Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Untreated hyperglycemia associated with tau pathology and worse cognitive performance in older adults
(USC Thesis Other)
Untreated hyperglycemia associated with tau pathology and worse cognitive performance in older adults
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Untreated Hyperglycemia Associated with Tau Pathology and Worse Cognitive Performance in
Older Adults
Elissa C. McIntosh
Clinical Science Program
Master of Arts (Psychology)
University of Southern California
August 2017
2
Table of Contents
Introduction ……………………………………………………………………………………… 3
Methods ………………………………………………………………………………………….. 8
Results ………………………………………………………………………………………….. 14
Discussion ……………………………………………………………………………………… 19
References ……………………………………………………………………………………… 32
Figures ………………………………………………………………………………………….. 46
Figure 1 …………………………………………………………………………………….. 46
Figure 2 …………………………………………………………………………………….. 47
Figure 3 …………………………………………………………………………………….. 48
Figure 4 …………………………………………………………………………………….. 49
Figure 5 …………………………………………………………………………………….. 50
3
Introduction
In the United States alone, it is estimated that 29.1 million people have type 2 diabetes
(T2DM) with 8.1 million people undiagnosed (Center for Disease Control and Prevention, 2014).
T2DM is a chronic metabolic disease characterized by high blood sugar (hyperglycemia) that is
caused by either deficient insulin or the inability to use insulin efficiently in the body. T2DM is
most common in older age, but this health condition has been on the rise in young and middle-
aged adults (Alberti et al., 2004). Population-based studies have consistently linked T2DM to
increased risk for dementia (Biessels, Staekenborg, Brunner, Brayne, & Scheltens, 2006; Lu,
Lin, & Kuo, 2009; Profenno, Porsteinsson, & Faraone, 2010). Longitudinal studies have found
T2DM to be associated with both vascular dementia (VaD) (Hassing et al., 2002; Hayden et al.,
2006; MacKnight, Rockwood, Awalt, & McDowell, 2002; Xu, von Strauss, Qiu, Winblad, &
Fratiglioni, 2009) and Alzheimer’s disease (AD) (Irie et al., 2008; Ott et al., 1999; Xu et al.,
2009). A recent meta-analysis of longitudinal studies reported that the relative risk for AD was
1.46 (95% confidence interval (CI): 1.20-1.77) for people with T2DM and 2.48 for VaD (95%
CI: 2.08-2.96) (Cheng, Huang, Deng, & Wang, 2012). The relationship between T2DM and
dementia is thought to be a result of cerebrovascular and/or neurodegenerative disease (Biessels
et al., 2006; Ott et al., 1999; Peila, Rodriguez, & Launer, 2002). While it is widely agreed upon
that T2DM substantially increases risk for dementia, the mechanisms responsible for this
association are not well understood.
Alzheimer’s disease is the most common cause of dementia in older adults, accounting
for about 60-80% of dementia cases (Alzheimer’s Association, 2016). In terms of
neuropathology, AD is characterized by neuritic plaques, neurofibrillary tangles, synaptic
dysfunction, and brain atrophy, especially in the medial temporal lobes (Weiner et al., 2010).
4
Some AD models state that the disease begins with amyloid beta (Aβ) accumulation in the brain,
characterized by extracellular deposits of beta-amyloid peptides that form insoluble amyloid
plaques (Mayeda, Whitmer, & Yaffe, 2015; Weiner et al., 2010) that are thought to contribute to
neurodegeneration (Reitz, 2012). In living patients, Aβ is quantified through measures of
cerebrospinal fluid (CSF) and by positron emission tomography (PET) amyloid imaging. Levels
of the 42 amino acid form of amyloid beta (Aβ-42) in the CSF are a sensitive diagnostic marker
for AD (Andreasen et al., 1999). Low levels of Aβ-42 in the CSF are interpreted as an indicator
of amyloid retention in the brain (Fagan & Holtzman, 2010). In AD, a second marker is tau
protein; tau protein in the brain is abnormally hyperphosphorylated leading to phosphorylated tau
(p-tau) that accumulates into neurofibrillary tangles (NFTs) (Iqbal et al., 2005). Elevated levels
of tau and p-tau in the CSF are in vivo markers of tauopathy and correlate with tau-related AD
pathology in the brain (Apostolova et al., 2010; Braak & Braak, 1991). Increased tau and p-tau in
the CSF likely result from neuronal death. When neurons die they release tau and p-tau in the
extracellular space which eventually gets released into the CSF. Since neuronal death occurs
fairly late relative to synapse loss and loss of cellular function, tau and p-tau build up in the CSF
indicates more significant tau pathology in the brain. Elevated CSF levels of tau and p-tau have
been positively correlated with NFT burden in autopsy studies (Buerger et al., 2006; Tapiola et
al., 2009). The process of neurodegeneration in living patients is also reflected in synaptic
dysfunction on FDG-PET, and atrophy (Weiner et al., 2010).
The relationship between T2DM and AD and dementia has been studied using various
methods and outcome measures, including neuropsychological and neuroimaging research.
Neuropsychological research has repeatedly demonstrated individuals with T2DM have worse
cognitive performance in multiple domains such as verbal memory, information processing
5
speed and attention, and executive functioning compared to those without T2DM (Reijmer, van
den Berg, Ruis, Kappelle, & Biessels, 2010; van den Berg, Kloppenborg, Kessels, Kappelle, &
Biessels, 2009). Though the relationship between T2DM and cognitive dysfunction is well
documented, the mechanism underlying this relationship is poorly understood. Magnetic
resonance imaging (MRI) has been another tool used to study the relationship between T2DM
and dementia. Studies have consistently linked T2DM and atrophy, including in medial temporal
areas known to be affected in AD (e.g., hippocampus, medial temporal lobe). Additionally,
T2DM has been related to markers of cerebrovascular damage such as cerebral infarcts, and
white matter hyperintensities (WMH). One study reported that T2DM was associated with
worse performance in tasks evaluating visuospatial construction, planning, visual memory and
processing speed. These relationships were attenuated after adjustment for hippocampal and
gray matter volume, but not cerebrovascular lesions, suggesting that T2DM-related cognitive
impairment may be driven by neurodegeneration, not cerebrovascular damage (Moran et al.,
2013).
T2DM pathophysiology is thought to be related to AD pathophysiology via several
mechanisms. Insulin resistance, and subsequent hyperglycemia, seem integral in explaining this
relationship. Insulin resistance is often a causal factor in T2DM, and often leads to
hyperinsulinemia in the periphery due to reduced insulin clearance. Seemingly contradictory,
chronic peripheral hyperinsulinemia due to insulin resistance is associated with decreased insulin
transport into the brain, which decreases insulin levels in the brain. Reduced insulin signaling in
the brain has been associated with AD pathophysiology (Craft & Watson, 2004; Zhao &
Townsend, 2009). AD has sometimes been described as “Type 3 Diabetes,” which refers to the
idea that AD represents a type of diabetes that only affects the brain (de la Monte & Wands,
6
2008). Other possible mechanisms are increased advanced glycation end products and oxidative
stress due to hyperglycemia, inflammation, and macrovascular and microvascular damage (Craft,
2009).
Animal studies using T2DM models and AD models have both found that T2DM and
related conditions have supported the link between T2DM and AD pathophysiologies. In T2DM
animal models, T2DM and related conditions such as hyperinsulinemia and insulin resistance
have shown AD pathological processes such as neurodegeneration, increased tau cleavage,
increased tau phosphorylation, and Aβ (Kim, Backus, Oh, Hayes, & Feldman, 2009; Li, Zhang,
& Sima, 2007; Park, 2011). In animal work using AD models, diet-induced insulin resistance and
high fat diets have exacerbated amyloidosis (Ho et al., 2004; Kohjima, Sun, & Chan, 2010).
Human studies have utilized CSF AD biomarkers or post-mortem autopsy studies to
elucidate the connection between T2DM and AD neuropathology. Recently, one study
demonstrated that T2DM was associated with greater CSF tau and p-tau (Moran et al., 2015).
They also reported that T2DM was associated with decreased cortical thickness; this relationship
was attenuated when adjusting for p-tau suggesting that tau hyperphosphorylation may mediate
the relationship between T2DM and neurodegeneration. T2DM was not related to CSF levels of
Aβ-42 in this sample; the authors suggested that this null finding supports the idea that
neurodegenerative processes in T2DM are more likely driven by tau than Aβ (Moran et al.,
2015). In contrast, human autopsy studies have often failed to find a significant association
between T2DM and amyloid plaque and neurofibrillary tangle burden (Alafuzoff, Aho,
Helisalmi, Mannermaa, & Soininen, 2009; Arvanitakis et al., 2006; Heitner & Dickson, 1997).
At the same time, there is more consistent evidence from autopsy studies that people with T2DM
have more cerebral infarcts (Ahtiluoto et al., 2010; Arvanitakis et al., 2006; Nelson et al., 2009).
7
Thus, despite these interesting findings, more human research on the relationship between T2DM
and CSF AD biomarkers are needed in light of contradictory evidence. Taking into consideration
the evidence from autopsy studies, the relationship between T2DM and AD may be better
understood by the hypothesis that vascular brain injury associated with T2DM (e.g., lacunar
infarcts, WMH, cerebral small vessel disease) decreases the threshold at which AD pathology
manifests as cognitive impairment (Exalto, Whitmer, Kappele, & Biessels, 2012). T2DM
increases vascular brain injury by damaging endothelial and smooth muscle cells which leads to
atherosclerosis and related pathologies. Specifically, hyperglycemia decreases nitric oxide (NO)
bioavailability via the free radical superoxide anion, and increases reactive oxygen species
(ROS) causing endothelial dysfunction. Reductions in NO bioavailability are associated with
cardiovascular events, and accumulation of ROS is assumed to be the link between
hyperglycemia and the biochemical pathways involved in diabetes-related vascular
complications (Paneni, Beckman, Creager, & Cosentino, 2013).
Inconsistent findings in the literature investigating relationships between T2DM and AD
neuropathology may be partly explained by the use of heterogeneous T2DM groups in research
and not accounting for the role of glycemic control. Definitions of T2DM populations in research
often do not distinguish between treated and untreated diabetic patients thus there is great
heterogeneity in degrees of glycemic control. This is problematic, as previous research has
shown that diabetes-related cognitive dysfunction may be attenuated by glycemic control (Gao,
Matthews, Sargeant, & Brayne, 2008; Shorr et al., 2006). In addition, several longitudinal studies
have demonstrated that glycemic control (Xu et al., 2009) and medication use (Hsu, Wahlqvist,
Lee, & Tsai, 2011) may decrease risk for dementia. Glucose-lowering diabetes medications often
work by acting as insulin sensitizers, allowing the body to better absorb glucose to be used for
8
energy. By decreasing insulin resistance, these drugs help to decrease hyperinsulinemia and
hyperglycemia pathologies that may mediate the relationship between T2DM and AD pathology.
Given these findings, it is critical that research distinguish between untreated and treated T2DM
so that we may better understand the relationship between T2DM and dementia, and how
medication may moderate this relationship.
