Surface-based morphometry of the cerebral cortex in cognitive impairments of varying severity in patients with age-related cerebral small vessel disease

Full Text

BACKGROUND

Age-related cerebral microangiopathy constitutes a disease complex consisting of neuroimaging, morphologic, and associated clinical manifestations resulting from damage to small (≤500 µm diameter) cerebral blood vessels [1]. This disease plays a crucial role in the development of dementia (up to 45% dementia incidence) and in the disability and mortality of patients (up to 20%–25% of all strokes) [2]. Furthermore, our understanding of the mechanisms of the development and progression of cerebral microangiopathy is currently lacking. Therefore, there is a growing interest in neuroimaging, as technical advances and the widespread use of magnetic resonance imaging (MRI) enable the evaluation of the structural and functional aspects of brain changes in cerebral microangiopathy in vivo.

Most focused MRI signs in cerebral microangiopathy include white matter hyperintensities, small recent infarcts and lacunae, microbleeds, and perivascular spaces [3], with little attention being paid to cerebral atrophy. In this vascular disease, cerebral atrophy and thinning are believed to be caused by neuronal death, rarefaction and volume reduction of the white matter on a backdrop of arteriolosclerosis and venous collagenosis with secondary degenerative alterations [4]. Cortical atrophy in cerebral microangiopathy may also result from primary damage during microinfarctions with subsequent Wallerian degeneration and demyelination of axons [5]. It is probable that in most cases of cerebral microangiopathy, both primary and secondary cortical damage processes coexist [6].

Structural MRI analysis plays a vital role in assessing the underlying mechanism of cognitive impairment in cerebral microangiopathy, particularly in dementia. It is subjective to qualitatively assess cerebral atrophy based on cerebrospinal fluid (CSF) space dilation and ventriculo-cranial indices. Conversely, quantitative methods such as volumetry and morphometry provide accurate, operator-independent data. The total brain volume, gray and white matter volumes, and changes in selected regions and structures were measured.

Generalized cerebral atrophy in cerebral microangiopathy can be employed as a marker of disease progression and cognitive decline [7], even in the presence of natural age-related changes [8]. Morphometry is often used to evaluate regional changes and is currently understood to be a neuroimaging modality useful for the quantitative assessment of brain gray matter (with the ability to analyze selected convolutions/nuclei and their segments) using automated techniques for processing high-resolution isotropic T1-weighted images (T1-WI) (standard voxel size is 1 mm × 1 mm × 1 mm; MPR, MP-RAGE, MP2RAGE). The two most common morphometry techniques include voxel-based morphometry [9] and surface-based morphometry [10]. Voxel-based morphometry assesses gray matter volumes (both cortical and subcortical structures), but most voxel-based morphometry programs are also designed to isolate white matter and CSF volumes. Surface-based morphometry identifies and assesses the cortical surface, including the volume, thickness, area, and other quantitative parameters [11].

Based on morphometric data, studies on cerebral microangiopathy have evaluated a comparatively high number of distinct brain regions, the atrophy of which correlates with the degree of cognitive impairment. These regions include the frontal and temporal lobes [7]; occipital lobes and hippocampi; inferior parietal lobes; left precentral gyrus and right inferior frontal gyrus [12]; left superior parietal lobule and left insula [13]; right anterior cingulate gyrus; right cuneus; bilateral insula; and right middle temporal gyrus [14].

Because of the features of the modality, the diversity of MR morphometric results in cerebral microangiopathy requires further research and comparison with clinical data. To distinguish between neurodegenerative and mixed forms of dementia, it appears more promising to identify cortical regions that are most strongly linked to the overall degree of brain damage in cerebral microangiopathy as well as to the cognitive abnormalities characteristic of this condition.

AIM

To employ surface-based morphometry as a more reliable tool for group analysis than voxel-based morphometry in severe atrophy to examine the characteristics of brain atrophy-related cognitive impairment in sporadic age-related cerebral microangiopathy patients [15].

MATERIALS AND METHODS

Study Design

This was an experimental single-center, cross-sectional, randomized, controlled study.

