Disrupted White Matter Integrity and Cognitive Functions in Amyloid-β Positive Alzheimer’s Disease with Concomitant Lobar Cerebral Microbleeds

Abstract

Background:

Lobar cerebral microbleeds (CMBs), which can impair white matter (WM), are often concomitant with definite Alzheimer’s disease (AD).

Objective:

To explore the features of cognitive impairments and WM disruptions due to lobar CMBs in patients with AD.

Methods:

There were 310 participants who underwent Florbetapir F18 (AV45) amyloid PET and susceptibility-weighted imaging. Participants with cognitive impairment and amyloid-β positive (ADCI) were included into three groups: ADCI without CMBs, with strictly lobar CMBs (SL-CMBs), and with mixed CMBs (M-CMBs). Tract-based spatial statistics were performed to detect the group differences in WM integrity.

Results:

There were 82 patients and 29 healthy controls finally included. A decreasing tendency in memory and executive performance can be found among HCs > no CMBs (n = 16) >SL-CMBs (n = 41) >M-CMBs (n = 25) group. Compared to no CMBs, M-CMBs group had significantly decreased fractional anisotropy in left anterior thalamic radiation (ATR), forceps major, forceps minor and inferior longitudinal fasciculus, bilateral inferior fronto-occipital fasciculus (IFOF), and superior longitudinal fasciculus. M-CMBs group also had lower fractional anisotropy in left ATR, IFOF, uncinate fasciculus, and forceps minor compared with SL-CMBs. Furthermore, analysis of Pearson correlation indicated damages in discrepant WMs were positively associated with impairment of memory, executive function, and attention.

Conclusion:

This study showed lobar CMBs had intensively aggravated cognitive impairments associated with extensive WM damages in definite AD. These findings highlight that lobar CMBs play an important role in AD progression and need to be taken into consideration for the early detection of AD.

Keywords

Alzheimer’s disease cerebral microbleeds cognitive impairment diffusion tensor imaging Florbetapir F18 amyloid PET

INTRODUCTION

The insidious onset of Alzheimer’s disease (AD) makes accurate early clinical diagnosis nearly impossible [1]. Recently, molecular imaging using positron emission tomography (PET) to track the spread of amyloid-β (Aβ) and the microtubule tau protein has allowed clinicians and researchers to stage AD antemortem [2]. In the National Institute on Aging and Alzheimer’s Association (NIA-AA) Research Framework [3], individuals with the presence of Aβ (rated by CSF Aβ or amyloid PET as in vivo biomarkers) were acknowledged as having the evidence of AD pathologic change and regarded as the early phase of the “Alzheimer’s continuum”. Even at a given AD biomarker profile, there are also several variabilities in the clinical AD progression. Among the genetic or/and the presence of other critical morbidities, vascular insults have been shown to exacerbate cognitive impairment [4, 5]. Pathological studies have suggested that cerebrovascular disease and amyloid burden may synergistically affect cognitive impairment [6, 7].

Cerebral microbleeds (CMBs) have emerged as a novel marker of cerebral small vessel disease in addition to white matter hyperintensities (WMHs) and lacunes in the past decade [8] which can be recognized as small hypointense foci on susceptibility-weighted imaging (SWI) [9]. The association between the presence of CMBs and the risk of dementia seems to be conflicted according to different meta-analysis. Bos et al. meta-analyzed a total of 8,739 participants and found no significant increased risk for CMBs to dementia [10]. In contrast, Charidimou et al. found CMBs were independently associated with incident dementia among three population-based studies [11]. Thus, to refine the characteristics of CMBs is of great importance in elucidating the contributions of CMBs to AD risk. Location of CMBs may predict the primary pathology: strictly lobar CMBs (SL-CMBs) is thought to correlate with cerebral amyloid angiopathy (CAA) and AD, non-lobar (deep/infratentorial) CMBs is mostly related to hypertensive vasculopathy (HV), and mixed (lobar with deep/infratentorial) CMBs (M-CMBs) may be associated with both HV and CAA [12 –14].

Lobar CMBs are consist of SL-CMBs and M-CMBs, demonstrating a link between amyloid pathology and neurovascular changes [15]. Some studies found that even single lobar CMB (regardless of strictly lobar or mixed CMBs) has yielded more than 2-fold incidence risk of AD but no significant-risk of non-Alzheimer dementia, illustrating the predictive value of lobar CMBs in AD [16, 17]. In addition, mixed CMBs were also considered to have a correlation with cognitive decline, even can serve as a biomarker for predicting the treatment response of cholinesterase inhibitors [18]. However, the mechanisms why lobar CMBs aggravate cognitive decline in AD has not been fully understood. Recently, the damage of white matter (WM) integrity is garnering ever-increasing attention. A few studies using diffusion tensor imaging (DTI) indicated that CMBs were related to poorer microstructural integrity of WMs even when only a single microbleed is present [8, 19]. Other studies reported higher CMBs burdens might play a role in subcortical WM atrophy by interrupting brain network connectivity [20, 21]. However, few studies have investigated the effect of lobar CMBs distributions on WMs microstructure in patients with pathologically definite AD.

The non-lobar CMBs (deep/infratentorial CMBs) which was associated with HV may involve the white matter damage in elders, but in this study, we intend to explore the features of cognitive impairments and disruptions of WM integrity and their correlation due to lobar CMBs which was more related to CAA and AD in patients with Aβ positive results by PET (cognition impairment due to AD, ADCI). The effect of pure deep/infratentorial CMBs may confound the results so we excluded the CMBs only in deep basal. We hypothesize that lobar CMBs lead an aggravation in multi-cognitive domains which are associated with extensive WM damages in patients with ADCI. This may provide the evidence that how vascular factors incorporate and contribute to the multifactorial pathophysiology of AD, gaining a better understanding of AD and aid in detection and treatment efforts.

