O. Okusaga et al.
29. Launois R, Giroud M, Mégnigbeto A et al. Estimating the
cost-effectiveness of stroke units in France compared with
conventional care. Stroke 2004; 35: 770–5.
30. Sinclair SE, Frighetto L, Loewen PS et al. Cost-utility
analysis of tissue plasminogen activator therapy for acute ischaemic stroke: a Canadian healthcare perspective.
Pharmacoeconomics 2001; 19: 927–36.
31. Mar J, Begiristain JM, Arrazola A. Cost-effectiveness analysis
of thrombolytic treatment for stroke. Cerebrovasc Dis 2005;
20: 193–200.
32. Mar J, Sainz-Ezkerra M, Miranda-Serrano E. Calculation of
prevalence with Markov models: budget impact analysis of
Age and Ageing 2013; 42: 306–311
doi: 10.1093/ageing/afs193
Published electronically 9 January 2013
thrombolysis for stroke. Med Decis Making 2008; 28( Jul–
Aug): 481–90.
33. Yip TR, Demaerschalk BM. Estimated cost savings of
increased use of intravenous tissue plasminogen activator
for acute ischemic stroke in Canada. Stroke 2007; 38:
1952–5.
34. Pendlebury S, Rothwell PM, Algra A et al. Underfunding of
stroke research: a Europe-wide problem. Stroke 2004; 35:
2368–71.
Received 15 January 2012; accepted in revised form
16 November 2012
© The Author 2013. Published by Oxford University Press on behalf of the British Geriatrics Society.
All rights reserved. For Permissions, please email: [email protected]
Smoking, hypercholesterolaemia and
hypertension as risk factors for cognitive
impairment in older adults
OLAOLUWA OKUSAGA1, MARLENE C. W. STEWART2, ISABELLA BUTCHER2, IAN DEARY3,4, F. GERRY R. FOWKES2,5,
JACKIE F. PRICE2,4
1
Department of Psychiatry and Behavioral Sciences, The University of Texas Health Science Center at Houston, Houston, TX, USA
Centre for Population Health Sciences, University of Edinburgh, Edinburgh, UK
3
Department of Psychology, University of Edinburgh, Edinburgh, UK
4
Centre for Cognitive Ageing and Cognitive Epidemiology, University of Edinburgh, Edinburgh, UK
5
Wolfson Unit for the Prevention of Peripheral Vascular Diseases, Department of Public Health Sciences, The University of
Edinburgh, Teviot Place, Edinburgh EH8 9AG, UK
2
Address correspondence to: O. Okusaga. Tel: (+1) 7137413952; Fax: (+1) 7137416909. Email: [email protected]
Abstract
Background: the prevalence of all types of cognitive impairment, including dementia, is increasing but knowledge of aetiological factors is still evolving.
Objective: this study aimed to evaluate the association between cardiovascular risk factors and cognitive function in older
persons.
Design, setting and subjects: a population-based cohort design involving 2,312 men and women (aged 50–75) enrolled
in the University of Edinburgh Aspirin for Asymptomatic Atherosclerosis trial.
Methods: cognitive tests included the Mill Hill Vocabulary Scale, auditory verbal learning test (AVLT), digit symbol test,
verbal fluency test (VFT), Raven’s Progressive Matrices and the trail making test. A ‘g’ score (measure of general intelligence) was computed for each subject. Regression analysis was used to evaluate the association between relevant variables.
Results: higher diastolic BP was negatively associated with AVLT (β = −0.153, P < 0.01), and with an estimated decline on
AVLT (β = −0.125, P < 0.01). Smoking was negatively associated with all the cognitive variables except VFT. The total cholesterol level was not associated with cognitive function or estimated decline.
Conclusions: smoking and elevated blood pressure may be risk factors for cognitive decline, and thus potential targets for
preventive and therapeutic interventions.
Keywords: hypertension, smoking, cholesterol, cognition, older people
306
Cardiovascular risk factors and cognition
Introduction
Methods
The increasing proportion of older adults in the general
population means that the number of people with all types
of cognitive impairment, including dementia, is increasing.