Eligibility Criteria

Inclusion criteria (the main group): age of 46–75 years; MRI alterations consistent with the STRIVE signs for cerebral microangiopathy [3]; cognitive symptoms (decreased memory, concentration, attention, etc.).

Non-inclusion criteria (the main group): pre-existing and/or Alzheimer’s disease-related severe cognitive decline that prevents a patient from participating in the study; other causes (inflammatory, toxic, thrombophilic, systemic, and genetic etiology) of cerebral microangiopathy; history of severe migraine; other etiologies of stroke and concomitant cerebral diseases other than cerebral microangiopathy; atherosclerotic lesions of extracranial or intracranial arteries with stenosis >50%; supratentorial infarcts >15 mm in diameter; severe medical condition; MRI contraindications.

Inclusion criteria (the control group): healthy volunteers free of clinical or neuroimaging evidence of brain disease, severe medical conditions, and MRI contraindications.

Study Setting

The study was conducted at the Federal State Budgetary Scientific Institution Research Center of Neurology (Research Center of Neurology, Moscow) supported by the Department of Radiology and the 3rd Department of Neurology.

Study Duration

The study was carried out between 2016 and 2022.

Intervention Description

The clinical evaluation comprised a review of medical history, major clinical vascular risk factors, physical examination, and an evaluation of the physical, neurological, and cognitive status of the participants. The Montreal Cognitive Assessment (MoCA) questionnaire, including independence in activities of daily living, was utilized to assess the degree of cognitive impairment. Groups of patients with subjective cognitive impairment were identified, including patients with cognitive complaints (MoCA ≥ 26), moderate cognitive impairment (MoCA < 26, independence in activities of daily living), and dementia (MoCA < 26, loss of independence in activities of daily living).

The brain MRI was performed using a Siemens MAGNETOM Verio 3T scanner (Siemens AG, Erlangen, Germany) with an 8-channel head coil. The scanning protocol included T2WI (axial plane, slice thickness: 5 mm), 3D-T2-FLAIR (sagittal view, slice thickness: 1 mm), SWI (axial view, thickness: 1.2 mm), diffusion-weighted imaging (axial view, slice thickness: 5 mm) to evaluate the primary indicators of structural brain damage in cerebral microangiopathy, and 3D T1 gradient echo (T1 MPR) in the sagittal view (thickness: 1 mm) for subsequent morphometry (basic parameters: TR 1900 ms, TE 2.47 ms, 176 slices, field of view: 250 mm × 250 mm, matrix: 256 pixels × 246 pixels).

Using eFilm Workstation 4.2.2 (IBM Watson Health, USA) and the standard brain MRI in accordance with the STRIVE criteria [3], a qualitative evaluation of the MRI indicators of cerebral microangiopathy was carried out in each patient. These data were used to compute an overall cerebral microangiopathy index based on the combination of the following indicators [16]: white matter hyperintensity of Fazekas grade 2 or 3, ≥ 1 microbleeds, ≥ 1 lacunae, and subcortical perivascular spaces ≥ 2 mm in diameter. The total index of cerebral microangiopathy had a four-point scale, with 0 indicating an absence of signs with the specified parameters and 4 indicating the presence of all four signs with the specified parameters.

Volumetric and morphometric data were obtained by preprocessing each participant’s original data (3D-T1-WI) using CAT12 [17] based on the SPM12 software. The initial step was to segment (divide) the original 3D T1-WI into gray matter, white matter, and CSF and extract a surface mesh of the cerebral hemispheres to create brain surface files with the estimated cortical thickness. For each participant, a report was generated that included the volumes of each compartment (separately for white matter hyperintensities) and the intracranial volume. The indicated volumes were subsequently standardized to eliminate the influence of intersex differences (volumetry of major intracranial components). To assess changes in cortical thickness, the resulting surfaces were then co-registered with a FreeSurfer FsAverage template (https://surfer.nmr.mgh.harvard.edu/fswiki/FsAverage) and further smoothed using a 15 mm × 15 mm × 15 mm kernel for group analysis. In addition, an automated ROI analysis was performed when the hemisphere surface files were acquired, calculating the cortical thickness of the selected areas of the hemispheres in accordance with the Desikan-Killiany-40 atlas [18]. The surface-based morphometry results were presented in CAT12 as 3D brain surface reconstructions, with color-coded areas of statistically significant variations in cortical thickness between the groups (FWE analysis at p < 0.05, adjusted for multiple comparisons). The white matter hyperintensity and lateral ventricle volumes were calculated using the white matter and CSF hyperintensity files acquired in the first preprocessing step. After that, the white matter hyperintensity was manually adjusted and the lateral ventricles were selected in ITKSnap.

