Article

Clinical predictors of cognitive decline in patients with mild cognitive impairment: The Chongqing aging study

Department of Neurology, Third Military Medical University, Daping, Chongqing, China.
Journal of Neurology (Impact Factor: 3.38). 12/2011; 259(7):1303-11. DOI: 10.1007/s00415-011-6342-0
Source: PubMed

ABSTRACT

Mild cognitive impairment (MCI) is considered as the early stage of dementia which currently has no effective treatments. Reducing progression of cognitive decline at the MCI stage could be an important strategy for preventing conversion to dementia. The goal of this work was to screen for clinical predictors indicating the prognosis of MCI comprehensively; therefore, we assumed vascular risk factors (VRFs), carotid stenosis, and white matter changes (WMC) to be independent predictors. A total of 257 patients with MCI underwent collection of VRF information, neuropsychological evaluation, computed tomography angiography (CTA) to investigate carotid stenosis, and magnetic resonance imaging (MRI) to identify severity of WMC. After a 3-year follow-up period, the neuropsychological evaluation, CTA, and MRI were repeated to assess the progression of cognitive decline, carotid stenosis, and WMC. The conversion rate from MCI to dementia was 11.65% per year, and the conversion rate from MCI to Alzheimer's disease was 7.05% per year in our cohort. Cognitive decline (in terms of changes in Mini Mental State Examination scores) was associated with diabetes mellitus (p = 0.004), baseline WMC severity (p < 0.001), baseline carotid stenosis (p < 0.001), and WMC severity change (p < 0.001). Besides, diabetes, baseline WMC severity, baseline moderate-to-severe carotid stenosis, and carotid stenosis change during follow-up were predictors of conversion from MCI to dementia. Given the potential clinical predictors, our findings could imply that controlling blood glucose, removing carotid stenosis, and improving cerebral perfusion could be effective measures to delay cognitive decline in patients with MCI and prevent conversion from MCI to dementia.

Full-text

Available from: Xu Yi, Nov 07, 2015
ORIGINAL COMMUNICATION
Clinical predictors of cognitive decline in patients with mild
cognitive impairment: the Chongqing aging study
Ling Li
Yanjiang Wang
Jiachuan Yan
Yang Chen
Rui Zhou
Xu Yi
Qianqian Shi
Huadong Zhou
The Chongqing Aging Study Group
Received: 27 October 2011 / Revised: 18 November 2011 / Accepted: 24 November 2011 / Published online: 21 December 2011
Ó Springer-Verlag 2011
Abstract Mild cognitive impairment (MCI) is considered
as the early stage of dementia which currently has no
effective treatments. Reducing progression of cognitive
decline at the MCI stage could be an important strategy for
preventing conversion to dementia. The goal of this work
was to screen for clinical predictors indicating the prog-
nosis of MCI comprehensively; therefor, we assumed
vascular risk factors (VRFs), carotid stenosis, and white
matter changes (WMC) to be independent predictors.
A total of 257 patients with MCI underwent collection of
VRF information, neuropsychological evaluation, com-
puted tomography angiography (CTA) to investigate car-
otid stenosis, and magnetic resonance imaging (MRI) to
identify severity of WMC. After a 3-year follow-up period,
the neuropsychological evaluation, CTA, and MRI were
repeated to assess the progression of cognitive decline,
carotid stenosis, and WMC. The conversion rate from MCI
to dementia was 11.65% per year, and the conversion rate
from MCI to Alzheimer’s disease was 7.05% per year in
our cohort. Cognitive decline (in terms of changes in Mini
Mental State Examination scores) was associated with
diabetes mellitus (p = 0.004), baseline WMC severity
(p \ 0.001), baseline carotid stenosis (p \ 0.001), and
WMC severity change (p \ 0.001). Besides, diabetes,
baseline WMC severity, baseline moderate-to-severe car-
otid stenosis, and carotid stenosis change during follow-up
were predictors of conversion from MCI to dementia.
Given the potential clinical predictors, our findings could
imply that controlling blood glucose, removing carotid
stenosis, and improving cerebral perfusion could be
effective measures to delay cognitive decline in patients
with MCI and prevent conversion from MCI to dementia.
Keywords Mild cognitive impairment (MCI)
Dementia Vascular risk factors (VRFs)
Carotid stenosis White matter changes (WMC)
Introduction
Mild cognitive impairment (MCI) refers to an intermediate
state of cognitive decline between the changes observed in
aging and those fulfilling the criteria for dementia and often
Alzheimer’s disease (AD) [1]. More than half of patients
with MCI progress to dementia within 5 years [2]. Thus,
MCI can be regarded as an increased risk for progression to
dementia, and its identification could lead to secondary
prevention of dementia by controlling related somatic
factors. Vascular risk factors (VRFs), such as hypertension
[3, 4], diabetes [5, 6], hyperlipidemia [7, 8], and stroke
[9, 10], play critical roles in the development of cognitive
decline and AD among the treatable somatic factors [11].
However, there is still no certainty as to whether VRFs are
simply additive elements compounding cognitive decline
or whether they play a causal role by directly affecting
cognitive impairment [12]. Most epidemiological studies
have encountered difficulties in precisely gauging the role
of VRF severity, and recent research has focused on
identifying practical parameters which may precisely
predict cognitive impairment. Indeed, chronic cerebral
hypoperfusion (CCH) as a result of VRFs is a common
vascular component among cognitive impairment risk factors
L. Li Y. Wang J. Yan Y. Chen R. Zhou
X. Yi Q. Shi H. Zhou (&)
Department of Neurology and Center for Clinical Neuroscience,
Daping Hospital, Third Military Medical University,
No. 10 Changjiang Branch Road, Daping,
Chongqing 400042, China
e-mail: zhouhuad@163.com
123
J Neurol (2012) 259:1303–1311
DOI 10.1007/s00415-011-6342-0
Page 1
[13, 14]. Most studies investigating the possibility that
CCH may predispose to reduced cognitive function have
focused on carotid steno-occlusive disease [15, 16] and
white matter changes (WMC) [1719] in order to deter-
mine possible practical quantitative predictors of cognitive
decline. However, there are relatively few reports which
have assumed VRFs, carotid stenosis, and WMC to be the
independent risk factors of cognitive decline in patients
with MCI and which have explored the correlation between
above factors and cognitive decline.
