Asymptomatic carotid artery stenosis (CAS) and occlusion (CAO) disrupt cerebral hemodynamics. There are few studies on the brain network changes and compensation associated with the progression from chronic CAS to CAO. In the current study, our goal is to improve the understanding of the specific abnormalities and compensatory phenomena associated with the functional connection in patients with CAS and CAO. In this prospective study, 27 patients with CAO, 29 patients with CAS, and 15 healthy controls matched for age, sex, education, handedness, and risk factors underwent neuropsychological testing and resting-state functional magnetic resonance (rs-fMRI) imaging simultaneously; graph theoretical analysis of brain networks was performed to determine the relationship between changes in brain network connectivity and the progression from internal CAS to CAO. The global properties of the brain network assortativity (
Asymptomatic carotid artery stenosis (CAS) and carotid artery occlusion (CAO) are characterized by the presence of extracranial internal carotid atherosclerotic stenosis in the ipsilateral carotid perfusion region in individuals without a recent history of ischemic stroke or transient ischemic attack (TIA) [
Functional neuroimaging can reveal brain activity, and thus, it has become an important tool in the study of neurological diseases [
The purpose of this study was to determine the relationship between changes in brain network connectivity and the progression from internal CAS to CAO through graphical theoretical analysis. Therefore, we used resting-state functional magnetic resonance imaging (rs-fMRI) to compare brain network connections in 27 patients with CAO, 29 patients with CAS, and 15 healthy controls (HCs).
This prospective study enrolled 27 patients with CAO and 29 patients with CAS from the Neurosurgery Department of Beijing Tiantan Hospital affiliated to Capital Medical University between March 2019 and December 2019.
The inclusion criteria were as follows: (1) findings of digital subtraction angiography (DSA) or carotid ultrasound examination which were consistent with the diagnostic criteria of CAS and CAO [
The exclusion criteria were as follows: (1) presence of posterior circulation diseases; (2) other causes of carotid stenosis including chronic inflammatory arteritis which were ruled out; (3) presence of other neuropsychiatric diseases and severe systemic diseases (e.g., Alzheimer’s disease, Parkinson’s disease, and history of stroke); (4) the manifestations of any medications that could affect the cognitive function; and (5) any contraindications for MR scan (e.g., metal implants). We also recruited 15 HCs whose age, sex, and education level were similar to those of the patient groups as possible.
Written informed consent was obtained from all participants. This study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Institutional Review Board of Beijing Tiantan Hospital, Capital Medical University (KYSQ2019-058-01).
MRI data were obtained using a 3.0-Tesla MR system (Verio A Tim+Dot System, Siemens, Germany). A standard 12-channel head coil (3T Head MATRIX, A Tim Coil, Siemens) was used for signal reception. Each subject lay supine with the head snugly secured by a belt and foam pads. In rs-fMRI scans, subjects were asked to close their eyes, not to fall asleep, and not to think about anything in particular. The scanning parameters were as follows: repetition time (TR), 2220 ms; echo time (TE), 30.0 ms; voxel size,
rs-fMRI data preprocessing was conducted using SPM 12 (Wellcome Department of Imaging Neuroscience, London, UK;
Graph theory analysis were performed according to the following steps implemented in GRETNA software [
For graph theory analysis, six node-based and three global parameters were obtained for each network. The node-based network parameters included nodal-clustering coefficient (
We used ICA to preprocess data using the Group ICA of fMRI toolbox (GIFT 4.0a,
One-way analysis of variance (ANOVA) was performed to compare continuous variables. Contingency tables of Pearson’s
Clinical variables of the patients with right internal carotid atherosclerosis and those of HCs are shown in Table
Basic characteristics of study participants.
CAS ( | CAO ( | HC ( | ||
---|---|---|---|---|
Age (years) | 0.145 | |||
Male : female | 2.2 | 6.5 | 2.75 | 0.570 |
Education (years) | 0.795 | |||
Risk factors (%) | ||||
Hypertension | 9 (56.3) | 7 (46.7) | 7 (46.7) | 0.871 |
Diabetes mellitus | 4 (25) | 5 (33.3) | 5 (33.3) | 0.851 |
Ischemic heart disease | 2 (12.5) | 1 (6.7) | 2 (13.3) | 1.000 |
Hypercholesterolemia | 6 (37.5) | 7 (46.7) | 4 (26.7) | 0.555 |
Smoking | 7 (43.8) | 8 (53.3) | 6 (40) | 0.812 |
Age and years of education are represented as the median and standard deviation. The risk factors are presented as the number of people and percentage. The chi-square test was used for the analyses.
Among the network attributes of the left CAO and CAS groups, the global attributes were not significantly different. However, the positive results of node indices are mostly consistent with the differences between the left and right hemispheres. In addition, there were only a few significant differences in edge attributes. Therefore, the results of the brain network analysis in the left CAO and CAS groups and their clinical basis variables are provided as supplementary material (Supplementary Table
Figure
Graphical representation of four global attributes.
The 6 node-based network parameters—
Graphical representation of node differences. ROL: Rolandic operculum; FFG: fusiform gyrus; TPOmid: temporal pole, middle temporal gyrus. The 45 nodes were subjected to repeated measurement ANOVA of
Figure
The connectivity maps of three groups (
Map of differences in brain functional connections. In the figure on the left, interactions between brain regions in the left hemisphere are higher in the CAO group than in the HC group (red node), and those in the right hemisphere are lower in the CAO group than in the HC group (blue node). In the figure on the right, interactions between brain regions in the left hemisphere are higher in the CAS group than in the HC group (red node), and those in the right hemisphere are lower in the CAS group than in the HC group (blue node).
