Abstract
Introduction
The search for a reliable neuroimaging biomarker in Alzheimer’s disease is a matter of intense research. The presence of cerebral microbleeds seems to be a potential biomarker. However, it is not clear if the presence of microbleeds has clinical usefulness to differentiate mild Alzheimer’s disease and amnestic mild cognitive impairment from normal aging. We aimed to verify if microbleed prevalence differs among three groups: mild Alzheimer’s disease, amnestic mild cognitive impairment due to Alzheimer’s disease, and normal controls. Moreover, we studied whether microbleeds were associated with apolipoprotein E allele ɛ4 status, cerebrospinal fluid amyloid-beta, total and phosphorylated tau protein levels, vascular factors, and cognition.
Methods
Twenty-eight mild Alzheimer’s disease patients, 34 with amnestic mild cognitive impairment and 36 cognitively normal elderly subjects underwent: magnetic resonance imaging with a susceptibility-weighted imaging sequence on a 3T scanner, apolipoprotein E genotyping and a full neuropsychological evaluation. Only amnestic mild cognitive impairment and mild Alzheimer’s disease underwent cerebrospinal fluid analysis. We compared the groups and verified if microbleeds were predicted by all other variables.
Results
Mild Alzheimer’s disease presented a higher prevalence of apolipoprotein E allele ɛ4 in relation to amnestic mild cognitive impairment and control group. No significant differences were found between groups when considering microbleed presence. Logistic regression tests failed to find any relationship between microbleeds and the variables. We performed three different regression models using different independent variables: Model 1 - amyloid-beta, phosphorylated tau protein, total tau, apolipoprotein E allele ɛ4 status, age, and sex; Model 2 - vascular risk factors, age, and sex; Model 3 - cognitive scores sex, age, and education.
Conclusion
Although microbleeds might be related to the Alzheimer’s disease process, their presence is not a good candidate for a neuroimaging biomarker of the disease, especially in its early phases.
Keywords: Microbleeds, biomarkers, Alzheimer’s disease, mild cognitive impairment
Introduction
Alzheimer’s disease (AD) is the most common neurodegenerative diseases and is characterized by cognitive and neuropsychiatric problems which lead to functional dependence. According to the Alzheimer’s Association,1 AD is among the top 10 causes of death in the last decade in the USA, and deaths attributed to AD increased 71%, while those attributed to all of the other most common causes have decreased their incidence. In this sense, the search for biomarkers that could improve diagnostic accuracy in the earliest disease phase is a matter for intense research. Structural neuroimaging has contributed with different and relevant methods such as brain volumetry, cortical thickness, voxel-based morphometry, diffusion tensor imaging techniques, among others.
In this context, cerebral microbleeds (MBs) have become a matter of research in the AD field, especially because they could combine different aspects of the pathophysiology of the disease such as vascular and amyloidosis hypotheses. Excessive and abnormal accumulation of extracellular neuritic plaques (NPs), composed of amyloid-beta (Aβ1-42), is thought to be one of the main pathological mechanisms of AD, together with the presence of neurofibrillary tangles (NFTs) and decreased synaptic density.2 Aβ1-42 peptides can also be deposited in the intravascular space causing cerebral amyloid angiopathy (CAA). Their progressive deposition in medial and adventitial layers of arteries, arterioles and brain capillaries, increases the chance of spontaneous MBs,3 most commonly with lobar distribution (cortical-subcortical regions and cerebellum).4,5 The anatomical co-occurrence of NP and CAA is common, both arising from the accumulation of Aβ1-42 in the brain, which makes lobar MBs a potential indicator of amyloidosis in patients in the AD spectrum.6,7
The presence of apolipoprotein E allele ɛ4 (APOE ɛ4) is another common risk factor for AD and CAA, and may be related to Aβ1-42 metabolism and MBs. In transgenic mice apolipoprotein E (APOE) knockout, for example, the deposition of Aβ1-42 in the form of NP and CAA showed considerably reductions, indicating a possible influence of APOE protein on this process.8 The Rotterdam Scan Study5 demonstrated that, even in healthy older individuals, there is an association between lobar MBs and APOE ɛ4, which is in agreement with increased APOE ɛ4 frequencies seen in patients with “probable CAA”.
However, other factors, such as arterial hypertension and diabetes, contribute to cerebral MBs in AD.9,10 When in association with hypertension, MBs distribution is most common in basal ganglia, thalamus, brainstem, and cerebellum,11 and lobar (mainly occipital lobe) when they are associated with amyloid angiopathy.
Regarding the effect of MBs on cognitive functions, it is more clear in vascular dementia than in AD,12 and the association between MBs and neuropsychological performance in AD patients is contradictory.13–15
Despite these close relationships between the pathophysiology of AD and MBs, it is not entirely clear if the presence of MBs has clinical usefulness to differentiate dementia (especially in the mild phase) and amnestic mild cognitive impairment (aMCI), a possible prodromal AD, from normal aging. Therefore, in this study, we aimed to verify if there are differences in the prevalence of MBs in three groups: mild dementia due to AD, aMCI due to AD and normal controls. In addition, we studied if MBs were associated with APOE ɛ4 status, cerebrospinal fluid (CSF) Aβ1-42, total and phosphorylated tau (t-tau and p-tau), vascular risk factors and cognition in these groups.
Methods
Subjects
We studied 98 subjects: 28 patients with mild AD, 34 with aMCI and 36 cognitive normal older adults. The patients (mild AD) were recruited from the Neuropsychology and Dementia clinic, Hospital de Clínicas, University of Campinas. The aMCI patients and cognitively normal older adults were invited to participate in the project in order to check their memory.
