How Do Different Forms of Vascular Brain Injury Relate to Cognition in a Memory Clinic Population: The TRACE-VCI Study

Abstract

Background:

Memory clinic patients frequently present with different forms of vascular brain injury due to different etiologies, often co-occurring with Alzheimer’s disease (AD) pathology.

Objective:

We studied how cognition was affected by different forms of vascular brain injury, possibly in interplay with AD pathology.

Methods:

We included 860 memory clinic patients with vascular brain injury on magnetic resonance imaging (MRI), receiving a standardized evaluation including cerebrospinal fluid (CSF) biomarker analyses (n = 541). The cognitive profile of patients with different forms of vascular brain injury on MRI (moderate/severe white matter hyperintensities (WMH) (n = 398), microbleeds (n = 368), lacunar (n = 188) and non-lacunar (n = 96) infarct(s), macrobleeds (n = 16)) was assessed by: 1) comparison of all these different forms of vascular brain injury with a reference group (patients with only mild WMH (n = 205) without other forms of vascular brain injury), using linear regression analyses also stratified for CSF biomarker AD profile and 2) multivariate linear regression analysis.

Results:

The cognitive profile was remarkably similar across groups. Compared to the reference group effect sizes on all domains were <0.2 with narrow 95% confidence intervals, except for non-lacunar infarcts on information processing speed (age, sex, and education adjusted mean difference from reference group (β: – 0.26, p = 0.05). Results were similar in the presence (n = 300) or absence (n = 241) of biomarker co-occurring AD pathology. In multivariate linear regression analysis, higher WMH burden was related to a slightly worse performance on attention and executive functioning (β: – 0.08, p = 0.02) and working memory (β: – 0.08, p = 0.04).

Conclusion:

Although different forms of vascular brain injury have different etiologies and different patterns of cerebral damage, they show a largely similar cognitive profile in memory clinic patients regardless of co-occurring AD pathology.

Keywords

Cerebral small vessel diseases cerebrovascular disorders cognitivedisorders neuropsychological test

INTRODUCTION

Patients presenting at a memory clinic often show vascular brain injury on magnetic resonance imaging (MRI) of the brain. Of note, vascular brain injury is a heterogeneous construct in this setting. Different forms of vascular brain injury have different etiologies and vascular lesions affect different parts of the brain. This heterogeneity of vascular injury, possible in interplay with co-occurring Alzheimer’s disease (AD) pathology, might be reflected in distinct cognitive profiles.

The literature on the relationship between vascular brain lesions on MRI and cognition largely concerns studies on non-memory clinic populations, mostly population-based cohorts. These studies often evaluate only one type of vascular brain injury in relation to cognition. Comparisons between different forms (for instance lacunar infarcts versus microbleeds) are generally not reported [1]. Other studies address different forms of vascular brain injury as one combined entity, generally as a construct for “small vessel disease” [2 –4]. Conventionally, small vessel disease, in particular with subcortical lesions, is considered to be related to impairment of the cognitive domains executive functioning and information processing speed, with relative preservation of memory [1 , 5]. If memory impairment is present, it typically involves retrieval problems, which may respond to cuing with preserved recognition [6]. AD, in contrast, mainly involves episodic memory impairment affecting the encoding and storage of information [6]. These concepts on etiology-function relationships are commonly used in clinical practice, when assigning potential clinical significance to vascular brain injury on MRI in the diagnostic work up of patients with cognitive complaints. However, thus far, there are only few studies in a memory clinic population that have actually assessed how different forms of vascular brain injury and possible co-occurring AD pathology relate to cognition, independent of the assigned clinical diagnosis or assumed etiology. This often unrecognized knowledge gap can lead to circular reasoning. Pre-existing concepts on functional impact of vascular brain lesions may bias the clinical diagnosis and the relevance that is assigned to these lesions. For example, in a patient with prominent episodic memory deficits we would consider a likely diagnosis of AD, and might consider moderate to severe white matter hyperintensities (WMH) on MRI as co-morbidity. However, if we would encounter exactly the same MRI in another patient with a deficit in executive functioning, we might label the WMH as the primary etiology.

This study compared the cognitive profile between patients with different forms of vascular brain injury (WHM, lacunar infarct(s), non-lacunar infarct(s), microbleed(s) or macrobleed(s)) in a memory clinic cohort. We also assessed whether this relation changed according to the presence or absence of cerebrospinal fluid (CSF) biomarker evidence of AD pathology. Patients were derived from the “Utrecht-Amsterdam clinical features and prognosis in vascular cognitive impairment” (TRACE-VCI) study population [7].

MATERIALS AND METHODS

Study population

Patients were included from the TRACE-VCI study population, using cross sectional data. The rationale and design of the TRACE-VCI study has been published previously [7]. In short, the aim of the TRACE-VCI study was to determine the influence of different forms of vascular brain injury on the cognitive profile and prognosis of patients with cognitive complaints in a memory clinic setting. We included patients regardless of cognitive severity, including patients with no objective cognitive impairment (NOCI), mild cognitive impairment (MCI), and dementia, attending the memory clinics between September 2009 and December 2013. All patients showed evidence of vascular brain injury (i.e., possible vascular cognitive impairment (VCI)) on MRI, which was operationalized as the presence of at least one of the following neuro-imaging markers: 1) mild WMH also known as Fazekas scale grade 1 [8] and an increased vascular risk defined as the presence of ≥2 vascular risk factors (hypertension, hypercholesterolemia, diabetes mellitus, obesity, current smoking or a reported history of a vascular event other than stroke; full definitions in design paper [7]); 2) moderate to severe WMH, also known as Fazekas scale grade ≥2; or 3) ≥1 lacunar infarct(s); 4) ≥1 non-lacunar (large vessel) infarct(s); 5) ≥1 cerebral microbleed(s); 6) ≥1 intracerebral hemorrhage(s) (ICH)/macrobleed(s).

Subjects were included from the outpatient clinic of the VU University Medical Center (VUMC), registered in the Amsterdam Dementia Cohort (N = 664) and from two outpatient memory clinics of the University Medical Center Utrecht (UMCU) (N = 196) [7]. One patient from the original reported cohort [7] was erroneously included (glioma with bi-hemispheric involvement) and therefore the current study included 860 patients. Patients with a presumed primary etiology other than vascular brain injury or neurodegeneration (e.g., brain tumors) were excluded. The presence of co-occurring etiologies, in addition to vascular injury, such as neurodegenerative pathology or depression was accepted, because many patients with VCI have neurodegenerative diseases as a co-morbid etiology and depression can be a manifestation of cerebrovascular disease [9, 10].

Each patient underwent a standardized extensive one-day memory clinic evaluation including an interview, physical and (cognitive) neurological examination, laboratory testing, extensive neuropsychological testing, and a MRI-scan of the brain. Lumbar puncture was performed in a subset of the study population [7]. In the VUMC, collection of CSF was a standard procedure for research purposes. Only patients with contraindications, such as the use of anticoagulants, an unsuccessful procedure or refusal of the puncture did not undergo the procedure. In the UMCU, collection of CSF biomarkers was not a standard procedure and only performed at the discretion of the doctor and the patient, also in the context of research. All patients provided informed consent prior to research related procedures.