In the UK for example, it is projected that 1.7 million individuals will be diagnosed with dementia by 2051 [1], an increase of 154% from the current estimate of 700,000. An
additional fifth of the population is affected by age-related
cognitive impairment, which falls short of frank dementia
[2]. The cost of caring for those with significant cognitive
impairment, in terms of the burden on caregivers and
financial expense is immense [3] and this underlies the
need for a better understanding of factors contributing to
the pathogenesis of cognitive impairment in older adults to
inform the development of preventive and therapeutic
measures.
Links between vascular disease and cognitive impairment in older age are increasingly recognised. Vascular
disease predisposes to cerebrovascular accidents
(strokes) which in turn impact negatively on cognitive
abilities and may lead to vascular dementia.
Atherogenic levels of vascular risk factors, including
cigarette smoking, hypercholesterolaemia and hypertension, may contribute to the development of both cardiovascular disease and cognitive impairment. If this is
the case, then treatment of such risk factors may help
reduce rates of cognitive decline in older age. To date,
few clinical trials have been performed to address this
possibility and these have produced inconclusive results
[4, 5]. Furthermore, previous studies have demonstrated
an association between worse cognitive ability and
increased smoking [6], hypercholesterolaemia [7] and
hypertension [6], but these findings notwithstanding, the
National Institute of Health State-of-the-Science
Conference Statement on Preventing Alzheimer Disease
and Cognitive Decline concluded that ‘currently, firm
conclusions cannot be drawn about the association of
any modifiable risk factor with cognitive decline or
Alzheimer disease’ [8] and additional studies have been
advocated for.
To expand the current understanding of the connection between cardiovascular risk factors and cognitive
function as well as to overcome the limitations of
small sample size and less comprehensive cognitive
testing in a number of previous studies [9, 10], we
examined the association of smoking status, raised
blood pressure/anti-hypertensive medication use and
raised plasma total cholesterol/cholesterol-lowering
medication with both cognitive ability and estimated
cognitive decline. We carried out a large, prospective
study of individuals free of symptomatic cardiovascular
disease at baseline. We used a battery of individual
cognitive tests applied 5 years after entry into the study
and estimated pre-morbid cognitive ability at baseline to
enable an estimation of decline in cognitive ability since
early adult life.
Study population
Data were derived from the Aspirin for Asymptomatic
Atherosclerosis (AAA) trial. The AAA trial is a randomised
controlled trial of low-dose aspirin in the prevention of cardiovascular events and deaths in asymptomatic subjects at
high risk of cardiovascular events (i.e. those with ankle brachial pressure index ≤0.95). Details of how subjects were
recruited into this trial and the characteristics of the subjects have been described elsewhere [5]. In brief, 3,350 individuals aged 50–75 years were recruited from general
practices in central Scotland. Subjects were excluded if they
had a history of cardiovascular disease, aspirin or other
anticoagulant use and any contraindication to aspirin. Of
the total 3,350 individuals, 2,312 initially recruited into the
AAA trial underwent detailed cognitive testing at the 5-year
follow-up. The remaining 1,038 did not undergo cognitive
testing for the following reasons: 160 died, 157 withdrew,
668 refused cognitive testing, 40 could not be contacted
and testing was incomplete in 13.
Clinical and cognitive examination
Details of physical and cognitive assessments have been
described previously [5]. Ankle brachial index (ABI) and
brachial blood pressure were measured at baseline. ABI was
calculated as the ratio of the lowest systolic blood pressure
in either ankle (measured with a Doppler probe) to the
higher of either arm systolic pressure (measured with a
standard sphygmomanometer). Plasma total cholesterol
levels and smoking status (current, previous and never,
where previous smokers were those who had stopped
smoking at least 6 months prior to examination) were measured at baseline. The Mill Hill Vocabulary Scale (MHVS)
( junior and senior synonyms combined) [11], also administered at baseline, was used to estimate peak prior (pre-morbid)
cognitive functioning, on the basis that vocabulary scores
have been shown to change very little during ageing [12].