MATERIALS AND METHODS

Participants

A total of 267 participants with cognitive impairment in the memory clinic of China-Japan Friendship Hospital as well as 43 healthy controls from community were recruited from October 2017 to October 2018. Standardized clinical evaluations including medical history interview, neurologic examination, and a battery of neuropsychological tests were performed by a team of neurologists that specializes in dementia for all participants. The neuropsychological tests assessed general mental status and cognitive domains including memory, attention, spatial processing, executive function, and language ability, which are provided in the Supplementary Material.

The diagnoses of cognitive impairment patients were made using the criteria for possible AD of the National Institute of Neurologic Disorders and Stroke-Alzheimer Disease and Related Disorders Association (NINCDS-ADRDA) [22] and the criteria of mild cognitive impairment due to AD from the NIA-AA [23], which were described as follows: 1) definite complaints of memory declined, preferably confirmed by an informant; 2) objective cognitive performances in single or multiple domains including memory documented by neuropsychological tests scores were below or equal to 1.5 SD of age- and education-adjusted norms; 3) a Clinical Dementia Rating (CDR) score≥0.5. Cognitively normal elders were those who had no complaints of cognition and normal objective cognitive performances as well as a CDR score of 0.

Exclusion criteria for all participants were as follows: 1) a history of stroke, traumatic brain injury, neurological/psychiatric diseases, and other central nervous system diseases that may lead to cognitive impairment, such as dementia with Lewy bodies, frontotemporal dementia, and so on; 2) major depression (Hamilton Depression Rating Scale score > 24 points); 3) other systemic diseases including thyroid dysfunction, syphilis, severe anemia, or human immunodeficiency virus (HIV) that may cause cognitive impairment; 4) addictions or treatments that would influence cognitive ability; 5) severe visual or auditory disabilities treatable dementia; 6) leukoencephalopathy as a result of other causes, such as subcortical ischemic vascular disease, normal pressure hydrocephalus, multiple sclerosis, brain irradiation, and metabolic diseases.

This study was approved by the Ethical Review Board of China-Japan Friendship Hospital. All the participants signed the informed consent. Data during the study are available from the corresponding author by request.

Amyloid PET

All patients underwent Florbetapir F18 (AV45) PET on the Discovery TM PET/CT Elite scanner (General Electrics) at Beijing Tiantan Hospital, Capital Medical University. The diagnosis of amyloid-β positive (Aβ+) was based on both visual interpretations of elevated binding in the neocortex and semi-quantitative assessment (standard uptake value ratios SUVR > 1.11) [24]. The visual reading was conducted by two nuclear medicine physicians with no access to clinical information, rating each PET image independently. SUVRs were calculated using the cerebellar gray matter reference region to normalize mean activity from 50 to 70 min. The detailed parameters are provided in the Supplementary Material.

Standard MRI protocol

All MRI data were acquired on a 3.0T Siemens Tim MRI scanner in the Imaging Center for Brain Research, Beijing Normal University. The time interval between MRI and amyloid PET was no longer than two weeks. T1-weighted, T2-weighted, fluid-attenuated inversion recovery (FLAIR), DTI, and SWI images were obtained. Two different radiologists assessed the anatomical MRI scans and gave the nearly same reports. The acquisitions of MRI data are detailed in the Supplementary Material.

Here, CMBs were defined as small hypointense foci < 10 mm in size on SWI according to the Standards for Reporting Vascular changes on neuroimaging (STRIVE) consensus [25]. CMBs were counted throughout the brain and their topographical distribution was classified as “deep,” “infratentorial,” or “lobar” according to the microbleed anatomical rating scale [26]. For this study, lobar CMBs were categorized as follows: 1) SL-CMBs, whereby CMBs were restricted to “lobar” locations, 2) M-CMBs, whereby CMBs were in both “lobar” and “deep” locations. CMBs only in deep basal (deep and/or infratentorial) locations were excluded.

In addition, the visual Fazekas scale was used on FLAIR images to rate the severity of WMHs into mild (grade 1), moderate (grade 2), and severe (grade 3) WMHs [27] and every participant was estimated to exclude the participants with Fazekas scales ≥2 grade and/or lacunas > 3 who met the brain imaging criteria of subcortical ischemic vascular disease [28].

DTI image analysis

Preprocessing

All of the DTI image preprocessing and analyses described below were implemented using a pipeline tool for diffusion MRI, named “Pipeline for Analyzing braiN Diffusion imAges” (PANDA) [29]. The similar procedure was shown in our previous studies [30, 31]. The detailed procedures are provided in the Supplementary Material.

Voxel-wise TBSS statistical analysis

The voxel-wise statistical analysis in the tract-based spatial statistics (TBSS) compares group differences only on the WM skeleton such that it provides better sensitivity, objectivity, and interpretability of analysis for multi-subject DTI studies [32]. The TBSS analyses of fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD) were performed using the FMRIB software library (FSL 4.1.4; http://www.fmrib.ox.ac.uk/fsl). Briefly, the following five-step process was first performed on the FA images: 1) the FA image of each subject was aligned to a pre-identified target FA image by nonlinear registration; 2) all of the aligned FA images were transformed onto the MNI152 template by affine registration; 3) a mean FA image and its skeleton (mean FA skeleton) was created from the images of all the subjects; 4) individual subjects’ FA images were projected onto the skeleton; 5) voxel-wise statistics across subjects were calculated for each point on the common skeleton. The voxel-wise statistics in TBSS were analyzed using a permutation-based inference tool for nonparametric statistical thresholding (the ‘randomize’ tool, part of FSL).

