Altered dynamic regional homogeneity and neurovascular coupling in multi-frequency bands in patients with cerebral small vascular disease
Original Article

Altered dynamic regional homogeneity and neurovascular coupling in multi-frequency bands in patients with cerebral small vascular disease

Jing Huang#, Wenwen Wang#, Runtian Cheng, Xiaoshuang Liu, Tianyou Luo

Department of Radiology, the First Affiliated Hospital of Chongqing Medical University, Chongqing, China

Contributions: (I) Conception and design: T Luo; (II) Administrative support: T Luo; (III) Provision of study materials or patients: T Luo; (IV) Collection and assembly of data: J Huang, W Wang, R Cheng, X Liu; (V) Data analysis and interpretation: J Huang, W Wang; (VI) Manuscript writing: All authors; (VII) Final approval of manuscript: All authors.

#These authors contributed equally to this work as co-first authors.

Correspondence to: Tianyou Luo, MD. Department of Radiology, the First Affiliated Hospital of Chongqing Medical University, No.1 Youyi Road, Yuanjiagang, Yuzhong District, Chongqing 400042, China. Email: ltychys@163.com.

Background: Cerebral small vessel disease (CSVD) can lead to neurologic deficits and cerebral perfusion injuries. However, the dynamic characteristics of neural function homogeneity and their coupling with cerebral perfusion are rarely elucidated in the sub-frequency bands. The purpose of this study is to explore these features for further understanding of the neural mechanisms of CSVD.

Methods: In this cross-sectional study, we obtained magnetic resonance imaging (MRI) data and a series of cognitive scores from 27 CSVD patients with mild cognitive impairment (CSVD-MCI), 32 CSVD patients with cognitive unimpaired (CSVD-CU) and 30 normal controls (NC). The dynamic regional homogeneity (dReHo) and cerebral blood flow (CBF) analyses were utilized to explore the dynamic neural function homogeneity and neurovascular coupling (dReHo-CBF correlation coefficient and CBF/dReHo ratio) in conventional frequency band (0.01−0.08 Hz) and two sub-frequency bands (slow-4: 0.027–0.073 Hz and slow-5: 0.01–0.027 Hz). The partial correlation analysis was used to investigate the relationships between the above MRI indicators and the cognitive scores in the patient group.

Results: CSVD-MCI group showed significantly abnormal dReHo values and CBF/dReHo ratios in multiple brain regions in conventional frequency band, including the left middle occipital gyrus, right inferior temporal gyrus, left anterior cingulate gyrus, and right thalamus (P<0.05). The brain regions with significantly abnormal dReHo and CBF/dReHo ratios in the sub-frequency bands were similar to those in the conventional frequency band. These abnormal values of the right inferior occipital gyrus, right inferior temporal gyrus, and left anterior cingulate gyrus in sub-frequency bands were significantly related to cognitive scores (P<0.05). Moreover, CSVD-CU group only showed a significant decrease in the CBF/dReHo ratio of the left occipital gyrus in the conventional frequency band (P<0.05).

Conclusions: Our findings showed that CSVD patients exhibited impaired dynamic patterns of neuronal synchronization and disrupted neurovascular coupling prior to the onset of dementia, which were associated with cognitive impairment in sub-frequency bands.

Keywords: Dynamic regional homogeneity (dReHo); neurovascular coupling; magnetic resonance imaging (MRI); multi-frequency; cerebral small vascular disease

Submitted Feb 12, 2025. Accepted for publication Jul 24, 2025. Published online Sep 13, 2025.

doi: 10.21037/qims-2025-344

Introduction

With the unprecedented increase in the aging population, the incidence of cerebral small vessel disease (CSVD) is rising, severely increasing the family economic and healthcare burdens in China (1,2). CSVD mainly affects cognitive function (ranging from subjective cognitive decline and mild cognitive impairment to dementia), characterized by executive and memory dysfunctions (3). The clinical diagnosis of CSVD primarily depended on neuroimaging features based on magnetic resonance imaging (MRI). However, the neuroimaging characteristics identified through conventional MRI were highly heterogeneous, which cannot fully explain the severity of cognitive impairment and brain damage in patients with CSVD (4). In addition, a study has pointed out that early identification and timely intervention of CSVD can delay disease progression and prevent the occurrence of dementia (5). Therefore, it is necessary to find sensitive neuroimaging markers for the early diagnosis of CSVD patients.

Resting state functional MRI (rs-fMRI), as an advanced neuroimaging technique, can reflect spontaneous brain activity by analyzing blood oxygen level dependent (BOLD) signals (6). The regional homogeneity (ReHo) analysis based on rs-fMRI technology was considered a potential neuroimaging marker for tracking changes in brain functional homogeneity, which can characterize local neural activity by measuring regional synchronizations of BOLD signals among neighboring brain voxels (7). Many studies have shown that cognitive dysfunction was closely related to the abnormal changes of cerebral ReHo (8,9). Patients with CSVD have also been found to exhibit abnormal ReHo in multiple brain regions, which were significantly correlated with cognitive scores (10). At present, ReHo analysis of CSVD mainly focuses on the static features and conventional frequency band (0.01–0.08 Hz) (10,11). Recently, a dynamic ReHo (dReHo) analysis utilizing a sliding window approach has been proposed to capture the dynamic changes of ReHo over time (12,13). Compared with static features, the dynamic changes of MRI indicators can provide more information about the brain disease (13,14). In addition, research showed that conventional frequency band can be divided into slow-4 (0.027–0.073 Hz) and slow-5 (0.01–0.027 Hz), which respectively reflect the functional activity of subcortical nuclei and cortical regions (15). Analyzing the functional activity of slow-4 and slow-5 may avoid interference from the physiological noise of other frequency bands, thereby increasing the accuracy of identifying brain regions with abnormal functional activity, especially improving the detection power of the spatial distribution of abnormal regions. However, to our knowledge, the changes in dReHo in sub-frequency bands and their relationship with cognitive function in CSVD patients in the early stage have not been explored.

