Key Points
Question
How common is pure vascular cognitive impairment (VCI) in the absence of Alzheimer disease and other neurodegenerative pathologies?
Findings
In this cohort study including postmortem data from 1767 participants, the frequency of participants without significant Alzheimer disease and other neurodegenerative pathologies was 20.9% (369 of 1767). Macroinfarcts, not Alzheimer disease or other neurodegenerative pathologies, were the main pathologies associated with cognitive impairment in these participants, an indication of pure VCI.
Meaning
In this study, pure VCI was not rare.
This cohort study identifies participants without Alzheimer disease and other neurodegenerative pathologies and determines the extent to which cerebrovascular disease pathologies were associated with cognitive impairment.
Abstract
Importance
It is not clear how common pure vascular cognitive impairment (VCI) is in the absence of Alzheimer disease (AD) and/or other neurodegenerative pathologies.
Objective
To identify participants without AD and other neurodegenerative pathologies and determine the extent to which cerebrovascular disease pathologies were associated with cognitive impairment.
Design, Setting, and Participants
This clinical pathological study included participants from 2 ongoing community-based cohorts that began enrollment in 1994 and 1997. Prior to death, participants were observed for a mean (SD) of 8.4 (5.3) years with annual assessments. From 2096 participants who died, 1799 (85.8%) underwent autopsy and 1767 had complete postmortem pathological examination data at the time of data analyses. To identify participants without neurodegenerative pathologies, we categorized them in 3 subgroups. A vascular subgroup was composed of participants without significant levels of neurodegenerative brain pathologies. A neurodegenerative subgroup was composed of participants without significant levels of cerebrovascular disease pathologies. A mixed subgroup was composed of the rest of the participants. Data were analyzed from May 2021 to July 2022.
Exposures
Brain pathology indices obtained by postmortem pathological assessments.
Main Outcomes and Measures
The primary outcome was cognitive impairment defined by presence of mild cognitive impairment or dementia. The secondary outcome was cognition assessed by 19 neuropsychological tests.
Results
Of 1767 included participants, 1189 (67.3%) were women, and the mean (SD) age at death was 89.4 (6.6) years. In the vascular subgroup (n = 369), cognitive impairment was present in 156 participants (42.3%) and was associated with cerebrovascular disease pathologies (macroinfarcts: odds ratio [OR], 2.05; 95% CI, 1.49-2.82; P < .001; arteriolosclerosis in basal ganglia: OR, 1.35; 95% CI, 1.04-1.76; P = .03) but not AD or other neurodegenerative pathologies, an indication of pure VCI. In mixed-effects models including all the pathologies, only macroinfarcts were associated with a faster cognitive decline rate (estimate, −0.019; SE, 0.005; P < .001) in the vascular subgroup. Further analyses identified macroinfarcts in the frontal white matter to be associated with faster cognitive decline rate when macroinfarcts of cortical and subcortical brain regions were examined in a single model.
Conclusions and Relevance
In this study, pure VCI was not rare. Macroinfarcts, specifically in frontal white matter, were the main cerebrovascular disease pathologies associated with cognitive decline in pure VCI.
Introduction
From the 1970s to 1990s, multiinfarct dementia1 and vascular dementia2 entities were developed to delineate dementia caused by vascular origins from dementia caused by Alzheimer disease (AD), and tools such as the Ischemic Scale3 were developed for this purpose. However, in the last 2 decades, doubts were raised about the existence of pure vascular dementia in individuals with sporadic cases4,5 because of the overwhelming effects of AD pathologies on cognitive impairment.6,7 Consequently, vascular cognitive impairment (VCI) was developed as a nosology indicating any contribution made by vascular causes to cognitive impairment.8 However, it remains unclear how common pure VCI is in the absence of AD and/or other neurodegenerative pathologies.
We previously examined contributions of neurodegenerative and cerebrovascular disease pathologies to cognitive decline and dementia by estimating additive effects of cerebrovascular disease to neurodegenerative pathologies on odds of dementia as well as rate of cognitive decline over multiple years prior to death.9,10 Here, we identify individuals without neurodegenerative pathologies and determine the extent to which cerebrovascular disease pathologies are associated with cognitive impairment.
Methods
The study participants were enrolled in either the Religious Orders Study (ROS) or the Rush Memory and Aging Project (MAP). The ROS began enrollment in 1994 and enrolls nuns, priests, sisters, and brothers. The MAP began enrollment in 1997 and recruits older persons from northeastern Illinois living in homes, retirement facilities, and subsidized housing. Inclusion criteria for both studies were older age (65 years and older), without known dementia, and agreeing to annual clinical evaluation and postmortem organ donation. Both studies used a large subset of harmonized protocols identical at the item level that were administered by the same trained personnel facilitating joint analyses. More details about the studies’ designs and components are provided elsewhere.11 Written informed consent was obtained from each participant as well as an Anatomic Gift Act. The Institutional Review Board of Rush University Medical Center approved each study. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
From 2096 participants who died, 1799 (85.8%) underwent autopsy, and 1767 had complete postmortem pathological examination data at the time of data analyses. To identify participants without neurodegenerative pathologies, we categorized participants in 3 subgroups. A vascular subgroup, the target group, was composed of 369 participants without significant levels of neurodegenerative brain pathologies, including intermediate to high likelihood of AD pathological diagnosis, extension of transactive response DNA binding protein 43 (TDP-43) proteinopathy to the hippocampus or neocortex, hippocampal sclerosis, and neocortical Lewy bodies. A neurodegenerative subgroup was composed of 407 participants without significant levels of cerebrovascular disease pathologies, including macroinfarcts, microinfarcts, and moderate to severe atherosclerosis or arteriolosclerosis. Participants with moderate to severe cerebral amyloid angiopathy (CAA) were not excluded because of association of CAA with AD pathology.12 A mixed subgroup was composed of the rest of the participants (n = 991).
Clinical Evaluation
Annually, 21 neuropsychological tests were administered. A neuropsychologist and a neurologist expert in dementia reviewed cognitive and clinical findings without access to neuropathologic data and adjudicated presence of mild cognitive impairment (MCI) and dementia according to established criteria.13,14 In this study, the main outcome was presence of cognitive impairment that was defined by the presence of either MCI or dementia. As secondary outcomes, neuropsychological tests’ scores were standardized and averaged (eMethods in the Supplement) to make global cognition (19 tests) and 5 specific cognitive functions: episodic memory (7 tests), perceptual speed (4 tests), semantic memory (3 tests), working memory (3 tests), and visuospatial ability (2 tests).15,16
Postmortem Brain Pathological Assessment
The median (IQR) time interval between death and autopsy was 6.9 (5.3-9.4) hours. As described previously,17 one hemisphere was frozen to be used for generation of multiomics atlas. The other hemisphere was fixed in 4% paraformaldehyde solution and was used for pathological assessments.