The purpose of the present study was to longitudinally investigate the relationships
between T2DM, glycemic control, and markers of dementia, particularly AD, in nondemented
individuals without any evidence of major cerebrovascular disease. Based on fasting blood
glucose and medication history, participants were designated as having euglycemia, untreated
hyperglycemia, or treated diabetes. The current study hypothesized 1) The untreated
hyperglycemia group would show the most evidence of AD pathology on CSF measures and the
euglycemia group would show the least evidence of pathology; 2) The untreated hyperglycemia
group would perform the worst on baseline cognitive measures and exhibit the fastest rate of
cognitive decline, while the euglycemia group would perform the best and show least evidence
of decline; 3) The untreated hyperglycemia and treated diabetes group would have lower brain
volumes in regions implicated in AD at baseline and show greater rates of atrophy compared to
the euglycemia group; and 4) The untreated hyperglycemia and treated diabetes group would
have greater WMH volumes at baseline compared to the euglycemia group. Further, the
untreated hyperglycemia group would show the greatest WMH volume progression, while the
euglycemia group would show the least WMH progression.
Methods
Data were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database
(adni.loni.usc.edu). ADNI was launched in 2003 with the primary purpose of investigating the
9
utility of MRI, PET, CSF, clinical, neuropsychological, and other biological markers in
predicting progression of mild cognitive impairment (MCI) to early AD dementia (Mueller et al.,
2005). Exclusion criteria from the ADNI study included a Hachinski ischemic score (Hachinski
et al., 1975) greater than four, inability to participate in MRI, presence of neurologic disorders,
current depression, history of psychiatric diagnosis, recent substance dependence, less than six
years of education, and lack of fluency in English or Spanish.
Participants
Participants were 907 nondemented ADNI-1, ADNI-Grand Opportunity (ADNI-GO), and
ADNI-2 participants who were classified as either cognitive normal or MCI at screening
evaluation. Criteria for MCI were 1) subjective memory complaint from the participant or
informant; 2) Mini-Mental State Examination scores between 24-30; 3) global Clinical Dementia
Rate score of 0.5; 4) scoring below education-adjusted cutoffs for delayed free recall on story A
of the Wechsler Memory Scale-Revised (WMS-R) Logical Memory II subtest; 5) mostly intact
general cognition and functional performance and did not qualify for dementia diagnosis
(Petersen et al., 2010). Participants were required to have baseline fasting blood glucose data
available and/or records of diabetic medication, and CSF biomarker measures to qualify for the
present study.
Glycemic groups. Participants were placed into one of three groups based on their
baseline fasting blood glucose values and reported medications. Glucose cut offs were groups
were established using guidelines recommended by the American Diabetes Association
("Standards of medical care in diabetes--2012," 2012). Seven-hundred ninety-one participants
were classified as the euglycemia group. Participants in the euglycemia group had fasting blood
glucose £ 125 mg/dL and did not report any use of medications used to treat T2DM. Fifty-one
10
participants were classified as the untreated hyperglycemia group. Participants in the untreated
hyperglycemia group had fasting blood glucose ≥ 126 mg/dL, and did not report any use of
medications used to treat T2DM. Fifty-nine participants were classified as the treated diabetes
group. Participants in the treated diabetes group reported use of one or more oral T2DM
medications at baseline. For the treated diabetes group, we collected information about each
medication’s drug class and whether they used monotherapy or combined therapies (Table 1).
Participants with insulin-dependent diabetes as evidence by the use of insulin medication (N = 7)
for excluded from the study as use of insulin medication may signify more serious cases of
T2DM and/or longer duration of T2DM (Exalto et al., 2012), or type 1 diabetes mellitus.
Table 1. Medication class breakdown and type of therapy for treated diabetes group.
Medication Class Number of Participants Percentage of Participants
Biguanides (Metformin) 42.0 71.0
Thiazolidinediones 9.0 15.0
Sulfonylureas 18.0 31.0
Dipeptidyl-peptidase-4
Inhibitors
5.0 8.5
Glucagon-like-peptide-1
(GLP-1)
1.0 1.7
Type of Treatment
Monotherapy 45.0 76.0
Combination Therapy 14.0 24.0
Physiological, Clinical, and Genetic Data
Subjects underwent fasting blood draws and physiological assessments including seated
brachial artery blood pressure assessment, and weight and height measurements. At the screening
visit, a thorough medical history, including medication use, was completed. BMI was calculated
as weight (kg) divided by height (meters) squared. Blood samples were used to measure fasting
blood glucose, and ApoE4 carrier status. ApoE genotyping was conducted by the ADNI
11
Biomarker Core at the University of Pennsylvania. Participants were designated as ApoE ε4
positive if they carried one or more ApoE ε4 alleles.
Vascular risk factors. Vascular risk factors were assessed at the baseline and screening
visits using medical history and physical examination. Vascular risk factor burden was measured
using criteria adapted from the Framingham Stroke Risk Profile (Wolf, D'Agostino, Belanger, &
Kannel, 1991) and Framingham Coronary Risk Profile (Wilson et al., 1998). The present study
evaluated group differences in history of hypertension, dyslipidemia, and cardiovascular disease.
Cognitive diagnosis. Cognitive diagnoses were assigned using criteria from the National
Institute on Neurologic and Communicative Disorders and Stroke/Alzheimer’s Disease and
Related Disorders Association (McKhann et al., 1984). Participants diagnosed with MCI had
memory complaints, but did not have significant functional impairment as assessed by the
Clinical Dementia Rating (CDR) (Morris, 1997). The CDR sum of boxes (CDR-SB) score was
calculated to further measure stage of dementia. For the purposes of this study, participants
diagnosed with early AD at baseline were excluded.
Neuropsychological measures. Global cognitive decline was measured using the Mini
Mental Status Exam (MMSE) and the Alzheimer's Disease Assessment Scale- Cognitive
Subscale test (ADAS-Cog). MMSE measures orientation to time and place, registration, attention
and calculation, recall, language, repetition, and complex commands. A higher score reflects
better cognitive functioning. The ADAS-Cog consists of 11 parts and primarily measures
language and memory. The CDR-SB was calculated to further measure stage of dementia.
Cognitive measures included: 1) Rey Auditory Verbal Learning Test (AVLT): total immediate
recall score for trials 1-5, delayed score; 2) Wechsler Adult Intelligence Scale-Revised (WAIS-
R) Digit Span: forward and backward scores; 3) Trail Making Test: parts A and B, times to
12
completion; 5) Animal Fluency: total score; 6) Boston Naming Test (BNT): total score; and 7)
Digit Symbol Substitution Test (DSST): total score. Measures were collected at three time
points: baseline, month 12, and month 24.
MRI measures. All ADNI participants underwent a 1.5 Tesla MRI scan at the screening
or baseline visit. ADNI scans were preprocessed (gradient warping, scaling, B1 correction, and
N3 inhomogeneity correct) to correct for different scanners at the study sites. Cortical volumes
were processed by the Center from Imaging of Neurodegenerative Disease at the University of
California, San Francisco. Cortical reconstruction and volumetric segmentation was completed
using the Freesurfer image analysis suite. More details on MRI acquisition and pre-processing
can be found elsewhere (Jack et al., 2008). The following measures were taken directly from the
ADNI databases: whole brain volume, ventricle volume, hippocampal volume, entorhinal cortex
volume, and intracranial volume (ICV).
White matter intensity (WMH) volumes were determined at the University of California,
Davis. In the ADNI1 cohort, WMH volumes were calculated using an automated segmentation
method with a Markov-Random Field approach (Schwarz, Fletcher, DeCarli, & Carmichael,
2009). In the ADNIGO and ADNI2 cohorts, WMH volumes were calculated with an updated
four-tissue segmentation that used a Bayesian approach to segment high resolution 3D T1 and
FLAIR sequences. Log transformation was applied to WMH volume values in both the ADNI1
and ADNIGO/ADNI2 data separately to correct for kurtosis in the raw data distribution.
Measures were collected at three time points: baseline, month 12, and month 24.
CSF biomarkers. Baseline fasting CSF biomarkers were collected and analyzed using a
Luminex platform (Luminex Corporation, Austin, TX) with an Innogenetics immunoassay kit
(INNO-BIA AlzBio3; Ghent, Belgium). CSF biomarkers for AD included Aβ1-42, tau, and p-tau
13
(pg/mL). More information on CSF collection and protocols are described elsewhere (Shaw et
al., 2009). Based on previous studies, biomarker profiles were determined using the following
cutoff values for CSF AD biomarkers: Aβ1-42 (≤192 pg/mL), p-tau (≥23 pg/mL), and tau (≥93
pg/mL) (Nation et al., 2015; Shaw et al., 2009). Participants were categorized as either
biomarker positive or biomarker negative for each biomarker.
Statistical Analyses
Demographic, clinical, and cognitive comparisons. ANOVA and Chi-square
tests were
used to test for group differences in demographic (age, sex, education), clinical (ApoE genotype,
BMI, fasting blood glucose, systolic and diastolic blood pressure, WMH volume), and cognitive
variables (diagnosis, MMSE score, ADAS-Cog score). Groups were also assessed for differences
in history of cardiovascular disease, hypertension, and dyslipidemia.
Baseline analyses. All raw data were screened for departures from normality (skewness,
kurtosis). Due to significant departures from the normal distribution, the following variables
were log-transformed: CDR-SB, ADAS-Cog, Trails A, Trails B, and WMH volume. The log-
transformation was judged to adequately improve skewness and/or kurtosis. ANCOVA was used
to assess group differences in CSF AD biomarkers (Aβ1-42, tau, and p-tau), cognitive
performance on all neuropsychological measures, and MRI variables. For most analyses,
covariates included age, sex, years of education, and ApoE4 carrier status. Additionally, for MRI
analyses, with the exception of WMH volumes, estimated intracranial volume (ICV) was used a
covariate to account for differences in head size. For WMH analyses, all log-transformed data
were analyzed together, but WMH cohort was added as a covariate to control for differing
methods of measuring WMH. Pairwise comparisons were investigated using post-hoc least
14
significant difference tests. All analyses used two-tailed significance testing with an alpha of
0.05.
To assess for group differences in biomarker profiles, logistic regression analyses
controlling for age, sex, and ApoE4 carrier status were used. Using the guidelines specified
above for biomarker cutoffs, pairwise comparisons of biomarker profiles were performed for
each biomarker.
Longitudinal analyses. To examine time x group interactions and group differences for
cognitive and MRI measures, mixed model analyses were used with compound structure
covariance structure and maximum likelihood estimation. Group, time, time x group, age, sex,
and ApoE4 carrier status were entered as fixed factors in all models. Years of education was
added as fixed factor in cognitive analyses. Time was added as a repeated factor. Baseline ICV
was also included as a fixed factor in MRI analyses with the exception of WMH volume.
Analyses used two-tailed significance testing with an alpha of 0.05. All analyses were performed
with SPSS for Mac OS X version 24.
