UBC Faculty Research and Publications

Exploring the effects of coexisting amyloid in subcortical vascular cognitive impairment Dao, Elizabeth; Hsiung, Ging-Yuek R; Sossi, Vesna; Jacova, Claudia; Tam, Roger; Dinelle, Katie; Best, John R; Liu-Ambrose, Teresa Oct 12, 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12883_2015_Article_459.pdf [ 503.41kB ]
JSON: 52383-1.0361943.json
JSON-LD: 52383-1.0361943-ld.json
RDF/XML (Pretty): 52383-1.0361943-rdf.xml
RDF/JSON: 52383-1.0361943-rdf.json
Turtle: 52383-1.0361943-turtle.txt
N-Triples: 52383-1.0361943-rdf-ntriples.txt
Original Record: 52383-1.0361943-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessExploring the effects of coexisting amyloidin subcortical vascular cognitive impairmentElizabeth Dao1, Ging-Yuek Robin Hsiung2, Vesna Sossi 3,7, Claudia Jacova4, Roger Tam5,6, Katie Dinelle7,John R. Best1,8 and Teresa Liu-Ambrose1,8*AbstractBackground: Mixed pathology, particularly Alzheimer’s disease with cerebrovascular lesions, is reported as thesecond most common cause of dementia. Research on mixed dementia typically includes people with a primaryAD diagnosis and hence, little is known about the effects of co-existing amyloid pathology in people with vascularcognitive impairment (VCI). The purpose of this study was to understand whether individual differences in amyloidpathology might explain variations in cognitive impairment among individuals with clinical subcortical VCI (SVCI).Methods: Twenty-two participants with SVCI completed an 11C Pittsburgh compound B (PIB) position emissiontomography (PET) scan to quantify global amyloid deposition. Cognitive function was measured using: 1) MOCA; 2)ADAS-Cog; 3) EXIT-25; and 4) specific executive processes including a) Digits Forward and Backwards Test, b)Stroop-Colour Word Test, and c) Trail Making Test. To assess the effect of amyloid deposition on cognitive functionwe conducted Pearson bivariate correlations to determine which cognitive measures to include in our regressionmodels. Cognitive variables that were significantly correlated with PIB retention values were entered in a hierarchicalmultiple linear regression analysis to determine the unique effect of amyloid on cognitive function. We controlled forage, education, and ApoE ε4 status.Results: Bivariate correlation results showed that PIB binding was significantly correlated with ADAS-Cog (p < 0.01) andMOCA (p < 0.01); increased PIB binding was associated with worse cognitive function on both cognitive measures. PIBbinding was not significantly correlated with the EXIT-25 or with specific executive processes (p > 0.05).Regression analyses controlling for age, education, and ApoE ε4 status indicated an independent association betweenPIB retention and the ADAS-Cog (adjusted R-square change of 15.0 %, Sig F Change = 0.03). PIB retention was alsoindependently associated with MOCA scores (adjusted R-Square Change of 27.0 %, Sig F Change = 0.02).Conclusion: We found that increased global amyloid deposition was significantly associated with greater memory andexecutive dysfunctions as measured by the ADAS-Cog and MOCA. Our findings point to the important role of co-existingamyloid deposition for cognitive function in those with a primary SVCI diagnosis. As such, therapeutic approachestargeting SVCI must consider the potential role of amyloid for the optimal care of those with mixed dementia.Trial registration: NCT01027858Keywords: Vascular cognitive impairment, Amyloid, Mixed dementia, Cognitive function* Correspondence: teresa.ambrose@ubc.ca1Djavad Mowafaghian Centre for Brain Health, University of British Columbia,2215 Wesbrook Mall, Vancouver, BC V6S 0A9, Canada8Department of Physical Therapy, University of British Columbia, 212-2177Wesbrook Mall, Vancouver, BC V6T 1Z3, CanadaFull list of author information is available at the end of the article© 2015 Dao et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Dao et al. BMC Neurology  (2015) 15:197 DOI 10.1186/s12883-015-0459-1BackgroundThe world’s population is rapidly aging and the total num-ber of people living with dementia is projected to increaseglobally from 24.3 million in 2001 to 81.1 million in 2040[1]. Alzheimer’s disease (AD) and vascular cognitive im-pairment (VCI) are the two most common causes of cogni-tive dysfunction [2]. AD is a neurodegenerative diseasecharacterized by amyloid-beta (Aβ) plaques and neurofib-rillary tangles (NFT) [3]. Individuals with AD often presentwith impaired episodic memory, defined as the consciousretrieval of autobiographical events [3, 4]. Vascular cogni-tive impairment can be associated with both large vesseldisease and small vessel disease [5]; this study will focus onthose with subcortical VCI (SVCI) as this group is sug-gested to be a more homogenous group of patients thatare expected to show greater predictability in their clin-ical picture, natural history, outcome, and treatment re-sponse [6]. SVCI is caused by small vessel damage that istypically associated with chronic and diffuse hypoperfu-sion causing cerebral white matter lesions (WML) andlacunes [7]. People with SVCI display relatively intactepisodic memory, but show impairments on measures ofexecutive functions, defined as higher-order cognitive pro-cesses underpinning goal directed behaviors [8].AD and SVCI are often reported as two distinct diseasesin epidemiological studies; however, evidence fromneuropathological studies indicate a high rate of mixedAD-vascular pathology, generally referred to as “mixed de-mentia”. Mixed pathology is present in approximately halfof all clinically diagnosed AD cases [9–13], including partic-ipants of AD clinical trials who were extensively screenedfor pure AD [14]. An autopsy study reported AD with cere-brovascular lesions to be the second most common path-ology after AD [15]; thus, mixed pathology may often bethe rule rather than the exception in clinical diagnosis. Re-cently, efforts were made in understanding the manifest-ation of AD with cerebrovascular disease [16]. For example,at the early AD pathology stage of entorhinal cortical in-volvement–which is generally clinically asymptomatic–thepresence of cerebrovascular lesions is associated with cog-nitive impairment. This suggests that cerebrovascular le-sions may lower the threshold for dementia [17]. Inaddition, among those with AD, the presence of ischemiclesions is associated with a greater degree of cognitive