In this prospective study, our aim was to ascertain
whether VRFs could be used to predict cognitive decline in
an independent elderly cohort with MCI, taking into
account carotid stenosis and WMC as two independent risk
factors. We also aimed to identify the risk factors pro-
moting conversion from MCI to dementia.
Methods
Study subjects
A total of 257 subjects were selected from inpatients in the
Department of Neurology of Daping Hospital in the city of
Chongqing during March–September 2008. Eligibility
requirements included subjects (1) who were 60 years and
older, (2) who were long-term residents of these commu-
nities, and (3) who were diagnosed as MCI. Exclusion
criteria were (1) a diagnosis of dementia, or Hachinski
Ischemic Score (HIS) B4 and HIS C7, (2) a concomitant
neurologic disorder potentially affecting cognitive function
(e.g., severe Parkinson’s disease), (3) a history of stroke,
(4) the degree of other cerebral arteriostenosis is greater
than carotid stenosis, (5) being unable to comply with the
study assessment, (6) enduring mental illness, or the score
of Hamilton Depression Rating Scale (HDRS) [17, (7)
drug abuse, (8) moving away or declining to participate.
This study was approved by the Institutional Review Board
of the Third Military Medical University, and all subjects
and their caregivers provided informed consent.
Baseline data
These data included demographic data and VRFs.
Demographic data comprised age, sex, and educational
level (lower educational level refers to the education time
B6 years; higher educational level refers to the education
time [6 years).
VRFs including hypertension, diabetes, hyperlipidemia,
previous transient ischemic attack, and treatment were
collected from a structured clinical interview (with care-
givers’ involvement), physical examination, and blood
tests. Hypertension was defined as systolic blood pressure
C140 mm Hg and/or diastolic blood pressure C90 mm Hg
[20]. Diabetes was a concentration of fasting plasma glu-
cose C7.0 mmol/l (126 mg/dl) [21], and hyperlipidemia
was a total cholesterol concentration C5.2 mmol/l
(200 mg/dl) and a low-density lipoprotein cholesterol
concentration C3.4 mmol/l (130 mg/dl) [22]. These values
were confirmed by repeated determinations before a
definitive diagnosis was made. In addition, VRFs included
smoking and drinking status. The smoking status was
classified as follows: past smokers who had quit smoking
for at least 6 months, current smokers, or nonsmokers. The
drinking status was classified as drinking daily, weekly,
monthly, occasionally, or never drinking previously as
defined at the time the subjects were enrolled [23].
The treatment of VRFs included the following: use of
diuretic, angiotensin-converting enzyme inhibitor, angio-
tensin receptor blocker, b-blocker, calcium channel
blocker, or other antihypertensive medication for hyper-
tension; oral antihyperglycemic or insulin for diabetes;
statins for hyperlipidemia; smoking or drinking were con-
sidered ‘treated’’ if the patient ceased smoking or drinking
during the follow-up period.
Neuropsychological evaluation
The cognitive and functional status was assessed using the
Chinese version of the MMSE, Clinical Dementia Rating
(CDR), and the Barthel Index of Activities of Daily Living
(Barthel ADL Index), which had been previously validated
in Chinese elderly people [23,
24]. The subjects with
abnormal MMSE score were administered HDRS for
measuring emotional status [25], and HIS for evaluating
significant vascular disease [26]. Subsequently, a set of
neuropsychological tests were applied, including Fuld
Object Memory Evaluation (FOM) for detecting extensive
cognitive dysfunction mainly composed of memory [27],
Rapid Verbal Retrieve (RVR) for detecting the function of
semantic memory [28], Wechsler Adult Intelligence Scale
(WAIS) for evaluating immediate memory and function of
graphical recognition [29], and the Pfeiffer Outpatient
Disability Questionnaire (POD) for assessing ability of
social activities [30]. We chose the MMSE to assess cog-
nitive decline in this study because of its larger clinical
application and high correlation with other indices such as
ADAS-Cog which has been reported by Silvestrini et al.
[31].
Diagnosis of MCI
The clinical diagnosis of MCI was made according to the
established Petersen criteria [32], including (1) subjective
complaint of memory deficits, (2) abnormal memory
functioning for age (tests claim 1.5 SD below normative
1304 J Neurol (2012) 259:1303–1311
123
Page 2
values), (3) absence of dementia according to the diag-
nostic examination (MMSE C24 in subjects with higher
educational level; MMSE C20 in subjects with lower
educational level; CDR B0.5), (4) and normal everyday
functioning on ADL (ADL [60). The subjects with a
depressive disorder were excluded [33].
Diagnosis of dementia
Dementia was diagnosed based on criteria according to the
Diagnostic and Statistical Manual of Mental Disorders, 4th
edition (DSM-IV) [34]. Alzheimer’s disease (AD), vascu-
lar dementia (VaD), and mixed dementia (MD) were all
included. The diagnosis of AD was made according to the
criteria for probable AD published by the National Institute
of Neurological and Communicative Disorders and Stroke-
Alzheimer’s disease and Related Disorders Association
Work Group (NINCDS-ADRDA) [35]; the diagnosis of
VaD was based on the criteria about probable vascular
dementia published by the National Institute of Neurolog-
ical Disorders and Stroke-Association Internationale pour
la Recherche et l’Enseignement en Neurosciences
(NINDS-AIREN) [36]; MD refers to Alzheimer’s disease
with cerebrovascular disease (CVD) which was diagnosed
when the clinical picture of the subject presented aspects of
both AD and VaD [37].