Results
Physiological, Clinical, and Genetic Data
ANOVA and chi-square analyses revealed significant group main effect differences in
age, sex, fasting blood glucose, and BMI at baseline (see Table 2). For age, pairwise
comparisons showed the untreated hyperglycemia group was significantly older than the treated
diabetes group (p = 0.013). For sex differences, pairwise comparisons demonstrated that the
treated diabetes group had significantly more males than both the euglycemia (c
2
= 8.67, p =
0.003) and untreated hyperglycemia (c
2
= 6.00, p = 0.014) groups, but the euglycemia and
untreated hyperglycemia groups did not differ (p = 0.589). For fasting blood glucose, pairwise
15
comparisons showed the untreated hyperglycemia group had significantly higher glucose values
compared to the euglycemia (p < 0.001) group, and the treated diabetes group had significantly
higher glucose values than the euglycemia group (p < 0.001). For BMI, the treated diabetes
group had significantly higher BMIs compared to both the euglycemia (p < 0.001) and untreated
hyperglycemia (p = 0.001) groups. There were no group differences in cognitive diagnosis, years
of education, ApoE genotype, systolic blood pressure, or diastolic blood pressure.
According to medical histories, there were significant group differences in histories of
hypertension and dyslipidemia, but not cardiovascular disease (see Table 3). The treated diabetes
group had significantly more history of hypertension compared to both the euglycemia (c
2
=
14.98, p < 0.001) and untreated hyperglycemia (c
2
= 9.55, p = 0.002) groups. With regard to
history of dyslipidemia, the treated diabetes group had significantly more history of dyslipidemia
compared to both the euglycemia (c
2
= 13.24, p < 0.001) and untreated hyperglycemia (c
2
= 9.45,
p = 0.002) groups. The untreated diabetes group did not differ from the euglycemia group in
histories of hypertension (p = 0.550), dyslipidemia (p = 0.626), or cardiovascular disease (p =
0.617).
Table 2. Participant Characteristics.
Euglycemia
(N = 791)
Treated
Diabetes
(N = 59)
Untreated
Hyperglycemia
(N = 51)
P-
value
Age 73.03 71.31 74.69 0.044
Sex, male (%) 434 (55) 44 (75) 26 (51) 0.010
Education 16.14 16.02 16.20 0.934
ApoE, positive (%) 335 (42) 27 (46) 20 (39) 0.784
Baseline Diagnosis, MCI
(%)
532 (67) 46 (78) 34 (67) 0.231
MMSE 28.16 28.17 27.59 0.075
Fasting Glucose 95.51 121.64 145.29 <0.001
BMI 26.92 30.08 27.24 <0.001
Systolic BP 133.79 133.81 134.96 0.899
Diastolic BP 74.48 74.02 73.16 0.641
16
Table 3. Vascular Risk Factors.
Euglycemia
(N = 791)
Treated
Diabetes (N
= 59)
Untreated
Hyperglycemia
(N = 51)
c
2
P-Value
Hypertension
(%)
344 (43) 41 (69) 20 (40) 15.72 < 0.001
Dyslipidemia
(%)
369 (47) 42 (71) 21 (42) 13.77 0.001
Cardiovascular
Disease (%)
44 (<1) 7 (12) 2 (<1) 4.31 0.116
CSF Biomarker Analyses
ANCOVA analyses revealed a significant group main effect for p-tau (p = 0.018),
controlling for age, sex, and ApoE4 carrier status. Pairwise comparisons showed the untreated
hyperglycemia group had significantly higher levels of CSF p-tau compared to the euglycemia
group (p = 0.007), but not the treated diabetes group (p = 0.170). There were no differences
between the euglycemia and treated diabetes groups (p = 0.341). For CSF tau, there was a
marginally significant group main effect (p = 0.074). Pairwise comparisons showed the untreated
hyperglycemia had significantly higher levels of CSF tau compared to the euglycemia group (p =
0.025), and showed a trend for higher levels of CSF tau compared to the treated diabetes group
(p = 0.063). Due to significant group differences in history of hypertension and dyslipidemia,
history of hypertension and dyslipidemia were added as covariates in additional analyses. Group
differences in p-tau (p = 0.018) remained unchanged after adding hypertension history to the
model, but the group differences in tau were slightly attenuated (p = 0.087). Group differences in
p-tau (p = 0.019) remained relatively unchanged after adding dyslipidemia history to the model,
and group differences in tau (p = 0.074) were unchanged. All significant pairwise comparisons
remained significant after controlling for history of hypertension and dyslipidemia. There was no
significant group difference for Aβ1-42 (p = 0.715). Additionally, there were no differences in
Aβ1-42 between the untreated hyperglycemia group and the euglycemia group (p = 0.238) or the
17
treated diabetes group (p = 0.484), or between the treated diabetes group and the euglycemia
group (p = 0.792).
Logistic regression controlling for age, sex, and ApoE4 carrier status showed a similar
pattern of findings as ANCOVA analyses (Figures 1 and 2). The untreated hyperglycemia group
had significantly more p-tau positive (p = 0.028) and tau positive (p = 0.018) individuals
compared to the euglycemia group, but not the treated diabetes group. The untreated
hyperglycemia group had larger proportions of p-tau positive (p = 0.095) and tau-positive (p =
0.087) individuals, but these differences did not reach significance. The treated diabetes group
did not significantly differ from the euglycemia group in either p-tau (p = 0.960) or tau (p =
0.578) profiles. History of hypertension did not attenuate differences between the untreated
hyperglycemia and euglycemia groups in proportions of p-tau (p = 0.025) and tau (p = 0.016)
positive profiles. Similarly, history of cholesterol did not affect differences between the untreated
hyperglycemia and euglycemia groups in proportions of p-tau (p = 0.025) and tau (p = 0.016)
positive profiles. There were no significant findings for Ab profiles. Specifically, there were no
differences in proportions of Ab positive profiles between the untreated hyperglycemia group
and the euglycemia group (p = 0.836), between the untreated hyperglycemia group and the
treated diabetes group (p = 0.616), or between the euglycemia group and treated diabetes group
(p = 0.421)
Baseline Analyses: Cognitive and MRI Data
For baseline neuropsychological measures, ANCOVA analyses showed there were
significant group main effects for DSST (p = 0.045), and a non-significant trend for Animal
Fluency (p = 0.065), and MMSE (p = 0.072), controlling for age, sex, education, and ApoE4
carrier status. On the DSST, pairwise comparisons showed the untreated hyperglycemia group
18
scored significantly worse than the euglycemia group (p = 0.013). On Animal Fluency, the
untreated hyperglycemia group scored significantly worse compared to the euglycemia group (p
= 0.031). On the MMSE, the untreated hyperglycemia group scored significantly worse than the
euglycemia group (p = 0.023). Though not significant, the untreated hyperglycemia group scored
worse on the MMSE relative to the treated diabetes group (p = 0.067).
For baseline MRI measures, ANCOVA analyses showed no main effect for group on
baseline WMH volume (p = 0.135) controlling for age, sex, ApoE4 carrier status, and WMH
methodology. There were no differences in any of the other MRI volumes analyzed (entorhinal,
p = 0.987; hippocampus, p = 0.817; middle temporal, p = 0.413; whole brain, p = 0.523;
ventricles, p = 0.368).
Longitudinal Analyses: Cognitive and MRI Data
Group x time interactions. For neuropsychological data, linear mixed models showed a
significant group x time interaction for CDR-SB (F = 4.502, p = 0.001) (Figure 3) and a
marginally significant group x time interaction for MMSE (F = 2.292, p = 0.058) (Figure 4). On
the CDR-SB, all groups showed significant decline (increased scores on the CDR-SB) between
baseline and month 24 (Euglycemia: b = 0.061, p < 0.001; Treated Diabetes: b = 0.103, p <
0.001; Untreated Hyperglycemia: b = 0.143, p < 0.001). The untreated hyperglycemia group
showed accelerated decline on the CDR-SB compared to the euglycemia group (p = 0.001), but
not the treated diabetes group (p = 0.213). At month 24, the euglycemia group had significantly
better scores compared to both the untreated hyperglycemia group (p = 0.004) and the treated
diabetes group (p = 0.019). There was no difference in rate of decline between the euglycemia
and treated diabetes group (p = 0.102). For MMSE, all groups showed significant decline in
MMSE scores over the two-year period (Euglycemia: b = -0.750, p < 0.001; Treated Diabetes: b
19
= -1.310, p < 0.001; Untreated Hyperglycemia: b = -1.465, p < 0.001). The rate of decline for the
untreated hyperglycemia group was significantly greater compared to the euglycemia group (p =
0.041), but not the treated diabetes group (p = 0.734). At month 24, the untreated hyperglycemia
group had significantly worse performance on the MMSE compared to the euglycemia group (p
< 0.001), but not the treated diabetes group (p = 0.125).
For MRI data, there was a non-significant group x time interaction for WMH volume (F
= 2.141, p = 0.074) (Figure 5). WMH volume significantly increased between baseline and
month 24 for the euglycemia group (b = 0.066, p < 0.001) and marginally increased for the
untreated hyperglycemia group (b = 0.129, p = 0.058), but there was no significant change in
WMH volume in the treated diabetes group (b = -0.042, p = 0.493). The rate of change in WMH
volume was marginally greater in the untreated hyperglycemia group compared to the treated
diabetes group (p = 0.062), but was not different than the euglycemia group (p = 0.365). At
month 24, there were no group differences in WMH volume (p = 0.557).
Group main effects. There were significant effects of group for CDR-SB (p = 0.031),
MMSE (p = 0.021).
Time main effects. Additionally, there were significant main effects of time for WMH
volume (p = 0.019), CDR-SB (p < 0.001), MMSE (p < 0.001), Animal Fluency (p = 0.030),
Trails B (p = 0.050), BNT total (p = 0.027), ADAS-Cog 11 (p = 0.016), and volumes for whole
brain (p < 0.001), ventricles (p < 0.001), hippocampus (p < 0.001), entorhinal cortex (p < 0.001),
and middle temporal area (p < 0.001).
Discussion
In the present study, several hypotheses were investigated to better understand how
treatment of hyperglycemia and T2DM may influence risk for dementia and accelerated
20
cognitive decline. We propose that untreated hyperglycemia may increase the risk for cognitive
decline and dementia through its associations with tauopathy, and cognitive dysfunction. Results
of CSF analyses provided evidence that oral medication management of hyperglycemia or
T2DM may impact CSF levels of p-tau and tau, important biomarkers of AD. Thus, treatment of
hyperglycemia may affect tau-mediated processes. Moran et al. (2015) was the first to show that
T2DM was associated with greater p-tau in their ADNI sample of cognitively normal, MCI, and
AD patients. When looking at each diagnostic group separately, they reported that only the MCI
group showed a relationship between T2DM and p-tau. In contrast to the current study, Moran et
al. defined people as having T2DM based on elevated fasting blood glucose or medication use.
Thus, while providing very interesting findings about a relationship between T2DM and p-tau,
this study did not explore how medication use may moderate this relationship. By distinguishing
between untreated hyperglycemia and treated diabetes in this study, we showed that the use of
glucose-lowering medications may affect the relationship between hyperglycemia and tauopathy.