In the right cerebral hemisphere of patients with CAO and CAS of the right carotid artery, connections between the Rolandic operculum and supplementary motor area, middle frontal gyrus and insula, supplementary motor area and insula, insula and median cingulate and paracingulate gyri, insula and superior parietal gyrus, lingual gyrus and superior parietal gyrus, and fusiform gyrus and superior parietal gyrus were reduced. However, three connections, that is, connections between the precentral gyrus and insula, inferior frontal gyrus, triangular part with median cingulate and paracingulate gyri, and insula and inferior temporal gyrus, that reduced in the CAO group were more significant than those in the CAS group.
In the left cerebral hemisphere of patients with CAO and CAS of the right carotid artery, connections between the Rolandic operculum and insula, median cingulate and paracingulate gyri and postcentral gyrus, and superior parietal gyrus and precuneus were stronger than those in normal controls. However, two connections, that is, connections between the precentral gyrus and insula and between insula and postcentral gyrus that appeared in the CAS group, were more significant than those that appeared in the CAO group. The connection between the middle frontal gyrus and anterior cingulate and paracingulate gyri is even more significant in the CAO group than in the CAS group.
We show the results of FNC analysis in Figure
Map of differences in functional network connectivity.
This rs-fMRI-based prospective study is the first study to fully elucidate the similarities and differences and compensatory connections in the brain between patients with asymptomatic CAS and CAO. Our findings suggest that changes in brain network connectivity indicators are more sensitive to hemispheric detection. Our results show that the global attributes of the patient’s brain network and the efficiency of nodes in multiple brain regions decreased, while the affected hemisphere lost many key functional connections among patients with CAS and CAO.
Because of the effect of the dominant hemisphere, this study focuses more on the main effects of groups than on the differences between the left and right sides of the brain. We included the results of differences between the brain regions in the supplementary material. (Supplementary Figure
In the analysis of the global network parameters, no significant differences were observed between the controls and patients. Only when data on each hemisphere were processed independently (by using
In patients with affected right carotid artery, there were significant differences in four global attribute indices (Figure
Networks that are cheap to build but still efficient in propagating information are called economic small-world networks. Small-worldness is an attractive model to characterize brain networks because the combination of high local clustering and short-path length supports the two fundamental organizational principles in the brain: functional segregation and functional integration [
We assume that in an organization’s command communication network, if all the commands are one way from top to bottom, the whole network is considered very hierarchical; therefore, the number of two-way connections is small, and the hierarchy is closer to 1 (number of edges actually connected in both directions)/(number of edges that can be connected in both directions, i.e.,
Among the indicators of global attribute, we found several specific indicators of the cerebral ischemia group. In patients with CAS, brain networks are affected very severely, and the indicators related to the speed of network information transmission have declined significantly, while still showing a fluctuating and reconfigurable hierarchy phenomenon. However, in the chronic CAO group, although the unilateral blood flow was completely occluded, the performance of the brain network was significantly better than that of the CAS group because of the existence of collateral circulation compensation, which indicated the presence of a strong plasticity and self-mediation.
Besides, the indices of nodal efficiency were significantly different among the three groups with regard to different brain regions, namely, the Rolandic operculum, fusiform gyrus, and temporal pole (middle temporal gyrus). In terms of anatomical functions, the temporal pole and fusiform gyrus are often used as supplementary language functional areas outside Wernicke’s area. Recent evidence indicates that this region is not critical for speech perception or for word comprehension. Rather, it supports retrieval of phonological information, which is used for speech output and short-term memory tasks [
As Figure
There are several similarities between the results of functional connectivity analysis of the brain network in these two chronic ischemic diseases (Figure
Our results contribute to the understanding of normal brain function networks, explore changes in brain connectivity in asymptomatic patients with chronic ischemic encephalopathy, and identify potential compensation mechanisms for changes in brain hemodynamics.
Our study has some limitations. Brain networks are complex and diverse, and further studies with a large sample size are thus needed to determine possible differences in functional connectivity between patients and controls and to understand the effects of changes in blood circulation in CAS and CAO on the brain network and advanced nervous function. In addition, patients with different handedness and stenosis may show different types of cognitive impairments. Therefore, further studies including more patients with different handedness are needed to confirm our current findings.
In conclusion, a comparison of the differences among the three groups using graph theory analysis showed four indicators of abnormal cerebral network (assortativity, hierarchy, network efficiency, and small-worldness), which occur as a result of the disruption of hemodynamics in the brains of patients with CAS and CAO. Furthermore, the nodal efficiency of key nodes in multiple brain regions of patients with CAS and CAO decreased, while the affected hemisphere lost many key functional connections. In the FNC analysis, the decline of the connection between multiple functional networks was also found. However, partial compensation occurred in the contralateral cerebral hemisphere, which may be the reason for the clinical asymptomatic manifestations. And the increase of functional network connectivity between LECN and DMN and primary visual network reflects the strong plasticity of the brain. The above-mentioned results indicate a correlation between impaired functional connectivity and clinical higher neurological function before the occurrence of real clinical symptoms because of the existing functional connectivity impairment in the local brain regions in the asymptomatic state. Out future studies will be focused on the exploration of methods for predicting the condition, providing interventions in advance, and reducing the impairment of higher brain functions.
All data that support the findings of this study are available upon request from the corresponding author.
The authors declare that they have no competing interests.
Shihao He and Ziqi Liu contributed equally to this work.
This study was supported by the Beijing Municipal Science & Technology Commission (Z151100004015077-DR) and by the Beijing Municipal Health System High-Level Health Technical Personnel Training Program (2015-3-041-DR).
Table 1: basic characteristics of study participants (left). Alterations in regional nodal characteristics. Alterations in functional connectivity. Supplementary Figure A: CAO—the left