Ethics approval

The study was approved by the local ethics committee of the Federal State Budgetary Scientific Institution “Research Center of Neurology” (Protocol No. 2-4/16 dated February 17, 2016). Informed consent was obtained from all participants.

Statistical Analysis

The volumetric parameters and regions of interest (ROI) were statistically calculated using SPSS Statistics 26.0 software (IBM). Two-sided statistical tests were applied. The null hypothesis was rejected at p < 0.05. The Shapiro–Wilk and Kolmogorov–Smirnov tests were used to evaluate the hypothesis that the characteristics had a normal distribution. Frequency histograms were also constructed and analyzed. Parametric methods were preferred due to the sample size. Quantitative parameters with a normal distribution were compared using the Student’s t-test or one-way analysis of variance. Within-group analysis was performed using the Bonferroni correction. The Mann–Whitney test or the Kruskal–Wallis test was used to compare quantitative parameters with a non-normal distribution. The Mann–Whitney test was then used for an intragroup analysis. The predictive value of some parameters for the development of the expected outcomes was assessed using binary logistic regression, and receiver operator characteristic (ROC) analysis was then performed. The area under the curve (AUC), optimal threshold, and its sensitivity and specificity were also determined.

RESULTS

Participant Characteristics

The main group comprised 173 patients with cerebral microangiopathy (mean age: 60.5 ± 7.5 years; 54% female) and cognitive impairment of varying severity: there were 54 patients with subjective cognitive impairment, 78 with moderate cognitive impairment, and 41 with dementia. The control group included 47 healthy volunteers (mean age: 56.8 ± 6.0 years; 66% women).

There was no significant difference between the groups in terms of sex (p = 0.07) and age (p = 0.06). Women predominated in both groups (66% and 54%, respectively). Regarding white matter hyperintensity among main group patients, Fazekas grade 1 lesions were documented in 19 patients, Fazekas grade 2 lesions in 45, and Fazekas grade 3 lesions in 109.

Primary Findings

Volumetry of intracranial brain lesions in the cerebral microangiopathy and control groups

Comparative analysis of the primary intracranial volumetric parameters for the cerebral microangiopathy and control groups revealed significant differences in all parameters (p < 0.05) with a decreased volume of the brain and its compartments (gray and white matter) in the cerebral microangiopathy group (p < 0.0001) and vicarious enlargement of the CSF spaces (p < 0.0001) (Table 1).

 

Table 1. Comparative analysis of the major intracranial volumetric parameters among the study groups, Me [Q25%; Q75%]

Parameter	Cerebral microangiopathy n=173	Control n=47	p
White matter hyperintensities, cm3	30.293 [12.07; 52.16]	–	–
Brain volume / ICV	0.73 [0.69; 0.75]	0.79 [0.77; 0.81]	<0.0001
Gray matter / ICV	0.41 [0.38; 0.43]	0.44 [0.43; 0.45]	<0.0001
White matter / ICV	0.32 [0.29; 0.35]	0.35 [0.34; 0.37]	<0.0001
Cerebrospinal fluid / ICV	0.28 [0.25; 0.31]	0.2 [0.18; 0.22]	<0.0001
Volume of lateral ventricles, cm3	29.26 [20.05; 40.8]	13.78 [10.83; 16.86]	<0.0001

Note. ICV, intracranial volume.

