jou rn al hom epage: www.elsev ier .com/ locate /ar r
eview
rterial stiffness, the brain and cognition: A systematic review
oel Singera,c, Julian N. Trollora,c, Bernhard T. Bauned,erminder S. Sachdevb,c, Evelyn Smithc,e,∗
Department of Developmental Disability Neuropsychiatry, School of Psychiatry, University of New South Wales, Sydney, AustraliaNeuropsychiatric Institute, Prince of Wales Hospital, Randwick, NSW, AustraliaCentre of Healthy Brain Ageing (CHeBA), School of Psychiatry, University of New South Wales, Sydney, AustraliaDiscipline of Psychiatry, University of Adelaide, Adelaide, AustraliaSchool of Psychology, University of Sydney, Australia
r t i c l e i n f o
rticle history:eceived 3 February 2014ccepted 5 February 2014vailable online 15 February 2014
Background: Arterial stiffness is a known predictor of cardiovascular disease, and has also been associatedwith markers of cerebral small vessel disease as well as poor cognitive function and cognitive decline.The consistency of these associations and their relationship to each other are unclear.Method: We conducted a systematic review of the evidence associating arterial stiffness with cognitivefunction and cognitive decline, and with makers of cerebral small vessel disease, specifically lacunarinfarcts and white matter hyperintensities.Results: Thirteen cross-sectional studies examining arterial stiffness and white matter hyperintensitiesor lacunar infarctions reported a positive association between increased arterial stiffness and radio-logical findings of cerebral small vessel disease. Two longitudinal studies examining the relationshipbetween arterial stiffness and white matter hyperintensities found increased pulse wave velocity tobe an independent predictor of white matter hyperintensity volume. Fifteen cross-sectional and sevenlongitudinal studies examining arterial stiffness and cognition were identified. Fourteen of the fifteen
cross-sectional studies associated increased arterial stiffness with lower cognitive function, and six ofthe seven longitudinal studies found arterial stiffness to be predictive of cognitive decline.Conclusion: Arterial stiffness is associated with cerebral small vessel disease and decreased cognitivefunction. However methodological limitations such as differing covariates between studies and an over-reliance on the MMSE to measure cognition are a concern across much of the literature.
∗ Corresponding author at: Barker Street, Euroa Centre, The University of New South WE-mail addresses: [email protected], [email protected] (E. Smith).
Cognitive impairment is becoming increasingly common withhe aging of societies, and causes an immense social, economicnd emotional burden (Gorelick et al., 2011a). A greater under-tanding of the mechanisms leading to cognitive impairment withging may hold the potential to prevent cognitive decline ando maintain cognitive ability into old age. Cardiovascular fac-ors have long been recognized as playing a prime role in theascular pathogenesis of cognitive decline, and recent studiesave found several cardiovascular risk factors such as hyper-ension, diabetes mellitus and increased left ventricular masso be linked with cognitive decline (Gorelick et al., 2011b;anon et al., 2005; Henskens et al., 2007). Arterial stiffening,
hallmark of vascular aging caused by structural and cellularhange within vessel walls, is one such pathophysiological pro-ess that has also been associated with cognitive decline (Benetost al., 2012; Gorelick et al., 2011b; Waldstein and Elias, 2001;aldstein et al., 2007; Watson et al., 2011), but it remains
less well-studied risk factor of cognitive impairment in lateife.
The understanding of the pathophysiology of arterial stiffnessas improved in recent years. During aging, increased stiffening
n the central elastic arteries due to haemodynamically inducedragmentation of elastin increases the speed of arterial wave prop-gation, leading reflected waves to arrive at the aorta prematurelyuring systole rather than diastole (O’Rourke et al., 2010). Thisesults in increased systolic pressure, high resting flow and higherow pulsations down the vascular tree (O’Rourke and Hashimoto,007; Zieman et al., 2005). Arterial stiffness is predictive of totalnd cardiovascular mortality and end-stage renal disease (Lacolleyt al., 2008).
Additionally, arterial stiffness has been found to be associ-ted with structural change in the brain, primarily white matteryperintensities, cerebral lacunar infarction and cortical braintrophy (Bateman et al., 2008; Henry Feugeas et al., 2005; Henry-eugeas et al., 2009; Nichols et al., 2011). Microvascular brainesions, and in particular white matter hyperintensities havendependently been associated with cognitive decline (Bozzalit al., 2011; Eckerstrom et al., 2011; Gili et al., 2011; Grambaitet al., 2011) and as such may be part of the mechanism throughhich arterial stiffness is propagated into cognitive decline. In
he literature it has been hypothesized that increased flow pul-ations through the carotid and vertebral arteries extend deepnto the microvasculature of the brain leading to vascular rupturend subsequent micro-hemorrhages, endothelial denudation andhrombotic obstruction (Henskens et al., 2008; O’Rourke, 2007).
hite matter hyperintensities in tracts supplied by the ante-ior and middle cerebral arteries have been related to functionalmpairments, indicating a relationship between cerebrovascularathology, white matter change and cognitive ability (Wang et al.,011).
Despite much advancement in the understanding of arterialtiffness, the role of arterial stiffness in structural brain changesnd as a clinical correlate or predictor of cognitive decline is stillnclear. Clinically, arterial stiffness could be related to acceler-ted cognitive aging and cognitive impairment and decline dueo small vessel disease and could be a risk factor for dementia.n order to address the possible aetiological and clinical role ofrterial stiffness in cognitive aging and related structural brainhanges, we conducted a systematic review combining both clini-al and brain-imaging data. This systematic review has three aims:
1) to examine the associations between arterial stiffness and struc-ural change in the brain, specifically lacunar infarction and white
atter hyperintensities as markers of small vessel disease, (2) toxamine the cross-sectional and longitudinal clinical associations
Reviews 15 (2014) 16–27 17
between arterial stiffness and cognitive impairment and declineand (3) to summarize the reported relationship between arte-rial stiffness and dementia to have a full understanding ofthe possible role of arterial stiffness on the brain. The resultsof this systematic review are discussed taking into accountthe findings of previous reviews on arterial stiffness anddementia.
2. Methods
A search of Medline, PubMed and PsycInfo was conducteduntil February 2012 using the following terms: “arterial stiff-ness”, “arterial stiffening”, “arterial ageing”, “pulse wave velocity”,“augmentation index”, “PWV”, “pulse pressure” or “arterial com-pliance.” For literature dealing with brain structure, the followingterms were combined with arterial stiffness terms: “brain”,“leukoaraiosis”, “small vessel”, “white matter hyperintensities”,“white matter”, “lacunar infarct”, “microvascular”, “MRI”, “cere-bral” or “infarction”. For literature dealing with cognition, thepreviously listed terms were combined with one of the follow-ing: “cognition”, “cognitive decline”, “cognitive function”, “brain”,“mini mental state examination”, or “memory”. In this reviewwe included articles in English that examined arterial stiffnessand cognition with their primary aim to identify the relation-ship between these two constructs. Articles examining arterialstiffness with another condition such as diabetes mellitus, renalfailure or hypertension were excluded. Similarly, studies thatexamined levels of arterial stiffness in patients who had clin-ical dementia were excluded, as a systematic review on thistopic was recently published on this topic. Studies that exam-ined subjects with subjective memory complaints were includeddue to the review’s focus on cognitive decline. Longitudinal stud-ies were only included if they had at least 2 measurements,taken at least 12 months apart. The following measures of arte-rial stiffness were considered acceptable: Carotid-femoral pulsewave velocity (cf-PWV), brachial-ankle pulse wave velocity (ba-PWV), aortic pulse wave velocity, aortic pulse pressure andaugmentation index. Cf-PWV has been established as the gold-standard non-invasive measure of arterial stiffness (Laurent et al.,2006). The measurement is determined via applanation tonom-etry an estimation of the velocity of the propagation of theforward and backward pressure waves between the carotid andfemoral arteries (Lacolley et al., 2008). While (cf-PWV) is thegold standard, PWV may also be measured at other points inthe arterial tree such as brachial-ankle or at the aorta directlyvia ultrasound and both these measures were included due toobserved correlations with cf-PWV of 0.75–0.89 (Sugawara et al.,2010) and 0.68 (Vappou et al., 2011) respectively. The indi-rect measurements of arterial stiffness of aortic pulse pressureand augmentation index were also included as they have beenidentified as predictive of cardiovascular mortality (Vlachopouloset al., 2010). Pulse pressure is measured as systolic blood pres-sure minus diastolic blood pressure (Waldstein et al., 2007), andaugmentation index is the augmentation pressure (a measureof the contribution of wave reflection to systolic arterial pres-sure) divided by pulse pressure as taken as a percentage (Janneret al., 2012). Papers examining small vessel disease were onlyincluded if they used magnetic-resonance imaging. The initialsearch identified fifty-one relevant studies, twenty-nine of whichmet the above inclusion criteria. The rest were excluded dueto examining cohorts defined by the existence of pre-existing
health problems such metabolic disease. However, the meta-analyses and systematic reviews already published examining therelationship between arterial stiffness and dementia are also sum-marized.