The diagnosis of dementia due to AD fulfilled the National Institute of Aging and Alzheimer’s Association (NIA/AA)16 and all AD patients had a Clinical Dementia Rating (CDR)6 score of 0.5 or 1. aMCI patients were diagnosed using the core criteria of the NIA/AA for MCI17 and had pathophysiological evidence of AD (Aβ1-42 < 416 pg/ml) and/or low Aβ1-42/p-tau (<9.53).18 All aMCI participants had a CDR score of 0.5 (with an obligatory memory score of 0.5). CDR was performed using a semi-structured interview.
Controls were identified as cognitively normal: they did not exhibit any neurological or psychiatric disorders or require psychoactive medication; they performed as normal in the Mini Mental State Examination (MMSE), considering age and education;19 and their structural images were normal. The control group had no memory complaints, performed as normal on neuropsychological tests and had a CDR score of zero.
Exclusion criteria for all subjects: other neurological or psychiatric diseases or a head injury with loss of consciousness; use of sedative drugs in the last 24 h before the neuropsychological assessment, drug or alcohol addiction, prior chronic exposure to neurotoxic substances, Fazekas scale>120 and Hachinski ischemic score>4.21 The Medical Research Ethics Committee of UNICAMP Hospital approved this study, and written informed consent form (either from the subjects or from their responsible caretakers, if incapable) was obtained from all participants before study initiation, according to the Declaration of Helsinki. Pre-diagnostic procedures also comprised laboratory tests including Vitamin B12, folate, and thyroid hormones.
Neuropsychological, functional and neuropsychiatric assessment
An neuropsychologist of six-years experience (AFMCC), blinded to the biomarkers data, performed the neuropsychological evaluations. Global cognitive status was measured by MMSE;22 episodic memory by the Rey Auditory Verbal Learning Test (RAVLT; sub-items encoding, delayed recall - A7, and recognition - RC-FP);23 attention/working memory by forward digit span (FDS) and backward digit span (BDS). For visual perception, we used the following tests: subtests of Luria’s Neuropsychological Investigation (LNI),24 using items G12, G13, G14 and G17 (item from Raven’s test), one item for mental rotation of figures25 and a copy of the Rey-Osterrieth Complex Figure Test.26 For executive functions we used the Trail Making Test A (TMT-A), and B (TMT-B),27 the Stroop test (congruent – Stroop C, and incongruent – Stroop I)28 and Clock Drawing Test (CDT).26 Language tests included the Boston Naming Test (BNT),29 semantic verbal fluency (SVF) - for category (animals), and phonological fluency (PVF) for letters (items beginning with F, A, S in one minute for each letter);30 neuropsychiatric symptoms were evaluated by neuropsychiatric inventory (NPI).31 Functional performance was measured by CDR and Pfeffer functional activities questionnaire.32
Vascular risk factors assessment
Information about vascular risk factors (arterial hypertension, diabetes and dyslipidemia) of the participants was obtained during the interview or accessed in their medical report.
Image acquisition
All magnetic resonance (MR) images were acquired on a 3.0 T magnetic resonance imaging (MRI) Philips Achieva scanner. The following protocol was applied to each subject:
sagittal high-resolution T1-weighted (isotropic voxels of 1 mm3, TR/TE = 7/3.2 ms, FOV = 240 × 240 mm, 180 slices);
coronal and axial fluid-attenuated inversion recovery (FLAIR) T2-weighted images, anatomically aligned to the hippocampus (reconstructed voxel size = 0.45 × 0.45 × 4.00 mm3, TR/TE/TI = 12000/140/2850 ms and FOV = 220 × 206 mm, gap = 1 mm);
coronal IR (inversion recovery) T1-weighted images (reconstructed voxel size = 0.42 × 0.42 × 3.00 mm3, TR/TE/TI = 3550/15/400 ms and FOV = 180 × 180 mm);
coronal multi-echo (five echos) T2-weighted image (reconstructed voxel size = 0.42 × 0.42 × 3.00 mm3, TR/TE = 3300/30 ms and FOV = 180 × 180 mm).
To evaluate the MBs, susceptibility-weighted imaging (SWI) were acquired with the parameters: FOV 230 mm × 182 mm × 130 mm, voxel size 1 mm × 1 mm × 1 mm, reconstructed voxel size 0.44 mm × 0.44 mm × 1 mm, TE:18, TR:13, flip angle 15°, no averages. A neuroradiologist with 10 years of experience (ACSAF), blinded for clinical data, analyzed the images.
Blood and CSF sample collection and handling
Peripheral blood samples collected from the three groups were centrifuged at 2500 rpm for 10 min, then the serum was, subsequently, aliquoted in 1 ml Eppendorf microtube and stored at −80℃ until analysis. CSF samples, only from aMCI and AD patients, were collected by lumbar puncture and stored in a polypropylene tube of 1 ml. Then the samples were centrifuged at 700 rpm for 10 min and stored at −80℃ until analysis.