Specifically, our findings suggest that the use of medication may decrease the phosphorylation of
tau in persons with T2DM. As noted above, the ANCOVA analysis showed that the untreated
hyperglycemia group had greater p-tau and tau than the euglycemia group, but not the treated
diabetes group. This suggests that medication may only ameliorate the effects of tau pathology
associated with hyperglycemia. While the untreated hyperglycemia group had higher levels of
CSF p-tau and tau compared to the treated diabetes group, these differences were not significant.
Therefore, it appears that glycemic control via medication use may reduce or prevent tau
pathology, but individuals with treated T2DM still demonstrate more tauopathy than euglycemic
individuals.
21
The finding that untreated hyperglycemia was associated with increased p-tau is
consistent with the animal research using T2DM models. Spontaneous models of T2DM are
obesity-induced diabetes models. In these models, which include Bio-Breeding Zucker diabetic
rat/Wor rats (BBZDR/Wor-rat) and db/db mice models, T2DM has been linked to AD
pathological processes such as neurodegeneration, increased tau cleavage, and increased tau
hyperphosphorylation (Kim et al., 2009; Li et al., 2007). In the BBZDR/Wor-rat model, obesity
precedes T2DM. The model produces peripheral insulin resistance with hyperinsulinemia, as
well as hypercholesterolemia and hyperlipidemia (Li et al., 2007). In the db/db mouse model, the
diabetes db gene mutation causes mice to be develop hyperinsulinemia and insulin resistance in
early life, then obesity. Subsequently, hyperglycemia becomes apparent due to beta cell failure
and the hyperinsulinemia abates (Wang et al., 2013).
Animal studies have shown evidence that T2DM produces more AD pathology than type
1 diabetes. Compared to streptozotocin (STZ) models that cause type 1 diabetes through
destroying insulin-secreting pancreatic beta cells, T2DM animal models that have
hyperinsulinemia and insulin resistance, conditions that often precede and accompany T2DM,
show more severe AD pathology that includes increased tau phosphorylation, increased tau
cleavage, and more extensive neuronal and synaptic damage. These findings suggest that the
insulin deficiency seen in type 1 diabetes (e.g. STZ models) does not produce as much AD
pathology as T2DM models that also produce hyperinsulinemia and insulin resistance (Park,
2011). Similarly, one study showed that Ab and related markers were significantly increased in
T2DM model rats (BBZDR-Wor-rat) compared to type 1 diabetes rats (BB/Wor-rat). (Li et al.,
2007). Overall, these animal study findings support the connection between insulin resistance,
hyperinsulinemia, and pathophysiological processes seen in AD (Craft, 2009).
22
Consistent with Moran et al. (2015) and counter to animal research, we did not find any
group differences in Ab pathology. Animal studies using T2DM and AD models have reported
evidence for a relationship between T2DM and Ab (Li et al., 2007). In a transgenic mouse model
of AD, diet-induced insulin resistance exacerbated amyloidosis (Ho et al., 2004). A more recent
study reported similar findings of increased Aβ in the brain in their AD model with a high-fat
diet (Kohjima et al., 2010). The incorporation of a high-fat diet in AD models leads to insulin
resistance and obesity, thereby mimicking T2DM pathologies. Our results are more consistent
with T2DM animal models that have demonstrated a potential relationship between T2DM and
tauopathy.
There are several potential mechanisms through which hyperglycemia and T2DM may
lead to increased levels of p-tau in the brain. These mechanisms include hyperinsulinemia,
inflammation, vascular factors, and oxidative stress (Yarchoan & Arnold, 2014).
Hyperinsulinemia, a process closely tied to T2DM, may induce brain insulin resistance by
reducing insulin receptor expression and receptor kinase activity (Kim, Sullivan, Backus, &
Feldman, 2011) which may then promote tau pathology (Yarchoan & Arnold, 2014). Abnormal
insulin signaling the brain may increase hyperphosphorylated tau through increased glycogen
synthase kinase 3 (GSK-3) activity. Specifically, dysfunctional insulin signaling in the brain
through the insulin receptor substrate 1 (IRS-1) à AKT pathway in AD may increase GSK-3
activity and indirectly hyperphosphorylated tau (Yarchoan & Arnold, 2014). Additionally,
chronic hyperglycemia is associated with increased advanced glycation end-products (AGEs),
which are associated with intracellular NFTs. Research suggests that glycation may promote tau
aggregation in paired helical filaments, leading to subsequent tangle formation (Munch,
Westcott, Menini, & Gugliucci, 2012). One study showed that AD and T2DM were associated
23
with decreases in brain glucose transporters (GLUT3) and protein O-GlcNAcylation, and these
decreases corresponded to increases in tau phosphorylation. These results suggest that T2DM
may impair brain glucose uptake, which causes a downregulation of the protein O-
GlcNAcylation thus promoting tau hyperphosphorylation (Liu, Liu, Grundke-Iqbal, Iqbal, &
Gong, 2009).
Given that untreated hyperglycemia, but not treated diabetes, was related to increased p-
tau, we may speculate that it is possible that T2DM medications may influence tau
phosphorylation. Diabetes medications may indirectly affect AD pathology by altering
circulating levels of glucose, insulin, inflammatory markers, and by creation of reactive oxygen
species and AGEs (Yarchoan & Arnold, 2014). The specific mechanisms regarding how
different classes of diabetes drug affects AD neuropathology are largely unknown, however there
are some clues as to how these different medications affect the brain. Metformin, a biguanide
medication, was used by about 71 percent of the treated diabetes group. Metformin is thought to
sensitize the body to insulin and may have protective effect on vascular endothelial cells
(Umegaki, 2016). It is thought that Metformin is able to cross the blood-brain barrier and may
exert anti-inflammatory and other neuroprotective effects (Labuzek et al., 2010). Sulfonylureas
were the next most popular prescribed drug in the treated diabetes group, and stimulates insulin
release to lower blood sugar. It has been posited that glimepiride may reduce synapse damage
and thus indirectly protect against cognitive decline in AD (Osborne et al., 2016; Umegaki,
2016). Lastly, thiazolidinediones improve insulin resistance and inflammation, which may
protect against AD.
Despite the potential benefits of these drugs, others have argued that T2DM medications
may have null or adverse effects on AD pathology, cognition, or cerebrovascular damage. For
24
example, one study challenged the benefits of Metformin by showing Metformin promotes Ab
production in mice (Chen et al., 2009). While several clinical trials have shown that glycemic
control is beneficial to cognition, there is an absence of large-scale randomized trials
demonstrating these effects (Umegaki, 2016). There is still much unknown about how these
medications may indirectly improve, or in some cases maybe worsen, dementia pathology.
The current study suggests that glycemic control may attenuate T2DM-related cognitive
dysfunction. Overall, there was consistent evidence for a pattern in which the euglycemia group
performed the best on neuropsychological measures, followed by the treated diabetes group, and
then the untreated hyperglycemia group. The pattern of findings in the baseline data supports our
hypothesis, and is consistent with prior studies that have shown diabetes-related cognitive
dysfunction may be attenuated by medication treatments (Gradman, Laws, Thompson, &
Reaven, 1993; Meneilly, Cheung, Tessier, Yakura, & Tuokko, 1993; Naor, Steingruber,
Westhoff, Schottenfeld-Naor, & Gries, 1997). Though T2DM treatment may ameliorate
cognitive dysfunction, treatment of T2DM does not seem to effectively resolve all cognitive
dysfunction as many studies have shown relationships between T2DM and cognitive
dysfunction. In cross-sectional studies, T2DM is most consistently associated with worse
performance in verbal memory, information processing speed and attention, and executive
functioning (Awad, Gagnon, & Messier, 2004; Reijmer et al., 2010). Therefore, the findings for
differences in the DSST and Trails A in the patient groups may be interpreted as consistent with
past findings as the DSST and Trails A test both represent tasks of processing speed, as well as
visual attention in Trails A.
In the literature, longitudinal studies have been less consistent in relating T2DM to
accelerated cognitive decline; many studies have reported a relationship between T2DM and
25
decline in one or more domains, but the identified domains have been inconsistent across studies
(Mayeda et al., 2015). The group by time interactions reported for CDR-SB and MMSE in the
current study suggest that the untreated hyperglycemia group is deteriorating more quickly in
overall functioning and overall cognitive ability. The CDR-SB score reflects the sum of
impairment scores for six domains: memory, orientation, judgment and problem solving,
community affairs, home and hobbies, and personal care. The CDR-SB has been shown to be
comparable to the global CDR score for dementia staging (O'Bryant et al., 2008), and CDR-SB
scores have demonstrated utility in diagnosing patients with mild dementia (Lynch et al., 2006).
Thus, there is evidence that the untreated hyperglycemia group may have accelerated loss in both
cognitive function and overall function. Interestingly, there were no group x time interactions for
other neuropsychological measures, which may be due to small sample size and drop out.
The cognitive findings presented in this study lend further support to the hypothesis that
untreated hyperglycemia can negatively affect cognitive functioning. Several population studies
have shown people with untreated T2DM performed worse than non-diabetic or treated diabetes
populations on cognitive measures, including the MMSE (Awad et al., 2004; Grodstein, Chen,
Wilson, & Manson, 2001; Hiltunen, Keinanen-Kiukaanniemi, & Laara, 2001; U'Ren, Riddle,
Lezak, & Bennington-Davis, 1990). Further, the results add to existing evidence that shows
T2DM medications may ameliorate cognitive decline or even improve cognition. Metformin has
been shown to decrease the risk of dementia in at least two studies (Hsu et al., 2011; Ng et al.,
2014). Sulfonylureas has also been shown to decrease risk for dementia (Hsu et al., 2011), and
improve cognition (Gradman et al., 1993). Also, several studies have shown thiazolidinediones
may ameliorate cognitive decline (Umegaki, 2016). In addition, the pattern of findings where the
treated diabetes group tended to consistently perform worse than the euglycemia group and were
26
not significantly different from the untreated hyperglycemia group on most measures is
consistent with the literature that shows T2DM negatively affects cognition, whether treated or
untreated.
Magnetic resonance imaging (MRI) has been used extensively in studying the
relationship between T2DM and dementia. MRI is an important research tool that may help
better understand the relationship between T2DM and dementia since MRI can provide in vivo
measures that are associated with both cerebrovascular lesions and neurodegeneration. Studies
have primarily focused on cerebral infarcts, WMH, and atrophy. T2DM has been fairly
consistently associated with cerebral infarcts in structural MRI studies (Manschot et al., 2006;
Moran et al., 2013; van Harten, de Leeuw, Weinstein, Scheltens, & Biessels, 2006). However,
the relationship between T2DM and other indicators of cerebrovascular damage such as cerebral
WMH (Manschot et al., 2006; Moran et al., 2013; van Harten et al., 2006) and cerebral
microbleeds (Cordonnier, Al-Shahi Salman, & Wardlaw, 2007; Moran et al., 2013; Qiu et al.,
2008) has been less clear. In addition, MRI studies have assessed the association between T2DM
and neurodegeneration by measuring atrophy. Many studies have shown T2DM is associated
with global atrophy (Jongen & Biessels, 2008; van Harten et al., 2006). Furthermore, there has
been ample evidence for regional atrophy in structures implicated in the AD such as the medial
temporal lobe, including the hippocampus and amygdala (den Heijer et al., 2003; Kamiyama et
al., 2010; Korf et al., 2007). Though there is much support for the association between T2DM
and atrophy, it is important to note that the observed atrophy in these studies could be due to
either cerebrovascular disease, AD, or both (Jack, 2011).