18 J. Singer et al. / Ageing Research Reviews 15 (2014) 16–27
Table 1Summary of the studies examining the cross-sectional association between arterial stiffness and white matter hyperintensities or lacunar infarction.
Authors Study population Brain pathologyexamined
Arterial stiffness Covariates Results
Coutinho et al. (2011) 812 subjects with nohistory of myocardialinfarction or stroke, meanage 58. Sourced from theSibships Study in which atleast 2 family membersreceived a diagnosis ofhypertension before theage of 60.
White matterhyperintensities
PWV (carotid-femoral) Age, sex, BMI, pulsepressure, total andhigh-density lipoproteincholesterol, glomerularfiltration rate, history ofhypertension, diabetes,smoking, aspirin use, statinuse, and brain volume.
A higher PWV wasassociated with greaterwhite matterhyperintensity volume(p = 0.002).
Hatanaka et al. (2011) 363 subjects withouthistory of cerebrovasculardisease, mean age 65.9.Sourced from communitybased Ohasama Study.
White matterhyperintensities andlacunar infarcts
PWV (brachial-ankle) Age, sex, BMI, 24-h systolicblood pressure, totalcholesterol, HbA1c, historyof cardiovascular disease,smoking, alcohol, use ofantihypertensives, statinuse and use of medicationfor diabetes.
PWV was significantlyhigher in subjects withwhite materhyperintensities (p = 0.003)and lacunar infarcts(p = 0.02).
Kennedy and Raz(2009)
52 subjects withouthypertension, mean age52.4. Sourced fromcommunity.
White matterhyperintensities
Pulse pressure N/A Higher pulse pressure wasassociated with whitematter change inprefrontal region for olderadults, the anterior limb ofthe internal capsule formiddle-aged adults and noregion in the young(p = 0.05).
Kim et al. (2011) 692 subjects over 55without history of stroke,transient ischemic attackor symptoms or signs ofneurological disease, meanage 63.2. Communitypatients presenting forgeneral check up.
PP over 54mmHG wasassociated with thepresence of advancedwhite matterhyperintensities (OR 2.55;95% CI: 1.03, 6.3)
Kuo et al. (2010) 93 subjects, withouthistory of myocardialinfarction or stroke, meanage 72.5. Sourced fromcommunity.
White matterhyperintensities
PWV (brachial-ankle). Age, sex, metabolic riskfactor index (systolic anddiastolic blood pressure,waist circumference,fasting glucose,triglycerides, high densitylipoprotein), presence ofsilent infarcts, use ofantihypertensives, statinuse and use of medicationfor diabetes.
Those with deep whitematter hyperintensitieshad higher PWV (p = 0.018)than those without whitematter hyperintensities.
Matsumoto et al.(2007)
480 subjects asymptomaticof disease, mean age 51.5.Sourced from community.
Cerebral infarcts PWV (brachial-ankle) Age, sex and history ofhypertension.
Increased PWV wasassociate with silentcerebral infarction(p < 0.01)
Mitchell et al. (2011) 668 subjects aged between69 and 93 with no historyof stroke, transientischemic attack ordementia, mean age 75.4.Sourced from communitybased Reykjavik Study.
White matterhyperintensities andlacunar infarcts
PWV (carotid-femoral),pulse pressure,augmentation index
PWV was associated withincreased white matterhyperintensities(p = 0.018). Carotid pulsepressure and PWV wereassociated with increasedrisk for silent subcorticalinfarcts (p < 0.002).
Ochi et al. (2010) 500 subjects withoutsymptomaticcardiovascular disease,mean age 66.9. Sourcedfrom hospital.
Those with silent cerebrallacunar infarcts hadsignificantly higher PWV(p < 0.05)
Ohmine et al. (2008) 132 subjects asymptomaticof disease, mean age 70.3.Sourced from community.
White matterhyperintensities
PWV (brachial-ankle) N/A Increased PWV wasassociated with theappearance ofperiventricularhyperintensities (p = 0.015)but not deep matterhyperintensities.
J. Singer et al. / Ageing Research Reviews 15 (2014) 16–27 19
Table 1 (Continued)
Authors Study population Brain pathologyexamined
Arterial stiffness Covariates Results
Poels et al. (2012) 1460 subjects, mean age58.2. Sourced fromcommunity basedRotterdam Study.
White matterhyperintensities,lacunar infarcts andcerebral microbleeds.
Increased PWV wasassociated with largerwhite matter volume(difference in WMHvolume per standarddeviation increase in PWV,0.07; 95% CI: 0.02, 0.12).
Saji et al. (2011) 240 subjects, withouthistory of stroke, transientischemic attack andwithout neurologicalabnormalities onexamination, mean age 69.Sourced from hospital.
White matterhyperintensities
PWV (brachial-ankle) Age, sex, hypertension,total cholesterol, diabetes,angina, myocardialinfarction and smoking.
Increased PWV wasindependently associatedwith the presence of whitematter hyperintensities(odds ratio 1.12, 95% CI:1.02, 1.23)
Saji et al. (2012) 240 subjects withouthistory of stroke, transientischemic attack andwithout neurologicalabnormalities onexamination, mean age 69.Sourced from hospital.
Tsao et al. (2013) 1587 subjects, mean age61. Sourced from thecommunity basedFramingham OffspringCohort Study.
White matterhyperintensities andsilent brain infarcts.
PWV (carotid-femoral),pulse pressure.
Years of education,diabetes, atrial fibrillation,left ventricularhypertrophy, smoking,coronary heart disease,congestive heart failure,waist-hip ratio, depression,total and high-densitylipoprotein cholesterol,triglycerides,homocysteine, ApoEgenotype, use ofantihypertensives.
Increased PWV wasassociated with greaterwhite matterhyperintensity volume,and greater prevalence ofsilent cerebral infarcts(p < 0.05). Increased pulsepressure was associatedgreater white matterhyperintensity volume(p = 0.05).
Pulse wave velocity (PWV) is determined via an estimation of the velocity of the propagation of the forward and backward pressure waves between two points on the arterialtree. Pulse pressure (PP) is defined as systolic pressure minus diastolic pressure. Body Mass Index (BMI) is defined as mass (kg)/height m2.M re − dh
3
3w
aoairwwtata(cwsaaat
g
ean arterial pressure (MAP) is defined as diastolic pressure + 1/3(systolic pressuemoglobin (HbA1c) is a measure of blood glucose.
. Results
.1. Cross-sectional association between arterial stiffness andhite matter hyperintensities or lacunar infarctions
Table 1 outlines studies examining the association betweenrterial stiffness and one or more of white matter hyperintensitiesr lacunar infarctions. All thirteen studies reported a significantssociation and all were published in the last 6 years, suggesting its a developing area. However, studies showed conflicting resultsegarding the location of white matter hyperintensities associatedith arterial stiffness. One study found ba-PWV to be associatedith the presence of only peri-ventricular white matter hyperin-
ensities (Ohmine et al., 2008), while another found ba-PWV to bessociated with the presence of only deep white matter hyperin-ensities (Kuo et al., 2010), and one simply confirmed PWV to bessociated with the presence of total white matter hyperintensitiesSaji et al., 2011). One study (Kennedy and Raz, 2009) examining aohort aged 19–84 found increased pulse pressure to be associatedith white matter change in the anterior limb of the internal cap-
ule in middle-aged adults, but in the pre-frontal region in olderdults. Another study also used pulse pressure as a measure ofrterial stiffness and found that pulse pressure over 54 mmHG was
ssociated with the presence of advanced white matter hyperin-ensities (Kim et al., 2011).