CSF biomarkers quantification
Dosages of Aβ1-42, t-tau and p-tau were carried out with amyloid-β INNOTEST kits (1-42), h-TAU INNOTEST Ag and INNOTEST Phospho-tau (181 P) (INNO-BIA AlzBio3, Fujirebio, USA), based on enzyme immunoassay solid phase. This multiparameter immunoassay allows the simultaneous quantification of Aß1-42, t-tau, and p-tau181P in CSF using xMAP technology. Suspensions with immunoglobulins against Aβ1-42, t-tau and p-tau were added under inert surface plates, previously washed. A mixture containing CSF samples of research subjects and monoclonal antibodies – to recognize specific immunoglobulins on the plates – was added and incubated for one night, and protected from light exposure. The plates were then washed and subsequently added to streptavidin-phycoerythrin. The set was incubated for one hour. A scanning solution (phosphate buffer saline) was added to the plates following a final wash. Finally, the immunoassay was analyzed to quantify biomarkers.
Genotyping of the APOE gene
Genomic DNA for genotyping the main APOE polymorphisms was extracted from leukocytes obtained from peripheral blood. The presence/absence of haplotypes was either determined by differential amplification with the three specific amplification setups for ɛ2, ɛ3, or ɛ4 allele. This genotyping analysis was performed by polymerase chain reaction (PCR) in real time.
Statistical analysis
We used IBM’s SPSS software (Version 21, SPSS Inc., Chicago, Illinois, USA) for statistical analysis. We first performed the Kolmogorov–Smirnov to test for normality. All variables (MBs, Aβ1-42, p-tau, t-tau, APOE ɛ4 status, neuropsychological and demographic data) were compared between the groups. When data followed a normal distribution, we performed parametric tests; otherwise, we used non-parametric tests. Results were considered significant when p < 0.05, corrected for multiple comparisons (Tukey test).
Initially, we intended to use the number of MBs as a continuous variable, since our hypothesis was that there would be a significant variance between individuals and between groups. However, we found a small variance of MBs (over 90% of the sample had up to two MBs). Thus, we chose to consider this variable as categorical: presence or absence of MBs and used the chi-square test for comparisons between groups. We also considered vascular risk factors as a categorical variable (presence of at least one factor or absence).
In order to verify if MBs were predicted by the variables of the study, we decided to perform logistic regression considering MBs as the dependent variable, in three different models: Model 1, Aβ1-42, p-tau, t-tau, APOE ɛ4 status, age and sex were considered independent variables; Model 2, vascular risk factors (arterial hypertension, diabetes, dyslipidemia) and age and sex as independent variables; Model 3, cognitive scores (episodic memory, visuospatial abilities, executive functions, and language tests), sex, age and education were used as independent variables. All analyses were done considering both the groups separately and together.
Results
Demographics and neuropsychological data are shown in Table 1. Mild AD patients were older and had less years of education than controls, but not in relation to aMCI. These variables were included as covariates in all the subsequent analysis.
Table 1.
Comparison between groups in demographic and neuropsychological data: mean (standard deviation (SD)).
| Control (n = 36) | aMCI (n = 34) | Mild AD (n = 28) | |
|---|---|---|---|
| Age | 66.53 (7.13) | 69.12 (6.41) | 73.00 (8.43)a** |
| Sex M/F | 8/28 | 10/24 | 8/20 |
| Education – years | 10.76 (5.28) | 8.41 (5.38) | 5.36 (4.89)a*** |
| MMSE | 28.0 (1.77) | 25.82 (2.57)a* | 18.81 (5.33)a***,b*** |
| RAVLT – ecoding | 44.67 (7.51) | 32.76 (9.31)a*** | 21.34 (13.58)a***,b*** |
| FDS | 5.14 (1.39) | 4.23 (0.89)a** | 4.07 (1.12)a*** |
| BDS | 4.38 (1.04) | 3.40 (1.0) | 2.76 (1.42)b* |
| Stroop C time | 39.23 (10.96) | 45.24 (13.45) | 69.84 (52.5)a* |
| Stroop I time | 104.79 (31.91) | 121.64 (42.11) | 159.9 (78.23)a***,b** |
| SVF | 17.91 (3.90) | 13.73 (3.18)a** | 8.69 (4.07)a***,b** |
| FAS | 35.09 (9.25) | 27.83 (7.33) | 18.92 (14.22) |
| LNI | 9.00 (1.43) | 7.01 (2.35) | 6.04 (2.93 |
| Rey copy | 31.91 (5.30) | 28.06 (7.40) | 18.46 (13.56)a***,b*** |
| TMT-A (s) | 63.58 (26.03) | 93.79 (53.32) | 158.4 (117.15)a***,b*** |
| TMT-B (s) | 130.46 (73.66) | 141.89 (71.72) | 212.68 (112.67)a***,b* |
| BNT | 56.53 (2.97) | 52.38 (6.85) | 40.21 (33.10)a**,b* |
AD: Alzheimer’s disease; aMCI: amnestic mild cognitive impairment; BDS: backward digit span; BNT: Boston Naming Test; FAS: phonological fluency for letters; FDS: forward digit span; LNI: visuospatial perception item of Luria’s neuropsychological investigation; MCI: mild cognitive impairment; M/F: male/female; MMSE: Mini-Mental Status Examination; RAVLT encoding: encoding of Rey Auditory Verbal Learning Test; Rey figure copy: copy of The Rey-Osterrieth Complex Figure Test; Stroop C: Stroop test congruent; Stroop I: Stroop test incongruent; SVF: semantic verbal fluency; TMT-A: Trail Making Test A; TMT-B: Trail Making Test B.
Significant difference from controls; bsignificant difference from MCI; *p < 0.05; **p < 0.01; ***p < 0.001.
