Cerebral small vessel disease unveils a vascular pathway to motoric cognitive risk in aging

Abstract

Background

Motoric cognitive risk (MCR) syndrome is characterized by subjective cognitive complaints and slow gait and confers a higher risk of dementia. Cerebral small vessel disease (CSVD) is associated with poor cognitive, functional, and survival outcomes in aging. Markers of CSVD seen on magnetic resonance imaging (MRI) include white matter hyperintensities (WMHs) and lacunes.

Objective

To examine associations between imaging markers of CSVD and the MCR syndrome.

Methods

Cross-sectional data from 4 cohorts in 4 countries were examined. WMHs and lacunes were quantified from brain MRIs manually, using a standardized grading scale. Regression models examined the associations between WMH and lacunes and MCR, gait speed, slow gait, and cognitive complaints. We also compared the prevalence of the outcomes of interest between participants with “confluent or diffuse” or “no or mild” WMH. Statistical models were adjusted for age, sex, study site, and vascular risk factors.

Results

Data from 1772 participants (M Age = 71.1 years, 49.9% female) was analyzed. Higher global WMH scores were associated with MCR (aOR = 1.07, p = 0.015). Frontal and basal ganglia WMH scores were associated with MCR (aOR = 1.23, p = 0.007, aOR = 1.31, p = 0.023, respectively). Participants with “confluent-diffuse” WMH had significantly higher prevalence of MCR (30.2% versus 19.2%, p = 0.003). Basal ganglia lacunes were associated with MCR (aOR = 1.57, p = 0.018).

Conclusions

In this multi-cohort study of older adults without cognitive impairment, we show that WMH and lacunes independently predict increased risk of MCR, after adjusting for key confounders. Our findings, based on a large multi-ethnic cohort, reveal region-specific CSVD patterns linked to MCR and related outcomes.

Keywords

aging Alzheimer's disease cerebral small vessel disease cognitive decline

Introduction

Motoric cognitive risk syndrome (MCR), a predementia syndrome characterized by the combination of subjective cognitive complaints and slow gait, can be used to identify individuals at increased risk of developing Alzheimer's disease (AD) and vascular dementia (VaD).¹ The association between MCR and dementia has been shown to be stronger for VaD than AD, indicating a potential vascular component to the underlying pathophysiology. Validation studies have shown that MCR can be a powerful clinical tool as a predictor of dementia, falls, and disability given its ease of use without the need for complex cognitive tests and costly imaging.^2–5 While the link between MCR and dementia risk has been established, the neuropathological mechanisms underlying MCR are not well understood.

Gait speed decline has been found to be significantly associated with poorer performance in multiple measures of cognitive function such as processing speed, short-term memory, and attention in older individuals with and without cognitive impairment.^6–8 In addition, accelerated gait speed decline is associated with increased risk of falls and disability.^9–11 Common neural substrates associated with gait speed, gait variability and cognition have been found in MRI studies involving multiple cohorts and involve frontal, temporal, cerebellar, and striatal regions, among others.^12,13 The slowing of gait speed also predicts cognitive decline and risk of dementia in longitudinal studies.^14–17 Similarly, subjective cognitive complaints, which reflect early, self-reported declines in memory and executive function, are predictive of both dementia and functional decline.^18,19 Taken together, MCR and its components are useful clinical tools to predict adverse outcomes in older patients.

Cerebral small vessel disease (CSVD) refers to pathology affecting intracranial arterioles, capillaries, and venules supplying the subcortical white matter and deep gray matter structures of the brain. The prevalence of CSVD increases with age, and CSVD is also associated with arterial hypertension, history of tobacco smoking, diabetes, and obesity. Less common forms of CSVD are caused by genetic, autoimmune, and infectious conditions.²⁰ While the pathogenesis of CSVD is not fully understood, histological evidence of affected vessels indicates arteriosclerosis and cerebral amyloid angiopathy.²¹ These lesions are thought to give rise to endothelial dysfunction of perforating arterioles, blood-brain barrier breakdown, and inflammation leading to visible lesions on brain imaging.²²

To date, studies of associations between CSVD markers and MCR have generated mixed results. For instance, one study found that MCR is associated with frontal lobe lacunes but not with WMH,²³ while other studies have shown that total WMH volume is increased in individuals with MCR compared to those without MCR.²⁴ In this multi-cohort study of older adults without dementia, we aim to further characterize the association between CSVD imaging markers and MCR. As secondary outcomes, we also explored the association between CSVD, and the clinical components of MCR: cognitive complaints and gait speed. Here, we focus on ischemic markers (WMH and lacunes) rather than hemorrhagic (CMB) given their previously shown stronger association with cognition and overall function.²⁵ By taking a regional and quantitative approach, we hope to provide insight on what regions of the brain affected by CSVD may drive these effects. Also, by examining both MCR and its components, slow gait, and cognitive complaints, we may be able to make inferences on mechanisms behind MCR pathophysiology. Given findings that more severe global WMH are associated with MCR,²⁴ we also looked at the potential effects of confluent-diffuse vs. none-focal WMH in each region. Our study population encompassed cohorts with heterogeneous demographics and vascular risk factors to increase the generalizability of our findings.