Cerebrovascular Disease Pathologies
Macroinfarcts
Slabs of fixed hemisphere and photographs of frozen hemisphere were observed without microscope for the presence of macroinfarcts, and suspected lesions were confirmed microscopically. The main objective of this study was examining the association of cerebrovascular disease pathologies with cognitive decline that occurred over years prior to death. Therefore, we included only chronic macroinfarcts that usually present as cavities with a few remaining macrophages surrounded by fibrillary gliosis,18 as done in our prior publications.17,19 We summarized chronic macroinfarcts as a semiquantitative variable with 3 levels (no, 1, and 2 or more macroinfarcts) and with a dichotomous (present vs absent) variable for descriptive purposes.
Furthermore, we categorized macroinfarcts by their location and size. Location categories included cortical and white matter regions of cerebral lobes, thalamus, caudate, lentiform nucleus, hippocampus, brainstem, and cerebellum. The categorization was not exclusive, and a macroinfarct could affect more than 1 brain region. At neuropathological examination, size of the macroinfarct was documented in x, y, z planes. For categorization by size, we used maximum diameter across the x, y, z planes. The median (IQR; range) maximum diameter of macroinfarcts was 7.5 (4.0-15.0; 1.0-132.0) mm. We categorized macroinfarcts by their maximum diameter into lacunar infarcts (10 mm or smaller) and nonlacunar infarcts (larger than 10 mm).
Microinfarcts
Microinfarcts are visible only under microscope. Besides cavities with perilesional fibrillary gliosis, chronic microinfarcts may appear as scarlike lesions with no cavity and with possible pial surface invagination.18 Tissue sections from 9 brain regions were examined for the presence of microinfarcts: 6 cortical brain regions (frontal, temporal, entorhinal, hippocampal, parietal, and anterior cingulate), 2 subcortical regions (anterior basal ganglia and thalamus), and the midbrain. Compared with macroinfarcts, chronic microinfarcts were generally much smaller, with a median (range) diameter of 246 (85-725) μm.20 For this study, chronic microinfarcts were summarized as a 3-level variable (0, 1, or 2 or more) or a dichotomous variable (present vs absent).
Atherosclerosis
Large vessels of the Circle of Willis and their proximal branches were examined for atherosclerosis. Presence and severity of atherosclerosis were scored based on the severity of atherosclerosis in each vessel and the number of vessels involved.17 The scores were summarized as a semiquantitative variable (none, mild, moderate, or severe). For descriptive purposes, we used a dichotomous variable indicating presence of moderate to severe atherosclerosis.
Arteriolosclerosis
Small arterioles in the basal ganglia were examined for presence and severity of arteriolosclerosis, including intimal injuries, smooth muscles degeneration, and concentric hyaline thickening. Severity of arteriolosclerosis was summarized using a semiquantitative scale (none, mild, moderate, severe) based on the thickness of the vessel wall and lumen narrowing.17 For descriptive purposes, we summarized presence of moderate to severe arteriolosclerosis as a dichotomous variable.
In a subset of participants, arteriolosclerosis was also assessed in anterior white matter deep to the middle frontal cortex and in white matter deep to the posterior parietal cortex.21
CAA
Parenchymal and meningeal vessels were examined by immunohistochemical methods using an antibody against amyloid-β. CAA assessment was summarized with a semiquantitative scale (none, mild, moderate, or severe) and a dichotomous variable indicating presence of moderate to severe CAA.
Neurodegenerative Pathologies
Pathological indices of AD, TDP-43, Lewy bodies, and hippocampal sclerosis were obtained, which are described elsewhere17 and in the eMethods in the Supplement.
Statistical Analysis
To compare participants in the 3 pathology subgroups, χ2, t, or Kruskal-Wallis tests were used. We used logistic regression models controlled for age, sex, and education (defined with other covariates; eMethods in the Supplement) to examine association of brain pathologies with odds of cognitive impairment prior to death. We used linear mixed-effects models to examine cognitive decline in the participants. The core model included time (rate of cognitive change), intercept (level of cognition prior to death), age, sex, education, and their interaction with time. The model also included random-effects terms for intercept and time, allowing person-specific level of cognition and rate of cognitive decline. Then, we added terms for the brain pathologies and their interaction with time to examine whether brain pathologies were associated with the level of cognition and rate of cognitive decline. Finally, we used the estimates of the variance of the person-specific rates of cognitive decline to calculate the percentage of variance in the rates of cognitive decline explained by brain pathologies. Two-tailed P values less than .05 were used for rejection of null hypotheses. The analyses were done using SAS version 9.4 (SAS Institute).
Results
Of 1767 included participants, 1189 (67.3%) were women, and the mean (SD) age at death was 89.4 (6.6) years; characteristics classified by the pathology subgroups are summarized in Table 1. Most of the participants (991 [56.1%]) were in the mixed subgroup, and approximately equal numbers composed the vascular subgroup (369 [20.9%]) and the neurodegenerative subgroup (407 [23.0%]). Cognitive impairment was present in 1201 participants (68.0%) prior to death, which was the most in the mixed subgroup (770 [77.7%]) followed by the neurodegenerative (275 [67.6%]) and the vascular (156 [42.3%]) subgroups (χ22 = 155.8; P < .001). Vascular diseases, including stroke, were more frequent in the mixed (244 [24.6%]) and vascular (78 [21.1%]) subgroups compared with the neurodegenerative subgroup (39 [9.6%]; χ22 = 40.0; P < .001).