We did not find significant differences between the groups when considering the presence of MBs as well as dosages of Aβ1-42, p-tau and t-tau between aMCI and mild AD. Mild AD patients had a higher prevalence of APOE ɛ4 in relation to controls and aMCI groups (Table 2). Comparing vascular risk factors, there was no difference regarding any variables between the groups.
Table 2.
Comparison between groups considering the study variables: mean (standard deviation (SD)).
| Control | aMCI | Mild AD | |
|---|---|---|---|
| Microbleeds | 16.66% | 29.41% | 25.00% |
| Presence of APOE ɛ4 | 22.22% | 35.29% | 67.85%a***,b** |
| Aβ1-42 (pg/ml) | NR | 335.15 (98.97) | 311.91 (103.96) |
| Total tau (pg/ml) | NR | 88.36 (69.04) | 135.33 (77.74) |
| Phospho-tau (pg/ml) | NR | 49.29 (32,65) | 48.69 (25.22) |
| Arterial hypertension | 6 (24%) | 15 (46,9%) | 12 (48%) |
| Diabetes | 3 (12%) | 10 (31.3%) | 9 (36%) |
| Dyslipidemia | 4 (16%) | 6 (18.8%) | 4 (16.7%) |
| Presence of VRFs | 9 (36%) | 17 (53.1%) | 15 (60%) |
Aβ1-42: Amyloid beta; AD: Alzheimer’s disease; aMCI: amnestic mild cognitive impairment; APOE ɛ4: apolipoprotein E allele 4; MCI: mild cognitive impairment; phospho-tau: phosphorylated tau protein; SD: standard deviation; total Tau: total Tau protein; VRF: vascular risk factors.
Significant difference from controls; bsignificant difference from MCI; *p < 0.05; **p < 0.01; ***p < 0.001
We conducted Fisher exact tests to assess whether there were differences in relation to MBs between the three groups given the presence of APOE ɛ4 (Table 3). We did not find significant differences between the groups. We also used Fisher exact test to analyze if there were intra-group differences considering the presence of the APOE ɛ4 and MBs. In the same way, we did not find intra-group differences (Table 3).
Table 3.
Prevalence of microbleeds, considering the presence of the apolipoprotein E allele 4 (APOE ɛ4).
| Control |
aMCI |
AD |
||||
|---|---|---|---|---|---|---|
| APOE ɛ4+ | APOE ɛ4− | APOE ɛ4+ | APOE ɛ4− | APOE ɛ4+ | APOE ɛ4− | |
| Presence of microbleeds | 0% | 21.43% | 33.33% | 27.27% | 26.32% | 22.22% |
| Absence of microbleeds | 100% | 78.57% | 66.67% | 72.73% | 73.68% | 77.78% |
AD: Alzheimer’s disease; aMCI: amnestic mild cognitive impairment; APOE ɛ4: apolipoprotein E allele 4.
APOE ɛ4+: patients who had at least one allele ɛ4 (homozygous and heterozygous).
To verify if MBs were predicted by the other variables of the study, we performed different logistic regression tests always considering MBs as the dependent variable. In the first regression model, Aβ1-42, p-tau, t-tau, APOE ɛ4 status, age, and sex were considered independent variables. In the second regression model, the independent variables were vascular risk factors (arterial hypertension, diabetes, dyslipidemia, age, and sex). In the last regression, cognitive scores (episodic memory, visuospatial abilities, executive functions, and language tests), sex and age were set as independent variables. We considered, for the analysis, both groups separately and together. There were no statistically significant results in any of the tests (for more details, see Appendix 1).
Discussion
One of the most relevant objectives in AD research is to find highly sensitive and specific biomarkers, especially before dementia is fully developed. In this sense, neuroimaging has been contributing with several different methods and approaches. The study of MBs as biomarkers has the theoretical advantage of combining different and complementary pathophysiological aspects of AD, such as amyloid and vascular hypotheses. However, our study did not detect any differences in the prevalence of MBs between normal aging, aMCI due to AD and mild dementia due to AD as well as any relationship between MBs and the most well established AD biomarkers and with cognition.
The MBs incidence in AD is around 23% of the cases (range of 17–31%),33 as only one MB presented in half of the patients and the other half with two or more brain MBs. According to Sepehry et al.,34 in a review that considered multiple population-based studies, the prevalence of MBs in cognitively normal older adults aged 60–69 years is 15–17% when evaluated by the SWI method. The prevalence found here was 16.66%. It is important to note that SWI has twice the sensitivity for detection of MBs than T2*-GRE acquisition.34 Although the MBs prevalence in our AD and aMCI patients was higher than in our controls, it did not reach statistical significance, even considering that our AD patients were older than controls. According to the same authors, the prevalence in cognitively normal older adults in the range of 70–79 years old is 30–31%.34 The prevalence in our AD patients (average 73 years old) was 25%, lower than that reported by the author. In the same way, Shams et al.35 evaluated 1504 patients (age 63 ± 10 years) with different diagnoses of dementia and found a prevalence of MBs of 22%.
Regarding the association between MBs, Aβ1-42 and APOE ɛ4, we also did not find significant results. It is important to state that our results in relation to Aβ1-42 levels and APOE ɛ4 prevalence are consistent with the literature.18,36 However, other recent studies with larger sample sizes found an association between MBs and Aβ1-42.15,35 Chiang et al.15 found this association only when the number of MBs was greater than three. They also found an association with CSF p-tau. Regarding APOE ɛ4, other authors failed to demonstrate an association with MBs.13,37,38 In a study comparing patients with AD with many brain MBs and those without any MB, patients with many brain MBs were more likely to be homozygous for APOE ɛ4.14 In addition, gender may influence the presence of MBs. Cacciottolo et al.39 found that only men with MCI and AD showed a higher risk of MBs.