Cognitive assessment

We used the Dutch version of the Mini-Mental State Examination (MMSE; maximum score of 30) as a cognitive screening test [11]. The severity of cognitive symptoms was assessed using the Clinical Dementia Rating (CDR; 0–3) global score [12]. All participants underwent an extensive neuropsychological examination, with some variation between centers and over time. Harmonization of the test battery had been established through a Dutch multicenter university memory clinic research program on diagnosis and prognosis of cognitive impairment and dementia [13]. Tasks that were available for the majority of patients (>70%) were included. The tasks were summarized in five major widely used cognitive domains to reduce the amount of neuropsychological variables for statistical analysis and clinical interpretation: 1) memory, 2) attention and executive functioning, 3) information processing speed, 4) perception and construction, and 5) working memory.

The domain memory was assessed by the Dutch version of the Rey Auditory Verbal Learning Test (RAVLT) [14] and the Visual Association Test (VAT) part A [15].

The domain attention and executive functioning was assessed using the ratio of the Trail Making Test part B and A (TMT-B and TMT-A) [16], the Stroop Color Word Test [17, 18] and the category naming tasks (animal naming, one minute) and lexical fluency tasks (one minute per letter) [19].

The domain information processing speed was assessed by the TMT-A, the Stroop Color Word Test I and II and the Digit Symbol-Coding Test (DSCT) of the WAIS-III or the Letter Digit Substitution Test (LDST) [20, 21]. The cognitive domain perception and construction was assessed using the Fragmented Letters and Dot Counting subtests of the Visual Object and Space Perception Battery (VOSP) [22].

The domain working memory was assessed by the Digit Span of the Wechsler Adult Intelligence Scale – 3rd edition (WAIS-III) [20]. After publication of the design article [7], more data regarding the Digit Span were collected. The availability and raw test scores are shown in the Supplementary Table 1. More information about the tests is provided in the design article of the TRACE-VCI study [7].

Z-scores based on the total study population were created for each individual test (inversed Z-scores for the TMT and Stroop Color Word Test). The test Z-scores were averaged to create Z-scores per domain. If patients were unable to perform a test for various reasons, the test was defined as a missing variable. If individual test scores were missing, the domain Z-score was based only on the available tests.

Laboratory testing

Plasma glucose level, HbA1c levels and CSF were collected in a subset of the study population. CSF concentrations of amyloid-β₄₂ (Aβ₄₂), tau and/or total tau phosphorylated at threonine 181 (p-tau) were measured at a central laboratory for clinics at the Department of Clinical Chemistry of the VUMC [23]. CSF samples were stored at – 20°C until biomarker analysis (within 2 months). Aβ₄₂, total tau, and p-tau were measured with commercially available ELISAs (Innotest β-amyloid(_1 - 42), Innotest hTAU-Ag and Innotest Phosphotau(181_ρ), respectively; Innogenetics, Ghent, Belgium) on a routine basis [23]. Patients with a ratio of total tau to amyloid-β₄₂ of more than 0.52 were classified as having a positive CSF biomarker AD profile [24].

MRI assessment

Brain MRI scans were performed on 3.0 Tesla (809 out of 860 patients (94%)) or 1.5 Tesla MRI scanners (51 patients (6%)). Most scans were performed on a General Electric (GE) (619 patients (72%)) or Philips (239 patients (28%)) MRI scanner. The MRI scan protocol included the following sequences: 3D T1-weighted, T2-weighted, T2*-weighted/susceptibility-weighted imaging (SWI) and fluid-attenuated inversion recovery (FLAIR) sequences. A total of 849 (99%) patients were scanned using all of these sequences. In 11 patients (1%) a 2D T1-weighted sequence was acquired instead of a 3D T1-weighted sequence and/or no FLAIR sequence was available. There were 12 patients with no microbleed rating because of a missing or unreadable T2*-weighted scan, therefore in 848 out of 860 patients (99%) microbleeds were scored. Further details of the MRI sequence parameters were described in the design article of the TRACE-VCI-study [7].

WMH were rated using the Fazekas scale (WMH grade 0–3: none or a single punctate lesion, multiple punctate lesions (mild WMH), beginning confluency of lesions (moderate WMH), large confluent lesions (severe WMH)) on FLAIR images [8]. Non-lacunar and lacunar infarct(s), microbleed(s) and ICH/macrobleed(s) were all rated in line with the STRIVE (standards for reporting vascular changes on neuroimaging) criteria [25]. Ratings were performed by or under the supervision of a neuroradiologist (in training).

Statistical analysis

Statistical analyses were performed using SPSS (version 21; SPSS, Chicago, IL, USA). The computed Z-scores of the five cognitive domains were used to evaluate the cognitive profile in relation to the vascular brain injury type in two ways.

Firstly, the mean cognitive domain Z-scores for each form of vascular brain injury were compared to a reference group, which was defined as patients with only mild WMH (Fazekas score 1), ≥2 vascular risk factors and no other forms of vascular brain injury. A linear regression model was used, adjusting for sex, age, and educational level.

Secondly, multivariate linear regression analysis was performed including the lesion burden for all forms of vascular brain injury (i.e., WMH entered as Fazekas scale 0, 1, 2, or 3, microbleed(s) as 0, 1–4, or ≥5, lacunar infarct(s) as 0, 1, or ≥2 and non-lacunar infarct(s) as 0 or ≥ 1) with sex, age and educational level, number of vascular risk factors (0–6), and CSF biomarker AD profile as covariates.

Presence or absence of AD pathology, as reflected in positive or negative CSF biomarker profile can influence the cognitive profile. Moreover, the relation of vascular lesion burden to cognition may vary according to disease stage (NOCI, MCI, and dementia). Therefore, we performed pre-specified post hoc multivariate linear regression analyses, stratified according to CSF biomarker status and in separate analyses also to disease stage, relating lesion burden for all forms of vascular brain injury to cognition.

A p-value of less than 0.05 was considered significant.

RESULTS

Characteristics of the study population

Mean age of the cohort was 67.7 years (SD±8.5) and 46% were female. The median MMSE score was 25 (IQR 22–28) points out of 30 and the median CDR was 0.5 (IQR 0.5–1) out of 3. Dementia was diagnosed in approximately half of patients (n = 449 (52%)), of whom 304 (68%) were clinically diagnosed with AD. CSF was collected in 541 (63%) patients. In total, 300 (56%) of these patients had a positive CSF biomarker AD profile (ratio of total tau to amyloid-β₄₂ >0.52). Table 1 provides details on demographic characteristics, vascular risk factors, severity of cognitive impairment, and the results of CSF biomarkers in the study population.

