Vascular Risk Factors and Brain Health in Aging: Insights from a Community-Based Cohort Study

Abstract

Background:

The aging population and high rates of Alzheimer’s disease (AD) create significant medical burdens, prompting a need for early prevention. Targeting modifiable risk factors like vascular risk factors (VRFs), closely linked to AD, may provide a promising strategy for intervention.

Objective:

This study investigates how VRFs influence cognitive performance and brain structures in a community-based cohort.

Methods:

In this cross-sectional study, 4,667 participants over 50 years old, drawn from the Beijing Ageing Brain Rejuvenation Initiative project, were meticulously examined. Cognitive function and VRFs (diabetes mellitus, hypertension, hyperlipidemia, obesity, and smoking), were comprehensively assessed through one-to-one interviews. Additionally, a subset of participants (n = 719) underwent MRI, encompassing T1-weighted and diffusion-weighted scans, to elucidate gray matter volume and white matter structural network organization.

Results:

The findings unveil diabetes as a potent detriment to memory, manifesting in atrophy within the right supramarginal gyrus and diminished nodal efficiency and degree centrality in the right inferior parietal lobe. Hypertension solely impaired memory without significant structural changes. Intriguingly, individuals with comorbid diabetes and hypertension exhibited the most pronounced deficits in both brain structure and cognitive performance. Remarkably, hyperlipidemia emerged as a factor associated with enhanced cognition, and preservation of brain structure.

Conclusions:

This study illuminates the intricate associations between VRFs and the varied patterns of cognitive and brain structural damage. Notably, the synergistic effect of diabetes and hypertension emerges as particularly deleterious. These findings underscore the imperative to tailor interventions for patients with distinct VRF comorbidities, especially when addressing cognitive decline and structural brain changes.

Keywords

Aging cohort Alzheimer’s disease brain structural cognitive performance vascular risk factors

INTRODUCTION

The aging population and high prevalence of Alzheimer’s disease (AD) are major concerns, posing significant medical, financial, and social burdens [1, 2]. Considering the lack of effective treatments and repeated failed clinical trials, it is therefore imperative to focus on efficient early prevention before clinical symptoms arise [3, 4]. One promising approach is to accurately assess and quantify modifiable risk factors, such as vascular risk factors (VRFs), which are closely linked to the occurrence and development of AD [5, 6]. VRFs are common comorbidities of AD [7 –9], vascular related damage is already present in the stages of AD, even preceding the AD classic biomarker changes [10, 11]. The hypothesis that the pathogenesis of AD is related to brain microcirculation damage has been proposed for decades [12, 13], the two-hit vascular hypothesis of AD also supports the association between VRFs and AD. In particular, experimental and autopsy studies have demonstrated that VRFs cause inflammation, endothelial dysfunction, mitochondrial dysfunction, and arterial stiffening in the brain, leading to changes in cerebrovascular alterations and chronic cerebral hypoperfusion [14 –16]. These pathophysiological processes can affect the neurodegenerative pathway, eventually resulting in cognitive impairment and AD [6 , 18]. Therefore, addressing VRFs could provide an effective approach to early intervention for AD [19].

Thus, VRFs may play a pivotal role in the prophylaxis and treatment of AD, they can be used both as early warnings and targets of intervention. Consequently, this can provide the basis for a relatively economical and yet impactful strategy for the preservation of brain health in older adults [11, 20]. A necessary precondition for the success of such a strategy is the accurate and quantitative clarification of the effects of VRFs on cognitive function and their relationship to alterations in brain structure. Numerous studies have revealed that VRFs are among the primary risk factors for brain structural and further cognitive dysfunctions [21, 22]. However, three key issues are still misunderstood.

First, the differential degrees of damage of these single VRFs to cognition (from a view of multi-domains) and brain structure (both morphologically and topologically) have not been thoroughly studied. It is proposed that the magnitude of damage elicited by different VRFs would not be the same, as well as the importance of these single VRFs should be ranked [23]. In other words, more attention should be paid to the VRFs that produce more severe injury, thus making the prevention and therapy against senile cognitive diseases more efficient and economical. Second, various combinations of VRFs have not been studied and compared adequately. In clinical situations, it is not uncommon that an individual may have two or more VRFs that synergistically produce effects [17, 24]. Only a few studies have addressed this topic, proposing that elevated vascular risks were related to poorer brain structural and diffusion MRI manifestation [25, 26]. However, these studies focused on the cumulative effects of the VRFs they included and did not attempt to elucidate the effects of varying VRF combinations or singular VRF, and in the aspect of brain imaging markers, few studies focused on brain connectome measures such as white matter topological properties. Third, given that the brain-vulnerable regions of VRFs and AD overlap [27, 28], the measurement and evaluation of the common vulnerable regions should be more targeted to accurately measure and generalize more reliable damage patterns.

In considering the above issues, we aimed to investigate the effects of VRFs on multidomain cognitive performance, gray matter volume, and white matter structural network topological organization. To achieve this, we examined the specific patterns associated with single VRFs and then clarified the effects of various combinations of VRFs. Our hypothesis of this community-based exploratory study was that older adults with different single VRFs may exhibit specific cognitive deficits and various brain morphological and topological patterns based on their different degrees of risk, and Combined VRFs may exacerbate cognitive and brain damage in different forms compared with single VRFs. By investigating these issues, we aimed to gain insights into how VRFs impact cognitive and brain health in aging.