In addition to changes in brain functional homogeneity, the cerebral blood flow (CBF) analysis method based on non-invasive arterial spin labeling (ASL) technique also identified perfusion abnormalities of brain regions in patients with CSVD (16). In fact, spontaneous neural activity is often closely coupled with CBF, known as neurovascular coupling, which is significant in maintaining normal brain function. More importantly, the disruption of neurovascular coupling has been increasingly believed to be associated with cognitive decline and may be an important potential neural mechanism (17,18). Therefore, it is of great significance to combine dReHo and CBF analyses to explore the changes in neurovascular coupling and its relationship with cognitive performance in CSVD patients, which has been rarely investigated in the past.

The aim of this study was to investigate the dReHo and neurovascular coupling alterations in conventional frequency band and two sub-frequency bands in patients with CSVD by using MRI technology, and to investigate their correlation with cognitive scores. This study hypothesized CSVD patients exhibit specific changes of the dReHo and neurovascular coupling in sub-frequency bands in the early stages, and these changes may be closely related to specific cognitive impairments. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-344/rc).

Methods

Participants

Our study included 59 CSVD patients in our hospital from 2017 to 2021 according to the relevant guidelines and regulations. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the ethics committee of the First Affiliated Hospital of Chongqing Medical University (No. K2023-266). All participants provided written informed consent prior to participating in the study.

Patients with CSVD were included according to the following criteria (19): (I) white matter lesions; (II) lacunar cases; (III) no hemorrhages, watershed infarcts and other white matter lesions with specific causes. Fifty-nine patients were further divided into two groups: a group of 27 patients with mild cognitive impairment (CSVD-MCI) and a group of 32 patients with cognitive unimpaired (CSVD-CU). The inclusion criteria for CSVD-MCI group, CSVD-CU group and 30 normal controls (NC), as well as the exclusion criteria for all participants, have been previously described (20).

Neuropsychological assessment

The mini-mental state examination (MMSE) test was conducted to assess global cognitive function; the auditory verbal learning test (AVLT), one-back, rey-osterrieth complex figure test (Rey-CFT), digit span test (DST), and reverse DST (R-DST) were conducted to assess memory function; the stroop color word test (stroop-1 and stroop-2) and clock drawing test (CDT) were conducted to assess executive function; the clinical dementia rating scale CDR test was conducted to assess dementia.

MRI acquisition

The GE Signa Hdxt (eight-channel phased-array head coil) 3.0T scanner was used to obtain MRI data. All participants were asked to hold still, close their eyes, and remain awake during the scanning. The MRI scan parameters were as follows: (I) 3D-T1 images: repetition time (TR) =8.3 ms, echo time (TE) =3.3 ms, flip angle =15°, slice thickness =1 mm, field of view (FOV) =240×240 mm2, matrix =240×240, and scanning time =6.45 min; (II) rs-fMRI images: TR =2,000 ms, TE =40 ms, flip angle =90°, slice thickness =4 mm, FOV =240×240 mm2, matrix = 64 × 64, scanning time point =240, and scanning time =8 min; (III) 3D pseudo-continuous arterial spin labeling (3D-pCASL) images: TR =5,216 ms, TE =9.8 ms, flip angle =90°, slice thickness =4 mm, FOV =240×240 mm2, matrix =512×8, delay time after marking =2,525 ms.

MRI data analysis

The dReHo analysis was performed on the rs-fMRI images with the DPARSF (http://www.rfmri.org/DPARSF) v4.3 for Statistical Parametric Mapping (SPM) 12 (http://www.fil.ion.ucl.ac.uk/spm) based on MATLAB R2020b platform. Based on previous research (21), the preprocessing steps included: deleting the first 10 volumes; correcting time layer and head movement (translation or rotation of >3 mm or 3°); regressing out nuisance covariates (linear trend, cerebral spinal fluid signals, white matter signals and Friston-24 parameters of head motions); using the Montreal Neurological Institute (MNI) space with DARTEL to normalize space and resampling to voxel size of 3×3×3 mm3; detrending; temporal filtering with conventional frequency band (0.01–0.08 Hz), slow-4 frequency band (0.027–0.073 Hz), and slow-5 frequency band (0.01–0.027 Hz). The dReHo maps in three frequency bands were generated by DPABI v4.3 (http://rfmri.org/dpabi) based on the sliding window method. Based on previous research (22,23), the selected parameters were as follows: setting the window length to 50 TR (100 s) and step size to five TR (10 s). Then, 37 windows were generated and the corresponding coefficient of variation (CV = standard deviation/mean) was obtained to evaluate the variability of dReHo. Finally, the CV maps were smoothed with 8-mm FWHM Gaussian kernel.

For the CBF analysis, the CBF and ASL images were obtained by an automated image postprocessing tool in the ADW workstation and were processed using SPM12 running in MATLAB R2020b. The data processing procedures (22,24) included: removing non-brain tissues; the ASL images were coregistered to corresponding 3D-T1 images; the 3D-T1 images were segmented to generated the deformation fields; then, the CBF images were spatially normalized to MNI space using the transformation fields and resampled to 3×3×3 mm3; the resulting CBF maps were standardized by dividing by the average CBF values of whole gray matter; smoothing with 8-mm FWHM Gaussian kernel. To quantitatively evaluate the neurovascular coupling, the correlation coefficients between dReHo maps in three frequency bands and CBF maps were computed at whole gray matter level, which was defined by the Automated Anatomical Labeling (AAL) 90 atlas (22,25). For regional level, the CBF/dReHo ratios (representing metabolic energy per unit of neuronal activity) in three frequency bands of each voxel within gray matter mask were obtained and further z-scored.