Table 1

Demographics, vascular risk factors, severity of cognitive impairment, and CSF biomarker analysis in the study population

	Number of patients (N = 860)
Demographic characteristics
Female, n (%)	398 (46)
Mean age (y) (SD)	67.7 (±8.5)
Level of education (Verhage scale range 1–7), median (IQR) (n = 856)^A	5 (4–6)
Vascular risk factors (n (%))
Hypertension	729 (85)
Hypercholesterolemia	386 (45)
Diabetes mellitus	159 (18)
Current smoker (n = 852)	173 (20)
Obesity (BMI ≥30) (n = 848)	176 (21)
History of reported vascular events
History of reported stroke	78 (9)
History of reported vascular events other than stroke^B	86 (10)
Severity of cognitive impairment and clinical diagnosis (n (%))
NOCI	198 (23)
MCI	213 (25)
Dementia	449 (52)
Vascular	37
Neurodegenerative	385
Alzheimer’s disease	304
Frontotemporal dementia	25
Lewy body dementia	20
Others^C	36
Unknown etiology	27
CSF biomarker analysis (N = 541)
Aβ₄₂, median (IQR) (pg/mL)	617 [463–927.5]
Tau, median [IQR] (pg/mL)	371 [254–651.5]
p-tau (N = 539), median (IQR) (pg/mL)	56 [39 –81]
Tau / Aβ₄₂ ratio, median (IQR) (pg/mL)	0.64 [0.28–1.29]
Positive CSF biomarker AD profile (ratio tau / Aβ₄₂ >0.52), n (%)	300 (56)

Aβ₄₂, amyloid-β₄; CSF, cerebrospinal fluid; IQR, interquartile range; MCI, mild cognitive impairment; NOCI, no objective cognitive impairment; p-tau, tau phosphorylated at threonine 181; SD, standard deviation; if there were missing data the number (n) is specifically mentioned. ^AVerhage scale [52]. ^BMyocardial infarction, surgery or endovascular treatment for coronary artery disease, any arterial occlusion or surgical intervention of a peripheral artery (such as an abdominal or leg artery) or carotid artery intervention (stenting or endarterectomy). ^cSuch as primary progressive aphasia [53], cortical basal syndrome [54], and progressive supranuclear palsy [55].

Distribution of different forms of vascular brain injury

A total of 500 (58%) patients presented with only one single form of vascular brain injury (Table 2). Hence, the other 360 (42%) patients showed mixed forms of vascular brain injury. WMH burden was Fazekas 0 in 71 (8%) patients, 1 in 391 (46%), 2 in 289 (34%), and 3 in 109 (13%) patients. Microbleeds were absent in 480 (56%) patients, whereas 276 (32%) had 1–4 and 91 (11%) patients had 5 or more microbleed(s). Lacunar infarcts were absent in 672 (78%) patients, whereas 81 (9%) had 1 and 106 (12%) patients had ≥2 lacunar infarct(s). Non-lacunar infarcts were absent in 764 (89%) patients and 96 (11%) patients had ≥1 infarct(s). Macrobleeds were infrequent (<2%) in this cohort. We therefore did not consider macrobleeds as a separate form of vascular brain injury in our analyses on cognitive profiles.

Table 2

Occurrence of different forms of vascular brain injury

Vascular brain injury	Mixed and single vascular brain injury				Single vascular brain injury
	Total	AD CSF+	AD CSF–	p	Total	AD CSF+	AD CSF–	p
	N = 860	N = 300	N = 241		N = 500	N = 195	N = 149
Mild WMH and ≥2 VRF	307 (36)	100 (33)	95 (39)	0.14	205 (41)^*	73 (37)^*	66 (44)^*	0.42
Moderate/severe WMH	398 (46)	141 (47)	97 (40)	0.12	157 (31)	68 (35)	39 (26)	0.06
≥1 Lacunar infarct(s)	188 (22)	41 (14)	57 (24)	0.00	13 (3)	5 (3)	4 (3)	1.0
≥1 Non-lacunar infarct(s)	96 (11)	15 (5)	18 (8)	0.23	4 (1)	1 (1)	1 (1)	1.0
≥1 Microbleed(s)¹	368 (43)	136 (45)	101 (42)	0.39	119 (24)	48 (25)	39 (26)	0.98
≥1 Macrobleed(s)	16 (2)	0 (0)	2 (1)	0.20	2 (0)	0 (0)	0 (0)	–

AD, Alzheimer’s disease; AD CSF+, positive cerebrospinal fluid biomarker AD profile; AD CSF –, negative cerebrospinal fluid biomarker AD profile; CSF, cerebrospinal fluid; VRF, vascular risk factors; WMH, white matter hyperintensities. Data are presented as n (%). CSF available from 541 patients. First the data for the overall cohort (both mixed and single forms of vascular brain injury) are shown, the column % is larger than 100% since a patient can have multiple types of vascular brain injury. P-values were calculated using chi-square tests, the Fisher’s exact test was used if a low number of patients was evaluated. ¹Microbleed(s) ratings were available for 848 (99%) patients. ^*these groups were used as reference groups in statistical analysis.

In the 541 patients with CSF biomarker AD analysis, lacunar infarcts were significantly more present in patients with a negative versus a positive CSF biomarker AD profile (24% and 14%), whereas the other forms of vascular brain injury showed similar numbers (Table 2).

Lesion type and cognitive profile

For all cognitive domains, data were available from at least 97% of participants, except for perception and construction (82%). The evaluated neuropsychological tests and the mean raw test scores in the whole study population are shown in Supplementary Table 1, including the number of each available test per cognitive domain. The overall cognitive profile of our study population, showing the mean unadjusted domain Z-scores for the different forms of vascular brain injury, is demonstrated in Fig. 1. Because Z-scores were calculated at test level across the whole population, mean domain z-scores across the whole population can deviate from 0. For all different forms of vascular brain injury, the difference in cognition compared to the reference group was remarkably small on all domains, with effect sizes of less than 0.2 with narrow 95% confidence intervals. Only patients with non-lacunar infarcts showed a slightly significant worse performance on information processing speed (age, sex, and education adjusted mean difference from reference group (β) – 0.26, (CI; – 0.52 – 0.01), p = 0.05; Fig. 1). Additional adjustments for the 15-item Geriatric Depression Scale (GDS) [26] (available in 814 out of 860 patients (95%)) did not materially change the results (data not shown).

Fig. 1.

Mean unadjusted domain Z-scores for the different forms of vascular brain injury in the total study population (n = 860). Reference group (only mild WHM and ≥2 VRF (n = 205)), moderate/severe WMH (n = 398), ≥1 microbleed(s) (n = 368), ≥1 lacunar infarct(s) (n = 188), ≥1 non-lacunar infarct(s) (n = 96). Note that a single patient can be represented in multiple groups (i.e., both in moderate/severe WMH and ≥1 microbleed(s)), but that the reference group by definition does not overlap with any of the other groups. Each form of vascular brain injury was compared to the reference group with regression analysis adjusted for age, sex and education. ^* p value = 0.05; estimated mean difference from reference group (β) – 0.26 (CI; – 0.52 – 0.01). VRF, vascular risk factors; WMH, white matter hyperintensities.

A post-hoc analysis, that excluded patients with mixed pathologies, showed similar results; the cognitive profile for the different single forms of vascular brain injury was largely similar to the reference group (Supplementary Figure 1), but the group of patients with non-lacunar infarcts was too small to evaluate (only 4 patients). We also evaluated the location of microbleeds (lobar, deeply located, or mixed) showing no significant effect.