METHODS

Study design and participants

Data used in this cross-sectional study were obtained from participants enrolled in a prospective, community-based cohort study, the Beijing Ageing Brain Rejuvenation Initiative (BABRI) [29]. Participants signed up for the BABRI project and the inclusion criteria were met: 1) age 50 or older; 2) completion of six or more years of education; 3) willingness to participate in face-to-face interviews. The exclusion criteria included: 1) diagnosed with dementia, any other neurological and psychiatric problem, and brain tumor according to medical records from the community healthcare center; 2) inability to complete cognitive examinations. In total, 4,667 participants (the entire group) completed a self-administered questionnaire regarding their medical history, cognitive function, and living habits. Among them, 3,834 participants (the multi-risk group) had records related to all interested VRFs of this study and thus, were used in combined VRFs analyses. For MRI scanning, 690 participants (the MRI group) underwent T1-weighted imaging scans. Among this population, 658 participants had diffusion-weighted imaging scans (Supplementary Figure 1). The institutional review board of the Imaging Center for Brain Research at Beijing Normal University approved this study and written informed consent was obtained from all participants.

Vascular risk factors

VRFs, including type 2 diabetes mellitus (DM), hypertension (HTN), hyperlipidemia (HPL), smoking, and obesity, were determined based on the patient’s self-reported medical history and records obtained from the community public health centers with the participants’ consent. Type 2 DM was defined as a self-reported diagnosis of DM and/or the use of insulin and/or other hypoglycemic agents. HTN was determined by self-reported diagnosis of hypertension and/or antihypertensive medication use. HPL was defined by self-reported diagnosis of hyperlipidemia and/or consummation of lipid-lowering medicine. Participants who smoked were included in the smoking group, which excluded ex-smokers and passive smokers. In addition, participants were considered obese if their body mass index (BMI) was 30 or above, in accordance with the World Health Organization Global Health Observatory data.

Neuropsychological tests and MRI acquisition and preprocessing

The neuropsychological test battery evaluated five cognitive domains, including episodic memory, visuospatial ability, language, attention, and executive function (Supplementary Material). Participants who scored 1.0 (SD) below the age- and education-adjusted norms on two tests within the same cognitive domain were considered to have domain-specific cognitive impairment. For combined VRF analyses, the test scores were transformed into Z-scores, with the sum of Z-scores being in the same domain taken as the measurement of this domain’sperformance.

MRI data (T1-weighted and diffusion-weighted images) acquisition was performed using a Siemens Trio 3.0 Tesla scanner (Trio; Siemens, Erlangen, Germany) at the Imaging Center for Brain Research at Beijing Normal University (Supplementary Material).

Voxel-based morphometry analysis

We conducted voxel-based morphometry (VBM) analysis on voxels in regions that were associated with both VRFs and AD-related cognitive impairment [27, 30]. These regions comprise the frontal, parietal, temporal, and subcortical regions (VRF ROIs in Fig. 1A and Supplementary Table 1) in accordance with the Automated anatomical labeling (AAL) atlas [31]. To clarify the specific effect of each VRF on gray matter volume (GMV), we conducted multi-variable regression analyses with five VRFs as regressors, age, gender, level of education, and total intracranial volume (TIV) as covariants. Then, to investigate GM changes from different combinations of VRFs, in eight groups (Supplementary Table 2), we performed the analysis of covariance (ANCOVA) with age, gender, level of education, and TIV serving as covariants. The Gaussian random field (GRF) theory correction was used, with the threshold of voxel-level being p < 0.001 and cluster-level p < 0.05. The GMV values of regions with VRF-related GMV reduction or expansion were then extracted, and partial correlations were performed to evaluate the relationships between GMV and multidomain cognitive performances, controlling for the covariates. The false discovery rate (FDR) correction was used for correcting possible multiple comparisons (p < 0.05).

Fig. 1

The regions of interest (ROIs) that defined by the AAL atlas for voxel-based morphometry and white matter network analyses. SFGdor, dorsolateral superior frontal gyrus; ORBsup, orbital superior frontal gyrus; MFG, middle frontal gyrus; ORBmid, orbital middle frontal gyrus; IFGoperc, opercularis inferior frontal gyrus; IFGtriang, triangularis inferior frontal gyrus; ORBinf, orbital inferior frontal gyrus; ACG, anterior cingulum; DCG, dorsal and middle cingulum; PCG, posterior cingulum; HIP, hippocampus; PHG, parahippocampal gyrus; AMYG, amygdala; SPG, superior parietal gyrus; IPL, inferior parietal lobe; SMG, supramarginal gyrus; PCUN, precuneus; CAU, caudate; PUT, putamen; PAL, pallidum; MTG, middle temporal gyrus; ITG, inferior temporal gyrus.

White matter (WM) network construction and topological analysis

The two fundamental elements of the WM structural network are nodes and edges. The AAL atlas was used to classify the network nodes, and edges were obtained by performing deterministic tractography. Briefly, the T1-weighted image of each participant was coregistered to the first b0 image in individual diffusion tensor image (DTI) space. Following this, the registered T1-weighted images were then nonlinear transformed to the ICBM 152 template in the MNI space, and the inverse transformation was applied to warp the AAL atlas from the MNI space into the native DTI space. We then obtained 90 nodes of the WM network, of which 44 nodes are of interest (Fig. 1B, Supplementary Table 1). Considering the network edges, all the tracts in the FA data set were computed by seeding each voxel with an FA greater than 0.2, the tractography was terminated if it turned an angle greater than 45 degrees or reached a voxel with an FA less than 0.2 [32]. As a result, all the fiber pathways in the brain were constructed. Two GM regions were reported connected if there were at least three fiber streamlines with the two endpoints located in these two regions. We defined the average fiber number (FN) along the pathways of the interconnecting streamlines as the weights of the network edges. As a result, the FN-weighted WM network (a 90 × 90 matrix) for each participant was constructed (Supplementary Figure 2).