Methods

Cohorts

Data from several independent studies spanning 4 different countries were examined: USA (LonGenity), Australia (Tasmanian Study of Cognition and Gait, TASCOG), Japan (National Center for Geriatrics and Gerontology-Study of Geriatric Syndromes, NCGG-SGS), and India (Kerala-Einstein study, KES). All cohorts included community-dwelling older adults and participants without brain imaging were excluded for this study. The LonGenity study includes participants 65 years or older and excludes individuals with a diagnosis of dementia (>8 on the Blessed Mental Status Examination and >2 on the AD8-item Informant Questionnaire) at initial screening, or severe visual or hearing impairment (n = 122).²⁶ TASCOG participants were included if aged 60–85 years. Exclusion criteria were inability to walk without aid, contraindication to MRI, diagnosis of dementia, or residence in an aged care facility (n = 383).^27,28 NCGG-SGS includes participants aged 60 years and over. Participants were recruited by letter invitation and excluded those with a diagnosis of neurological disease, impaired independence in activities of daily living (ADLs), and severe cognitive impairment (n = 1088).^29,30 KES participants were included if individuals were 60 years of age and older. Exclusion criteria were severe audiovisual loss, cognitive impairment, and medical, neurological, or psychiatric illnesses that would interfere with completion of study procedures (n = 179).³¹ In total, data from 1772 participants was included.

MRI acquisition and processing

Each study site independently collected the MRIs, which were then transferred to Albert Einstein College of Medicine for harmonized or centralized processing. Images were generated by a Philips 3 T MRI scanner (Elition multinuclear MRI/MRS system, Amsterdam, Netherlands) for LonGenity, a General Electric (GE) 1.5 T MRI scanner (GE LX Horizon, Milwaukee, Wisconsin) for TASCOG, a Siemens 3 T MRI scanner (TIM Trio, Siemens, Germany) for NCGG-SGS and a Siemens 3 T scanner (Magnatom Skyra, Siemens, Germany) for KES. FLAIR images were generated with the following sequences: a) LonGenity; TR/TI = 4300/1650 ms, TE = 325 ms, 240 × 238 acquisition matrix, and 0.46 mm voxel size b) TASCOG; TR/TI = 8802/2200 ms, TE = 130 ms, flip angle 30 ˚, voxel size 0.93 × 0.93 × 7 mm c) NCGG-SGS; TR/TI = 9000/2500 ms, TE = 100 ms d) KES; TR/TI = 4500/1800 ms, TE = 387 ms, voxel size = 0.89 mm. T1-weighted images were generated with the following sequences: a) LonGenity; TR/TE = 9.8/4.6 ms; voxel size = 1 mm.) b) TASCOG; TR/TE = 35/7 ms, flip angle 35°, field of view: 24 cm c) KES; TR/TE = 2400/2.26 ms, voxel size = 1 mm, d) NCGG- SGS; TR = 1800 ms; TE = 1.98 ms; TI = 800 ms; slice thickness, 1.1 mm.

CSVD marker quantification

Ratings were manually performed independently by two of the authors: JV and GA. JV, a study investigator, was trained in CSVD quantification by GA, a board-certified neurologist over a period of 2 months. Interrater reliability was calculated using a random subset of the Japan cohort with substantial agreement between both raters (n = 30, linearly weighted κ = 0.80, κ = 0.66 for WMH and lacunes, respectively).

White matter hyperintensities

WMH were defined as lesions appearing hyperintense on T2-weighted FLAIR sequences in subcortical white matter, deep gray matter (basal ganglia), and brainstem. We used the Age-related White Matter Changes scale validated elsewhere.³² Briefly, focal lesions in subcortical white matter measuring >5 mm were given a score of 1, confluent lesions were given a score of 2, and diffuse lesions a score of 3. For the basal ganglia, the presence of 1 focal lesion was given a score of 1, > 1 focal lesions a score of 2, and confluent lesions a score of 3. Lack of lesions were given a score of 0 for all regions. Regions of interest included frontal, parieto-occipital, temporal, basal ganglia, and infratentorial (brain stem and cerebellum). The presence of confluent-diffuse WMH in each region was defined as having a rating ≥2 bilaterally (Figure 1). Presence of severe global WMH was defined as having an overall WMH rating >10 (threshold set given a score of 2 in each region would give an overall score of 10).