Multivariate analysis, including lesion burden

In multivariate analysis, WMH burden was related to worse performance on attention and executive functioning (– 0.08 per Fazekas point, (CI; – 0.14 – 0.01), p = 0.02; Table 3) and working memory (– 0.08 per Fazekas point, (CI; – 0.15 – 0.00), p = 0.04; Table 3), with similar result after adjusting for other forms of vascular brain injury, vascular risk factors, and CSF biomarker AD profile. This implies that in patients with severe WMH (Fazekas 3) versus no WMH (Fazekas 0), the effect size of the deficits on these domains is around 0.25, which equals around 10 percentile points of the Gaussian distribution curve. Patients with lacunar and non-lacunar infarcts had worse performance on information processing speed than patients without infarcts, but these effects were attenuated after additional adjustment for other forms of vascular brain injury (Table 3). Higher numbers of microbleeds (as a categorical variable) tended to be related to worse cognitive performance, but the results were not statistically significant (Table 3).

Table 3

Multivariate linear regression analysis in all forms of vascular brain injury including lesion burden

Total study population (N = 860)
	Memory	Attention and executive functioning	Information processing speed	Perception and construction	Working memory
Model A: adjusted for age, sex and education
WMH	–0.05 (– 0.12 0.03)	– 0.08 (– 0.14 – 0.01) ^*	–0.02 (– 0.10 0.07)	–0.04 (– 0.13 0.05)	– 0.08 (– 0.15 – 0.00) ^*
Microbleed(s)	–0.03 (– 0.11 0.06)	–0.07 (– 0.14 0.01)	–0.09 (– 0.19 0.01)	–0.09 (– 0.19 0.02)	–0.03 (– 0.11 0.06)
Lacunar infarct(s)	0.04 (– 0.04 0.13)	–0.02 (– 0.10 0.06)	– 0.10 (– 0.20 – 0.00) ^*	0.02 (– 0.09 0.12)	–0.03 (– 0.11 0.05)
Non-lacunar infarct(s)	–0.08 (– 0.26 0.11)	0.01 (– 0.16 0.17)	– 0.25 (– 0.46 – 0.04) ^*	0.08 (– 0.15 0.30)	–0.15 (– 0.33 0.03)
Model B: model A, each lesion type additionally adjusted for all other forms of vascular brain injury
Patients included	N = 835	N = 829	N = 819	N = 693	N = 816
WMH	–0.07 (– 0.14 0.01)	– 0.09 (– 0.16 – 0.02) ^*	0.01 (– 0.09 0.10)	–0.05 (– 0.14 0.05)	– 0.08 (– 0.16 – 0.00) ^*
Microbleed(s)	–0.04 (– 0.12 0.05)	–0.06 (– 0.14 0.02)	–0.09 (– 0.19 0.02)	–0.09 (– 0.19 0.02)	–0.03 (– 0.11 0.06)
Lacunar infarct(s)	0.08 (– 0.02 0.17)	0.02 (– 0.06 0.10)	–0.07 (– 0.18 0.04)	0.04 (– 0.08 0.15)	0.02 (– 0.08 0.11)
Non-lacunar infarct(s)	–0.11 (– 0.30 0.08)	–0.01 (– 0.18 0.16)	–0.21 (– 0.43 0.02)	0.04 (– 0.20 0.28)	–0.14 (– 0.32 0.05)
Model C: model B, with the number of vascular risk factors (VRF) (0–6)
Patients included	N = 835	N = 829	N = 819	N = 693	N = 816
WMH	–0.06 (– 0.14 0.02)	– 0.08 (– 0.15 – 0.01) ^*	0.01 (– 0.09 0.10)	–0.05 (– 0.14 0.05)	– 0.08 (– 0.16 – 0.01) ^*
Microbleed(s)	–0.03 (– 0.11 0.06)	–0.06 (– 0.13 0.03)	–0.09 (– 0.19 0.02)	–0.09 (– 0.19 0.02)	–0.04 (– 0.13 0.05)
Lacunar infarct(s)	0.07 (– 0.03 0.16)	0.02 (– 0.07 0.10)	–0.07 (– 0.19 0.04)	0.04 (– 0.08 0.15)	0.02 (– 0.07 0.12)
Non-lacunar infarct(s)	–0.12 (– 0.31 0.07)	–0.02 (– 0.19 0.16)	–0.21 (– 0.43 0.01)	0.04 (– 0.19 0.28)	–0.14 (– 0.33 0.05)
VRF	0.03 (– 0.02 0.09)	0.02 (– 0.03 0.07)	0.00 (– 0.06 0.07)	–0.00 (– 0.07 0.06)	–0.02 (– 0.07 0.04)
Model D: Model B, with CSF biomarker Alzheimer profile (N = 541)
Patients included	N = 527	N = 524	N = 519	N = 445	N = 518
WMH	–0.11 (– 0.20 – 0.02) ^*	–0.10 (– 0.19 – 0.01) ^*	–0.05 (– 0.17 0.07)	–0.06 (– 0.17 0.06)	–0.10 (– 0.20 – 0.00) ^*
Microbleed(s)	0.04 (– 0.06 0.14)	–0.04 (– 0.14 0.06)	–0.03 (– 0.16 0.11)	0.03 (– 0.09 0.16)	0.02 (– 0.08 0.13)
Lacunar infarct(s)	–0.03 (– 0.15 0.09)	–0.02 (– 0.14 0.09)	–0.19 (– 0.35 – 0.04) ^*	–0.04 (– 0.19 0.11)	–0.00 (– 0.13 0.12)
Non-lacunar infarct(s)	–0.28 (– 0.57 0.01)	–0.27 (– 0.54 0.01)	–0.40 (– 0.78 – 0.03) ^*	–0.19 (– 0.58 0.20)	–0.17 (– 0.48 0.14)
CSF AD profile	–0.78 (– 0.92 – 0.63) ^*	–0.39 (– 0.53 – 0.25) ^*	–0.50 (– 0.69 – 0.32) ^*	–0.45 (– 0.62 – 0.27) ^*	–0.19 (– 0.34 – 0.03) ^*

CSF, cerebrospinal fluid; VRF, vascular risk factors; WMH, white matter hyperintensities. WMH burden entered as; Fazekas scale 0–3. Lacunar infarct(s); 0 = no lacune, 1 = one lacune, 2 = 2 lacunes or more. Non-lacunar infarct(s); 0 = no infarct and 1 = 1 or more infarct(s). Microbleed(s); 0 = no microbleed, 1 = 1–4 microbleed(s) and 2 = 5 or more microbleeds. CSF biomarker AD profile; 0 = negative CSF biomarker AD profile and 1 = positive. CSF biomarker AD profile. Vascular risk factor(s) as amount of factors 0–6. ^* p < 0.05.

Influence of disease stage

We performed stratified analyses based on the severity of cognitive impairment (NOCI, MCI, and dementia). Comparing the different groups of vascular brain injury with the reference group showed similar cognitive profiles to the overall study sample in the subgroup of patients with NOCI albeit with some variation in effect sizes and bigger confidence intervals due to smaller samples compared to the total study population (Supplementary Figure 2). The same was also true for patients with dementia (Supplementary Figure 4). In patients with MCI, however, those with moderate/severe WMH or lacunar infarcts showed a slightly significant better performance on memory (age, sex, and education adjusted mean difference from reference group (β) 0.22, (CI; – 0.00 0.43), p = 0.05 and (β) 0.28, (CI; 0.06 0.49), p = 0.01; Supplementary Figure 3). Patients with MCI and non-lacunar infarcts showed a slightly significant worse performance on information processing speed (age, sex, and education adjusted mean difference from reference group (β) – 0.35, (CI; – 0.62 – 0.09), p = 0.01; Supplementary Figure 3).