Diagnosis of cerebral arteriostenosis
The cerebral arteries were evaluated by means of CTA
(Light Speed VCT 64-slice Scanner, General Electric,
Milwaukee, WI), and the degree of cerebral arteriostenosis
was assessed on an advanced workstation (Advantage 4.2,
General Electric, Milwaukee, WI); the data were recorded,
based on rounding (e.g., 0, 10, 20%). The final degree of
carotid stenosis for each subject was defined as the maxi-
mum value of all sites of the common carotid arteries,
carotid bulbs, and internal carotid arteries. The severity of
carotid stenosis was grouped as mild (0–29%), moderate
(30–69%), or severe (C70%), according to the NASCET
method [38].
Diagnosis of WMC
MRI was performed using a 3.0 T magnet (MAGNETOM
Verio 3.0 T, Siemens) with T1- and T2-weighted and fluid-
attenuated inversion recovery sequences (FLAIR) at entry
and at the end of the study. The degree of WMC severity
was rated on FLAIR by the practitioners in the department
of radiology who were blind to the clinical data, using the
three classes in the revised version of the visual scale of
Fazekas and colleagues [39]. Taking into account only
deep and subcortical white matter, lesions were classified
into three categories: mild WMC (single lesions below
10 mm; areas of grouped lesions smaller than 20 mm in
any diameter); moderate WMC (single lesions between 10
and 20 mm; areas of grouped lesions more than 20 mm in
any diameter; no more than connecting bridges between
individual lesions); or severe WMC (single lesions or
confluent areas of hyperintensity 20 mm or more in any
diameter) [40].
Follow-up
A total of 257 patients who were enrolled into the present
study accepted follow-up for 3 years from 2008–2011.
Demographic data and VRFs were collected at baseline.
The same neuropsychological tests were administered at
entry and at the end of follow-up by a neuropsychologist
blinded to the medical records of the subjects. All subjects
were examined by CTA and MRI at entry and at the end of
follow-up. The type of CT machine, the software used to
assess the degree of cerebral arteriostenosis, and the
practitioners who assessed the WMC severity and were
blinded to the cognitive status were kept the same at entry
and at the end of follow-up. In addition, phone interviews
were performed at 3-month intervals to obtain information
concerning cognition status and check compliance.
Statistical analysis
A sample size calculation was performed briefly because
this was an explorative study of possible risk factors for the
progression of cognitive decline. According to baseline
data, WMC severity, and carotid stenosis, the potential risk
factors were less than 20; thus, 200 cases (20 9 10) would
be the recommended sample size for a reliable analysis.
Considering drop-outs, our final sample size was 257.
Because this study aimed to explore clinical predictors for
progression of cognitive decline (measured by a MMSE
score decrease, a continuous variable, as dependent vari-
able) related to WMC severity, carotid stenosis, and VRFs,
our goal was best served by a linear regression model.
Simple regression analyses were performed first to evaluate
the bivariate association between MMSE decrease (sub-
jects with stroke during follow-up were excluded for the
confounding effect) and each potential predictor. Then the
independent variables which had significant linear rele-
vance with the MMSE decrease were entered into a mul-
tiple regression analysis; thus, some potential confounding
effect of risk factors would be removed and a regression
formula describing the MMSE decrease can be worked out.
For those who converted to dementia (subjects with stroke
during follow-up were included), we analyzed the differ-
ence between the dementia and subjects remaining in
MCI by the
t test for independent normally distributed
J Neurol (2012) 259:1303–1311 1305
123
Page 3
continuous data and the v
2
test for categorical data. Then,
the independent variables that were significantly different
in the two groups were used to analyze the association for
conversion to dementia by means of the Cox proportional
hazard regression model. The statistical analyses were
performed using SPSS 18.0 for Windows.
Results
A total of 257 subjects with MCI were enrolled into the
study at baseline (mean age 70.05 years, SD 6.78; 43.19%
women, 40.15% subjects with lower education level), 246
(95.72%) subjects completed the follow-up process, and
only 11 (4.28%) subjects dropped out, 8 died and 3
declined. For these 11 subjects, no cognitive diagnosis was
attributed. During follow-up, 36 subjects experienced a
stroke, and these 36 subjects were excluded from linear
regression. Considering the cognitive diagnosis performed
at the last clinical visit, dementia was diagnosed in 86
(34.96%) subjects (among those who progressed to
dementia, 60.47% progressed to AD, 30.23% to VaD, and
9.30% to MD). The conversion rate from MCI to dementia
was 11.65% per year, and the conversion rate from MCI to
AD was 7.05% per year which are similar to the values we
previously reported and other investigations [32, 41]. The
MMSE score decrease in subjects diagnosed with MD
(9.00 ± 1.69) was larger than VaD (5.92 ± 2.45), and the
smallest in subjects diagnosed with AD (5.58 ± 1.98).
Baseline characteristics
The baseline characteristics, including demographic data,
VRFs and related treatments, WMC severity, and degree of
carotid stenosis in 257 subjects, are presented in Table 1.
MMSE scores exhibited a narrow baseline variability
(coefficient of variation 6.16%), indicating that the sample
had fairly homogeneous cognitive impairment. In this
study, hypertension and previous transient ischemic attack
were common; diabetes mellitus and hyperlipidemia were
poorly treated. Mild WMC and moderate carotid stenosis
were identified in a large proportion of patients.
Effect of demographic characteristics, VRFs,
WMC severity, and carotid stenosis on the 3-year
decrease in MMSE
According to simple regression analysis (Table 2), the
decrease of MMSE scores over the 3-year study period was
mainly attributed to gender, diabetes mellitus, alcohol
consumption, WMC severity, and degree of carotid ste-
nosis at baseline and deterioration of WMC severity and
carotid stenosis. Gender was related to the decreasing
MMSE scores in the simple regression analysis because
most of the male subjects reported smoking or drinking
habits. We examined gender as a possible confounder and
excluded it.
In addition, the decrease of the MMSE scores had to be
adjusted for baseline cognitive status as well as age [23].