In this study, voxel-wise group comparisons were performed between the ADCI without CMBs versus SL-CMBs, and the ADCI without CMBs versus M-CMBs, and the SL-CMBs versus M-CMBs, separately. The covariance includes age, gender, education, numbers of lacunar infarcts, presence of hypertension, and WMHs which were rated visually on axial FLAIR images using the Fazekas scale. The mean FA skeleton was used as a mask (threshold at a mean FA value of 0.2), and the number of permutations was set to 5000. The significance threshold for between-group differences was set at p < 0.05 (family-wise error, FWE corrected for multiple comparisons) using the threshold-free cluster enhancement option in the ‘randomize’ permutation-testing tool in FSL.

Region-of-interest-wise statistical analysis

We used the digital WM atlas JHU ICBM-DTI-81 (http://cmrm.med.jhmi.edu/), a probabilistic atlas generated by mapping DTI data to a template image. The JHU-WM atlas was overlaid on the WM skeleton of each subject in the CBM-DTI-81 space, such that each skeleton voxel could be categorized into one of the major tracts. Next, the values of at the skeleton voxels within each tract can be calculated. The regional FA, MD, AD, and RD were calculated by averaging the values within regions of the WM atlas with significant results based on the above voxel-wise TBSS statistical analysis.

Statistical analysis

Statistical analyses were performed using Windows SPSS software package (version 24.0, IBM). Categorical variables were compared with the chi-square test and Fisher’s exact test. Group differences on demographic measures were tested using the one-way analysis of variance (ANOVA) for continuous data. Analysis of covariance (ANCOVA) using age, sex, and education as covariates were performed to evaluate the group differences in neuropsychological assessments. The significance threshold for post-hoc pairwise comparisons were performed Tukey’s corrected for multiple comparisons. Pearson correlation analyses were performed to explore the relationship between regional metrics (FA, MD, AD, and RD) of white matters and neuropsychological tests with significant group differences (age, sex, and education were included as covariates). A significance level of p < 0.05 was applied in all these comparisons and correlations.

RESULTS

Demographic and neuropsychological measurements

Among a total of 267 participants with cognitive impairment, we excluded 112 amyloid-β negative (Aβ-) patients, and the remaining 155 patients with Aβ+ were defined as cognitive impairment due to AD (ADCI). Then we excluded the patients with CMBs only in deep basal locations (n = 17) and patients with Fazekas scales≥2 (n = 40), and/or lacunas > 3 (n = 16) which may influence the WM and cognitive functions besides lobar CMBs. Finally, based on the appearance of lobar CMBs and Aβ+ PET, we divided ADCI into three different groups: ADCI without CMBs (n = 16), ADCI with SL-CMBs (n = 41), and ADCI with M-CMBs (n = 25). Among the 43 cognitively normal elders, we excluded 5 subjects with Aβ+, 4 with Fazekas scales ≥2, 4 with lacunas > 3, and 1 with CMBs. 29 participants were finally included as HCs. The detailed selection of subjects is shown in Fig. 1.

Fig. 1

Flowchart shows the selection of subjects. A) The results of ¹⁸F-AV45 PET (A-1: Aβ positive, A-2: Aβ negative). B) The scores of Fazaekas scale (B-1 Fazekas score = 1, B-2: Fazekas score = 2, B-3: Fazekas score = 3). ADCI, cognitive impairment due to Alzheimer’s disease; CMBs, cerebral microbleeds; SL-CMBs, strictly lobar cerebral microbleeds; M-CMBs, mixed cerebral microbleeds; PET, positron emission tomography; FLAIR, fluid attenuated inversion recovery; SWI, susceptibility-weighted imaging.

Table 1 summarized the baseline characteristics of all participants. There were no significant differences among four groups in age, sex, education level, and the prevalence of APOE ɛ4. History of hypertension, existence of WMHs (Fazekas scale = 1), and numbers of lacunes (1–3) were significantly different among four groups. Post-hoc comparisons also showed differences in history of hypertension between ADCI with M-CMBs and ADCI without CMBs. WMHs and numbers of lacunes between ADCI with M-CMBs and ADCI without CMBs, as well as M-CMBs and HCs were also significantly different after post-hoc analysis.

Table 1

Baseline characteristics of all participants

	HCs (n = 29)	ADCI without CMBs (n = 16)	ADCI with SL-CMBs (n = 41)	ADCI with M-CMBs (n = 25)	F-value/χ²	p
Age (y)	69.4±8.3	73.4±6.4	71.1±8.7	70.6±9.1	0.801	0.496
APOEɛ4 (+/-, %)	5/18	7/8	23/18	11/14	7.091	0.071
Education (y)	11.8±0.7	11.6±1.1	12.0±0.7	12.0±0.9	0.092	0.962
Gender (M/F)	12/17	6/10	15/26	11/14	0.433	0.936
Cardiovascular risk factors
Hypertension¹ (%)	7 (24%)	2 (13%)	13 (32%)	13 (52%)	8.279	0.041^*b
Diabetes mellitus¹ (%)	2 (7%)	3 (19%)	5 (12%)	5 (20%)	2.587	0.482
Hyperlipidemia¹ (%)	9 (31%)	1 (6%)	10 (24%)	6 (24%)	3.676	0.305
Small vessel disease markers
WMHs¹ (%)	13 (45%)	6 (38%)	29 (71%)	21 (84%)	16.52	0.001^*a,b
Lacunas¹ (%)	5 (17%)	2 (13%)	16 (39%)	20 (80%)	28.71	0.000^*a,b
Number of lacunas	0.4±1.0	0.2±0.6	1.1±1.6	1.4±1.6	11.03	0.000^*a,b

Values are mean±standard deviation or Nos. of participants. ^*p < 0.05; ¹The group differences were detected by Fisher’s exact test; Post-hoc showed ^athe differences between M-CMBs and HCs; ^bthe differences between M-CMBs and ADCI without CMB. HCs, healthy controls; ADCI, cognitive impairment due to Alzheimer’s disease; CMBs, cerebral microbleeds; SL-CMBs, strictly lobar cerebral microbleeds; M-CMBs, mixed cerebral microbleeds; APOE, apolipoprotein E; WMHs, white matter hyperintensities.