defi-cits compared with pure AD pathology. Overall, it is hy-pothesized that vascular lesions may magnify the effect ofmild AD pathology, result in more severe cognitive impair-ment, and accelerate disease progression [18]. Currently,much of our knowledge on mixed dementia stems fromthe perspective of a primary AD diagnosis and hence, littleis known on the effects of secondary AD pathology in a pri-mary SVCI diagnosis. Specifically, it is unclear how co-existing amyloid pathology may affect cognitive function inpeople with a primary clinical SVCI diagnosis.To investigate cognitive function in mixed pathology it isimportant to include cognitive domains associated withboth SVCI and AD to understand the full spectrum ofcognitive impairment. There is consensus that cognitivemeasures such as the Alzheimer’s Disease AssessmentScale-cognitive subscale (ADAS-Cog) [19], the ExecutiveInterview Test (EXIT-25) [20], and the Montreal CognitiveAssessment (MOCA) [21] should be included for an opti-mal assessment battery in AD and VCI trials [22]. TheADAS-Cog is sensitive to a wide range of disease severityspecific to the central dysfunctions experienced in AD in-cluding memory, praxis, and language; it is regarded as thestandard instrument for use in clinical trials as a primaryindex of cognitive change in AD [23]. The EXIT-25 and theMOCA provide a standardized clinical assessment of ex-ecutive control functions relevant to SVCI [20, 21]. Thoughit is important to use clinically relevant measures, thesegeneralized tests may not capture specific processes thatmay be impaired in mixed dementia. As such, additionaltests of specific executive functions–i.e. working memory(Digits Forward and Backward Test), attention and re-sponse inhibition (Stroop Test), and set shifting (Trail Mak-ing Test)–may be more sensitive to subtle change [24].The neurocognitive profile of SVCI with co-existingamyloid pathology remains to be elucidated. A better un-derstanding of the cognitive dysfunctions associated withamyloid pathology in SVCI may be a useful adjunct in theclinical assessment of mixed SVCI-AD dementia. Thus, thepurpose of this study was to understand whether individualdifferences in amyloid pathology might explain variationsin cognitive impairment among individuals with clinicalSVCI, using a clinically relevant neuropsychological testbattery that is sensitive to both pathologies.MethodsStudy design and participantsWe conducted a cross-sectional analysis of baseline dataacquired from a proof-of-concept randomized controlledtrial of aerobic exercise (i.e., NCT01027858) [25].This study consisted of adults with a clinical diagnosisof mild SVCI. We recruited from the University of BritishColumbia Hospital Clinic for AD and Related Disorders,the Vancouver General Hospital Stroke Prevention Clinic,and specialized geriatric clinics in Metro Vancouver, BC.Clinical diagnosis of SVCI was made by neurologists andgeriatricians based on the presence of both small vesselischemic disease and cognitive syndrome. Small vessel is-chemic disease was defined as evidence of relevant cere-brovascular disease on MRI brain imaging that included:1) Periventricular and deep WML: patchy areas of low at-tenuation or diffuse symmetrical areas of low attenuationwith ill defined margins extending to the centrum semio-vale, and at least one lacune; 2) Absence of cortical and orcortico-subcortical non-lacunar territorial infarcts andDao et al. BMC Neurology  (2015) 15:197 Page 2 of 10watershed infarcts, hemorrhages indicating large vesseldisease, signs of normal pressure hydrocephalus, orother specific causes of WML (i.e. multiple sclerosis,leukodystrophies, sarcoidosis, and brain irradiation). Cogni-tive syndrome was defined as a MOCA score < 26/30 atbaseline–the MOCA is a brief screening tool for mild cog-nitive impairment with high sensitivity and specificity [26].Furthermore, study participants exhibited progressive cog-nitive decline (compared with previous level of cognitivefunction) as confirmed through medical records or care-giver/family member interviews. Overall, participants weregenerally functioning independently and living in the com-munity with minimal assistance by family or caregiver. Allparticipants underwent a physician assessment to confirmcurrent health status and eligibility for the study. Ethicalapproval was obtained from the Vancouver CoastalHealth Research Institute (V07-01160) and the Universityof British Columbia’s Clinical Research Ethics Board (H07-01160). All participants provided written informed consent.Individuals were eligible for study entry if they met thefollowing criteria: 1) aged 55 years or older; 2) MOCAscore < 26/30 at screening [26]; 3) Mini-Mental StateExamination score ≥ 20 at screening [27]; 4) lived in MetroVancouver, Canada and was able to read, write, and speakEnglish; 5) if participants are on cognitive medications(i.e. donepezil, galantamine, rivastigmine, memantine,etc.) they must be on a fixed dose for the duration ofthe trial; 6) must be in sufficient health to participate inthe study’s aerobic-based exercise training program; and7) provide informed consent. Exclusion criteria included:1) absence of small vessel ischemic lesions such as WMLor lacunes on brain CT or MRI; 2) diagnosed with anothertype of dementia (e.g. AD, dementia with lewy bodies, orfrontal-temporal dementia) or other neurological condi-tions (e.g. multiple sclerosis or Parkinson’s disease); 3)taking medications that may negatively affect cognitivefunction (e.g. anticholinergics); and 4) people whoplanned to participate in a clinical drug trial concurrentto this study. This analysis included a sub-set of 22 par-ticipants who met the overall study eligibility criteriaand volunteered to complete a positron emission tom-ography (PET) scan.Descriptive variablesDemographic variablesInformation regarding age, sex, education level, bodymass index (BMI), and waist-hip ratio (WHR) was col-lected at study entry.WML quantificationScanning protocolStructural MRI data was acquired on a Philips 3 TAchieva MRI scanner (Philips Medical Systems, Best,The Netherlands) at the UBC MRI Research Centre. AT2-weighted scan and a Proton-Density-weighted (PD-weighted) scan were acquired. The repetition time (TR)and echo time (TE) for the T2-weighted images wereTR = 5431 and TE = 90 ms and for the PD-weighted im-ages were TR = 2000 ms and TE = 8 ms. Dimensions forthe T2-and