The adjusted decrease of the MMSE scores as a dependent
variable and the above 6 risk factors as independent vari-
ables were entered into the main regression model (step-
wise regression), which indicated that the MMSE score
decrease was significantly related to diabetes mellitus
(p = 0.004), baseline WMC severity (p \ 0.001), baseline
carotid stenosis (p \ 0.001), and WMC severity change
(p \ 0.001) (Table 2, right side).
The MMSE score decrease can be described by the
following formula: MMSE score decrease =-1.180 ?
1.611 (baseline WMC severity) ? 5.805 (baseline carotid
stenosis) ? 1.221 (WMC severity change) ? 0.957 (1 if
with diabetes mellitus; 0 if without) (F = 51.583;
p \ 0.001; R
2
= 0.674).
Table 1 Baseline patient characteristics (n = 257)
Characteristics Values
Demographic data
Age, years, mean ± SD 70.05 ± 6.78
Female, n (%) 111 (43.19)
Lower education level (B6 years), n (%) 103 (40.07)
VRFs of the study cohort, n (%)
Hypertension 161 (62.64)
Diabetes mellitus 76 (29.57)
Hyperlipidemia 110 (42.80)
Previous transient ischemic attack 165 (64.20)
Antihypertensives 106 (41.24)
Oral antihyperglycemic or insulin 32 (12.45)
Statins 34 (13.23)
Daily drinking 54 (21.01)
Current smoking 55 (21.40)
Alcohol withdrawal 24 (9.34)
Smoking cessation 18 (7.00)
Cognition assessment
MMSE score, mean ± SD 25.17 ± 1.55
WMC severity, n (%)
None 44 (17.12)
Mild 110 (42.80)
Moderate 67 (26.07)
Severe 36 (14.01)
Carotid stenosis, N (%)
Mild 113 (43.97)
Moderate 132 (51.36)
Severe 12 (4.67)
1306 J Neurol (2012) 259:1303–1311
123
Page 4
Cox proportional hazards model with dementia at last
clinical evaluation as a dependent variable
Using Cox proportional hazards model (Table 3; indepen-
dent variables in Cox proportional model differed signifi-
cantly between the subjects developing dementia and
subjects remaining in MCI by means of a t test or v
2
test;
data not shown, available on request), it was found that
diabetes, baseline WMC severity (severe), baseline carotid
stenosis (moderate to severe), and carotid stenosis change
during follow-up were predictors of conversion from MCI
to dementia.
Cox proportional hazards model with AD at last clinical
evaluation as a dependent variable
Using Cox proportional hazards model (Table 4), inde-
pendent variables in this Cox proportional model did not
include stroke during follow-up or severe carotid stenosis
because the two conditions did not exist among the subjects
who converted to AD. Finally, it was found that the pre-
dictors of conversion from MCI to AD were same as the
result among subjects diagnosed as dementia except for
carotid stenosis change during follow-up.
Discussion
AD is one of the most disabling and burdensome health
conditions worldwide, but no effective treatment is available
for the 4.6 million new patients who will be affected by AD
this year [42]. Thus, it is of great importance to recognize and
treat patients with MCI because this is an early stage of
dementia [43] and is associated with an increased risk for
progression to AD (10–15% per year), which is 10 times that
in the normal population [32]. Most studies about MCI have
pursued two goals: (1) to find the predictors which can
indicate the prognosis of MCI and then recognize the patients
with poor prognosis; (2) to identify the critical risk factors
promoting the conversion from MCI to dementia. Thus, the
identification of those patients with poor prognosis and
subsequent management of critical risk factors at the MCI
stage could be an important strategy for preventing and
delaying progression to dementia.
The clinical predictors studies mainly focused people’s
attention on carotid atherosclerosis (carotid stenosis, pla-
que and intima-media thickness), WMC, and VRFs, such
as hypertension and diabetes. In addition, VRFs have
become the target of treatable risk factors [44, 45]. In the
present prospective 3-year study, we enrolled 257 subjects
Table 2 Regression of 3 years decrease in MMSE on demographic characteristics, VRFs, WMC severity and carotid stenosis (n = 210)
Simple linear regression Multiple linear regression
B SE tpvalue B SE tpvalue
Gender (male vs. female) 0.728* 0.238* 3.057* 0.003*
Age (years) 0.058 0.036 1.609 0.111
Education level (years) 0.093 0.070 1.330 0.186
Hypertension 0.675 0.499 1.353 0.179
Antihypertensives -0.360 0.498 -0.722 0.472
Diabetes mellitus 1.679* 0.526* 3.192* 0.002* 0.957* 0.326* 2.934* 0.004*
Oral antihyperglycemic or insulin 1.332 0.829 1.606 0.111
Hyperlipidemia 0.462 0.497 0.929 0.355
Statins -0.811 0.744 -1.090 0.278
Previous TIA 0.356 0.504 0.707 0.481
Alcohol consumption 1.297* 0.627* 2.068* 0.041*
Alcohol withdrawal 0.709 0.926 0.765 0.446
Smoking 0.655 0.613 1.068 0.288
Smoking cessation -1.020 0.983 -1.038 0.302
Baseline MMSE score -0.186 0.161 -1.146 0.254
Baseline WMC severity 1.896* 0.207* 9.181* 0.000* 1.611* 0.182* 8.827* 0.000*
Baseline carotid stenosis 10.977* 1.479* 7.421* 0.000* 5.805* 1.213* 4.786* 0.000*
WMC severity change 1.141* 0.480* 2.377* 0.019* 1.221* 0.310* 3.933* 0.000*
Carotid stenosis change 6.310* 2.766* 2.282* 0.025*
B = slope, SE = standard error of slope, t = B/SE
* With significant (p \ 0.05) effect on MMSE change
J Neurol (2012) 259:1303–1311 1307
123
Page 5
with MCI in Chongqing, the biggest municipality in
southwest China. With the purpose of using advanced
imaging technology and determining the practical clinical
predictors comprehensively, we assumed VRFs, carotid
stenosis, and WMC to be the independent risk factors of
cognitive decline in patients with MCI in this study.