Table 1. Characteristics of Study Participantsa.
| Characteristic | No. (%) | |||
|---|---|---|---|---|
| Vascular subgroup (n = 369) | Neurodegenerative subgroup (n = 407) | Mixed subgroup (n = 991) | Total (N = 1767) | |
| Age at death, mean (SD), y | 87.0 (6.9) | 88.5 (6.6) | 90.7 (6.2) | 89.4 (6.6)b |
| Age at the last visit, mean (SD), y | 86.2 (7.0) | 87.5 (6.7) | 89.9 (6.3) | 88.6 (6.7)b |
| Sex | ||||
| Female | 220 (60) | 270 (66) | 699 (71) | 1189 (67)b |
| Male | 149 (40) | 137 (34) | 292 (29) | 578 (33)b |
| Years of education, mean (SD) | 16.4 (3.6) | 16.8 (3.6) | 16.1 (3.6) | 16.3 (3.6)c |
| Clinical characteristics at the last visit | ||||
| Hypertension | 265 (72) | 248 (61) | 664 (67) | 1177 (67)c |
| Diabetes | 95 (26) | 92 (23) | 194 (20) | 381 (22)d |
| Smoking history | 122 (33) | 138 (34) | 282 (29) | 542 (31) |
| No. of vascular risk factors, median (IQR)e | 1.0 (1.0 to 2.0) | 1.0 (0 to 2.0) | 1.0 (0 to 2.0) | 1.0 (0 to 2.0)d |
| History of stroke | 78 (21) | 39 (10) | 244 (25) | 361 (21)b |
| History of myocardial infarction | 92 (25) | 62 (15) | 206 (21) | 360 (20)c |
| History of lower extremities claudication | 104 (28) | 101 (25) | 298 (30) | 503 (29) |
| No. of vascular diseases, median (IQR)f | 1.0 (0 to 1.0) | 0 (0 to 1.0) | 1.0 (0 to 1.0) | 1.0 (0 to 1.0)b |
| Cognition status | ||||
| No cognitive impairment | 213 (58) | 132 (32) | 220 (22) | 565 (32)b |
| Mild cognitive impairment | 95 (26) | 102 (25) | 210 (21) | 407 (23)b |
| Dementia | 61 (17) | 173 (43) | 560 (57) | 794 (45)b |
| Mini-Mental State Examination score, mean (SD) | 25.4 (6.0) | 21.1 (9.3) | 18.6 (9.7) | 20.6 (9.4)b |
| Global cognition, mean (SD)g | −0.30 (0.86) | −0.97 (1.22) | −1.31 (1.20) | −1.02 (1.21)b |
| Episodic memory, mean (SD)g | −0.05 (1.00) | −0.98 (1.45) | −1.32 (1.39) | −0.98 (1.42)b |
| Perceptual speed, mean (SD)g | −0.74 (1.01) | −1.07 (1.08) | −1.42 (1.06) | −1.20 (1.09)b |
| Working memory, mean (SD)g | −0.35 (0.90) | −0.70 (1.13) | −0.97 (1.14) | −0.78 (1.12)b |
| Semantic memory, mean (SD)g | −0.27 (0.93) | −0.86 (1.48) | −1.17 (1.46) | −0.91 (1.41)b |
| Visuospatial ability, mean (SD)g | −0.28 (0.97) | −0.61 (1.09) | −0.81 (1.08) | −0.65 (1.08)b |
| Years of cognitive assessment prior to death, mean (SD) | 8.4 (5.0) | 8.3 (5.6) | 8.5 (5.3) | 8.4 (5.3) |
| Postmortem indices of brain pathologies | ||||
| Macroinfarcts (≥1) | 128 (35) | 0 | 503 (51) | 631 (36)b |
| Microinfarcts (≥1) | 103 (28) | 0 | 449 (45) | 552 (31)b |
| Atherosclerosis | ||||
| None | 75 (20) | 128 (31) | 109 (11) | 312 (18)b |
| Mild | 175 (47) | 279 (69) | 407 (41) | 861 (49)b |
| Moderate | 100 (27) | 0 | 384 (39) | 484 (28)b |
| Severe | 19 (5) | 0 | 83 (8) | 102 (6)b |
| Arteriolosclerosis in basal ganglia | ||||
| None | 136 (37) | 235 (58) | 243 (25) | 614 (35)b |
| Mild | 125 (34) | 172 (42) | 285 (29) | 582 (33)b |
| Moderate | 78 (21) | 0 | 342 (35) | 420 (24)b |
| Severe | 30 (8) | 0 | 109 (11) | 139 (8)b |
| Cerebral amyloid angiopathy | ||||
| None | 158 (43) | 68 (17) | 156 (16) | 382 (22)b |
| Mild | 145 (39) | 169 (43) | 403 (42) | 717 (42)b |
| Moderate | 41 (11) | 109 (27) | 242 (25) | 392 (23)b |
| Severe | 25 (7) | 51 (13) | 159 (17) | 235 (14)b |
| Intermediate to high likelihood of AD pathology | 0 | 333 (82) | 802 (81) | 1135 (64)b |
| Global AD pathology score, mean (SD) | 0.37 (0.21) | 0.88 (0.37) | 0.88 (0.36) | 0.77 (0.40)b |
| TDP-43 stages | ||||
| None | 278 (77) | 160 (42) | 358 (39) | 796 (48)b |
| Stage 1 (amygdala) | 85 (23) | 68 (18) | 150 (16) | 303 (18)b |
| Stage 2 (hippocampus) | 0 | 51 (13) | 113 (12) | 164 (10)b |
| Stage 3 (neocortex) | 0 | 100 (26) | 291 (32) | 391 (24)b |
| Lewy bodies | ||||
| None | 320 (87) | 273 (67) | 718 (72) | 1311 (74)b |
| Other (nigral, limbic) | 49 (13) | 51 (13) | 116 (12) | 216 (12)b |
| Neocortex | 0 | 83 (20) | 157 (16) | 240 (14)b |
| Hippocampal sclerosis | 0 | 36 (9) | 120 (12) | 156 (9)b |
Abbreviation: AD, Alzheimer disease.
To compare cognitively impaired and unimpaired subgroups in the characteristics, χ2 (for categorical characteristics) and analysis of variance (for continuous characteristics) were used. Number of vascular risk factors and diseases were compared using the Kruskal-Wallis test.
P < .001.
P < .01.
P < .05.
Number of vascular risk factors ranged from 0 to 3 and included hypertension, diabetes, and smoking.
Number of vascular diseases ranged from 0 to 3 and included stroke, myocardial infarction, and claudication of lower extremities.
Global cognition and the 5 cognitive domain scores were derived from standardized scores of 19 neuropsychological tests: immediate and delayed recall of the East Boston Story and Logical Memory Story A, Word List Memory, and Recall, and Recognition (episodic memory); Symbol Digit Modalities Test, Number Comparison, and Stroop Color and Word Test (perceptual speed); Boston Naming Test, Verbal Fluency, and Word Reading (semantic memory); Digit Span Forward and Backward, and Digit Ordering (working memory); and Judgement of Line Orientation and Standard Progressive Matrices (visuospatial ability).