In the present study, baseline MRI analyses did not show strong evidence of a
relationship between T2DM and cerebrovascular damage as assessed by WMH volume.
27
However, a pairwise comparison did show that the treated diabetes group had marginally greater
WMH volume than the euglycemia group. While some studies have failed to find a relationship
between T2DM and WMH (den Heijer et al., 2003; Schmidt et al., 2004), other studies have
reported relationships between T2DM and increased WMH measures (Gouw et al., 2008; Jongen
et al., 2007; Saczynski et al., 2009; Taylor et al., 2003), and deep WMH more specifically
(Manschot et al., 2006; van Harten, Oosterman, Potter van Loon, Scheltens, & Weinstein, 2007).
Additionally, one study showed that glycated hemoglobin A1c was positively associated with
WMH volume (Murray et al., 2005). One autopsy study that compared brains of dementia
patients with and without diabetes found that dementia patients with T2DM had significantly
more microvascular infarcts compared to non-diabetic dementia patients, and treated diabetic
patients had more deep microvascular infarcts than untreated diabetic patients. On the other
hand, the treated diabetics had lower amyloid plaque load than the untreated diabetics. The
treated and untreated diabetic cases with dementia did not differ on measures of cerebral amyloid
angiopathy severity or atherosclerosis (Sonnen et al., 2009). It should be noted that one problem
with dementia pathology studies is that we often see less AD pathology in populations with
vascular disease because of the fact that vascular disease lowers the threshold at which dementia
manifests clinically. With respect to our findings, while there were no statistically significant
findings with respect to cerebrovascular damage in our baseline data, the treated diabetes group
showed the most cerebrovascular damage and the euglycemia group showed the least damage.
This pattern is both consistent with our hypotheses and with the finding that the treated diabetes
group had more vascular risk factors than the other groups. It is unsurprising that the treated
diabetes group had larger WMH volumes given that they also had more hypertension and
dyslipidemia, which can also cause cerebrovascular damage. The lack of statistically significant
28
findings may be due to the differing methods in measuring WMH volume, and the need to
control for that in the statistical model. It is unclear whether the treatment of diabetes itself
causes cerebrovascular injury, or if there is a confound by indication where diabetes treatment is
indicative of greater severity of illness and thus we would expect more vascular damage. Though
untreated hyperglycemia certainly appears to negatively affect neural processes, it is not
surprising that the untreated hyperglycemia group did not differ from the other groups as there is
some evidence that disease duration of diabetes is positively correlated with WMH volume
(Saczynski et al., 2009; van Harten et al., 2007). Though there is no data regarding disease
duration for this dataset, it is very likely that the treated diabetes group had a longer history of
hyperglycemia compared to the untreated hyperglycemia group. It is also possible that the treated
diabetes group initially had more severe hyperglycemia than the untreated hyperglycemia group,
which prompted medical intervention. Alternatively, it is possible that the small sample size of
the untreated hyperglycemia group resulted in insufficient power to show a relationship.
Looking at WMH volume progression, the results suggest diabetes medications may
protect against WMH progression through glycemic control. Unlike the euglycemia group and
the untreated hyperglycemia group, the treated diabetes group did not show WMH progression
over the two-year period. The treatment of T2DM may be protective against further small vessel
damage. Other prospective cohort studies have shown mixed findings regarding T2DM and
WMH progression with some studies citing a relationship between T2DM and WMH
progression (Gouw et al., 2008; Taylor et al., 2003), and other studies finding no difference in
WMH changes between diabetic and non-diabetic populations (de Bresser et al., 2010; van
Elderen et al., 2010). The conflicting findings in the literature suggest that this topic deserves
29
more attention, and future research should consider the role of medication in the relationship
between T2DM and WMH progression.
There were no baseline group differences or group by time interactions for any of the
volumes examined in this study. This was somewhat surprising given the strong relationship
between T2DM and atrophy, but it is important to recognize the limited nature of the analyses. In
an effort to only focus on brain areas known to be affected early in AD, analyses only looked at
two global brain measures (whole brain volume, ventricle volume), and three regional volumes
(hippocampus, entorhinal, middle temporal).
As with any study, there were several limitations in the current study. The most important
limitation of the current study is confound by indication. Confound by indication is a bias where
the severity of the condition (e.g. hyperglycemia) is likely associated with both the outcome
measures, and treatment group. This makes it difficult to disentangle the effects of a treatment
(e.g. diabetes medication) with those of the severity of illness (Skelly, Dettori, & Brodt, 2012).
The use of diabetes medications may be a risk indicator for the severity of hyperglycemia and the
presence of other vascular risk factors. However, if there had been confound by indication, we
would predict that the treated diabetes groups would have worse CSF biomarker levels, cognitive
performance, or atrophy compared to the untreated hyperglycemia group given that medication
users had potentially more severe hyperglycemia. The generalizability of this study may be
compromised due to the nature of the ADNI participant selection and potential recruitment
biases specific to each study center. Participants were screened for many vascular issues, which
means that the sample may have less vascular risk factors than the general population. At the
same time, this is a strength of the study because we can better isolate how T2DM and
hyperglycemia are related to markers of aging and neurodegenerative disease independent of
30
other vascular risk factors. Another limitation is the definition of the different groups based on
limited data. Groups were assigned based only on fasting blood glucose and self-reported
medication use at baseline. Fasting blood glucose was not available after baseline thus we cannot
be certain which, if any, euglycemia participants became hyperglycemic over the two-years or if
any untreated hyperglycemia participants lowered their glucose levels to normal levels. In
addition, duration of illness data was unavailable, which we might expect to influence the results
as longer duration of T2DM is associated with increased risk of dementia (Bruce, Davis, Casey,
Starkstein, Clarnette, Almeida, et al., 2008; Bruce, Davis, Casey, Starkstein, Clarnette, Foster, et
al., 2008; Ott et al., 1999; Parikh et al., 2011; Peila et al., 2002). Microinfarct data were also
unavailable, which has been shown to be related to T2DM, and would have been useful in further
exploration how medication may affect cerebrovascular damage (Sonnen et al., 2009). Another
limitation is that we cannot draw any causal conclusions because this was not a clinical trial. We
can only infer the role of medication based on differences between the untreated hyperglycemia
and treated diabetes groups. Lastly, the study may have been limited statistically because of both
small sample sizes and missing data. To mitigate the problem of missing data, mixed models
were used to conduct the longitudinal analyses.
Despite several limitations in the study, there were also several strengths. To our
knowledge, this is potentially the first study to investigate the role of T2DM medication on CSF
AD biomarkers in a human population. The majority of research on this topic has been limited to
animal research. Another strength of the current study was the use of several different outcome
measures. While the relationship between T2DM medication management and cognition and risk
for dementia has received attention, few, if any, of these studies have used CSF biomarkers, MRI
31
measures, or both. Lastly, the longitudinal design of this study allowed us to examine how
glycemic control influences cognition, cerebrovascular damage, and brain atrophy over time.
In conclusion, the present study demonstrated that untreated hyperglycemia was
associated with increased tau pathology, and this group tended to perform worse than the other
groups on cognitive measures lending further support for a relationship between glycemic
control and cognition. The untreated hyperglycemia group also showed greater cognitive and
functional decline and more cerebrovascular damage progression compared to the treated
diabetes group. Given previous evidence that glycemic control and T2DM treatment can
decrease risk for dementia and ameliorate cognitive decline, it is important that we continue to
explore how T2DM medications may indirectly influence dementia pathologies using various
methods in both animal and human research.
32
References
Ahtiluoto, S., Polvikoski, T., Peltonen, M., Solomon, A., Tuomilehto, J., Winblad, B., . . .
Kivipelto, M. (2010). Diabetes, Alzheimer disease, and vascular dementia: a population-
based neuropathologic study. Neurology, 75(13), 1195-1202.
doi:10.1212/WNL.0b013e3181f4d7f8
Alafuzoff, I., Aho, L., Helisalmi, S., Mannermaa, A., & Soininen, H. (2009). Beta-amyloid
deposition in brains of subjects with diabetes. Neuropathol Appl Neurobiol, 35(1), 60-68.
doi:10.1111/j.1365-2990.2008.00948.x
Alberti, G., Zimmet, P., Shaw, J., Bloomgarden, Z., Kaufman, F., & Silink, M. (2004). Type 2
Diabetes in the Young: The Evolving Epidemic. Diabetes Care, 27(7), 1798-1811.
doi:10.2337/diacare.27.7.1798
Andreasen, N., Hesse, C., Davidsson, P., Minthon, L., Wallin, A., Winblad, B., . . . Blennow, K.
(1999). Cerebrospinal fluid beta-amyloid(1-42) in Alzheimer disease: differences
between early- and late-onset Alzheimer disease and stability during the course of
disease. Arch Neurol, 56(6), 673-680.
Apostolova, L. G., Hwang, K. S., Andrawis, J. P., Green, A. E., Babakchanian, S., Morra, J. H., .
. . Thompson, P. M. (2010). 3D PIB and CSF biomarker associations with hippocampal
atrophy in ADNI subjects. Neurobiol Aging, 31(8), 1284-1303.
doi:10.1016/j.neurobiolaging.2010.05.003
Arvanitakis, Z., Schneider, J. A., Wilson, R. S., Li, Y., Arnold, S. E., Wang, Z., & Bennett, D. A.
(2006). Diabetes is related to cerebral infarction but not to AD pathology in older
persons. Neurology, 67(11), 1960-1965. doi:10.1212/01.wnl.0000247053.45483.4e
33
Awad, N., Gagnon, M., & Messier, C. (2004). The relationship between impaired glucose
tolerance, type 2 diabetes, and cognitive function. J Clin Exp Neuropsychol, 26(8), 1044-
1080. doi:10.1080/13803390490514875
Biessels, G. J., Staekenborg, S., Brunner, E., Brayne, C., & Scheltens, P. (2006). Risk of
dementia in diabetes mellitus: a systematic review. Lancet Neurol, 5(1), 64-74.
doi:10.1016/s1474-4422(05)70284-2
Braak, H., & Braak, E. (1991). Neuropathological stageing of Alzheimer-related changes. Acta
Neuropathol, 82(4), 239-259. doi:10.1007/BF00308809
Bruce, D. G., Davis, W. A., Casey, G. P., Starkstein, S. E., Clarnette, R. M., Almeida, O. P., &
Davis, T. M. (2008). Predictors of cognitive decline in older individuals with diabetes.