Statistical analysis

The between-group comparisons in the demographic and neuropsychological data were performed by the chi-square test, Kruskal-Wallis H test, analysis of variance (ANOVA) with the SPSS 26.0 software. The ANOVA and post-hoc analyses (Bonferroni correction with P<0.05) were applied to assess the between-group differences in the dReHo-CBF correlation coefficients in three frequency bands. In addition, the between-group differences in the dReHo maps and CBF/dReHo ratios in three frequency bands were examined by analysis of covariance (ANCOVA) and post-hoc analysis with false discovery rate (FDR) correction with DPABI controlling for age, sex and education as covariates (FDR correction was used for multiple comparison with P<0.05 and the cluster size >20 for removing small clusters). Finally, to explore the relationships between the above results with significant difference and the neuropsychological scores, the partial correlation analyses (controlling for age, sex and education as covariates) were performed in the two patient groups (the FDR correction was used for multiple comparison with P<0.05).

Results

Demographic data and neuropsychological scores

There was no significant difference in sex, age, education and head motion among the three groups (P>0.05). Apart from the DST scores, the other neuropsychological scores of CSVD-MCI group were significantly lower than those of NC group (P<0.05). The MMSE, AVLT-immediate recall (AVLT-IR), AVLT-delayed recall (AVLT-DR), AVLT-recognition recall (AVLT-RR), CDT, stroop-1 and stroop-2 scores of CSVD-MCI group were significantly lower than those of CSVD-CU group (P<0.05). The AVLT-IR, AVLT-DR, Rey-CFT-immediate recall (Rey-CFT-IR) and one-back scores of CSVD-CU group exhibited significant decreases compared to NC group (P<0.05). Table 1 displays the detailed results.

Table 1

Comparison results of demographic data and neuropsychological scores between groups

Data NC (n=30) CSVD-CU (n=32) CSVD-MCI (n=27) F/χ2 P
Sex (female/male) 16/14 13/19 11/16 1.287 0.525†
Age (years) 67.50 (65.00–72.00) 70.50 (68.25–75.00) 70.00 (66.00–73.00) 4.618 0.099‡
Education (years) 9.00 (9.00–11.75) 9.00 (9.00–14.25) 9.00 (9.00–12.00) 1.822 0.225‡
Head motion 0.07 (0.04–0.11) 0.08 (0.05–0.10) 0.07 (0.05–0.09) 1.477 0.478‡
MMSE 28.00 (27.00–30.00) 28.00 (27.00–29.00) 25.00 (24.00–26.00)* 57.497 <0.001‡
AVLT-IR 8.32±1.78 6.86±2.14* 5.27±2.28* 15.378 <0.001§
AVLT-DR 9.20±2.63 7.44±2.94* 5.07±2.83* 15.424 <0.001§
AVLT-RR 17.50 (11.25–20.00) 16.00 (14.00–20.75) 10.00 (6.00–14.00)* 20.125 <0.001‡
Rey-CFT-IR 35.00 (33.25–36.00) 33.00 (30.00–36.00)* 32.00 (29.00–35.00)* 15.381 <0.001‡
Rey-CFT-DR 13.75 (10.25–17.75) 12.50 (5.25–18.63) 5.00 ( 2.00–12.00)* 13.813 0.001‡
DST 8.00 (7.00–8.00) 8.00 (7.00–9.75) 8.00 (7.00–8.00) 0.461 0.794‡
R-DST 4.00 (4.00–5.00) 4.00 (3.00–4.00) 3.00 (2.00–4.00)* 19.331 <0.001‡
CDT 3.50 (3.00–4.00) 4.00 (3.00–4.00) 3.00 (2.00–3.00)* 15.507 <0.001‡
Stroop-1 107.50 (103.75–110.25) 105.00 (96.50–110.75) 96.00 (80.00–103.00)* 15.916 <0.001‡
Stroop-2 100.00 (90.00–107.00) 95.00 (79.50–103.00) 74.00 (60.00–91.00)* 15.595 <0.001‡
One-back 58.00 (56.00–60.00) 55.00 (39.00–59.00)* 40.00 (24.00–53.00)* 26.573 <0.001‡

Data are presented as n, median (range) or mean ± standard deviation. †, χ2; ‡, Kruskal-Wallis H test; §, analysis of variance. *, significant difference compared to the NC group (P<0.05); , significant difference compared to the CSVD-CU group (P<0.05). AVLT, auditory verbal learning test; CDT, clock drawing test; CSVD-CU, cerebral small vessel disease patients with cognition unimpaired; CSVD-MCI, cerebral small vessel disease patients with mild cognition impairment; DR, delayed recall; DST, digit span test; IR, immediate recall; MMSE, mini-mental state examination; NC, normal controls; R-DST, reverse DST; Rey-CFT, rey-osterrieth complex figure test; RR, recognition recall.

Changes of dReHo

Compared with the NC group, the dReHo values of left middle occipital gyrus, left middle temporal gyrus, bilateral thalamus, bilateral putamen, right calcarine, bilateral postcentral gyrus and left supplementary motor area in conventional frequency band also exhibited significant increases (P<0.05), and the dReHo values of bilateral inferior occipital gyrus, left middle occipital gyrus and left supplementary motor area in slow-5 frequency band exhibited significant increases in CSVD-MCI group (P<0.05). No statistically significant differences in the dReHo values of three frequency bands were observed between the CSVD-CU group and the NC or CSVD-MCI groups (P>0.05). Table 2 and Figure 1 display the significant comparison results.