Brain Pathologies and Cognitive Impairment
In separate logistic regression models, we examined the association of the brain pathologies with cognitive impairment in each of the pathology subgroups. In the vascular subgroup, cerebrovascular disease pathologies, including macroinfarcts, atherosclerosis, and arteriolosclerosis in basal ganglia, were associated with higher odds of cognitive impairment when each pathology was separately examined in one model (Table 2; Figure 1). Of note, low levels of neurodegenerative pathologies, including AD, were not associated with cognitive impairment. Because cerebrovascular disease pathologies were correlated, we examined whether they were independently associated with cognitive impairment. In a logistic regression model including all the pathologies, only macroinfarcts (odds ratio [OR], 2.05; 95% CI, 1.49-2.82; P < .001) and arteriolosclerosis in basal ganglia (OR, 1.35; 95% CI, 1.04-1.76; P = .03) were associated with higher odds of cognitive impairment (Table 2). These findings indicated that neurodegenerative diseases were not significant contributor to cognitive impairment in the vascular subgroup and supported presence of pure VCI.
Table 2. Association of Brain Pathologies With Cognitive Impairment in 3 Pathology Subgroups.
| Pathology index | Vascular subgroup | Neurodegenerative subgroup | Mixed subgroup | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Each pathology in a separate model | All the pathologies in 1 model | Each pathology in a separate model | All the pathologies in 1 model | Each pathology in a separate model | All the pathologies in 1 model | |||||||
| OR (95% CI)a | P value | OR (95% CI)a | P value | OR (95% CI)a | P value | OR (95% CI)a | P value | OR (95% CI)a | P value | OR (95% CI)a | P value | |
| Macroinfarcts | 1.98 (1.50-2.63) | <.001 | 2.05 (1.49-2.82) | <.001 | NA | NA | NA | NA | 1.23 (1.01-1.49) | .04 | 1.19 (0.95-1.49) | .13 |
| Atherosclerosis | 1.67 (1.27-2.20) | <.001 | 1.29 (0.94-1.76) | .12 | 0.93 (0.58-1.46) | .74 | 0.86 (0.51-1.46) | .57 | 1.32 (1.08-1.60) | .006 | 1.36 (1.08-1.72) | .009 |
| Arteriolosclerosis in basal ganglia | 1.55 (1.24-1.95) | <.001 | 1.35 (1.04-1.76) | .03 | 1.68 (1.08-2.63) | .02 | 1.83 (1.09-3.06) | .02 | 1.27 (1.08-1.49) | .004 | 1.18 (0.97-1.43) | .10 |
| Microinfarcts | 0.96 (0.72-1.28) | .79 | 0.80 (0.57-1.11) | .18 | NA | NA | NA | NA | 1.16 (0.94-1.43) | .16 | 1.24 (0.97-1.60) | .09 |
| Cerebral amyloid angiopathy | 0.91 (0.72-1.16) | .44 | 0.87 (0.66-1.13) | .30 | 1.13 (0.89-1.44) | .32 | 0.84 (0.62-1.12) | .24 | 1.32 (1.12-1.57) | .001 | 1.13 (0.94-1.38) | .20 |
| Global AD pathology score | 0.81 (0.30-2.15) | .67 | 0.91 (0.31-2.68) | .87 | 7.14 (3.76-13.58) | <.001 | 8.90 (4.06-19.50) | <.001 | 4.64 (2.99-7.21) | <.001 | 6.39 (3.62-11.28) | <.001 |
| TDP-43 | 1.12 (0.68-1.85) | .66 | 1.12 (0.65-1.93) | .68 | 1.53 (1.26-1.87) | <.001 | 1.50 (1.19-1.88) | <.001 | 1.41 (1.23-1.61) | <.001 | 1.33 (1.14-1.55) | <.001 |
| Lewy bodies | 1.61 (0.87-2.97) | .13 | 1.51 (0.77-2.95) | .23 | 1.78 (1.11-2.86) | .02 | 1.43 (0.82-2.48) | .20 | 2.18 (1.47-3.23) | <.001 | 2.24 (1.44-3.47) | <.001 |
| Hippocampal sclerosis | NA | NA | NA | NA | 3.65 (1.25-10.65) | .02 | 2.13 (0.65-6.93) | .21 | 4.87 (2.22-10.67) | <.001 | 4.95 (2.10-11.70) | <.001 |
Abbreviations: AD, Alzheimer disease; NA, not applicable; OR, odds ratio.
Estimates are derived from logistic regression models with cognitive impairment as the outcome, controlled for age at death, sex, and education.
Figure 1. Association of Macroinfarcts, Atherosclerosis, and Arteriolosclerosis in Basal Ganglia With Pure Vascular Cognitive Impairment.
Each plot illustrates the association of age at death with probability of cognitive impairment of average women in this study (age 87 years and with 16 years of education) with different levels of macroinfarcts (A), atherosclerosis (B), and arteriolosclerosis at basal ganglia (C).
We also examined associations of brain pathologies with cognitive impairment in the neurodegenerative and mixed subgroups. In the neurodegenerative subgroup, besides AD and other neurodegenerative pathologies, low levels of arteriolosclerosis in basal ganglia were also associated with cognitive impairment (Table 2), which might be because of the association of arteriolosclerosis with neurodegenerative pathologies.22,23 In the mixed subgroup, both neurodegenerative and cerebrovascular disease pathologies were associated with cognitive impairment (Table 2).
Next, we examined associations of the brain pathologies with odds of dementia in the 3 subgroups. In the vascular subgroup, cerebrovascular disease pathologies, including macroinfarcts, atherosclerosis, and arteriolosclerosis in basal ganglia, were associated with higher odds of dementia when examined in separate models (eTable 1 in the Supplement). In a model including all the pathologies, only macroinfarcts (OR, 2.06; 95% CI, 1.42-2.98; P < .001) remained associated with dementia (eTable 1 in the Supplement), an indication of vascular dementia. Neurodegenerative and both neurodegenerative and cerebrovascular disease pathologies were associated with dementia in the neurodegenerative and mixed subgroups, respectively (eTable 1 in the Supplement).
Brain Pathologies and Cognitive Decline
Participants were observed for a mean (SD) of 8.4 (5.3) years prior to death, which enabled examining cognitive decline rate and the associated pathologies in the 3 pathology subgroups. In a mixed-effects model including terms for the subgroups, cognitive decline was fastest in the mixed subgroup, followed by the neurodegenerative and vascular subgroups (eFigure and eTable 2 in the Supplement).
