as statistically significant. And the K trans value (0.039 ± 0.014) was reduced in the same region of interest six months after treatment. For the three patients who were in the phase of chronic cerebral infarction, the K trans value (0.044 ± 0.012) was higher than that of the mirror side.
Two of the three patients were reviewed six months after treatment, and the K trans value (0.018 ± 0.009) was lower than before in the same region of interest. For the 10 patients who were without cerebrovascular events, the K trans value (0.007 ± 0.001) was not statistically significant compared with the mirror side. Four of the 10 patients were reviewed six months after treatment, and the K trans value (0.006 ± 0.001) has no significant change compared with the previous one in the same region of interest. Two of the 10 patients had a significantly higher K trans value (0.066 ± 0.019) in the other region of interest than the mirror side. Statistical analysis of the K trans value between the three groups showed that the K trans value of the acute-subacute infarction group was higher than that of the noninfarction group, and there was no statistical difference between the any other two groups (acute-subacute infarction group vs.
The representative images were shown in Figures 6 and 7.

| The results of HR-MRI
The HR-MRI finds 23 patients with CNS vasculitis were thickened and strengthened with different degrees of wall obstruction in the MCA-M1 segment in our study. They were reconstructed vertically and parallel to the direction of the responsible vessel, and the vessel wall was ring-shaped, with the wall thickening and strengthening more uniform, there is no obvious difference between the front and rear walls and the upper and lower walls; and there are many small branches at the distal end, and some branches can be seen with diffuse mild reinforcement. Li Mingli and other studies of Peking Union Medical College Hospital found that the wall of CNS vasculitis showed diffuse annular thickening with obvious and uniform enhancement, and the wall and lumen structure of the entire affected vessel were narrowed (Li, Xu, Feng, & Jin, 2012); some researchers speculated that this is related to the compression of the vascular lumen by the lining of the lining of the wall after the inflammatory reaction occurs (Gomes, 2010). After the treatment with hormones and/or immunosuppressive agents, the inflammatory reaction is weakened or disappeared, and the wall compression and the stenosis are alleviated or eliminated. This is also confirmed by the inflammatory enhancement of the responsible vessel wall eliminated after treatment with hormones and/or immunosuppressive agents in our study.
It was reported that CNS vasculitis often presents as a single lesion with multiple vessel involvement, and multiple lesions involving a single artery are relatively rare (Hajj-Ali & Calabrese, 2014). In TA B L E 2 Distribution of stenosis (the degree and range) and cerebrovascular events our cases, most of them were involved in single vessel involvement, which may be related to the chosen of standard unilateral cerebral artery M1 stenosis. In this study, 10 patients were included in the acute/subacute phase of cerebral infarction, three patients with chronic cerebral infarction, and 10 patients without cerebral infarction. The relationship between the extent of stenosis and the degree of stenosis and cerebral infarction was not statistically significant (p = .253). There is no clear correlation between the extent and extent of simple stenosis and whether acute ischemic stroke occurs.
The occurrence of cerebrovascular accident is the result of multiple factors.

| The results of DCE-MRI
In this study, the vascular permeability parameters of the cerebral arteries in the affected side and the mirror side of the selected subjects were measured. The K trans value (0.098 ± 0.038) of the region of interest in the acute-subacute phase of cerebral infarction was higher than that of the mirror side (0.003 ± 0.001) is significantly elevated; the literature reports that when the acute cerebrovascular accident occurs, the blood-brain barrier is destroyed by a series of complex mechanisms, the contrast agent infiltrates into the intercellular space, and DCE-MRI captures the microcirculation. The change was characterized by an increase in K trans value; 10 patients were reviewed after 6 months, and the K trans value was reduced in the same region of interest (0.098 ± 0.038 vs. 0.039 ± 0.014, p = .016). The K trans value (0.044 ± 0.012) in the three patients (100%) in the chronic phase of cerebral infarction was higher than that in the mirror side; after 6 months, two patients were reviewed and the K trans value was measured in the same region of interest (0.018 ± 0.009) lower than before. The K trans value (0.007 ± 0.001) of the 10 patients (100%) without cerebrovascular events was not statistically significant compared with the mirror side; four patients were reviewed after 6 months, and the K trans value (0.006 ± 0.001) was measured in the same region of interest. No significant change compared with the previous one. This study found the K trans value of the lesion side of the patient gradually decreased over time, suggesting the repair of the blood-brain barrier; however, the K trans value was still higher than the normal level after re-examination of patients with ischemic stroke. There have reports indicate the self-repair of the blood-brain barrier is a long process after the cerebrovascular event (Obermeier, Daneman, & Ransohoff, 2013), In this study, it was found that two patients (12.5%) had a higher K trans value (0.066 ± 0.019) in the other region of interest than the mirror side after review. However, the DWI sequence did not find abnormal changes in brain parenchyma in this area; it has been reported in the literature that a significant increase in K trans value suggests the potential for cerebrovascular accidents in this area (Merali, Huang, Mikulis, Silver, & Kassner, 2017). Therefore, follow-up observation of these two patients is needed. We compared the K trans value of the patients in the acute-subacute infarction group with chronic infarction group and without cerebrovascular events group and found that the acute-subacute infarction would lead to significant destruction of the blood-brain barrier, while the K trans value of the chronic infarction group was not statistically significant compared with the other two groups, which may be related to the fact that there were fewer cases included in the chronic infarction group.
The monitoring of K trans value by DCE-MRI can microscopically reflect the process of repair and change of blood-brain barrier after acute cerebrovascular events, thus effectively guiding clinical treatment and prognosis, and predicting cerebrovascular accident.

| CON CLUS ION
HR-MRI can provide important diagnostic information for central vasculitis and can accurately assess the degree and extent of vascular stenosis. DCE-MRI can quantitatively measure the permeability of the blood-brain barrier, so as to reflect the situation of the distal collateral circulation of vascular stenosis, which provides important information for the evaluation of the treatment and prognosis of cerebral infarction, and can predict the possibility of cerebrovascular events to a certain extent.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

E TH I C S A PPROVA L A N D CO N S E NT TO PA RTI CI PATE
The study protocol was approved by the Ethics Committee of The Second Hospital of Hebei Medical University. Written informed consent was obtained from all the study subjects before enrollment.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.10.1002/brb3.1732.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.