Diabetes Care, 31(11), 2103-2107. doi:10.2337/dc08-0562
Bruce, D. G., Davis, W. A., Casey, G. P., Starkstein, S. E., Clarnette, R. M., Foster, J. K., . . .
Davis, T. M. (2008). Predictors of cognitive impairment and dementia in older people
with diabetes. Diabetologia, 51(2), 241-248. doi:10.1007/s00125-007-0894-7
Buerger, K., Ewers, M., Pirttila, T., Zinkowski, R., Alafuzoff, I., Teipel, S. J., . . . Hampel, H.
(2006). CSF phosphorylated tau protein correlates with neocortical neurofibrillary
pathology in Alzheimer's disease. Brain, 129(Pt 11), 3035-3041.
doi:10.1093/brain/awl269
Chen, Y., Zhou, K., Wang, R., Liu, Y., Kwak, Y. D., Ma, T., . . . Liao, F. F. (2009). Antidiabetic
drug metformin (GlucophageR) increases biogenesis of Alzheimer's amyloid peptides via
up-regulating BACE1 transcription. Proc Natl Acad Sci U S A, 106(10), 3907-3912.
doi:10.1073/pnas.0807991106
34
Cheng, G., Huang, C., Deng, H., & Wang, H. (2012). Diabetes as a risk factor for dementia and
mild cognitive impairment: a meta-analysis of longitudinal studies. Intern Med J, 42(5),
484-491. doi:10.1111/j.1445-5994.2012.02758.x
Cordonnier, C., Al-Shahi Salman, R., & Wardlaw, J. (2007). Spontaneous brain microbleeds:
systematic review, subgroup analyses and standards for study design and reporting.
Brain, 130(Pt 8), 1988-2003. doi:10.1093/brain/awl387
Craft, S. (2009). The role of metabolic disorders in Alzheimer disease and vascular dementia:
two roads converged. Arch Neurol, 66(3), 300-305. doi:10.1001/archneurol.2009.27
Craft, S., & Watson, G. S. (2004). Insulin and neurodegenerative disease: shared and specific
mechanisms. Lancet Neurol, 3(3), 169-178. doi:10.1016/s1474-4422(04)00681-7
de Bresser, J., Tiehuis, A. M., van den Berg, E., Reijmer, Y. D., Jongen, C., Kappelle, L. J., . . .
Biessels, G. J. (2010). Progression of cerebral atrophy and white matter hyperintensities
in patients with type 2 diabetes. Diabetes Care, 33(6), 1309-1314. doi:10.2337/dc09-
1923
de la Monte, S. M., & Wands, J. R. (2008). Alzheimer's Disease Is Type 3 Diabetes–Evidence
Reviewed. J Diabetes Sci Technol, 2(6), 1101-1113.
den Heijer, T., Vermeer, S. E., van Dijk, E. J., Prins, N. D., Koudstaal, P. J., Hofman, A., &
Breteler, M. M. (2003). Type 2 diabetes and atrophy of medial temporal lobe structures
on brain MRI. Diabetologia, 46(12), 1604-1610. doi:10.1007/s00125-003-1235-0
Exalto, L. G., Whitmer, R. A., Kappele, L. J., & Biessels, G. J. (2012). An update on type 2
diabetes, vascular dementia and Alzheimer's disease. Exp Gerontol, 47(11), 858-864.
doi:10.1016/j.exger.2012.07.014
35
Fagan, A. M., & Holtzman, D. M. (2010). Cerebrospinal fluid biomarkers of Alzheimer’s
disease. Biomark Med, 4(1), 51-63. doi:10.2217/bmm.09.83
Gao, L., Matthews, F. E., Sargeant, L. A., & Brayne, C. (2008). An investigation of the
population impact of variation in HbA1c levels in older people in England and Wales:
from a population based multi-centre longitudinal study. BMC Public Health, 8, 54.
doi:10.1186/1471-2458-8-54
Gouw, A. A., van der Flier, W. M., Fazekas, F., van Straaten, E. C., Pantoni, L., Poggesi, A., . . .
Barkhof, F. (2008). Progression of white matter hyperintensities and incidence of new
lacunes over a 3-year period: the Leukoaraiosis and Disability study. Stroke, 39(5), 1414-
1420. doi:10.1161/strokeaha.107.498535
Gradman, T. J., Laws, A., Thompson, L. W., & Reaven, G. M. (1993). Verbal learning and/or
memory improves with glycemic control in older subjects with non-insulin-dependent
diabetes mellitus. J Am Geriatr Soc, 41(12), 1305-1312.
Grodstein, F., Chen, J., Wilson, R. S., & Manson, J. E. (2001). Type 2 diabetes and cognitive
function in community-dwelling elderly women. Diabetes Care, 24(6), 1060-1065.
Hachinski, V. C., Iliff, L. D., Zilhka, E., Du Boulay, G. H., McAllister, V. L., Marshall, J., . . .
Symon, L. (1975). Cerebral blood flow in dementia. Arch Neurol, 32(9), 632-637.
Hassing, L. B., Johansson, B., Nilsson, S. E., Berg, S., Pedersen, N. L., Gatz, M., & McClearn,
G. (2002). Diabetes mellitus is a risk factor for vascular dementia, but not for Alzheimer's
disease: a population-based study of the oldest old. Int Psychogeriatr, 14(3), 239-248.
Hayden, K. M., Zandi, P. P., Lyketsos, C. G., Khachaturian, A. S., Bastian, L. A., Charoonruk,
G., . . . Welsh-Bohmer, K. A. (2006). Vascular risk factors for incident Alzheimer disease
36
and vascular dementia: the Cache County study. Alzheimer Dis Assoc Disord, 20(2), 93-
100. doi:10.1097/01.wad.0000213814.43047.86
Heitner, J., & Dickson, D. (1997). Diabetics do not have increased Alzheimer-type pathology
compared with age-matched control subjects. A retrospective postmortem
immunocytochemical and histofluorescent study. Neurology, 49(5), 1306-1311.
Hiltunen, L. A., Keinanen-Kiukaanniemi, S. M., & Laara, E. M. (2001). Glucose tolerance and
cognitive impairment in an elderly population. Public Health, 115(3), 197-200.
doi:10.1038/sj/ph/1900758
Ho, L., Qin, W., Pompl, P. N., Xiang, Z., Wang, J., Zhao, Z., . . . Pasinetti, G. M. (2004). Diet-
induced insulin resistance promotes amyloidosis in a transgenic mouse model of
Alzheimer's disease. Faseb j, 18(7), 902-904. doi:10.1096/fj.03-0978fje
Hsu, C. C., Wahlqvist, M. L., Lee, M. S., & Tsai, H. N. (2011). Incidence of dementia is
increased in type 2 diabetes and reduced by the use of sulfonylureas and metformin. J
Alzheimers Dis, 24(3), 485-493. doi:10.3233/jad-2011-101524
Iqbal, K., Alonso Adel, C., Chen, S., Chohan, M. O., El-Akkad, E., Gong, C. X., . . . Grundke-
Iqbal, I. (2005). Tau pathology in Alzheimer disease and other tauopathies. Biochim
Biophys Acta, 1739(2-3), 198-210. doi:10.1016/j.bbadis.2004.09.008
Irie, F., Fitzpatrick, A. L., Lopez, O. L., Kuller, L. H., Peila, R., Newman, A. B., & Launer, L. J.
(2008). Enhanced risk for Alzheimer disease in persons with type 2 diabetes and APOE
epsilon4: the Cardiovascular Health Study Cognition Study. Arch Neurol, 65(1), 89-93.
doi:10.1001/archneurol.2007.29
Jack, C. R., Jr. (2011). Alliance for aging research AD biomarkers work group: structural MRI.
Neurobiol Aging, 32 Suppl 1, S48-57. doi:10.1016/j.neurobiolaging.2011.09.011
37
Jack, C. R., Jr., Bernstein, M. A., Fox, N. C., Thompson, P., Alexander, G., Harvey, D., . . .
Weiner, M. W. (2008). The Alzheimer's Disease Neuroimaging Initiative (ADNI): MRI
methods. J Magn Reson Imaging, 27(4), 685-691. doi:10.1002/jmri.21049
Jongen, C., & Biessels, G. J. (2008). Structural brain imaging in diabetes: a methodological
perspective. Eur J Pharmacol, 585(1), 208-218. doi:10.1016/j.ejphar.2007.11.085
Jongen, C., van der Grond, J., Kappelle, L. J., Biessels, G. J., Viergever, M. A., & Pluim, J. P.
(2007). Automated measurement of brain and white matter lesion volume in type 2
diabetes mellitus. Diabetologia, 50(7), 1509-1516. doi:10.1007/s00125-007-0688-y
Kamiyama, K., Wada, A., Sugihara, M., Kurioka, S., Hayashi, K., Hayashi, T., . . . Kitagaki, H.
(2010). Potential hippocampal region atrophy in diabetes mellitus type 2: a voxel-based
morphometry VSRAD study. Jpn J Radiol, 28(4), 266-272. doi:10.1007/s11604-009-
0416-2
Kim, B., Backus, C., Oh, S., Hayes, J. M., & Feldman, E. L. (2009). Increased tau
phosphorylation and cleavage in mouse models of type 1 and type 2 diabetes.
Endocrinology, 150(12), 5294-5301. doi:10.1210/en.2009-0695
Kim, B., Sullivan, K. A., Backus, C., & Feldman, E. L. (2011). Cortical neurons develop insulin
resistance and blunted Akt signaling: a potential mechanism contributing to enhanced
ischemic injury in diabetes. Antioxid Redox Signal, 14(10), 1829-1839.
doi:10.1089/ars.2010.3816
Kohjima, M., Sun, Y., & Chan, L. (2010). Increased food intake leads to obesity and insulin
resistance in the tg2576 Alzheimer's disease mouse model. Endocrinology, 151(4), 1532-
1540. doi:10.1210/en.2009-1196
38
Korf, E. S., van Straaten, E. C., de Leeuw, F. E., van der Flier, W. M., Barkhof, F., Pantoni, L., .
. . Scheltens, P. (2007). Diabetes mellitus, hypertension and medial temporal lobe
atrophy: the LADIS study. Diabet Med, 24(2), 166-171. doi:10.1111/j.1464-
5491.2007.02049.x
Labuzek, K., Suchy, D., Gabryel, B., Bielecka, A., Liber, S., & Okopien, B. (2010).
Quantification of metformin by the HPLC method in brain regions, cerebrospinal fluid
and plasma of rats treated with lipopolysaccharide. Pharmacol Rep, 62(5), 956-965.
Li, Z. G., Zhang, W., & Sima, A. A. (2007). Alzheimer-like changes in rat models of
spontaneous diabetes. Diabetes, 56(7), 1817-1824. doi:10.2337/db07-0171
Liu, Y., Liu, F., Grundke-Iqbal, I., Iqbal, K., & Gong, C. X. (2009). Brain glucose transporters,
O-GlcNAcylation and phosphorylation of tau in diabetes and Alzheimer's disease. J
Neurochem, 111(1), 242-249. doi:10.1111/j.1471-4159.2009.06320.x
Lu, F. P., Lin, K. P., & Kuo, H. K. (2009). Diabetes and the risk of multi-system aging
phenotypes: a systematic review and meta-analysis. PLoS One, 4(1), e4144.
doi:10.1371/journal.pone.0004144
Lynch, C. A., Walsh, C., Blanco, A., Moran, M., Coen, R. F., Walsh, J. B., & Lawlor, B. A.