Table 2

Significant comparison results of dReHo between groups

Brain regions Coordinates in MNI (mm) Voxels Peak intensity Groups
x y z
Slow-5 frequency band
   Left inferior occipital gyrus −42 −75 −9 28 5.03 CSVD-MCI > NC
   Right inferior occipital gyrus 42 −75 −3 50 5.44 CSVD-MCI > NC
   Left middle occipital gyrus −21 −99 9 22 4.90 CSVD-MCI > NC
   Left supplementary motor area 0 15 63 38 5.66 CSVD-MCI > NC
Conventional frequency band
   Left middle occipital gyrus −18 −96 9 290 5.66 CSVD-MCI > NC
   Left middle temporal gyrus −69 −21 0 38 4.68 CSVD-MCI > NC
   Left putamen −30 9 −3 40 5.79 CSVD-MCI > NC
   Right thalamus 12 −12 0 30 6.47 CSVD-MCI > NC
   Left thalamus −15 −9 0 28 5.00 CSVD-MCI > NC
   Right putamen 21 6 6 42 5.16 CSVD-MCI > NC
   Right calcarine 18 −87 12 179 5.55 CSVD-MCI > NC
   Left middle temporal gyrus −54 −60 15 25 4.74 CSVD-MCI > NC
   Left postcentral gyrus −57 −9 36 79 4.82 CSVD-MCI > NC
   Right postcentral gyrus 36 −30 63 25 4.32 CSVD-MCI > NC
   Left supplementary motor area 0 21 63 29 4.47 CSVD-MCI > NC

The between-group differences in the dReHo in three frequency bands were examined by analysis of covariance and post-hoc analysis (false discovery rate correction, P<0.05). CSVD-MCI, cerebral small vessel disease patients with mild cognition impairment; dReHo, dynamic regional homogeneity; MNI, Montreal Neurological Institute; NC, normal controls.

Figure 1 Significant comparison results of dReHo between groups. The between-group differences in the dReHo in three frequency bands were examined by ANCOVA and post-hoc analysis (false discovery rate correction, P<0.05). ANCOVA, analysis of covariance; CSVD-MCI, cerebral small vessel disease patients with mild cognition impairment; dReHo, dynamic regional homogeneity; L, left; NC, normal controls; R, right.

Changes of dReHo-CBF correlation coefficients at the whole gray matter level

No statistically significant differences in the dReHo-CBF correlation coefficient values of slow-4 (F=1.378, P=0.502), slow-5 (F=2.793, P=0.247) and conventional frequency bands (F=0.577, P=0.750) at the whole gray matter level were observed among the three groups.

Changes of CBF/dReHo ratios at the regional level

In conventional frequency band, the CSVD-MCI group exhibited significantly increased CBF/dReHo values in right thalamus but significantly decreased CBF/dReHo values in right inferior temporal gyrus, right medial orbital superior frontal gyrus, bilateral middle occipital gyrus, bilateral heschl gyrus, left anterior cingulate gyrus, left middle frontal gyrus and left postcentral gyrus compared with the NC group (P<0.05), and the CSVD-CU group showed significantly decreased CBF/dReHo values in left middle occipital gyrus compared with the NC group (P<0.05). In slow-4 frequency band, the CSVD-MCI group exhibited significantly increased CBF/dReHo values in bilateral thalamus but significantly decreased CBF/dReHo values in bilateral insula, bilateral middle occipital gyrus, left angular, left anterior cingulate gyrus, left middle frontal gyrus and left postcentral gyrus compared with the NC group (P<0.05). In slow-5 frequency band, the CSVD-MCI group showed significantly decreased CBF/dReHo values in the left inferior occipital gyrus, right inferior temporal gyrus, bilateral middle occipital gyrus, bilateral angular, right medial superior frontal gyrus and left inferior parietal gyrus compared with the NC group (P<0.05). No statistically significant differences in the CBF/dReHo values of three frequency bands were observed between the CSVD-MCI group and the CSVD-CU group (P>0.05). Table 3 and Figure 2 show the significant comparison results.

Table 3

Significant comparison results of CBF/dReHo ratios between groups

Brain regions Coordinates in MNI (mm) Voxels Peak intensity Groups
x y z
Slow-4 frequency band
   Left thalamus −18 −27 6 29 4.88 CSVD-MCI > NC
   Right thalamus 18 −21 6 32 4.51 CSVD-MCI > NC
   Left insula −45 6 −3 33 −4.86 CSVD-MCI < NC
   Left middle occipital gyrus −36 −84 15 271 −5.84 CSVD-MCI < NC
   Right insula 42 −15 12 56 −5.95 CSVD-MCI < NC
   Right middle occipital gyrus 33 −78 21 262 −5.47 CSVD-MCI < NC
   Left angular −42 −57 33 180 −4.79 CSVD-MCI < NC
   Left insula −39 −15 6 45 −5.02 CSVD-MCI < NC
   Left anterior cingulate gyrus 0 39 27 28 −5.11 CSVD-MCI < NC
   Left middle frontal gyrus −42 27 36 24 −5.23 CSVD-MCI < NC
   Left postcentral gyrus −48 −18 51 37 −4.85 CSVD-MCI < NC
Slow-5 frequency band
   Left inferior occipital gyrus −42 −75 −9 26 −4.35 CSVD-MCI < NC
   Right inferior temporal gyrus 42 −75 −3 47 −5.04 CSVD-MCI < NC
   Left middle occipital gyrus −18 −96 6 121 −5.03 CSVD-MCI < NC
   Right middle occipital gyrus 30 −87 15 24 −4.76 CSVD-MCI < NC
   Right angular 36 −66 30 99 −5.24 CSVD-MCI < NC
   Left angular −33 −69 30 36 −4.85 CSVD-MCI < NC
   Right medial superior frontal gyrus 9 36 45 26 −4.84 CSVD-MCI < NC
   Left inferior parietal gyrus −39 −39 51 67 −4.57 CSVD-MCI < NC
Conventional frequency band
   Right thalamus 21 −27 3 24 4.79 CSVD-MCI > NC
   Right inferior temporal gyrus 42 −66 0 31 −4.24 CSVD-MCI < NC
   Right medial orbital superior frontal gyrus 6 57 −6 27 −4.07 CSVD-MCI < NC
   Right middle occipital gyrus 33 −66 33 535 −6.76 CSVD-MCI < NC
   Left middle occipital gyrus −33 −84 15 787 −6.54 CSVD-MCI < NC
   Left heschl gyrus −42 −18 9 59 −4.53 CSVD-MCI < NC
   Right heschl gyrus 42 −18 12 58 −4.49 CSVD-MCI < NC
   Left anterior cingulate gyrus 0 39 18 49 −4.48 CSVD-MCI < NC
   Left middle frontal gyrus −39 30 33 100 −5.38 CSVD-MCI < NC
   Left postcentral gyrus −48 −15 48 91 −5.58 CSVD-MCI < NC
   Left middle occipital gyrus −30 −87 15 70 −5.47 CSVD-CU < NC

The between-group differences in the CBF/dReHo ratios in three frequency bands were examined by analysis of covariance and post-hoc analysis (false discovery rate correction, P<0.05). CBF, cerebral blood flow; CSVD-CU, cerebral small vessel disease patients with cognition unimpaired; CSVD-MCI, cerebral small vessel disease patients with mild cognition impairment; dReHo, dynamic regional homogeneity; MNI, Montreal Neurological Institute; NC, normal controls.