Figure 1.

(A-D) examples of rating scores for WMH on axial FLAIR imaging. Arrows point to white matter lesions of interest for demonstration purposes. Panel A shows no visible WMH (WMH score of 0). Panel B shows bilateral focal WMHs in frontal regions (WMH score of 1 for each side). Panel C shows bilateral confluent WMHs in parieto-occipital regions (WMH score of 2 for each side). Panel D shows diffuse bilateral WMHs spanning both frontal and parieto-occipital regions (WMH score of 3 for both regions).

Lacunes

Lacunes were defined as ovoid or circular lesions measuring 3–15 mm in diameter appearing hypointense on FLAIR/T1 sequences with an associated hyperintense rim on FLAIR. A score of 1 was given for the presence of a lacune. Regions of interest for lacunes included frontal, parietal, occipital, temporal, brain stem, cerebellum, basal ganglia, thalamus, internal and external capsule, corpus callosum, and deep/periventricular (see Figure 2).

Figure 2.

Parietal and basal ganglia lacunes on FLAIR imaging indicated by blue arrows. Figure on the left shows a left- sided parietal region lacune. Figure on the right shows a left-sided basal ganglia lacune.

MCR, slow gait, and cognitive complaints

Participants were categorized as having slow gait if their normal walking pace (gait speed) was measured to be one standard deviation below age- and sex-standardized means for each cohort. MCR was defined as previously described.² Briefly, participants with slow gait and the presence of cognitive complaints (defined as cognitive impairment of independent activities of daily living (iADLs) and/or a positive response to the Geriatric Depression Scale (GDS) item “Do you feel that you have more problems with memory than most?”) were categorized as having MCR.

Statistical analyses

Associations between markers of CSVD and outcomes were assessed with linear and logistic regression models for continuous and categorical outcome variables, respectively. Outcome variables of interest included MCR (categorical), slow gait (categorical), cognitive complaints (categorical), and gait speed (continuous). Statistical models were adjusted for baseline covariates including age, sex, cohort, history of diabetes, hypertension, and stroke. To compare prevalence of outcomes between participants with severe global WMH burden (defined as overall score >10) and regional confluent-diffuse WMH (i.e., A score ≥2 in any brain region), Pearson's chi square test was used for categorical variables (MCR, slow gait, and cognitive complaints). Statistical analyses were performed using Stata version 17.0, and an alpha level of p < 0.05 was adopted for all analyses. Finally, we tested for significant interactions between measures of CSVD and age and sex to examine for any potential underlying biological effects.

Sensitivity analyses

Given the Japanese cohort contributed the largest sample size, we conducted weighted regressions to ensure that associations were not disproportionately driven by that site. A weight variable was created by dividing the size of the largest cohort by each participant's cohort size, thereby balancing the effective contribution of each cohort. Weighted logistic and linear regressions were then performed with the same covariate adjustments (age, sex, diabetes, hypertension, stroke, and study site) as the main models. This approach preserved the full sample size while accounting for cohort imbalance.

Results

Baseline characteristics for all cohorts are summarized in Table 1. Cohorts differed in terms of age, sex and history of diabetes, hypertension, and stroke and were therefore adjusted for in all upcoming analyses. To reduce the influence of other unmeasured differences between cohorts, all analyses were additionally adjusted for cohort status (or study site).

Table 1.

Cohort demographics. M: mean; SD: standard deviation.

	All Cohorts	USA	Australia	India	Japan
N	1772	122	383	179	1088
Age, y M ± SD	71.1 ± 6.8	78.7 ± 6.4	72.2 ± 7.1	68.9 ± 5.5	70.3 ± 6.4
Sex, % female ± SD	49.9 ± 0.5	46.9 ± 0.5	57.7 ± 0.47	37.3 ± 0.48	49.8 ± 0.5
Hypertension, % ±SD	45.88%	29.8%	49.6%	51.9%	45.4%
Diabetes, % ±SD	16.3%	9.1%	12.5%	43.6%	13.9%
Stroke, % ±SD	5.6%	0%	7.8%	7.8%	4.4%
Gait velocity, cm/s M ± SD	107.6 ± 24.1	102.6 ± 19.7	113.2 ± 22.4	74.6 ± 23.0	111.6 ± 20.1
Presence of WMH, % ±SD	69.5%	69.5%	48.0%	86.6%	72.2%
Presence of Lacunes, % ±SD	13.3%	21.3%	14.1%	8.4%	13.0%
Slow Gait, % ±SD	15.2%	10.3%	12.0%	21.2%	15.8%
Cognitive Complaints
iADLs, % ±SD	22%	4.9%	4.2%	12.9%	31.7%
GDS memory Item, % ±SD	31.8%	4.1%	9.4%	38.6%	41.5%
Any Cognitive Complaint, % ±SD	41.4%	9.0%	11.5%	44.7%	55%
MCR
MCR/iADLs, % ±SD	4.6%	0%	1.6%	6.7%	5.9%
MCR/GDS memory item, % ±SD	4.8%	0%	1.6%	8.4%	5.8%
Any MCR	7.3 ± 0.26	0	2.4 ± 0.15	10.1% ± 0.31	9.4% ± 0.29