At the 5-year follow-up a cognitive test battery was
administered by trained researchers, including: the
Mini-Mental State Examination (screening test for dementia) [13], the verbal fluency test (VFT) (a test of executive
function) [14], Raven’s Progressive Matrices (assesses nonverbal reasoning) [11], auditory verbal learning test (AVLT)
(assesses verbal learning, immediate and delayed memory)
[14], trail making test (TMT) (visual search, scanning, speed
of processing, mental flexibility and executive functions)
[14] and the digit symbol test (assesses processing speed)
[15]. Principal components analysis [16] was used to generate a general cognitive ability score (designated ‘g’) using information from the test battery (excluding Mini-Mental
State Examination). With the aid of a scree plot, one component was identified and extracted as the first unrotated
principal component. The principal component (‘g’)
accounted for 56.9% of the variance in the cognitive test
battery data. All five cognitive tests loaded strongly on ‘g’.
307
O. Okusaga et al.
Serum total cholesterol levels were measured on a
venous blood sample at the Biochemistry Department,
Gartnavel General Hospital, Glasgow on a Beckman CX4
analyser using the cholesterol oxidase method.
Table 1. Characteristics of the AAA trial population by the
cognitive test status at the follow-up [figures are means
(SD) unless stated otherwise]
Data analysis
........................................
Statistical analysis was carried out using SPSS version 14
(SPSS, Inc., Chicago IL, USA) and all significance levels
reported are two-sided with P < 0.01 considered statistically
significant (the alpha level was lowered to maintain adequate control over type I error rate). The TMT was
skewed to the right and natural log of this variable was,
therefore, computed and used for all analyses. Age- and
sex-adjusted Pearson correlation coefficients were calculated
for the cognitive function variables with total cholesterol,
systolic and diastolic blood pressure. Multiple linear regression was used to model the relationship of cognitive variables with risk factor variables, where the covariates age,
systolic blood pressure, diastolic blood pressure, cholesterol,
smoking, gender, cholesterol medication, anti hypertensive
medication and deprivation were fitted together. The deprivation variable was the Scottish Index of Multiple
Deprivation (SIMD), an index used to classify the degree
of deprivation in a geographical region based on 38 different indicators including education [17]. The smoking status
was fitted as a binary variable (current versus previous/
never smoker). Models adjusting for prior cognitive ability
(MHVS) were used to analyse the association of the cardiovascular risk factor variables with estimated decline in cognitive function. The use of a regression-based methodology
(adjusting for vocabulary scores) in estimating cognitive
change has been validated previously [18]. Furthermore,
adjusting for estimated prior cognitive ability circumvents
confounding by regression-to-the-mean and practice effect.
In the regression models, cognitive function variables were
the dependent variables and the cardiovascular risk factors
were predictor variables.
Age, years
Sex, males [n (%)]
SIMD deprivation category [n (%)]
1 Most deprived
2
3
4
5 Least deprived
Ankle brachial index
Systolic blood pressure (mmHg)
Diastolic blood pressure (mmHg)
Total plasma cholesterol
(mmol/L)
Smoking status [n (%)]
Current
Previous
Never
On cholesterol-lowering medication [n (%)]
On anti-hypertensive medication [n (%)]
Characteristic
Cognitively
tested
(n = 2,312)
Not cognitively
tested
(n = 1,038)
61.7 (6.5)
623 (26.9)
62.6 (6.8)
331 (31.9)
565 (24.4)
490 (21.2)
354 (15.3)
347 (15.0)
556 (24.0)
0.86 (0.09)
147.2 (21.4)
83.8 (10.7)
6.2 (1.1)
323 (31.1)
261 (25.1)
151 (14.5)
136 (13.1)
167 (16.1)
0.85 (0.09)
149.4 (21.9)
83.4 (10.4)
6.2 (1.1)
665 (28.8)
779 (33.7)
868 (37.5)
95 (4.1)
603 (26.1)
420 (40.5)
327 (31.5)
291 (28.0)
46 (4.4)
284 (27.4)
variables (Table 2). There were small but statistically significant correlations of diastolic and systolic blood pressure
with estimated pre-morbid cognitive ability (MHVS), ‘g’
and all of the individual age-sensitive cognitive ability tests.