Fig. 2.

Mean unadjusted domain Z-scores for the different forms of vascular brain injury in patients stratified by CSF biomarker AD profile (n = 241). Note that a single patient can be represented in multiple groups (i.e., both in moderate/severe WMH and microbleeds), but that the reference group by definition does not overlap with any of the other groups. Each form of vascular brain injury was compared to the reference group with regression analysis adjusted for age, sex, and education. CSF, cerebrospinal fluid; VRF, vascular risk factors; WMH, white matter hyperintensities. 2a: Reference group (only mild WHM and ≥2 VRF (n = 73)), moderate/severe WMH (n = 141), ≥1 microbleed(s) (n = 136), ≥1 lacunar infarct(s) (n = 41), ≥1 non-lacunar infarct(s) (n = 15). ^* p value = 0.04; estimated mean difference from reference group (β) – 0.79 (CI; – 1.56 – 0.22). 2b: Reference group (only mild WHM and ≥2 VRF (n = 66)), moderate/severe WMH (n = 97), ≥1 microbleed(s) (n = 101), ≥1 lacunar infarct(s) (n = 57), ≥1 non-lacunar infarct(s) (n = 18). ^* p = 0.04; estimated mean difference from reference group (β) – 0.39 (– 0.75 0.02). ^** p = 0.002; estimated mean difference from reference group (β) – 0.61 (– 0.99 – 0.24).

Fig. 3.

Mean unadjusted domain Z-scores for the different forms of vascular brain injury in patients with a positive or negative CSF biomarker AD profile (n = 541). Univariate analyses of variance were performed with CSF biomarker AD profile as fixed factor and age, sex and education as covariates. ^* p < 0.0005; ^** p = 0.04. CSF, cerebrospinal fluid.

Influence of lesion type considering co-occurring AD pathology

Patients with a positive CSF biomarker AD profile performed significantly worse on all cognitive domains, especially memory (– 0.78, (CI; – 0.92 – 0.63), p < 0.0005; Table 3; also Fig. 3), compared to patients with a negative profile (Table 3). Effect sizes substantially exceeded those of any form of vascular brain injury (Table 3; model D).

We repeated the analysis on the relation between vascular lesion type and cognition after stratification for a positive or negative CSF biomarker AD profile. This showed very similar results as for the total study population, although the sample sizes resulted in somewhat more variation in effects sizes and wider 95% confidence intervals. Again, compared to the reference group, the largest effect sizes were seen for patients with non-lacunar infarcts, regardless of the CSF biomarker AD profile being negative or positive (Fig. 2a, b).

DISCUSSION

This study demonstrates that memory clinic patients with different forms of vascular brain injury on MRI showed largely similar cognitive profiles, regardless of CSF biomarker AD status or severity of cognitive impairment. Patients with non-lacunar infarcts performed worse than the reference group on the cognitive domain information processing speed. Moreover, in the multivariate models WMH burden was associated with worse attention and executive functioning and working memory, but effect sizes were relatively small. On the other hand, a positive CSF biomarker AD profile by itself did markedly affect cognitive performance on all domains. These observations may certainly impact our appreciation of vascular brain lesions on MRI in clinical practice. At memory clinics neuropsychological tests in combination with patterns of vascular and non-vascular brain injury on MRI scans guide clinical diagnoses on the etiology of cognitive dysfunction. For each individual patient, the clinician assigns relevance to potential vascular brain injury in the context of cognitive dysfunction. However, the current results showed that, in this setting, patterns and severity of vascular brain injury actually explain little of the interindividual variation in cognition.

Our study is different from most of the existing literature in that we evaluated patients referred for suspected cognitive problems in a memory clinic setting. The majority of previous studies are population based, stroke clinic based or involve hospital populations selected on the presence of certain forms of vascular brain injury, rather than cognitive complaints. Moreover, previous studies mainly address one particular form of vascular brain injury, without making comparisons between the cognitive profile of different forms of vascular brain injury. WMH are the form of vascular brain injury that have been most widely investigated. Studies in large population-based cohorts or cognitively healthy elderly generally do report a relation between WMH burden and cognition [27, 28]. Yet, a meta-analysis showed that effect sizes were generally small: general intelligence (Fisher z-score – 0.10 (CI; – 0.19 to – 0.04)), memory (– 0.08 (CI; – 0.13 to – 0.06)), processing speed (– 0.11 (CI; – 0.17 to – 0.07)), attention and executive functions (– 0.11 (CI; – 0.16 to – 0.07)), and perception/construction (– 0.15 (CI; – 0.21 to – 0.07)) [29]. In a hospital-based cohort, including patients with minor cognitive complaints without formal cognitive impairment or patients with incidental vascular brain injury on brain imaging, higher burden of WMH was also associated with worse performance on these domains [30]. Although effect sizes were slightly larger than reported in the meta-analysis [29], they were also small to modest [30]. The small effect sizes that we observe for WMH burden are in line with these observations.

Regarding microbleeds, population-based studies generally do report an association with worse cognitive functioning [31 –33]. Of note, effect sizes and explained variance generally also were small. The Rotterdam Scan Study, for example, showed that only subjects with at least 5 microbleeds (i.e., less than 2% of the study population) had significantly worse performance than subjects without microbleeds, again with modest effects (i.e., 0.5 MMSE points) [32]. Similarly, a meta-analysis including both population and hospital based cohorts, observed small effects for the presence versus absence of microbleeds (0.3 MMSE points) [34]. Another meta-analysis evaluating the effect of microbleeds on the cognitive profile in patients diagnosed with AD showed nonsignificant small effects (effect size (hedges g) = –0.155 (CI; – 0.465 0.155)) [35].

Previous studies on the impact of non-lacunar and lacunar infarcts and ICH on cognition largely focused on acute post-stroke deficits [36 –40]. Fewer studies reported on so called “silent infarcts” that appear on CT or MRI without a patient having experienced acute stroke symptoms connected to this lesion. The latter are clearly most relevant for the present study, but mainly focused on cognitive decline over time [41 –43] or just used the MMSE as outcome measurement. A population-based cohort, showed that patients with silent brain infarcts scored 1.6 MMSE points less than patients without silent brain infarcts [44]. The results of the ISSYS study, investigating silent strokes on brain imaging in hypertensive patients have not been published yet [45]. We show that also on top of co-existing AD pathology, presence of non-lacunar infarcts is associated with worse performance on information processing speed.

Previous cross-sectional studies evaluating the relation between vascular brain injury and cognition in a memory clinic setting mostly included patients according to specific diagnoses, such as probable AD or vascular dementia. In general, the presence of WMH was associated with a small decrease in cognitive functioning, observed across different cognitive domains, mostly with effect sizes around 0.1 in patients with AD and 0.25 in patients with MCI [46]. Regarding microbleeds in memory clinic populations, only very high burden (e.g., 8 or more microbleeds, which occurred in a subset of only 5% of patients with AD [47]) has been associated with worse cognitive performance [47, 48]. The other forms of vascular brain injury are less studied in memory clinic populations, only showing a slightly worse cognitive performance for lacunar infarcts, not statistically significant [49].