Table 2 showed the neuropsychological measurements for all participants. The scores in Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), episodic memory domain including auditory verbal learning test immediate recall (AVLT-I), delay recall (AVLT-D) and total recall (AVLT-T), executive domain including Trail Making Test Parts B (TMT-B) and Stroop Color and Word Test-right percent (SCWT-right), Symbol Digit Modalities Test (SDMT) which indicated attention performance, and category verbal fluency test (CVFT) which indicated language ability were different among four groups. Post-hoc showed there were significant differences between M-CMBs group and HCs in MMSE, MoCA, AVLT-I, AVLT-D, AVLT-T, SCWT-right, TMT-B, SDMT, and CVFT, which are mainly all of the cognitive domains. The results after comparing SL-CMBs and HCs were similar as those when compared M-CMBs with HCs, except the scores of TMT-B. Comparing ADCI without CMBs and HCs, the differences mainly existed in global mental status (MMSE, MoCA) and episodic memory domain (AVLT-I, AVLT-D, AVLT-T). SCWT-right, TMT-B, and SDMT also showed significant differences between M-CMBs and ADCI without CMBs, while there were no differences between SL-CMBs and the other two ADCI groups. Of note, we found a decreased tendency in memory and executive performances among HCs > ADCI without CMBs > SL-CMBs > M-CMBs (Fig. 2), indicating more severe disruption in executive function with the appearance of lobar and mix CMBs in AD.

Table 2

Neuropsychological measurements for all participants

	HCs (n = 29)	ADCI without CMBs (n = 16)	ADCI with SL-CMBs (n = 41)	ADCI with M-CMBs (n = 25)	F-value (partial Eta squared)	Effect sizes	p
MMSE	25.5±3.9	19.2±7.4	18.8±7.6	20.6±6.3	6.21	0.16	0.001^*a,b,c
MoCA	23.9±4.6	15.9±7.1	14.9±7.1	16.5±6.7	5.08	0.13	0.003^*a,b,c
Memory domain
AVLT-I	13.4±4.7	9.6±3.9	9.6±4.4	9.7±4.6	4.87	0.13	0.003^*a,b,c
AVLT-D	3.0±2.0	1±1.3	1.4±1.7	1.2±2.2	5.78	0.15	0.001^*a,b,c
AVLT-Total	19.6±8.3	11.1±5.7	11.9±6.9	11.9±8.5	6.97	0.18	0.000^*a,b,c
R-O-delayed	12.3±1.7	7.9±2.1	8.9±1.5	8.7±1.9	1.21	0.05	0.313
Executive domain
SCWT - right	44.4±6.2	43.4±6.4	35.8±15.1	33.0±16.9	5.21	0.16	0.002^*a,b,d
TMT-B	204.6±87.7	214.1±85.1	272.8±124.2	419.8±141.1	3.31	0.12	0.012^*a,d
Attention domain
SDMT	23.9±10.4	21.7±12.7	15.9±9.3	16.1±10.7	4.12	0.14	0.009^*a,b,d
Spatial processing
Clock drawing	22.3±6.2	19.6±8.5	19.6±8.9	16.5±10.2	2.24	0.07	0.089
Language domain
Verbal Fluency Test	33.7±11.3	28.1±9.9	25.8±12.3	19.7±6.3	3.76	0.11	0.013^*a,b
Boston naming	22.7±4.8	22.1±5.4	19.3±6.9	–1.3±1.6	2.16	0.07	0.099

^*p < 0.05; Values are mean±standard deviation. The comparison of neuropsychological scores among the four groups was performed with an analysis of covariance (ANCOVA), with age, education, and gender as covariate. Post-hoc showed ^a,b,cthe differences between M-CMBs/HCs, SL-CMBs/HCs, ADCI without CMBs/HCs; ^dthe differences between M-CMBs and ADCI without CMBs. HCs, healthy controls; ADCI, cognitive impairment due to Alzheimer’s disease; CMBs, cerebral microbleeds; SL-CMBs, strictly lobar cerebral microbleeds; M-CMBs, mixed cerebral microbleeds; MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment; AVLT-I, D, T, auditory verbal learning test- immediate recall; -delay recall; -total recall; R-O delayed, Rey-Osterrieth Complex Figure-delayed recall; TMT-B, Trail Making Test Parts B; SDMT, Symbol Digit Modalities Test; SCWT-right, Stroop Color and Word Test-right percent

Fig. 2

Group differences among ADCI without CMBs, with SL-CMBs, with M-CMBs and HCs groups (^*p < 0.05, Tukey’s corrected). A) Group differences in AVLT immediate recall; B) group differences in TMT-B time; C) group differences in SCWT-right performance. HCs, healthy controls; CMBs, cerebral microbleeds; SL-CMBs, strictly lobar cerebral microbleeds; M-CMBs, mixed cerebral microbleeds; AVLT, auditory verbal learning test; SDMT, Symbol Digit Modalities Test; TMT-B, Trail Making Test Parts B; SCWT-right, Stroop Color and Word Test-right percent.