Higher PWV (carotid-femoral and aortic) was associated withreater white matter hyperintensity volume (Coutinho et al., 2011;
iastolic pressure). C-reactive protein (CRP) is an inflammatory marker. Glycated
Mitchell et al., 2011; Poels et al., 2012; Tsao et al., 2013), and sub-jects with white matter hyperintensities or lacunar infarcts hadhigher ba-PWV (Hatanaka et al., 2011; Ochi et al., 2010). Finally,ba-PWV is associated with silent brain infarction (Matsumoto et al.,2007; Saji et al., 2012), as is cf-PWV (Tsao et al., 2013). Overall, a con-sistent association between PWV and hyperintensities and infarctswas found.
3.2. Longitudinal association between arterial stiffness and whitematter hyperintensities or lacunar infarctions
Table 3 described the two studies that examine the longi-tudinal relationship between arterial stiffness and white matterhyperintensities. Both identified studies found pulse wave veloc-ity measured at baseline to be an independent predictor of whitematter hyperintensity volume at follow-up after adjusting fordemographic and cardiovascular factors. However while one studyfound PWV to predict overall white matter hyperintensity vol-ume (King et al., 2013) the other found increased baseline arterialstiffness to only be associated with hyperintensity volume in theleft superior longitudinal fasiculus at follow up, but not othertracts (Rosano et al., 2013). It is of note that neuroimaging was
not available at the time of baseline PWV measurements in bothstudies, making it impossible to determine the predictive value ofPWV independent of baseline white matter hyperintensity volume(Table 2).
20 J. Singer et al. / Ageing Research Reviews 15 (2014) 16–27
Table 2Summary of the studies examining the longitudinal association between arterial stiffness and white matter hyperintensities or lacunar infarction.
Authors Study population Follow-up Brain pathologyexamined
Arterialstiffness
Covariates Results
Rosano et al.(2013)
303 subjects, mean age82.9 at follow-up MRIassessment. Sourcedfrom the communitybased Health, aging,and Body Composition(Health ABC) Study.
10 years White matterhyperintensities
PWV (carotid-femoral)
Age, sex, years of education,ethnicity, BMI, physicalactivity, systolic and diastolicblood pressure, pulse pressure,stroke, hypertension,cardiovascular disease,diabetes, smoking, alcohol use,use of antihypertensives.
PWV was anindependent predictorof hyperintensityvolume in the leftsuperior longitudinalfasiculus, but not inother tracts (p = 0.023).
King et al.(2013)
1270 subjects, meanage 51.4 at follow-upMRI assessment.Sourced from thecommunity basedDallas Heart Study.
7 years White matterhyperintensities
PWV (aortic) Age, sex, ethnicity, systolic anddiastolic blood pressure, heartrate, BMI, waist circumference,abnormal glucose tolerance,triglycerides, total cholesterol,low-density lipoprotein,high-density lipoprotein,hypertension, myocardialinfarction, congestive heartfailure, diabetes, smoking,statin use, use of
PWV was anindependent predictorof subsequent whitematter hyperintensityvolume (p < 0.0001).
3c
inos
foHetcis22ic2t(tvpvM2adm
3c
bs2f
.3. Cross sectional association between arterial stiffness andognitive function
Table 3 contains a description of the cross-sectional studiesnvestigating the relationship between arterial stiffness and cog-itive function. The evidence is highly consistent, with fourteenut of fifteen studies reporting an association between high arterialtiffness and low cognitive function.
All studies that examined the relationship between carotid-emoral PWV and global cognitive function as measured by an arrayf neuropsychological tests found an association (Elias et al., 2009;anon et al., 2005; Singer et al., 2013; Tarumi et al., 2013; Watsont al., 2011; Zhong et al., 2014). Studies that examined the rela-ionship between measurements of arterial stiffness and globalognitive function as measured via the Mini-Mental State Exam-nation (MMSE) (Folstein et al., 1975) were inconsistent betweentudies (Fujiwara et al., 2005; Fukuhara et al., 2006; Hanon et al.,005; Mitchell et al., 2011; Poels et al., 2007; Scuteri et al., 2005,013; Sugawara et al., 2010; Zhong et al., 2014). Of these studies,
t was those that examined relatively healthy community basedohorts that found no association (Mitchell et al., 2011; Poels et al.,007; Sugawara et al., 2010; Tarumi et al., 2013). However, three ofhese studies found an association with specific cognitive domainsMitchell et al., 2011; Poels et al., 2007; Tarumi et al., 2013). Studieshat examined the association between arterial stiffness and indi-idualized domains of cognitive function showed an inconsistentattern, with executive function, memory, processing speed andisual-spatial all identified in at least one study (Elias et al., 2009;itchell et al., 2011; Pase et al., 2010; Poels et al., 2007; Singer et al.,
013; Tarumi et al., 2013; Zhong et al., 2014). One study found notssociation between PWV and measures of cognitive function, butid find an association between pulse pressure and a measure ofemory (Tsao et al., 2013).
.4. Longitudinal association between arterial stiffness andognitive function
Table 4 describes the studies that examined the relationship
etween arterial stiffness and cognitive decline, with six out ofeven finding PWV as predictive of cognitive decline (Poels et al.,007). Studies that examined the association between carotid-emoral PWV and cognitive function as measured by MMSE found
antihyptensives and leftventricular hypertrophy.
increased PWV as predictive of decline in MMSE score, with allutilizing nursing-facility or hospital based cohorts (Benetos et al.,2012; Scuteri et al., 2007, 2013; Zeki Al Hazzouri et al., 2013). How-ever one of these studies found a non-linear relationship betweenPWV and cognition, with only subjects in the upper quartile of PWVat an increased risk for developing cognitive dysfunction (Scuteriet al., 2013). An additional study, which measured cognition via themodified MMSE, found arterial stiffness to be associated with fasterrates of cognitive decline when their cohort was divided via tertiles(Zeki Al Hazzouri et al., 2013).
Findings regarding specific cognitive domains were mixed. TheBaltimore Longitudinal Study of Aging (Waldstein et al., 2007)examined the link between both pulse pressure and carotid-femoral PWV with cognition in a younger healthy sample (meanage 57) and found accelerated increases in pulse pressure or higherPWV at baseline to predict decline on tests relating to overall cog-nition, memory and concentration 11 years later. This is consistentwith the cross-sectional studies. However, examination of a healthelderly sub sample (mean age 73) of the Health, Aging and BodyComposition Study (Watson et al., 2011) found carotid-femoralPWV to be predictive of decline in psychomotor speed after 6 years.A second paper examining a larger sample within the same cohortfound PWV to be predictive of faster rates of cognitive decline asmeasured by the modified MMSE, but only when the cohort wasdivided into tertiles.
No longitudinal association between arterial stiffness and cog-nitive decline was found in the Rotterdam Study, a large elderlysample (mean age 70) without dementia (Poels et al., 2007). As theauthors suggest, the results of this study are likely explained by theinclusion of a more comprehensive list of cardiovascular covariates,in particular carotid intima-media thickness (IMT) that is seen tobe an indicator of atherosclerosis. This covariate was not includedin any other studies, as it is argued that atherosclerotic change,which also indicated arterial aging, is linearly associated with arte-rial stiffening and should be viewed in a continuum (Scuteri, 2010;Wykretowicz et al., 2009). Another limitation of the study was thatinformation on cognitive decline was only measured for subjectswho participated in the third and fourth round of examination in
the study. As such, selective attrition may have affected the resultwith older patients with poorer cardiovascular risk factors lesslikely to participate in the third and fourth examinations of theRotterdam Study (Poels et al., 2007).
J. Singer et al. / Ageing Research Reviews 15 (2014) 16–27 21
Table 3Summary of the studies examining the cross-sectional relationship between arterial stiffness and cognitive function.
Authors Study population Arterial stiffnessmeasure
Cognitive tests and outcomesassessed
Covariates Results
Elias et al.(2009)
409 subjects aged24–92, mean age 61.3.Sourced fromCommunity BasedMaine-SyracuseLongitudinal Study.
PWV(carotid-femoral)
Visual-spatial ability: HooperVisual Organization TestProcessing speed: Trail makingTests A, digit substitution andsymbol searchVerbal memory: Logicalmemory (immediate anddelayed) and Hopkins Verballearning Test Workingmemory: Digit span forwardand backward, letter-numbersequence and controlled oralword associationsExecutive function/problemsolving: trail making test B,matrix reasoning
Age, sex, years ofeducation, ethnicity,height, weight, heart rate,MAP, cholesterol,creatinine, plasmahomocysteine, ApoEgenotype, history ofcardiovascular disease(myocardial infarction,coronary artery disease,congestive heart failure,angina, transient ischemicattack), diabetes,depression, anxiety,smoking and use ofantihypertensives.