In all cases, the direction of the correlation indicated worse
cognitive ability associated with raised blood pressure.
Table SA1, Supplementary data available in Age and
Ageing online, shows the result of fitting regression models
in which all of the covariates were included simultaneously
(but without MHVS). Being a current smoker was significantly associated with worse performance on all the cognitive function tests except VFT. Higher diastolic blood
pressure was significantly associated with worse performance on AVLT (β = −0.153, P < 0.01).
Results
Description of cohort
Table 1 shows baseline characteristics of the AAA trial participants who were cognitively tested compared with those not
tested. Those not tested were slightly older, more likely to be
men, more socially deprived and more likely to be current
smokers, but overall differences were not large. 804 of the
2,312 subjects cognitively tested were 65 years and above,
while 319 were 70 years or older. For the 1,038 not tested, 428
were 65 years and above while 207 were 70 years or older.
Association of cardiovascular risk factors with
cognitive function
After adjustment for age and sex, the total cholesterol level
was not correlated with any of the cognitive function
308
Association of cardiovascular risk factors with
estimated cognitive decline
After adjusting for MHVS diastolic BP was significantly
associated with worse performance on AVLT (β = −0.125,
P < 0.01) (Table 3). As in the model unadjusted for MHVS,
the ‘current smoker’ status was negatively associated with
scores on all the cognitive function variables except VFT.
Discussion
We evaluated the effects of blood cholesterol, cholesterollowering medication, systolic and diastolic blood pressure,
anti-hypertensive medication and smoking on a broad
range of cognitive domains in men and women aged over
50 years from the general population. We found no
Cardiovascular risk factors and cognition
Table 2. Age- and sex-adjusted Pearson correlation coefficients between baseline cardiovascular risk factors and follow-up
cognitive test scores
MHVS
AVLT
RAVENS
DST
VFT
lnTMTa
g
0.022
−0.051*
−0.058*
0.016
−0.102**
−0.126**
0.007
−0.083**
−0.079**
−0.011
−0.072**
−0.057*
0.013
−0.065*
−0.083**
0.002
0.079**
0.083**
0.007
−0.097**
−0.103**
....................................................................................
Total cholesterol
Systolic blood pressure
Diastolic blood pressure
a
Timed test in which a greater score indicates poorer ability (for all other tests, greater score indicates better ability).
*P < 0.01.
**P < 0.001.
MHVS, Mill Hill Vocabulary Scale; AVLT, auditory verbal learning test; DST, digit symbol test; VFT, verbal fluency test; TMT, trail making test; ‘g’, measure of
general intelligence.
Table 3. Multivariate model with MHVS (unstandardised regression coefficients)
AVLT
RAVENS
DST
VFT
lnTMT
g
−0.587**
8.364**
−0.065*
−0.023
−0.125*
0.051
−2.265*
0.237
−1.361
−0.468**
−1.557**
−0.073**
−0.023
−0.009
−0.095
−1.605*
0.986
−0.544
−0.669**
4.518**
−0.104**
−0.025
0.013
−0.269
−2.672**
0.798
−1.044
−0.128**
0.523
−0.051*
−0.005
−0.051
−0.124
−0.303
0.392
−0.837
0.020**
−0.080**
0.003**
0.0005
0.001
0.004
0.095**
0.003
0.011
−0.056**
0.251**
−0.009**
−0.002
−0.004
−0.015
−0.223**
0.059
−0.085
....................................................................................
Covariates
Age
Gender
SIMD
Systolic BP
Diastolic BP
Total cholesterol
Smoking status (Current versus previous/never)
On cholesterol Medication (yes versus no)
On anti-hypertensive medication (yes versus no)
*P < 0.01.