PD-weighted scans were 256 × 256 × 60 voxelswith a voxel size of 0.937 × 0.937 × 3.000 mm.Image analysisPrior to lesion identification and segmentation, each MRimage was preprocessed using standard and publicly avail-able neuroimaging tools that included: 1) MR intensity in-homogeneity correction using a multiscale version ofthe nonparametric non-uniform intensity normalizationmethod (N3) [28]; 2) a structure-preserving noise-removalfilter (SUSAN) was applied [29]; and 3) all non-brain tissueswere removed using the brain extraction tool (BET) [30].WML were identified and digitally marked by a singleradiologist with extensive experience in WML identifica-tion. The radiologist used the following guidelines in theseeding procedure, which was designed to be simple whileenabling subsequent automated processing:1. Mark all distinct WML regardless of size.2. Place more than one point on a lesion if the additionalpoints would help define the extent of the lesion.3. Place at least one point near the center of each lesion.WML were then segmented by a method that auto-matically computed the extent of each marked lesion[31]. This segmentation method has been extensivelyvalidated in large data sets with a large range of lesionloads, and was found to be highly accurate compared toradiologist segmentations and also robust to variationsin the placement of the seed points [31]. Full details onthe point placement procedure and subsequent auto-matic segmentation are described in previous work [31],but briefly, the seed points were processed by a custom-ized Parzen windows classifier [32] to estimate the inten-sity distribution of the lesions. The algorithm includedheuristics to optimize the accuracy of the estimated dis-tributions by dynamically adjusting the position and thenumber of seed points used for the Parzen window com-putation, as well as a spatial method that approximatedvisual shape partitioning to identify areas that were likelyto be false positives. The lesion masks were then used toquantify WML volumes in mm3.Dependent variablesGlobal cognitive functionMOCA This is a cognitive screening tool that includes anassessment of set shifting, visuospatial abilities, short-termand working memory, attention, concentration, language,Dao et al. BMC Neurology  (2015) 15:197 Page 3 of 10abstraction, and orientation to time and place–generally, itplaces emphasis on executive functions [33]. The MOCAcomputes a score out of 30 with lower scores indicatinggreater cognitive impairment. The MOCA is a sensitivetool for detecting both mild cognitive impairment [26] andVCI, including SVCI [34].ADAS-Cog This scale assesses memory, language, andpraxis. There are 11 tests and scores range from 0 to 70with higher scores indicating greater cognitive dysfunc-tion. The inter-rater reliability of the ADAS-Cog is 0.989and its test-retest reliability is 0.915 [35].Executive functionsEXIT-2 This is a standardized clinical assessment of globalexecutive functions [20, 36] and is designed to detectfrontal systems pathology [37]. This test contains 25 itemsand scores range from 0 to 50 with higher scores indicatinggreater global executive dysfunctions. This measure can ac-curately separate non-demented subjects from those withcortical or subcortical dementias [38]. Its inter-rater reli-ability is 0.90 [39].Specific executive processesThree specific executive processes were measured: 1)Working memory was assessed with the Verbal DigitsForward and Backward Tests [40]. Participants repeatedprogressively longer random number sequences in thesame order as presented (forward) and in the reversedorder (backward). The difference in score between thetwo tests was calculated, with smaller differences indi-cating better performance; 2) Selective attention andconflict resolution was assessed by the Stroop test [41],which involved three different conditions (80 trials each).First, participants read out words printed in black ink;second, they named the display colour of coloured-X’s;and third, they were shown a page with colour-wordsprinted in incongruent coloured inks (e.g., the word“BLUE” printed in red ink). Participants were asked toname the ink colour in which the words were printed(while ignoring the word itself ). We recorded the timeparticipants took to read the items in each conditionand calculated the time difference between the thirdcondition and the second condition. Smaller time dif-ferences indicate better selective attention and conflictresolution; 3) Set shifting was assessed by the TrailMaking Test (Part A and B) [42]. First, participantsdrew lines connecting encircled numbers sequentially(Part A) then they were asked to alternate betweennumbers and letters (Part B). The difference in time tocomplete Part B and Part A was calculated, smaller dif-ference scores indicated better performance.Independent variablesAmyloid imagingScanning protocol PET scans were performed using11C-PIB produced at UBC TRIUMF. Scans were per-formed in 3-D mode using the GE Advance tomograph(General Electric, Canada/USA). Prior to injection, a10-minute transmission scan with a 68Ge rod was col-lected for attenuation correction. After the transmissionscan, 555 to 560 MBq of 11C-PIB was injected as a bolusinto the antecubital vein and flushed with saline. A 90-minute dynamic acquisition started at tracer injectionand data were framed into an 18×300 sec imagingsequence.Image analysis Parametric images of the non-displaceablebinding potential (BPND) [43] were generated using tissueinput Logan graphical analysis [44, 45] with the cerebellumas the reference region. This method has been validated asreliable for quantification of amyloid deposition [46, 47]. Amean PIB-PET image, i.e. radiotracer concentration aver-aged over the entire scan duration was also formed forimage co-registration and ROI definition purposes. UsingSPM 8 (Wellcome Department of Cognitive Neurology, In-stitute of Neurology, University College London) each sub-ject’s MRI image was co-registered to the correspondingmean PIB-PET image. Each subject’s MRI image was thennormalized to the SPM MNI305 template and the corre-sponding transformation parameters were applied to thesubject’s PET images (mean and parametric images). Forthose without MRI scans (5 subjects did not scan due toMR contraindications), the subject’s mean PIB-PET imagewas normalized to a mean PIB-PET image template inMNI space. This PIB-PET image