308 subjects withcomplaints of memoryloss, mean age 78.Geriatric outpatientsclinic.
PWV(carotid-femoral)
Cognitive impairment: MMSEImmediate and delayedmemory (free and cued recall),language, visual-spatial,executive function, attentionand judgment: cognitiveefficiency profileIndependence: activities ofdaily living
Age, sex, years ofeducation, systolic bloodpressure, presence ofcardiovascular diseasesand use ofantihypertensives.
PWV was inverselycorrelated with MMSE(p < 0.001) and cognitiveefficiency profile scores(p < 0.001).
Fujiwaraet al.(2005)
352 subjects aged 70 orover without stroke orischemic heart diseasemean age 75. Sourcedfrom community.
PWV(brachial-ankle)
Cognitive impairment: MMSEIndependence: activities ofdaily living
Age, sex, education, systolic& diastolic blood pressure,pulse pressure, heart rate,HbA1c, albumin, totalcholesterol, high densitylipoprotein, triglyceride,albumin, hypertension,history of hyperlipidaemia,diabetes, smoking, alcohol,antihypertensives.
Poor cognitive functionwas independentlyassociated with the middletertile of PWV (odds ratio9.66, 95% CI: 1.04, 1.22 andthe highest tertile of pulsepressure (odds ratio 4.70,95% CI: 1.08, 20.48)
Fukuharaet al.(2006)
203 subjects, all 85years old. Sourced fromcommunity.
PWV(brachial-ankle)
Cognitive impairment: MMSEProcessing speed: trail-makingtests A
Sex, years of education,BMI, systolic bloodpressure, pulse pressure,total cholesterol, SV +RVvoltage, smoking andalcohol use.
PWV was negativelycorrelated with the MMSEscore (p = 0.001)
Mitchell et al.(2011)
668 subjects agedbetween 69 and 93with no history ofstroke, transientischemic attack ordementia, mean age75.4. Sourced fromcommunity basedReykjavik Study.
Age, sex, years ofeducation, height, weight,heart rate, MAP,high-density lipoprotein,cholesterol glucose,depression, use ofantihypertensives andstatin use.
Increased PWV and pulsepressure were associatedwith lower memory score(p = 0.028, p = 0.013).Augmentation Index wasnot associated withcognitive function.
Pase et al.(2010)
92 healthymiddle-agedvolunteers all agedbetween 40 and 65,mean age 53. Sourcedfrom community.
Pulse pressure andaugmentationindex
Working memory, episodicmemory, power of attention,speed of memory andcontinuity of attention:Cognitive Drug Researchcomputerized-assessment
Age, sex, years ofeducation, BMI and MAP.
Increased Pulse Pressurewas correlated with poorepisodic memory (p < 0.05)and speed of memory(p < 0.05). Augmentationindex was a predictor ofspeed of memory (p < 0.01)
Poels et al.(2007)
2767 subjects aged 55and over withoutdementia, mean age70.7. Sourced fromcommunity basedRotterdam Study.
PWV(carotid-femoral)
Cognitive impairment: MMSEExecutive function: StroopTest; Letter Digit SubstitutionTaskSemantic memory: WordFluency Test
Age, sex, years ofeducation, BMI, MAP, heartrate, total cholesterol,high-density lipoproteincholesterol, diabetes, ApoEgenotype and intima mediathickening of the carotidarteries.
After adjustment forcardiovascular factors,PWV was associated withpoor performance on theStroop Test (p < 0.05).
22 J. Singer et al. / Ageing Research Reviews 15 (2014) 16–27
Table 3 (Continued)
Authors Study population Arterial stiffnessmeasure
Cognitive tests and outcomesassessed
Covariates Results
Scuteri et al.(2005)
84 subjects withmemory problems,mean age 78. Excludedif evidence of stroke,atrial fibrillation orvascular pathology.Sourced from hospital.
PWV(carotid-femoral)
Cognitive impairment: MMSEPersonal independence:activities of daily living
Age, sex, years ofeducation, diabetes,low-density lipoproteinand high-densitylipoprotein cholesterol,history of cardiovasculardisease, use ofantihypertensives andnitrate use.
PWV was inverselycorrelated with MMSEscores (p < 0.05) and withpersonal independence(p < 0.01).
Scuteri et al.(2013)
280 subjects withmemory problems,mean age 78.3.Excluded if they hadcancer or acutemyocardial infarctionin the past 6 months,hepatic or cardiacfailure, serumcreatinine > 2 mg/dl,secondaryhypertension, thyroiddisease or vascularpathology. Sourcedfrom hospital.
PWV(carotid-femoral)
Cognitive impairment: MMSE Age, sex, years ofeducation, BMI, systolicblood pressure, diastolicblood pressure, heart rate,total cholesterol,HDL-cholesterol,LDL-cholesterol,triglycerides, glucose,creatinine, depression,previous cardiovasculardisease, use ofantihypertensives, use ofhyperglycaemicmedication, use ofantiplatelet medication,use of statins and nitrateuse.
PWV was significantlyassociated with poorercognitive function(p < 0.05).
Global cognitive function:Aggregate score of belowtests, Wechsler BlockDesignProcessing speed:Wechsler Digit SymbolCoding, Trial Making Test AMemory: Animal naming,30-item Boston namingTestExecutive function:Phonemic Fluency, TrailMaking Test B, Stroop Test
26 sedentary subjects,mean age 54 and 32endurance-trainedsubjects, mean age 52.Excluded if currentsmoker, hypertensive,usingcardiovascular-actingmedications orcontraindications toMRI. Sourced fromcommunity.
PWV(carotid-femoral)
Cognitive impairment: MMSEGlobal cognitive function:Wechsler Test for AdultReading, aggregate score ofbelow testsMemory: California VerbalLearning Test-II immediaterecall and delayed recallExecutive function:Trail-making Test B, ControlledOral Word Association Test,Wechsler Adult IntelligenceScale-III Digit Span Subtest
Age, sex and years ofeducation.
Lower arterial stiffness wasassociated with highertotal composite cognitivescore and executivefunction (p < 0.05).
Tsao et al.(2013)
1587 subjects, meanage 61. Sourced fromthe community basedFramingham OffspringCohort Study
PWV(carotid-femoral),pulse pressure.
Alzheimer-type aging: Logicalmemory-delayed recallExecutive function: TrailMaking Test Part B minus PartA score.
Years of education,diabetes, atrial fibrillation,ECG left ventricularhypertrophy, smoking,coronary heart disease,congestive heart failure,waist-hip ratio, depression,total and high-densitylipoprotein cholesterol,triglycerides,homocysteine, ApoEgenotype, use ofantihypertensives.
Increased PWV was notassociated with measuresof cognitive function.Greater pulse pressure wasassociated with lowerlogical memory-delayedscores (p = 0.02).
J. Singer et al. / Ageing Research Reviews 15 (2014) 16–27 23
Table 3 (Continued)
Authors Study population Arterial stiffnessmeasure
Cognitive tests and outcomesassessed
Covariates Results
Watson et al.(2011)
552 subjects, mean age73.1. Sourced fromcommunity basedHealth, Aging and BodyComposition (HealthABC) Study.
PWV(carotid-femoral)
Cognitive impairment: TheModified Mini-mental StatusExamVerbal learning and memory:The Buschke SelectiveReminding TestPsychomotor speed: The Boxesand Digit Copying TestProcessing speed: The Patternand letter Comparison Test
Age, sex, years ofeducation, ethnicity, BMI,physical activity, MAP,heart rate, totalcholesterol, depression,history of hypertension,coronary heart disease,cerebrovascular disease,diabetes and smoking.
PWV was inverselyassociated with globalcognitive function(beta = −0.55, 95%CI = −0.91, −0.19),psychomotor (−1.59, 95%CI = −3.03, −0.15) andprocessing speed (−0.60,95% CI = −0.98, −0.22).
Zhong et al.(2014)
1433 subjects, meanage 75. Sourced fromthe community basedEpidemiology ofHearing Loss Study.