**P < 0.001.
SIMD, Scottish Index of Multiple Deprivation; MHVS, Mill Hill Vocabulary Scale; AVLT, auditory verbal learning test; DST, digit symbol test; VFT, verbal fluency
test; TMT, trail making test; ‘g’, measure of general intelligence.
evidence of an association of blood cholesterol or
cholesterol-lowering medication with cognitive ability after
the 5-year follow-up. There was evidence of an association
between raised blood pressure and poorer cognition, in that
both higher diastolic and systolic blood pressures were correlated with poorer general cognitive ability and with each
of the individual cognitive tests, after controlling for the
effects of age and sex. After adjusting for all other
co-variates, including deprivation, and estimated prior cognitive ability, statistically significant negative association was
retained for elevated diastolic blood pressure and AVLT
only. However, the most consistent association was found
between cognition and smoking status, which persisted
across all measures of cognitive ability (except VFT) and
after multivariate adjustment including estimated prior cognitive ability, suggesting that smoking may be associated
with both cognitive impairment and with age-related cognitive decline.
The lack of evidence of an association between cholesterol and cognition observed in this study is consistent with
findings from a number of previous studies [19]. A systematic review and meta-analysis by Anstey et al. [20] indicated
that there is stronger evidence of an association between
cholesterol ( particularly in mid-life) and an increased risk of
dementia than there is between cholesterol and cognitive
impairment which falls short of frank dementia. Animal
models suggest that cholesterol may increase the production of amyloid precursor protein (APP) and subsequent
beta cleavage of APP to beta amyloid [21]. Overall, there is
currently no consensus on the relationship between serum
cholesterol and age-related cognitive decline. In our study,
cognition was not related to the use of cholesterol-lowering
medication, although the number of individuals taking
these medications was relatively small. A number of randomised controlled trials have also not observed any significant effect of cholesterol-lowering medication (statins) on
cognitive function in individuals without dementia [22]. A
Cochrane review [23] did not show any beneficial effect of
statins in the prevention of dementia and concluded that
‘statins should not be prescribed to prevent dementia by
health care professionals practicing evidence based medicine’. Furthermore, a randomised double blind placebo
controlled trial of low-dose aspirin in the AAA trial sample
revealed that aspirin did not affect cognition [5].
Our findings add to the growing body of evidence suggesting that elevated blood pressure in mid-life impacts
negatively on cognition and may predispose to the development of dementia [24]. High blood pressure is a recognised
risk factor for white matter (subcortical) damage which
results in impaired cognition, in particular reduced psychomotor speed, attention, working memory and executive
function [25]. However, knowledge of the precise
309
O. Okusaga et al.
mechanism by which elevated blood pressure may contribute to cognitive change over time is still evolving. Available
evidence indicates that hypertension can cause direct
damage to cerebral vascular endothelium, and alter cerebral
autoregulatory mechanisms [26] ultimately resulting in cerebral hypoperfusion—an important common pathway by
which vascular pathology is felt to impact negatively on
cognitive function [27].
The negative effect of smoking on cognition observed
in the current study is consistent with the results of a systematic review [28], suggesting that smoking may result in
poorer cognitive ability in middle to late age. However, it is
also worth considering whether poor cognitive ability in
early to mid-life could lead to a disadvantageous cardiovascular risk factor profile in individuals later in life, i.e. the
association between cognitive ability and altered risk factor
levels may be bidirectional. For example, individuals with
lower cognitive ability have been shown to have clustering
of cardiovascular disease risk factors and low IQ has been
associated with a raised mortality rate from cardiovascular
disease in middle-aged men and women [29]. Detailed data
on the cognitive function of the AAA trial population prior
to inclusion in the trial were not available and it was therefore not possible to fully control for the possibility that
smoking initiation in their earlier years may have been
mediated by cognitive ability. However, we were able to
adjust for estimated prior cognitive ability, following which
the association between smoking and cognitive ability persisted. Cigarette smoking, apart from acting synergistically
with systolic blood pressure to cause vascular endothelial
damage, has also been found to reduce neurogenesis and
increase death of brain cells [30].