WM skeleton voxel-wise TBSS comparisons

After adjusted for age, sex, education years, history of HT, numbers of lacunes, and Fazekas scores of WMHs, compared to ADCI without CMBs, patients with M-CMBs had significantly decreased FA in left anterior thalamic radiation (ATR), forceps major, forceps minor and inferior longitudinal fasciculus (ILF), bilateral inferior fronto-occipital fasciculus (IFOF), and superior longitudinal fasciculus (FWE corrected, p < 0.05, see Fig. 3A). Besides, the FAs of left ATR, IFOF, uncinate fasciculus, and forceps minor which were mainly in left hemisphere were much lower in patients with ADCI with M-CMBs when compared with SL-CMBs (FWE corrected, p < 0.05, see Fig. 3B). When compared ADCI without CMBs and ADCI with SL-CMBs, we found no significant differences in FA across the global white matters (FWE corrected, p > 0.05, see Fig. 3C). In addition, we found no significant difference in the measures of MD, AD, and RD between any group comparisons.

Fig. 3

TBSS results of the FA between three ADCI groups after adjusted for age, sex, education years, history of hypertension, numbers of lacunes, and Fazekas scores of white matter hyperintensities. Green represents the mean white matter skeleton of all subjects. Red voxels (thickened for better visibility) represent the white matter regions with increased fractional anisotropy in the former compared with the latter (p < 0.05, family-wise error corrected). ADCI, cognitive impairment due to Alzheimer’s disease; CMBs, cerebral microbleeds; SL-CMBs, strictly lobar cerebral microbleeds; M-CMBs, mixed cerebral microbleeds.

Correlations between FA in discrepant WMs and behavior performances

When we compared the relationship with significantly different FA of the white matters and cognitive performances between ADCI without CMBs and patients with M-CMBs, there was a significant positive correlation between the score of AVLT immediate recall and the FA value of the left ATR (r = 0.39, p = 0.048, Fig. 4A). The decreasing of SCWT-right performance was correlated with the lowering FA values of left ATR (r = 0.53, p = 0.01, Fig. 4B) and left IFOF (r = 0.48, p = 0.03, Fig. 4C). In addition, the performance of SDMT were positively correlated with FA values of left ATR (r = 0.61, p = 0.007, Fig. 4D), forceps major (r = 0.48, p = 0.04; Fig. 4E), forceps minor (r = 0.53, p = 0.03, Fig. 4F), left IFOF (r = 0.56, p = 0.02, Fig. 4G) and left ILF (r = 0.52, p = 0.03, Fig. 4H).

Fig. 4

The correlations between neuropsychological scores and fractional anisotropy (FA) value of discrepant white matters between ADCI without CMBs and ADCI with M-CMBs. A) The correlation between AVLT-IR and FA of left ATR; B, C) the correlations between SCWT-right performance and FA of left ATR and left IFOF; D–H) the correlations between SDMT performance and FA of left ATR, forceps major, forceps minor, left IFOF and left ILF. ATR, anterior thalamic radiation; IFOF, inferior fronto-occipital fasciculus; ILF, inferior longitudinal fasciculus; AVLT-IR, auditory verbal learning test immediate recall; SCWT-right, Stroop Color and Word Test-right percent; SDMT, Symbol Digit Modalities Test; ADCI, cognitive impairment due to Alzheimer’s disease; CMBs, cerebral microbleeds; SL-CMBs, strictly lobar cerebral microbleeds; M-CMBs, mixed cerebral microbleeds.

Compared the relationship with significantly different WMs and cognitive performance between ADCI with SL-CMBs and M-CMBs, we found significant positive correlations between the scores of AVLT immediate recall and left ATR (r = 0.33, p = 0.04, Fig. 5A) and forceps minor (r = 0.41, p = 0.009, Fig. 5B), as well as the performance of SDMT in the same white matters as above (r_ATR . L=0.36, p = 0.048; r_{forceps minor} = 0.43, p = 0.02, Fig. 5C, D).

Fig. 5

The correlations between neuropsychological scores and fractional anisotropy (FA) value of discrepant white matter between ADCI with SL-CMBs and ADCI with M-CMBs. A, B) The correlation between AVLT-IR and FA of left ATR and Forceps minor; C, D) the correlation between SDMT and FA of left ATR and Forceps minor. ATR, anterior thalamic radiation; AVLT-IR, auditory verbal learning test immediate recall; SDMT, Symbol Digit Modalities Test. ADCI, cognitive impairment due to Alzheimer’s disease; CMBs, cerebral microbleeds; SL-CMBs, strictly lobar cerebral microbleeds; M-CMBs, mixed cerebral microbleeds.

DISCUSSION

This study evaluated the characteristics of cognitive impairments in multi-domains and the related disruptions of WM integrity due to lobar cerebral microbleeds (strictly lobar or mixed CMBs) in patients with Aβ+ AD. Firstly, there was a decreasing tendency in memory and executive functions in patients with lobar CMBs compared to no CMBs. Secondly, we found extensive damages in WM integrity across the whole brain in patients with AD with lobar CMBs, which are correlated with the declines in cognitive functions. Intriguingly, compared to strictly lobar CMBs, the presence of mixed CMBs resulted in the more widely cognitive function and WMs impairment which was restricted in the left hemisphere.

Currently, few studies have investigated the effect of lobar CMBs on cognition in patients with Aβ+ AD who were considered as the AD with definite diagnosis. In community studies, numerous SL-CMBs were associated with worse performance on cognitive tests [33], while M-CMBs had the greatest impact on the declines of global cognitive function, executive function, and visuo-construction compared with SL-CMBs and deep CMBs [34, 35]. In our study, we found a downward tendency in memory and executive functions gradually in no CMBs group > SL-CMBs group > M-CMBs, as well as significant differences between M-CMBs and ADCI without CMBs in executive function and attention which indicated that overlapping of the different pathologies of microbleeds may aggravate cognitive impairment in patients with AD.