White matter hyperintensities

Table 2 shows that global WMH score was found to be associated with MCR (aOR = 1.07, p = 0.015). Global WMH was also associated with secondary outcomes: slower gait speed (β = −0.48 cm/s, p = 0.002) and cognitive complaints (aOR = 1.04, p = 0.018). Regionally, white matter hyperintensity scores in the frontal and basal ganglia regions were also associated with higher risk of MCR (aOR = 1.23, p = .007, aOR = 1.31, p = 0.023, respectively). Frontal, parieto-occipital, and temporal WMH score were associated with slower gait speed (β = −1.48, p < 0.001, β = −0.77 cm/s, p = 0.041, β = −1.57 cm/s, p = 0.021, respectively). WMH score in the temporal lobes, basal ganglia, and infratentorial region were found to be associated with higher risk of cognitive complaints (aOR = 1.15, p = 0.042, aOR = 1.18, p = 0.015, aOR = 1.40, p = 0.005). WMH were not found to be associated with slow gait criteria.

Table 2.

Associations between regional and global WMH score and slow gait, cognitive complaints, and MCR.

	Velocity (β*), (95% CI), p	Slow Gait (aOR*), (95% CI), p	Cognitive Complaints, (aOR), (95% CI), p	MCR (aOR), (95% CI), p
Frontal	−1.48 cm/s, (−2.32, −0.65), <0.001**	1.08, (0.98, 1.22), 0.12	1.07, (0.98, 1.11), 0.16	1.23, (1.06, 1.42), 0.007
Parieto-occipital	−0.77 cm/s, (−1.51, −0.033), 0.041	1.03, (0.94, 1.14), 0.50	1.04, (0.96, 1.13) 0.33	1.11, (0.97, 1.26), 0.12
Temporal	−1.57 cm/s, (−2.91, −0.24), 0.021	0.95, (0.79, 1.14), 0.58	1.15, (1.01, 1.33), 0.042	1.05, (0.84, 1.33), 0.63
Basal Ganglia	−1.05 cm/s, (−2.28, 0.18), 0.096	1.05, (0.90, 1.25), 0.48	1.18, (1.04, 1.35), 0.015	1.31, (1.04, 1.65), 0.023
Infratentorial	0.41 cm/s, (−2.56, 1.74), 0.71	0.96, (0.71, 1.29), 0.78	1.40, (1.11, 1.77), 0.005	1.29, (0.85, 1.97), 0.23
Global	−0.48 cm/s, (−0.79, −0.18) 0.002	1.02, (0.98, 1.06), 0.39	1.04, (1.01, 1.07), 0.018	1.07, (1.01, 1.12), 0.015

*β: regression coefficient; aOR: adjusted odds ratio. Linear and logistic regression models were adjusted for age, sex, cohort (study site), hypertension, diabetes, and stroke history. **Statistical significance at p < 0.05 is indicated in bold.

Lacunes

Table 3 shows that higher lacune score was associated with increased odds of meeting criteria for MCR (aOR = 1.57, p = 0.018). Parietal lacunes were associated with slower gait speed (β = −6.78 cm/s, p = 0.016). Overall lacunar burden was found to be associated with higher risk of cognitive complaints (aOR = 1.21, p = 0.005). Total deep (sum of score in basal ganglia, external and internal capsule, and periventricular) and basal ganglia lacune score were associated with cognitive complaints (aOR = 1.28, p = 0.016, OR = 1.42, p = 0.034).

Table 3.

Associations between regional and global lacune score and normal pace walking (NPW) velocity, slow gait, cognitive complaints, and MCR.