Taken together, the results of our study converge with those of previous studies reviewed above. Although there are significant associations between different forms of vascular brain injury and cognition, effect sizes were generally small. There may be exceptions, of course, particularly in individual patients with exceptionally high lesion burden. Such rare cases will have little impact on the overall results of a large cohort study such as ours, but could still be clinically relevant. Hence, overall, the variance in cognitive performance attributable to the presence or absence of different forms of vascular brain injury is relatively small, which raises questions to their diagnostic value in understanding the particular cognitive profile of patients presenting at a memory clinic. Yet, this clearly does not negate the value of MRI in a memory clinic population. Apart from understanding why a patient has a particular cognitive profile, the pattern of vascular brain injury on MRI can provide important leads on the underlying etiology, for example cerebral amyloid angiopathy. Moreover, the pattern and burden of vascular brain injury should guide decisions on vascular risk management [50], which may be particularly important in vulnerable individuals such as memory clinic patients. In light of the current results, the latter may currently represent the main clinical value of detecting vascular brain injury on MRI in memory clinic patients.

The strengths of our TRACE-VCI study include its large number of included memory clinic patients, with different levels of cognitive dysfunction, vascular brain injury and clinical diagnosis, with an extensive clinical and MRI evaluation. This inclusion procedure supports generalizability of our findings to patients who attend a memory clinic with any form of vascular brain injury and allow an unbiased assessment of the potential impact of vascular brain injury in this clinical setting. Of note, however, our findings might not be generalizable to VCI in other settings; such as for example post-stroke cognitive impairment. Moreover, the VUMC and UMCU are tertiary referral clinics (e.g., for a second opinion), with some overrepresentation of relatively young patients.

CSF was available in a substantial subset of patients, allowing us to explore the impact of vascular brain injury on cognition both in the absence and presence of biomarker evidence of concomitant AD pathology. CSF biomarkers amyloid-β₄₂ (reduced) and tau (increased) levels are sensitive and specific to identify AD pathology, also in de presence of vascular brain injury [51]. A limitation of the study is that there were some differences in neuropsychological tests that were applied in different patients. We decided not to impute or recode missing variables, since the available raw data was still relatively complete. Furthermore, in 31% of the patients a T2*-weighted sequence was performed without SWI, whereas SWI was performed in the others. Differences in sensitivity between these protocols may have affected microbleed detection. Moreover, the use of a categorical measure of microbleed burden (none, 1–4, ≥5) for analytical purposes may have underestimated the impact of microbleeds in a (small) subgroup of patients with very high microbleed counts.

Another limitation might be that post hoc analyses created small subgroups. Nonetheless, despite a lower power, these analyses showed the same pattern of cognitive profile in patients with a negative or positive CSF biomarker AD profile and also in patients with dementia, MCI, and NOCI. Finally, we did not have a reference group without any vascular injury, which is inherent to the design of TRACE-VCI study. However, it should be noted that our primary aim was to compare different forms of vascular brain injury rather than comparing the cognitive profile of patients with vascular brain injury to patients without any form of vascular brain injury. Yet, due to this design, the effect of vascular burden might be underestimated.

In conclusion, different forms of vascular brain injury on MRI explain relatively little of the variance in cognition in memory clinic patients, regardless of CSF biomarker AD status and severity of cognitive dysfunction. The diagnostic value of MRI in understanding the cognitive profile of individual patients according to type and severity of vascular brain injury in this setting may therefore often be limited. Nevertheless, MRI can help to identify underlying etiologies and direct patient management.

Footnotes

ACKNOWLEDGMENTS

The authors would like to express their thanks to all the members of the TRACE-VCI study group (in alphabetical order, per department);

VU University Medical Center, Amsterdam, The Netherlands:

Alzheimer Center and Department of neurology: M.R. Benedictus, J. Bremer, W.M. van der Flier, A.E. Leeuwis, J. Leijenaar, N.D. Prins, P. Scheltens, B.M. Tijms.

Department of Radiology and Nuclear Medicine: F. Barkhof, M.P. Wattjes.

Department of Clinical Chemistry: C.E. Teunissen.

Department of Medical Psychology: T. Koene.

University Medical Center Utrecht, Utrecht, The Netherlands:

Department of Neurology: E. van den Berg, G.J. Biessels, J.M.F. Boomsma, L.G. Exalto, D.A. Ferro, C.J.M. Frijns, O.N. Groeneveld, R. Heinen, N.M van Kalsbeek, J.H. Verwer.

Department of Radiology/Image Sciences Institute: J. de Bresser, H.J. Kuijf.

Department of Geriatrics: H.L. Koek.

Onze Lieve Vrouwe Gasthuis (OLVG) West, Amsterdam, The Netherlands:

Department of Neurology: J.M.F. Boomsma, H.C. Weinstein.

Erasmus MC University Medical Center, Rotterdam, The Netherlands:

Department of Neurology: E. van den Berg.

Leiden University Medical Center, Leiden, The Netherlands:

Department of Radiology: J. de Bresser.

National Institute for Health Research (NIHR) and University College London Hospitals NHS Foundation Trust (UCLH) biomedical research center, London, United Kingdom:

Department of Radiology: F. Barkhof.

The TRACE-VCI study is supported by Vidi grant 91711384 from ZonMw, The Netherlands, Organisation for Health Research and Development and grant 2010T073 from the Dutch Heart Association to Geert Jan Biessels.

Research of the VUMC Alzheimer center is part of the neurodegeneration research program of the Amsterdam Neuroscience. The VUMC Alzheimer Center is supported by Stichting Alzheimer Nederland and Stichting VUMC fonds. The clinical database structure was developed with funding from Stichting Dioraphte. F. Barkhof is supported by the National Institute for Health Research (NIHR) and University College London Hospitals NHS Foundation Trust (UCLH) biomedical research centre, London, United Kingdom.

Authors’ disclosures available online ().

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

References

Gorelick

, Scuteri

, Black

, Decarli

, Greenberg

, Iadecola

, Launer

, Laurent

, Lopez

, Nyen-huis

, Petersen

, Schneider

, Tzourio

, Arnett

, Bennett

, Chui

, Higashida

, Lindquist

, Nilsson

, Roman

, Sellke

, Seshadri

; 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 (2011) Vascular contributions to cognitive impairment and dementia: A statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 42, 2672–2713.

Biesbroek

, Weaver

, Biessels

(2017) Lesion location and cognitive impact of cerebral small vessel disease. Clin Sci (Lond) 131, 715–728.

Lawrence

, Patel

, Morris

, MacKinnon

, Rich

, Barrick

, Markus

(2013) Mechanisms of cognitive impairment in cerebral small vessel disease: Multimodal MRI results from the St George’s cognition and neuroimaging in stroke (SCANS) study. PLoS One 8, e61014.