Strengths of this study include the large number of participants and the selection of participants from the community who were therefore representative of individuals seen
routinely by health-care professionals. Also informative was
the use of a comprehensive battery of cognitive assessment
tools, and our ability to adjust for estimated prior cognitive
ability. The use of a detailed cognitive assessment ensured
that several cognitive domains were considered when studying associations with risk factors. This is in contrast to a
number of studies that have employed very basic cognitive
assessment tools such as the Mini-Mental State
Examination alone.
The study is limited in that cholesterol, blood pressure
and smoking status were measured only once, at baseline,
and smoking status and number of cigarettes smoked were
self-reported. Another limitation is the fact that the cognitive test battery was administered only at the 5-year followup and we estimated baseline ( pre-morbid) cognitive
function with the use of the MHVS. We also did not
control for apolipoprotein-E alleles/genotypes. In addition,
participants were recruited from general practice surgeries
in the Lanarkshire, Glasgow and Edinburgh districts of
Scotland and may therefore not be generalisable to other
areas and/or countries with different demographic characteristics. Ideally the potential causal association between
310
cognitive decline and the vascular risk factors studied here
should be tested in randomised controlled trials, as has
started to be done for hypercholesterolaemia and hypertension. However, such trials are resource intensive and observational studies such as the one described here are useful
to indicate which risk factors should be targeted for retention of cognitive function in specific domains.
In conclusion, we have demonstrated that both elevated
diastolic blood pressure and smoking are associated with
poorer cognitive function in individuals older than 50 years
and may also be associated with an age-related cognitive
decline. The outcome of this study supports the recommendation of good control of blood pressure and smoking
cessation to prevent an accelerated cognitive decline in
middle-aged individuals.
Key points
• Smoking and elevated blood pressure are associated with
poorer cognitive function.
• Cholesterol was not associated with cognitive function.
• Good control of blood pressure and smoking cessation
may prevent accelerated cognitive decline.
Conflicts of interest
None declared.
Funding
Funding for the Aspirin for Asymptomatic Atherosclerosis
(AAA) trial, including cognitive testing, was provided by
the Wellcome Trust, the British Heart Foundation and the
Chief Scientist Office, Scotland.
Supplementary data
Supplementary data mentioned in the text is available to
subscribers in Age and Ageing online.
References
1. Spooner MH. Britain grappling with dementia prevalence.
CMAJ 2009; 180: E13.
2. Bischkopf J, Busse A, Angermeyer MC. Mild cognitive impairment: a review of prevalence, incidence and outcome according to current approaches. Acta Psychiatr Scand 2002; 106:
403–14.
3. Nguyen M. Nurse’s assessment of caregiver burden. Medsurg
Nurs 2009; 18: 147–51.
4. Lithell H, Hansson L, Skoog I et al. The Study on Cognition
and Prognosis in the Elderly (SCOPE): principal results of a
randomized double-blind intervention trial. J Hypertens 2003;
21: 875–86.
Cardiovascular risk factors and cognition
5. Price JF, Stewart MC, Deary IJ et al. Low dose aspirin and
cognitive function in middle aged to elderly adults: randomised controlled trial. BMJ 2008; 337: a1198.
6. Cherbuin N, Reglade-Meslin C, Kumar R et al. Risk factors of
transition from normal cognition to mild cognitive disorder:
the PATH through Life Study. Dement Geriatr Cogn Disord
2009; 28: 47–55.
7. Solomon A, Kareholt I, Ngandu TBM et al. Serum cholesterol
changes after midlife and late-life cognition, twenty-one-year
follow-up study. Neurology 2007; 68: 751–6.
8. Daviglus ML, Bell CC, Berrettini W et al. National Institutes
of Health State-of-the-Science Conference statement: preventing Alzheimer disease and cognitive decline. Ann Intern Med
2010; 153: 176–81.