	Velocity β*, (95% CI), p	Slow Gait aOR*, (95% CI), p	Cognitive Complaints aOR, (95% CI), p	MCR aOR, (95% CI), p
Frontal	0.73 cm/s, (−3.32, 4.78), 0.72	0.89, (0.59, 1.59), 0.89	1.38, (0.85, 2.24), 0.20)	1.06, (0.60, 1.88), 0.84
Parietal	−6.78 cm/s, (−12.3, −1.25), 0.016**	1.50, (0.84, 2.67) 0.17	0.96, (0.54, 1.70), 0.89	1.69, (0.93, 3.07) 0.087
Temporal	−0.46 cm/s, (−12.46, 11.56), 0.94	N/A	N/A	N/A
Occipital	−2.84 cm/s, (−11.34, 5.67), 0.51	1.76, (0.69, 4.48), 0.24	1.39, (0.55, 3.50), 0.48	1.39, (0.42, 4.61), 0.59
Total Cortical	−1.60 cm/s, (−4.29, 1.08), 0.24	1.15, (0.84, 1.56), 0.39	1.18, (0.86, 1.61), 0.31	1.24, (0.88, 1.14), 0.22
Basal Ganglia	−1.41 cm/s, (−4.26, 1.44), 0.33	1.28, (0.92, 1.78), 0.14	1.42, (1.03, 1.96) 0.034	1.57, (1.08, 2.27), 0.018
Total Deep	−0.33 cm/s, (−2.03, 1.38), 0.71	1.12, (0.91, 1.39), 0.29	1.28, (1.05, 1.56), 0.016	1.28, (0.99, 1.67), 0.059
Infratentorial	0.26 cm/s, (−7.12, 7.65), 0.94	0.73, (0.24, 2.24), 0.59	0.96, (0.38, 2.46), 0.93	0.54, (0.088, 3.31), 0.51
Overall	−0.83 cm/s, (−1.93, 0.28), 0.14	1.06, (0.92, 1.21), 0.41	1.21, (1.06, 1.38) 0.005	1.16, (0.98, 1.38), 0.077

Age and sex effect modification

Significant two-way interactions between measures of CSVD sex and age were not found in our models.

Confluent-diffuse WMH

The presence of confluent-diffuse WMH (defined as WMH score ≥2 in any region) in any of the five brain regions of interest (frontal, parieto-occipital, temporal, basal ganglia, and/or infratentorial) was found to be associated with MCR and secondary outcomes slow gait and cognitive complaints (30.2% versus 19.2%, p = 0.003, 25.8% versus 19.0%, p = 0.012, 23.1% versus 17.9%, p = 0.008, respectively) by Pearson's chi-squared test. The percentage of participants with confluent-diffuse WMH in the frontal and parieto-occipital regions was also found to be significantly higher among those with MCR, slow gait, and cognitive complaints than without (Table 4).

Table 4.

Presence of severe WMH score (total rating > 10), global and regional confluent-diffuse WMH and slow gait, cognitive complaints, and MCR. SD: standard deviation.

	Slow Gait		p*	Cognitive Complaints		p*	MCR		p*
	Yes	No	p*	Yes	No	p*	Yes	No	p*
N	268	1497		730	1035		129	1643
Presence of Confluent- Diffuse WMH, %±SD	25.8% ± 0.44	19.0% ± 0.39	0.012	23.1% ± 0.42	17.9% ± 0.38	0.008	30.2% ± 0.46	19.2% ± 0.39	0.003
Frontal	16.0% ± 0.37	11.4% ± 0.32	0.03	14.7% ± 0.35	10.2% ± 0.30	0.004	20.2% ± 0.40	11.4% ± 0.32	0.003
Parieto-occipital	19.0% ± 0.39	14.2% ± 0.35	0.04	17.1% ± 0.38	13.4% ± 0.34	0.03	22.5% ± 0.42	14.4% ± 0.35	0.01
Temporal	2.6% ± 0.14	1.9% ± 0.16	0.47	2.3% ± 0.15	1.8% ± 0.13	0.47	3.9% ± 0.19	1.9% ± 0.14	0.12
Basal Ganglia	3.4% ± 0.18	2.6% ± 0.16	0.49	3.0% ± 0.17	2.5% ± 0.16	0.52	3.9% ± 0.19	2.6% ± 0.16	0.40
Infratentorial	0%	0.47% ± 0.07	0.26	0.4% ± 0.06	0.4% ± 0.06	0.94	0%	0.43% ± 0.066	0.46
Overall Severe	7.1% ± 0.26	4.8% ± 0.22	0.13	6.1% ± 0.24	4.5% ± 0.21	0.13	10.1% ± 0.30	4.8% ± 0.21	0.009

*Pearson's chi-square test for categorical variable. **Statistical significance at p < 0.05 is indicated in bold.

Sensitivity analyses

The findings from the weighted regression models generally aligned with those of the unweighted pooled analyses. For MCR, frontal WMH (aOR = 1.21, p = 0.040) and overall WMH burden (OR = 1.07, p = 0.033) remained significantly associated with MCR. Parieto-occipital WMH showed a trend toward significance (OR = 1.15, p = 0.068). While associations with basal ganglia WMH were no longer significantly associated in the weighted models for both MCR and cognitive complaints, basal ganglia lacunes were. Full results can be found in Supplemental Tables 1 and 2.