To characterize the underlying impacts of VRFs on WM network organization, we quantified graph theory-based topological properties on the nodal level, including the nodal efficiency (Enodal) and the degree centrality (Dc). These reflect the information communication ability of each node[33]. Network analyses were conducted using GRETNA software (https://www.nitrc.org/projects/gretna/), and the results were visualized using BrainNet Viewer (https://www.nitrc.org/projects/bnv/).

Statistical analysis

Demographic characteristics, cognitive performance, and VRF prevalence were reported individually for the entire group (n = 4,667), the multi-risk group (n = 3,893), and the MRI group (n = 690). One-way ANOVA or the χ² test was used to test for significant differences among the three groups. To clarify the specific relationship of single VRFs to each cognitive domain, we performed binary logistic regression (domain-specific cognitive impairment was labeled as “1”. Furthermore, the five VRFs, and age, gender, and level of education were proposed as predictors for the entire group, and the impacts of VRFs’ on cognition were represented by odds ratios (OR). Focusing solely on VRFs that significantly affect cognitive function, we then sorted the multi-risk group into eight combined VRF subgroups: 0, normal control (NC); 1, HPL; 2, HTN; 3, DM; 4, HPL & HTN; 5, DM & HPL; 6, DM & HTN; 7, DM, HTN, & HPL (Supplementary Table 2). Based on these subgroups, we evaluated the combined effects of the VRFs on each cognitive domain using one-way ANCOVA, with age, gender, and level of education as covariants. Following VBM and topological analyses, the GMV, Enodal, and Dc of brain regions with significant intergroup differences were extracted and partially correlated with the mean Z-score of each cognitive domain. This enabled us to determine the relationship between brain structural properties and cognitive performance.

RESULTS

Demographics

4,667 participants (age range 50–91 years, mean age 65.65 years, 62.8% female) were included in the current study. The demographic characteristics and cognitive functions of the participants and the distribution of VRF prevalence in the entire, multi-risk, and MRI groups are shown in (Table 1). The prevalence of hypertension (53.4%) was reported as the highest, with hyperlipidemia being the second highest (42.8%).

Table 1

Demographic profiles and VRF prevalence in three groups

	Entire group	Multi-risk group	MRI group	F/χ² (p)
Demographic profile
Number	4,667	3,834	690
Age (y)	65.65±7.56	65.26±7.32	65.16±7.00	3.485 (p = 0.031)
Education (y)	11.31±2.96	11.41±2.92	11.5±3.01	2.288 (p = 0.102)
Gender (F, %)	2,932 (62.8%)	2,393 (62.4%)	430 (62.3%)	0.178 (p = 0.915)
BMI	24.83±3.43	24.87±3.42	24.8±3.43	0.272 (p = 0.762)
VRF morbidity
Diabetes mellitus	1,184 (25.4%)	964 (25.1%)	175 (25.4%)	0.060 (p = 0.970)
Hypertension	2,493 (53.4%)	2,042 (53.3%)	346 (50.1%)	2.648 (p = 0.266)
Hyperlipidemia	1,999 (42.8%)	1,733 (45.2%)	298 (43.2%)	0.992 (p = 0.609)
Obesity	284 (6.1%)	250 (6.5%)	52 (7.5%)	0.854 (p = 0.652)
Smoking	557 (11.9%)	512 (13.4%)	97 (14.1%)	0.295 (p = 0.863)
DM&HTN	828 (17.7%)	673 (17.6%)	113 (16.4%)	0.774 (p = 0.679)
DM&HPL	673 (14.4%)	563 (14.7%)	103 (14.9%)	0.218 (p = 0.897)
HTN&HPL	1,251 (26.8%)	1,065 (27.8%)	174 (25.2%)	2.302 (p = 0.259)
HPL&HTN&DM	481 (10.3%)	395 (10.3%)	75 (10.9%)	0.519 (p = 0.771)

Values are mean±SD or Nos. of participants (percentage). The comparisons of age, education, and BMI among the three groups were performed with ANOVA. The p-values for gender, diabetes, hyperlipidemia, hypertension, obesity, and smoking ratio were obtained using a Chi-square test.

Single vascular risk factors

In analyses of single VRFs and cognitive function, we first considered the outcome of general cognition and then, of the five separate cognitive domains. The binary logistic regression results indicated that patients with diabetes and hypertension tended to have poorer cognitive performance, especially in the memory domain. In contrast, hyperlipidemia was shown to be manifested by the maintenance of cognitive function, especially in the language domain. No significant association was found between obesity or smoking and cognitive functions (Table 2; Fig. 2A).

Table 2

Associations of vascular risk factors with multi-domain cognitive performance

Factors		General	Memory	Visuospatial	Language	Attention	Executive
Age	OR	1.33^***	1.09^***	1.19	2.47^***	1.84^***	1.36^*
	95% CI	1.15–1.53	0.93–1.27	0.82–1.71	1.84–3.32	1.46–2.32	1.03–1.80
Education	OR	0.56^***	0.56^***	0.23^***	0.36^***	0.83	0.49^***
	95% CI	0.47–0.66	0.46–0.68	0.12–0.44	0.24–0.53	0.65–1.07	0.34–0.69
Gender	OR	0.80^**	0.69^***	1.49	1.20	0.96	0.75
	95% CI	0.68–0.95	0.58–0.83	0.95–2.33	0.88–1.65	0.75–1.24	0.55–1.02
DM	OR	1.25^**	1.29^**	1.31	0.95	1.05	1.24
	95% CI	1.06–1.46	1.08–1.54	0.88–1.97	0.69–1.30	0.81–1.35	0.92–1.69
HTN	OR	1.23^**	1.32^**	1.15	1.07	1.09	1.06
	95% CI	1.06–1.42	1.12–1.54	0.79–1.68	0.81–1.42	0.87–1.37	0.80–1.41
HPL	OR	0.86^*	0.96	0.71	0.74^*	0.80	0.78
	95% CI	0.74–0.99	0.82–1.12	0.49–1.04	0.56–0.99	0.64–1.01	0.59–1.03
Obesity	OR	1.27	1.18	1.34	0.66	1.11	1.29
	95% CI	0.97–1.67	0.88–1.58	0.72–2.48	0.35–1.23	0.73–1.70	0.80–2.10
Smoking	OR	1.03	0.91	0.76	0.77	1.39	1.34
	95% CI	0.82–1.29	0.72–1.16	0.38–1.52	0.47–1.26	0.99–1.93	0.91–1.98

OR, odds ratio; CI: confidence interval. Significance: ^*p < 0.05, ^**p < 0.01, ^***p < 0.001.