template was created byaveraging 6 healthy control PIB-PET scans that had all beenwarped with their own MRI to the SPM MNI305 template.Regions of interest analysis A custom set of regions ofinterest (ROIs) was defined on the coronal view of theMNI305 template [48]. These ROIs were transposed toeach subject’s warped MRI and mean-PET images (in MNIspace). ROIs were adjusted as necessary using both theMRI and mean PIB-PET image for guidance (1–2 pixelsmaximum). The modified set of ROIs was then applied tothe parametric PIB-PET image and the average BPNDwithin each ROI was extracted. Global PIB binding was de-termined by averaging values in bilateral frontal (combinedorbitofrontal and medial prefrontal cortex), parietal (com-bined angular gyrus, superior parietal, precuneus, andsupramarginal gyrus), temporal (combined lateral temporaland middle temporal gyrus), and occipital cortices, and an-terior and posterior cingulate gyrus [49].Dao et al. BMC Neurology  (2015) 15:197 Page 4 of 10PIB-positive vs PIB-negative categorization To deter-mine PIB-positive/negative categorization we used stan-dardized uptake values (SUV–tracer concentration/(injecteddose/body weight)) normalized to the cerebellar cortex SUV,referred to as SUV ratio (SUVR–global SUV/cerebellarcortex SUV). An SUVR threshold of 1.50 was imple-mented–this PIB threshold was based on studies in alarge group of cognitively normal subjects studied at theUniversity of Pittsburgh [50] and is used by ADNI [51].Participants with global SUVR above 1.50 were catego-rized as PIB-positive; participants with global SUVR below1.50 were categorized as PIB-negative.Amyloid and cognitive function To determine the effectof amyloid on cognitive function we used Logan graphicalanalysis [44, 45] as it is more accurate when compared withSUVR [52].CovariatesAPOE ε4 genotypeApoE genotype was determined using TaqMan assay sys-tems for the single nucleotide polymorphisms–219G/T.DNA was extracted from whole blood using an auto-mated DNA extraction machine (AutogenflexStar, Auto-gen Inc, Hollisten, MA). Because the ApoE ε4 genotypeis relatively rare, the ApoE ε4 genotype odds ratios wascollapsed into 2 main categories: those with at least 1 ε4allele and those with no ε4 allele [53].Statistical analysisAll statistical analyses were performed using StatisticalPackage for the Social Sciences 22.0. Initial data inspec-tion determined that the distributions were normal (allskew values were less than the absolute value of 1). Wefirst conducted Pearson bivariate correlations to deter-mine which cognitive measures to include in our regres-sion models–cognitive variables that were significantlycorrelated with PIB BPND values (p ≤ .05) were then en-tered in a hierarchical multiple linear regression analysisto determine the unique effect of amyloid on cognitivefunction. In the regression age, education, and ApoE ε4status were entered in the first step as covariates, andPIB BPND was entered in the second step to determinethe unique contribution of amyloid of cognitive function.However, due to our small sample size we conducted aregression analysis without covariates and a regressionwith covariates to ensure the robustness of our results.Also, we report adjusted R2 values, which penalizesthe explained variance for each additional covariate.For each hierarchical regression model, we computedcollinearity statistics (tolerance and variance inflationfactor), histograms of the residuals, and scatterplotsof the predicted versus residual values to ensure thatthe assumptions of linear regression were met. In allmodels, mutlicollinearity was not an issue amongpredictor variables, and the residuals were normallydistributed and homoscedastic. These analyses alsoconfirmed a linear association between the predictorsand outcome variables for a hierarchical multiple linearregression.ResultsDescriptive variablesTwenty-two participants (8 females, 14 males) completedPIB-PET imaging. The mean age was 72.95 ± 7.76 yearswith an average MOCA score of 23.54 ± 2.34. Five out of22 participants did not complete an MRI scan due to MRcontraindications (e.g. presence of coronary stent or artifi-cial optic lens). One scan was discarded from WML quanti-fication analysis due to severe motion artifacts–with asubset of 16 participants, WML volume ranged from 23.49-3093.39 mm3 with an average of 616.41 ± 849.13 mm3. Of22 participants, six were PIB-positive and 16 were PIB-negative. The global PIB BPND was 0.07 ± 0.23. Detaileddemographic characteristics, neuropsychological test re-sults, WML volume, and PIB BPND values are presented inTable 1. Compared with the total participants in the ran-domized controlled trial, this subset was not different inage (mean difference = 2.42, p > 0.05), but displayed higherMOCA scores (mean difference = 3.07, p < 0.01).Bivariate correlationsPIB binding was significantly correlated with ADAS-Cog(r = 0.58, p < 0.01) and MOCA (r = −0.55, p < 0.01)–spe-cifically, increased PIB BPND was associated with worsecognitive performance on the ADAS-Cog and the MOCA(Table 2). PIB BPND was not significantly correlated withthe EXIT-25 or with any of the specific executive pro-cesses (p > 0.05).Linear regressionTo determine the independent association between PIBBPND and ADAS-Cog and MOCA scores we conducteda hierarchical multiple linear regression.ADAS-CogWithout covariates in the model, PIB rentention accountedfor 31.0 % (adjusted R-square) of the variance in ADAS-Cog scores (F [1, 20] = 10.27, p = 0.00–Table 3). When con-trolled for age, education, and APOE ε4, this accounted for27.0 % of the variance in ADAS-Cog scores. Adding PIBBPND to the model resulted in a significant adjusted R-square change of 15.0 % (F Change [1, 17] = 5.78, Sig FChange = 0.03–Table 3). The total adjusted varianceaccounted by the final model was 42.0 %.Dao et al. BMC Neurology  (2015) 15:197 Page 5 of 10MOCAWithout covariates in the model, PIB retention accountedfor 27.0 % (adjusted R-square) of the variance in MOCAscores (F [1, 20] = 8.66, p = 0.01–Table 4). When controlledfor age, education, and APOE ε4 this accounted for–7.0 %of the variance in MOCA scores, which suggests that thepenalty for adding these covariates outweighed their ex-plained variance in MOCA scores. Adding PIB BPND tothe model resulted in a significant adjusted R-square changeof 27.0 % (F Change [1, 17] = 6.98, Sig F Change = 0.02–Table 4). The total adjusted