There were only a few studies have found multiple CMBs (regardless of location) associated with disruptions of the cerebral network in patients with possible or probable AD [21, 36]. Few studies have focused on the effect of lobar microbleeds on WMs in patients with pathologically diagnosed AD. Here, after adjusting for age, sex, education years, history of HT, numbers of lacunes, and Fazekas scales of WMHs which may give effect on changes of WM, by using TBSS we found that in patients with Aβ+ AD, the WM tracts are more extensively disrupted in the M-CMBs group than SL-CMBs and no-CMBs groups. Compared to no CMBs, M-CMBs had significantly decreased FA in WM projecting to the prefrontal lobe such as left ATR, superior longitudinal fasciculus, forceps minor, and long course WM including bilateral IFOF and ILF. While compared to SL-CMBs, M-CMBs showed obvious damage on WM tracts of the left hemisphere such as left ATR, IFOF, and uncinate fasciculus. These findings indicated that the mixed CMBs give rise to more serious damage in WM. The above results of our study are consistent with previous studies which showed M-CMBs were associated with higher values of presence of lacunes, microbleeds, and degree of WMHs compared to CAA-CMBs and HV-CMBs [37, 38]. Most previous studies tended to show that different degrees of both CAA and HV are involved in this process [12, 14]. However, the underlying pathologies are still controversial in M-CMBs.

In M-CMBs groups, we found multiple WM tracts such as left ATR, IFOF, ILF, and forceps major and minor are correlated with the decline of attention. Meanwhile, left ATR was also found highly associated with declines in memory and executive and attention functions. CMBs were proved to predominantly affect executive and attention functions which are maintained by the integrity of frontal-subcortical circuits in previous studies. Liu et al. also showed the SL-CMBs were associated with lower FA in the internal capsule and corpus callosum which are mainly connected with the frontal lobe by TBSS [19]. Another TBSS study also found M-CMBs lead to disruptions of thalamocortical connectivity when compared with healthy controls [34]. A recent study [39] has shown that SL-CMBs has poorer performances in verbal memory and visuospatial and executive functions which were associated with significantly abnormal FA in frontal (32%) and subcortical (30%) regions compared to the control group. ATR, forceps major, and ILF are the major WM tracts of frontal-subcortical circuits [40 –42]. They project to the hippocampus, anterior cingulate, and prefrontal cortex which control long-term memory storage and executive and attention functions.

An interesting point in our results was that compared with SL-CMBs, there was a lateralization injury of WM integrity restricted to left hemisphere in M-CMBs group. Previous studies have found the network of left hemisphere exhibited much less efficient WM connections and age-related improvement compared with right side [43, 44]. These findings suggested that the networks in left hemisphere seem not to be as efficient, developed, and compensable as that in right side. These theories may partly explain why the integrity of WMs in the left hemisphere is more vulnerable to damage in our study when there are multiple pathologies involved in M-CMBs group. However, in our previous study, we found the significantly damaged tracts in the patients at the early stage of subcortical vascular cognitive impairment were mainly concentrated in the right hemisphere [45]. The distributions and connections of white matters in the bilateral hemisphere seem to be affected by various pathologies.

This is the initial study to detect additive effects of lobar CMBs which may be involved with different pathologies in company with Aβ on cognition and WMs integrity in patients with ADCI. Here, ¹⁸F-AV45 PET was used to screen patients with cognitive impairments which make the enrollment of patients with ADCI much more precise. In addition, a unified protocol and high-quality MRI have increased the reliability of this study. However, there are still several limitations. First, our results are based on cross-sectional data, which prevent us from making a causal inference. Second, it was a small sample study on a Chinese population. A long-term follow-up with multiple centers in a large sample can be conducted further.

In conclusion, lobar cerebral microbleeds (CMBs) may result in more damages of WM integrity and cognition functions in patients with Aβ+ AD and need to be taken into account for the early detection of AD. DTI may serve as an imaging marker in AD with multifactorial pathophysiology including cerebrovascular factors.

Footnotes

ACKNOWLEDGMENTS

The authors thank X.B.Z and Q.C for giving supports on reading PET images.

Authors’ disclosures available online ().

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

References

Hane

, Robinson

, Lee

, Bai

, Leonenko

, Albert

(2017) Recent progress in Alzheimer’s disease research, part 3: Diagnosis and treatment. J Alzheimers Dis 57, 645–665.

Dubois

, Feldman

, Jacova

, Hampel

, Molinuevo

, Blennow

, DeKosky

, Gauthier

, Selkoe

, Bateman

, Cappa

, Crutch

, Engelborghs

, Frisoni

, Fox

, Galasko

, Habert

, Jicha

, Nordberg

, Pasquier

, Rabinovici

, Robert

, Rowe

, Salloway

, Sarazin

, Epelbaum

, de Souza

, Vellas

, Visser

, Schneider

, Stern

, Scheltens

, Cummings

(2014) Advancing research diagnostic criteria for Alzheimer’s disease: The IWG-2 criteria. Lancet Neurol 13, 614–629.

Jack

Jr. , Bennett

, Blennow

, Carrillo

, Dunn

, Haeberlein

, Holtzman

, Jagust

, Jessen

, Karlawish

, Liu

, Molinuevo

, Montine

, Phelps

, Rankin

, Rowe

, Scheltens

, Siemers

, Snyder

, Sperling

(2018) NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement 14, 535–562.