Our study showed that (1) in the present study, the
conversion rate from MCI to dementia was 11.65% per
year, and the conversion rate from MCI to AD was 7.05%
per year. (2) After being adjusted for baseline cognitive
status and age, cognitive decline of the patients with MCI
during the 3-year follow-up were associated with diabetes,
baseline WMC severity, baseline carotid stenosis, and
WMC severity change; (3) among the subjects diagnosed
as dementia at the end of follow-up, diabetes, baseline
severe WMC, baseline moderate-to-severe carotid stenosis,
and carotid stenosis change during follow-up were pre-
dictors of conversion from MCI to dementia; (4) among the
subjects diagnosed as AD at the end of follow-up, predic-
tors of conversion from MCI to AD were the same as those
for subjects diagnosed with dementia except for carotid
stenosis change during follow-up.
Table 3 Cox proportional hazards model, dependent variable: dementia at last clinical evaluation (n = 86)
b Hazard ratio p 95% Confidence interval
Age 0.013 1.013 0.595 0.966–1.063
Education level -0.727 0.484 0.083 0.212–1.101
Hypertension -0.350 0.705 0.550 0.224–2.221
Diabetes mellitus 0.870* 2.387* 0.034* 1.069–5.333*
Antihypertensives 0.199 1.220 0.679 0.476–3.130
Oral antihyperglycemic or insulin 0.730 2.075 0.213 0.658–6.544
Baseline WMC severity 0.001*
WMC severity (none vs. severe) -15.019 0 0.935 0–3.0471E150
WMC severity (mild vs. severe) -2.810* 0.060* 0.000* 0.018–0.202*
WMC severity (moderate vs. severe) -0.796* 0.317* 0.047* 0.113–1.035*
Baseline carotid stenosis 0.004*
Carotid stenosis (moderate vs. mild) 1.433* 4.190* 0.006* 1.517–11.572*
Carotid stenosis (severe vs. mild) 2.455* 11.641* 0.002* 2.496–54.292*
WMC severity change 0.622 1.863 0.115 0.859–4.040
Carotid stenosis change 5.069* 159.055* 0.005* 4.568–5,537.669*
Stroke during follow-up 0.614 1.848 0.110 0.871–3.922
* With significant (p \ 0.05) effect on conversion from MCI to dementia
Table 4 Cox proportional hazards model, dependent variable: AD at last clinical evaluation (n = 52)
b Hazard ratio p 95% Confidence interval
Age 0.49 1.050 0.133 0.985–1.119
Education level -0.046 0.955 0.936 0.307–2.966
Hypertension -0.291 0.747 0.723 0.150–3.733
Diabetes mellitus 1.072* 2.921* 0.028* 1.123–7.595*
Antihypertensives -0.168 0.846 0.813 0.210–3.402
Oral antihyperglycemic or insulin 1.029 2.798 0.173 0.636–12.307
Baseline WMC severity 0.002*
WMC severity (none vs. severe) -14.496 0 0.962 0–1.894E253
WMC severity (mild vs. severe) -3.280* 0.038* 0.001* 0.006–0.242*
WMC severity (moderate vs. severe) -0.680 0.506 0.349 0.122–2.101
Baseline carotid stenosis 0.011*
Carotid stenosis (moderate vs. mild) 2.135* 8.458* 0.003* 2.096–34.138*
WMC severity change 0.561 1.753 0.272 0.645–4.767
Carotid stenosis change 4.821 124.087 0.052 0.950–16,209.684
* With significant (p \ 0.05) effect on conversion from MCI to AD
1308 J Neurol (2012) 259:1303–1311
123
Page 6
The predictors for MMSE score decrease, conversion
from MCI to dementia, and conversion from MCI to AD
were different in our study. However, diabetes, baseline
WMC severity, and carotid stenosis were still identified as
common predictors for every outcome of cognitive decline
and all predictors for progression to AD. In addition to
these common predictors, WMC severity change during
follow-up was a predictor for the MMSE score decease,
while carotid stenosis change during follow-up predicted
progression to dementia. The differential results would be
attributed to different populations, because subjects having
a stroke during follow-up were excluded for the con-
founding effect when we analyzed the association factors
of MMSE score decrease, while all subjects were included
in Cox proportional hazards model with dementia and AD.
We did not analyze predictors of progression to VaD and
MD due to the limited number of subjects, but we could
still infer a conclusion from the above results, i.e., carotid
stenosis change during follow-up would mainly predict
progression to VaD in addition to the common predictors
of cognitive decline and WMC severity change during
follow-up.
We previously found that hypertension, diabetes, and
hypercholesterolemia were associated with a higher risk of
incident AD dementia [41]. In this study, taking into
account carotid stenosis and WMC in addition to VRFs, we
found that diabetes was the only independent predictor of
cognitive decline among all VRFs studied. The same result
was reported in the LADIS study [40]. The Chinese people
have paid more attention to hypertension during the recent
3 years, which can account for the discrepancy in our two
studies. Controversial results have been published on the
relationship between hypertension and dementia. Some
studies suggested a higher incidence of dementia in
patients with hypertension [4], but results of trials using
antihypertensive medication have been inconclusive
[40, 44]. Nevertheless, hypertension was not a predictor of
cognitive decline during follow-up in our cohort, while
diabetes was the only independent predictor of conversion
from MCI to dementia among all VRFs studied.
Extracranial carotid Doppler ultrasound is commonly
used to study carotid artery disease [46] due to its conve-
nience and low price. However, it has limitations in eval-
uating the cerebral arteries more comprehensively; in fact,
steno-occlusive disease of other cerebral arteries may be
the confounding factor for the analysis of an association
between carotid stenosis and cognitive decline. Therefore,
we assessed cerebral arteries using CTA in order to exclude
subjects with a higher degree of other cerebral arterioste-
nosis than carotid stenosis. After controlling for con-
founding factors, we found that moderate-to-severe carotid
stenosis and deterioration of carotid stenosis are predictors
of conversion from MCI to dementia.