An additional aspect that may be related to both AD and MBs is of vascular risk factors such as arterial hypertension, diabetes, and dyslipidemia. Epidemiological studies indicate that elderly patients with vascular risk factors have a higher risk of developing AD.40–42 MBs are much more common in hypertensive subjects (both in the general population and in subjects with cerebrovascular disease). Nakata-Kudo et al.10 showed a relationship between brain MBs with arterial hypertension, and the association was weaker in AD patients than in the cerebrovascular disease population and healthy people. MBs seen in hypertensive individuals are typically located in the deep gray matter and infra-tentorial brain, whereas those seen in AD are most commonly lobar, at the cortico-subcortical junction. We did not find any association between vascular risk factors and MBs. One of the possible reasons is that only three of 98 subjects had MBs in the cerebellum, brainstem, or deep gray nuclei. Another possible explanation is due to our exclusion criteria, that excluded patients with cerebrovascular disease (Fazekas > 1 and Hachinski > 4). Due to hypertension, MBs are most commonly associated with ischemic white matter disease.43
We also did not demonstrate associations between cognition and MBs. Other authors also had similar results when considering MMSE.10,13,44 However, Akoudad et al.,45 in a prospective population-based study with a much larger sample size, found that the presence of more than four MBs affected cognition in all domains and represented an increased risk for developing dementia. In our study, only four subjects (one control, two aMCI, and one AD) had more than three MBs.
The main limitation of our study was the sample size that could explain the lack of associations between MBs and the other variables. Another possible explanation for our results is that only patients with mild dementia stage (CDR = 1) and patients without dementia (aMCI) were included in this study. It is possible that patients in advanced stages of the disease present more MBs and lower levels of Aβ1-42 in the CSF, which would increase this association. However, if this was true, the prevalence of MBs would not be a good option to become a biomarker in the earlier phases of the disease.
Despite the limitations, it is important to highlight that, in our sample, both mild AD and aMCI patients showed abnormal values of Aβ1-42, showing that these patients actually present the pathophysiological process of AD. Additionally, our AD patients presented a higher prevalence of APOE ɛ4 compared with aMCI and controls. These findings (neuropsychological assessment, change in Aβ1-42 and presence of APOE ɛ4) increase the quality of selection of our population. Moreover, MBs were evaluated by an experienced neuroradiologist using SWI protocol at 3 T, a more sensitive image contrast than T2*-GRE.
Conclusion
We suggest that, although the prevalence of MBs might be related to the AD process, it is not a good candidate for a neuroimaging biomarker of the disease. The prevalence of MBs is too small, which indicates a low sensitivity; it was not related neither to the main AD biomarkers, nor to cognition, except when larger sample sizes are studied.
Acknowledgments
The following author contributions were made: conception and design; acquisition of data, analysis/interpretation; writing of article, critical review of article (AGBR and CVLT); acquisition of data, critical review of article (TNCM, HPGJ, LLT, OF, PAORAA, RS, ÍTLC, FC); acquisition of data, analysis/interpretation, critical review of article (AFMCC, ACSAF); conception and design; analysis/interpretation; writing of article, critical review of article (MLFB).
Appendix 1
Regression models using the presence of microbleeds as dependent variable, considering the groups together and separately.
Model 1
| Group (r2) | Predictor | p value | B | Beta | t |
|---|---|---|---|---|---|
| APOE | 0.616 | −0.063 | −0.072 | −0.505 | |
| All (0.019) | AB | 0.425 | 0.000 | 0.115 | 0.803 |
| Total tau | 0.989 | −0.00001501 | −0.003 | −0.013 | |
| Phospho tau | 0.786 | −0.001 | −0.053 | −0.272 | |
| Age | 0.762 | 0.000 | −0.007 | −0.047 | |
| Gender | 0.762 | 0.042 | 0.044 | 0.305 | |
| AD | |||||
| (0.084) | APOE | 0.390 | −0.187 | −0.219 | −0.879 |
| AB | 0.802 | 0.000 | −0.063 | −0.255 | |
| Total tau | 0.622 | 0.001 | 0.169 | 0.501 | |
| Phospho tau | 0.993 | −0.00004251 | −0.003 | −0.008 | |
| Age | 0.378 | 0.012 | 0.263 | 0.902 | |
| Gender | 0.488 | 0.141 | 0.165 | 0.708 | |
| aMCI | |||||
| (0.060) | APOE | 0.592 | −0.105 | −0.110 | −0.543 |