Montine

, Larson

(2009) Late-life dementias: Does this unyielding global challenge require a broader view? JAMA 302, 2593–2594.

Kapasi

, DeCarli

, Schneider

(2017) Impact of multiple pathologies on the threshold for clinically overt dementia. Acta Neuropathol 134, 171–186.

Luchsinger

, Reitz

, Honig

, Tang

, Shea

, Mayeux

(2005) Aggregation of vascular risk factors and risk of incident Alzheimer disease. Neurology 65, 545–551.

de la Torre

(2018) The vascular hypothesis of Alzheimer’s disease: A key to preclinical prediction of dementia using neuroimaging. J Alzheimers Dis 63, 35–52.

Akoudad

, de Groot

, Koudstaal

, van der Lugt

, Niessen

, Hofman

, Ikram

, Vernooij

(2013) Cerebral microbleeds are related to loss of white matter structural integrity. Neurology 81, 1930–1937.

Fazekas

, Kleinert

, Roob

, Kleinert

, Kapeller

, Schmidt

, Hartung

(1999) Histopathologic analysis of foci of signal loss on gradient-echo T2*-weighted MR images in patients with spontaneous intracerebral hemorrhage: Evidence of microangiopathy-related microbleeds. AJNR Am J Neuroradiol 20, 637–642.

10.

Bos

, Wolters

, Darweesh

SKL

, Vernooij

, de Wolf

, Ikram

, Hofman

(2018) Cerebral small vessel disease and the risk of dementia: A systematic review and meta-analysis of population-based evidence. Alzheimers Dement 14, 1482–1492.

11.

Charidimou

, Shams

, Romero

, Ding

, Veltkamp

, Horstmann

, Eiriksdottir

, van Buchem

, Gudnason

, Himali

, Gurol

, Viswanathan

, Imaizumi

, Vernooij

, Seshadri

, Greenberg

, Benavente

, Launer

, Shoamanesh

(2018) Clinical significance of cerebral microbleeds on MRI: A comprehensive meta-analysis of risk of intracerebral hemorrhage, ischemic stroke, mortality, and dementia in cohort studies (v1). Int J Stroke 13, 454–468.

12.

Vernooij

, van der Lugt

, Ikram

, Wielopolski

, Niessen

, Hofman

, Krestin

, Breteler

(2008) Prevalence and risk factors of cerebral microbleeds: The Rotterdam Scan Study. Neurology 70, 1208–1214.

13.

van Rooden

, van der Grond

, van den Boom

, Haan

, Linn

, Greenberg

, van Buchem

(2009) Descriptive analysis of the Boston criteria applied to a Dutch-type cerebral amyloid angiopathy population. Stroke 40, 3022–3027.

14.

Tsai

, Chen

, Tsai

, Pasi

, Lo

, Chen

, Tang

, Jeng

(2021) Long-term vascular outcomes in patients with mixed location intracerebral hemorrhage and microbleeds. Neurology 96, e995–e1004.

15.

Cordonnier

, van der Flier

(2011) Brain microbleeds and Alzheimer’s disease: Innocent observation or key player? Brain 134, 335–344.

16.

Staekenborg

, Koedam

, Henneman

, Stokman

, Barkhof

, Scheltens

, van der Flier

(2009) Progression of mild cognitive impairment to dementia: Contribution of cerebrovascular disease compared with medial temporal lobe atrophy. Stroke 40, 1269–1274.

17.

Kirsch

, McAuley

, Holshouser

, Petersen

, Ayaz

, Vinters

, Dickson

, Haacke

, Britt

3rd , Larseng

, Kim

, Mueller

, Schrag

, Kido

(2009) Serial susceptibility weighted MRI measures brain iron and microbleeds in dementia. J Alzheimers Dis 17, 599–609.

18.

Gyanwali

, Shaik

, Tan

, Vrooman

, Venketasubramanian

, Chen

, Hilal

(2019) Mixed-location cerebral microbleeds as a biomarker of neurodegeneration in a memory clinic population. Aging (Albany NY) 11, 10581–10596.

19.

Liu

, Zhou

, Zhai

, Han

, Zhou

, Ni

, Yao

, Zhang

, Jin

, Cui

, Zhu

(2020) Cerebral microbleeds are associated with loss of white matter integrity. AJNR Am J Neuroradiol 41, 1397–1404.

20.

, Liang

, Yao

, Zhou

, Wang

, Jin

, Zhang

, Cui

, Gong

, Zhu

, Ni

(2019) Cerebral microbleeds correlated with white matter and hippocampal volumes in community-dwelling populations. J Alzheimers Dis 71, 559–567.

21.

Heringa

, Reijmer

, Leemans

, Koek

, Kappelle

, Biessels

(2014) Multiple microbleeds are related to cerebral network disruptions in patients with early Alzheimer’s disease. J Alzheimers Dis 38, 211–221.

22.

McKhann

, Drachman

, Folstein

, Katzman

, Price

, Stadlan

(1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34, 939–944.

23.

Albert

, DeKosky

, Dickson

, Dubois

, Feldman

, Fox

, Gamst

, Holtzman

, Jagust

, Petersen

, Snyder

, Carrillo

, Thies

, Phelps

(2011) 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 7, 270–279.

24.

Shi

, Fu

, Zhang

, Zhao

, Liu

, Zuo

, Cai

, Wang

, Gao

, Ai

, Guan

, Xu

, Ji

(2020) Amyloid PET in dementia syndromes: A Chinese multicenter study. J Nucl Med 61, 1814–1819.

25.