Fig. 2

Associations of single and combined VRFs with multi-domain cognitive performance. (A) Odds ratios (OR) of single VRFs predict cognition status. OR > 1 indicates possible risk factors, and OR < 1 indicates possible protective factors. (B) The between-group difference of eight combined VRF sub-groups on multi-domain cognitive performance. Significance: ^* p < 0.05, ^**p < 0.01, ^***p < 0.001.

We further investigated the effect of single VRFs on GMV. DM is significantly associated with GMV atrophy in the right supramarginal gyrus (SMG.R), and hyperlipidemia is related to increased GMV in the right hippocampus (HIP.R). Meanwhile, no significant relationship was found in other single VRFs (Fig. 3A). The mean GMV of the SMG.R cluster is found positively correlated with attention performance (r = 0.099, p = 0.016). Regarding the hyperlipidemia-related HIP.R cluster, the GMV positively correlated with all cognitive domains (memory, r = 0.199, p < 0.001; visual-spatial ability, r = 0.082, p = 0.046; attention, r = 0.147, p < 0.001; executive function, r = 0.144, p < 0.001; language, r = 0.184, p < 0.001).

Fig. 3

VBM results of the single (A) and combined (B) VRF-related changes in GMV. The regions shown in blue are regions in which VRF-positive/combined VRF participants have more atrophied GMV than VRF-negative/single VRF ones; the regions in red show the opposite situation. Threshold: Gaussian random field (GRF) corrected, voxel-level p < 0.001, cluster-level p < 0.05. SMG.R, right supramarginal gyrus; HIP.R, right hippocampus; PCUN.R, right precuneus; SFGdor.R, right dorsolateral superior frontal gyrus; CAU.R, right caudate; PUT.L, left putamen.

Regression analyses demonstrate that multiple regions had DM-related decreases and HPL-related increases on both Enodal and Dc (Fig. 4A, B; Supplementary Table 3). Following the FDR correction for multiple comparisons, we found that the right inferior parietal lobe (IPL.R) was a significant region with decreased Enodal in DM patients (standardized beta = –0.127, p = 0.001, FDR corrected p = 0.044). In contrast, the Enodal of IPL.R was closely correlated with memory (r = 0.085, p = 0.030), language (r = 0.137, p < 0.001), attention (r = 0.194, p < 0.001), and executive (r = 0.154, p < 0.001) performance. The Dc of the left amygdala (AMYG.L, standardized beta = 0.141, p < 0.001, FDR corrected p < 0.001) and inferior temporal gyrus (ITG.L, standardized beta = 0.124, p = 0.001, FDR corrected p = 0.022) was found increased for HPL patients. Moreover, the Dc of ITG.L is significantly associated with language (r = 0.107, p = 0.006), attention (r = 0.123, p = 0.002), and executive (r = 0.160, p < 0.001) performance.

Fig. 4

Single and combined VRFs related topological properties of WM network. (A) DM-related E_nodal decreases (nodes in blue) and HPL-related E_nodal increases (nodes in red) in multiple regions. (B) DM-related Dc decreases (nodes in blue) and HPL-related Dc increases (nodes in red) in multiple regions. For (A) and (B), the statistical threshold of nodes in blue and red is uncorrected p < 0.05, and for nodes in brilliant blue and red, the threshold is FDR corrected p < 0.05. (C) significant intergroup differences on Enodal among eight combined VRF subgroups. Significance: ^*p < 0.05; ^**p < 0.01; ^***p < 0.001. IPL.R, right inferior parietal lobe; AMYG.L, left amygdala; ITG.L, left inferior temporal gyrus; PCG.L, left posterior cingulate gyrus; SPG.L, left superior parietal gyrus.

Combined vascular risk factors

To explore whether different combined VRFs show specific influences on multi-domain cognition, we performed a one-way ANCOVA among the eight combined VRF subgroups. Memory was found to be the most influenced cognitive domain, followed by visuospatial ability, attention, and executive function (Fig. 2B). All these domains indicated that the combined DM & HTN subgroup performed the worst, and HPL performed best. To further explore the positive effect of hyperlipidemia, we found that the statin-use group performed better than the untreated group in the executive function and language domain (Supplementary Table 4).

Exploring the effect of combined VRFs on cognitive function (Fig. 3B). The DM & HPL combined subgroup presented a larger GM volume in the right precuneus (PCUN.R) than the single DM subgroup and the GMV of the PCUN. The PCUN.R cluster positively correlates with attention function (r = 0.090, p = 0.028). Regarding the DM& HTN subgroup, multiple regions with GM atrophy were found located in the right superior frontal gyrus (SFGdor.R) compared to the single DM subgroup, and in the right caudate (CAU.R) and left putamen (PUT.L) compared to the single HTN subgroup. The GMV of SFGdor. The R cluster correlated positively with memory (r = 0.101, p = 0.014), visual-spatial ability (r = 0.123, p = 0.003), attention (r = 0.084, p = 0.040) and language performance (r = 0.150, p < 0.001). However, on the contrary, both CAU.R and PUT.L presented GMV to be negatively correlated with visual-spatial ability (r=–0.110, p = 0.007; r=–0.106, p = 0.010).