variance accounted by the finalmodel was 20.0 %.DiscussionTo date, few studies have focused on the role of co-existingamyloid pathology in a primary SVCI diagnosis. Our studyfound that six out of twenty-two participants with clinicalSVCI were PIB-positive. In assessing the effect of amyloidon cognitive function, we found that increased global amyl-oid deposition–suggestive of co-existing Alzheimer path-ology–was significantly associated with worse cognitivefunction as indicated by the ADAS-Cog and MOCA. Ourfindings concur with and extend the results of previousliterature assessing the role of amyloid on cognitive func-tion in people with mild cognitive impairments (MCI) andhealthy older adults.The ADAS-Cog primarily assesses episodic memoryand has been linked to amyloid deposition [54]. This as-sociation is present in both healthy older adults andpeople with MCI. Longitudinal studies in healthy olderadults found that increased PIB binding was associatedwith greater memory decline over time [55, 56] and maybe indicative of preclinical AD [56]. A similar associationis found in people with MCI. Several studies have foundincreased PIB binding to be strongly correlated with epi-sodic memory impairments in amnestic MCI subtypes[57, 58]; furthermore, PIB-positive amnestic MCI patientsare more likely to progress to AD [59, 60]. Together, theseprevious studies have established the association betweenamyloid and memory impairments within an AD context.The current study extends previous knowledge of amyloiddeposition by showing that greater amyloid deposition onPIB-PET screening is associated with greater memory im-pairment in a SVCI cohort.Our study also found increased amyloid to be associ-ated with lower MOCA scores, which assesses a mix ofTable 1 Descriptive CharacteristicVariable Mean SDAge 72.95 7.77Female Sex, No. (%) 8 (36 %)Education, No. (%)High school education 1 (5 %)Trade or professional certificate or diploma 12 (55 %)University education 9 (41 %)MMSE (max. score 30) 27.50 1.95MOCA (max. score 30) 23.55 2.34WHR 0.91 0.08BMI 26.95 4.78PIB-positive, No. (%) 6 (27 %)Global PIB BPND 0.07 0.23WML volume (mm3), n = 16 616.41 849.13Cognitive AssessmentsADAS-Cog (max. score 70) 8.99 3.30Exit-25 (max. score 50) 10.59 4.38Stroop CW-C, sec. 61.44 26.01Trails B-A, sec. 50.51 24.84Digits F-B, sec. 3.23 2.79SD = Standard Deviation, MMSE =Mini-Mental State Examination, MOCA =Montreal Cognitive Assessment, WHR =Waist-to-Hip Ratio, BMI = Body MassIndex, ADAS-Cog = Alzheimer’s Disease Assessment Scale – Cognitive subscale,Exit-25 = Executive Interview Test, Stroop CW-W = Stroop Color Words minusStroop colored x’s, Trails B-A = Trails B (numbers and letters) minus Trails A(numbers), Digits F-B = Digits F-B = Digits Forwards minus Digits BackwardsTable 2 Correlation MatrixPIB BPND Age Education APOE ε4 ADAS-Cog MOCA EXIT 25 Digits Stroop TrailsPIB BPNDAge −0.01Education 0.09 0.21APOE ε4 0.36 0.13 0.04ADAS-Cog 0.58** 0.15 0.04 0.61**MOCA −0.55* −0.12 −0.25 −0.16 −0.41EXIT-25 0.13 0.18 0.22 0.25 0.43* −0.40Digits 0.02 0.12 0.34 −0.45* −0.26 0.17 −0.04Stroop 0.17 −0.04 0.04 −0.11 0.34 −0.18 0.34 0.04Trails 0.19 −0.02 0.12 0.48* 0.28 −0.18 0.29 0.01 0.13*significant at p < 0.05; **significant at p < 0.01Dao et al. BMC Neurology  (2015) 15:197 Page 6 of 10cognitive functions with an emphasis on executive func-tions. Though AD is typically associated with memoryimpairments, people in the early stages of AD displayexecutive dysfunctions [61, 62]; thus, it is plausible thatamyloid deposition would also be associated with execu-tive dysfunctions. However, we note that we did not finda significant association with specific executive measures(i.e., Digit Span Test, Stroop test, and Trail Making Test)and the EXIT-25 test. No other studies have reporteddata on the EXIT-25 and few studies have examined theeffect of amyloid deposition on specific executive pro-cesses. One reason for these non-significant results maylie in the minimal power of these tests to detect an ef-fect. A complex statistical study conducted by ADNIfound a composite score (ADNI-EF included: CategoryFluency, Clock Drawing, WAIS-R Digit Symbol, DigitSpan Backwards, and the Trail Making Test includingTrails A, Trails B, and Trails B minus Trails A) to besuperior to any independent measure of executive func-tioning. Specifically, ADNI-EF was sensitive to capturingchanges in cognitive function over time and was thestrongest baseline predictor of conversion to AD [63].Although the MOCA does include a memory subtest, itplaces greater emphasis on tasks of executive functionand has similar components to ADNI-EF (includes ClockDrawing, Digit Span Backwards, and Trail Making andadditionally includes a phonemic fluency task, a two-itemverbal abstraction task, a sustained attention task, and aconcentration task). Thus, the MOCA–as a global com-posite measure–may be more sensitive compared withspecific executive processes.Overall, we found that amyloid was associated withimpairments in multiple domains of cognitive functionincluding memory and executive dysfunctions in peopleTable 3 Multiple linear regression models assessing the contribution of PIB retention on ADAS-CogIndependent variables R2 Adjusted R2 R2 Change Unstandardized B (Standard error) Standardized β P- ValueModel 1PIB BPND 0.34 0.31 - 8.56 (2.67) 0.58 0.00Model 2Step 1 0.37 0.27 0.37*Age 0.03 (0.08) 0.070 0.73Education 0.02 (0.68) 0.01 0.97APOE ε4 4.14 (1.30) 0.60 0.01Step 2 0.53 0.42 0.16*Age 0.04 (0.07) 0.10 0.56Education −0.11 (0.61) −0.03 0.86APOE ε4 3.07 (1.24) 0.44 0.02PIB BPND 6.31 (2.62) 0.43 0.03*significant at p < 0.05Table 4 Multiple linear regression models assessing the contribution of PIB retention on MOCAIndependent variables R2 Adjusted R2 R2 Change Unstandardized B (Standard error) Standardized β P- ValueModel 1PIB BPND 0.30 0.27 - −5.75 (1.95) −0.55 0.01Model 2Step 1 0.09 −0.07 0.09Age −0.02 (0.07) −0.06 0.81Education −0.58 (0.59) −0.23 0.34APOE ε4 −0.72 (1.12) −0.15 0.53Step 2 0.35 0.20 0.27*Age −0.03 (0.06) −0.10 0.63Education −0.46 (0.51) −0.18 0.39APOE ε4 0.27 (1.04) 0.06 0.80PIB BPND −5.80 (2.20) −0.56 0.02*significant at p < 0.05Dao et al. BMC Neurology  (2015) 15:197 Page 7 of 10with clinical SVCI. A similar study conducted by Leeand colleagues [64] found that PIB retention inpeople with small vessel MCI was associated withimpairments in multiple domains of cognitive func-tion including