 

The main intracranial volumetric parameters were compared between the groups of patients with different levels of cognitive impairment. The dementia and subjective cognitive impairment groups showed substantial (p < 0.0001) differences in CSF parameters, with higher values in the dementia group. The brain volume (total volume, gray matter, and white matter volumes) in the cerebral microangiopathy group differed significantly only from the control group, but no differences were reported between the groups of patients with varying degrees of cognitive impairment. There was a significant difference in the white matter hyperintensity volume between the dementia and moderate cognitive impairment groups and the dementia and subjective cognitive impairment groups (p < 0.0001).

Cerebral cortex thickness and severity of structural brain damage in patients with cerebral microangiopathy

When groups with various white matter hyperintensities were compared pairwise, no significant (FWE, p_corr < 0.05) variations in cortical thickness were found between patients in the Fazekas grade 1 group and the control group. Compared to the control group, the most severe lesions were noted in the central anterior and central posterior regions of the right cingulate gyrus in the Fazekas grade 2 group patients (Fig. 1). The greatest difference in cortical thickness between the Fazekas grade 2 and Fazekas grade 3 groups was observed in the middle frontal gyri and temporoparietal regions, as well as in the posterior cingulate gyri and medial frontal lobes. When compared to the control group, the progression of white matter hyperintensities to Fazekas grade 3 was marked by lesions in extended cortical regions (Fig. 1). The most severe lesions were observed in the same regions that were affected in the Fazekas grade 2 and Fazekas grade 3 groups, i.e., in the middle frontal gyri and temporoparietal regions, as well as in the posterior parts of the cingulate gyri and medial parts of the frontal lobes.

 

Fig. 1. Visual representation of surface-based morphometry results following statistical analysis of images in the control and Fazekas grade 2 (first row, F2), control and Fazekas grade 3 (middle row, F3), and Fazekas grade 2 and Fazekas grade 3 (bottom row) groups utilizing CAT12. Regions with significantly major differences in cortical thickness between groups (FWE, p_corr < 0.05) are highlighted in red and yellow (according to the scale). L, the left hemisphere; R, the right hemisphere.

 

Binary logistic regression was used to assess digital measurements of the thickness of specific regions of the cerebral cortex (ROI analysis) to determine those that were most consistent with the overall burden of cerebral microangiopathy, with a score of 4 interpreted as 1 and all other grades as 0 (Table 2). A decrease in the cortex of certain frontal lobe regions, the cuneus, the right isthmus of the cingulate gyrus, and the left paracentral lobule most closely correlated with the overall score of cerebral microangiopathy.

 

Table 2. Cortical regions with thinning exhibited strongest correlation with the total cerebral microangiopathy score (binary logistic regression)

Cortical regions	В	р
Posterior parts of the middle frontal gyrus, L	-5.501	0.015
Cuneus, R	8.201	0.000
Isthmus of the cingulate gyrus, R	-3.079	0.011
Paracentral lobule, L	-2.874	0.027
Inferior frontal gyrus, pars opercularis, R	-11.268	0.000
Inferior frontal gyrus, pars orbitalis, R	6.001	0.004

Note. B, the coefficient to multiply the predictor parameter for estimating the linear exponential function in assessing the probability of severe brain damage in cerebral microangiopathy; R, the right hemisphere; L, the left hemisphere.

 

Cerebral cortex thickness by cognitive impairment degree in cerebral microangiopathy

When differences in the thickness of selected cortical regions were assessed between the control and cognitive impairment groups, the pattern of differences for the control versus moderate cognitive impairment and control versus dementia group comparisons predominantly included the medial frontal cortex, posterior cingulate gyri, middle frontal gyri, and temporoparietal areas. Significant differences in the control versus subjective cognitive impairment group comparison involved the pre- and postcentral gyri, the cingulate gyrus, and selected frontal regions (p < 0.05) (Fig. 2).

 

Fig. 2. Surface-based morphometry results: a, comparative cortical thickness in the control, subjective cognitive impairment (subCI), and dementia groups; b, in the cerebral microangiopathy group, between moderate cognitive impairment (MCI) and dementia, subjective (subCI) and moderate (MCI) cognitive impairment. Regions with significantly greater cortical thickness in the control (a), moderate cognitive impairment versus dementia (b, top), and subjective cognitive impairment versus moderate cognitive impairment (b, bottom) groups are highlighted in yellow and red, respectively (FWE, p_corr < 0.05). L, the left hemisphere; R, the right hemisphere.