WMC is considered as a manifestation of cerebral small
vessel disease, which is one of the common causes of
dementia [47]. Verdelho et al. [40] found that severe WMC
at baseline was a predictor for cognitive decline in the
LADIS sample40. A recent study reported that while pro-
gression of WMC was associated with progression for
cognitive impairment and dementia, baseline, WMC mea-
sure was not [48]. In our findings, both baseline WMC
severity and progression of WMC were associated with
cognitive decline in patients with MCI, but just baseline
WMC severity was a predictor of conversion from MCI to
dementia.
The mechanism underpinning the association of VRFs,
CVD, and cognitive impairment may mainly involve CCH
[49, 50]. Various mechanisms of neuronal injury in CCH,
including formation of free radicals, oxidative stress,
mitochondrial dysfunction, inflammatory processes and
apoptosis, have been proposed [51, 52]. These factors may
interact and amplify each other, finally leading to clinical
presentation of cognitive impairment. Cumulative animal
experiment results have provided the evidence that altered
Ab metabolism appears to be a central initiating factor for
cognitive impairment in animal models with CCH. Ab
peptides are generated from amyloid precursor protein
(APP) by sequential actions of two proteolytic enzymes,
the b-site APP cleavage enzyme 1 (BACE1) and the
c-secretase [53]. APP and BACE1 are both up-regulated in
chronic cerebral hypoperfusion, and then the elevation of
BACE1 contributing to APP processing and the increased
APP lead to a high level of Ab formation and eventual
deposition [
54]. Furthermore, the failure of Ab clearance is
a possible cause of Ab deposition and amyloid plaque
formation in CCH. Meanwhile, the formation of Ab and
accumulation in the vasculature may act to perpetuate
further vascular dysfunction and accelerate white matter
pathology [14], thus, forming a vicious cycle.
Our study is an observational study with inherent sub-
stantial limitations. We did not divide cognitive impair-
ment or dementia into vascular and non-vascular type,
because the focus of this study was cognitive decline.
Carotid stenosis and WMC were selected to be candidates
in our study because their assessments are more feasible
than other types of cerebral atherosclerosis. With the aim
of reducing bias induced by the impact of other cerebral
atherosclerosis, we excluded subjects with higher degrees
of other cerebral arteriostenosis than carotid stenosis.
However, atherosclerosis is considered a systemic disease
and we can not simplify the situation of the patient accu-
rately. In addition, the propensity score and intention to
treat analysis strategy used in our previous study were
adopted in the present research, in order to reduce bias
induced by changes in the VRFs and their treatment during
follow-up [41]. Our findings indicate that carotid stenosis
J Neurol (2012) 259:1303–1311 1309
123
Page 7
change during follow-up is a predictor of conversion from
MCI to dementia, but the threshold of carotid stenosis
change still needs further research to be identified. Nev-
ertheless, our study has implications for both clinical
practice and research, emphasizing the need to identify
patients with diabetes, severe WMC or moderate to severe
carotid stenosis, but also the need to treat VRFs and carotid
stenosis in order to prevent cognitive decline in patients
with MCI.
Conflicts of interest None.
References
1. Petersen RC (2011) Clinical practice. Mild cognitive impairment.
N Engl J Med 364:2227–2234
2. Gauthier S, Reisberg B, Zaudig M et al (2006) Mild cognitive
impairment. Lancet 367:1262–1270
3. Ruitenberg A, Skoog I, Ott A et al (2001) Blood pressure and risk
of dementia: results from the Rotterdam study and the Gothen-
burg H-70 study. Dement Geriatr Cogn Disord 12:33–39
4. Tzourio C, Dufouil C, Ducimetie
`
re P, Alpe
´
rovitch A (1999)
Cognitive decline in individuals with high blood pressure: a
longitudinal study in the elderly. EVA Study Group. Epidemi-
ology of vascular aging. Neurology 53:1948–1952
5. Xu WL, Qiu CX, Wahlin A, Winblad B, Fratiglioni L (2004)
Diabetes mellitus and risk of dementia in the Kungsholmen
project: a 6-year follow-up study. Neurology 63:1181–1186
6. Biessels GJ, Staekenborg S, Brunner E, Brayne C, Scheltens P
(2006) Risk of dementia in diabetes mellitus: a systematic review.
Lancet Neurol 5:64–74
7. Allouche A, Royer L, Escanye MC et al (2011) Dyslipidemia is a
modifiable risk factor for cardiovascular and alzheimer’s dis-
eases. Alzheimers Dement 7(Suppl 1):607–608
8. Panza F, Frisardi V, Seripa D et al (2011) Metabolic syndrome,
mild cognitive impairment and dementia. Curr Alzheimer Res
8:492–509
9. He
´
non H, Pasquier F, Leys D (2006) Poststroke dementia.
Cerebrovasc Dis 22:61–70
10. Rastas S, Pirttila
¨
T, Mattila K et al (2010) vascular risk factors
and dementia in the general population aged [85 years: pro-
spective population-based study. Neurobiol Aging 31:1–7
11. Debette S, Seshadri S, Beiser A et al (2011) Midlife vascular risk
factor exposure accelerates structural brain aging and cognitive
decline. Neurology 77:461–468
12. Solfrizzi V, Panza F, Colacicco AM et al (2004) Vascular risk
factors, incidence of MCI, and rates of progression to dementia.
Neurology 63:1882–1891
13. de la Torre JC (2002) Alzheimer disease as a vascular disorder:
nosological evidence. Stroke 33:1152–1162
14. Horsburgh K, Reimer MM, Holland P et al (2011) Axon-glial
disruption: The link between vascular disease and Alzheimer’s
disease? Biochem Soc Trans 39:881–885
15. Silvestrini M, Viticchi G, Falsetti L et al (2011) The role of
carotid atherosclerosis in alzheimer’s disease progression. J
Alzheimers Dis 25:719–726
16. Fergenbaum JH, Bruce S, Spence JD et al (2009) Carotid athero-
sclerosis and a reduced likelihood for lowered cognitive perfor-
mance in a Canadian first nations population. Neuroepidemiology
33:321–328
17. Fernando MS, Simpson JE, Matthews F et al (2006) White matter
lesions in an unselected cohort of the elderly: molecular
pathology suggests origin from chronic hypoperfusion injury.