language, visuospatial, memory, andexecutive functions. Furthermore, these results con-cur with a study published by Nordlund and col-leagues [65] who found that people with cognitiveimpairments in multiple domains (i.e. memory and execu-tive dysfunctions) were more likely to convert to mixeddementia and vascular dementia compared with peoplewho exhibited either memory or executive dysfunctionalone.However, our conclusions are not without limitations.First, this is an exploratory study, and thus, is limited byits small sample size; therefore, we are limited in ourability to detect smaller effects and future studies are re-quired to confirm and extend our current results. Wealso did not account for the effect of NFT on cognitivefunction. This is of particular importance as neocorticalNFT is more consistently correlated with dementia sever-ity, and it is suggested that NFT may mediate the associ-ation of amyloid on cognitive function [66]. In addition,we were not able to acquire MRI scans in all participantsand did not have the sample size to include WML as a co-variate. As a result, it is not clear how WML may haveuniquely contributed to performance on the ADAS-Cogand MOCA. This is particularly important as declines inmemory and executive functions have been linked toincreased subcortical white matter disease [67–69].However, a study published by Park and colleagues[49] investigating the relationship between cerebrovas-cular disease, amyloid, and cognitive function in SVCIsuggested that amyloid burden and SVCI pathologywere largely unrelated and that the effects of amyloidon cognition is independent of markers of SVCI path-ology. The unique impact of amyloid on cognitivefunction is further supported by evidence in cerebrospinalfluid (CSF). A study assessing the role of amyloid beta (the42-amino-acid form–Aβ1–42) and neurofilament light(NF-L)–elevated concentration of NF-L in CSF are as-sociated with WML and small vessel disease—in CSFfound that only Aβ1–42 was associated with worse cog-nitive outcomes in people with cerebral vascular disease[70]. Overall, the results of our study and previous studies[64, 70, 71] suggest that increased amyloid is independ-ently associated with worse cognitive outcomes in peoplewith vascular disease.ConclusionIt is reported that approximately 33 % of those with SVCIexhibit amyloid pathology [71]. Yet, few studies have in-vestigated the effect of amyloid deposition in people withSVCI and fewer studies have assessed PIB rentention as acontinuous variable. Categorical diagnostic classificationsof AD, SVCI, and mixed AD/SVCI in clinical practice failto take into account the reality of a gradient of brain amyl-oid deposition across these disease states. Our findingspoint to the important role of amyloid deposition for cog-nitive function even among those with a primary SVCIdiagnosis. As such, therapeutic approaches targeting SVCImust consider the potential role of amyloid for the opti-mal care of those with mixed dementia.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsED, G-YR H, VS, CJ, RT, KD, JRB, TLA contributed to study design, statisticalanalyses, data interpretation, and manuscript preparation. All authors read andapproved the final manuscript.AcknowledgementsThis study was jointly funded by the Canadian Stroke Network and the Heartand Stroke Foundation of Canada. Teresa Liu-Ambrose is a Canada ResearchChair in Physical Activity, Mobility, and Cognitive Neuroscience, a Michael SmithFoundation for Health Research (MSFHR) Scholar, a Canadian Institutes of HealthResearch (CIHR) New Investigator, and a Heart and Stroke Foundation of Canada’sHenry J.M. Barnett Scholarship recipient. John Best is CIHR and Michael SmithFoundation for Health Research Post-Doctoral Fellow. Elizabeth Dao is a CIHRDoctoral Trainee. TRIUMF is gratefully acknowledged for PET tracer production.Author details1Djavad Mowafaghian Centre for Brain Health, University of British Columbia,2215 Wesbrook Mall, Vancouver, BC V6S 0A9, Canada. 2Department ofMedicine, University of British Columbia, 2211 Wesbrook Mall, Vancouver, BCV6T 2B5, Canada. 3Department of Physics and Astronomy, University ofBritish Columbia, 6224 Agricultural Road, Vancouver, BC V6T 1Z1, Canada.4School of Professional Psychology, Pacific University, 190 SE 8th Avenue,Hillsboro, OR 97123, USA. 5Department of Radiology, University of BritishColumbia, 3350-950 W 10th Avenue, Vancouver, BC V5Z 1 M9, Canada. 6MS/MRI Research Group, University of British Columbia, 2215 Wesbrook Mall,Vancouver, BC V6S 0A9, Canada. 7UBC PET, Brain Research Centre, 2211Westboork Mall, Vancouver, BC V6T 2B5, Canada. 8Department of PhysicalTherapy, University of British Columbia, 212-2177 Wesbrook Mall, Vancouver,BC V6T 1Z3, Canada.Received: 23 April 2015 Accepted: 4 October 2015References1. Ferri CP, Prince M, Brayne C, Brodaty H, Fratiglioni L, Ganguli M, et al. Globalprevalence of dementia: a Delphi consensus study. Lancet. 2005;366(9503):2112–7. doi:10.1016/S0140-6736(05)67889-0.2. Dementia: a public health priority. World Health Organization. 2012.3. Mattson MP. Pathways towards and away from Alzheimer’s disease. Nature.2004;430(7000):631–9. doi:10.1038/nature02621.4. Tulving E. Episodic memory: from mind to brain. Annu Rev Psychol. 2002;53:1–25.doi:10.1146/annurev.psych.53.100901.135114.5. Moorhouse P, Rockwood K. Vascular cognitive impairment: current conceptsand clinical developments. Lancet Neurol. 2008;7(3):246–55. doi:10.1016/S1474-4422(08)70040-1.6. Erkinjuntti T. Subcortical ischemic vascular disease and dementia. IntPsychogeriatr. 2003;15 Suppl 1:23–6. doi:10.1017/S1041610203008925.7. Gorelick PB, Scuteri A, Black SE, Decarli C, Greenberg SM, Iadecola C, et al.Vascular contributions to cognitive impairment and dementia: a statement forhealthcare professionals from the american heart association/american strokeassociation. Stroke. 2011;42(9):2672–713. doi:10.1161/STR.0b013e3182299496.8. Breteler MM, van Swieten JC, Bots ML, Grobbee DE, Claus JJ, van den HoutJH, et al. Cerebral white matter lesions, vascular risk factors, and cognitivefunction in a population-based study: the Rotterdam Study. Neurology.1994;44(7):1246–52.Dao et al. BMC Neurology  (2015) 15:197 Page 8 of 109. Jellinger KA, Attems J. Neuropathological evaluation of mixed dementia.J Neurol Sci. 2007;257(1–2):80–7. doi:10.1016/j.jns.2007.01.045.10. Kalaria RN. The role of cerebral ischemia in Alzheimer’s disease. NeurobiolAging. 