The intergroup topological differences among eight combined VRF subgroups were observed in the left posterior cingulate gyrus (PCG.L, F = 3.657, p = 0.001, FDR corrected p = 0.044) and left superior parietal gyrus (SPG.L, F = 3.297, p = 0.002, FDR corrected p = 0.044) on the property of Enodal (Fig. 4 C), whereas the Enodal of these two regions significantly correlated with cognitive function: for PCG.L with, memory, r = 0.094, p = 0.016; language function, r = 0.134, p = 0.001, attention, r = 0.169, p < 0.001; executive function, r = 0.188, p < 0.001; for SPG.L with, attention, r = 0.145, p < 0.001; executive function, r = 0.129, p = 0.001. The intergroup differences were presented to be primarily driven by the declined Enodal of the DM& HTN subgroup and the DM& HPL subgroup. Nevertheless, the Enodal of the bilateral dorsal cingulate gyrus (DCG) and the left middle frontal gyrus (MFG.L), together with the Dc of the PCG.L, illustrated a marginally significant intergroup difference, possibly resulting from the topological dysfunction in the DM& HTN subgroup (Supplementary Figure 3).

DISCUSSION

This study aimed to investigate the heterogeneity of cognitive function and brain structure associated with various combinations of VRFs in the health population in a large-scale aging cohort. To our knowledge, this is the first study to investigate the ranking of the impact of different VRFs on brain health, as well as the associations between various combinations of VRFs and multidomain cognitive function and AD-related brain structure. Our findings support the hypothesis that single VRFs have varying degrees of risk, with a progressively lower degree of risk from DM to HTN to HPL. However, the combinations of certain VRFs were associated with a more significant negative impact on cognitive function and brain structure, particularly the combination of DM and HTN. This suggests that these two VRFs may synergistically elevate the risk of cognitive and brain structural degeneration. Interestingly, we also found that hyperlipidemia alone, or in combination with other VRFs may have a beneficial effect on the brain. Overall, our study highlights the importance of considering VRF combinations in assessing cognitive and brain health in older adults.

A study from the UK biobank explored associations of multiple VRFs with the whole brain structure, smoking pack-years, HTN, and DM were found associated with poorer performance in all brain measures [25]. Given that VRFs and AD are closely related, our study focused on brain ROIs that were previously shown associated with both VRFs and AD. Our study indicated that the risk of different single VRFs varies, with DM and HTN having the highest degree of risk, and HPL showing contrasting results, however, our findings for smoking and obesity were not significant. Specifically, DM showed a negative effect on both cognition and brain measures, while HTN solely presented a significant cognitive decline. Previous research has also confirmed that DM and HTN are both risk factors for AD, with DM having a greater risk of dementia than HTN [23, 34]. Our study adds to the existing literature by exploring the association between various combinations of VRFs and cognitive function and brain structure perspectives. As a result of the complex etiology and challenging treatment of AD, a continued integrated approach and comprehensive evaluation are necessary to move beyond clinical trials and instead, find potential effective intervention methods[19]. The two-hit vascular hypothesis of AD has been proposed[19], whereby VRFs act as the initiating factor of “one-hit”, leading to blood-brain barrier dysfunction and reduced cerebral blood flow, thus establishing the pathological basis of Aβ production and clearance. The “second hit” is vascular risk-induced Aβ imbalance that exacerbates neuronal dysfunction and neurodegeneration. Previous studies have demonstrated that the incorporation of more VRFs leads to poorer cognitive performance [35], brain [25, 36]. and WM network [26]. The direct cause of brain structural and microstructural damage due to VRFs may be attributed to decreased brain vascular reactivity, leading to a subsequent imbalance in cerebral hemodynamics [37]. These lead to sustained impairment of cerebral perfusion and neurovascular dysfunction. When cerebral perfusion fails to adequately meet the normal operational demands of synapses or neurons, apoptosis of neurons and glial cells occurs [38], resulting in structural brain damage observable from a radiological perspective. Additionally, many studies have demonstrated that brain structural damage in patients with mild cognitive impairment or AD typically occurs following compromised cerebral perfusion [39 –41]. Building on the above evidence, our research findings provide further evidence of brain susceptibility patterns in different combinations of VRFs. we demonstrate that there are significant differences in the pattern of effects of different VRF combinations on brain health, particularly in the DM and HTN group. Moreover, the comorbidity of HTN and DM was shown to synergistically affect poor cognition and all brain measures. Compared with the single DM and HTN, the DM and HTN group illustrated damage in the frontal and subcortical structures, which are both susceptible to aging and AD [42], This therefore, suggests that DM and HTN synergize the pathological process of aging and AD. The potential shared defects in DM and HTN perhaps include insulin resistance [43], arterial stiffness, and complicated hemodynamic feedback cycles [44]. This would consequently lead to the aggravation of vascular structural and functional dysfunction, occurrences of adverse outcomes such as coronary heart disease, stroke, etc. [45], increasing the longitudinal risk of adverse cardiovascular and non-cardiovascular outcomes [46]. In our study, although the absolute number of DM and HTN comorbidities seems relatively small, two-thirds of DM patients suffer from HTN. This is given that DM and HTN comorbidities can lead to accelerated vascular pathological injury and target organ outcomes. For patients suffering from DM, more attention should be paid to their blood pressure, and the same for patients with HTN. Interestingly, our findings suggest that both single and combined HPL were associated with better cognitive performance and brain structural integrity, particularly in the temporal lobe and limbic system, being regions affected by early AD neuropathology [47]. However, the relationship between HPL and AD is proposed confounding, unreliable, or non-rivalrous, as it may involve complex lipid metabolism mechanisms [48, 49] or the use of statins [50]. Although the underlying protective mechanism due to HPL and its treatment is not yet clear, regular control of VRFs is undoubtedly worthwhile [51].