Discussion

In this cross-sectional, multi-cohort study of individuals older than 60 years of age, we show multiple associations between markers of CSVD, namely white matter hyperintensities (WMH) and lacunes, in distinct brain regions and MCR, as well as its components: slow gait and cognitive complaints. Our findings also extended to associations between these imaging markers and normal pace walking (NPW) velocity. Global WMH score was associated with an increased risk of MCR and regionally, higher WMH score in the frontal lobes were associated with increased risk of MCR. Cortical WMH scores (frontal, parieto-occipital, and temporal cortex) were all found to be associated with slower NPW velocity.

Given that CSVD is a significant contributor to cortical thinning,³³ these findings are in line with previous studies showing that increased WMH burden^23,34,35 or decreased cortical thickness are associated with worsened gait performance.^24,25 Recent work has shown that cortical thickness is lower in prefrontal regions among individuals with MCR,²⁶ again suggesting a link between CSVD-driven atrophy and gait and cognitive impairment. We also show that WMH score in temporal regions is associated with slow gait and cognitive complaints. Fan et al. recently showed an association between global CSVD burden, temporal lobe atrophy and cognitive impairment.²⁷ Interestingly, CSVD lesions (both WMH and lacune score) in the basal ganglia were found to be associated with cognitive complaints and MCR but not slow gait. The basal ganglia are a group of deep grey matter nuclei classically associated with motor function though recent evidence suggests a role in cognition as well.²⁸ In addition, patients with lacunar strokes in the basal ganglia have been shown to develop cognitive impairment.^29,30 Participants with higher WMH burden in frontal and parieto-occipital regions, defined as having regional confluent-diffuse WMH, were found to be more likely to have slow gait, cognitive complaints, and MCR. In addition to frontal lobe atrophy, evidence exists for decreased parietal grey matter volume being associated with worsened gait function possibly due to this brain region's role in visuospatial function and sensory integration.^25,31 Here, we have also shown that lacunes of the parietal region were associated with significantly slower NPW velocity.

Previous studies have also shown that more severe global WMH scores are correlated with worse cognitive and gait performance;^32,33 however, this is among the first to our knowledge showing that these effects may be localized to pathology affecting the cortical white matter in the frontal, parieto-occipital, and temporal regions as well as subcortical regions like the basal ganglia. This may indicate areas underlying MCR pathology. Given the strong association between MCR and incident vascular dementia,³⁶ our study also supports a possible vascular mechanism for MCR. The significant but modest associations, however, suggest other mechanisms must be considered as well.

Key strengths of this study include the large sample size and the broad variability in clinical characteristics across cohorts, allowing robust assessment of associations despite differing prevalence of slow gait, cognitive complaints, and MCR. Additionally, both measures of CSVD in this study, WMH and lacunes, were measured by the same rater for each participant. We also observed variation in the prevalence of cognitive complaints and MCR across cohorts, with the largest cohort (NCGG-SGS) showing higher rates. Possible reasons for this difference include recruitment strategies, cultural differences in subjective assessment of cognition, and baseline health. Models were adjusted by cohort to reduce likelihood of results being skewed. To further account for these differences, we performed a sensitivity analysis using weighted variables for each cohort proportional to their size. Results largely coincided and confirmed the robustness of the main findings: frontal and global WMH scores remained significantly associated with MCR. Associations between basal ganglia WMH and MCR were attenuated after weighting, suggesting these effects were, in part, cohort-driven. Overall, the consistency of findings across both weighted and unweighted models reinforces the conclusion that regional ischemic CSVD burden, particularly within the frontal and parietal networks, contributes to the pathophysiology of MCR and its clinical components.

Our study has several potential limitations. First, its cross-sectional design precludes causal inference. Longitudinal studies are needed to clarify the temporal progression of CSVD lesions and their potential contribution to incident brain atrophy, gait decline, cognitive impairment, and dementia. Second, we did not assess other CSVD markers such as cerebral microbleeds and enlarged perivascular spaces, which may also play a role. Third, certain potential confounders—such as APOE4 status, intracranial volume, physical activity and smoking status—were not included in our models. Finally, some CSVD features (e.g., temporal lobe lacunes and confluent WMH in the temporal, basal ganglia, and infratentorial regions) were rare in our sample and could not be examined in depth.

In conclusion, this study is among the first to establish robust associations between regional CSVD markers and early indicators of neurodegenerative risk—slower gait, cognitive complaints, and MCR—in a large, multi-ethnic cohort of older adults. These findings provide compelling evidence that clinically silent vascular brain injury, detectable through imaging, may precede and predict functional and cognitive decline. By identifying region-specific CSVD patterns linked to these early symptoms, our results lay critical groundwork for advancing dementia risk stratification and targeting early interventions before irreversible damage occurs.