Staals

, Booth

, Morris

, Bastin

, Gow

, Corley

, Redmond

, Starr

, Deary

, Wardlaw

(2015) Total MRI load of cerebral small vessel disease and cognitive ability in older people. Neurobiol Aging 36, 2806–2811.

Vasquez

, Zakzanis

(2015) The neuropsychological profile of vascular cognitive impairment not demented: A meta-analysis. J Neuropsychol 9, 109–136.

Suades-Gonzalez

Jodar-Vicente

, Perdrix-Solas

(2009) Memory deficit in patients with subcortical vascular cognitive impairment versus Alzheimer-type dementia: The sensitivity of the ‘word list’ subtest on the Wechsler Memory Scale-III. Rev Neurol 49, 623–629.

Boomsma

JMF

, Exalto

, Barkhof

, van den Berg

, de Bresser

, Heinen

, Koek

, Prins

, Scheltens

, Weinstein

, van der Flier

, Biessels

(2017) Vascular cognitive impairment in a memory clinic population: Rationale and design of the “Utrecht-Amsterdam Clinical Features and Prognosis in Vascular Cognitive Impairment” (TRACE-VCI) study. JMIR Res Protoc 6, e60.

Fazekas

, Chawluk

, Alavi

, Hurtig

, Zimmerman

(1987) MR signal abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. AJR Am J Roentgenol 149,351–356.

O’Brien

, Erkinjuntti

, Reisberg

, Roman

, Sawada

, Pantoni

, Bowler

, Ballard

, DeCarli

, Gorelick

, Rockwood

, Burns

, Gauthier

, DeKosky

(2003) Vascular cognitive impairment. Lancet Neurol 2, 89–98.

10.

Firbank

, Teodorczuk

, van der Flier

, Gouw

, Wallin

, Erkinjuntti

, Inzitari

, Wahlund

, Pantoni

, Poggesi

, Pracucci

, Langhorne

, O'Brien

, LADIS group (2012) Relationship between progression of brain white matter changes and late-life depression: 3-year results from the LADIS study. Br J Psychiatry 201, 40–45.

11.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

12.

Hughes

, Berg

, Danziger

, Coben

, Martin

(1982) A new clinical scale for the staging of dementia. Br J Psychiatry 140, 566–572.

13.

Aalten

, Ramakers

, Biessels

, de Deyn

, Koek

, OldeRikkert

, Oleksik

, Richard

, Smits

, van Swieten

, Teune

, van der Lugt

, Barkhof

, Teu-nissen

, Rozendaal

, Verhey

, van der Flier

(2014) The Dutch Parelsnoer Institute-Neurodegenerative diseases; methods, design and baseline results. BMC Neurol 14, 254.

14.

Van der Elst

, van Boxtel

, van Breukelen

, Jolles

(2005) Rey’s verbal learning test: Normative data for 1855 healthy participants aged 24-81 years and the influence of age, sex, education, and mode of presentation. J Int Neu-ropsychol Soc 11, 290–302.

15.

Lindeboom

, Schmand

, Tulner

, Walstra

, Jonker

(2002) Visual association test to detect early dementia of the Alzheimer type. J Neurol Neurosurg Psychiatry 73,126133.

16.

Corrigan

, Hinkeldey

(1987) Relationships between parts A and B of the Trail Making Test. J Clin Psychol 43, 402–409.

17.

JRS (1935) Studies of interference in serial verbal reactions. J Exp Psychol 18, 643–662.

18.

Van der Elst

, Van Boxtel

, Van Breukelen

, Jolles

(2008) Detecting the significance of changes in performance on the Stroop Color-Word Test, Rey's Verbal Learning Test, and the Letter Digit Substitution Test: The regression-based change approach. J Int Neuropsychol Soc 14, 71–80.

19.

Deelman

, Liebrand

, Koning-Haanstra

, van den Burg

(1980) Measurements of aphasic disorders. A brief description of the SAN-battery. Gerontologie 11, 17–21.

20.

Moses

JA Jr

, Pritchard

, Adams

(1997) Neuropsy-chological information in the Wechsler Adult Intelligence Scale-Revised. Arch Clin Neuropsychol 12, 97–109.

21.

van der Elst

, van Boxtel

, van Breukelen

, Jolles

(2006) The Letter Digit Substitution Test: Normative data for 1,858 healthy participants aged 24-81 from the Maastricht Aging Study (MAAS): Influence of age, education, and sex. J Clin Exp Neuropsychol 28, 998–1009.

22.

Warrington

, James

(1991) A new test of object decision: 2D silhouettes featuring a minimal view. Cortex 27, 370–383.

23.

Mulder

, Verwey

, van der Flier

, Bouwman

, Kok

, van Elk

, Scheltens

, Blankenstein

(2010) Amyloid-beta(1-42), total tau, and phosphorylated tau as cerebrospinal fluid biomarkers for the diagnosis of Alzheimer disease. Clin Chem 56, 248–253.

24.

Duits

, Teunissen

, Bouwman

, Visser

, Matts-son

, Zetterberg

, Blennow

, Hansson

, Minthon

, Andreasen

, Marcusson

, Wallin

, Rikkert

, Tso-laki

, Parnetti

, Herukka

, Hampel

, De Leon

, Schroder

, Aarsland

, Blankenstein

, Scheltens

, van der Flier

(2014) The cerebrospinal fluid “Alzheimer profile”: Easily said, but what does it mean? Alzheimers Dement 10, 713–723 e712.

25.

Wardlaw

, Smith

, Biessels

, Cordonnier

, Fazekas

, Frayne

, Lindley

, O'Brien

, Barkhof

, Benavente

, Black

, Brayne

, Breteler

, Chabriat

, Decarli

, de Leeuw

, Doubal

, Duering

, Fox

, Green-berg

, Hachinski

, Kilimann

, Mok

, Oostenbrugge

, Pantoni

, Speck

, Stephan

, Teipel

, Viswanathan

, Werring

, Chen

, Smith

, van Buchem

, Norrv-ing

, Gorelick

, Dichgans

, STandards for ReportIng Vascular changes on nEuroimaging (STRIVE v1) (2013) Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol 12, 822–838.

26.

Yesavage

, Brink

, Rose

, Lum

, Huang

, Adey

, Leirer

(1982) Development and validation of a geriatric depression screening scale: A preliminary report. J Psychiatr Res 17, 37–49.

27.

Gunning-Dixon

, Raz

(2000) The cognitive correlates of white matter abnormalities in normal aging: A quantitative review. Neuropsychology 14, 224–232.

28.

Lampe

, Kharabian-Masouleh

, Kynast

, Arelin

, Steele

, Loffler

, Witte

, Schroeter

, Villringer

, Bazin

(2019) Lesion location matters: The relationships between white matter hyperintensities on cognition in the healthy elderly. J Cereb Blood Flow Metab 39, 36–43.

29.

Kloppenborg

, Nederkoorn

, Geerlings

, van den Berg

(2014) Presence and progression of white matter hyperintensities and cognition: A meta-analysis. Neurology 82, 2127–2138.

30.