(2006). The clinical dementia rating sum of box score in mild dementia. Dement Geriatr
Cogn Disord, 21(1), 40-43. doi:10.1159/000089218
MacKnight, C., Rockwood, K., Awalt, E., & McDowell, I. (2002). Diabetes mellitus and the risk
of dementia, Alzheimer's disease and vascular cognitive impairment in the Canadian
Study of Health and Aging. Dement Geriatr Cogn Disord, 14(2), 77-83. doi:64928
39
Manschot, S. M., Brands, A. M., van der Grond, J., Kessels, R. P., Algra, A., Kappelle, L. J., &
Biessels, G. J. (2006). Brain magnetic resonance imaging correlates of impaired
cognition in patients with type 2 diabetes. Diabetes, 55(4), 1106-1113.
Mayeda, E. R., Whitmer, R. A., & Yaffe, K. (2015). Diabetes and Cognition. Clin Geriatr Med,
31(1), 101-ix. doi:10.1016/j.cger.2014.08.021
McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., & Stadlan, E. M. (1984).
Clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA Work Group
under the auspices of Department of Health and Human Services Task Force on
Alzheimer's Disease. Neurology, 34(7), 939-944.
Meneilly, G. S., Cheung, E., Tessier, D., Yakura, C., & Tuokko, H. (1993). The effect of
improved glycemic control on cognitive functions in the elderly patient with diabetes. J
Gerontol, 48(4), M117-121.
Moran, C., Beare, R., Phan, T. G., Bruce, D. G., Callisaya, M. L., & Srikanth, V. (2015). Type 2
diabetes mellitus and biomarkers of neurodegeneration. Neurology.
doi:10.1212/wnl.0000000000001982
Moran, C., Phan, T. G., Chen, J., Blizzard, L., Beare, R., Venn, A., . . . Srikanth, V. (2013).
Brain atrophy in type 2 diabetes: regional distribution and influence on cognition.
Diabetes Care, 36(12), 4036-4042. doi:10.2337/dc13-0143
Morris, J. C. (1997). Clinical dementia rating: a reliable and valid diagnostic and staging
measure for dementia of the Alzheimer type. Int Psychogeriatr, 9 Suppl 1, 173-176;
discussion 177-178.
Mueller, S. G., Weiner, M. W., Thal, L. J., Petersen, R. C., Jack, C. R., Jagust, W., . . . Beckett,
L. (2005). Ways toward an early diagnosis in Alzheimer's disease: the Alzheimer's
40
Disease Neuroimaging Initiative (ADNI). Alzheimers Dement, 1(1), 55-66.
doi:10.1016/j.jalz.2005.06.003
Munch, G., Westcott, B., Menini, T., & Gugliucci, A. (2012). Advanced glycation endproducts
and their pathogenic roles in neurological disorders. Amino Acids, 42(4), 1221-1236.
doi:10.1007/s00726-010-0777-y
Murray, A. D., Staff, R. T., Shenkin, S. D., Deary, I. J., Starr, J. M., & Whalley, L. J. (2005).
Brain white matter hyperintensities: relative importance of vascular risk factors in
nondemented elderly people. Radiology, 237(1), 251-257.
doi:10.1148/radiol.2371041496
Naor, M., Steingruber, H. J., Westhoff, K., Schottenfeld-Naor, Y., & Gries, A. F. (1997).
Cognitive function in elderly non-insulin-dependent diabetic patients before and after
inpatient treatment for metabolic control. J Diabetes Complications, 11(1), 40-46.
Nation, D. A., Edmonds, E. C., Bangen, K. J., Delano-Wood, L., Scanlon, B. K., Han, S. D., . . .
Bondi, M. W. (2015). Pulse pressure in relation to tau-mediated neurodegeneration,
cerebral amyloidosis, and progression to dementia in very old adults. JAMA Neurol,
72(5), 546-553. doi:10.1001/jamaneurol.2014.4477
Nelson, P. T., Smith, C. D., Abner, E. A., Schmitt, F. A., Scheff, S. W., Davis, G. J., . . .
Markesbery, W. R. (2009). Human cerebral neuropathology of Type 2 diabetes mellitus.
Biochim Biophys Acta, 1792(5), 454-469. doi:10.1016/j.bbadis.2008.08.005
Ng, T. P., Feng, L., Yap, K. B., Lee, T. S., Tan, C. H., & Winblad, B. (2014). Long-term
metformin usage and cognitive function among older adults with diabetes. J Alzheimers
Dis, 41(1), 61-68. doi:10.3233/jad-131901
41
O'Bryant, S. E., Waring, S. C., Cullum, C. M., Hall, J., Lacritz, L., Massman, P. J., . . . Doody,
R. (2008). Staging Dementia Using Clinical Dementia Rating Scale Sum of Boxes
Scores: A Texas Alzheimer's Research Consortium Study. Arch Neurol, 65(8), 1091-
1095. doi:10.1001/archneur.65.8.1091
Osborne, C., West, E., Nolan, W., McHale-Owen, H., Williams, A., & Bate, C. (2016).
Glimepiride protects neurons against amyloid-beta-induced synapse damage.
Neuropharmacology, 101, 225-236. doi:10.1016/j.neuropharm.2015.09.030
Ott, A., Stolk, R. P., van Harskamp, F., Pols, H. A., Hofman, A., & Breteler, M. M. (1999).
Diabetes mellitus and the risk of dementia: The Rotterdam Study. Neurology, 53(9),
1937-1942.
Paneni, F., Beckman, J. A., Creager, M. A., & Cosentino, F. (2013). Diabetes and vascular
disease: pathophysiology, clinical consequences, and medical therapy: part I. Eur Heart
J, 34(31), 2436-2443. doi:10.1093/eurheartj/eht149
Parikh, N. M., Morgan, R. O., Kunik, M. E., Chen, H., Aparasu, R. R., Yadav, R. K., . . .
Johnson, M. L. (2011). Risk factors for dementia in patients over 65 with diabetes. Int J
Geriatr Psychiatry, 26(7), 749-757. doi:10.1002/gps.2604
Park, S. A. (2011). A common pathogenic mechanism linking type-2 diabetes and Alzheimer's
disease: evidence from animal models. J Clin Neurol, 7(1), 10-18.
doi:10.3988/jcn.2011.7.1.10
Peila, R., Rodriguez, B. L., & Launer, L. J. (2002). Type 2 diabetes, APOE gene, and the risk for
dementia and related pathologies: The Honolulu-Asia Aging Study. Diabetes, 51(4),
1256-1262.
42
Petersen, R. C., Aisen, P. S., Beckett, L. A., Donohue, M. C., Gamst, A. C., Harvey, D. J., . . .
Weiner, M. W. (2010). Alzheimer's Disease Neuroimaging Initiative (ADNI): clinical
characterization. Neurology, 74(3), 201-209. doi:10.1212/WNL.0b013e3181cb3e25
Profenno, L. A., Porsteinsson, A. P., & Faraone, S. V. (2010). Meta-analysis of Alzheimer's
disease risk with obesity, diabetes, and related disorders. Biol Psychiatry, 67(6), 505-512.
doi:10.1016/j.biopsych.2009.02.013
Qiu, C., Cotch, M. F., Sigurdsson, S., Garcia, M., Klein, R., Jonasson, F., . . . Launer, L. J.
(2008). Retinal and cerebral microvascular signs and diabetes: the age, gene/environment
susceptibility-Reykjavik study. Diabetes, 57(6), 1645-1650. doi:10.2337/db07-1455
Reijmer, Y. D., van den Berg, E., Ruis, C., Kappelle, L. J., & Biessels, G. J. (2010). Cognitive
dysfunction in patients with type 2 diabetes. Diabetes Metab Res Rev, 26(7), 507-519.
doi:10.1002/dmrr.1112
Reitz, C. (2012). Alzheimer's disease and the amyloid cascade hypothesis: a critical review. Int J
Alzheimers Dis, 2012, 369808. doi:10.1155/2012/369808
Saczynski, J. S., Siggurdsson, S., Jonsson, P. V., Eiriksdottir, G., Olafsdottir, E., Kjartansson, O.,
. . . Launer, L. J. (2009). Glycemic status and brain injury in older individuals: the age
gene/environment susceptibility-Reykjavik study. Diabetes Care, 32(9), 1608-1613.
doi:10.2337/dc08-2300
Schmidt, R., Launer, L. J., Nilsson, L. G., Pajak, A., Sans, S., Berger, K., . . . Hofman, A. (2004).
Magnetic resonance imaging of the brain in diabetes: the Cardiovascular Determinants of
Dementia (CASCADE) Study. Diabetes, 53(3), 687-692. doi:10.2337/diabetes.53.3.687
43
Schwarz, C., Fletcher, E., DeCarli, C., & Carmichael, O. (2009). Fully-automated white matter
hyperintensity detection with anatomical prior knowledge and without FLAIR. Inf
Process Med Imaging, 21, 239-251.
Shaw, L. M., Vanderstichele, H., Knapik-Czajka, M., Clark, C. M., Aisen, P. S., Petersen, R. C.,
. . . Trojanowski, J. Q. (2009). Cerebrospinal fluid biomarker signature in Alzheimer's
disease neuroimaging initiative subjects. Ann Neurol, 65(4), 403-413.
doi:10.1002/ana.21610
Shorr, R. I., de Rekeneire, N., Resnick, H. E., Yaffe, K., Somes, G. W., Kanaya, A. M., . . .
Harris, T. B. (2006). Glycemia and cognitive function in older adults using glucose-
lowering drugs. J Nutr Health Aging, 10(4), 297-301.
Skelly, A. C., Dettori, J. R., & Brodt, E. D. (2012). Assessing bias: the importance of considering
confounding. Evid Based Spine Care J, 3(1), 9-12. doi:10.1055/s-0031-1298595
Sonnen, J. A., Larson, E. B., Brickell, K., Crane, P. K., Woltjer, R., Montine, T. J., & Craft, S.
(2009). Different patterns of cerebral injury in dementia with or without diabetes. Arch
Neurol, 66(3), 315-322. doi:10.1001/archneurol.2008.579
Standards of medical care in diabetes--2012. (2012). Diabetes Care, 35 Suppl 1, S11-63.
doi:10.2337/dc12-s011
Tapiola, T., Alafuzoff, I., Herukka, S. K., Parkkinen, L., Hartikainen, P., Soininen, H., & Pirttila,
T. (2009). Cerebrospinal fluid {beta}-amyloid 42 and tau proteins as biomarkers of
Alzheimer-type pathologic changes in the brain. Arch Neurol, 66(3), 382-389.
doi:10.1001/archneurol.2008.596
Taylor, W. D., MacFall, J. R., Provenzale, J. M., Payne, M. E., McQuoid, D. R., Steffens, D. C.,
& Krishnan, K. R. (2003). Serial MR imaging of volumes of hyperintense white matter
44
lesions in elderly patients: correlation with vascular risk factors. AJR Am J Roentgenol,
181(2), 571-576. doi:10.2214/ajr.181.2.1810571
U'Ren, R. C., Riddle, M. C., Lezak, M. D., & Bennington-Davis, M. (1990). The mental
efficiency of the elderly person with type II diabetes mellitus. J Am Geriatr Soc, 38(5),
505-510.