Supplemental Material

sj-docx-1-alz-10.1177_13872877251405448 - Supplemental material for Cerebral small vessel disease unveils a vascular pathway to motoric cognitive risk in aging

Supplemental material, sj-docx-1-alz-10.1177_13872877251405448 for Cerebral small vessel disease unveils a vascular pathway to motoric cognitive risk in aging by Juan P Vazquez, Gilles Allali, Olivier Beauchet, Michele Callisaya, Takehiko Doi, VG Pradeep Kumar, Sofiya Milman, Hiroyuki Shimada, Velandai Srikanth, Joe Verghese and Helena M Blumen in Journal of Alzheimer's Disease

Footnotes

Acknowledgements

The authors would like to acknowledge the following people for their contributions to the work in this manuscript: Dachel Sanchez-Castellanos and Shanika Oshadi Jayakody.

ORCID iDs

Juan P Vazquez

Helena M Blumen

Ethical considerations

This study protocol was reviewed and approved by an institutional review board at Albert Einstein College of Medicine, NY, approval number 2007-272.

Consent to participate

Written informed consent to participate and to utilize data gathered from participation for publication was obtained from all participants prior to study enrollment.

Author contribution(s)

Juan P Vazquez: Conceptualization; Data curation; Formal analysis; Writing – original draft; Writing – review & editing.

Gilles Allali: Data curation; Formal analysis; Investigation; Supervision; Writing – review & editing.

Olivier Beauchet: Data curation; Writing – review & editing.

Michele Callisaya: Data curation; Writing – review & editing.

Takehiko Doi: Data curation; Writing – review & editing.

VG Pradeep Kumar: Data curation; Writing – review & editing.

Sofiya Milman: Conceptualization; Data curation; Funding acquisition; Project administration; Writing – review & editing.

Hiroyuki Shimada: Data curation; Writing – review & editing.

Velandai Srikanth: Data curation; Writing – review & editing.

Joe Verghese: Conceptualization; Funding acquisition; Methodology; Project administration; Writing – review & editing.

Helena M Blumen: Funding acquisition; Investigation; Methodology; Supervision; Validation; Visualization; Writing – review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported, in part, by grants from National Institute of Health (NIH) R01AG062659-01A1, NIH/National Institute on Aging (PI: Helena M. Blumen), R01AG057548-01A1 (PI: Joe Verghese), R01AG061155-01 (PI: Sofiya Milman).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

All data generated or analyzed during this study are included in this article. Further enquiries can be directed at the corresponding author.

Supplemental material

Supplemental material for this article is available online.

References

Verghese

Wang

Lipton

, et al. Motoric cognitive risk syndrome and the risk of dementia. J Gerontol A Biol Sci Med Sci 2013; 68: 412–418.

Verghese

Annweiler

Ayers

, et al. Motoric cognitive risk syndrome. Neurology 2014; 83: 718–726.

Callisaya

Ayers

Barzilai

, et al. Motoric cognitive risk syndrome and falls risk: a multi-center study. J Alzheimers Dis 2016; 53: 1043–1052.

Doi

Shimada

Makizako

, et al. Motoric cognitive risk syndrome: association with incident dementia and disability. J Alzheimers Dis 2017; 59: 77–84.

Chhetri

Chan

Vellas

, et al. Motoric cognitive risk syndrome: predictor of dementia and age-related negative outcomes. Front Med (Lausanne) 2017; 4: 166.

Killane

Donoghue

Savva

, et al. Relative association of processing speed, short-term memory and sustained attention with task on gait speed: a study of community-dwelling people 50 years and older. J Gerontol Ser A 2014; 69: 1407–1414.

Peel

Alapatt

Jones

, et al. The association between gait speed and cognitive status in community-dwelling older people: a systematic review and meta-analysis. J Gerontol Ser A 2019; 74: 943–948.

Toots

ATM

Taylor

Lord

, et al. Associations between gait speed and cognitive domains in older people with cognitive impairment. J Alzheimers Dis 2019; 71: S15–S21.

Jayakody

Blumen

Ayers

, et al. Longitudinal associations between falls and risk of gait decline: results from the Central Control of Mobility and Aging Study. Arch Phys Med Rehabil 2023; 104: 245–250.

10.

Adam

Fitzpatrick

Leary

, et al. Change in gait speed and fall risk among community-dwelling older adults with and without mild cognitive impairment: a retrospective cohort analysis. BMC Geriatr 2023; 23: 328.

11.

Artaud

Singh-Manoux

Dugravot

, et al. Decline in fast gait speed as a predictor of disability in older adults. J Am Geriatr Soc 2015; 63: 1129–1136.