| AB | 0.490 | 0.000 | 0.145 | 0.700 | |
| Total tau | 0.625 | 0.001 | 0.167 | 0.495 | |
| Phospho tau | 0.394 | −0.004 | −0.297 | −0.866 | |
| Age | 0.714 | −0.006 | −0.080 | −0.371 | |
| Gender | 0.954 | −0.012 | −0.012 | −0.058 | |
| Control | APOE | 0.184 | 0.203 | 0.226 | 1.357 |
| (0.114) | Age | 0.705 | −0.003 | −0.065 | −0.382 |
| Gender | 0.212 | 0.194 | 0.217 | 1.272 |
Model 2
| Group (r2) | Predictor | p value | B | Beta | t |
|---|---|---|---|---|---|
| Age | 0.340 | 0.007 | 0.117 | 0.960 | |
| All(0.05) | Gender | 0.340 | 0.198 | 0.207 | 1.692 |
| Hypertension | 0.398 | 0.095 | 0.108 | 0.850 | |
| Diabetes | 0.292 | −0.126 | −0.130 | −1.062 | |
| Dyslipidemia | 0.065 | 0.285 | 0.250 | 2.042 | |
| AD | |||||
| (0.088) | Age | 0.340 | 0.007 | 0.117 | 0.960 |
| Gender | 0.095 | 0.198 | 0.207 | 1.692 | |
| Hypertension | 0.398 | 0.095 | 0.108 | 0.850 | |
| Diabetes | 0.292 | −0.126 | −0.130 | −1.062 | |
| Dyslipidemia | 0.055 | 0.285 | 0.250 | 2.042 | |
| aMCI | |||||
| (0.092) | Age | 0.640 | −0.007 | −0.107 | −0.473 |
| Gender | 0.823 | −0.054 | −0.056 | −0.227 | |
| Hypertension | 0.787 | 0.057 | 0.063 | 0.273 | |
| Diabetes | 0.379 | −0.183 | −0.189 | −0.869 | |
| Dyslipidemia | 0.440 | 0.227 | 0.197 | 0.785 | |
| Control | |||||
| (0.142) | Age | 0.623 | 0.006 | 0.119 | 0.500 |
| Gender | 0.138 | 0.349 | 0.406 | 1.550 | |
| Hypertension | 0.551 | 0.139 | 0.162 | 0.607 | |
| Diabetes | 0.533 | 0.178 | 0.158 | 0.636 | |
| Dyslipidemia | 0.822 | 0.060 | 0.060 | 0.228 |
Model 3
| Group (r2) | Predictor | p value | B | Beta | t |
|---|---|---|---|---|---|
| Age | 0.776 | −0.002 | −0.043 | −0.286 | |
| All (0.192) | Gender | 0.877 | 0.023 | 0.023 | 0.155 |
| MMSE | 0.978 | −0.001 | −0.007 | −0.027 | |
| RAVLT | 0.438 | −0.009 | −0.263 | −0.781 | |
| A7 | 0.419 | 0.026 | 0.266 | 0.815 | |
| RC-FP | 0.929 | −0.001 | −0.021 | −0.089 | |
| FDS | 0.737 | −0.018 | −0.055 | −0.337 | |
| BDS | 0.441 | −0.047 | −0.149 | −0.776 | |
| Stroop C - time | 0.802 | 0.000 | −0.063 | −0.255 | |
| Stroop C - errors | 0.991 | 0.002 | 0.002 | 0.011 | |
| Stroop I - time | 0.116 | 0.003 | 0.346 | 1.597 | |
| Stroop I - errors | 0.369 | −0.008 | −0.173 | −0.905 | |
| SVF | 0.636 | 0.099 | 0.110 | 0.461 | |
| PVF | 0.967 | 0.000 | −0.008 | −0.041 | |
| LNI | 0.343 | −0.024 | −0.191 | −0.956 | |
| Clock | 0.561 | −0.019 | −0.104 | −0.584 | |
| Rey copy | 0.431 | 0.007 | 0.168 | 0.794 | |
| TMT-A | 0.616 | 0.001 | 0.098 | 0.504 | |
| TMT-B | 0.083 | −0.001 | −0.286 | −1.762 | |
| BNT | 0.350 | 0.008 | 0.215 | 0.942 | |
| aMCI (0.123) | Age | 0.952 | −0.003 | −0.043 | −0.062 |
| Gender | 0.228 | −0.707 | −0.657 | −1.343 | |
| MMSE | 0.414 | −0.144 | −0.698 | −0.878 | |
| RAVLT | 0.857 | −0.011 | −0.187 | −0.188 | |
| A7 | 0.313 | 0.153 | 1.04 | 1.102 | |
| RC-FP | 0.840 | −0.021 | −0.257 | −0.210 | |
| FDS | 0.587 | 0.148 | 0.282 | 0.574 | |
| BDS | 0.649 | 0.137 | 0.278 | 0.479 | |
| Stroop C - time | 0.140 | 0.027 | 0.776 | 1.702 | |
| Stroop C - errors | 0.931 | −0.076 | −0.077 | −0.090 | |
| Stroop I - time | 0.915 | −0.001 | −0.128 | −0.112 | |
| Stroop I - errors | 0.887 | −0.009 | −0.080 | −0.148 | |
| SVF | 0.841 | 0.024 | 0.134 | 0.209 | |
| PVF | 0.812 | −0.015 | −0.232 | −0.248 | |
| LNI | 0.852 | −0.060 | −0.246 | −0.195 | |
| Clock | 0.596 | 0.123 | 0.575 | 0.560 | |
| Rey copy | 0.844 | −0.012 | −0.175 | −0.206 | |
| TMT-A | 0.442 | −0.008 | −0.630 | −0.823 | |
| TMT-B | 0.944 | 0.000 | 0.042 | 0.073 | |
| BNT | 0.305 | −0.055 | −0.697 | −1.122 | |
| Control (0.239) | Age | 0.991 | 0.000 | −0.003 | −0.011 |
| Gender | 0.756 | 0.109 | 0.116 | 0.317 | |
| MMSE | 0.709 | 0.029 | 0.133 | 0.382 | |
| RAVLT | 0.381 | −0.020 | −0.387 | −0.907 | |
| A7 | 0.174 | 0.116 | 0.763 | 1.437 | |
| RC-FP | 0.907 | 0.006 | 0.044 | 0.119 | |
| FDS | 0.065 | 0.261 | 0.942 | 2.351 | |
| BDS | 0.061 | −0.577 | −1.557 | −2.416 | |
| Stroop C - time | 0.177 | 0.011 | 0.321 | 1.427 | |
| Stroop C - errors | 0.132 | −2.041 | −0.905 | −1.607 | |
| Stroop I - time | 0.286 | −0.010 | −0.853 | −1.112 | |
| Stroop I - errors | 0.631 | 0.022 | 0.178 | 0.491 | |
| SVF | 0.640 | 0.020 | 0.199 | 0.479 | |
| PVF | 0.970 | 0.000 | 0.011 | 0.038 | |
| LNI | 0.308 | −0.082 | −0.353 | −1.062 | |
| Clock | 0.338 | 0.090 | 0.332 | 0.995 | |
| Rey copy | 0.975 | −0.001 | −0.010 | −0.032 | |
| TMT-A | 0.139 | 0.012 | 0.797 | 1.578 | |
| TMT-B | 0.330 | 0.002 | 0.376 | 1.013 | |
| BNT | 0.250 | 0.116 | 0.887 | 2.541 |
Funding
The authors would like to thank the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) for financial support of the present research. Grant support: FAPESP #2014/25429-2 and #2013/07559-3.