Wardlaw

, Smith

, Biessels

, Cordonnier

, Fazekas

, Frayne

, Lindley

, O’Brien

, Barkhof

, Benavente

, Black

, Brayne

, Breteler

, Chabriat

, Decarli

, de Leeuw

, Doubal

, Duering

, Fox

, Greenberg

, Hachinski

, Kilimann

, Mok

, Oostenbrugge

, Pantoni

, Speck

, Stephan

, Teipel

, Viswanathan

, Werring

, Chen

, Smith

, van Buchem

, Norrving

, Gorelick

, Dichgans

(2013) Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol 12, 822–838.

26.

Gregoire

, Chaudhary

, Brown

, Yousry

, Kallis

, Jäger

, Werring

(2009) The Microbleed Anatomical Rating Scale (MARS): Reliability of a tool to map brain microbleeds. Neurology 73, 1759–1766.

27.

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.

28.

Román

, Erkinjuntti

, Wallin

, Pantoni

, Chui

(2002) Subcortical ischaemic vascular dementia. Lancet Neurol 1, 426–436.

29.

Cui

, Zhong

, Xu

, He

, Gong

(2013) PANDA: A pipeline toolbox for analyzing brain diffusion images. Front Hum Neurosci 7, 42.

30.

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.

31.

Zhang

, Chen

, Liu

, Zhang

, Li

, Cui

, Zhang

(2015) Association of white matter integrity and cognitive functions inchinese non-demented elderly with the APOE ɛ4 allele. J Alzheimers Dis 48, 781–791.

32.

Smith

, Jenkinson

, Johansen-Berg

, Rueckert

, Nichols

, Mackay

, Watkins

, Ciccarelli

, Cader

, Matthews

, Behrens

(2006) Tract-based spatial statistics: Voxelwise analysis of multi-subject diffusion data. Neuroimage 31, 1487–1505.

33.

Ding

, Sigurðsson

, Jónsson

, Eiriksdottir

, Meirelles

, Kjartansson

, Lopez

, van Buchem

, Gudnason

, Launer

(2017) Space and location of cerebral microbleeds, cognitive decline, and dementia in the community. Neurology 88, 2089–2097.

34.

Wang

, Jiang

, Suo

, Yuan

, Xu

, Yang

, Tang

, Zhang

, Zhu

, Tian

, Fan

, Li

, Ye

, Dong

, Jin

, Cui

, Chen

(2019) Deep/mixed cerebral microbleeds are associated with cognitive dysfunction through thalamocortical connectivity disruption: The Taizhou Imaging Study. Neuroimage Clin 22, 101749.

35.

Tan

, Hilal

, Xu

, Vrooman

, Cheng

, Wong

, Venketasubramanian

, Chen

(2020) MRI markers of mixed pathology and cognitive impairment in multiethnic Asians. J Alzheimers Dis 73, 1501–1509.

36.

Kuroda

, Honma

, Mori

, Futamura

, Sugimoto

, Yano

, Kinno

, Murakami

, Ono

(2020) Increased presence of cerebral microbleeds correlates with ventricular enlargement and increased white matter hyperintensities in Alzheimer’s disease. Front Aging Neurosci 12, 13.

37.

Pasi

, Charidimou

, Boulouis

, Auriel

, Ayres

, Schwab

, Goldstein

, Rosand

, Viswanathan

, Pantoni

, Greenberg

, Gurol

(2018) Mixed-location cerebral hemorrhage/microbleeds: Underlying microangiopathy and recurrence risk. Neurology 90, e119–e126.

38.

Blanc

, Viguier

, Calviere

, Planton

, Albucher

, Rousseau

, Sommet

, Bonneville

, Pariente

, Olivot

, Raposo

(2019) Underlying small vessel disease associated with mixed cerebral microbleeds. Front Neurol 10, 1126.

39.

Chou

, Lee

, Peng

, Lee

, Wang

, Chen

, Lin

, Chung

(2021) Classification differentiates clinical and neuroanatomic features of cerebral small vessel disease. Brain Commun 3, fcab107.

40.

Valenti

, Del Bene

, Poggesi

, Ginestroni

, Salvadori

, Pracucci

, Ciolli

, Marini

, Nannucci

, Pasi

, Pescini

, Diciotti

, Orlandi

, Cosottini

, Chiti

, Mascalchi

, Bonuccelli

, Inzitari

, Pantoni

(2016) Cerebral microbleeds in patients with mild cognitive impairment and small vessel disease: The Vascular Mild Cognitive Impairment (VMCI)-Tuscany study. J Neurol Sci 368, 195–202.

41.

Wang

, Wong

, Liu

, Yang

, Chu

, Au

, Lau

, Chan

, Xiong

, Soo

, Leung

, Wong

, Mok

(2015) Cerebral microbleeds and cognitive function in ischemic stroke or transient ischemic attack patients. Dement Geriatr Cogn Disord 40, 130–136.

42.

, 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.

43.

Iturria-Medina

, Pérez Fernández

, Morris

, Canales-Rodríguez

, Haroon

, García Pentón

, Augath

, Galán García

, Logothetis

, Parker

, Melie-García

(2011) Brain hemispheric structural efficiency and interconnectivity rightward asymmetry in human and nonhuman primates. Cereb Cortex 21, 56–67.

44.

Wei

, Zhong

, Nie

, Gong

(2018) Aberrant development of the asymmetry between hemispheric brain white matter networks in autism spectrum disorder. Eur Neuropsychopharmacol 28, 48–62.

45.

Qiao

, He

, Zhang

, Liang

, Shao

, Zhang

, Peng

(2021) The associations between white matter disruptions and cognitive decline at the early stage of subcortical vascular cognitive impairment: A case-control study. Front Aging Neurosci 13, 681208.