 

In patients with cerebral microangiopathy with different levels of cognitive impairment, intragroup comparisons of metrics revealed more regions with a significant difference in cortical thickness across all groups. The most significant areas that varied between the groups included the posterior cingulate cortex and left supramarginal gyrus for the dementia versus moderate cognitive impairment group comparisons (Fig. 2), the supramarginal gyrus for the moderate cognitive impairment versus subjective cognitive impairment group comparisons, and the posterior right cingulate gyrus for the subjective cognitive impairment and control group comparisons. Differences in cortical thickness between the dementia and subjective cognitive impairment group patients (p < 0.0001) included a significant number of lesions in both hemispheres (superior, middle and inferior frontal gyri, cuneus, precuneus, inferior and superior parietal lobules, posterior cingulate gyrus, parahippocampus, middle temporal gyri, paracentral lobules, supramarginal gyri, insula; right medial orbitofrontal cortex, left postcentral gyrus).

Cortical areas where atrophy was the most significant for the degree of cognitive impairment in cerebral microangiopathy were identified through ROI analysis employing binary logistic regression, where the onset of dementia was taken as 1. This was followed by ROC analysis. The left isthmus of the cingulate gyrus (AUC 0.826, confidence interval (CI): 0.8–0.9; threshold: 2.23 mm; sensitivity 73%, specificity 85%) and the left supramarginal gyrus (AUC 0.778, CI: 0.7–0.9; threshold: 2.26 mm; sensitivity 77%, specificity 73%) displayed the best characteristics, consistent with the high predictive value of cortical thickness in these regions in predicting dementia in cerebral microangiopathy.

DISCUSSION

Summary of the Primary Study Results

Our study used surface-based morphometry to assess the cerebral cortex thickness in age-related cerebral microangiopathy patients with varying degrees of cognitive impairment. As the severity of the structural brain lesions and cognitive impairment increases, the cortical thickness of the selected regions decreases in a similar pattern (cingulate gyri, especially their posterior parts; medial and middle frontal lobes; various areas of the insular cortex; temporoparietal regions, especially supramarginal gyri).

Discussion of the Primary Study Results

Our study discovered that the cortical atrophy pattern in cerebral microangiopathy was consistent with that described in earlier morphometric studies [12–14, 19]. Smith et al. [19] revealed that cerebral microangiopathy affects the superior and inferior frontal gyri, posterior parts of the superior and middle temporal gyri, supramarginal gyri, and inferior parietal lobules in its early stages, which worsens with age. In our study, all these regions exhibited significant differences in the subgroups of the main group, while the potential impact of age-related changes was balanced by comparing subgroups with varying degrees of cognitive impairment at comparable ages.

Lesions in chosen locations were not specific to cerebral microangiopathy and are observed in other age-related disorders, despite the high sensitivity and specificity of the ROC analysis performed. However, the combination of lesions in these regions correlates well with our measurements using other MRI techniques and with clinical manifestations, indicating the pattern of structural and functional brain damage in cerebral microangiopathy [20, 21]. A decrease in the cortical thickness of the middle and posterior cingulate gyri with the advancement of cognitive impairment in cerebral microangiopathy has also been documented using voxel-based MR morphometry evaluating the cortical volumes of the corresponding regions [20]. Furthermore, diffusion MRI assessing indicators of white matter microstructural lesions in different models also revealed the most severe lesions in the right middle cingulate gyrus [21]. The cingulate gyrus actively regulates memory, emotional behavior, and executive brain functions [22]. Changes in cortical thickness and regional homogeneity on resting-state functional MRI (ReHo) are associated with memory decline [14]. In addition to the medial frontal lobes, which also showed significant changes with the progression of cerebral microangiopathy and its clinical manifestations, the cingulate gyri were discovered to be involved in the executive brain functions affected by cerebral microangiopathy [23]. In functional neuroradiology and neuropsychology, the anterior parts of the insula and cingulate gyrus form a functional salience network linked to the evaluation of novel and significant stimuli. An executive control network (frontoparietal network) made up of the components of the frontal lobes and the inferior parietal cortex ensures dynamic regulation, attentional switching to important stimuli, and decision-making based on objectives and anticipated outcomes. As in our study, the described networks and their components are combined into a multiple-demand network affected by cerebral microangiopathy [23], which is reflected in cognitive function alterations and is further supported by functional MRI [24] and structural MR morphometry data. Slower information transfer is believed to be linked to changes in the frontal and parietal lobes, which impacts the neuropsychological profile of patients with cerebral microangiopathy [25].