PWV(carotid-femoral)
Cognitive impairment: MMSEExecutivefunction/attention/speed: TrailMaking Test A Trail makingtest B, Digit symbolSubstitution TestMemory: Rey Auditory VerbalLearning TestSemantic memory: VerbalFluency Test
Age, sex, BMI, physicalactivity, heart rate,high-density lipoproteincholesterol, HbA1c,depression, history ofhypertension,cardiovascular history(angina, myocardialinfarction, stroke, TIA),smoking and alcohol use.
PWV >12 m/s wasassociated with a lowerMMSE score (p = 0.005),fewer words recalled onthe Auditory VerbalLearning Test (p = 0.01) andlower score on a compositecognition score (p = 0.04).
Pulse wave velocity (PWV) is determined via an estimation of the velocity of the propagation of the forward and backward pressure waves between two points on the arterialtree. Pulse pressure (PP) is defined as systolic pressure minus diastolic pressure. Body Mass Index (BMI) is defined as mass (kg)/height m2.Mean arterial pressure (MAP) is defined as diastolic pressure + 1/3(systolic pressure − diastolic pressure). C-reactive protein (CRP) is an inflammatory marker. Glycatedhemoglobin (HbA1c) is a measure of blood glucose.
3d
nttsdmwbtorAhbM
4
afccibeathdiat2
.5. Summary of the association between arterial stiffness andementia
A meta-analysis of six studies showed arterial stiffness to be sig-ificantly higher in subjects with vascular dementia as comparedo subjects with Alzheimer’s disease and controls without demen-ia (Rabkin and Jarvie, 2011). One study also demonstrated arterialtiffness to be significantly higher in subjects with Alzheimer’sisease than in subjects with mild cognitive impairment or nor-al cognitive function, and significantly higher again in subjectsith vascular dementia (Hanon et al., 2005). While the relationship
etween arterial stiffness and vascular dementia is consistent withhe previously discussed hypotheses, observations of an associationf arterial stiffness with Alzheimer’s dementias are novel and war-ant further research (Hanon et al., 2005; Rabkin and Jarvie, 2011).
recent study found increased ba-PWV to be associated with aigher risk of having beta-amyloid plaques as measured by Pitts-urgh compound B, a marker of amyloid plaques in the brain, onRI and PET imaging (Hughes et al., 2013)
. Discussion
The findings of this systematic review suggest that higherrterial stiffness is consistently associated with poor cognitiveunction and small vessel disease in the elderly, and predictsognitive decline and white matter hyperintensities. Specifically,ross-sectional studies unanimously suggest that arterial stiffnesss associated with small vessel disease and two recent community-ased longitudinal studies seem to confirm this relationship (Kingt al., 2013; Rosano et al., 2013). However, both prospective studiesre limited by their failure to measure white matter hyperintensi-ies at baseline and thus delineate what component of white matteryperintensity volume found at follow-up is not just incident andue to age-related change. Among cross-sectional studies, there
s also an over-reliance on imprecise measurement like brachial-nkle PWV and pulse pressure as measurement, which are nothe gold standard measurement of arterial stiffness (Laurent et al.,006).
While the data examining arterial stiffness and cognitive func-tion are highly suggestive of an association, especially in globalcognition, executive function, processing speed, and memory,several methodological limitations prevent it from being conclu-sive. Within the literature there is an over-reliance on imprecisemeasurements, in regards to both arterial stiffness and cogni-tive function. Similar to the studies examining arterial stiffnessand small vessel disease, many of the studies used measurementssuch as brachial-ankle PWV that are less reflective of chang-ing central arterial stiffness with age (Nichols et al., 2011) andshould be avoided, especially in the elderly. In regards to cogni-tion, the literature has shown an over-reliance on the Mini-MentalState Exam (MMSE), which is a screening tool, rather than aprecise test of cognitive function (Folstein et al., 1975). Whilethe MMSE is a useful tool, it may not be sensitive enough tomeasure subtle cognitive changes in healthy populations. Thisis evidenced by two identified studies using community-basedcohorts that did not report an association between arterial stiff-ness and MMSE score, but did find an association between arterialstiffness and more specific cognitive measurements (Mitchell et al.,2011; Poels et al., 2007). Among the studies investigating indi-vidual cognitive domains, there is substantial inconsistency inthe cognitive tests used to define each domain. Some studies usefactor analysis to identify the tests that define a domain (Eliaset al., 2009), some use a single cognitive test to define a domain(Watson et al., 2011), while others use multiple tests (Singer et al.,2013). While cognitive domains have traditionally been definedby more than one specific cognitive test, some tests have beenvalidated to individually measure specific domains (Lezak et al.,2004).
Additionally, the covariates used across the literature vary con-siderably, with some studies controlling for only basic demographicand cardiovascular risk factors, and others employing multipleadjustments for medical psychological, genetic and lifestyle factors.
Few studies accounted for factors such as depression (Ronnlundet al., 2011); and ApoE genotype (Striepens et al., 2011), factorsthat have been shown to have an independent effect on cogni-tion. It is imperative that a comprehensive list of covariates that
24 J. Singer et al. / Ageing Research Reviews 15 (2014) 16–27
Table 4Summary of the studies examining the longitudinal relationship between arterial stiffness and cognitive function.
Authors Study population Follow-up Arterialstiffness
Cognitive tests andoutcomes assessed
Covariates Results
Benetos et al.(2012)
873 subjects, mean age 87at baseline. Sourced fromnursing home basedPARTAGE Study.
1 year PWV (carotid-femoral)
Cognitive impairment:MMSE
Age, years of education,mean blood pressure andactivities of daily living,baseline MMSE.
According to PWVtertiles, negativechange in MMSE wasprogressively greaterfrom the lowest PWVtertile to the highest(p < 0.03)
Poels et al.(2007)
2767 subjects aged 55 andover without dementia,mean age 70.7 at baseline.Sourced from communitybased Rotterdam Study.
Average: 5years
PWV (carotid-femoral)
Cognitive impairment:MMSEExecutive function:Letter DigitSubstitution Task,Stroop TestSemantic memory:Word Fluency Test
Age, sex, years ofeducation, BMI, MAP, heartrate, total cholesterol,high-density lipoproteincholesterol, smoking,intima media thickening ofthe carotid arteries anddiabetes.
Increased PWV did notpredict cognitivedecline.
Scuteri et al.(2007)
102 subjects whopresented with memoryloss. Individuals withevidence of vessel stroke oratrial fibrillation wereexcluded. Mean age 79 atbaseline. Sourced fromhospital.
Median: 1year
PWV (carotid-femoral)
Cognitive impairment:MMSEPersonalindependence:Activities of DailyLiving
PWV wasindependentlyassociated with annualchange in MMSE scores(p < 0.01).
Scuteri et al.(2013)
105 subjects whopresented with memoryproblems. Excluded if theyhad cancer or acutemyocardial infarction inthe past 6 months, hepaticor cardiac failure, serumcreatinine > 2 mg/dl,secondary hypertension,thyroid disease or vascularpathology. Mean age 77.2at baseline. Sourced fromhospital
Median:15 months
PWV (carotid-femoral)
Cognitive impairment:MMSE
Age, sex, years ofeducation, BMI, systolicblood pressure, diastolicblood pressure, heart rate,total cholesterol,HDL-cholesterol,LDL-cholesterol,triglycerides, glucose,creatinine, depression,previous cardiovasculardisease, use ofantihypertensives, use ofhyperglycaemicmedication, use ofantiplatelet medication,use of statins and nitrateuse.
PWV in the upperquartile of the studydistribution wassignificantly associatedwith cognitiveimpairment (p < 0.05).
Waldstein et al.(2007)
PP Analyses: 1748 subjects,mean age 57.1 at baseline.PWV: 582 subjects, meanage 54.3 at baseline.Sourced from communitybased Baltimore Study.
Maximum:11 years
PWV (carotid-femoral) andpulse pressure
Cognitive impairment:MMSE, BlessedInformation MemoryConcentration Workingmemory: Digit spanForward andBackwardsVerbal learning:California Verballearning Test, Bostonnaming TestNonverbal Memory:Benton VisualRetention TestExecutive function andspeed: TMT A and BPhonetic and semanticassociation: Letter &Category Fluency
Age, sex, years ofeducation, BMI, MAP, heartrate, fasting totalcholesterol, depression,cardiovascularco-morbidity index(history of coronary arterydisease, myocardialinfarction, peripheralarterial disease, congestiveheart failure or diabetes),smoking, alcohol use, useof antihypertensives.