Figure 2 Significant comparison results of CBF/dReHo between groups. The between-group differences in the CBF/dReHo in three frequency bands were examined by ANCOVA and post-hoc analysis (false discovery rate correction, P<0.05). ANCOVA, analysis of covariance; CBF, cerebral blood flow; CSVD-CU, cerebral small vessel disease patients with cognition unimpaired; CSVD-MCI, cerebral small vessel disease patients with mild cognition impairment; dReHo, dynamic regional homogeneity; L, left; NC, normal controls; R, right.

Correlation analysis

In CSVD-MCI group, the dReHo values of right inferior occipital gyrus (r=−0.580, P=0.040) in slow-5 frequency band were negatively related to Rey-CFT-DR scores; the CBF/dReHo values of right inferior temporal gyrus (r=0.577, P=0.036) in slow-5 frequency band were positively associated with Rey-CFT-DR scores; the CBF/dReHo values of left anterior cingulate gyrus in slow-4 frequency band were negatively correlated with Rey-CFT-DR (r=−0.505, P=0.045) and AVLT-DR (r=−0.497, P=0.030) scores, respectively. All significant correlation analysis results are displayed in Figure 3.

Figure 3 Significant correlation analysis results (this figure was drawn by Figdraw). The partial correlation analyses (controlling for age, sex and education as covariates) were performed to evaluate the relationships in the two patient groups (false discovery rate correction, P<0.05). Red line: positive correlation; blue line: negative correlation. AVLT, auditory verbal learning test; CBF, cerebral blood flow; dReHo, dynamic regional homogeneity; DR, delayed recall; Rey-CFT, rey-osterrieth complex figure test.

Discussion

Compared with the NC group, we found that the CSVD-MCI group showed abnormal dReHo and CBF/dReHo values in the conventional frequency band and two sub-frequency bands (dReHo mainly increased significantly, but CBF/dReHo ratio mainly decreased significantly), which mainly occurred in the brain regions of the default network, salience network, and visual network. The brain regions with significantly abnormal dReHo and CBF/dReHo ratios in the sub-frequency bands were similar to those in the conventional frequency band. More importantly, in the sub-frequency bands, the dReHo and CBF/dReHo values with significant changes in specific brain regions were significantly correlated with cognitive scores in CSVD-MCI group. For CSVD-CU group, only a significant decrease of CBF/dReHo ratios in the left middle occipital gyrus was observed in the conventional frequency band compared to the NC group.

The dReHo analysis can reflect the dynamic characteristics of local brain functional homogeneity, such as excessive variability (increased dReHo) and excessive stability (decreased dReHo), indicating brain dysfunction (12,13). In this study, the CSVD-MCI group showed significantly increased dReHo in the left middle occipital gyrus, left middle temporal gyrus, bilateral thalamus, and left supplementary motor area in the conventional frequency band, while no abnormal change in dReHo was observed in the CSVD-CU group. This may indicate that CSVD-MCI patients had excessive ReHo variability in multiple brain regions, representing damage to the homogeneity of local neuronal function activity. This was consistent with previous research using dynamic amplitude of low frequency fluctuations analysis, finding that CSVD patients exhibit increased variability in spontaneous brain activity (26,27). Notably, the brain regions with significantly increased dReHo in the slow-5 frequency band were similar to those in the conventional frequency band in the CSVD-MCI group, but relatively fewer and mainly occurred in the occipital lobe. A pathological anatomical study on CSVD patients demonstrated that cortical micro-infarctions were also more common in the occipital lobe, which was also related to the severity of CSVD (28). These may confirm the damage to the occipital lobe in CSVD patients. In addition, the increased dReHo value in the right inferior occipital gyrus in the slow-5 frequency band was significantly negatively related to the Rey-CFT-DR scores. The occipital lobe is the main area of the visual network, involved in the processing of visual information (29). Therefore, the excessive variability of dReHo in the occipital lobe may affect visual memory (Rey-CFT-DR scores) by hindering the formation of visual information. Visual memory impairment has been found to be associated with decreased CBF in the occipital cortex in patients with diabetes (30). Furthermore, Wang et al. analyzed spontaneous brain activity and suggested that the occipital lobe was one of the main brain regions involved in consolidating visual memory processes (31). Therefore, the abnormality of occipital dynamic functional homogeneity in the slow-5 frequency may be a potential mechanism for visual memory decline in CSVD patients.