2000;21(2):321–30.11. Rockwood K, Macknight C, Wentzel C, Black S, Bouchard R, Gauthier S, et al.The diagnosis of “mixed” dementia in the Consortium for the Investigation ofVascular Impairment of Cognition (CIVIC). Ann N Y Acad Sci. 2000;903:522–8.12. Schneider JA, Arvanitakis Z, Bang W, Bennett DA. Mixed brain pathologiesaccount for most dementia cases in community-dwelling older persons.Neurology. 2007;69(24):2197–204. doi:10.1212/01.wnl.0000271090.28148.24.13. Zekry D, Hauw JJ, Gold G. Mixed dementia: epidemiology, diagnosis, andtreatment. J Am Geriatr Soc. 2002;50(8):1431–8.14. Wang BW, Lu E, Mackenzie IR, Assaly M, Jacova C, Lee PE, et al. Multiplepathologies are common in Alzheimer patients in clinical trials. Can JNeurol Sci. 2012;39(5):592–9.15. Jellinger KA, Attems J. Prevalence of dementia disorders in the oldest-old:an autopsy study. Acta Neuropathol. 2010;119(4):421–33. doi:10.1007/s00401-010-0654-5.16. Dubois B, Feldman HH, Jacova C, Cummings JL, Dekosky ST, Barberger-Gateau P, et al. Revising the definition of Alzheimer’s disease: a new lexicon.Lancet Neurol. 2010;9(11):1118–27. doi:10.1016/S1474-4422(10)70223-4.17. Esiri MM, Nagy Z, Smith MZ, Barnetson L, Smith AD. Cerebrovascular diseaseand threshold for dementia in the early stages of Alzheimer’s disease.Lancet. 1999;354(9182):919–20. doi:10.1016/S0140-6736(99)02355-7.18. Nagy Z, Esiri MM, Jobst KA, Morris JH, King EM, McDonald B, et al. Theeffects of additional pathology on the cognitive deficit in Alzheimer disease.J Neuropathol Exp Neurol. 1997;56(2):165–70.19. Rosen WG, Mohs RC, Davis KL. A new rating scale for Alzheimer’s disease.Am J Psychiatry. 1984;141(11):1356–64.20. Royall DR, Mahurin RK, Gray KF. Bedside assessment of executive cognitiveimpairment: the executive interview. J Am Geriatr Soc. 1992;40(12):1221–6.21. Koski L. Validity and applications of the Montreal cognitive assessment forthe assessment of vascular cognitive impairment. Cerebrovasc Dis.2013;36(1):6–18. doi:10.1159/000352051.22. O’Brien J, Lilienfeld S. Relevant clinical outcomes in probable vasculardementia and Alzheimer’s disease with cerebrovascular disease. J NeurolSci. 2002;203–204:41–8.23. Mohs RC, Kawas C, Carrillo MC. Optimal design of clinical trials for drugsdesigned to slow the course of Alzheimer’s disease. Alzheimers Dement.2006;2(3):131–9. doi:10.1016/j.jalz.2006.04.003.24. Hachinski V, Iadecola C, Petersen RC, Breteler MM, Nyenhuis DL, Black SE,et al. National Institute of Neurological Disorders and Stroke-CanadianStroke Network vascular cognitive impairment harmonization standards.Stroke. 2006;37(9):2220–41. doi:10.1161/01.STR.0000237236.88823.47.25. Liu-Ambrose T, Eng JJ, Boyd LA, Jacova C, Davis JC, Bryan S, et al. Promotion ofthe mind through exercise (PROMoTE): a proof-of-concept randomizedcontrolled trial of aerobic exercise training in older adults with vascularcognitive impairment. BMC Neurol. 2010;10:14. doi:10.1186/1471-2377-10-14.26. Nasreddine ZS, Phillips NA, Bedirian V, Charbonneau S, Whitehead V, Collin I,et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool formild cognitive impairment. J Am Geriatr Soc. 2005;53(4):695–9. doi:10.1111/j.1532-5415.2005.53221.x.27. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical methodfor grading the cognitive state of patients for the clinician. J Psychiatr Res.1975;12(3):189–98.28. Sled JG, Zijdenbos AP, Evans AC. A nonparametric method for automaticcorrection of intensity nonuniformity in MRI data. IEEE Trans Med Imaging.1998;17(1):87–97. doi:10.1109/42.668698.29. Smith SMB JM. SUSAN-a new approach to low level image processing. Int JComput Vis. 1997;23:1.30. Smith SM. Fast robust automated brain extraction. Hum Brain Mapp.2002;17(3):143–55. doi:10.1002/hbm.10062.31. McAusland J, Tam R, Wong E, Riddehough A, Li D. Optimizing the Use ofRadiologist Seed Points for Improved Multiple Sclerosis Lesion Segmentation.IEEE Trans Biomed Eng. 2010. doi:10.1109/TBME.2010.2055865.32. Parzen E. On estimation of a probability density function and mode. AnnMath Stat. 1962;33:1065–76.33. Lam B, Middleton LE, Masellis M, Stuss DT, Harry RD, Kiss A, et al. Criterionand convergent validity of the Montreal cognitive assessment withscreening and standardized neuropsychological testing. J Am Geriatr Soc.2013;61(12):2181–5. doi:10.1111/jgs.12541.34. Dong Y, Sharma VK, Chan BP, Venketasubramanian N, Teoh HL, Seet RC, et al.The Montreal Cognitive Assessment (MoCA) is superior to the Mini-MentalState Examination (MMSE) for the detection of vascular cognitive impairmentafter acute stroke. J Neurol Sci. 2010;299(1–2):15–8. doi:10.1016/j.jns.2010.08.051.35. Kirk A. Target symptoms and outcome measures: cognition. Can J NeurolSci. 2007;34 Suppl 1:S42–6.36. Stokholm J, Vogel A, Gade A, Waldemar G. The executive interview as ascreening test for executive dysfunction in patients with mild dementia.J Am Geriatr Soc. 2005;53(9):1577–81. doi:10.1111/j.1532-5415.2005.53470.x.37. Royall DR, Rauch R, Roman GC, Cordes JA, Polk MJ. Frontal MRI findingsassociated with impairment on the Executive Interview (EXIT25). Exp AgingRes. 2001;27(4):293–308. doi:10.1080/03610730109342350.38. Roman GC, Royall DR. Executive control function: a rational basis for thediagnosis of vascular dementia. Alzheimer Dis Assoc Disord. 1999;13 Suppl3:S69–80.39. Royall DR, Chiodo LK, Polk MJ. Executive dyscontrol in normal aging:normative data, factor structure, and clinical correlates. Curr Neurol NeurosciRep. 2003;3(6):487–93.40. Wechsler D. Wechsler adult intelligence scale. 4edth ed. New York: NCSPearson; 2008.41. Trenerry M. Stroop neuropsychological screening test manual. Odessa, FL:Psychological Assessment Resources; 1989.42. Allen DN, Haderlie MM. Trail-making test. The Corsini Encyclopedia ofPsychology: Wiley; 2010.43. Innis RB, Cunningham VJ, Delforge J, Fujita M, Gjedde A, Gunn RN, et al.Consensus nomenclature for in vivo imaging of reversibly bindingradioligands. J Cereb Blood Flow Metab. 2007;27(9):1533–9. doi:10.1038/sj.jcbfm.9600493.44. Logan J, Fowler JS, Volkow ND, Wang GJ, Ding YS, Alexoff DL. Distributionvolume ratios without blood sampling from graphical analysis of PET data.J Cereb Blood Flow Metab. 1996;16(5):834–40. doi:10.1097/00004647-199609000-00008.45. Logan J, Fowler JS, Volkow ND, Wolf AP, Dewey SL, Schlyer DJ, et al.Graphical analysis of reversible radioligand binding from time-activitymeasurements applied to [N-11C-methyl]-(−)-cocaine PET studies in humansubjects. J Cereb Blood Flow Metab. 