PP was associated withdecline on the BlessedIMC test, CaliforniaVerbal Learning Test,Benton VisualRetention Test anddigits backward(p < 0.05).PWV was associatedwith decline on theBlessed IMC test,California VerbalLearning Test andBenton VisualRetention Test(p < 0.05)
J. Singer et al. / Ageing Research Reviews 15 (2014) 16–27 25
Table 4 (Continued)
Authors Study population Follow-up Arterialstiffness
Cognitive tests andoutcomes assessed
Covariates Results
Watson et al.(2011)
552 subjects, mean age73.1 at baseline. Sourcedfrom community basedHealth, Aging and BodyComposition (Health ABC)Study.
6 years PWV (carotid-femoral)
Cognitive impairment:The ModifiedMini-mental StatusExam.Verbal learning andmemory: The BuschkeSelective RemindingTestPsychomotor speed:The Boxes and DigitCopying TestPerceptual speed: ThePattern and letterComparison Test
Age, sex, years ofeducation, ethnicity, BMI,MAP, total cholesterol,coronary heart disease,hypertension, coronaryheart disease, diabetes,ApoE genotype, smokingand alcohol use.
Higher arterial stiffnesswas associated withfaster rates of cognitivedecline when thecohort was dividedinto tertiles by PWV.Annual change by PWVtertile (95% CI):Low: −0.30 (−0.37,−0.22)Medium: −0.46 (−0.54,0.39)High: −0.53 (−0.53,−0.38)
Pulse wave velocity (PWV) is determined via an estimation of the velocity of the propagation of the forward and backward pressure waves between two points on the arterialtree. Pulse pressure (PP) is defined as systolic pressure minus diastolic pressure. Body Mass Index (BMI) is defined as mass (kg)/height m2.M ure–dh
hri
ahbaeta(nps2aaK
afintm2wlcaaic
ean arterial pressure (MAP) is defined as diastolic pressure + 1/3(systolic pressemoglobin (HbA1c) is a measure of blood glucose.
ave been theoretically linked or empirically shown to affect theelationship between arterial stiffness and cognition emerges, ands utilized in future research.
Despite the above limitations, the findings of this review suggest positive association between arterial stiffness and white matteryperintensities and cerebral infarcts, and a negative associationetween arterial stiffness and cognitive performance. These datare consistent with the data on arterial stiffness and dementia,specially with Vascular Dementia. Overall, these findings supporthe hypothesis that pulsations and flow extend through the carotidnd vertebral arteries into the microcirculation as mentioned aboveHatanaka et al., 2011; Mitchell et al., 2011). It is thought that lacu-ar infarcts are caused by deficient absorption of the increasedulse wave cause by arterial stiffening, which augments small ves-el disease of the brain through high pulsatile flow (Brandts et al.,009). However it does remain a distinct possibility that the associ-tion between arterial stiffness and small vessel disease is indirectnd perhaps due to shared vascular risk factors (Brandts et al., 2009;ing et al., 2013).
It is hypothesized that the relationship between arterial stiffnessnd cognition may be mediated by small vessel disease. One identi-ed study found an association between cf-PWV and memory to beo longer significant when measures of brain structure (white mat-er hyperintensity volume, subcortical infarcts and gray and white
atter volume) were included in cognitive models (Mitchell et al.,011). Studies have found white matter change to be associatedith cognitive function, however it is of note that these studies are
argely reliant on correlational methods, and as such cannot prove aausal relationship between finding white matter hyperintensities
nd poor cognitive performance (Salthouse, 2011). Further studiesre needed to determine the degree to which small vessel disease ismplicated in explaining the relationship between arterial stiffnessognition.
iastolic pressure). C-reactive protein (CRP) is an inflammatory marker. Glycated
Although the majority of studies identified an associationbetween arterial stiffness and cognitive function, arterial stiffnesshas yet to be proven applicable as a predictor of cognitive declineto be utilized by clinicians. The procedure of PWV measurement issufficiently inexpensive and easy to perform routinely in a clin-ical setting. However measuring arterial stiffness will only be auseful tool to assess cognitive aging if arterial stiffness is foundto be definitively predictive of cognitive decline, and treatmentsare available that both reduce arterial stiffness and slow or haltthe deterioration of cognitive function. As previously mentioned,further studies are needed to both clarify and confirm the strengthand characteristics of the association between arterial stiffness andcognitive decline. Further longitudinal studies examining arterialstiffness and small vessel disease that have baseline brain imagingwould be of benefit, as would further studies examining arterialstiffness and cognitive function that use a comprehensive array ofcovariates. If arterial stiffness were found to be definitively predic-tive of cognitive decline, large population studies identifying levelsof arterial stiffness that indicate increased risk of cognitive declinewould also be necessary in order to establish clinical referenceranges.
More research is also necessary to determine the efficacy oftreatments for arterial stiffness. While antihypertensive agentssuch as ACE-inhibitors, angiotensin II receptor antagonist, cal-cium channel blockers and nitrates have been proven to markedlyreduce wave reflection and pulse pressure (Ait-Oufella et al., 2010;O’Rourke and Safar, 2005), it is uncertain whether reducing arte-rial stiffness has the potential to prevent age-related cognitivedecline. A recent Cochrane Review examining the effectiveness of
anti-hypertensives in preventing dementia and cognitive declinein patients with hypertension found no evidence of efficacy(McGuinness et al., 2009). However, there were many limitationsto these studies and the authors suggested further investigations
2 earch
aabvca2srIaei2
5
baoe
A5
R
A
B
B
B
B
B
C
E
E
F
F
F
G
G
6 J. Singer et al. / Ageing Res
re needed (McGuinness et al., 2009). It should also be noted thatlthough the effects of different classes of anti-hypertensives onrachial blood pressure may be similar, their effects centrally arearied (Mackenzie et al., 2009). In turn it is unlikely that the effi-acy of pharmacological treatments for arterial stiffness will beble to be monitored via brachial blood pressure (Hirata et al.,005). In addition to pharmacological treatment, an upcomingtudy hopes to evaluate the effects of diet and exercise on arte-ial stiffness and neurocognitive function (Blumenthal et al., 2013).ndeed one recent study reported arterial stiffness to be lowernd neuropsychological performance to be greater in a group ofndurance-trained individuals as compared to a group of sedentaryndividuals independent of age, sex and education (Tarumi et al.,013).
. Conclusions
In conclusion the review of the evidence suggests an associationetween arterial stiffness and cerebral small vessel disease, andrterial stiffness and decreased cognitive function. The clinical usef arterial stiffness as a predictor of cognitive decline is yet to bestablished.Acknowledgements
This work was supported by a Dementia Research Grant from theustralian National Health and Medical Research Council (grant ID10124).
eferences
it-Oufella, H., Collin, C., Bozec, E., Laloux, B., Ong, K.-T., Dufouil, C., Boutouyrie, P.,Laurent, S., 2010. Long-term reduction in aortic stiffness: a 5.3-year follow-upin routine clinical practice. J. Hypertens. 28, 2336–2341.
ateman, G.A., Levi, C.R., Schofield, P., Wang, Y., Lovett, E.C., 2008. The venous man-ifestations of pulse wave encephalopathy: windkessel dysfunction in normalaging and senile dementia. Neuroradiology 50, 491–497.
enetos, A., Watfa, G., Hanon, O., Salvi, P., Fantin, F., Toulza, O., Manckoundia, P.,Agnoletti, D., Labat, C., Gautier, S., 2012. Pulse wave velocity is associated with1-year cognitive decline in the elderly older than 80 years: the PARTAGE study.J. Am. Med. Dir. Assoc. 13, 239–243.
lumenthal, J.A., Smith, P.J., Welsh-Bohmer, K., Babyak, M.A., Browndyke, J., Lin,P.H., Doraiswamy, P.M., Burke, J., Kraus, W., Hinderliter, A., Sherwood, A., 2013.Can lifestyle modification improve neurocognition? Rationale and design of theENLIGHTEN clinical trial. Contemp. Clin. Trials 34, 60–69.
ozzali, M., Padovani, A., Caltagirone, C., Borroni, B., 2011. Regional grey matter lossand brain disconnection across Alzheimer disease evolution. Curr. Med. Chem.18, 2452–2458.
randts, A., van Elderen, S.G., Westenberg, J.J., van der Grond, J., van Buchem, M.A.,Huisman, M.V., Kroft, L.J., Tamsma, J.T., de Roos, A., 2009. Association of aorticarch pulse wave velocity with left ventricular mass and lacunar brain infarcts inhypertensive patients: assessment with MR imaging. Radiology 253, 681–688.