The previously described association between cortical thinning in certain regions and MRI micro- and macrostructural white matter lesions in the early stages of the disease [21] suggests that brain atrophy in cerebral microangiopathy is primarily secondary or mixed in origin. This is supported by quantitative measurements, indicating cerebral microangiopathy was associated with a significant decrease in brain, gray matter, and white matter volumes when compared to the control group, with increased CSF and lateral ventricle volumes. Other studies have found a correlation between the severity of cerebral microangiopathy (assessed mainly by subcortical infarcts) and generalized brain atrophy, as well as atrophy of some brain regions and corresponding involvement of adjacent internal and external cerebrospinal fluid spaces (corpus callosum, basal ganglia, midbrain, hippocampi), as well as local cortical thinning linked to areas where the lacunae were located [26]. Independent of white matter hyperintensities, medial temporal atrophy, or generalized hemisphere atrophy, the population-based LADIS study [27] identified the role of generalized brain atrophy and the atrophy of the cortex, subcortical structures, and corpus callosum in cognitive impairment.

Our results revealed that despite significant volumetric differences, patients in the cerebral microangiopathy group exhibited more severe cognitive impairment than the control group, which was not accompanied by a significant decline in volumetric parameters. Additionally, the CSF volume increased in the cerebral microangiopathy group relative to the control group. The median CSF/intracranial volume ratios for the dementia and subjective cognitive impairment groups were 0.31 and 0.27, respectively, indicating a substantial increase in CSF volume in the dementia group only. The fact that there is no decrease in white matter volume in cerebral microangiopathy despite an increase in white matter hyperintensities can be explained by the simultaneous occurrence of two processes: demyelination and myelin loss on the one hand, and neuroinflammation and edema on the other [28]. Cortical atrophy in vascular disease is still viewed as more of a secondary process due to the significant volume of white matter damage or cortical microinfarcts [6]. Consequently, the reduction of gray matter and the vicarious increase of the external CSF spaces are less severe than in classical neurodegenerative processes [5]. Cerebral microangiopathy is characterized by the prevalence of ventricular enlargement over the external CSF spaces. The volume of the lateral ventricles in dementia patients was three times greater than in those without cognitive impairment and 1.5 times greater than in those with moderate cognitive impairment (median volume is 14 cm³ in normal condition, 29 cm³ in moderate cognitive impairment, and 42 cm³ in dementia) [29]. Such enlargement can result from both primary and secondary atrophy of the subcortical structures due to multiple lacunae and microbleeds [3], as well as their vicarious enlargement from damage to the periventricular white matter.

Our data can be utilized to monitor the process, assess treatment efficacy, or make a differential diagnosis of various diseases, especially neurodegenerative diseases [30]. The most evident advantages of surface-based morphometry include process automation and standardization, speedy processing and analysis of large amounts of data, and the ability to evaluate specific gray matter areas based on the atlases. Most researchers favor this technique for assessing age-related cognitive impairment and cerebral microangiopathy [14, 19, 30] because of its higher sensitivity compared to voxel-based morphometry.