Verdelho

, Madureira

, Ferro

, Basile

, Chabriat

, Erkinjuntti

, Fazekas

, Hennerici

, O'Brien

, Pantoni

, Salvadori

, Scheltens

, Visser

, Wahlund

, Waldemar

, Wallin

, Inzitari

, LADIS Study (2007) Differential impact of cerebral white matter changes, diabetes, hypertension and stroke on cognitive performance among non-disabled elderly The LADIS study. J Neurol Neurosurg Psychiatry 78, 1325–1330.

31.

Charidimou

, Werring

(2012) Cerebral microbleeds and cognition in cerebrovascular disease: An update. JNeu-rol Sci 322, 50–55.

32.

Poels

, Ikram

, van der Lugt

, Hofman

, Niessen

, Krestin

, Breteler

, Vernooij

(2012) Cerebral microbleeds are associated with worse cognitive function: The Rotterdam Scan Study. Neurology 78, 326333.

33.

Qiu

, Cotch

, Sigurdsson

, Jonsson

, Jonsdottir

, Sveinbjrnsdottir

, Eiriksdottir

, Klein

, Harris

, van Buchem

, Gudnason

, Launer

(2010) Cerebral microbleeds, retinopathy, and dementia: The AGES-Reykjavik Study. Neurology 75, 2221–2228.

34.

, Yuan

, Yang

, Qin

, Yang

, Li

, Fan

, Hu

(2017) The significant effects of cerebral microbleeds on cognitive dysfunction: An updated meta-analysis. PLoS One 12, e0185145.

35.

Sepehry

, Rauscher

, Hsiung

, Lang

(2016) Microbleeds in Alzheimer’s disease: A neuropsychological overview and meta-analysis. Can J Neurol Sci 43, 753–759.

36.

Stebbins

, Nyenhuis

, Wang

, Cox

, Freels

, Bangen

, deToledo-Morrell

, Sripathirathan

, Moseley

, Turner

, Gabrieli

, Gorelick

(2008) Gray matter atrophy in patients with ischemic stroke with cognitive impairment. Stroke 39, 785–793.

37.

Gorelick

, Bowler

(2008) Advances in vascular cognitive impairment 2007. Stroke 39, 279–282.

38.

Edwards

, Jacova

, Sepehry

, Pratt

, Benavente

(2013) A quantitative systematic review of domain-specific cognitive impairment in lacunar stroke. Neurology 80, 31539?

39.

Planton

, Raposo

, Danet

, Albucher

, Peran

, Pariente

(2017) Impact of spontaneous intracerebral hemorrhage on cognitive functioning: An update. Rev Neurol (Paris) 173, 481–489.

40.

Xiong

, Davidsdottir

, Reijmer

, Shoamanesh

, Roongpiboonsopit

, Thanprasertsuk

, Martinez-Ramirez

, Charidimou

, Ayres

, Fotiadis

, Gurol

, Blacker

, Greenberg

, Viswanathan

(2016) Cognitive profile and its association with neuroimaging markers of non-demented cerebral amyloid angiopathy patients in a stroke unit. J Alzheimers Dis 52, 171–178.

41.

Troncoso

, Zonderman

, Resnick

, Crain

, Plet-nikova

, O'Brien

(2008) Effect of infarcts on dementia in the Baltimore longitudinal study of aging. Ann Neurol 64, 168–176.

42.

Renjen

, Gauba

, Chaudhari

(2015) Cognitive impairment after stroke. Cureus 7, e335.

43.

Vermeer

, Prins

, den Heijer

, Hofman

, Koudstaal

, Breteler

(2003) Silent brain infarcts and the risk of dementia and cognitive decline. N Engl J Med 348, 12151222.

44.

Chen

, Wang

, Tang

, Wei

, Li

, Chen

, Wang

, Zhang

(2015) Association of white matter integrity and cognitive functions in patients with subcortical silent lacunar infarcts. Stroke 46, 1123–1126.

45.

Riba-Llena

, Jarca

, Mundet

, Tovar

, Orfila

, Lopez-Rueda

, Nafria

, Fernandez

, Castane

, Domingo

, Alvarez-Sabin

, Fernandez-Cortinas

, Maisterra

, Montaner

, Delgado

(2013) Investigating silent strokes in hypertensives: A magnetic resonance imaging study (ISSYS): Rationale and protocol design. BMC Neurol 13, 130.

46.

Van den Berg

, Geerlings

, Biessels

, Nederkoorn

, Kloppenborg

(2018) White matter hyperintensities and cognition in mild cognitive impairment and Alzheimer’s disease: A domain specific meta-analysis. J Alzheimers Dis 63, 515–527.

47.

Goos

, Kester

, Barkhof

, Klein

, Blankenstein

, Scheltens

, van der Flier

(2009) Patients with Alzheimer disease with multiple microbleeds: Relation with cerebrospinal fluid biomarkers and cognition. Stroke 40, 3455–3460.

48.

Seo

, Hwa Lee

, Kim

, Chin

, Sun Cho

, Yoon

, Na

(2007) Clinical significance of microbleeds in subcortical vascular dementia. Stroke 38, 1949–1951.

49.

Reijmer

, Leemans

, Caeyenberghs

, Heringa

, Koek

, Biessels

, Utrecht Vascular Cognitive Impairment Study Group (2013) Disruption of cerebral networks and cognitive impairment in Alzheimer disease. Neurology 80, 1370–1377.

50.

Smith

, Saposnik

, Biessels

, Doubal

, Fornage

, Gorelick

, Greenberg

, Higashida

, Kasner

, Seshadri

, American Heart Association Stroke Council; Council on Cardiovascular Radiology and Intervention; Council on Functional Genomics and Translational Biology; and Council on Hypertension (2017) Prevention of stroke in patients with silent cerebrovascular disease: A scientific statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 48, e44–e71.

51.

Wallin

, Kapaki

, Boban

, Engelborghs

, Hermann

, Huisa

, Jonsson

, Kramberger

, Lossi

, Malo-jcic

, Mehrabian

, Merighi

, Mukaetova-Ladinska

, Paraskevas

, Popescu

, Ravid

, Traykov

, Tsiv-goulis

, Weinstein

, Korczyn

, Bjerke

, Rosenberg

(2017) Biochemical markers in vascular cognitive impairment associated with subcortical small vessel disease - A consensus report. BMC Neurol 17, 102.

52.

Verhage

(1964) Intelligentie en leeftijd: Onderzoek bij Nederlanders van twaalf tot zevenenzeventig jaar. Van Gor-cum, Assen.

53.

Neary

, Snowden

, Gustafson

, Passant

, Stuss

, Black

, Freedman

, Kertesz

, Robert

, Albert

, Boone

, Miller

, Cummings

, Benson

(1998) Frontotemporal lobar degeneration: A consensus on clinical diagnostic criteria. Neurology 51, 1546–1554.

54.

Wadia

, Lang

(2007) The many faces of corticobasal degeneration. ParkinsonismRelatDisord 13 Suppl 3, S336–S340.

55.

Litvan

, Agid

, Calne

, Campbell

, Dubois

, Duvoisin

, Goetz

, Golbe

, Grafman

, Growdon

, Hallett

, Jankovic

, Quinn

, Tolosa

, Zee

(1996) Clinical research criteria for the diagnosis of progressive supranuclear palsy (Steele-Richardson-Olszewski syndrome): Report of the NINDS-SPSP international workshop. Neurology 47, 1–9.