We explored pathologies associated with global cognition decline rate in each pathology subgroup. In the vascular subgroup, cerebrovascular disease pathologies, including macroinfarcts, atherosclerosis, and arteriolosclerosis in basal ganglia, were associated with a faster cognitive decline rate when examined in separate models (Figure 2; eTable 3 in the Supplement). The neurodegenerative pathologies, including AD, were not associated with rate of cognitive decline. We also examined all the pathologies together in a mixed-effects model. The analysis indicated that only macroinfarcts (estimate, −0.019; SE, 0.005; P < .001) were associated with a faster cognitive decline rate (eTable 4 in the Supplement) and could explain 10.0% of the variance of cognitive decline rate. Neurodegenerative pathologies alone or together with cerebrovascular disease pathologies were associated with faster cognitive decline in the neurodegenerative and mixed subgroups, respectively (eTables 3 and 4 in the Supplement).
Figure 2. Association of Macroinfarcts With Pure Vascular Cognitive Decline Prior to Death.
Each plot illustrates cognitive decline trajectories of 3 average women in this study (age 87 years and with 16 years of education) with none, 1, and 2 or more macroinfarcts. Global cognition and cognitive domains scores are composite measures derived from standardized scores of 19 neuropsychological tests described in the eMethods in the Supplement.
Macroinfarcts and Pure VCI
The above findings indicated that macroinfarcts were the main pathologies associated with cognitive impairment and decline in the vascular subgroup. In sensitivity analyses, we further examined whether the association of macroinfarcts with cognitive impairment was confounded by additional covariates. In subsequent models, we added terms for vascular risk factors score, body mass index, physical activity, and history of stroke. Addition of these covariates to the core model examining association of macroinfarcts with cognitive impairment did not change this association (eTable 5 in the Supplement). Furthermore, examining the core model in a subgroup of 288 participants without history of stroke showed that macroinfarcts were associated with higher odds of cognitive impairment (OR, 1.84; 95% CI, 1.29-2.62; P < .001), indicating that the association of macroinfarcts with cognitive impairment was independent of history of stroke.
Global cognition is composed of cognitive functions. We examined whether macroinfarcts were associated with decline rate in specific cognitive functions. In 5 separate mixed-effects models, we replaced global cognition with 5 cognitive functions. Macroinfarcts were associated with a faster decline rate in all the cognitive functions except visuospatial ability (Figure 2; eTable 6 in the Supplement).
Imaging studies have suggested that infarcts at some strategic brain locations are more strongly associated with cognitive impairment.24 Therefore, we examined macroinfarcts at different brain regions in relation to cognitive impairment. Examining brain regions in separate logistic regressions found that macroinfarcts at white matters of frontal, parietal, and temporal lobes together with macroinfarcts at lentiform nucleus and thalamus were associated with cognitive impairment (Table 3). However, macroinfarcts only at frontal and parietal white matter were associated with cognitive impairment when all regions were included in a single model (Table 3). When all brain regions were examined in one mixed-effects model, frontal white matter was the only brain region at which presence of macroinfarcts was associated with faster cognitive decline (eTable 7 in the Supplement). These findings indicate that subcortical macroinfarcts, specifically at frontal white matter, were the main pathologies associated with pure VCI.
Table 3. Association of Macroinfarcts in Different Brain Regions With Cognitive Impairment in 369 Participants in the Vascular Subgroup.
| Brain regiona | Participants, No. (%) | Association with cognitive impairmentb | |||
|---|---|---|---|---|---|
| Each region in a separate model | All regions in 1 model | ||||
| OR (95% CI) | P value | OR (95% CI) | P value | ||
| Frontal lobe cortex | 19 (5) | 1.46 (0.57-3.73) | .43 | 0.41 (0.11-1.50) | .18 |
| Frontal lobe white matter | 36 (10) | 4.30 (1.97-9.38) | <.001 | 3.25 (1.38-7.66) | .007 |
| Parietal lobe cortex | 13 (4) | 3.07 (0.92-10.24) | .07 | 2.64 (0.63-11.04) | .18 |
| Parietal lobe white matter | 18 (5) | 11.59 (2.61-51.42) | .001 | 8.18 (1.69-39.65) | .009 |
| Temporal lobe cortex | 14 (4) | 1.90 (0.64-5.65) | .25 | 0.85 (0.21-3.47) | .82 |
| Temporal lobe white matter | 13 (4) | 4.75 (1.27-17.76) | .02 | 2.44 (0.48-12.34) | .28 |
| Hippocampus | 6 (2) | 6.53 (0.75-56.83) | .09 | 4.38 (0.37-51.56) | .24 |
| Thalamus | 23 (6) | 3.51 (1.40-8.84) | .008 | 1.31 (0.42-4.09) | .65 |
| Caudate | 33 (9) | 1.94 (0.99-3.80) | .05 | 1.29 (0.60-2.74) | .52 |
| Lentiform nucleus | 41 (11) | 2.96 (1.38-6.36) | .005 | 1.67 (0.69-4.07) | .26 |
| Occipital lobe cortex | 13 (4) | 1.88 (0.59-5.95) | .29 | 0.88 (0.19-4.11) | .87 |
| Occipital lobe white matter | 7 (2) | 7.98 (0.94-68.09) | .06 | 3.87 (0.34-44.14) | .28 |
Abbreviation: OR, odds ratio.
Macroinfarcts in the brainstem and cerebellum were not examined in association with cognitive impairment as they were present in 1 and 3 participants, respectively.
Estimates are derived from logistic regression models, with cognitive impairment as the outcome and brain region(s) as the models’ terms, either in separate models or all in 1 model. All the models were controlled for age at death, sex, and education.
Next, we examined size of the macroinfarcts in relation to cognitive impairment. Both lacunar and nonlacunar macroinfarcts were present in 15 participants, while 66 had only lacunar and 22 only nonlacunar macroinfarcts. In a logistic regression including terms for both sizes of macroinfarcts, lacunar macroinfarcts (OR, 2.28; 95% CI, 1.35-3.86; P = .002) and nonlacunar macroinfarcts (OR, 2.73; 95% CI, 1.30-5.75; P = .008) were both associated with cognitive impairment.
In 246 participants of vascular subgroup, arteriolosclerosis was also assessed in frontal and parietal white matter, brain regions whose macroinfarcts were associated with pure VCI. We examined if arteriolosclerosis at these brain regions was also associated with pure VCI. Moderate to severe arteriolosclerosis was present in 91 participants (37.0%) and 81 participants (32.9%) in frontal and parietal white matters, respectively. In a model controlled for all the pathologies, arteriolosclerosis at neither of white matter regions was associated with cognitive impairment (eTable 8 in the Supplement).