Given the strengths, there are, however, several potential limitations to this study. Firstly, due to the use of cross-sectional data, this study can only provide insights into the correlations between VRFs, cognition, and brain health at a single point in time. Unable to determine the long-term impact of vascular risk on structural brain decline and delineate the trajectory of decline Further longitudinal studies are vital in establishing causality and determining the temporal relationship between VRFs and brain health. Secondly, this study did not stratify participants by age, so it is unclear whether there are age-related effects. Thirdly, our study failed to find a relationship between smoking, obesity, and brain health, which may be related to certain flaws in the design of data collection for these two risk factors. Apart from VRFs, further exploration could be conducted on the co-occurrence and cumulative effects of mental disorders and other risk factors on cognition and brain health in the elderly population. Finally, the study may be subject to potential bias, given that all the participants are from the Beijing community, which may have a higher medical treatment rate and better compliance than other areas of China. This could, therefore, limit the generalizability of the results and weaken the effects of VRFs on brain structure and function.

Conclusion

In conclusion, the study underscores the nuanced impact of both individual vascular risk factors (VRFs) and their combinations on cognitive performance and brain structure in the elderly. Significantly, the coexistence of diabetes and hypertension manifests more pronounced detrimental effects than any singular VRF. These findings underscore the critical importance of proactively addressing VRFs, particularly in individuals with multiple comorbidities, as a strategic approach to safeguard cognitive and brain health during the aging process. Looking ahead, it is imperative for future research endeavors to delve deeper into the mechanistic underpinnings of these associations, aiming to unveil targeted interventions that effectively manage VRFs and mitigate the onset and progression of age-related cognitive decline and neurodegeneration.

AUTHORS CONTRIBUTIONS

Wenxiao Wang (Conceptualization; Data curation; Investigation; Methodology; Writing – original draft; Writing – review & editing); Yiru Yang (Methodology; Writing – original draft; Writing – review & editing); Feng Sang (Methodology); Yaojing Chen (Validation); Xin Li (Supervision; Writing – review & editing); Kewei Chen (Validation; Visualization); Jun Wang (Writing – review & editing); Zhanjun Zhang (Conceptualization).

Footnotes

ACKNOWLEDGMENTS

We thank all the volunteers for their participation in the study.

FUNDING

This work was supported by State Key Program of National Natural Science of China (grant number 82130118), STI2030-Major Projects (2022ZD0211600) and the Natural Science Foundation of China (grant number 32171085).

CONFLICT OF INTEREST

Zhanjun Zhang is an Editorial Board Member of this journal but was not involved in the peer-review process of this article nor had access to any information regarding its peer review.

All other authors have no conflict of interest to report.

DATA AVAILABILITY

The data supporting the findings of this study are available from the corresponding author on reasonable request.

The supplementary material is available in the electronic version of this article: .

References

GBD 2016 Neurology Collaborators (2019) Global, regional, and national burden of neurological disorders, 1990-2016: A systematic analysis for the Global Burden of Disease Study 2016, Lancet Neurol 18, 459–480.

Jia

, Quan

, Fu

, Zhao

, Li

, Wei

, Tang

, Qin

, Wang

, Qiao

, Shi

, Wang

, Du

, Zhang

, Luo

, Qu

, Zhou

, Gauthier

, Jia

, Group for the Project of Dementia Situation in China (2020) Dementia in China: Epidemiology, clinical management, and research advances, Lancet Neurol 19, 81–92.

Long

, Holtzman

(2019) Alzheimer disease: An update on pathobiology and treatment strategies, Cell 179, 312–339.

Karran

, De Strooper

(2022) The amyloid hypothesis in Alzheimer disease: New insights from new therapeutics, Nat Rev Drug Discov 21, 306–318.

Livingston

, Huntley

, Sommerlad

, Ames

, Ballard

, Banerjee

, Brayne

, Burns

, Cohen-Mansfield

, Cooper

, Costafreda

, Dias

, Fox

, Gitlin

, Howard

, Kales

, Kivimäki

, Larson

, Ogunniyi

, Orgeta

, Ritchie

, Rockwood

, Sampson

, Samus

, Schneider

, Selbæk

, Teri

, Mukadam

(2020) Dementia prevention, intervention, and care: 2020 report of the Lancet Commission, Lancet 396, 413–446.

Bangen

, Nation

, Delano-Wood

, Weissberger

, Hansen

, Galasko

, Salmon

, Bondi

(2015) Aggregate effects of vascular risk factors on cerebrovascular changes in autopsy-confirmed Alzheimer’s disease, Alzheimers Dement 11, 394–403e391.

Biessels

, Despa

(2018) Cognitive decline and dementia in diabetes mellitus: Mechanisms and clinical implications, Nat Rev Endocrinol 14, 591–604.

Attems

, Jellinger

(2014) The overlap between vascular disease and Alzheimer’s disease-lessons from pathology, BMC Med 12, 206.

Kučikienėq

, Costa

, Banning

, Van Gils

, Schulz

, Ramakers

, Verhey

, Vos

, Reetz

(2022) The role of vascular risk factors in biomarker-based AT (N) groups: A German-Dutch memory clinic study, J Alzheimers Dis 87, 185–195.

10.