Stroke 37:1391–1398
18. Prins ND, van Dijk EJ, den Heijer T et al (2004) Cerebral white
matter lesions and the risk of dementia. Arch Neurol
61:1531–1534
19. Juma WM, Lira A, Marzuk A et al (2011) C-reactive protein
expression in a rodent model of chronic cerebral hypoperfusion.
Brain Res 1414:85–93
20. Chobanian AV, Bakris GL, Black HR et al (2003) The seventh
report of the joint national committee on prevention, detection,
evaluation, and treatment of high blood pressure: the JNC 7
report. JAMA 289:2560–2572
21. Alberti KG, Zimmet PZ (1998) Definition, diagnosis and classi-
fication of diabetes mellitus and its complications. Part 1: diag-
nosis and classification of diabetes mellitus provisional report of a
WHO consultation. Diabet Med 15:539–553
22. Executive Summary of The Third Report of The National Cho-
lesterol Education Program (NCEP) (2001) EXPERT panel on
detection, evaluation and treatment of high blood cholesterol in
adults (adult treatment panel III). JAMA 285:2486–2497
23. Zhou H, Deng J, Li J et al (2003) Study of the relationship
between cigarette smoking, alcohol drinking and cognitive
impairment among elderly people in China. Age Ageing
32:205–210
24. Zhou DH, Wang JY, Li J et al (2004) Study on frequency and
predictors of dementia after ischemic stroke: the Chongqing
stroke study. J Neurol 251:421–427
25. Korner A, Lauritzen L, Abelskov K et al (2006) The Geriatric
Depression Scale and the Cornell Scale for Depression in
Dementia: a validity study. Nord J Psychiatry 60:360–364
26. Hachinski V, Iadecola C, Petersen RC et al (2006) National
institute of neurological disorders and stroke: Canadian stroke
network vascular cognitive impairment harmonization standards.
Stroke 37:2220–2241
27. Fuld PA, Masur DM, Blau AD, Crystal H, Aronson MK (1990)
Object-memory evaluation for prospective detection of dementia
in normal functioning elderly: predictive and normative data.
J Clin Exp Neuropsychol 12:520–528
28. Zhang M (1990) Prevalence study on dementia and Alzheimer
disease. Zhonghua Yi Xue Za Zhi 70: 424–428, 430
29. Welsh KA, Butters N, Hughes JP, Mohs RC, Heyman A (1992)
Detection and staging of dementia in Alzheimer’s disease: use of
the neuropsychological measures developed for the Consortium
to Establish a Registry for Alzheimer’s disease. Arch Neurol
49:448–452
30. Pfeiffer E (1975) A short portable mental status questionnaire for
the assessment of organic brain deficit in elderly patients. J Am
Geriatr Soc 23:433–441
31. Silvestrini M, Pasqualetti P, Baruffaldi R et al (2006) Cerebro-
vascular reactivity and cognitive decline in patients with
Alzheimer disease. Stroke 37:1010–1015
32. Petersen RC, Smith GE, Waring SC et al (1999) Mild cognitive
impairment: clinical characterization and outcome. Arch Neurol
56:303–308
33. Palmer K, Di Iulio F, Varsi AE et al (2010) Neuropsychiatric
predictors of progression from amnestic-mild cognitive impair-
ment to Alzheimer’s disease: the role of depression and apathy.
J Alzheimers Dis 20:175–183
34. American Psychiatric Association (1994) Diagnostic and statis-
tical manual of mental disorders: DSM-IV. American Psychiatric
Association, Washington, DC
35. McKhann G, Drachman D, Folstein M et al (1984) Clinical
diagnosis of Alzheimer’s disease: report of the NINCDS-AD-
RDA Work Group under the auspices of Department of Health
and Human Services Task Force on Alzheimer’s disease.
Neurology 34:939–944
1310 J Neurol (2012) 259:1303–1311
123
Page 8
36. Roman GC, Tatemichi TK, Erkinjuntti T et al (1993) Vascular
dementia: diagnostic criteria for research studies: report of the
NINDS-AIREN international workshop. Neurology 43:250–260
37. Rockwood K (2003) Mixed dementia: Alzheimer’s and cerebro-
vascular disease. Int Psychogeriatr 15(Suppl 1):39–46
38. North American Symptomatic Carotid Endarterectomy Trial
Collaborators (1991) Beneficial effect of carotid endarterectomy
in symptomatic patients with high-grade carotid stenosis. N Engl
J Med 325:445–453
39. Fazekas F, Chawluk JB, Alavi A et al (1987) MR signal abnor-
malities at 1.5T in Alzheimer’s dementia and normal aging.
AJNR Am J Neuroradiol 8:421–426
40. Verdelho A, Madureira S, Moleiro C et al (2010) White matter
changes and diabetes predict cognitive decline in the elderly: the
LADIS study. Neurology 75:160–167
41. Li J, Wang YJ, Zhang M et al (2011) Vascular risk factors pro-
mote conversion from mild cognitive impairment to Alzheimer
disease. Neurology 76:1485–1491
42. Ferri CP, Prince M, Brayne C et al (2005) Global prevalence of
dementia: a Delphi consensus study. Lancet 366:2112–2117
43. Morris JC, Storandt M, Miller JP et al (2001) Mild cognitive
impairment represents early-stage Alzheimer disease. Arch
Neurol 58:397–405
44. Peters R, Beckett N (2009) Hypertension, dementia, and antihy-
pertensive treatment: implications for the very elderly. Curr
Hypertens Rep 11:277–282
45. Yaffe K, Fiocco AJ, Lindquist K et al (2009) Predictors of
maintaining cognitive function in older adults: the Health ABC
Study. Neurology 72:2029–2035
46. Silvestrini M, Gobbi B, Pasqualetti P et al (2009) Carotid ath-
erosclerosis and cognitive decline in patients with Alzheimer’s
disease. Neurobiol Aging 30:1177–1183
47. Black SE (2005) Vascular dementia. Stroke risk and sequelae
define therapeutic approaches. Postgrad Med 117(15–6):19–25
48. Silbert LC, Howieson DB, Dodge H, Kaye JA (2009) Cognitive
impairment risk: white matter hyperintensity progression matters.