Discussion
In a cohort of 1767 older adults, 369 (20.9%) were without significant neurodegenerative brain pathologies. In this subgroup, we found that cerebrovascular disease pathologies, specifically macroinfarcts, were associated with higher odds of cognitive impairment and dementia (pure VCI and vascular dementia) and a faster rate of cognitive decline. By examining brain regions, we found that the main associates of pure VCI were macroinfarcts in white matters that possibly resulted in cognitive impairment and dementia by interruption of communication between cortical and subcortical brain regions.
Several community-based clinical pathological studies have reported that most dementia cases in the community are caused by combinations of neurodegenerative and cerebrovascular disease pathologies.25,26,27,28 However, AD was the prominent pathology, explaining 30% to 36% of cognitive decline variance compared with 3% to 8% of the variance explained by cerebrovascular disease pathologies.10 Therefore, it was uncertain if cerebrovascular disease pathologies alone can be the attributed pathologies of sporadic cognitive decline and dementia in older adults in the absence of AD and other neurodegenerative pathologies.4,5 This uncertainty is reflected in a recent guideline from the Vascular Impairment of Cognition Classification Consensus Study.29 The guideline classifies vascular dementia into 4 subtypes, one being mixed dementia. It is unclear if the 3 other subtypes will have noticeable frequency in the community, as their subtype will change into mixed dementia if evidence of comorbid AD and other neurodegenerative pathologies are present. Our findings extend prior research by showing that pure VCI is not uncommon. Indeed, frequency of participants at risk of pure VCI is comparable with frequency of participants at risk of cognitive impairment attributed to pure AD and other neurodegenerative pathologies. However, participants at risk of cognitive impairment attributed to mixed neurodegenerative and cerebrovascular disease pathologies were the most frequent. Future studies should investigate biological markers for in vivo identification of these subgroups.30,31
Our findings indicate that it is mainly white matter disruptions in the frontal and, to a lower extent, parietal lobe that is associated with cognitive decline in pure VCI, findings supported by imaging studies that reported strategic locations of brain infarcts and disrupted white matter tracts.24,32 As these brain regions are specifically associated with executive functions and processing speed, our findings are also in line with prior research that identified processing speed impairment as a prominent manifestation of cerebrovascular disease pathologies.33 The mechanism linking cerebrovascular disease pathologies with frontal white matter disruption in not clear. Computational models34 and imaging studies35 have identified that cerebral perfusion pressure is lower in deep white matter compared with vascular centerncephalon, including the basal ganglia and thalamus. We hypothesize that low perfusion pressure causes hypoxia and damages the integrity of the deep white matter, which is more severe in individuals with long-lasting vascular risk factors having narrowed vessels. Other mechanisms are also possible, including endothelial failure.36 However, the cerebrovascular disease pathologies could explain only 10% of the variance in the cognitive decline rate of pure VCI, and other pathologies and proteins37 should be examined in future studies with larger samples.
Our study findings could have clinical applications. Most randomized clinical trials examining vascular treatments for prevention of dementia have been futile.38,39,40,41 However, these treatments have been examined in samples derived from the general population, where the main cognitive decline driving pathology is AD, which is unrelated to the burden of vascular risk factors.42 We hypothesize that vascular treatments, including statins,43 will be effective in prevention of dementia in subgroups of older adults who do not have significant neurodegenerative pathologies, and cerebrovascular disease pathologies are the main cognitive driving pathologies. An observational study using patients’ registries found a lower risk of dementia in patients with atrial fibrillation treated with anticoagulants.44
Strengths and Limitations
The findings are supported by several study strengths. The data were derived from 2 community-based clinical pathological cohorts with high follow-up and autopsy rates, making attrition bias less possible. The pathological data were obtained by personnel blinded to clinical data, and experts adjudicating presence of cognitive impairment and dementia were not aware of the pathological findings. Thorough structured pathological examination protocols were used that gathered 9 pathological indices, enabling us delineation of participants with and without cerebrovascular disease and neurodegenerative pathologies.
However, several limitations require our findings to be examined in future complementary studies. The analyses, by design, were limited to deceased participants with postmortem pathological assessments. The deceased are older and more impaired than living ROS and MAP participants. Further, participants with neurodegenerative diseases were, by necessity, separately examined from the participants with cerebrovascular disease pathologies, with the potential for additional bias. The cross-sectional nature of the analyses precludes causal inferences as the timing of the infarcts is not known, other than being chronic which is typically at least six months old. Additionally, participants were mainly White with high educational levels, not representative of the general population of older adults. Cerebrovascular disease pathologies, including arteriolosclerosis, were examined in selected brain regions and in only 1 hemisphere except macroinfarcts, which were examined in both hemispheres. As cerebrovascular disease pathologies may not be distributed equally across the 2 hemispheres,45 future pathological studies that examine both hemispheres are needed. Neuropathological evaluation did not include diseased white matter or standardized measures for brain hemorrhages, including microhemorrhages.
Conclusions
In this study, pure VCI, defined as cognitive impairment contributed mainly by cerebrovascular disease pathologies, was not uncommon in a cohort of 1767 older adults. Macroinfarcts, specifically in the white matter, was the main cerebrovascular disease pathology associated with cognitive decline and contributing to cognitive impairment. Future studies are required to uncover whether vascular treatments can reduce cognitive impairment in this subgroup of the population.
eMethods.
eFigure. Cognitive Decline in the 3 Pathology Subgroups
eTable 1. Association of Brain Pathologies With Dementia in 3 Pathology Subgroups
eTable 2. Cognitive Decline in the Vascular and Mixed Subgroups Compared With the Neurodegenerative Subgroup
eTable 3. Pathologies Separately Examined in Association With Longitudinal Changes of Global Cognition in the 3 Pathology Subgroups
eTable 4. Pathologies Examined Together in Association With Longitudinal Changes of global Cognition in the 3 Pathology Subgroups
eTable 5. Association of Macroinfarcts With Cognitive Impairment in the Vascular Subgroup After Controlling for Additional Covariates
eTable 6. Association of Macroinfarcts With the Decline in Specific Cognitive Functions
eTable 7. Macroinfarcts in Different Brain Regions in Association With Cognitive Decline in the Vascular Subgroup
eTable 8. Association of Arteriolosclerosis at Frontal and Parietal White Matter With Cognitive Impairment in the Vascular Subgroup
eReferences.