outinho, T., Turner, S.T., Kullo, I.J., 2011. Aortic pulse wave velocity is associatedwith measures of subclinical target organ damage. JACC Cardiovasc. Imaging 4,754–761.
ckerstrom, C., Olsson, E., Klasson, N., Bjerke, M., Gothlin, M., Jonsson, M., Rolstad,S., Malmgren, H., Wallin, A., Edman, A., 2011. High white matter lesion load isassociated with hippocampal atrophy in mild cognitive impairment. Dement.Geriatr. Cogn. Disord. 31, 132–138.
olstein, M.F., Folstein, S.E., McHugh, P.R., 1975. Mini-mental state. A practicalmethod for grading the cognitive state of patients for the clinician. J. Psychiatr.Res. 12, 189–198.
ujiwara, Y., Chaves, P.H., Takahashi, R., Amano, H., Yoshida, H., Kumagai, S., Fujita,K., Wang, D.G., Shinkai, S., 2005. Arterial pulse wave velocity as a marker of poorcognitive function in an elderly community-dwelling population. J. Gerontol. A.Biol. Sci. Med. Sci. 60, 607–612.
ukuhara, M., Matsumura, K., Ansai, T., Takata, Y., Sonoki, K., Akifusa, S., Wakisaka,M., Hamasaki, T., Fujisawa, K., Yoshida, A., Fujii, K., Iida, M., Takehara, T., 2006.Prediction of cognitive function by arterial stiffness in the very elderly. Circ. J.70, 756–761.
ili, T., Cercignani, M., Serra, L., Perri, R., Giove, F., Maraviglia, B., Caltagirone, C.,Bozzali, M., 2011. Regional brain atrophy and functional disconnection across
Alzheimer’s disease evolution. J. Neurol. Neurosurg. Psychiatry 82, 58–66.
cognitive impairment and dementia: a statement for healthcare professionalsfrom the American Heart Association/American Stroke Association. Stroke 42,2672–2713.
Gorelick, P.B., Scuteri, A., Black, S.E., DeCarli, C., Greenberg, S.M., Iadecola, C., Launer,L.J., Laurent, S., Lopez, O.L., Nyenhuis, D., Petersen, R.C., Schneider, J.A., Tzourio,C., Arnett, D.K., Bennett, D.A., Chui, H.C., Higashida, R.T., Lindquist, R., Nilsson,P.M., Roman, G.C., Sellke, F.W., Seshadri, S., 2011b. Vascular contributions tocognitive impairment and dementia: a statement for healthcare professionalsfrom the American Heart Association/American Stroke Association. Stroke 42,2672–2713.
Grambaite, R., Reinvang, I., Selnes, P., Fjell, A.M., Walhovd, K.B., Stenset, V., Fladby, T.,2011. Pre-dementia memory impairment is associated with white matter tractaffection. J. Int. Neuropsychol. Soc. 17, 143–153.
Hanon, O., Haulon, S., Lenoir, H., Seux, M.L., Rigaud, A.S., Safar, M., Girerd, X., Forette,F., 2005. Relationship between arterial stiffness and cognitive function in elderlysubjects with complaints of memory loss. Stroke 36, 2193–2197.
Hatanaka, R., Obara, T., Watabe, D., Ishikawa, T., Kondo, T., Ishikura, K., Aikawa, T.,Aono, Y., Hara, A., Metoki, H., Asayama, K., Kikuya, M., Mano, N., Ohkubo, T., Izumi,S., Imai, Y., 2011. Association of arterial stiffness with silent cerebrovascularlesions: the Ohasama study. Cerebrovasc. Dis. 31, 329–337.
Henry Feugeas, M.C., De Marco, G., Peretti, I.I., Godon-Hardy, S., Fredy, D., Claeys,E.S., 2005. Age-related cerebral white matter changes and pulse-wave encepha-lopathy: observations with three-dimensional MRI. Magn. Reson. Imaging 23,929–937.
Henry-Feugeas, M.C., Roy, C., Baron, G., Schouman-Claeys, E., 2009. Leukoaraiosisand pulse-wave encephalopathy: observations with phase-contrast MRI in mildcognitive impairment. J. Neuroradiol. 36, 212–218.
Henskens, L.H.G., Kroon, A.A., van Oostenbrugge, R.J., Gronenschild, E.H.B.M., Fuss-Lejeune, M.M.J.J., Hofman, P.A.M., Lodder, J., de Leeuw, P.W., 2008. Increasedaortic pulse wave velocity is associated with silent cerebral small-vessel diseasein hypertensive patients. Hypertension 52, 1120–1126.
Henskens, L.H.G., van Oostenbrugge, R.J., Kroon, A.A., de Leeuw, P.W., Lodder, J., 2007.Brain microbleeds are associated with ambulatory blood pressure levels in ahypertensive population. Hypertension 51, 62–68.
Hirata, K., Vlachopoulos, C., Adji, A., O’Rourke, M.F., 2005. Benefits from angiotensin-converting enzyme inhibitor ‘beyond blood pressure lowering’: beyond bloodpressure or beyond the brachial artery. J. Hypertens. 23, 551–556.
Hughes, T.M., Kuller, L.H., Barinas-Mitchell, E.J., Mackey, R.H., McDade, E.M., Klunk,W.E., Aizenstein, H.J., Cohen, A.D., Snitz, B.E., Mathis, C.A., Dekosky, S.T., Lopez,O.L., 2013. Pulse wave velocity is associated with beta-amyloid deposition in thebrains of very elderly adults. Neurology 81, 1711–1718.
Janner, J.H., Godtfredsen, N.S., Ladelund, S., Vestbo, J., Prescott, E., 2012. The asso-ciation between aortic augmentation index and cardiovascular risk factors in alarge unselected population. J. Hum. Hypertens. 26, 476–484.
Kennedy, K.M., Raz, N., 2009. Pattern of normal age-related regional differences inwhite matter microstructure is modified by vascular risk. Brain Res. 1297, 41–56.
Kim, C.K., Lee, S.-H., Kim, B.J., Ryu, W.-S., Yoon, B.-W., 2011. Age-independent asso-ciation of pulse pressure with cerebral white matter lesions in asymptomaticelderly individuals. J. Hypertens. 29, 325–329.
King, K.S., Chen, K.X., Hulsey, K.M., McColl, R.W., Weiner, M.F., Nakonezny, P.A.,Peshock, R.M., 2013. White matter hyperintensities: use of aortic arch pulsewave velocity to predict volume independent of other cardiovascular risk fac-tors. Radiology 267, 709–717.
Lacolley, P., Challande, P., Osborne-Pellegrin, M., Regnault, V., 2008. Genetics andpathophysiology of arterial stiffness. Cardiovasc. Res. 81, 637–648.
Laurent, S., Cockcroft, J., Van Bortel, L., Boutouyrie, P., Giannattasio, C., Hayoz, D.,Pannier, B., Vlachopoulos, C., Wilkinson, I., Struijker-Boudier, H., 2006. Expertconsensus document on arterial stiffness: methodological issues and clinicalapplications. Eur. Heart J. 27, 2588–2605.
Lezak, M., Howieson, D., Bigler, E., Tranel, D., 2004. Neuropsychological Assessment,5th ed. Oxford University Press, New York, USA.
Mackenzie, I.S., McEniery, C.M., Dhakam, Z., Brown, M.J., Cockcroft, J.R., Wilkinson,I.B., 2009. Comparison of the effects of antihypertensive agents on central bloodpressure and arterial stiffness in isolated systolic hypertension. Hypertension54, 409–413.
Matsumoto, M., Inoue, K., Moriki, A., 2007. Associations of brachial-ankle pulsewave velocity and carotid atherosclerotic lesions with silent cerebral lesions.Hypertens. Res. 30, 767–773.
McGuinness, B., Todd, S., Passmore, P., Bullock, R., 2009. Blood pressure loweringin patients without prior cerebrovascular disease for prevention of cognitiveimpairment and dementia. Cochrane Database Syst. Rev., CD004034.
Mitchell, G.F., van Buchem, M.A., Sigurdsson, S., Gotal, J.D., Jonsdottir, M.K., Kjar-tansson, O., Garcia, M., Aspelund, T., Harris, T.B., Gudnason, V., Launer, L.J., 2011.Arterial stiffness, pressure and flow pulsatility and brain structure and func-tion: the age, gene/environment susceptibility – Reykjavik Study. Brain 134,3398–3407.