Iturria-Medina

, Sotero

, Toussaint

, Mateos-Pérez

, Evans

(2016) Early role of vascular dysregulation on late-onset Alzheimer’s disease based on multifactorial data-driven analysis, Nat Commun 7, 11934.

11.

Hachinski

, Einhaupl

, Ganten

, Alladi

, Brayne

, Stephan

BCM

, Sweeney

, Zlokovic

, Iturria-Medina

, Iadecola

, Nishimura

, Schaffer

, Whitehead

, Black

, Ostergaard

, Wardlaw

, Greenberg

, Friberg

, Norrving

, Rowe

, Joanette

, Hacke

, Kuller

, Dichgans

, Endres

, Khachaturian

(2019) Preventing dementia by preventing stroke: The Berlin Manifesto, Alzheimers Dement 15, 961–984.

12.

De la Torre

, Mussivan

(1993) Can disturbed brain microcirculation cause Alzheimer’s disease? Neurol Resh 15, 146–153.

13.

De la Torre

(1994) Impaired brain microcirculation may trigger Alzheimer’s disease, Neurosci Biobehav Rev 18, 397–401.

14.

Lee

, Raghavan

, Bhattarai

, Siddiqui

, Sariya

, Reyes-Dumeyer

, Flowers

, Cardoso

SAL

, De Jager

, Bennett

, Schneider

, Menon

, Wang

, Lantigua

, Medrano

, Rivera

, Jimenez-Velazquez

, Kukull

, Brickman

, Manly

, Tosto

, Kizil

, Vardarajan

, Mayeux

(2022) FMNL2 regulates gliovascular interactions and is associated with vascular risk factors and cerebrovascular pathology in Alzheimer’s disease, Acta Neuropathol 144, 59–79.

15.

, De Silva

, Chen

, Faraci

(2017) Cerebral vascular disease and neurovascular injury in ischemic stroke, Circ Res 120, 449–471.

16.

Sims-Robinson

, Kim

, Rosko

, Feldman

(2010) How does diabetes accelerate Alzheimer disease pathology? Nat Rev Neurol 6, 551–559.

17.

Gottesman

, Seshadri

(2022) Risk Factors, lifestyle behaviors, and vascular brain health, Stroke 53, 394–403.

18.

Cheng

, Chiu

, Chen

, Cheng

, Lai

, Chen

(2020) The contribution of vascular risk factors in neurodegenerative disorders: From mild cognitive impairment to Alzheimer’s disease, Alzheimers Res Ther 12, 91.

19.

Zlokovic

(2011) Neurovascular pathways to neurodegeneration in Alzheimer’s disease and other disorders, Nat Rev Neurosci 12, 723–738.

20.

Gardener

, Wright

, Rundek

, Sacco

(2015) Brain health and shared risk factors for dementia and stroke, Nat Rev Neurol 11, 651–657.

21.

Kern

, Wright

, Bergfield

, Fitzhugh

, Chen

, Moeller

, Nabizadeh

, Elkind

MSV

, Sacco

, Stern

, DeCarli

, Alexander

(2017) Blood pressure control in aging predicts cerebral atrophy related to small-vessel white matter lesions, Front Aging Neurosci 9, 132.

22.

, Ma

, Sun

, Zhang

, Chen

, Zhang

(2016) Disrupted white matter structure underlies cognitive deficit in hypertensive patients, Eur Radiol 26, 2899–2907.

23.

Fan

, Hsu

, Tung

, Chou

, Bai

(2017) Increased dementia risk predominantly in diabetes mellitus rather than in hypertension or hyperlipidemia: A population-based cohort study, Alzheimers Res Ther 9, 7.

24.

Terentes-Printzios

, Vlachopoulos

, Xaplanteris

, Ioakeimidis

, Aznaouridis

, Baou

, Kardara

, Georgiopoulos

, Georgakopoulos

, Tousoulis

(2017) Cardiovascular risk factors accelerate progression of vascular aging in the general population: Results from the CRAVE Study (Cardiovascular Risk Factors Affecting Vascular Age), Hypertension 70, 1057–1064.

25.

Cox

, Lyall

, Ritchie

, Bastin

, Harris

, Buchanan

, Fawns-Ritchie

, Barbu

, de Nooij

, Reus

, Alloza

, Shen

, Neilson

, Alderson

, Hunter

, Liewald

, Whalley

, McIntosh

, Lawrie

, Pell

, Tucker-Drob

, Wardlaw

, Gale

, Deary

(2019) Associations between vascular risk factors and brain MRI indices in UK Biobank, Eur Heart J 40, 2290–2300.

26.

Shen

, Tozer

, Markus

, Tay

(2020) Network efficiency mediates the relationship between vascular burden and cognitive impairment: A diffusion tensor imaging study in UK Biobank, Stroke 51, 1682–1689.

27.

Lamar

, Boots

, Arfanakis

, Barnes

, Schneider

(2020) Common brain structural alterations associated with cardiovascular disease risk factors and Alzheimer’s dementia: Future directions and implications, Neuropsychol Rev 30, 546–557.

28.

Raz

, Knoefel

, Bhaskar

(2016) The neuropathology and cerebrovascular mechanisms of dementia, J Cereb Blood Flow Metab 36, 172–186.

29.

Yang

, Li

, Zhang

, Chen

, Li

, Wei

, Lu

, Liang

, Liu

, Shu

, Wang

, Guan

, Tao

, Wang

, Jia

, Ai

, Cui

, Wang

, Peng

, Zhang

, Chen

, Wang

, Zhao

, Wang

, Dong

, Wang

, Zhang

, Beijing Aging Brain Rejuvenation Initiative workgroup (2021) Early prevention of cognitive impairment in the community population: The Beijing Aging Brain Rejuvenation Initiative, Alzheimers Dement 17, 1610–1618.