1990;10(5):740–7. doi:10.1038/jcbfm.1990.127.46. Lopresti BJ, Klunk WE, Mathis CA, Hoge JA, Ziolko SK, Lu X, et al. Simplifiedquantification of Pittsburgh Compound B amyloid imaging PET studies: acomparative analysis. J Nucl Med. 2005;46(12):1959–72.47. Price JC, Klunk WE, Lopresti BJ, Lu X, Hoge JA, Ziolko SK, et al. Kineticmodeling of amyloid binding in humans using PET imaging and PittsburghCompound-B. J Cereb Blood Flow Metab. 2005;25(11):1528–47. doi:10.1038/sj.jcbfm.9600146.48. Collins DL, Neelin P, Peters TM, Evans AC. Automatic 3D intersubject registrationof MR volumetric data in standardized Talairach space. J Comput Assist Tomogr.1994;18(2):192–205.49. Park JH, Seo SW, Kim C, Kim SH, Kim GH, Kim ST, et al. Effects of cerebrovasculardisease and amyloid beta burden on cognition in subjects with subcorticalvascular cognitive impairment. Neurobiol Aging. 2014;35(1):254–60. doi:10.1016/j.neurobiolaging.2013.06.026.50. Aizenstein HJ, Nebes RD, Saxton JA, Price JC, Mathis CA, Tsopelas ND, et al.Frequent amyloid deposition without significant cognitive impairmentamong the elderly. Arch Neurol. 2008;65(11):1509–17. doi:10.1001/archneur.65.11.1509.51. Jagust WJ, Bandy D, Chen K, Foster NL, Landau SM, Mathis CA, et al. TheAlzheimer’s Disease Neuroimaging Initiative positron emission tomographycore. Alzheimers Dement. 2010;6(3):221–9. doi:10.1016/j.jalz.2010.03.003.52. Yaqub M, Tolboom N, Boellaard R, van Berckel BN, van Tilburg EW,Luurtsema G, et al. Simplified parametric methods for [11C] PIB studies.NeuroImage. 2008;42(1):76–86. doi:10.1016/j.neuroimage.2008.04.251.53. Hsiung GY, Sadovnick AD, Feldman H. Apolipoprotein E epsilon4 genotype asa risk factor for cognitive decline and dementia: data from the Canadian Studyof Health and Aging. CMAJ. 2004;171(8):863–7. doi:10.1503/cmaj.1031789.54. Doraiswamy PM, Sperling RA, Johnson K, Reiman EM, Wong TZ, SabbaghMN, et al. Florbetapir F 18 amyloid PET and 36-month cognitive decline: aprospective multicenter study. Mol Psychiatry. 2014;19(9):1044–51.doi:10.1038/mp.2014.9.55. Resnick SM, Sojkova J, Zhou Y, An Y, Ye W, Holt DP, et al. Longitudinalcognitive decline is associated with fibrillar amyloid-beta measured by [11C]PiB. Neurology. 2010;74(10):807–15. doi:10.1212/WNL.0b013e3181d3e3e9.Dao et al. BMC Neurology  (2015) 15:197 Page 9 of 1056. Villemagne VL, Pike KE, Darby D, Maruff P, Savage G, Ng S, et al. Abetadeposits in older non-demented individuals with cognitive decline areindicative of preclinical Alzheimer’s disease. Neuropsychologia.2008;46(6):1688–97. doi:10.1016/j.neuropsychologia.2008.02.008.57. Forsberg A, Engler H, Almkvist O, Blomquist G, Hagman G, Wall A, et al. PETimaging of amyloid deposition in patients with mild cognitive impairment.Neurobiol Aging. 2008;29(10):1456–65. doi:10.1016/j.neurobiolaging.2007.03.029.58. Pike KE, Savage G, Villemagne VL, Ng S, Moss SA, Maruff P, et al. Beta-amyloidimaging and memory in non-demented individuals: evidence for preclinicalAlzheimer’s disease. Brain. 2007;130(Pt 11):2837–44. doi:10.1093/brain/awm238.59. Okello A, Koivunen J, Edison P, Archer HA, Turkheimer FE, Nagren K, et al.Conversion of amyloid positive and negative MCI to AD over 3 years: an 11C-PIBPET study. Neurology. 2009;73(10):754–60. doi:10.1212/WNL.0b013e3181b23564.60. Wolk DA, Price JC, Saxton JA, Snitz BE, James JA, Lopez OL, et al. Amyloidimaging in mild cognitive impairment subtypes. Ann Neurol. 2009;65(5):557–68.doi:10.1002/ana.21598.61. Collette F, Van der Linden M, Salmon E. Executive dysfunction in Alzheimer’sdisease. Cortex. 1999;35(1):57–72.62. Sgaramella TM, Borgo F, Mondini S, Pasini M, Toso V, Semenza C. Executivedeficits appearing in the initial stage of Alzheimer’s disease. Brain Cogn.2001;46(1–2):264–8.63. Gibbons LE, Carle AC, Mackin RS, Harvey D, Mukherjee S, Insel P, et al.A composite score for executive functioning, validated in Alzheimer’sDisease Neuroimaging Initiative (ADNI) participants with baseline mildcognitive impairment. Brain Imaging Behav. 2012;6(4):517–27.doi:10.1007/s11682-012-9176-1.64. Lee MJ, Seo SW, Na DL, Kim C, Park JH, Kim GH, et al. Synergistic effects ofischemia and beta-amyloid burden on cognitive decline in patients withsubcortical vascular mild cognitive impairment. JAME Psychiatry.2014;71(4):412–22. doi:10.1001/jamapsychiatry.2013.4506.65. Nordlund A, Rolstad S, Klang O, Edman A, Hansen S, Wallin A. Two-year outcomeof MCI subtypes and aetiologies in the Goteborg MCI study. J Neurol NeurosurgPsychiatry. 2010;81(5):541–6. doi:10.1136/jnnp.2008.171066.66. Bennett DA, Schneider JA, Wilson RS, Bienias JL, Arnold SE. Neurofibrillarytangles mediate the association of amyloid load with clinical Alzheimerdisease and level of cognitive function. Arch Neurol. 2004;61(3):378–84.doi:10.1001/archneur.61.3.378.67. Carmichael O, Schwarz C, Drucker D, Fletcher E, Harvey D, Beckett L, et al.Longitudinal changes in white matter disease and cognition in the first yearof the Alzheimer disease neuroimaging initiative. Arch Neurol.2010;67(11):1370–8. doi:10.1001/archneurol.2010.284.68. Pasi M, Salvadori E, Poggesi A, Ciolli L, Del Bene A, Marini S, et al. Whitematter microstructural damage in small vessel disease is associated withMontreal Cognitive Assessment but not with mini mental state examinationperformances: Vascular Mild Cognitive Impairment Tuscany study. Stroke.2015;46(1):262–4. doi:10.1161/STROKEAHA.114.007553.69. Van Petten C, Plante E, Davidson PS, Kuo TY, Bajuscak L, Glisky EL. Memory andexecutive function in older adults: relationships with temporal and prefrontalgray matter volumes and white matter hyperintensities. Neuropsychologia.2004;42(10):1313–35. doi:10.1016/j.neuropsychologia.2004.02.009.70. Rolstad S, Berg AI, Eckerstrom C, Johansson B, Wallin A. Differential Impactof Neurofilament Light Subunit on Cognition and Functional Outcome inMemory Clinic Patients with and without Vascular Burden. J Alzheimers Dis.2015;45(3):873–81. doi:10.3233/JAD-142694.71. Lee JH, Kim SH, Kim GH, Seo SW, Park HK, Oh SJ, et al. Identification of puresubcortical vascular dementia using 11C-Pittsburgh compound B. Neurology.2011;77(1):18–25. doi:10.1212/WNL.0b013e318221acee.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitDao et al. BMC Neurology  (2015) 15:197 Page 10 of 10


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items