Study Limitations

As with any automated method of MRI data analysis, surface-based morphometry has limitations relating to the structure of the central gyri, which influences the results of surface-based morphometry for these regions [17]. During follow-up, areas of the entorhinal cortex, medial orbitofrontal cortex, lingual gyri, and rostral portions of middle frontal gyri were the most challenging regions in terms of reproducibility of repeat scans [31]. Results and their reproducibility are also impacted by the magnetic field intensity, MRI type, and parameters (e.g., parallel scanning decreases the reliability of cortical thickness measurements in the parietal lobes and cingulate gyri [31]). Given the multiple preprocessing steps and large amounts of data, statistical analysis is usually subject to false positive and false negative results; therefore, many researchers advise employing multiple morphometric techniques and critically interpreting the data [11].

CONCLUSION

Our data supports the secondary/mixed origin of cerebral microangiopathy-related brain atrophy. The wide range of regions beyond the multitask network structures involved in cerebral microangiopathy limits the ability to explain the progression of cognitive impairment by their atrophy. This allows the quantitative measurement of the cortex only as an auxiliary method for determining the prognosis of cerebral microangiopathy.

ADDITIONAL INFORMATION

Funding source. This work was supported by the Russian Science Foundation grant № 22-15-00183 (https://rscf.ru/project/22-15-00183/.

Competing interests. The authors declare that they have no competing interests.

Authors’ contribution. All authors made a substantial contribution to the conception of the work, acquisition, analysis, interpretation of data for the work, drafting and revising the work, final approval of the version to be published and agree to be accountable for all aspects of the work. E.I. Kremneva ― processing of morphometric data, analysis and interpretation of MRI data, writing and preparation of the article; L.A. Dobrynina ― general concept of the study, supervision of patients, discussion of results, writing the article; K.V. Shamtieva ― clinical assessment of patients, statistical data processing, preparation of the article; V.V. Trubitsyna ― conducting MRI studies, collecting and preparing MRI data for analysis; Z.Sh. Gadzhieva ― clinical assessment of patients, research methodology, discussion of results; A.G. Makarova ― clinical assessment of patients, control of MRI and data collection, M.M. Tsypushtanova ― clinical assessment of patients, control of MRI and data collection; M.V. Krotenkova ― development of MRI research methodology, supervising the collection of MRI data, discussion and analysis of the results, writing the article.

About the authors

Elena I. Kremneva

Research Center of Neurology

Author for correspondence.
Email: kremneva@neurology.ru
ORCID iD: 0000-0001-9396-6063
SPIN-code: 8799-8092

MD, Dr. Sci. (Medicine)

Russian Federation, Moscow

Larisa A. Dobrynina

Research Center of Neurology

Email: dobrla@mail.ru
ORCID iD: 0000-0001-9929-2725
SPIN-code: 2824-8750

MD, Dr. Sci. (Medicine), Assistant Professor

Russian Federation, Moscow

Kamila V. Shamtieva

Research Center of Neurology

Email: kamila.shamt@gmail.com
ORCID iD: 0000-0002-6995-1352
SPIN-code: 5645-8768

MD, Cand. Sci. (Medicine)

Russian Federation, Moscow

Victoria V. Trubitsyna

Research Center of Neurology

Email: pobeda-1994@mail.ru
ORCID iD: 0000-0001-7898-6541
Russian Federation, Moscow

Zukhra S. Gadzhieva

Research Center of Neurology

Email: zuhradoc@mail.ru
ORCID iD: 0000-0001-7498-4063
SPIN-code: 7015-5970

MD, Cand. Sci. (Medicine)

Russian Federation, Moscow

Angelina G. Makarova

Research Center of Neurology

Email: angelinagm@mail.ru
ORCID iD: 0000-0001-8862-654X

MD, Cand. Sci. (Medicine)

Russian Federation, Moscow

Maria M. Tsypushtanova

Research Center of Neurology

Email: tzipushtanova@mail.ru
ORCID iD: 0000-0002-4231-3895

MD, Cand. Sci. (Medicine)

Russian Federation, Moscow

Marina V. Krotenkova

Research Center of Neurology

Email: krotenkova_mrt@mail.ru
ORCID iD: 0000-0003-3820-4554
SPIN-code: 9663-8828

MD, Dr. Sci. (Medicine), Assistant Professor

Russian Federation, Moscow

References

Supplementary files