Neurology 73:120–125
49. Sarti C, Pantoni L, Bartolini L et al (2002) Cognitive impairment
and chronic cerebral hypoperfusion: what can be learned from
experimental models. J Neurol Sci 203–204:263–266
50. Yoshizaki K, Adachi K, Kataoka S et al (2008) Chronic cerebral
hypoperfusion induced by right unilateral common carotid artery
occlusion causes delayed white matter lesions and cognitive
impairment in adult mice. Exp Neurol 210:585–591
51. Aliev G, Smith MA, Obrenovich ME et al (2003) Role of vas-
cular hypoperfusion-induced oxidative stress and mitochondria
failure in the pathogenesis of Azheimer disease. Neurotoxic Res
5:491–504
52. De la Torre JC, Pappas BA, Prevot V et al (2003) Hippocampal
nitric oxide upregulation precedes memory loss and A beta 1–40
accumulation after chronic brain hypoperfusion in rats. Neurol
Res 25:635–641
53. Antoine V, Rigaud AS (2006) Alzheimer’s disease: cardiovas-
cular risk factors must be assessed. Rev Med Interne 27:21–31
54. Zhiyou C, Yong Y, Shanquan S et al (2009) Upregulation of
BACE1 and b-amyloid protein mediated by chronic cerebral
hypoperfusion contributes to cognitive impairment and patho-
genesis of alzheimer’s disease. Neurochem Res 34:1226–1235
J Neurol (2012) 259:1303–1311 1311
123
Page 9
  • Source
    • "Mild cognitive impairment (MCI) is characterized as an early but measurable stage of cognitive impairment [8] [9]. MCI is predictive of progression to dementia and the conversion rate from MCI to dementia was 11.65% per year [10]. One study suggests that individuals with diabetes are 1.5 times more likely to experience cognitive decline and have 1.6 times greater risk of future dementia than individuals without diabetes [11]. "
    [Show abstract] [Hide abstract] ABSTRACT: Is synchronization altered in amnestic mild cognitive impairment (aMCI) and normal cognitive functions subjects in type 2 diabetes mellitus (T2DM)? Resting eye-closed EEG data were recorded in 8 aMCI subjects and 11 age-matched controls in T2DM. Three multivariate synchronization algorithms (S-estimator (S), synchronization index (SI), and global synchronization index (GSI)) were used to measure the synchronization in five ROIs of sLORETA sources for seven bands. Results showed that aMCI group had lower synchronization values than control groups in parietal delta and beta2 bands, temporal delta and beta2 bands, and occipital theta and beta2 bands significantly. Temporal (r = 0.629; P = 0.004) and occipital (r = 0.648; P = 0.003) theta S values were significantly positive correlated with Boston Name Testing. In sum, each of methods reflected that the cortical source synchronization was significantly different between aMCI and control group, and these difference correlated with cognitive functions.
    Full-text · Article · Aug 2014
  • [Show abstract] [Hide abstract] ABSTRACT: Recent data increasingly support a complex interplay between the metabolic condition diabetes mellitus and the pathologically defined nonalcoholic fatty liver disease (NAFLD). NAFLD predicts the development of type 2 diabetes and vice versa, and each condition may serve as a progression factor for the other. Although the association of diabetes and NAFLD is likely to be partly the result of a "common soil," it is also probable that diabetes interacts with NAFLD through specific pathogenic mechanisms. In particular, through interrelated metabolic pathways currently only partly understood, diabetes appears to accelerate the progression of NAFLD to nonalcoholic steatohepatitis, defined by the presence of necroinflammation, with varying degrees of liver fibrosis. In the research setting, obstacles that have made the identification of clinically significant NAFLD, and particularly nonalcoholic steatohepatitis, difficult are being addressed with the use of new imaging techniques combined with risk algorithms derived from peripheral blood profiling. These techniques are likely to be used in the diabetes population in the near future. This review examines the pathogenic links between NAFLD and diabetes by exploring the epidemiological evidence in humans and also through newer animal models. Emerging technology to help screen noninvasively for differing pathological forms of NAFLD and the potential role of preventive and therapeutic approaches for NAFLD in the setting of diabetes are also examined.
    No preview · Article · Dec 2012 · Endocrine reviews
  • [Show abstract] [Hide abstract] ABSTRACT: To identify some risk factors of MCI among patients with type 2 diabetes(T2DM) and to find if there is any correlation between these factors and the degree of cognitive decline. A total of 155 patients with T2DM referred to the Department of Endocrinology at First Hospital of Qinhuangdao were enrolled. To assess MCI the Montreal Cognitive Assessment (MoCA) scoring system was used. There were 66 patients with MCI and 89 patients without MCI (control). HbAlc, blood lipid, liver and renal functions were measured in all subjects. Compared with the control group, type 2 diabetic patients with MCI had a longer duration of diabetes; higher non-high-density lipoprotein cholesterol (non-HDL-C), triglycerides, total cholesterol, HbA1c, and BMI; and lower high-density lipoprotein cholesterol (HDL-C) (P<0.05). The rates of patients with a history of habitual light-to-moderate alcohol consumption, a high proportion of Mediterranean-type diet, and regular physical activity were lower; and the rate of current smoking was higher in type 2 diabetic patients with MCI than the control group (P<0.05). Among patients with MCI, the results indicated that MoCA score was negatively correlated with non-HDL-C (r=-0.761 P<0.001). Our results suggest that non-HDL-C can act as a readily available method for estimating risk of MCI in Chinese type 2 diabetic patient in routine clinical practice. Good lifestyle likely reduces MCI risk in diabetic patients.
    No preview · Article · Jul 2013 · Journal of diabetes and its complications
Show more

Similar Publications