Nichols, W., O’Rourke, M., Vlachopoulos, C., 2011. McDonald’s Blood Flow in Arteries,
’Rourke, M.F., Safar, M.E., 2005. Relationship between aortic stiffening andmicrovascular disease in brain and kidney: cause and logic of therapy. Hyper-tension 46, 200–204.
’Rourke, M.F., Hashimoto, J., 2007. Mechanical factors in arterial aging. J. Am. Coll.Cardiol. 50, 1–13.
chi, N., Kohara, K., Tabara, Y., Nagai, T., Kido, T., Uetani, E., Ochi, M., Igase, M., Miki,T., 2010. Association of central systolic blood pressure with intracerebral smallvessel disease in Japanese. Am. J. Hypertens. 23, 889–894.
hmine, T., Miwa, Y., Yao, H., Yuzuriha, T., Takashima, Y., Uchino, A., Takahashi-Yanaga, F., Morimoto, S., Maehara, Y., Sasaguri, T., 2008. Association betweenarterial stiffness and cerebral white matter lesions in community-dwellingelderly subjects. Hypertens. Res. 31, 75–81.
ase, M.P., Pipingas, A., Kras, M., Nolidin, K., Gibbs, A.L., Wesnes, K.A., Scholey, A.B.,Stough, C., 2010. Healthy middle-aged individuals are vulnerable to cognitivedeficits as a result of increased arterial stiffness. J. Hypertens. 28, 1724–1729.
oels, M.M.F., van Oijen, M., Mattace-Raso, F.U.S., Hofman, A., Koudstaal, P.J., Witte-man, J.C.M., Breteler, M.M.B., 2007. Arterial stiffness, cognitive decline, and riskof dementia: the Rotterdam study. Stroke 38, 888–892.
oels, M.M., Zaccai, K., Verwoert, G.C., Vernooij, M.W., Hofman, A., van der Lugt, A.,Witteman, J.C., Breteler, M.M., Mattace-Raso, F.U., Ikram, M.A., 2012. Arterialstiffness and cerebral small vessel disease: the Rotterdam scan study. Stroke 43,2637–2642.
abkin, S.W., Jarvie, G., 2011. Comparison of vascular stiffness in vascular dementia,Alzheimer dementia and cognitive impairment. Blood Press. 20, 274–283.
onnlund, M., Vestergren, P., Mantyla, T., Nilsson, L.G., 2011. Predictors of self-reported prospective and retrospective memory in a population-based sampleof older adults. J. Genet. Psychol. 172, 266–284.
osano, C., Watson, N., Chang, Y., Newman, A.B., Aizenstein, H.J., Du, Y., Venkatraman,V., Harris, T.B., Barinas-Mitchell, E., Sutton-Tyrrell, K., 2013. Aortic pulse wavevelocity predicts focal white matter hyperintensities in a biracial cohort of olderadults. Hypertension 61, 160–165.
aji, N., Kimura, K., Shimizu, H., Kita, Y., 2012. Association between silent brain infarctand arterial stiffness indicated by brachial-ankle pulse wave velocity. Intern.Med. 51, 1003–1008.
aji, N., Shimizu, H., Kawarai, T., Tadano, M., Kita, Y., Yokono, K., 2011. Increasedbrachial-ankle pulse wave velocity is independently associated with white mat-ter hyperintensities. Neuroepidemiology 36, 252–257.
cuteri, A., 2010. Evaluating arterial aging in the clinical setting: a tentative agendafor critical appraisal. Aging Health 6, 243–249.
cuteri, A., Brancati, A.M., Gianni, W., Assisi, A., Volpe, M., 2005. Arterial stiffness isan independent risk factor for cognitive impairment in the elderly: a pilot study.J. Hypertens. 23, 1211–1216.
cuteri, A., Tesauro, M., Appolloni, S., Preziosi, F., Brancati, A.M., Volpe, M., 2007. Arte-rial stiffness as an independent predictor of longitudinal changes in cognitive
function in the older individual. J. Hypertens. 25, 1035–1040.
cuteri, A., Tesauro, M., Guglini, L., Lauro, D., Fini, M., Di Daniele, N., 2013. Aortic stiff-ness and hypotension episodes are associated with impaired cognitive functionin older subjects with subjective complaints of memory loss. Int. J. Cardiol. 169,371–377.
Reviews 15 (2014) 16–27 27
Singer, J., Trollor, J.N., Crawford, J., O’Rourke, M.F., Baune, B.T., Brodaty, H., Sama-ras, K., Kochan, N.A., Campbell, L., Sachdev, P.S., Smith, E., 2013. The associationbetween pulse wave velocity and cognitive function: the Sydney memory andageing study. PLoS One 8, e61855.
Striepens, N., Scheef, L., Wind, A., Meiberth, D., Popp, J., Spottke, A., Kolsch, H., Wag-ner, M., Jessen, F., 2011. Interaction effects of subjective memory impairmentand ApoE4 genotype on episodic memory and hippocampal volume. Psychol.Med. 41, 1997–2006.
Sugawara, N., Yasui-Furukori, N., Umeda, T., Kaneda, A., Sato, Y., Takahashi, I., Mat-suzaka, M., Danjo, K., Nakaji, S., Kaneko, S., 2010. Comparison of ankle-brachialpressure index and pulse wave velocity as markers of cognitive function in acommunity-dwelling population. BMC Psychiatry 10, 46.
Tarumi, T., Gonzales, M.M., Fallow, B., Nualnim, N., Pyron, M., Tanaka, H., Haley,A.P., 2013. Central artery stiffness, neuropsychological function, and cerebralperfusion in sedentary and endurance-trained middle-aged adults. J. Hypertens.31, 2400–2409.
Tsao, C.W., Seshadri, S., Beiser, A.S., Westwood, A.J., Decarli, C., Au, R., Himali, J.J.,Hamburg, N.M., Vita, J.A., Levy, D., Larson, M.G., Benjamin, E.J., Wolf, P.A., Vasan,R.S., Mitchell, G.F., 2013. Relations of arterial stiffness and endothelial functionto brain aging in the community. Neurology 81, 984–991.
Vappou, J., Luo, J., Okajima, K., Di Tullio, M., Konofagou, E., 2011. Aortic pulse wavevelocity measured by pulse wave imaging (PWI): a comparison with applanationtonometry. Artery Res. 5, 65–71.
Vlachopoulos, C., Aznaouridis, K., Stefanadis, C., 2010. Prediction of cardiovascularevents and all-cause mortality with arterial stiffness: a systematic review andmeta-analysis. J. Am. Coll. Cardiol. 55, 1318–1327.
Waldstein, S., Elias, M., 2001. Neuropsychology of Cardiovascular Disease. LawrenceErlbaum Associates, Mahwah, NJ.
Waldstein, S.R., Rice, S.C., Thayer, J.F., Najjar, S.S., Scuteri, A., Zonderman, A.B., 2007.Pulse pressure and pulse wave velocity are related to cognitive decline in thebaltimore longitudinal study of aging. Hypertension 51, 99–104.
Wang, L., Goldstein, F.C., Levey, A.I., Lah, J.J., Meltzer, C.C., Holder, C.A., Mao, H., 2011.White matter hyperintensities and changes in white matter integrity in patientswith Alzheimer’s disease. Neuroradiology 53, 373–381.
Watson, N.L., Sutton-Tyrrell, K., Rosano, C., Boudreau, R.M., Hardy, S.E., Simonsick,E.M., Najjar, S.S., Launer, L.J., Yaffe, K., Atkinson, H.H., Satterfield, S., Newman,A.B., 2011. Arterial stiffness and cognitive decline in well-functioning olderadults. J. Gerontol. A. Biol. Sci. Med. Sci. 66, 1336–1342.
Wykretowicz, A., Gerstenberger, P., Guzik, P., Milewska, A., Krauze, T., Adamska,K., Rutkowska, A., Wysocki, H., 2009. Arterial stiffness in relation to subclinicalatherosclerosis. Eur. J. Clin. Invest. 39, 11–16.
Zeki Al Hazzouri, A., Newman, A.B., Simonsick, E., Sink, K.M., Sutton Tyrrell, K., Wat-son, N., Satterfield, S., Harris, T., Yaffe, K., 2013. Pulse wave velocity and cognitivedecline in elders: the Health, Aging, and Body Composition study. Stroke 44,388–393.