Conflict of interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
References
- 1.Alziheimer’s Association. 2016 Alzheimer’s disease facts and figures report. Alzheimers Dement 2016; 12: 459–509. [DOI] [PubMed]
- 2.Eckman CB, Eckman EA. An update on the amyloid hypothesis. Neurol Clin 2007; 25: 669–682. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Den Heijer T, van der Lijn F, Koudstaal PJ, et al. A 10-year follow-up of hippocampal volume on magnetic resonance imaging in early dementia and cognitive decline. Brain 2010; 133: 1163–1172. [DOI] [PubMed] [Google Scholar]
- 4.Martinez-Ramirez S, Greenberg SM, Viswanathan A. Cerebral microbleeds: Overview and implications in cognitive impairment. Alzheimers Res Ther 2014; 6: 33 DOI: 10.1186/alzrt263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Vernooij MW, van der Lugt A, Ikram MA, et al. Prevalence and risk factors of cerebral microbleeds: The Rotterdam Scan Study. Neurology 2008; 70: 1208–1214. [DOI] [PubMed] [Google Scholar]
- 6.Morris JC. The Clinical Dementia Rating (CDR): Current version and scoring rules. Neurology 1993; 43: 2412–2414. [DOI] [PubMed] [Google Scholar]
- 7.Attems J. Sporadic cerebral amyloid angiopathy: Pathology, clinical implications, and possible pathomechanisms. Acta Neuropathol 2005; 110: 345–359. [DOI] [PubMed] [Google Scholar]
- 8.Irizarry MC. Biomarkers of Alzheimer disease in plasma. NeuroRx 2004; 1: 226–234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Cordonnier C, Al-Shahi Salman R, Wardlaw J. Spontaneous brain microbleeds: Systematic review, subgroup analyses and standards for study design and reporting. Brain 2007; 130: 1988–2003. [DOI] [PubMed] [Google Scholar]
- 10.Nakata-Kudo Y, Mizuno T, Yamada K, et al. Microbleeds in Alzheimer disease are more related to cerebral amyloid angiopathy than cerebrovascular disease. Dement Geriatr Cogn Disord 2006; 22: 8–14. [DOI] [PubMed] [Google Scholar]
- 11.Fazekas F, Kleinert R, Roob G, et al. 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 1999; 20: 637–642. [PMC free article] [PubMed] [Google Scholar]
- 12.Seo SW, Hwa Lee B, Kim EJ, et al. Clinical significance of microbleeds in subcortical vascular dementia. Stroke 2007; 38: 1949–1951. [DOI] [PubMed]
- 13.Pettersen JA, Sathiyamoorthy G, Gao FQ, et al. Microbleed topography, leukoaraiosis, and cognition in probable Alzheimer disease from the Sunnybrook dementia study. Arch Neurol 2008; 65: 790–795. [DOI] [PubMed] [Google Scholar]
- 14.Goos JD, Kester MI, Barkhof F, et al. Patients with Alzheimer disease with multiple microbleeds: Relation with cerebrospinal fluid biomarkers and cognition. Stroke 2009; 40: 3455–3460. [DOI] [PubMed] [Google Scholar]
- 15.Chiang GC, Cruz Hernandez JC, Kantarci K, et al. Cerebral microbleeds, CSF p-tau, and cognitive decline: Significance of anatomic distribution. AJNR Am J Neuroradiol 2015; 36: 1635–1641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 2011; 7: 263–269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Albert MS, DeKosky ST, Dickson D, et al. The diagnosis of mild cognitive impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 2011; 7: 270–279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Forlenza OV, Radanovic M, Talib LL, et al. Cerebrospinal fluid biomarkers in Alzheimer’s disease: Diagnostic accuracy and prediction of dementia. Alzheimers Dement (Amst) 2015; 1: 455–463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Brucki SM, Nitrini R, Caramelli P, et al. Suggestions for utilization of the mini-mental state examination in Brazil. Arq Neuropsiquiatr 2003; 61: 777–781. [DOI] [PubMed] [Google Scholar]
- 20.Fazekas F, Chawluk J, Alavi A, et al. MRI signal abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. AJNR Am J Neuroradiol 1987: 8: 421–426. [DOI] [PubMed]
- 21.Hachinski V, Iadecola C, Petersen RC, et al. National Institute of Neurological Disorders and Stroke-Canadian Stroke Network. Stroke 2006; 37: 2220–2241. [DOI] [PubMed] [Google Scholar]
- 22.Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: 189–198. [DOI] [PubMed] [Google Scholar]
- 23.Malloy-Diniz LF, Lasmar VA, Gazinelli Lde S, et al. The Rey Auditory-Verbal Learning Test: Applicability for the Brazilian elderly population. Rev Bras Psiquiatr 2007; 29: 324–329. [DOI] [PubMed] [Google Scholar]
- 24.Christensen A-L. Luria’s neuropsychological investigation, manual and test material, 4th ed Copenhagen: Munksgaard, 1975. [Google Scholar]
- 25.Ratcliff G. Spatial thought, mental rotation and the right cerebral hemisphere. Neuropsychologia 1979; 17: 49–54. [DOI] [PubMed] [Google Scholar]
- 26.Osterrieth P. The test of copying a complex figure: A contribution to the study of perception and memory. Arch Psychol 1944; 30: 206–356.