Umegaki, H. (2016). Therapeutic Potential of Antidiabetic Medications in the Treatment of
Cognitive Dysfunction and Dementia. Drugs Aging, 33(6), 399-409. doi:10.1007/s40266-
016-0375-0
van den Berg, E., Kloppenborg, R. P., Kessels, R. P., Kappelle, L. J., & Biessels, G. J. (2009).
Type 2 diabetes mellitus, hypertension, dyslipidemia and obesity: A systematic
comparison of their impact on cognition. Biochim Biophys Acta, 1792(5), 470-481.
doi:10.1016/j.bbadis.2008.09.004
van Elderen, S. G., de Roos, A., de Craen, A. J., Westendorp, R. G., Blauw, G. J., Jukema, J. W.,
. . . van der Grond, J. (2010). Progression of brain atrophy and cognitive decline in
diabetes mellitus: a 3-year follow-up. Neurology, 75(11), 997-1002.
doi:10.1212/WNL.0b013e3181f25f06
van Harten, B., de Leeuw, F. E., Weinstein, H. C., Scheltens, P., & Biessels, G. J. (2006). Brain
imaging in patients with diabetes: a systematic review. Diabetes Care, 29(11), 2539-
2548. doi:10.2337/dc06-1637
van Harten, B., Oosterman, J. M., Potter van Loon, B. J., Scheltens, P., & Weinstein, H. C.
(2007). Brain lesions on MRI in elderly patients with type 2 diabetes mellitus. Eur
Neurol, 57(2), 70-74. doi:10.1159/000098054
45
Wang, Y. W., Sun, G. D., Sun, J., Liu, S. J., Wang, J., Xu, X. H., & Miao, L. N. (2013).
Spontaneous type 2 diabetic rodent models. J Diabetes Res, 2013, 401723.
doi:10.1155/2013/401723
Weiner, M. W., Aisen, P. S., Jack, C. R., Jr., Jagust, W. J., Trojanowski, J. Q., Shaw, L., . . .
Schmidt, M. (2010). The Alzheimer's disease neuroimaging initiative: progress report and
future plans. Alzheimers Dement, 6(3), 202-211.e207. doi:10.1016/j.jalz.2010.03.007
Wilson, P. W., D'Agostino, R. B., Levy, D., Belanger, A. M., Silbershatz, H., & Kannel, W. B.
(1998). Prediction of coronary heart disease using risk factor categories. Circulation,
97(18), 1837-1847.
Wolf, P. A., D'Agostino, R. B., Belanger, A. J., & Kannel, W. B. (1991). Probability of stroke: a
risk profile from the Framingham Study. Stroke, 22(3), 312-318.
Xu, W. L., von Strauss, E., Qiu, C. X., Winblad, B., & Fratiglioni, L. (2009). Uncontrolled
diabetes increases the risk of Alzheimer's disease: a population-based cohort study.
Diabetologia, 52(6), 1031-1039. doi:10.1007/s00125-009-1323-x
Yarchoan, M., & Arnold, S. E. (2014). Repurposing diabetes drugs for brain insulin resistance in
Alzheimer disease. Diabetes, 63(7), 2253-2261. doi:10.2337/db14-0287
Zhao, W. Q., & Townsend, M. (2009). Insulin resistance and amyloidogenesis as common
molecular foundation for type 2 diabetes and Alzheimer's disease. Biochim Biophys Acta,
1792(5), 482-496. doi:10.1016/j.bbadis.2008.10.014
46
FIGURES
Figure 1. Logistic regression analyses showed that the untreated hyperglycemia group had
significantly greater proportions of p-tau positive individuals compared to the euglycemia group
(p = 0.028).
0
10
20
30
40
50
60
70
80
90
P-Tau + P-Tau -
Percent
P-Tau Profiles
Group Differences in P-Tau Profiles
Euglycemia
Treated Diabetes
Untreated
Hyperglycemia
47
Figure 2. Logistic regression analyses showed that the untreated hyperglycemia group had
significantly greater proportions of tau positive individuals compared to the euglycemia group (p
= 0.018).
0
10
20
30
40
50
60
70
80
Tau + Tau -
Percent
Tau Profiles
Group Differences in Tau Profiles
Euglycemia
Treated Diabetes
Untreated Hyperglycemia
*
48
Figure 3. Changes in CDR-SB scores over two years. There was a significant group x time
interaction on CDR-SB score (p = 0.001) All groups showed significant decline (increased scores
on the CDR-SB) between baseline and month 24 (Euglycemia: b = 0.061, p < 0.001; Treated
Diabetes: b = 0.103, p < 0.001; Untreated Hyperglycemia: b = 0.143, p < 0.001). The untreated
hyperglycemia group showed accelerated decline compared to the euglycemia group (p = 0.001),
but not the treated diabetes group (p = 0.213). At month 24, the euglycemia group had
significantly better scores compared to both the untreated hyperglycemia group (p = 0.004) and
the treated diabetes group (p = 0.019). There was no difference in rate of decline between the
euglycemia and treated diabetes group (p = 0.102).
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
Baseline Month 12 Month 24
Log Transformed CDR-SB Scores
Changes in CDR-SB Scores
Euglycemia
Treated Diabetes
Untreated Hyperglycemia
49
Figure 4. Changes in MMSE scores over two years. There was a marginally significant group x
time interaction for MMSE (p = 0.058). All groups showed significant decline in MMSE scores
over the two-year period (Euglycemia: b = -0.750, p < 0.001, Treated Diabetes: b = -1.310, p <
0.001; Untreated Hyperglycemia: b = -1.465, p < 0.001). The rate of decline for the untreated
hyperglycemia group was significantly greater compared to the euglycemia group (p = 0.041),
but not the treated diabetes group (p = 0.734). At month 24, the untreated hyperglycemia group
had significantly worse performance on the MMSE compared to the euglycemia group (p <
0.001), but not the treated diabetes group (p = 0.125).
25
25.5
26
26.5
27
27.5
28
28.5
Baseline Month 12 Month 24
MMSE Score
Changes in MMSE Scores
Euglycemia
Treated Diabetes
Untreated Hyperglycemia
50
Figure 5. WMH Progression Over Two Years. There was a marginally significant group x time
interaction for WMH progression over the two-year follow up period (p = 0.074). WMH volume
significantly increased over the two-year period for the euglycemia group (b = 0.066, p < 0.001)
and marginally increased for the untreated hyperglycemia group (b = 0.129, p = 0.058). There
was no change in WMH volume in the treated diabetes group (b = -0.042, p = 0.493). The rate of
change in WMH volume was marginally greater in the untreated hyperglycemia group compared
to the treated diabetes group (p = 0.062), but was not different than the euglycemia group (p =
0.365).
0
0.05
0.1
0.15
0.2
0.25
0.3
Baseline Month 24
Log Transformed WMH
WMH Progression
Euglycemia
Treated Diabetes
Untreated Hyperglycemia
Abstract (if available)
Abstract
Objective: Type 2 diabetes mellitus (T2DM) is associated with increased risk for Alzheimer’s dementia. Past studies have shown that glycemic control and medication use may attenuate the risk for dementia in diabetics. Further, recent research indicates increased cerebral spinal fluid (CSF) phosphorylated tau (p-tau) concentrations in older adults with T2DM. To date no studies have investigated these associations in treated T2DM versus untreated hyperglycemia. The present study investigated cerebrospinal fluid (CSF) biomarkers, neuropsychological performance, and brain structure differences between non-diabetic older adults and those with treated T2DM versus untreated hyperglycemia. ❧ Participants and Methods: Non-demented older adults from the Alzheimer’s Disease Neuroimaging Initiative (age 55
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Affective neuropsychiatric symptoms and neural connectivity in the early stages of Alzheimer’s disease
PDF
Relationships between lifetime chronic stress exposure, vascular risk, cognition, and brain structure in HIV
PDF
Dementia, co-morbid depressive symptoms, and mortality in older adults: a twin study
PDF
Indexing cerebrovascular resistance in cognitive decline and Alzheimer's disease
PDF
Brainstem structural integrity in the progression of Alzheimer's disease
PDF
Optimism when imagining the future in old age influences the positivity effect in memory
PDF
The role of blood pressure variability in cognitive decline, cerebrovascular disease and Alzheimer’s disease
PDF
Communication and home-based interventions in Alzheimer’s disease -- a review
PDF
Accuray of subjective cognitive complaints in a longitudinal context: the effect of depression and dementia status
PDF
Averting dementia: renin-angiotensin system and angiogenic cells in cognitive decline
PDF
The role of the locus coeruleus in Alzheimer’s disease and cerebrovascular function: insights from neuroimaging, neuropsychology, and biofluid markers
PDF
Crosscultural differences in dementia diagnosis and care-seeking in Hispanic and non-Hispanic white outpatients
PDF
Are life events differentially associated with dementia risk by gender? A twin study
PDF
Neuropsychological functioning in older adult long-term cancer survivors
PDF
Effects of AT-1 receptor blockers on cognitive decline and Alzheimer's disease
PDF
Long-term blood pressure variability across the clinical and biomarker spectrum of Alzheimer’s disease
PDF
A fotonovela for improving dementia literacy among Latinos
PDF
Dedifferentiation of emotion regulation strategies in the aging brain: an MVPA investigation
PDF
How emotional arousal influences memory and learning in younger and older adults
PDF
Angiotensin (1-7): a novel treatment for diabetes-induced kidney and heart dysfunction
Asset Metadata
Creator
McIntosh, Elissa Charney
(author)
Core Title
Untreated hyperglycemia associated with tau pathology and worse cognitive performance in older adults
School
College of Letters, Arts and Sciences
Degree
Master of Arts
Degree Program
Psychology
Publication Date
07/28/2017
Defense Date
04/24/2017
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
Alzheimer's disease,dementia,hyperglycemia,neuropsychology,OAI-PMH Harvest,type 2 diabetes mellitus
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Nation, Daniel Addison (
committee chair
), Gatz, Margaret (
committee member
), Mather, Mara (
committee member
)
Creator Email
ecmcinto@usc.edu,mcintosh.elissa@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-419248
Unique identifier
UC11265042
Identifier
etd-McIntoshEl-5655.pdf (filename),usctheses-c40-419248 (legacy record id)
Legacy Identifier
etd-McIntoshEl-5655.pdf
Dmrecord
419248
Document Type
Thesis
Rights
McIntosh, Elissa Charney
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
Alzheimer's disease
dementia
hyperglycemia
neuropsychology
type 2 diabetes mellitus