References
- 1.Hachinski VC, Lassen NA, Marshall J. Multi-infarct dementia. a cause of mental deterioration in the elderly. Lancet. 1974;2(7874):207-210. doi: 10.1016/S0140-6736(74)91496-2 [DOI] [PubMed] [Google Scholar]
- 2.Román GC, Tatemichi TK, Erkinjuntti T, et al. Vascular dementia: diagnostic criteria for research studies. report of the NINDS-AIREN International Workshop. Neurology. 1993;43(2):250-260. doi: 10.1212/WNL.43.2.250 [DOI] [PubMed] [Google Scholar]
- 3.Hachinski VC, Iliff LD, Zilhka E, et al. Cerebral blood flow in dementia. Arch Neurol. 1975;32(9):632-637. doi: 10.1001/archneur.1975.00490510088009 [DOI] [PubMed] [Google Scholar]
- 4.Emrani S, Lamar M, Price CC, et al. Alzheimer’s/vascular spectrum dementia: classification in addition to diagnosis. J Alzheimers Dis. 2020;73(1):63-71. doi: 10.3233/JAD-190654 [DOI] [PubMed] [Google Scholar]
- 5.van der Flier WM, Skoog I, Schneider JA, et al. Vascular cognitive impairment. Nat Rev Dis Primers. 2018;4:18003. doi: 10.1038/nrdp.2018.3 [DOI] [PubMed] [Google Scholar]
- 6.Gorelick PB, Scuteri A, Black SE, et al. ; American Heart Association Stroke Council, Council on Epidemiology and Prevention, Council on Cardiovascular Nursing, Council on Cardiovascular Radiology and Intervention, and Council on Cardiovascular Surgery and Anesthesia . Vascular contributions to cognitive impairment and dementia: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2011;42(9):2672-2713. doi: 10.1161/STR.0b013e3182299496 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Chui HC, Zarow C, Mack WJ, et al. Cognitive impact of subcortical vascular and Alzheimer’s disease pathology. Ann Neurol. 2006;60(6):677-687. doi: 10.1002/ana.21009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hachinski V, Iadecola C, Petersen RC, et al. National Institute of Neurological Disorders and Stroke-Canadian Stroke Network vascular cognitive impairment harmonization standards. Stroke. 2006;37(9):2220-2241. doi: 10.1161/01.STR.0000237236.88823.47 [DOI] [PubMed] [Google Scholar]
- 9.Boyle PA, Yu L, Leurgans SE, et al. Attributable risk of Alzheimer’s dementia attributed to age-related neuropathologies. Ann Neurol. 2019;85(1):114-124. doi: 10.1002/ana.25380 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Boyle PA, Wang T, Yu L, et al. To what degree is late life cognitive decline driven by age-related neuropathologies? Brain. 2021;144(7):2166-2175. doi: 10.1093/brain/awab092 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bennett DA, Buchman AS, Boyle PA, Barnes LL, Wilson RS, Schneider JA. Religious Orders Study and Rush Memory and Aging Project. J Alzheimers Dis. 2018;64(s1):S161-S189. doi: 10.3233/JAD-179939 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Boyle PA, Yu L, Nag S, et al. Cerebral amyloid angiopathy and cognitive outcomes in community-based older persons. Neurology. 2015;85(22):1930-1936. doi: 10.1212/WNL.0000000000002175 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Albert MS, DeKosky ST, Dickson D, et al. The diagnosis of mild cognitive impairment due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7(3):270-279. doi: 10.1016/j.jalz.2011.03.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7(3):263-269. doi: 10.1016/j.jalz.2011.03.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Oveisgharan S, Buchman AS, Yu L, et al. APOE ε2ε4 genotype, incident AD and MCI, cognitive decline, and AD pathology in older adults. Neurology. 2018;90(24):e2127-e2134. doi: 10.1212/WNL.0000000000005677 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Oveisgharan S, Wilson RS, Yu L, Schneider JA, Bennett DA. Association of early-life cognitive enrichment with Alzheimer disease pathological changes and cognitive decline. JAMA Neurol. 2020;77(10):1217-1224. doi: 10.1001/jamaneurol.2020.1941 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Oveisgharan S, Arvanitakis Z, Yu L, Farfel J, Schneider JA, Bennett DA. Sex differences in Alzheimer’s disease and common neuropathologies of aging. Acta Neuropathol. 2018;136(6):887-900. doi: 10.1007/s00401-018-1920-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kapasi A, Leurgans SE, Arvanitakis Z, Barnes LL, Bennett DA, Schneider JA. Aβ (amyloid beta) and tau tangle pathology modifies the association between small vessel disease and cortical microinfarcts. Stroke. 2021;52(3):1012-1021. doi: 10.1161/STROKEAHA.120.031073 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Oveisgharan S, Yu L, Capuano A, et al. Late-life vascular risk score in association with postmortem cerebrovascular disease brain pathologies. Stroke. 2021;52(6):2060-2067. doi: 10.1161/STROKEAHA.120.030226 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Westover MB, Bianchi MT, Yang C, Schneider JA, Greenberg SM. Estimating cerebral microinfarct burden from autopsy samples. Neurology. 2013;80(15):1365-1369. doi: 10.1212/WNL.0b013e31828c2f52 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kapasi A, Yu L, Petyuk V, Arfanakis K, Bennett DA, Schneider JA. Association of small vessel disease with tau pathology. Acta Neuropathol. 2022;143(3):349-362. doi: 10.1007/s00401-021-02397-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Agrawal S, Yu L, Kapasi A, et al. Limbic-predominant age-related TDP-43 encephalopathy neuropathologic change and microvascular pathologies in community-dwelling older persons. Brain Pathol. 2021;31(3):e12939. doi: 10.1111/bpa.12939 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Neltner JH, Abner EL, Baker S, et al. Arteriolosclerosis that affects multiple brain regions is linked to hippocampal sclerosis of ageing. Brain. 2014;137(pt 1):255-267. doi: 10.1093/brain/awt318 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Weaver NA, Kuijf HJ, Aben HP, et al. Strategic infarct locations for post-stroke cognitive impairment: a pooled analysis of individual patient data from 12 acute ischaemic stroke cohorts. Lancet Neurol. 2021;20(6):448-459. doi: 10.1016/S1474-4422(21)00060-0 [DOI] [PubMed] [Google Scholar]
- 25.White L, Petrovitch H, Hardman J, et al. Cerebrovascular pathology and dementia in autopsied Honolulu-Asia Aging Study participants. Ann N Y Acad Sci. 2002;977:9-23. doi: 10.1111/j.1749-6632.2002.tb04794.x [DOI] [PubMed] [Google Scholar]