- 27.Reitan R. Validity of the Trail Making Test as an indicator of organic brain damage. Percept Mot Skills 1958; 8: 271–276. [Google Scholar]
- 28.Stroop JR. Studies of interference in serial verbal reactions. J Exp Psychol 1935; 18: 643–662. [Google Scholar]
- 29.Kaplan E, Goodglass H, Weintraub S. The Boston Naming Test, 2nd ed. Philadelphia: Lea & Febiger, 1983. [Google Scholar]
- 30.Christensen P, Guilford J. Manual for the Christensen Guilford Fluency Tests, 2nd ed Beverly Hills, California: Sheridan Supply, 1959. [Google Scholar]
- 31.Cummings JL, Mega M, Gray K, et al. The Neuropsychiatric Inventory: Comprehensive assessment of psychopathology in dementia. Neurology 1994; 44: 2308–2314. [DOI] [PubMed] [Google Scholar]
- 32.Pfeffer RI, Kurosaki TT, Harrah CH, Jr, et al. Measurement of functional activities in older adults in the community. J Gerontol 1982; 37: 323–329. [DOI] [PubMed] [Google Scholar]
- 33.Cordonnier C, van der Flier WM. Brain microbleeds and Alzheimer’s disease: Innocent observation or key player? Brain 2011; 134: 335–344. [DOI] [PubMed] [Google Scholar]
- 34.Sepehry AA, Lang D, Hsiung GY, et al. Prevalence of brain microbleeds in Alzheimer Disease: A systematic review and meta-analysis on the influence of neuroimaging techniques. AJNR Am J Neuroradiol 2016; 37: 215–222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Shams S, Martola J, Granberg T, et al. Cerebral microbleeds: Different prevalence, topography, and risk factors depending on dementia diagnosis–the Karolinska Imaging Dementia Study. AJNR Am J Neuroradiol 2015; 36: 661–666. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Belinson H, Michaelson DM. Pathological synergism between amyloid-beta and apolipoprotein E4–the most prevalent yet understudied genetic risk factor for Alzheimer’s disease. J Alzheimers Dis 2009; 17: 469–481. [DOI] [PubMed] [Google Scholar]
- 37.Hanyu H, Tanaka Y, Shimizu S, et al. Cerebral microbleeds in Alzheimer’s disease. J Neurol 2003; 250: 1496–1497. [DOI] [PubMed] [Google Scholar]
- 38.Petersen RC, Parisi JE, Dickson DW, et al. Neuropathologic features of amnestic mild cognitive impairment. Arch Neurol 2006; 63: 665–672. [DOI] [PubMed] [Google Scholar]
- 39.Cacciottolo M, Morgan T and Finch C. Rust on the brain from microbleeds and its relevance to Alzheimer Studies: Invited commentary on Cacciottolo Neurobiology of Aging, 2016. J Alzheimers Dis Parkinsonism 2016; 6. DOI: 10.4172/2161-0460.1000287. [DOI] [PMC free article] [PubMed]
- 40.Skoog I, Lernfelt B, Landahl S, et al. 15-Year longitudinal study of blood pressure and dementia. Lancet 1996; 347: 1141–1145. [DOI] [PubMed] [Google Scholar]
- 41.Biessels GJ, Deary IJ, Ryan CM. Cognition and diabetes: A lifespan perspective. Lancet Neurol 2008; 7: 184–190. [DOI] [PubMed] [Google Scholar]
- 42.Li G, Shofer JB, Kukull WA, et al. Serum cholesterol and risk of Alzheimer disease: A community-based cohort study. Neurology 2005; 65: 1045–1050. [DOI] [PubMed] [Google Scholar]
- 43.Greenberg SM, Vernooij MW, Cordonnier C, et al. Cerebral microbleeds: A field guide to their detection and interpretation. Lancet Neurol 2009; 8: 165–174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Nakata Y, Shiga K, Yoshikawa K, et al. Subclinical brain hemorrhages in Alzheimer’s disease: Evaluation by magnetic resonance T2*-weighted images. Ann N Y Acad Sci 2002; 977: 169–172. [DOI] [PubMed] [Google Scholar]
- 45.Akoudad S, de Groot M, Koudstaal PJ, et al. Cerebral microbleeds are related to loss of white matter structural integrity. Neurology 2013; 81: 1930–1937. [DOI] [PubMed] [Google Scholar]