30.

Boots

, Zhan

, Dion

, Karstens

, Peven

, Ajilore

, Lamar

(2019) Cardiovascular disease risk factors, tract-based structural connectomics, and cognition in older adults, Neuroimage 196, 152–160.

31.

Tzourio-Mazoyer

, Landeau

, Papathanassiou

, Crivello

, Etard

, Delcroix

, Mazoyer

, Joliot

(2002) Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain, Neuroimage 15, 273–289.

32.

Mori

, Kaufmann

, Davatzikos

, Stieltjes

, Amodei

, Fredericksen

, Pearlson

, Melhem

, Solaiyappan

, Raymond

(2002) Imaging cortical association tracts in the human brain using diffusion-tensor-based axonal tracking, Magn Reson Med 47, 215–223.

33.

Rubinov

, Sporns

(2010) Complex network measures of brain connectivity: Uses and interpretations, Neuroimage 52, 1059–1069.

34.

Gottesman

, Albert

, Alonso

, Coker

, Coresh

, Davis

, Deal

, McKhann

, Mosley

, Sharrett

, Schneider

ALC

, Windham

, Wruck

, Knopman

(2017) Associations between midlife vascular risk factors and 25-year incident dementia in the Atherosclerosis Risk in Communities (ARIC) Cohort, JAMA Neurol 74, 1246–1254.

35.

Rabin

, Schultz

, Hedden

, Viswanathan

, Marshall

, Kilpatrick

, Klein

, Buckley

, Yang

, Properzi

, Rao

, Kirn

, Papp

, Rentz

, Johnson

, Sperling

, Chhatwal

(2018) Interactive associations of vascular risk and beta-amyloid burden with cognitive decline in clinically normal elderly individuals: Findings from the Harvard Aging Brain Study, JAMA Neurol 75, 1124–1131.

36.

Suzuki

, Venkataraman

, Bai

, Guitton

, Guo

, Dehghan

, Matthews

, Alzheimer’s Disease Neuroimaging Initiative (2019) Associations of regional brain structural differences with aging, modifiable risk factors for dementia, and cognitive performance, JAMA Netw Open 2, e1917257.

37.

Dai

, Lopez

, Carmichael

, Becker

, Kuller

, Gach

(2008) Abnormal regional cerebral blood flow in cognitively normal elderly subjects with hypertension, Stroke 39, 349–354.

38.

de la Torre

(2012) Cerebral hemodynamics and vascular risk factors: Setting the stage for Alzheimer’s disease, J Alzheimers Dis 32, 553–567.

39.

Edman

, Edenbrandt

, Fredén-Lindqvist

, Nilsson

, Wallin

(2011) Asymmetric cerebral blood flow in patients with mild cognitive impairment: Possible relationship to further cognitive deterioration, Dement Geriatr Cogn Disord Extra 1, 228–236.

40.

Korte

, Nortley

, Attwell

(2020) Cerebral blood flow decrease as an early pathological mechanism in Alzheimer’s disease, Acta Neuropathol 140, 793–810.

41.

Bracko

, Cruz Hernández

, Park

, Nishimura

, Schaffer

(2021) Causes and consequences of baseline cerebral blood flow reductions in Alzheimer’s disease, J Cereb Blood Flow Metab 41, 1501–1516.

42.

Mostafavi

, Gaiteri

, Sullivan

, White

, Tasaki

, Xu

, Taga

, Klein

, Patrick

, Komashko

, McCabe

, Smith

, Bradshaw

, Root

, Regev

, Yu

, Chibnik

, Schneider

, Young-Pearse

, Bennett

, De Jager

(2018) A molecular network of the aging human brain provides insights into the pathology and cognitive decline of Alzheimer’s disease, Nat Neurosci 21, 811–819.

43.

Ferrannini

, Cushman

(2012) Diabetes and hypertension: The bad companions, Lancet 380, 601–610.

44.

Safar

(2018) Arterial stiffness as a risk factor for clinical hypertension, Nat Rev Cardiol 15, 97–105.

45.

Climie

, van Sloten

, Bruno

, Taddei

, Empana

, Stehouwer

CDA

, Sharman

, Boutouyrie

, Laurent

(2019) Macrovasculature and microvasculature at the crossroads between type 2 diabetes mellitus and hypertension, Hypertension 73, 1138–1149.

46.

Jiang

, Li

, Shi

, Wang

, Qian

, Yan

, Pang

, Yang

(2022) The additive effect of essential hypertension on coronary artery plaques in type 2 diabetes mellitus patients: A coronary computed tomography angiography study, Cardiovasc Diabetol 21, 1.

47.

Braak

, Braak

(1991) Neuropathological stageing of Alzheimer-related changes, Acta Neuropathol 82, 239–259.

48.

Wood

, Li

, Muller

, Eckert

(2014) Cholesterol as a causative factor in Alzheimer’s disease: A debatable hypothesis, J Neurochem 129, 559–572.

49.

Jimenez-Conde

, Biffi

, Rahman

, Kanakis

, Butler

, Sonni

, Massasa

, Cloonan

, Gilson

, Capozzo

, Cortellini

, Ois

, Cuadrado-Godia

, Rodriguez-Campello

, Furie

, Roquer

, Rosand

, Rost

(2010) Hyperlipidemia and reduced white matter hyperintensity volume in patients with ischemic stroke, Stroke 41, 437–442.

50.

Power

, Weuve

, Sharrett

, Blacker

, Gottesman

(2015) Statins, cognition, and dementia-systematic review and methodological commentary, Nat Rev Neurol 11, 220–229.

51.

de la Torre

(2010) Vascular risk factor detection and control may prevent Alzheimer’s disease, Ageing Res Rev 9, 218–225.