- 26.James BD, Bennett DA, Boyle PA, Leurgans S, Schneider JA. Dementia from Alzheimer disease and mixed pathologies in the oldest old. JAMA. 2012;307(17):1798-1800. doi: 10.1001/jama.2012.3556 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kovacs GG, Alafuzoff I, Al-Sarraj S, et al. Mixed brain pathologies in dementia: the BrainNet Europe Consortium experience. Dement Geriatr Cogn Disord. 2008;26(4):343-350. doi: 10.1159/000161560 [DOI] [PubMed] [Google Scholar]
- 28.Schneider JA, Arvanitakis Z, Bang W, Bennett DA. Mixed brain pathologies account for most dementia cases in community-dwelling older persons. Neurology. 2007;69(24):2197-2204. doi: 10.1212/01.wnl.0000271090.28148.24 [DOI] [PubMed] [Google Scholar]
- 29.Skrobot OA, Black SE, Chen C, et al. ; VICCCS Group . Progress toward standardized diagnosis of vascular cognitive impairment: guidelines from the Vascular Impairment of Cognition Classification Consensus Study. Alzheimers Dement. 2018;14(3):280-292. doi: 10.1016/j.jalz.2017.09.007 [DOI] [PubMed] [Google Scholar]
- 30.Wilcock D, Jicha G, Blacker D, et al. ; MarkVCID Consortium . MarkVCID cerebral small vessel consortium: I. enrollment, clinical, fluid protocols. Alzheimers Dement. 2021;17(4):704-715. doi: 10.1002/alz.12215 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Lu H, Kashani AH, Arfanakis K, et al. ; MarkVCID Consortium . MarkVCID cerebral small vessel consortium: II. neuroimaging protocols. Alzheimers Dement. 2021;17(4):716-725. doi: 10.1002/alz.12216 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Duering M, Zieren N, Hervé D, et al. Strategic role of frontal white matter tracts in vascular cognitive impairment: a voxel-based lesion-symptom mapping study in CADASIL. Brain. 2011;134(Pt 8):2366-2375. doi: 10.1093/brain/awr169 [DOI] [PubMed] [Google Scholar]
- 33.Yang FM, Grigorenko A, Tommet D, et al. AD pathology and cerebral infarctions are associated with memory and executive functioning one and five years before death. J Clin Exp Neuropsychol. 2013;35(1):24-34. doi: 10.1080/13803395.2012.740001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Blanco PJ, Müller LO, Spence JD. Blood pressure gradients in cerebral arteries: a clue to pathogenesis of cerebral small vessel disease. Stroke Vasc Neurol. 2017;2(3):108-117. doi: 10.1136/svn-2017-000087 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Hachinski V, Østergaard L. The ambibaric brain: pathophysiological and clinical implications. Stroke. 2021;52(6):e259-e262. doi: 10.1161/STROKEAHA.120.033492 [DOI] [PubMed] [Google Scholar]
- 36.Wardlaw JM, Smith C, Dichgans M. Mechanisms of sporadic cerebral small vessel disease: insights from neuroimaging. Lancet Neurol. 2013;12(5):483-497. doi: 10.1016/S1474-4422(13)70060-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Yu L, Tasaki S, Schneider JA, et al. Cortical proteins associated with cognitive resilience in community-dwelling older persons. JAMA Psychiatry. 2020;77(11):1172-1180. doi: 10.1001/jamapsychiatry.2020.1807 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Ryan J, Storey E, Murray AM, et al. ; ASPREE Investigator Group . Randomized placebo-controlled trial of the effects of aspirin on dementia and cognitive decline. Neurology. 2020;95(3):e320-e331. doi: 10.1212/WNL.0000000000009277 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Williamson JD, Pajewski NM, Auchus AP, et al. ; SPRINT MIND Investigators for the SPRINT Research Group . Effect of intensive vs standard blood pressure control on probable dementia: a randomized clinical trial. JAMA. 2019;321(6):553-561. doi: 10.1001/jama.2018.21442 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.McGuinness B, Craig D, Bullock R, Passmore P. Statins for the prevention of dementia. Cochrane Database Syst Rev. 2016;(1):CD003160. doi: 10.1002/14651858.CD003160.pub3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Peters R, Yasar S, Anderson CS, et al. Investigation of antihypertensive class, dementia, and cognitive decline: a meta-analysis. Neurology. 2020;94(3):e267-e281. doi: 10.1212/WNL.0000000000008732 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Oveisgharan S, Capuano AW, Kapasi A, et al. Association of low systolic blood pressure with postmortem amyloid-β and tau. J Alzheimers Dis. 2020;78(4):1755-1764. doi: 10.3233/JAD-200412 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Oveisgharan S, Yu L, Barnes LL, et al. Association of statins with cerebral atherosclerosis and incident parkinsonism in older adults. Neurology. 2022;98(19):e1976-e1984. doi: 10.1212/WNL.0000000000200182 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Friberg L, Rosenqvist M. Less dementia with oral anticoagulation in atrial fibrillation. Eur Heart J. 2018;39(6):453-460. doi: 10.1093/eurheartj/ehx579 [DOI] [PubMed] [Google Scholar]
- 45.Ryu WS, Schellingerhout D, Ahn HS, et al. Hemispheric asymmetry of white matter hyperintensity in association with lacunar infarction. J Am Heart Assoc. 2018;7(22):e010653. doi: 10.1161/JAHA.118.010653 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods.
eFigure. Cognitive Decline in the 3 Pathology Subgroups
eTable 1. Association of Brain Pathologies With Dementia in 3 Pathology Subgroups
eTable 2. Cognitive Decline in the Vascular and Mixed Subgroups Compared With the Neurodegenerative Subgroup
eTable 3. Pathologies Separately Examined in Association With Longitudinal Changes of Global Cognition in the 3 Pathology Subgroups
eTable 4. Pathologies Examined Together in Association With Longitudinal Changes of global Cognition in the 3 Pathology Subgroups
eTable 5. Association of Macroinfarcts With Cognitive Impairment in the Vascular Subgroup After Controlling for Additional Covariates
eTable 6. Association of Macroinfarcts With the Decline in Specific Cognitive Functions
eTable 7. Macroinfarcts in Different Brain Regions in Association With Cognitive Decline in the Vascular Subgroup
eTable 8. Association of Arteriolosclerosis at Frontal and Parietal White Matter With Cognitive Impairment in the Vascular Subgroup
eReferences.