9. Emery CF, Huppert FA, Schein RL. Do pulmonary function
and smoking behavior predict cognitive function? Finding
from a British sample. Psychol Health 1997; 12: 265–75.
10. Benton D. Do low cholesterol levels slow mental processing?
Psychosom Med 1995; 57: 50–3.
11. Raven J, Raven JC, Court JH. Manual for Raven’s Progressive
Matrices and Vocabulary Scales. Oxford: Oxford
Psychologists Press, 1998.
12. Salthouse TA. Localizing age-related individual differences in
a hierarchical structure. Intelligence 2004; 32: 541–61.
13. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”.
A practical method for grading the cognitive state of patients
for the clinician. J Psychiatr Res 1975; 12: 189–98.
14. Lezak MD. Neuropsychological Assessment. New York:
Oxford University Press, 1995.
15. Wechsler D. Wechsler Adult Intelligence Scale—Revised.
New York: The Psychological Corporation, 1981.
16. Norusis MJ. Factor Analysis in SPSS 15.0 Statistical Procedures
Companion. New Jersey: Prentice Hall Inc, 2006, pp. 394–7.
17. Scottish Index of Multiple Deprivation. http://www.scotland.
gov.uk/Topics/Statistics/Browse/Social-Welfare/TrendSIMD.
(8 January 2012, date last accessed).
18. Deary IJ, Whalley LJ, Crawfor JR. An ‘instantaneous’ estimate
of a lifetime’s cognitive change. Intelligence 2004; 32: 113–9.
19. Van Exel E, de Craen JMA, Gusesekloo J et al. Association
between high density lipoprotein and cognitive impairment in
the oldest old. Ann Neurol 2002; 51: 716–21.
20. Anstey KJ, Lipnicki DM, Low L-F. Cholesterol as a risk
factor for dementia and cognitive decline: a systematic review
of prospective studies with meta-analysis. Am J Geriatr
Psychiatry. 2008; 16: 343–54.
21. Simons M, Keller P, De Strooper B et al. Cholesterol depletion inhibits the generation of beta-amyloid in hippocampal
neurons. Proc Natl Acad Sci USA 1998; 95: 6460–4.
22. Summers MJ, Oliver KR, Coombes JS, Fassett RG. Effect of
atorvastatin on cognitive function in patients from the Lipid
Lowering and Onset of Renal Disease (LORD) trial.
Pharmacotherapy 2007; 27: 183–90.
23. McGuinness B, Craig D, Bullock R, Passmore P. Statins for
the prevention of dementia. Cochrane Database of Syst Rev
2009; 2: CD003160.
24. Qiu C, Winblad B, Fratiglioni L. The age-dependent relation
of blood pressure to cognitive function and dementia. Lancet
Neurol 2005; 4: 487–99.
25. Hannesdottir K, Nitkunan A, Charlton RA, Barrick TR,
MacGregor GA, Markus HS. Cognitive impairment and
white matter damage in hypertension: a pilot study. Acta
Neurol Scand 2009; 119: 261–8.
26. Manolio TA, Olson J, Longstreth WT. Hypertension and
cognitive function: pathophysiologic effects of hypertension
on the brain. Curr Hypertens Rep 2003; 5: 255–61.
27. Obisesan TO. Hypertension and cognitive function. Clin
Geriatr Med 2009; 25: 259–88.
28. Peters R, Poulter R, Warner J, Beckett N, Burch L, Bulpitt C.
Smoking, dementia and cognitive decline in the elderly, a systematic review. BMC Geriatr 2008; 8: 36.
29. Batty GD, Deary IJ, Benzeval M, Der G. Does IQ predict
cardiovascular disease mortality as strongly as established risk
factors? Comparison of effect estimates using the West of
Scotland ‘Twenty-07’ cohort study. Eur J Cardiovasc Prev
Rehabil 2010; 17: 24–7.
30. Abrous DN, Adriani W, Montaron M-F et al. Nicotine selfadministration impairs hippocampal plasticity. J Neurosci
2002; 22: 3656–62.
Received 15 January 2012; accepted in revised form
16 November 2012
311