Neurovascular units, composed of neurons, astrocytes, smooth muscle cells, and endothelial cells, are the basis for functions such as blood oxygen exchange and signalling in the brain (32). Research has shown that endothelial cell injuries may be the early pathological basis for neuronal damage in cerebrovascular diseases (33). So, neurovascular unit injury may be the key pathological mechanism of cerebrovascular diseases. The correlation coefficient and ratio between CBF and dReHo have been recommended as neuroimaging markers for evaluating the function of neurovascular units (22). In a healthy brain, neurovascular coupling maintains coordination to support normal brain function, and deviation from this coordination may lead to an increase or decrease in the correlation coefficient and ratio between CBF and dReHo. This study used CBF and dReHo to analyze the changes in neurovascular coupling in CSVD patients, and found in the three frequency bands, the CBF/dReHo ratios in the bilateral middle occipital gyrus, right inferior temporal gyrus, left anterior cingulate gyrus, right insula, and right medial superior frontal gyrus were significantly reduced, while the CBF/dReHo ratios in the right thalamus were significantly increased in CSVD-MCI group. These brain regions were the main components of the default network, salience network, and visual network, indicating CSVD may mainly affect the neurovascular coupling of these three brain networks. Previous studies have also shown that these networks were indeed susceptible to CSVD (4,34). More importantly, the significantly decreased CBF/dReHo ratios of the right inferior temporal gyrus in the slow-5 frequency band were positively correlated with Rey-CFT-DR scores, as well as the decreased CBF/dReHo ratios of the left anterior cingulate gyrus in the slow-4 frequency band were negatively associated with Rey-CFT-DR and AVLT-DR scores in CSVD-MCI group. The inferior temporal gyrus was a part of the visual pathway and was considered to play an important role in advanced cognitive functions related to vision, such as memory function (35). This was confirmed by the study of diabetes patients with mild cognitive impairment, demonstrating the reduction of function activities of the inferior temporal gyrus was closely related to memory dysfunction (36,37). The anterior cingulate gyrus was a core component of the salience network and correlated with memory performance (38). A task state study suggested that memory tasks can activate the anterior cingulate cortex (39). Therefore, we inferred that the decrease in CBF/dReHo ratios (reduced coordination of neurovascular coupling) of the inferior temporal gyrus and anterior cingulate gyrus in sub-frequency bands may play an important role in memory dysfunction related to CSVD.

Contrary to the hypothesis, no statistical difference was observed in the dReHo-CBF correlation coefficient values among the three groups. This may be attributed to the fact that in the early stage of CSVD disease, in addition to significantly decreased correlation coefficients, some brain regions may have increases in correlation coefficient (which may represent a compensatory mechanism), resulting in no change in the correlation coefficient of the whole gray matter level. This was also observed in diabetes patients with mild cognitive impairment (22). But this requires future studies to confirm. In addition, compared with the results of dReHo analysis, there were more brain regions with abnormal CBF/dReHo ratios in the CSVD-MCI group, and the CSVD-CU group showed significantly decreased CBF/dReHo values in the left middle occipital gyrus. These findings indicated that, compared to dReHo and CBF-dReHo correlation coefficients, CBF/dReHo ratio may be a more sensitive marker for detecting brain function changes in the early stage of the disease (40). It is worth mentioning that the CBF/dReHo ratios of the left middle occipital gyrus of the CSVD-CU group in the conventional frequency band were significantly reduced compared to the NC group, while there was no significant between-group difference in dReHo. This demonstrated that abnormal neurovascular coupling has already occurred in CSVD patients before their cognitive function declined, and as the disease progressed, the number of abnormal brain regions gradually increased.

The present study had several shortcomings: (I) a relatively small-sample size cross-sectional study, so a longitudinal study with a larger cohort was required. (II) Although the sliding window parameters were strictly set according to the standards of previous research and previous studies have shown that changing the sliding window parameters has a limited impact on dynamic indicators (13,41), the optimal parameters still need further validation. (III) This study did not consider the impacts of white matter lesions, cavities, enlargement of perivascular spaces, and microbleeds on results. In the future, after further excluding or controlling the influence of these MRI features, direct functional changes in the gray matter itself should be analyzed, especially stratified analysis based on the total MRI feature load of CSVD. (IV) Not considering the impact of extensive white matter lesions on image normalization and not considering the registration of CBF maps by dividing by the average CBF value of the entire gray matter may risk inflating differences in focal CBF. Further research should improve registration quality by masking WML and using alternative normalization methods (e.g., identifying a reference region with minimal group differences) to validate our results. (V) Although the correlation and ratio between CBF and dReHo images were recommended for characterizing neurovascular coupling, they were indirect methods that may not directly and accurately measure neurovascular coupling. Employing direct techniques such as CO2 inhalation for assessing cerebrovascular reactivity, cerebral metabolic rate measurement of oxygen with positron emission tomography, or electroencephalography in future research may help differentiate between neurovascular coupling breakdown and primary neural dysfunction. (VI) This study did not include patients with dementia, who should be included for a comprehensive analysis of the neural mechanisms underlying CSVD.

Conclusions

This study employed dReHo and CBF analyses to find that CSVD patients exhibited impaired dynamic patterns of neuronal synchronization and disrupted neurovascular coupling in conventional frequency bands and sub-frequency bands prior to the onset of dementia. These abnormalities in sub-frequency bands were associated with cognitive impairment. This emphasized the importance of dynamic features in sub-frequency bands for supplementing pathological information of brain diseases, as well as the potential role of neurovascular decoupling in sub-frequency bands in the occurrence and development of cognitive deficits related to CSVD.

Acknowledgments

The authors thank the study participants for providing data.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-344/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-344/dss

Funding: This work was supported by the National Natural Science Foundation of China (grant number 81671666); the Chongqing Municipal Youth Science and Technology Talent Training Project (grant number 2022QNXM055); and the Chongqing Science and Technology Innovation Leading Talent Support Program (grant number cstc2022ycjh-bgzXM0230).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-344/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the ethics committee of the First Affiliated Hospital of Chongqing Medical University (No. K2023-266). All participants provided written informed consent prior to participating in the study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

  1. Rajeev V, Fann DY, Dinh QN, Kim HA, De Silva TM, Lai MKP, Chen CL, Drummond GR, Sobey CG, Arumugam TV. Pathophysiology of blood brain barrier dysfunction during chronic cerebral hypoperfusion in vascular cognitive impairment. Theranostics 2022;12:1639-58. [Crossref] [PubMed]
  2. Xin H, Wen H, Feng M, Gao Y, Sui C, Zhang N, Liang C, Guo L. Disrupted topological organization of resting-state functional brain networks in cerebral small vessel disease. Hum Brain Mapp 2022;43:2607-20. [Crossref] [PubMed]
  3. Tap L, Vernooij MW, Wolters F, van den Berg E, Mattace-Raso FUS. New horizons in cognitive and functional impairment as a consequence of cerebral small vessel disease. Age Ageing 2023;52:afad148. [Crossref] [PubMed]
  4. Ter Telgte A, van Leijsen EMC, Wiegertjes K, Klijn CJM, Tuladhar AM, de Leeuw FE. Cerebral small vessel disease: from a focal to a global perspective. Nat Rev Neurol 2018;14:387-98. [Crossref] [PubMed]
  5. Sanford AM. Mild Cognitive Impairment. Clin Geriatr Med 2017;33:325-37. [Crossref] [PubMed]
  6. Fox MD, Raichle ME. Spontaneous fluctuations in brain activity observed with functional magnetic resonance imaging. Nat Rev Neurosci 2007;8:700-11. [Crossref] [PubMed]
  7. Zang Y, Jiang T, Lu Y, He Y, Tian L. Regional homogeneity approach to fMRI data analysis. Neuroimage 2004;22:394-400. [Crossref] [PubMed]
  8. Zhang Z, Liu Y, Jiang T, Zhou B, An N, Dai H, Wang P, Niu Y, Wang L, Zhang X. Altered spontaneous activity in Alzheimer's disease and mild cognitive impairment revealed by Regional Homogeneity. Neuroimage 2012;59:1429-40. [Crossref] [PubMed]
  9. Cai S, Wang Y, Kang Y, Wang H, Kim H, von Deneen KM, Huang M, Jiang Y, Huang L. Differentiated Regional Homogeneity in Progressive Mild Cognitive Impairment: A Study With Post Hoc Label. Am J Alzheimers Dis Other Demen 2018;33:373-84. [Crossref] [PubMed]
  10. Diciotti S, Orsolini S, Salvadori E, Giorgio A, Toschi N, Ciulli S, Ginestroni A, Poggesi A, De Stefano N, Pantoni L, Inzitari D, Mascalchi M. Resting state fMRI regional homogeneity correlates with cognition measures in subcortical vascular cognitive impairment. J Neurol Sci 2017;373:1-6. [Crossref] [PubMed]
  11. Liu X, Cheng R, Chen L, Gong J, Luo T, Lv F. Altered Neurovascular Coupling in Subcortical Ischemic Vascular Disease. Front Aging Neurosci 2021;13:598365. [Crossref] [PubMed]
  12. Li K, Tian Y, Chen H, Ma X, Li S, Li C, Wu S, Liu F, Du Y, Su W. Temporal Dynamic Alterations of Regional Homogeneity in Parkinson's Disease: A Resting-State fMRI Study. Biomolecules 2023;13:888. [Crossref] [PubMed]
  13. Xue K, Liang S, Yang B, Zhu D, Xie Y, Qin W, Liu F, Zhang Y, Yu C. Local dynamic spontaneous brain activity changes in first-episode, treatment-naïve patients with major depressive disorder and their associated gene expression profiles. Psychol Med 2022;52:2052-61. [Crossref] [PubMed]
  14. Cui Q, Sheng W, Chen Y, Pang Y, Lu F, Tang Q, Han S, Shen Q, Wang Y, Xie A, Huang J, Li D, Lei T, He Z, Chen H. Dynamic changes of amplitude of low-frequency fluctuations in patients with generalized anxiety disorder. Hum Brain Mapp 2020;41:1667-76. [Crossref] [PubMed]
  15. Zuo XN, Di Martino A, Kelly C, Shehzad ZE, Gee DG, Klein DF, Castellanos FX, Biswal BB, Milham MP. The oscillating brain: complex and reliable. Neuroimage 2010;49:1432-45. [Crossref] [PubMed]
  16. Sun Y, Cao W, Ding W, Wang Y, Han X, Zhou Y, Xu Q, Zhang Y, Xu J. Cerebral Blood Flow Alterations as Assessed by 3D ASL in Cognitive Impairment in Patients with Subcortical Vascular Cognitive Impairment: A Marker for Disease Severity. Front Aging Neurosci 2016;8:211. [Crossref] [PubMed]
  17. Li P, Mu J, Ma X, Ding D, Ma S, Zhang H, Liu J, Zhang M. Neurovascular coupling dysfunction in end-stage renal disease patients related to cognitive impairment. J Cereb Blood Flow Metab 2021;41:2593-606. [Crossref] [PubMed]
  18. Xing C, Feng J, Yao J, Xu XM, Wu Y, Yin X, Salvi R, Chen YC, Fang X. Neurovascular coupling dysfunction associated with cognitive impairment in presbycusis. Brain Commun 2024;6:fcae215. [Crossref] [PubMed]
  19. Román GC, Erkinjuntti T, Wallin A, Pantoni L, Chui HC. Subcortical ischaemic vascular dementia. Lancet Neurol 2002;1:426-36. [Crossref] [PubMed]
  20. Huang J, Wang W, Cheng R, Liu X, Chen L, Luo T. A multi-parametric MRI study on changes in the structure, function, and connectivity of thalamic subregions and their relationship with cognitive impairment in patients with subcortical ischemic vascular disease. Brain Res 2025;1850:149420. [Crossref] [PubMed]
  21. Zhang X, Wang W, Bai X, Mei Y, Tang H, Yuan Z, Zhang X, Li Z, Zhang P, Hu Z, Zhang Y, Yu X, Sui B, Wang Y. Alterations in regional homogeneity and multiple frequency amplitudes of low-frequency fluctuation in patients with new daily persistent headache: a resting-state functional magnetic resonance imaging study. J Headache Pain 2023;24:14. [Crossref] [PubMed]
  22. Shang S, Ye J, Wu J, Zhang H, Dou W, Krishnan Muthaiah VP, Tian Y, Zhang Y, Chen YC, Yin X. Early disturbance of dynamic synchronization and neurovascular coupling in cognitively normal Parkinson's disease. J Cereb Blood Flow Metab 2022;42:1719-31. [Crossref] [PubMed]
  23. Jiang Y, Du W, Li Y, Gao B, Liu N, Song Q, Wang N, Wu J, Miao Y. Disturbed Dynamic Brain Activity and Neurovascular Coupling in End-Stage Renal Disease Assessed With MRI. J Magn Reson Imaging 2025;61:1831-44. [Crossref] [PubMed]
  24. Hu B, Yan LF, Sun Q, Yu Y, Zhang J, Dai YJ, Yang Y, Hu YC, Nan HY, Zhang X, Heng CN, Hou JF, Liu QQ, Shao CH, Li F, Zhou KX, Guo H, Cui GB, Wang W. Disturbed neurovascular coupling in type 2 diabetes mellitus patients: Evidence from a comprehensive fMRI analysis. Neuroimage Clin 2019;22:101802. [Crossref] [PubMed]
  25. Rolls ET, Huang CC, Lin CP, Feng J, Joliot M. Automated anatomical labelling atlas 3. Neuroimage 2020;206:116189. [Crossref] [PubMed]
  26. Zhang X, Wang Z, Zheng D, Cao X, Qi W, Yuan Q, Zhang D, Liang X, Ruan Y, Zhang S, Tang W, Huang Q, Xue C. Aberrant spontaneous static and dynamic amplitude of low-frequency fluctuations in cerebral small vessel disease with or without mild cognitive impairment. Brain Behav 2023;13:e3279. [Crossref] [PubMed]
  27. Song J, Lei T, Li Y, Zhou L, Yan W, Li H, Chen L. Dynamic alterations in the amplitude of low-frequency fluctuation in patients with cerebral small vessel disease. Front Mol Neurosci 2023;16:1200756. [Crossref] [PubMed]
  28. Ii Y, Ishikawa H, Shindo A, Matsuyama H, Matsuura K, Matsuda K, Yoshimaru K, Satoh M, Kogue R, Umino M, Maeda M, Tomimoto H. Association between cortical microinfarcts and total small vessel disease burden in cerebral amyloid angiopathy on 3-Tesla magnetic resonance imaging. Eur J Neurol 2021;28:794-9. [Crossref] [PubMed]
  29. Machielsen WC, Rombouts SA, Barkhof F, Scheltens P, Witter MP. FMRI of visual encoding: reproducibility of activation. Hum Brain Mapp 2000;9:156-64. [Crossref] [PubMed]
  30. Cui Y, Liang X, Gu H, Hu Y, Zhao Z, Yang XY, Qian C, Yang Y, Teng GJ. Cerebral perfusion alterations in type 2 diabetes and its relation to insulin resistance and cognitive dysfunction. Brain Imaging Behav 2017;11:1248-57. [Crossref] [PubMed]
  31. Wang K, Jiang T, Yu C, Tian L, Li J, Liu Y, Zhou Y, Xu L, Song M, Li K. Spontaneous activity associated with primary visual cortex: a resting-state FMRI study. Cereb Cortex 2008;18:697-704. [Crossref] [PubMed]
  32. Lecrux C, Hamel E. The neurovascular unit in brain function and disease. Acta Physiol (Oxf) 2011;203:47-59. [Crossref] [PubMed]
  33. Wong SM, Jansen JFA, Zhang CE, Hoff EI, Staals J, van Oostenbrugge RJ, Backes WH. Blood-brain barrier impairment and hypoperfusion are linked in cerebral small vessel disease. Neurology 2019;92:e1669-77. [Crossref] [PubMed]
  34. Dey AK, Stamenova V, Turner G, Black SE, Levine B. Pathoconnectomics of cognitive impairment in small vessel disease: A systematic review. Alzheimers Dement 2016;12:831-45. [Crossref] [PubMed]
  35. Lin YH, Young IM, Conner AK, Glenn CA, Chakraborty AR, Nix CE, Bai MY, Dhanaraj V, Fonseka RD, Hormovas J, Tanglay O, Briggs RG, Sughrue ME. Anatomy and White Matter Connections of the Inferior Temporal Gyrus. World Neurosurg 2020;143:e656-66. [Crossref] [PubMed]
  36. Xiong Y, Chen X, Zhao X, Fan Y, Zhang Q, Zhu W. Altered regional homogeneity and functional brain networks in Type 2 diabetes with and without mild cognitive impairment. Sci Rep 2020;10:21254. [Crossref] [PubMed]
  37. Zhou X, Zhang J, Chen Y, Ma T, Wang Y, Wang J, Zhang Z. Aggravated cognitive and brain functional impairment in mild cognitive impairment patients with type 2 diabetes: a resting-state functional MRI study. J Alzheimers Dis 2014;41:925-35. [Crossref] [PubMed]
  38. Zhang J, Andreano JM, Dickerson BC, Touroutoglou A, Barrett LF. Stronger Functional Connectivity in the Default Mode and Salience Networks Is Associated With Youthful Memory in Superaging. Cereb Cortex 2020;30:72-84. [Crossref] [PubMed]
  39. Buckner RL, Koutstaal W. Functional neuroimaging studies of encoding, priming, and explicit memory retrieval. Proc Natl Acad Sci U S A 1998;95:891-8. [Crossref] [PubMed]
  40. Guo X, Zhu J, Zhang N, Zhang L, Qi Y, Cai H, Zhang X, Sun J, Wang Q, Yang L, Shi FD, Yu C. Altered neurovascular coupling in neuromyelitis optica. Hum Brain Mapp 2019;40:976-86. [Crossref] [PubMed]
  41. Chen J, Wei Y, Xue K, Gao X, Zhang M, Han S, Wen B, Wu G, Cheng J. Static and temporal dynamic changes of intrinsic brain activity in early-onset and adult-onset schizophrenia: a fMRI study of interaction effects. Front Neurol 2024;15:1445599. [Crossref] [PubMed]
Cite this article as: Huang J, Wang W, Cheng R, Liu X, Luo T. Altered dynamic regional homogeneity and neurovascular coupling in multi-frequency bands in patients with cerebral small vascular disease. Quant Imaging Med Surg 2025;15(10):9805-9817. doi: 10.21037/qims-2025-344

Article Options

Download Citation

Share