12.

Blumen

Brown

Habeck

, et al. Gray matter volume covariance patterns associated with gait speed in older adults: a multi-cohort MRI study. Brain Imaging Behav 2019; 13: 446–460.

13.

Oshadi

Breslin

Beare

, et al. The associations between grey matter volume covariance patterns and gait variability—the tasmanian study of cognition and gait. Brain Topogr 2021; 34: 478–488.

14.

Darweesh

SKL

Licher

Wolters

, et al. Quantitative gait, cognitive decline, and incident dementia: the rotterdam study. Alzheimers Dement 2019; 15: 1264–1273.

15.

Dumurgier

Artaud

Touraine

, et al. Gait speed and decline in gait speed as predictors of incident dementia. J Gerontol A Biol Sci Med Sci 2017; 72: 655–661.

16.

Skillbäck

Blennow

Zetterberg

, et al. Slowing gait speed precedes cognitive decline by several years. Alzheimers Dement 2022; 18: 1667–1676.

17.

Verghese

Wang

Lipton

, et al. Quantitative gait dysfunction and risk of cognitive decline and dementia. J Neurol Neurosurg Psychiatry 2007; 78: 929–935.

18.

Jessen

Wiese

Bachmann

, et al. Prediction of dementia by subjective memory impairment: effects of severity and temporal association with cognitive impairment. Arch Gen Psychiatry 2010; 67: 414–422.

19.

Jonker

Geerlings

Schmand

. Are memory complaints predictive for dementia? A review of clinical and population-based studies. Int J Geriatr Psychiatry 2000; 15: 983–991.

20.

Cannistraro

Badi

Eidelman

, et al. CNS Small vessel disease. Neurology 2019; 92: 1146–1156.

21.

Pantoni

. Cerebral small vessel disease: from pathogenesis and clinical characteristics to therapeutic challenges. Lancet Neurol 2010; 9: 689–701.

22.

Østergaard

Engedal

Moreton

, et al. Cerebral small vessel disease: capillary pathways to stroke and cognitive decline. J Cereb Blood Flow Metab 2016; 36: 302–325.

23.

Wang

Allali

Kesavadas

, et al. Cerebral small vessel disease and motoric cognitive risk syndrome: results from the Kerala-Einstein Study. J Alzheimers Dis 2016; 50: 699–707.

24.

Doi

Nakakubo

Tsutsumimoto

, et al. The association of white matter hyperintensities with motoric cognitive risk syndrome. Cereb Circ Cogn Behav 2022; 3: 100150.

25.

Zanon Zotin

Sveikata

Viswanathan

, et al. Cerebral small vessel disease and vascular cognitive impairment: from diagnosis to management. Curr Opin Neurol 2021; 34: 246–257.

26.

Gubbi

Schwartz

Crandall

, et al. Effect of exceptional parental longevity and lifestyle factors on prevalence of cardiovascular disease in offspring. Am J Cardiol 2017; 120: 2170–2175.

27.

Srikanth

Phan

Chen

, et al. The location of white matter lesions and gait—A voxel-based study. Ann Neurol 2010; 67: 265–269.

28.

Callisaya

Beare

Phan

, et al. Global and regional associations of smaller cerebral gray and white matter volumes with gait in older people. PLoS One 2014; 9: e84909.

29.

Doi

Blumen

Verghese

, et al. Gray matter volume and dual-task gait performance in mild cognitive impairment. Brain Imaging Behav 2017; 11: 887–898.

30.

Shimada

Tsutsumimoto

Lee

, et al. Driving continuity in cognitively impaired older drivers. Geriatr Gerontol Int 2016; 16: 508–514.

31.

Verghese

Noone

Johnson

, et al. Picture-based memory impairment screen for dementia. J Am Geriatr Soc 2012; 60: 2116–2120.

32.

Wahlund

Barkhof

Fazekas

, et al. A new rating scale for age-related white matter changes applicable to MRI and CT. Stroke 2001; 32: 1318–1322.

33.

Wardlaw

Smith

Dichgans

. Small vessel disease: mechanisms and clinical implications. Lancet Neurol 2019; 18: 684–696.

34.

Rosano

Brach

Longstreth

Jr ., et al. Quantitative measures of gait characteristics indicate prevalence of underlying subclinical structural brain abnormalities in high-functioning older adults. Neuroepidemiology 2005; 26: 52–60.

35.

Verghese

LeValley

Hall

, et al. Epidemiology of gait disorders in community-residing older adults. J Am Geriatr Soc 2006; 54: 255–261.

36.

Staals

Booth

Morris

, et al. Total MRI load of cerebral small vessel disease and cognitive ability in older people. Neurobiol Aging 2015; 36: 2806–2811.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB