University of Southern California Dissertations and Theses

Contributions of dynamic cerebrovascular function to cognitive decline and dementia: development and validation of a novel neuroimaging approach

(USC Thesis Other)

Contributions of dynamic cerebrovascular function to cognitive decline and dementia: development and validation of a novel neuroimaging approach

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content CONTRIBUTIONS OF DYNAMIC CEREBROVASCULAR FUNCTION TO
COGNITIVE DECLINE AND DEMENTIA:
DEVELOPMENT AND VALIDATION OF A NOVEL NEUROIMAGING APPROACH
by
Belinda Yew, M.A.
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
PSYCHOLOGY
August 2020
Copyright 2020 Belinda Yew

iv
List of Tables

Table 1. Baseline Demographic and Physiological Characteristics of Age Groups ..................... 36
Table 2. Age and Breathing Condition Differences ...................................................................... 37
Table 3. Effects of Paced Breathing on CVRi, CO2R, and CBF .................................................. 38
Table 4. Effects of Breath Holding on CVRi, CO2R, and CBF .................................................... 39
Table 5. Age Differences in CVRi, CO2R, and CBF Response to Paced Breathing .................... 40
Table 6. Age Differences in CVRi, CO2R, and CBF Response to Breath Holding ...................... 41
Table 7. Relationships Between Global Cognition and Resting-State Cerebrovascular
Function ........................................................................................................................................ 42

Table 8. Relationships Between Global Cognition and Cerebrovascular Response to Breathing
Manipulations ............................................................................................................................... 43

Table 9. Relationships Between Word-List Memory and Resting-State Cerebrovascular
Function ........................................................................................................................................ 44

Table 10. Relationships Between Word-List Memory and Cerebrovascular Response to
Breathing Manipulations ............................................................................................................... 45

Table 11. Relationships Between Story Memory and Resting-State Cerebrovascular Function . 46
Table 12. Relationships Between Story Memory and Cerebrovascular Response to Breathing
Manipulations ............................................................................................................................... 47

Table 13. Relationships Between Executive Function and Resting-State Cerebrovascular
Function ....................................................................................................................................... 48

Table 14. Relationships Between Executive Function and Cerebrovascular Response to
Breathing Manipulations ............................................................................................................... 49

Table 15. Baseline Demographic and Physiological Characteristics of APOE ε4 Carriers and
Non-Carriers ................................................................................................................................. 50

v
Table 16. Comparison of APOE ε4 Non-Carriers to Carriers for Resting-State Cerebrovascular
Function ........................................................................................................................................ 51

Table 17. Comparison of APOE ε4 Non-Carriers to Carriers for Cerebrovascular Response to
Breathing Manipulations ............................................................................................................... 52

vi
List of Figures

Figure 1. Experimental Setup ....................................................................................................... 14
Figure 2. Effects of Breathing Manipulations on CBF and CVRi ................................................ 16
Figure 3. Relationships Between CVRi and Global Cognition .................................................... 20
Figure 4. Relationships Between CO2R and Global Cognition .................................................... 21
Figure 5. Relationships Between CVRi and Word List Memory ................................................. 22
Figure 6. Relationships Between CVRi and Story Memory ......................................................... 23
Figure 7. Relationships Between CO2R and Story Memory ......................................................... 24

Figure 8. Resting-state CVRi and WMH Severity ........................................................................ 25

Figure 9. Resting-state CBF and WMH Severity ......................................................................... 26
Figure 10. CVRi Response to Paced Breathing and APOE e4 ..................................................... 27
Figure 11. CVRi Response to Breath Holding and APOE e4 ....................................................... 28

vii
Abstract
Cerebrovascular contributions to cognitive decline have been increasingly recognized.
Evaluation of dynamic cerebrovascular function, however, has been restricted to animal
paradigms, invasive manipulations, and/or methods offering suboptimal measurement of
microvascular function. Breath holding and hyperventilation elicit alterations in end-tidal carbon
dioxide (etCO2) and associated hemodynamic variables. We therefore developed and validated a
neuroimaging paradigm incorporating breathing manipulations for assessment of dynamic
cerebrovascular function. Community-dwelling younger (n = 27) and older (n = 40) adults
performed guided breath holds and hyperventilatory paced breathing while undergoing pseudo-
continuous arterial spin labelling (pCASL) MRI. Continuous blood pressure and capnography
were simultaneously collected and synchronized with pCASL-derived cerebral blood flow
(CBF). Cerebrovascular resistance index (CVRi) was calculated by dividing mean arterial blood
pressure by CBF. Cerebrovascular reactivity (CO2R) was calculated by dividing change in CBF
by change in etCO2. Older adults also completed cognitive testing, T2 FLAIR imaging (from
which white matter hyperintensity [WMH] burden was determined), and APOE genotyping.
Across age groups, breath holding led to increased etCO2, CBF, and CO2R, and decreased CVRi.
Paced breathing led to decreased etCO2, CBF, and CO2R, and increased CVRi. Older adults
exhibited diminished hemodynamic response to both breathing manipulations. Attenuated
response was associated with poorer cognitive performance and APOE ε4. Lower resting-state
CO2R, CVRi, and CBF were associated with poorer cognitive performance, APOE ε4, and
greater WMH severity. Findings validate our paradigm as an effective measure of hemodynamic
changes associated with dementia risk factors in non-clinical populations. This may facilitate
detection of cerebrovascular changes characterizing the earliest stages of pathological aging.

1
Introduction
Vascular dysfunction has be increasingly implicated in cognitive decline and
neurodegenerative disease (Arvanitakis, Capuano, Leurgans, Bennett, & Schneider, 2016; Qiu &
Fratiglioni, 2015; Snyder et al., 2015; Toth, Tarantini, Csiszar, & Ungvari, 2017). Pre-existing
vascular conditions such as cerebrovascular disease, hypertension, and diabetes, have been
linked to greater age-related cognitive decline (Qiu, Xu, Winblad, & Fratiglioni, 2010; Raz &
Rodrigue, 2006) and development of neurodegenerative lesions in the brain (Rabin et al., 2019;
Rodrigue et al., 2013). Older adults bearing greater vascular risk also exhibit increased atrophy
(Cardenas et al., 2012) and less efficient structural and functional connectivity (Sun et al., 2011;
Tuladhar et al., 2016) throughout the brain. Furthermore, there has been some success in
mitigating cognitive decline through management of vascular risk (e.g. antihypertensive
medication use)(I. Hajjar, Zhao, Alsop, & Novak, 2010; Ligthart, Moll van Charante, Van Gool,
& Richard, 2010; Mearns & Fuster, 2010). Associations of vascular function with cognition and
dementia are thus well established.
Vascular pathways also appear to play an integral role in genetic contributions to
cognitive decline. The strongest known genetic risk factor for Alzheimer’s disease (AD), the ε4
variant of the apolipoprotein gene (APOE4), has been associated with cerebrovascular
dysfunction through a number of avenues (Tai et al., 2016). Although APOE is primarily
responsible for cholesterol metabolism, the ε4 polymorphism has also been linked to
neurovascular abnormalities (Rasmussen et al., 2019) including greater blood brain barrier
(BBB) permeability (Donahue & Johanson, 2008), neuroinflammation (Tai et al., 2015),
attenuated cerebrovascular reactivity (Rasmussen et al., 2019; Suri et al., 2015), impaired
amyloid clearance (Castellano et al., 2011), and reduced expression of amyloid-degrading

2
enzymes neprilysin and insulin-degrading enzyme (Jiang et al., 2008). Relative to non-carriers,
ε4 carriers also experience elevated risk for vascular disease (including cerebral amyloid
angiopathy), and heightened effects of cardiovascular risk on cognition (Caselli et al., 2011) and
brain function (Filippini et al., 2011; C. E. Wierenga et al., 2013). APOE4 may thus drive AD
onset through promotion of both of amyloid accumulation and cerebrovascular dysfunction (Bu,
2009; Tai et al., 2016). Notably, recent work has linked APOE4 to BBB breakdown, independent
of amyloid or tau. APOE also seems to mediate links between BBB breakdown and future
cognitive decline (Montagne et al., 2020; Nation et al., 2019).
Both separate and in conjunction with APOE4, cerebrovascular dysfunction has been
implicated in pathological aging. Cerebral blood flow (CBF) abnormalities have been detected in
mild cognitive impairment (MCI) and dementia of varying etiologies, both at rest and during
cognitive performance (Dai, 2009; Johnson et al., 2005; Toth et al., 2017). Medial temporal,
precuneus, posterior cingulate, and prefrontal hypoperfusion have been detected in AD cases
(Fleisher et al., 2009; Wang, 2014), as has diminished activation of neural regions comprising
the default-mode network (Greicius, Srivastava, Reiss, & Menon, 2004). Reduced CBF has also
been evinced by individuals with mild cognitive impairment (typically viewed as a transitional
phase between normal cognition and dementia)(Chao et al., 2010) and elevated cerebral amyloid
load (Mattson et al., 2014). Notably, CBF increases have also been linked to pathological aging
and cognitive decline (Bangen et al., 2017; Dai, 2009; Dickerson et al., 2005; C. E. Wierenga et
al., 2013), sometimes accompanied by superior cognitive performance (C.E. Wierenga et al.,
2012). Such increases may represent compensatory mechanisms (e.g. hyperperfusion to
accommodate increased metabolic demands in the face of neuronal inefficiency)(Dai, 2009; C.E.
Wierenga et al., 2012) or reduced precision in neuronal recruitment (i.e. dedifferentiation, more

3
generalized activation)(S. D. Han, Bangen, & Bondi, 2009). CBF elevation in vulnerable
populations may therefore index the same vascular dysfunction through which hypoperfusion
associated with severe pathology arises.
This connection of bi-directional CBF changes to cognitive decline has been paralleled in
studies of blood pressure. Elevated blood pressure and hypertension have been linked to greater
neurodegenerative disease pathology and brain atrophy (Langbaum et al., 2012; Petrovitch et al.,
2000), poorer cognitive performance (Muela et al., 2017) and elevated risk for later dementia
(Qiu, Winblad, & Fratiglioni, 2005). Extremely low or steep (i.e. > 10 mmHg) decreases in
diastolic blood pressure, however, are also associated with greater dementia risk (Qiu, von
Strauss, Fastbom, Winblad, & Fratiglioni, 2003) and cortical atrophy (den Heijer et al., 2003). It
is thus possible that cognitive decline may arise from both hypertension and hypotension.
Specifically, midlife blood pressure elevations may produce pathological changes through small
vessel disease and stroke, as well as atherosclerosis, which can itself induce hypoperfusion and
enduring drops in blood pressure. In contrast, low blood pressure in the very old (i.e. individuals
aged over 80), potentially caused by damage to neural structures involved in blood pressure
regulation (Skoog, 2003), may instigate hypoperfusion. This could in turn produce neuronal
dysfunction and dementia in later life (Guo, Viitanen, Fratiglioni, & Winblad, 1996; Morris,
1993).
The characterization of pathological aging by both decreases and increases in CBF
(Bangen et al., 2017) and blood pressure (Skoog, 2003), is consistent with deficits in
cerebrovascular autoregulation rather than hypoperfusion or hypertension alone. Investigations
of the relationship between CBF and blood pressure, as well as broader cerebrovascular
autoregulatory processes, may therefore better elucidate AD pathogenesis. Cerebrovascular

4
resistance (see Box 1) is the ratio of cerebral perfusion (Pα) to CBF, where Pα is the difference
between mean arterial pressure (MAP) and intracranial pressure (ICP). Adjustment of
cerebrovascular resistance enables maintenance (i.e. cerebrovascular autoregulation) of CBF
amid even significant changes in blood pressure. Specifically, MAP increases elicit constriction
of blood vessels, thereby reducing vessel diameter and blood volume. Conversely, MAP
decreases will trigger blood vessel dilation, enlarging vessel diameter and increasing blood
volume (Buxton, 2009). Steady CBF may therefore mask dramatic changes in cerebrovascular
resistance, detectable only when blood pressure is also taken into account.

Box 1. Cerebrovascular resistance

Transcranial doppler (TCD) and MRI approaches have revealed elevated static
cerebrovascular resistance in individuals with MCI, dementia, and neuropathological markers of
AD, across various brain regions (J. Liu et al., 2014; Rivera-Rivera et al., 2016; Roher et al.,
2011), as well as the cerebral arteries through which they are supplied (de Heus et al., 2018;
Gommer et al., 2012; Rivera-Rivera et al., 2016; Roher et al., 2011). Cerebrovascular resistance
increases have been linked to greater future cognitive declines, amyloid retention, brain atrophy,
and progression to dementia (Yew & Nation, 2017). Furthermore, cerebrovascular resistance
elevations in preclinical and early-stage AD cases are intermediate (i.e. higher than that of
controls but lower than that of AD patients), supporting involvement of increased resistance in
[1] Cerebral perfusion pressure (P
$
) = MAP – ICP

Under typical conditions, ICP is relatively stable and considerably lower than MAP.
Cerebrovascular resistance can therefore be approximated by dividing MAP by CBF.

[2] Cerebrovascular resistance =
P
$
CBF
≈
MAP
CBF

5
disease progression. Crucially, observed cerebrovascular resistance differences are observed in
more regions and at earlier disease stages than CBF itself, suggesting that resistance increases
represent more sensitive markers of early pathological change than blood flow alone (Nation et
al., 2013; Yew & Nation, 2017).
Despite these changes in static cerebrovascular resistance, autoregulatory processes
appear largely resistant to effects of heathy and pathological aging (Zazulia, Videen, Morris, &
Powers, 2010). However, a “rightward shift” in the autoregulatory curve is often observed in
older adults, such that the range of blood pressure values across which CBF can be maintained
spans higher (at the expense of lower) pressures. This adaptive increase in cerebrovascular
resistance aims to offset chronic hypertension and related mechanical injuries to the
cerebrovasculature. Accommodation of higher pressures nonetheless results in diminished
autoregulatory capacity at lower pressures, leaving the brain vulnerable to hypoperfusion during
hypotensive episodes (Paulson, Strandgaard, & Edvinsson, 1990; van Beek, Claassen, Rikkert, &
Jansen, 2008). Consistent with this, cerebrovascular autoregulation in response to more acute
changes is slower (Heckmann, Brown, Cheregi, Hilz, & Neundörfer, 2003), and diminished in
response to hypotension (Toth et al., 2017), for older versus younger adults.
Investigation of acute autoregulatory function has been largely restricted to animal
paradigms employing pharmacological manipulations. Angiotensin II-induced hypertension
results in reduced CBF, attenuated hyperemic response (i.e. reduced CBF increase following
whisker stimulation), and impaired learning and memory (Faraco et al., 2015; Toth et al., 2013).
Relatedly, transgenic mouse models of AD exhibit impaired response to endothelium-dependent
vasodilators (Faraco et al., 2015). Furthermore, application of amyloid, one of the
neuropathological hallmarks of AD, induces endothelium-dependent vasoconstriction, oxidative

6
stress (Thomas, Thomas, McLendon, Sutton, & Mullan, 1996), and reduced response to
endothelium-dependent vasodilators in the absence of changes to cerebral metabolism (Niwa,
Carlson, & Iadecola, 2000). Animal research thus implicates vascular dysfunction, particularly
that driven by blood pressure abnormalities, in pathological aging. More broadly, animal studies
evidence emergence of a hypercontractile phenotype (i.e. both causing and resulting from
amyloidosis), that likely interferes with CBF.
Cerebrovascular autoregulation is also affected by arterial partial pressure of carbon
dioxide (PaCO2). As with blood pressure, PaCO2 changes are met with corresponding changes in
cerebrovascular resistance. Hypercapnia elicits vasodilation and commensurate increases in
CBF. Conversely, hypocapnia triggers vasoconstriction and subsequent decreases in CBF
(Duffin et al., 2018; Kanno et al., 1988). Breathing alterations, which have been shown to alter
carbon dioxide (CO2) may therefore represent a non-invasive analog to pharmacological
modification of vascular function. More specifically, hyperventilation can be used to induce
hypocapnia (Bright, Bulte, Jezzard, & Duyn, 2009; Sousa, Vilela, & Figueiredo, 2014) and
breath holding to induce hypercapnia (Handwerker, Gazzaley, Inglis, & D'Esposito, 2007; H. L.
Liu, Huang, Wu, & Hsu, 2002).
Notably, the bulk of research examining acute/dynamic cerebrovascular function has
utilized TCD to measure CBF. Although this approach offers excellent temporal precision, it
lacks the spatial resolution of MRI-based methods and is therefore limited to larger cerebral
arteries in neuroanatomically superficial regions (van Beek et al., 2008). Given that
neurovascular changes underpinning age-related declines in cerebrovascular autoregulation
appear to occur in smaller, deeper lying structures (Iadecola, 2004), pCASL may be more
sensitive to early pathological processes. The pCASL signal also represents a more direct

7
measure of CBF than TCD, which measures CBF velocity, and BOLD fMRI, which indexes
neurovascular coupling. PCASL-derived CBF may therefore enable more precise quantification
of cerebrovascular resistance and cerebrovascular reactivity during CO2 manipulation (P. Liu, De
Vis, & Lu, 2019).

Aims and Hypotheses
In the present study we sought to apply breathing-based manipulations during MRI
acquisition to evaluate dynamic vascular function in groups vulnerable to cognitive decline and
dementia (i.e. older adults, APOE ε4-carriers, individuals with greater white matter
hyperintensity (WMH) burden, and those demonstrating potential cognitive dysfunction).
Our aims were to:
1) Validate a novel cerebrovascular reactivity paradigm, and assess its potential as a
preclinical marker for dementia risk through investigation of its relationships with
a. Age
b. Cognitive performance
c. APOE ε4 carrier status
d. WMH severity
2) Determine contributions of blood pressure and CO2 to pCASL-indexed measures of
hemodynamic response.
3) Evaluate the value of measures capturing reactivity versus mere static/baseline
indexes.
We hypothesized that:

8
1) Individuals vulnerable to dementia (i.e. those who were older, ε4-carrying, evincing
higher WMH severity, and/or demonstrating poorer cognitive performance) would
exhibit diminished reactivity to breathing manipulations.
2) Blood pressure and CO2 would significantly contribute to reactivity, as reflected in
relationships of cerebrovascular resistance and cerebrovascular reactivity to dementia
risk factors (age, APOE4, WMH severity, and cognitive dysfunction).
3) The importance of blood pressure and CO2 to reactivity would be reflected in greater
effects for reactivity measures incorporating blood pressure and CO2. Specifically,
relationships between dementia risk factors and cerebrovascular resistance and
reactivity, would be present in more brain regions than CBF alone.
4) Baseline and reactivity measures would both be associated with dementia risk.

Materials and Methods
Participants
Younger (n=27; aged between 18 and 34 years of age) and older (n=40; aged 54-92)
adults were recruited from the community and compensated financially for their participation.
Exclusion criteria included diagnosis of dementia, Mattis Dementia Rating Scale (DRS-2) score
≤ 126, genetic mutation known to produce dementia (i.e. PSEN1 and MAPT), diabetes,
HIV/AIDS, weight exceeding 270 pounds, MRI contraindication(s), current organ failure, and/or
history of stroke, myocardial infarction, neurological or major psychiatric disorder, head injury
with loss of consciousness exceeding 15 minutes, substance abuse resulting in hospitalization,
B12 deficiency, and/or hypothyroidism.

9
Neuroimaging and Physiological Measures
1) Static blood pressure (BP) was measured using a calibrated mercury
sphygmomanometer. Two supine measurements were collected and averaged.
2) Continuous (beat-to-beat) brachial BP was indexed during MRI acquisition using an
MR-safe device (Caretaker NIBP-MRI, BIOPAC Systems, Inc.) calibrated with
manual blood pressure measurements.
3) Mean arterial pressure (MAP) was calculated by multiplying diastolic blood pressure
(DBP) by two, adding systolic blood pressure (SBP), and dividing by three. This can
also be expressed as:
SBP+2(DBP)
3

Both static and dynamic MAP was calculated using static and dynamic BP values,
respectively.
4) Cerebral blood flow (CBF) was measured via arterial spin labeling (ASL) MRI,
collected using a 3T scanner and pseudo-continuous ASL (pCASL) method, with
background suppressed gradient and spin echo (GRASE). Parameters are as follows:
TR = 5000ms; TE = 36.3ms; slice thickness = 3.42mm; number of slices = 24;
number of measurements = 1 M0 image + 1 dummy image + 15 pairs of tag-control
images; total scan time = 5:25. CBF was derived by normalizing scaled, distortion-
corrected, co-registered, and partial volume-corrected perfusion weighted images to a
reference image estimating blood water magnetization. Resulting values indexed CBF
in units of arterial water density (mL/100g x 60s). Altogether, there were 3 scans
(baseline, paced breathing, and breath hold), each yielding 15 images (i.e. tag-control

10
pairs). Both global and regional (hippocampal, medial temporal, inferior temporal,
posterior cingulate, precuneus, and inferior frontal) CBF values were extracted.
5) Cerebrovascular resistance (CVRi) was calculated by dividing each dynamic MAP
value by its corresponding CBF value. This resulted in 15 CVRi measurements (i.e.
one for each pCASL pair) for each of the three pCASL scans. Both global and
regional (hippocampus, medial temporal, inferior temporal, posterior cingulate,
precuneus, and inferior frontal) CVRi values were calculated.
6) End-tidal carbon dioxide (etCO2) was measured during MRI acquisition, using an
M3015A sidestream carbon dioxide extension module (Philips Medical Systems),
connected to a combined oral-nasal cannula into which participants breathed.
7) Cerebrovascular reactivity (CO2R) was calculated by dividing change in CBF
between two time points (i.e. 2 pCASL pairs), by change in etCO2 between the same
two time points.
Physiological and cerebral blood flow data were synchronized using AcqKnowledge 4.0
software (BIOPAC Systems, Inc.).
8) White matter hyperintensities (WMH) were identified and evaluated using a T2 fluid-
attenuated inversion recovery (FLAIR) MRI scan.
Cognitive Measures
Cognitive measures were obtained for older adults as part of a comprehensive
neuropsychological battery completed during a separate study visit. For all measures, higher
scores reflect superior cognitive performance.

11
Global cognition
Global cognition was indexed using the Dementia Rating Scale (DRS-2) total score. The
DRS evaluates cognitive performance across five domains– Attention, Initiation and
Perseveration, Construction, Conceptualization, and Memory, scores from which are collated for
a total, global measure of cognition. (Hofer, Piccinin, & Douglas, 1996).
Memory
Memory was evaluated using the Rey Auditory Verbal Learning Test (RAVLT), in which
a word list must be learned and recalled after shorter and longer delays. Analyses utilized the
short-term percent retention and long-term percent retention scores (Rey, 1941). Memory was
also assessed with the Wechsler Memory Scale Logical Memory subtest, which requires
participants to learn and recall short stories after a short (Logical Memory I) and long (Logical
Memory 2) delay (Lezak, Howieson, & Loring, 2004).
Executive function
Executive function was assessed using FAS Fluency, a phonemic fluency test in which
participants must verbally generate as many words as possible beginning with specified letters,
within a given time limit (Tombaugh, Kozak, & Rees, 1999). Executive function was also
evaluated by the Delis-Kaplan Executive Function System (D-KEFS) Category Switching score,
which requires time-pressured generation of exemplars from alternating categories (Delis,
Kaplan, & Kramer, 2001).
Microvascular Pathology
WMHs were qualitatively scored by a trained rater using Fazekas’ scale, a 4-point scale
for assessing periventricular lesions and deep white matter lesions (Fazekas, Chawluk, Alavi,
Hurtig, & Zimmerman, 1987). Higher scores reflect greater WMH severity.

12
APOE Genotyping
In older adults, APOE genotyping of isolated DNA, extracted from blood cell pellet
fractions obtained during plasma separation, was conducted using the TaqMan SNP Genotyping
Assay (Thermo Fisher Scientific Inc.) on an Applied Biosystems 7300 Real Time PCR System.
APOE gene single nucleotide polymorphisms (SNPs) were inspected for SNP database (dbSNP)
IDs rs429358 and rs7412. Allelic discrimination was conducted using qPCR software. Incidence
of the APOE ε4 allele was defined by the presence of rs429358-C and rs7412-C. Participants
were classified as “carriers” if they possessed one or two ε4 alleles.
Study Design
Breathing manipulation procedures and associated collection of imaging and physiological
variables are depicted in Figure 1. During acquisition of pCASL scans, participants completed
three breathing exercises. Each breathing exercise was preceded by verbal and written
instructions. Participants were also provided with in-scanner visual stimuli to guide breathing
and ensure accurate completion of each exercise.
1) Baseline breathing: participants were shown a stationary green circle and instructed to
“breathe as you normally do”.
2) Paced breathing: participants were presented with a circle that filled with color over a
5 second interval. The color of the filling circle alternated between yellow and blue.
Participants were instructed to “inhale while the circle is yellow” and “exhale while
the circle is blue”.
3) Breath holding: participants were shown a circle that filled with color (green) over a
25 second interval. Once full, the circle “restarted” and began filling with another

13
color (red) over a 15 second interval. Participants were instructed to, “breathe
normally while the circle is green” and “hold your breath while the circle is red”.
All participants were trained until they were able to comfortably and reliably perform
each breathing exercise. PCASL, BP, and etCO2 data were then collected and synchronized
while participants completed breathing exercises in the scanner.
Statistical Analyses
Analyses were performed using IBM SPSS Statistics (Version 24). Age (young versus
old) comparisons of baseline demographic, BP, and CO2 variables were performed using
independent-samples T tests or chi-square tests of independence. Linear mixed models using
maximum likelihood estimation were employed to assess global and regional CBF, CVRi, and
CO2R changes in response to breathing conditions. Specifically, paced breathing was contrasted
with baseline, and breath holding was contrasted with paced breathing. We assessed effects of
age group (old versus young) on hemodynamic response paced breathing and breath holding. For
older adults, we analyzed relationships between hemodynamic response and APOE ε4 carrier
status, cognitive performance, and WMH burden. For all mixed models, time was entered as a
random variable with autoregressive covariance structure, and etCO2 and sex were entered as
covariates. For models evaluating relationships with cognitive performance and WMH burden,
additional covariates of age, APOE carrier status, and education were included. For models
assessing relationships with APOE, age, sex, and antihypertensive medication use were entered
as covariates. Relationships between baseline (resting state) hemodynamic and dementia risk
variables were also evaluated.

14

0 50 100 150
CBF
Baseline
Breath hold
c) b) a)
Hold your breath
while the circle is
red. Breathe
normally while the
circle is green.
You will see a
green circle.
Breathe as you
normally do.
Inhale while the
circle is yellow.
Exhale while
the circle is blue
Figure 1. Experimental setup. Continuous blood pressure (BP) and end-tidal CO2 (etCO2) were
measured during MRI acquisition and performance of breathing exercises. BP, etCO2, and CBF
measurements were synchronized and used to calculate cerebrovascular resistance (CVRi) and
cerebrovascular reactivity (CO2R). Participants engaged in i. “normal” baseline breathing, ii. paced
breathing, consisting of alternation between 5 second inhale and 5 second exhale, and iii. breath
holding, involving 15 seconds of breath hold followed by 20 seconds of spontaneous breathing.
Breathing conditions were preceded by verbal and written instructions, as well as visual stimuli
(animated, colored circles), guiding pacing and duration for paced breathing and breath holding. Paced
breathing led to decreased etCO2 and CBF (hypocapnic effect). Breath holding increased etCO2 and
CBF (hypercapnic effect).
etCO2
(mmHg)
Paced Breathing Baseline Breath Hold
0 5
CVRi
Baseline
Paced

15
Results
Baseline age comparisons
Results of comparisons between younger and older adult groups for baseline
demographic and physiological variables are shown in Table 1. Older adults had higher systolic
blood pressure, pulse pressure, and mean arterial pressure.
Effects of breathing manipulations
Effects of breathing manipulations on global physiological variables are displayed in
Table 2. Across age groups and relative to baseline, paced breathing was associated with reduced
etCO2 (ß = -4.88, SE = 0.73; P < 0.01). Across age groups and relative to paced breathing, breath
holding was associated with increased etCO2 (ß = 1.94, SE = 0.85; P < 0.05). No significant age
differences were detected in etCO2 during paced breathing or breath holding (all Ps > 0.05).
Older adults exhibited higher SBP than younger adults during paced breathing (ß = 21.48, SE =
4.07; P < 0.01) and breath holding (ß = 17.26, SE = 5.42; P < 0.01).
Changes to CBF and CVRi in response to breathing manipulations are depicted in Figure
2. Mixed models estimates for effects of paced breathing and breath holding on CBF, CVRi, and
CO2R are shown in Tables 3 and 4. Relative to baseline, paced breathing produced decreases in
global and regional CBF and CO2R, and increases in CVRi. In contrast, breath holding led to
increased global and regional CBF and CO2R, and decreased CVRi.
Age-differences in hemodynamic response to breathing manipulations
Results of age effects during paced breathing and breath hold (i.e. age x breathing
interaction effects) on CBF, CVRi, and CO2R are shown in Tables 5 and 6, respectively. Relative
to younger adults, older adults exhibited diminished CVRi increases, CBF decreases, and CO2R
changes, in response to paced breathing. Notably, age x paced breathing interactions were

16
detected in more brain regions for CO2R than CVRi and CBF, and more regions for CVRi than
CBF. Relative to younger adults, older adults exhibited smaller decreases in CVRi, and smaller
increases in CBF and CO2R, in response to breath holding. Age x breath holding interactions
were detected in more brain regions for CVRi and CBF than CO2R.

Relationship between cerebrovascular response and cognition
Global cognition
Results from mixed models assessing relationships between DRS total score and resting
state hemodynamic variables are shown in Table 7. Participants exhibiting higher baseline (i.e.
resting state) CO2R achieved higher scores on the DRS. Results reflecting relationships between
DRS score and hemodynamic response to breathing manipulations are shown in Table 8.
Figure 2. Cerebral blood flow (CBF) and cerebrovascular resistance index (CVRi) values for younger
and older adults during baseline, paced breathing, and breath holding.

17
Individuals demonstrating greater response to paced breathing (i.e. larger increases in CVRi and
CO2R) obtained higher DRS scores. No significant relationships were detected between CBF
response to paced breathing and global cognition. Individuals with superior response to breath
holding (i.e. greater decreases in CVRi and larger increases in CBF), performed better on the
DRS. No significant relationships were detected between CO2R response to breath holding and
global cognition. Associations between DRS score and CVRi in select regions, at baseline (i.e.
resting state) and during breathing manipulations, are depicted in Figure 3. Associations between
DRS score and baseline CO2R in select regions are shown in Figure 4.
Memory
Results from mixed models assessing relationships between RAVLT memory
performance and resting state hemodynamic variables are shown in Table 9. Participants
exhibiting lower CVRi, higher CO2R, and lower CBF at baseline showed better short- and long-
term retention scores. Results reflecting relationships between RAVLT scores and hemodynamic
response to breathing manipulations are shown in Table 10. Individuals demonstrating greater
response to paced breathing (i.e. larger increases in CVRi and CO2R) obtained higher RAVLT
scores. There were a small number of brain regions in which decreased CO2R in response to
paced breathing was associated with higher RAVLT scores. Individuals demonstrating greater
response to breath holding (i.e. larger decreases in CVRi and greater increases in CO2R)
generally obtained better RAVLT scores. There were a small number of brain regions in which
decreases in CO2R response to breath holding were associated with higher RAVLT scores. No
significant relationships were detected between CBF response to either breathing manipulation
and RAVLT performance.

18
Results from mixed models assessing relationships between Logical Memory
performance and resting state hemodynamic variables are shown in Table 11. Participants
exhibiting lower resting CVRi and higher resting CO2R, performed better on Logical Memory I
and II. No significant relationships were detected between resting-state CBF and Logical
Memory performance. Results reflecting relationships between Logical Memory scores and
hemodynamic response to breathing manipulations are shown in Table 12. Individuals
demonstrating greater response to paced breathing (i.e. larger increases in CVRi and decreases in
CBF) obtained higher Logical Memory scores. Participants exhibiting larger CO2R decreases in
response to paced breathing scored higher on Logical Memory. Individuals demonstrating
greater response to breath holding (i.e. larger decreases in CVRi and greater increases in CBF)
scored higher on Logical Memory. No significant associations between CO2R response to breath
hold and Logical Memory performance were detected.
Associations between memory scores and CVRi in select regions, at baseline (i.e. resting
state) and during breathing manipulations, are depicted in Figures 5 and 6. Associations between
memory scores and CO2R in select regions at baseline and during breathing manipulations are
shown in Figure 7.
Executive function
Results from mixed models assessing relationships between performance on tests of
executive function (FAS Verbal Fluency and Category Switching) and resting state
hemodynamic variables are shown in Table 13. Participants exhibiting lower CVRi and higher
CO2R at baseline obtained better FAS and Category Switching scores. No significant
relationships were detected between resting-state CBF and executive function performance.
Results reflecting relationships between executive function scores and hemodynamic response to

19
breathing manipulations are shown in Table 14. CVRi decreases and CBF increases in response
to paced breathing were associated with superior executive function performance. No significant
associations were detected between CO2R response to paced breathing and executive function.
No significant associations were detected between hemodyamic response to breath holding and
executive function.
Relationships between cerebrovascular response and WMH severity
Greater WMH severity (i.e. higher Fazekas score) was associated with higher CVRi and
lower CBF during baseline. Differences in resting state CVRi and CBF for WMH severity
groups are presented in Figures 8 and 9. No significant differences were detected between WMH
severity groups and hemodynamic response to breathing manipulations.
APOE e4 Carrier Status
APOE e4 carriers were younger [t(37) = 2.44, p = 0.02] and less likely to be hypertensive
and using antihypertensive medications [χ
2
(1, N= 39)=4.43, p = 0.04] than non-carriers. No
differences between carriers and non-carriers were detected for baseline blood pressure or
cerebrovascular resistance measures (see Table 15).
Relationships between cerebrovascular response and APOE e4 carrier status
As shown in Table 16, APOE e4 non-carriers had higher baseline CVRi and lower CO2R
than non-carriers. As shown in Table 17, however, non-carriers demonstrated superior
hemodynamic response (larger CVRi increases, CBF decreases, and CO2R increases) to paced
breathing. Non-carriers also exhibited stronger responses (larger CVRi decreases, CBF increases,
and CO2R increase) to breath holding than carriers. Reactivity differences between carriers and
non-carriers are depicted for select brain regions in Figures 10 and 11.

Figure 3. Relationships between cerebrovascular resistance (CVRi) and global cognition (DRS total score) during baseline, paced
breathing, and breath holding. * P < 0.05; ** P < 0.01.
r = -0.12 r = -0.27** r = -0.34**
r = -0.08 r = -0.18** r = -0.25**
20

Figure 4. Relationships between cerebrovascular reactivity (CO2R) and global cognition (DRS total score) at baseline. * P < 0.05; **
P < 0.01.
r = 0.18*
r = 0.23**
r = 0.17*
r = 0.19* r = 0.21*
r = 0.09
21

r = -0.06 r = -0.24** r = -0.24**
r = -0.24** r = -0.42** r = -0.16*
Figure 5. Relationships between cerebrovascular resistance (CVRi) and word list memory (RAVLT retention score) during baseline,
paced breathing, and breath holding. * P < 0.05; ** P < 0.01.
22

r = -0.01 r = -0.18**
r = -0.11
r = 0.00
r = -0.13*
r = -0.10
Figure 6. Relationships between cerebrovascular resistance (CVRi) and story memory (Logical Memory scores) during baseline,
paced breathing, and breath holding. * P < 0.05; ** P < 0.01.
23

Figure 7. Relationships between cerebrovascular reactivity (CO2R) and story memory (Logical Memory score) during baseline,
paced breathing, and breath holding. * P < 0.05; ** P < 0.01; CBF = cerebral blood flow; CO2 = carbon dioxide.
r = 0.21**
r = 0.17* r = 0.14*
r = 0.09 r = -0.09
r = -0.13
24

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Global L HC R HC L MTL R MTL L PC R PC L IFC R IFC L precun. R precun. L ITC R ITC
Baseline CVRi (mmHg/ml/100g/min)
Absent to minimal Evident
Figure 8. Resting-state cerebrovascular resistance (CVRi) in older adult participants with no/minimal versus moderate/severe white
matter hyperintensity burden. * P < 0.05; ** P < 0.01; L = left; R= right; HC = hippocampus; MTL = medial temporal lobe; PC =
posterior cingulate; IFC = inferior frontal cortex; precun. = precuneus; ITC = inferior temporal cortex.

WMH severity was analyzed using Fazekas scores. For illustrative purposes, this figure divides participants into no/minimal (Fazekas
score of 0 or 1) or moderate/severe (Fazekas score 2 or 3) WMH severity groups.

*
**
**
**
**
**
*
**
**
25

0
5
10
15
20
25
30
35
40
45
50
Global L HC R HC L MTL R MTL L PC R PC L IFC R IFC L precun. R precun. L ITC R ITC
Baseline CBF (ml/100g/min)
Absent to minimal Evident
**
**
**
**
**
**
Figure 9. Baseline cerebral blood flow (CBF) in older adult participants with minimal/no versus moderate to severe white matter
hyperintensity burden. * P < 0.05; ** P < 0.01; L = left; R= right; HC = hippocampus; MTL = medial temporal lobe; PC = posterior
cingulate; IFC = inferior frontal cortex; precun. = precuneus; ITC = inferior temporal cortex.

WMH severity was analyzed using Fazekas scores. For illustrative purposes, this figure divides participants into no/minimal (Fazekas
score of 0 or 1) or moderate/severe (Fazekas score 2 or 3) WMH severity groups.

**
**
26

27

**
**
**
Figure 10. Cerebrovascular resistance (CVRi) response to paced breathing (vs. baseline) in
APOE e4 carriers and non-carriers. Non-carriers exhibit greater hypocapnic reactivity. * P <
0.05; ** P < 0.01; BL = baseline.

**

**
**
**
Figure 11. Cerebrovascular resistance
(CVRi) response to breath holding (vs.
paced breathing) in APOE e4 carriers and
non-carriers. Non-carriers exhibit greater
hypercapnic reactivity.

* P < 0.05; ** P < 0.01; BH = breath hold.

**
**
28

29
Discussion
The present study employed a novel neuroimaging method to evaluate relationships
between acute cerebrovascular response and age, cognitive performance, microvascular
pathology, and APOE4 carrier status. Paced breathing and breath holding effectively elicited
changes in cerebral blood flow (CBF), cerebrovascular resistance (CVRi), and cerebrovascular
reactivity (CO2R), with attenuated responses observed for older versus younger adults.
Diminished hemodynamic response was also associated with dementia risk factors (white matter
hyperintensity [WMH] burden, poorer cognition, and APOE e4 possession) in older adults.
Notably, CVRi and CO2R evinced stronger relationships with dementia risk factors than CBF,
highlighting the importance of indexing blood pressure and CO2 in assessment of reactivity. It is
therefore significant that our study is the first to investigate the role of blood pressure in an MRI-
based measure of cerebrovascular reactivity (i.e. CVRi reactivity), as well as its relation to age,
microvascular pathology, APOE e4 carrier status, and cognition.
Our findings validate a novel neuroimaging paradigm through which cerebrovascular
reactivity can be used to index preclinical dementia risk. This approach offers several advantages
over existing methods, including its use of MRI, which delivers the anatomical resolution (van
Beek et al., 2008) necessary for imaging microvascular changes underpinning autoregulatory
deficits (Iadecola, 2004). Our employment of pCASL imaging also yields a more direct measure
of CBF than TCD or BOLD fMRI (P. Liu et al., 2019), improving the precision of
cerebrovascular resistance and reactivity measurements. In addition, breathing variations
represent a viable yet less invasive and better tolerated alternative (P. Liu et al., 2020) to
pharmacological and animal paradigms typically utilized in investigation of dynamic
cerebrovascular function. These advantages are further enhanced by our inclusion of visual

30
guidance during completion of breathing manipulations (versus unguided breathing and gas
inhalation paradigms), which promotes standardization and reproducibility across conditions and
participants.
Our incorporation of simultaneous etCO2 and blood pressure monitoring provides
considerable insights into unique and overlapping contributions of cerebrovascular resistance and
cerebrovascular reactivity to the autoregulatory profile. Blood pressure and CO2 reactivity have
been largely investigated as separate phenomena, with relatively few studies exploring their
interacting contributions. In the present study CVRi and CO2R were more sensitive to
hypo/hypercapnic effects and dementia risk factors than CBF. More specifically, CVRi and
CO2R effects were identified in more brain regions and to greater degrees of statistical
significance than CBF. This is consistent with previously reported superiority of CVRi and
CO2R over CBF in differentiating cognitively normal, prodromal (amyloid positive or MCI), and
probable AD cases (J. Liu et al., 2014; Nation et al., 2013; Yew & Nation, 2017). Taken
together, these findings suggest that measures capturing changes in cerebrovasculature are more
sensitive to early disease stages than CBF alone. This likely reflects the preservation of CBF
across broad blood pressure and CO2 changes, even in aging (Nagata et al., 2016; van Beek et al.,
2008), via substantial alterations to cerebrovascular resistance and reactivity (Duffin et al., 2018;
Kisler, Nelson, Montagne, & Zlokovic, 2017). Failure to account for blood pressure and CO2
changes may therefore mask considerable cerebrovascular dysfunction in healthy and
pathological aging.
Our inclusion of both hypo- and hypercapnic conditions captures a more comprehensive
profile of reactivity than prior work, the bulk of which focuses solely on hypercapnia. Indeed,
distinct relationships were detected between hemodynamic variables and hypo versus

31
hypercapnia. CBF and CVRi were more sensitive to age-related differences during hypercapnic
response, with relatively weaker effects observed during hypocapnia. In contrast, age differences
in CO2R were most pronounced during hypocapnic response and comparatively weaker during
hypercapnia. This is consistent with a recent TCD investigation of hypercapnic response, which
failed to detect age differences in cerebrovascular reactivity (i.e. CO2 driven response) but
showed greater blood pressure reactivity in older versus younger adults. Older individuals may
therefore rely more heavily on blood pressure to facilitate CBF increases during hypercapnia
(Miller, Howery, Harvey, Eldridge, & Barnes, 2018), with increased pressure sustaining CBF
despite compromised vasodilation (McKetton et al., 2018). Age effects were similarly
constrained to blood pressure-driven reactivity in the present study, which utilized a pCASL
MRI-derived index of cerebrovascular resistance, CVRi. This supports involvement of blood
pressure in age-related declines in vasodilatory capacity, not only for larger cerebral blood
vessels but also among penetrating microvasculature.
More broadly, CVRi and CO2R differences may reflect shifts in the impact of blood
pressure on CBF across increasing end-tidal CO2 values. Given that alterations in blood pressure
appear to occur only above a certain CO2 threshold, cerebrovascular resistance may play a more
crucial role during hypercapnic response and a less prominent role during hypocapnia (Battisti-
Charbonney, Fisher, & Duffin, 2011; Panerai, Deverson, Mahony, Hayes, & Evans, 1999). This
is supported by our findings of CVRi age differences predominantly during hyper- but not
hypocapnia. Similar results were reported for a recent TCD investigation of hypocapnic response
in younger and older adults following hyperventilation. Although both age groups exhibited
increased cerebrovascular resistance and decreased CBF velocity (the TCD correlate of CBF), no
age differences were observed for these variables (Minhas, Haunton, Robinson, & Panerai,

32
2019). These findings further emphasize the importance of blood pressure measurement for
MRI-based assessment of cerebrovascular reactivity in aging.
Interestingly, baseline CO2R was more strongly associated with cognitive performance
than reactivity in response to either breathing manipulation. This is consistent with TCD
research, which has shown equivalent and potentially stronger associations between cognition
and baseline fluctuations in cerebrovascular reactivity, relative to hyper/hypocapnic-driven
reactivity (Ihab Hajjar, Marmerelis, Shin, & Chui, 2014). TCD measures of cerebrovascular
reactivity derived from spontaneous, resting-state breathing also differentiate patients with AD
and MCI from cognitively normal controls (Marmarelis, Shin, Orme, & Zhang, 2013;
Marmarelis, Shin, Tarumi, & Zhang, 2017), supporting our findings of baseline CO2R sensitivity
to age-related cognitive decline. Given the relative simplicity of, and participant preference for
(P. Liu et al., 2020), resting state measurement over hypo/hypercapnic protocols, baseline
cerebroascular reactivity indexes may prove a superior alternative to reactivity paradigms. More
targeted investigation is needed, however, to clarify the sensitivity and clinical utility of such
measures (Marmarelis et al., 2017).
Among older adults in the present study, diminished cerebrovascular response was
associated with poorer cognition. Participants exhibiting attenuated CVR, CO2R, and CBF
changes performed more poorly on tests of global cognition and memory, even after controlling
for APOE ε4 carrier status. This is consistent with prior TCD and MRI research connecting
attenuated hypercapnic response to cognitive decline in AD (Silvestrini et al., 2006) mild
cognitive impairment (Cantin et al., 2011; McKetton et al., 2018) and dysexecutive cognitive
syndrome (Calviere et al., 2010). Age-related declines in cerebrovascular reactivity have also
been linked to declines in processing speed and episodic memory (Peng et al., 2018), as well as

33
greater subjective memory complaints (Catchlove et al., 2018). Notably, relationships between
cognition and hypocapnic cerebrovascular reactivity or cerebrovascular resistance remains
largely unstudied. In the present study, relationships between hypocapnic hemodynamic response
and cognition were present in more brain regions and at greater statistical significance than those
for hypercapnic response, suggesting that further investigation of hypocapnic/vasoconstrictive
changes in aging is warranted.
Diminished cerebrovascular response in older adults was also associated with greater
cerebral microvascular pathology and APOE ε4 carrier status. Participants with more severe
WMH burden exhibited higher resting-state CVRi and lower resting-state CBF. Similarly, APOE
ε4 carriers demonstrated attenuated cerebrovascular response to paced breathing and breath
holding relative to non-carriers, consistent with suggestions that APOE4 confers risk for
dementia via vascular pathways (Tai et al., 2016). This expands prior findings of attenuated
hypercapnic reactivity among carriers (Rasmussen et al., 2019), to hypocapnic/vasoconstrictive
contexts. Taken together, our WMH and APOE results connect acute cerebrovascular
dysfunction to markers for pathological aging, even among community-dwelling older adults.
This implicates autoregulatory deficits, potentially stemming from age-related increases in
cerebrovascular resistance, in the earliest pathological stages.
Our findings therefore shed light on potential mechanisms underlying pathological aging,
including AD. Greater resistance of vessels can, for example, decrease pulsation amplitude
(Weller, Subash, Preston, Mazanti, & Carare, 2008) and brain lymphatic function (Iliff,
Goldman, & Nedergaard, 2015; Nedergaard, 2013), compromising capacity for drainage of
pathological agents and promoting white matter disease (Qiu & Fratiglioni, 2015; Weller, Boche,
& Nicoll, 2009). Vasotoxicity resulting from accumulation of pathologic agents (e.g. amyloid)

34
can also induce vasoconstriction (B. H. Han et al., 2015), further disrupting autoregulatory
processes. Amid even mild CBF declines, protein synthesis is disturbed, impeding synaptic
plasticity crucial for memory formation (Mies, Ishimaru, Xie, Seo, & Hossmann, 1991). More
substantial CBF declines may instigate hypoxia, hampering enzymatic activity necessary for
initiation of action potentials, and disturbing brain pH, electrolyte, and water gradient
homeostasis. With prolonged disturbance, the brain is rendered vulnerable to white matter
lesions and neurotoxicity (Zlokovic, 2011). In addition, hypoxia-driven alterations to genes
underlying maintenance of the cerebrovascular endothelium, can trigger deterioration of cerebral
blood vessels and CBF deregulation (Chow et al., 2007; Wu et al., 2005). As cerebrovascular
function is increasingly disrupted, cognitive deficits arise, followed and exacerbated by the
neuronal loss that characterizes later stage dementia (Iadecola & Davisson, 2008).
Notably, our study is characterized by a number of limitations. The relatively small
sample size raises questions about statistical power, particularly given the modest effect sizes
detected. On a related note, the large number of statistical analyses performed may have inflated
Type I error rates. Calculation of the false discovery rate and/or adjustment for multiple
comparisons may therefore be warranted. Our use of CVRi and CO2R, though consistent with
existing work, limits comparisons of blood pressure and CO2 reactivity. More specifically, CVRi
captures cerebrovascular response at a given moment whereas CO2R indexes changes between
two time points. Assessment of a static CO2 reactivity metric (e.g. CO2/CBF) and/or a change-
based blood pressure index (i.e. change in CBF per unit change in blood pressure) is thus needed.
In addition, analysis of CO2R while controlling for blood pressure fluctuations may further
clarify age-related changes in autoregulatory response to CO2 versus blood pressure changes.
Our findings shed light on relationships between cerebrovascular dysfunction and dementia risk

35
factors. However, the temporal features of these associations is unclear and causal inferences
therefore limited. Through longitudinal investigation of these variables, the time course of
discussed changes may be determined, and mechanisms underlying cognitive decline further
illuminated.
Overall, our findings are consistent with suggestions that age-related arterial stiffening
may hamper the proficiency with which blood pressure and CO2 fluctuations are accommodated,
compromising CBF and in turn neuronal function and cognition. More broadly, these results
implicate cerebrovascular dysfunction in initial pathological processes associated with
preclinical cognitive decline and microvascular damage. Given our discovery of changes in
community-dwelling adults free of dementia, our method may facilitate identification of
hemodynamic dysfunction characterizing the earliest stages of cognitive decline and
neurodegenerative disease.

36
Table 1. Baseline demographic and physiological characteristics of age groups.

Younger Older t (df) or χ
2

n 27 40
Age 22.44 (0.67) 69.30 (1.41) 29.96 (65)**
Sex (F/M) 20/7 24/16 1.42
SBP 112.16 (4.18) 134.62 (6.50) 3.04 (65)**
DBP 63.19 (4.91) 71.77 (4.00) 1.30 (65)
PP 49.31 (2.51) 62.85 (4.69) 2.68 (65)*
MAP 79.23 (4.81) 92.72 (4.45) 2.02 (65)*
etCO2 42.41 (1.95) 41.44 (1.03) 0.49 (65)

Mean (SE) values for younger and older adults. Statistics reflect results of T tests and chi-square
tests.

* P < 0.05; ** P < 0.01; df = degrees of freedom; SBP = systolic blood pressure; DBP = diastolic
blood pressure; PP = pulse pressure; MAP = mean arterial pressure.

37
Table 2. Age and breathing condition differences

Paced Breathing Breath Hold Effects
SBP
Young 111.68 (4.79) 112.99 (4.99)
Old > young for both
conditions Old 131.73 (6.31) 128.93 (6.36)
DBP
Young 64.47 (4.29) 65.19 (4.34)
All P > 0.05.
Old 66.15 (5.65) 66.62 (5.53)
MAP
Young 80.21 (4.25) 81.12 (4.36)
All P > 0.05.
Old 88.01 (5.60) 87.39 (5.56)
etCO2
Young 38.43 (1.98) 41.46 (1.35) Paced < Baseline
Breath Hold > Paced Old 37.13 (1.74) 40.22 (1.19)

Mean (SE) hemodynamic values for younger and older adults during paced breathing and breath
holding. Significant (P < 0.05) age and breathing effects are noted. No significant age x
breathing interactions were detected. etCO2 = end-tidal CO2; MAP = mean arterial pressure; SBP
= systolic blood pressure; DBP = diastolic blood pressure.

Note: Values for each breathing condition reflect averages (i.e. across measurements collected
during designated breathing condition).

38
Table 3. Effects of paced breathing (vs. baseline) on CVRi, CO2R, and CBF.

CVRi CO2R CBF
Global 0.08 (0.02)** -0.61 (0.14)** -6.60 (1.66)**
L HC 0.14 (0.03)** -0.54 (0.15)** -6.48 (1.99)**
R HC 0.13 (0.03)** -0.65 (0.15)** -5.58 (1.59)**
L MTL 0.13 (0.03)** -0.72 (0.16)** -3.49 (2.91)
R MTL 0.13 (0.03)** -0.66 (0.15)** -5.76 (2.19)**
L PC 0.19 (0.03)** -0.64 (0.14)** -7.34 (1.52)**
R PC 0.16 (0.02)** -0.50 (0.14)** -6.68 (1.08)**
L IFC 0.12 (0.02)** -0.83 (0.15)** -7.11 (1.56)**
R IFC 0.10 (0.02)** -0.74 (0.28)** -5.01 (1.68)**
L precuneus 0.18 (0.02)** -0.50 (0.13)** -5.02 (1.03)**
R precuneus 0.16 (0.02)** -0.64 (0.14)** -7.69 (1.28)**
L ITC 0.14 (0.03)** -0.56 (0.14)** -6.54 (1.79)**
R ITC 0.13 (0.03)** -0.66 (0.14)** -6.79 (1.86)**

Estimated beta weights (standard errors) generated by mixed models evaluating effects of paced
breathing (vs. baseline) on cerebrovascular resistance (CVRi), cerebrovascular reactivity
(CO2R), and cerebral blood flow (CBF).

* P < 0.05; ** P < 0.01. SE = standard error; L= left; R = right; MTL = medial temporal lobe;
PC = posterior cingulate; IFC = inferior frontal cortex; ITC = inferior temporal cortex.

39
Table 4. Effects of breath holding (vs. paced breathing) on CVRi, CO2R, and CBF.

CVRi CO2R CBF
Global -0.19 (0.02)** 0.33 (0.15)* 15.97 (1.94)**
L HC -0.23 (0.05)** 0.38 (0.15)* 14.34 (2.18)**
R HC -0.19 (0.04)** 0.62 (0.15)** 11.27 (1.84)**
L MTL -0.28 (0.05)** 0.33 (0.15)* 18.49 (3.02)**
R MTL -0.24 (0.05)** 0.56 (0.15)** 14.42 (2.43)**
L PC -0.19 (0.04)** 0.12 (0.14) 13.79 (1.74)**
R PC -0.22 (0.04)** 0.23 (0.15) 9.47 (1.24)**
L IFC -0.24 (0.04)** 0.39 (0.15)* 15.80 (1.94)**
R IFC -0.28 (0.04)** 0.89 (0.31)** 16.06 (1.97)**
L precuneus -0.20 (0.03)** 0.24 (0.15) 9.60 (1.21)**
R precuneus -0.20 (0.03)** 0.19 (0.16) 13.59 (1.57)**
L ITC -0.28 (0.05)** 0.23 (0.13) 15.29 (1.94)**
R ITC -0.19 (0.05)** 0.46 (0.15)** 11.60 (1.90)**

Estimated beta weights (standard errors) generated by mixed models evaluating effects of breath
holding (vs. paced breathing) on cerebrovascular resistance (CVRi), cerebrovascular reactivity
(CO2R), and cerebral blood flow (CBF).

* P < 0.05; ** P < 0.01. SE = standard error; L= left; R = right; MTL = medial temporal lobe;
PC = posterior cingulate; IFC = inferior frontal cortex; ITC = inferior temporal cortex.

40
Table 5. Age (old vs. young) differences in CVRi, CO2R, and CBF response to paced
breathing.

CVRi CO2R CBF
Global -0.04 (0.03) 0.43 (0.18)* 1.80 (2.11)
L HC -0.05 (0.05) 0.22 (0.20) 2.02 (2.55)
R HC -0.06 (0.05) 0.53 (0.20)** 1.85 (2.03)
L MTL -0.01 (0.05) 0.47 (0.21)* -0.24 (3.72)
R MTL -0.02 (0.05) 0.46 (0.19)* 1.18 (2.80)
L PC -0.11 (0.04)** 0.56 (0.19)** 1.88 (1.92)
R PC -0.09 (0.04)* 0.38 (0.19)* 3.53 (1.36)*
L IFC -0.07 (0.03)* 0.47 (0.20)* 4.91 (1.99)*
R IFC -0.03 (0.04) 1.04 (0.35)** 2.28 (2.15)
L precuneus -0.08 (0.04)* 0.30 (0.18) 0.95 (1.31)
R precuneus -0.06 (0.03)
#
0.38 (0.18)* 2.18 (1.62)
L ITC -0.02 (0.05) 0.66 (0.18)** 2.23 (2.29)
R ITC -0.02 (0.05) 0.52 (0.18)** 0.39 (2.38)

Estimated beta weights (standard errors) generated by mixed models evaluating differences
between older and younger adults in cerebrovascular response to paced breathing.

* P < 0.05; ** P < 0.01;
#
trending towards significance (0.05 < P < 0.06). SE = standard error;
CVRi = cerebrovascular resistance index; CO2R = cerebrovascular reactivity; CBF = cerebral
blood flow; L= left; R = right; HC = hippocampus; MTL = medial temporal lobe; PC = posterior
cingulate; IFC = inferior frontal cortex; ITC = inferior temporal cortex.

41
Table 6. Age (old vs. young) differences in CVRi, CO2R, and CBF response to breath
holding.

CVRi CO2R CBF
Global 0.10 (0.03)** -0.40 (0.20)* -9.10 (2.51)**
L HC 0.20 (0.07)** -0.18 (0.20) -12.50 (2.82)**
R HC 0.15 (0.06)* -0.56 (0.19)** -9.04 (2.38)**
L MTL 0.22 (0.07)** -0.20 (0.19) -15.41 (3.91)**
R MTL 0.16 (0.07)* -0.39 (0.19)* -10.69 (3.15)**
L PC 0.13 (0.05)** 0.002 (0.18) -11.16 (2.25)**
R PC 0.16 (0.06)** -0.24 (0.19) -6.98 (1.60)**
L IFC 0.14 (0.05)** -0.28 (0.20) -11.31 (2.52)**
R IFC 0.19 (0.06)** -0.11 (0.40) -12.45 (2.56)**
L precuneus 0.11 (0.05)* -0.42 (0.19)* -5.89 (1.56)**
R precuneus 0.13 (0.05)** -0.24 (0.20) -9.25 (2.04)**
L ITC 0.13 (0.07)
#
-0.17 (0.17) -9.59 (2.51)**
R ITC 0.13 (0.07)* -0.52 (0.20)** -7.42 (2.46)**

Estimated beta weights (standard errors) generated by mixed models evaluating differences
between older and younger adults in cerebrovascular response to breath holding.

* P < 0.05; ** P < 0.01. SE = standard error; CVRi = cerebrovascular resistance index; CO2R =
cerebrovascular reactivity; CBF = cerebral blood flow; L= left; R = right; HC = hippocampus;
MTL = medial temporal lobe; PC = posterior cingulate; IFC = inferior frontal cortex; ITC =
inferior temporal cortex.

42
Table 7. Relationships between global cognition and resting-state cerebrovascular function.

Estimated beta weights (standard errors) generated by mixed models evaluating relationships
between Dementia Rating Scale (DRS) total score and resting state cerebrovascular reactivity
(CO2R) and cerebral blood flow (CBF). E.g. Higher DRS scores were associated with higher
global CO2R.

* P < 0.05; ** P < 0.01; L = left; R = right; HC = hippocampus; MTL = medial temporal lobe;
PC = posterior cingulate; IFC = inferior frontal cortex; ITC = inferior temporal cortex.

Baseline
variable
Brain region DRS total
CO2R
Global 0.05 (0.02)**
L HC 0.06 (0.02)**
R HC 0.06 (0.02)**
R MTL 0.04 (0.02)*
L PC 0.06 (0.02)**
R PC 0.05 (0.02)*
L IFC 0.08 (0.02)**
R IFC 0.05 (0.02)*
L precuneus 0.05 (0.02)**
R precuneus 0.08 (0.02)**
L ITC 0.04 (0.02)*
R ITC 0.04 (0.02)*
CBF L IFC -0.34 (0.15)*

43
Table 8. Relationships between global cognition and cerebrovascular response to breathing
manipulations.

Estimated beta weights (standard errors) generated by mixed models evaluating relationships
between Dementia Rating Scale (DRS) total score and cerebrovascular resistance (CVRi),
cerebrovascular reactivity (CO2R), and cerebral blood flow (CBF) response to paced breathing
(vs. baseline) and breath holding (vs. paced breathing). E.g. Individuals scoring higher on the
DRS showed larger CVRi increases in response to paced breathing, and larger CVRi decreases in
response to breath holding.

* P < 0.05; ** P < 0.01; L= left; R = right; PC = posterior cingulate.

Reactivity
variable
Brain region Paced Breath Holding
CVRi R PC 0.02 (0.01)* -0.03 (0.01)*
CO2R R PC 0.06 (0.03)*
CBF
L PC 0.60 (0.24)*
R PC 0.35 (0.17)*

44
Table 9. Relationships between word-list memory and resting-state cerebrovascular
function.

Estimated beta weights (standard errors) generated by mixed models evaluating relationships
between Rey Auditory Verbal Learning Test (RAVLT) scores and resting state cerebrovascular
resistance, cerebrovascular reactivity (CO2R), and cerebral blood flow (CBF). E.g. Higher
RAVLT scores were associated with lower resting state CVRi.

* P < 0.05; ** P < 0.01; L= left; R = right; HC = hippocampus; MTL = medial temporal lobe;
PC = posterior cingulate; ITC = inferior temporal cortex.

Baseline
variable
Brain region
Short-term %
retention
Long-term %
retention
CVRi
R MTL -0.003 (0.00)*
L PC -0.01 (0.00)** -0.01 (0.00)**
R PC -0.004 (0.00)* -0.01 (0.00)*
L ITC -0.004 (0.00)**
R ITC -0.007 (0.00)*
CO2R
L HC 0.01 (0.00)*
R PC 0.01 (0.00)*
R precuneus 0.02 (0.00)**
L ITC 0.02 (0.01)*
CBF
Global 0.01 (0.00)*
L PC 0.07 (0.03)*
R precuneus 0.01 (0.00)**
L ITC 0.06 (0.02)**
R ITC 0.04 (0.02)*

45
Table 10. Relationships between word-list memory and cerebrovascular response to breathing manipulations.

Estimated beta weights (standard errors) generated by mixed models evaluating relationships between Rey Auditory Verbal Learning
Test (RAVLT) scores and cerebrovascular resistance (CVRi) and cerebral blood flow (CBF) response to paced breathing (vs.
baseline) and breath holding (vs. paced breathing). E.g. Individuals scoring higher on the RAVLT showed larger CVRi increases in
response to paced breathing.

* P < 0.05; ** P < 0.01; L= left; R = right; HC = hippocampus; MTL = medial temporal lobe; PC = posterior cingulate; ITC = inferior
temporal cortex.
Reactivity
variable
Brain region
Short-term % retention Long-term % retention
Paced Breath Holding Paced Breath Holding
CVRi
Global 0.004 (0.00)**
L PC 0.01 (0.00)**
R PC 0.01 (0.00)** 0.004 (0.00)*
L precuneus 0.01 (0.00)**
R precuneus 0.01 (0.00)**
R ITC 0.01 (0.00)*
CO2R
L HC -0.02 (0.01)*
L PC -0.02 (0.01)*
R PC -0.02 (0.01)*
R precuneus -0.02 (0.01)**
L ITC -0.02 (0.01)**
R MTL 0.21 (0.11)*
L PC 0.16 (0.07)* 0.17 (0.07)*
R PC 0.10 (0.05)* 0.11 (0.05)*
L precuneus 0.14 (0.05)* 0.13 (0.05)*
R precuneus 0.15 (0.06)* 0.14 (0.06)*
45

46
Table 11. Relationships between story memory and resting-state cerebrovascular function.

Estimated beta weights (standard errors) generated by mixed models evaluating relationships
between Logical Memory scores and resting state cerebrovascular resistance (CVRi) and
cerebrovascular reactivity (CO2R). E.g. Higher Logical Memory scores were associated with
lower baseline CVRi.

* P < 0.05; ** P < 0.01; L= left; R = right; HC = hippocampus; PC = posterior cingulate; ITC =
inferior temporal cortex.

Baseline
variable
Brain region Logical Memory I Logical Memory II
CVRi L PC -0.01 (0.00)*
CO2R
Global 0.04 (0.01)** 0.03 (0.01)**
L HC 0.02 (0.01)*
R HC 0.02 (0.01)*
L PC 0.03 (0.01)** 0.02 (0.01)*
L precuneus 0.02 (0.01)* 0.02 (0.01)*
R precuneus 0.03 (0.01)** 0.03 (0.01)**
R ITC 0.02 (0.01)** 0.02 (0.01)*
L PC 0.29 (0.10)** 0.27 (0.09)**
R PC 0.19 (0.07)** 0.18 (0.06)**

47
Table 12. Relationships between story memory and cerebrovascular response to breathing manipulations.

Estimated beta weights (standard errors) generated by mixed models evaluating relationships between Logical. Memory scores and
cerebrovascular resistance (CVRi), cerebrovascular reactivity (CO2R), and cerebral blood flow (CBF) response to paced breathing (vs.
baseline) and breath holding (vs. paced breathing). E.g. Higher Logical Memory scores were associated with larger CVRi increases in
response to paced breathing, and larger CVRi decreases in response to breath holding.

* P < 0.05; ** P < 0.01; L= left; R = right; HC = hippocampus; PC = posterior cingulate.

Reactivity
variable
Brain region
Logical Memory I Logical Memory II
Paced Breath Holding Paced Breath Holding
CVRi
R PC 0.01 (0.00)** -0.01 (0.00)* 0.01 (0.00)* -0.01 (0.00)*
L precuneus 0.01 (0.00)* 0.01 (0.00)
#

R ITC 0.01 (0.00)* 0.01 (0.00)*
CO2R
L HC 0.10 (0.03)**
R PC -0.02 (0.01)
#

R precuneus -0.03 (0.02)* -0.03 (0.01)*
CBF
L HC -0.36 (0.18)* -0.37 (0.16)*
L precuneus 0.27 (0.11)* 0.23 (0.10)*
R precuneus 0.25 (0.12)* 0.21 (0.11)*
47

48
Table 13. Relationships between executive function and resting-state cerebrovascular
function.

Estimated beta weights (standard errors) generated by mixed models evaluating relationships
between scores on tests of executive function (FAS Verbal Fluency and D-KEFS Category
Switching) and resting state cerebrovascular resistance (CVRi) and cerebrovascular reactivity
(CO2R). E.g. Higher FAS scores were associated with lower baseline CVRi.

* P < 0.05; ** P < 0.01; L= left; R = right; IFC = inferior frontal cortex.
Baseline
variable
Brain region FAS Switching
CVRi
L IFC -0.003 (0.00)* -0.01 (0.00)**
R IFC -0.01 (0.00)**
CO2R
Global 0.03 (0.01)**
L IFC 0.04 (0.01)**
R IFC 0.02 (0.01)*
R IFC 0.29 (0.11)*

49
Table 14. Relationships between executive function and cerebrovascular response to breathing manipulations.

Estimated beta weights (standard errors) generated by mixed models evaluating relationships between scores on tests of executive
function (FAS Verbal Fluency and D-KEFS Category Switching), and cerebrovascular resistance (CVRi) and cerebral blood flow
(CBF) response to paced breathing (vs. baseline) and breath holding (vs paced breathing). E.g. Individuals scoring higher on Category
Switching showed smaller CVRi increases in response to paced breathing.

* P < 0.05; ** P < 0.01; L= left; R = right; IFC = inferior frontal cortex.

Reactivity
variable
Brain region
FAS Switching
Paced Breath Holding Paced Breath Holding
CVRi L IFC -0.01 (0.00)*
CBF
Global 0.16 (0.08)* 0.52 (019)**
L IFC 0.46 (0.18)*
49

50
Table 15. Baseline demographic and physiological characteristics of APOE e4 carriers and
non-carriers.
e4 non-carriers e4 carriers t or χ
2
(df) P
n 25 14
Age 73.04 (8.53) 66.14 (8.4) 2.44 (37) 0.02
Education 16.8 (2.18) 15.5 (3.44) 1.28 (19) 0.22
SBP 131.33 (10.44) 136.82 (22.58) 0.8 (13.31) 0.44
DBP 75.30 (13.03) 79.92 (18.1) 0.89 (35) 0.38
PP 56.03 (16.21) 56.90 (12.23) 0.16 (35) 0.87
MAP 93.98 (9.54) 98.89 (18.84) 0.85 (13.78) 0.41
Antihypertensives
¥
12/13 2/12 4.43 (1) 0.04

Statistics reflect results of T tests and chi-square tests.

* P < 0.05; ** P < 0.01; df = degrees of freedom; SBP = systolic blood pressure; DBP = diastolic
blood pressure; PP = pulse pressure; MAP = mean arterial pressure.
¥
Participants taking/not
taking antihypertensive medication.

51
Table 16. Comparison of APOE e4 non-carriers to carriers for resting-state
cerebrovascular function.

Estimated beta weights (standard errors) generated by mixed models comparing APOE e4 non-
carriers to carriers on baseline cerebrovascular resistance (CVRi) and cerebrovascular reactivity
(CO2R). E.g. Higher resting-state CVRi was observed in non-carriers than carriers.

* P < 0.05; ** P < 0.01; L = left; R = right; IFC = inferior frontal cortex; ITC = inferior temporal
cortex; HC = hippocampus; MTL = medial temporal lobe.

Baseline
variable
Brain region Non carriers vs. carriers
CVRi
Global 0.07 (0.03)
L IFC 0.09 (0.04)*
L precuneus 0.20 (0.06)**
R precuneus 0.12 (0.05)*
R ITC 0.20 (0.08)*
CO2R
R HC -0.37 (0.16)*
L MTL -0.48 (0.17)**

52
Table 17. Comparison of APOE e4 non-carriers to carriers for cerebrovascular response to
breathing manipulations.

Estimated beta weights (standard errors) generated by mixed models comparing APOE e4 non-
carriers to carriers for cerebrovascular resistance index (CVRi), cerebrovascular reactivity
(CO2R), and cerebral blood flow (CBF) response to paced breathing (vs. baseline) and breath
holding (vs paced breathing). E.g. Non-carriers show stronger CVRi increases in response to
paced breathing, and CVRi decreases in response to breath holding.

* P < 0.05; ** P < 0.01; L = left; R = right; PC = posterior cingulate; IFC = inferior frontal
cortex; ITC = inferior temporal cortex; MTL = medial temporal lobe; HC = hippocampus.

Reactivity
variable
Brain region Paced BH
CVRi
L PC 0.12 (0.06)* -0.15 (0.06)*
R caudate -0.20 (0.06)**
L IFC -0.11 (0.04)*
R IFC -0.10 (0.05)*
L precuneus -0.13 (0.05)*
R precuneus -0.20 (0.06)**
R ITC 0.23 (0.09)** -0.18 (0.09)*
CO2R
L MTL 0.29 (0.14)*
R IFC 0.48 (0.14)** 0.36 (0.15)*
R precuneus 0.29 (0.13)*
CBF
L HC
-3.45 (1.71)*
L PC
-4.45 (1.97)*
R PC
-2.83 (1.4)*
L precuneus
-2.88 (1.45)*
3.94 (1.55)*
R precuneus

3.65 (1.80)*

53
References
Alwatban, M., Murman, D. L., & Bashford, G. (2019). Cerebrovascular Reactivity Impairment in
Preclinical Alzheimer's Disease. Journal of Neuroimaging, 29(4).
Arvanitakis, Z., Capuano, A. W., Leurgans, S. E., Bennett, D. A., & Schneider, J. A. (2016).
Relation of cerebral vessel disease to Alzheimer's disease dementia and cognitive
function in elderly people: a cross-sectional study. The Lancet Neurology, 15(9), 934-
943. doi:10.1016/s1474-4422(16)30029-1
Bangen, K. J., Clark, A. L., Edmonds, E. C., Evangelista, N. D., Werhane, M. L., Thomas, K. R.,
. . . Delano-Wood, L. (2017). Cerebral Blood Flow and Amyloid-β Interact to Affect
Memory Performance in Cognitively Normal Older Adults. Frontiers in Aging
Neuroscience, 9, Article 181.
Battisti-Charbonney, A., Fisher, J., & Duffin, J. (2011). The cerebrovascular response to carbon
dioxide in humans. The Journal of physiology, 589(Pt 12), 3039-3048.
Bright, M. G., Bulte, D. P., Jezzard, P., & Duyn, J. H. (2009). Characterization of regional
heterogeneity in cerebrovascular reactivity dynamics using novel hypocapnia task and
BOLD fMRI. NeuroImage, 48(1), 166-175.
Bu, G. (2009). Apolipoprotein E and its receptors in Alzheimer's disease: pathways,
pathogenesis, and therapy. Nature Reviews Neuroscience, 10, 333-344.
Buxton, R. (2009). An introduction to functional magnetic resonance imaging: Principles and
techniques (2nd ed.). New York, NY: Cambridge University Press.
Calviere, L., Catalaa, I., Marlats, F., Viguier, A., Bonneville, F., Cognard, C., & Larrue, V.
(2010). Correlation between cognitive impairment and cerebral hemodynamic
disturbances on perfusion magnetic resonance imaging in European adults with

54
moyamoya disease. Clinical article. J Neurosurg, 113(4), 753-759.
doi:10.3171/2010.4.jns091808
Cantin, S., Villien, M., Moreaud, O., Tropres, I., Keignart, S., Chipon, E., . . . Krainik, A. (2011).
Impaired cerebral vasoreactivity to CO2 in Alzheimer's disease using BOLD fMRI.
Neuroimage, 58(2), 579-587. doi:10.1016/j.neuroimage.2011.06.070
Cardenas, V. A., Reed, B., Chao, L. L., Chui, H., Sanossian, N., DeCarli, C. C., . . . Weiner, M.
W. (2012). Associations among vascular risk factors, carotid atherosclerosis, and cortical
volume and thickness in older adults. Stroke, 43(2865-2870).
Caselli, R. J., Dueck, A. C., Locke, D. E., Sabbagh, M. N., Ahern, G. L., Rapcsak, S. Z., . . .
Reiman, E. M. (2011). Cerebrovascular risk factors and preclinical memory decline in
healthy APOE epsilon4 homozygotes. Neurology, 76(12), 1078-1084.
doi:10.1212/WNL.0b013e318211c3ae
Castellano, J. M., Kim, J., Stewart, F. R., Jiang, H., DeMattos, R. B., Patterson, B. W., . . .
Holtzman, D. M. (2011). Human apoE Isoforms Differentially Regulate Brain Amyloid-β
Peptide Clearance. Science Translational Medicine, 3(89), 89ra57.
Catchlove, S. J., Parrish, T. B., Chen, Y., Macpherson, H., Hughes, M. E., & Pipingas, A.
(2018). Regional Cerebrovascular Reactivity and Cognitive Performance in Healthy
Aging. Journal of experimental neuroscience, 12, 1179069518785151-
1179069518785151.
Chao, L. L., Buckley, S. T., Kornak, J., Schuff, N., Madison, C., Yaffe, K., . . . Weiner, M. W.
(2010). ASL perfusion MRI predicts cognitive decline and conversion from MCI to
dementia. Alzheimer's Disease & Associated Disorders, 24(1), 19-27.

55
Chow, N., Bell, R. D., Deane, R., Streb, J. W., Chen, J., Brooks, A., . . . Zlokovic, B. V. (2007).
Serum response factor and myocardin mediate arterial hypercontractility and cerebral
blood flow dysregulation in Alzheimer's phenotype. Proceedings of the National
Academy of Sciences of the United States of America, 104(3), 823-828.
doi:10.1073/pnas.0608251104
Dai, W. Y., Lopez, O. L., Carmichael, O. T., Becker, J. T., Kuller, L. H., & Gachm, H. M. .
(2009). Mild cognitive impairment and Alzheimer disease: Patterns of altered cerebral
blood flow at MR imaging. Radiology, 250(856-866).
de Heus, R. A. A., de Jong, D. L. K., Sanders, M. L., van Spijker, G. J., Oudegeest-Sander, M.
H., Hopman, M. T., . . . Claassen, J. (2018). Dynamic Regulation of Cerebral Blood Flow
in Patients With Alzheimer Disease. Hypertension, 72(1), 139-150.
doi:10.1161/hypertensionaha.118.10900
Delis, D. C., Kaplan, E., & Kramer, J. H. (2001). Delis-Kaplan Executive Function System. San
Antonio, TX: The Psychological Corporation.
den Heijer, T., Skoog, I., Oudkerk, M., de Leeuw, F., de Groot, J. C., Hofman, A., & Breteler,
M. M. B. (2003). Association between blood pressure levels over time and brain atrophy
in the elderly. Neurobiology of Aging, 24(2), 307-313.
Dickerson, B. C., Salat, D. H., Greve, D. N., Chua, E. F., Rand-Giovannetti, E., Rentz, D. M., . .
. Sperling, R. A. (2005). Increased hippocampal activation in mild cognitive impairment
compared to normal aging and AD. Neurology, 65(3), 404-411.
Donahue, J. E., & Johanson, C. E. (2008). Apolipoprotein E, Amyloid-β, and Blood-Brain
Barrier Permeability in Alzheimer Disease. Journal of Neuropathology & Experimental
Neurology, 67(4), 261-270.

56
Duffin, J., Sobczyk, O., McKetton, L., Crawley, A., Poublanc, J., Venkatraghavan, L., . . .
Fisher, J. A. (2018). Cerebrovascular Resistance: The Basis of Cerebrovascular
Reactivity. Frontiers in Neuroscience, 12, 409.
Faraco, G., Park, L., Zhou, P., Luo, W., Paul, S. M., Anrather, J., & Iadecola, C. (2015).
Hypertension enhances Aβ-induced neurovascular dysfunction, promotes β-secretase
activity, and leads to amyloidogenic processing of APP. Journal of Cerebral Blood Flow
& Metabolism, 36(1), 241-252.
Fazekas, F., Chawluk, J. B., Alavi, A., Hurtig, H. I., & Zimmerman, R. A. (1987). MR signal
abnormalities at 1.5 T in Alzheimer's dementia and normal aging. American Journal of
Roentgenology, 149(2), 351-356. doi:10.2214/ajr.149.2.351
Filippini, N., Ebmeier, K. P., MacIntosh, B. J., Trachtenberg, A. J., Frisoni, G. B., Wilcock, G.
K., . . . Mackay, C. E. (2011). Differential effects of the APOE genotype on brain
function across the lifespan. Neuroimage, 54(1), 602-610.
doi:10.1016/j.neuroimage.2010.08.009
Fleisher, A. S., Podraza, K. M., Bangen, K. J., Taylor, C., Sherzai, A., Sidhar, K., . . . Buxton, R.
B. (2009). Cerebral perfusion and oxygenation differences in Alzheimer's disease risk.
Neurobiology of Aging, 30(11), 1737-1748.
Glodzik, L., Randall, C., Rusinek, H., & de Leon, M. J. (2013). Cerebrovascular reactivity to
carbon dioxide in Alzheimer's disease. Journal of Alzheimer's disease : JAD, 35(3), 427-
440.
Glodzik, L., Rusinek, H., Brys, M., Tsui, W. H., Switalski, R., Mosconi, L., . . . de Leon, M. J.
(2011). Framingham cardiovascular risk profile correlates with impaired hippocampal
and cortical vasoreactivity to hypercapnia. Journal of cerebral blood flow and

57
metabolism : official journal of the International Society of Cerebral Blood Flow and
Metabolism, 31(2), 671-679.
Gommer, E. D., Martens, E. G., Aalten, P., Shijaku, E., Verhey, F. R., Mess, W. H., . . . Reulen,
J. P. (2012). Dynamic cerebral autoregulation in subjects with Alzheimer's disease, mild
cognitive impairment, and controls: Evidence for increased peripheral vascular resistance
with possible predictive value. Journal of Alzheimers Disease, 30, 805-813.
Greicius, M. D., Srivastava, G., Reiss, A. L., & Menon, V. (2004). Default-mode network
activity distinguishes Alzheimer's disease from healthy aging: Evidence from functional
MRI. Proceedings of the National Academy of Sciences of the United States of America,
101(13), 4637-4642.
Guo, Z., Viitanen, M., Fratiglioni, L., & Winblad, B. (1996). Low Blood Pressure And Dementia
In Elderly People: The Kungsholmen Project. British Medical Journal, 312(7034), 805-
808.
Hajjar, I., Marmerelis, V., Shin, D. C., & Chui, H. (2014). Assessment of cerebrovascular
reactivity during resting state breathing and its correlation with cognitive function in
hypertension. Cerebrovascular diseases (Basel, Switzerland), 38(1), 10-16.
Hajjar, I., Zhao, P., Alsop, D., & Novak, V. (2010). Hypertension and cerebral vasoreactivity: A
continuous arterial spin labeling magnetic resonance imaging study. Hypertension, 56,
859-864.
Han, B. H., Zhou, M.-l., Johnson, A. W., Singh, I., Liao, F., Vellimana, A. K., . . . Zipfel, G. J.
(2015). Contribution of reactive oxygen species to cerebral amyloid angiopathy,
vasomotor dysfunction, and microhemorrhage in aged Tg2576 mice. Proceedings of the
National Academy of Sciences, 112(8), E881-E890.

58
Han, S. D., Bangen, K. J., & Bondi, M. W. (2009). Functional magnetic resonance imaging of
compensatory neural recruitment in aging and risk for Alzheimer's disease: review and
recommendations. Dement Geriatr Cogn Disord, 27(1), 1-10. doi:10.1159/000182420
Handwerker, D. A., Gazzaley, A., Inglis, B. A., & D'Esposito, M. (2007). Reducing vascular
variability of fMRI data across aging populations using a breathholding task. Hum Brain
Mapp, 28(9), 846-859. doi:10.1002/hbm.20307
Heckmann, J. G., Brown, C. M., Cheregi, M., Hilz, M. J., & Neundörfer, B. (2003). Delayed
Cerebrovascular Autoregulatory Response to Ergometer Exercise in Normotensive
Elderly Humans. Cerebrovascular Diseases, 16(4), 423-429.
Hofer, S. M., Piccinin, A. M., & Douglas, H. (1996). Analysis of structure and discriminative
power of the Mattis Dementia Rating Scale. Journal of Clinical Psychology, 52, 395-409.
Iadecola, C. (2004). Neurovascular regulation in the normal brain and in Alzheimer's disease.
Nature Reviews Neuroscience, 5, 347-360.
Iadecola, C., & Davisson, R. L. (2008). Hypertension and cerebrovascular dysfunction. Cell
Metabolism, 7, 476-484.
Iliff, J. J., Goldman, S. A., & Nedergaard, M. (2015). Implications of the discovery of brain
lymphatic pathways. The Lancet. Neurology, 14(10), 977. doi:10.1016/s1474-
4422(15)00221-5
Jiang, Q., Lee, C. Y. D., Mandrekar, S., Wilkinson, B., Cramer, P., Zelcer, N., . . . Landreth, G.
E. (2008). ApoE promotes the proteolytic degradation of Abeta. Neuron, 58(5), 681-693.
Johnson, N. A., Jahng, G., Weiner, M. W., Miller, B. L., Chui, H. C., Jagust, W. J., . . . Schuff,
N. (2005). Pattern of cerebral hypoperfusion in Alzheimer disease and Mild Cognitive

59
Impairment measured with arterial spin-labeling MR imaging: Initial experience.
Radiology, 234(3), 851-859.
Kanno, I., Uemura, K., Higano, S., Murakami, M., Iida, H., Miura, S., . . . Sayama, I. (1988).
Oxygen Extraction Fraction at Maximally Vasodilated Tissue in the Ischemic Brain
Estimated from the Regional CO2 Responsiveness Measured by Positron Emission
Tomography. Journal of Cerebral Blood Flow & Metabolism, 8(2), 227-235.
Kisler, K., Nelson, A. R., Montagne, A., & Zlokovic, B. V. (2017). Cerebral blood flow
regulation and neurovascular dysfunction in Alzheimer disease. Nature Reviews
Neuroscience, 18(7), 419-434.
Langbaum, J. B. S., Chen, K., Launer, L. J., Fleisher, A. S., Lee, W., Liu, X., . . . Reiman, E. M.
(2012). Blood pressure is associated with higher brain amyloid burden and lower glucose
metabolism in healthy late middle-age persons. Neurobiology of Aging, 33(4), 827.e811-
827.e819.
Lezak, M. D., Howieson, D. B., & Loring, D. W. (2004). Neuropsychological Assessment (4th
ed.). New York: Oxford University Press.
Ligthart, S. A., Moll van Charante, E. P., Van Gool, W. A., & Richard, E. (2010). Treatment of
cardiovascular risk factors to prevent cognitive decline and dementia: a systematic
review. Vascular health and risk management, 6, 775-785.
Liu, H. L., Huang, J., Wu, C. T., & Hsu, Y. Y. (2002). Detectability of blood oxygenation level-
dependent signal changes during short breath hold duration. Magnetic Resonance
Imaging, 20, 643-648.

60
Liu, J., Zhu, Y.-S., Khan, M. A., Brunk, E., Martin-Cook, K., Weiner, M. F., . . . Zhang, R.
(2014). Global brain hypoperfusion and oxygenation in amnestic mild cognitive
impairment. Alzheimer's & Dementia, 10(2), 162-170.
Liu, P., De Vis, J. B., & Lu, H. (2019). Cerebrovascular reactivity (CVR) MRI with CO2
challenge: A technical review. NeuroImage, 187(104–115).
Liu, P., Xu, C., Lin, Z., Sur, S., Li, Y., Yasar, S., . . . Lu, H. (2020). Cerebrovascular reactivity
mapping using intermittent breath modulation. 215, 116787.
Marmarelis, V. Z., Shin, D. C., Orme, M. E., & Zhang, R. (2013). Model-based quantification of
cerebral hemodynamics as a physiomarker for Alzheimer's disease? Annals of biomedical
engineering, 41(11), 2296–2317.
Marmarelis, V. Z., Shin, D. C., Tarumi, T., & Zhang, R. (2017). Comparison of Model-Based
Indices of Cerebral Autoregulation and Vasomotor Reactivity Using Transcranial
Doppler versus Near-Infrared Spectroscopy in Patients with Amnestic Mild Cognitive
Impairment. Journal of Alzheimer's disease : JAD, 56(1), 89-105.
Mattson, N., Tosun, D., Insel, P. S., Simonson, A., Jack, C. R., Beckett, L. A., . . . Weiner, M.
W. (2014). Association of brain amyloid-ß with cerebral perfusion and structure in
Alzheimer's disease and mild cognitive impairment. Brain, 137, 1550-1561.
McKetton, L., Sobczyk, O., Duffin, J., Poublanc, J., Sam, K., Crawley, A. P., . . . Mikulis, D. J.
(2018). The aging brain and cerebrovascular reactivity. NeuroImage., 181, 132-141.
Mearns, B. M., & Fuster, V. (2010). Highlight dementia risk to reduce CVD. 7(5), 237-237.
Mies, G., Ishimaru, S., Xie, Y., Seo, K., & Hossmann, K. A. (1991). Ischemic Thresholds of
Cerebral Protein Synthesis and Energy State Following Middle Cerebral Artery
Occlusion in Rat. J Cereb Blood Flow Metab, 11(5), 753-761.

61
Miller, K. B., Howery, A. J., Harvey, R. E., Eldridge, M. W., & Barnes, J. N. (2018).
Cerebrovascular reactivity and central arterial stiffness in habitually exercising healthy
adults. Frontiers in Physiology, 9.
Minhas, J. S., Haunton, V. J., Robinson, T. G., & Panerai, R. B. (2019). Determining differences
between critical closing pressure and resistance-area product: responses of the healthy
young and old to hypocapnia. Pflugers Arch, 471(8), 1117-1126. doi:10.1007/s00424-
019-02290-3
Montagne, A., Nation, D. A., Sagare, A. P., Barisano, G., Sweeney, M. D., Chakhoyan, A., . . .
Zlokovic, B. V. (2020). APOE4 leads to blood–brain barrier dysfunction predicting
cognitive decline. 581(7806), 71-76.
Morris, J. C. (1993). The Clinical Dementia Rating (CDR): Current version and scoring rules.
Neurology, 43, 2412-2414.
Muela, H. C. S., Costa-Hong, V. A., Yassuda, M. S., Moraes, N. C., Memória, C. M., Machado,
M. F., . . . Bortolotto, L. A. (2017). Hypertension Severity Is Associated With Impaired
Cognitive Performance. Journal of the American Heart Association, 6(1).
Nagata, K., Yamazaki, T., Takano, D., Maeda, T., Fujimaki, Y., Nakase, T., & Sato, Y. (2016).
Cerebral circulation in aging. Ageing Research Reviews, 30(Supplement C), 49-60.
Nation, D. A., Sweeney, M. D., Montagne, A., Sagare, A. P., D’Orazio, L. M., Pachicano, M., . .
. Zlokovic, B. V. (2019). Blood–brain barrier breakdown is an early biomarker of human
cognitive dysfunction. 25(2), 270-276.
Nation, D. A., Wierenga, C. E., Clark, L. R., Dev, S. I., Stricker, N. H., Jak, A. J., . . . Bondi, M.
W. (2013). Cortical and subcortical cerebrovascular resistance index in Mild Cognitive
Impairment and Alzheimer's disease. Journal of Alzheimer's Disease, 36, 689-698.

62
Nedergaard, M. (2013). Garbage Truck of the Brain. Science, 340(6140), 1529.
Niwa, K., Carlson, G. A., & Iadecola, C. (2000). Exogenous Aβ1–40 Reproduces
Cerebrovascular Alterations Resulting from Amyloid Precursor Protein Overexpression
in Mice. Journal of Cerebral Blood Flow & Metabolism, 20(12), 1659-1668.
Panerai, R. B., Deverson, S. T., Mahony, P., Hayes, P., & Evans, D. H. (1999). Effects of CO2
on dynamic cerebral autoregulation measurement. Physiol Meas, 20(3), 265-275.
doi:10.1088/0967-3334/20/3/304
Paulson, O. B., Strandgaard, S., & Edvinsson, L. (1990). Cerebral autoregulation.
Cerebrovascular and Brain Metabolism Reviews, 2, 161-192.
Peng, S., Chen, X., Li, Y., Rodrigue, K. M., Park, D. C., & Lu, H. (2018). Age-related changes
in cerebrovascular reactivity and their relationship to cognition: A four-year longitudinal
study. NeuroImage, 174, 257-262.
Petrovitch, H., White, L. R., Izmirilian, G., Ross, G. W., Havlik, R. J., Markesbery, W., . . .
Launer, L. J. (2000). Midlife blood pressure and neuritic plaques, neurofibrillary tangles,
and brain weight at death: The HAAS. Neurobiology of Aging, 21(1), 57-62.
Qiu, C., & Fratiglioni, L. (2015). A major role for cardiovascular burden in age-related cognitive
decline. 12(5), 267-277.
Qiu, C., von Strauss, E., Fastbom, J., Winblad, B., & Fratiglioni, L. (2003). Low Blood Pressure
and Risk of Dementia in the Kungsholmen Project: A 6-Year Follow-up Study. 60(2),
223-228.
Qiu, C., Winblad, B., & Fratiglioni, L. (2005). The age-dependent relation of blood pressure to
cognitive function and dementia. The Lancet Neurology, 4(8), 487-499.

63
Qiu, C., Xu, W., Winblad, B., & Fratiglioni, L. (2010). Vascular risk profiles for dementia and
Alzheimer’s disease in very old people: a population-based longitudinal study. J
Alzheimers Dis, 20.
Rabin, J. S., Yang, H. S., Schultz, A. P., Hanseeuw, B. J., Hedden, T., Viswanathan, A., . . .
Chhatwal, J. P. (2019). Vascular Risk and beta-Amyloid Are Synergistically Associated
with Cortical Tau. Ann Neurol, 85(2), 272-279. doi:10.1002/ana.25399
Rasmussen, P. M., Aamand, R., Weitzberg, E., Christiansen, M., Østergaard, L., & Lund, T. E.
(2019). APOE gene-dependent BOLD responses to a breath-hold across the adult
lifespan. NeuroImage. Clinical, 24, 101955-101955.
Raz, N., & Rodrigue, K. M. (2006). Differential aging of the brain: Patterns, cognitive correlates
and modifiers. Neuroscience & Biobehavioral Reviews, 30(6), 730-748.
Rey, A. (1941). L'examen psychologique dans les cas d'encéphalopathie traumatique. Archives
de Psychologie, 28, 215-285.
Richiardi, J., Monsch, A. U., Haas, T., Barkhof, F., Van de Ville, D., Radü, E. W., . . . Haller, S.
(2015). Altered cerebrovascular reactivity velocity in mild cognitive impairment and
Alzheimer's disease. Neurobiology of Aging, 36, 33-41.
Rivera-Rivera, L. A., Turski, P., Johnson, K. M., Hoffman, C., Berman, S. E., Kilgas, P., . . .
Wieben, O. (2016). 4D flow MRI for intracranial hemodynamics assessment in
Alzheimer's disease. Journal of Cerebral Blood Flow & Metabolism, 36(10), 1718-1730.
Rodrigue, K. M., Rieck, J. R., Kennedy, K. M., Devous, M. D., Sr., Diaz-Arrastia, R., & Park, D.
C. (2013). Risk factors for beta-amyloid deposition in healthy aging: vascular and genetic
effects. JAMA Neurol, 70(5), 600-606. doi:10.1001/jamaneurol.2013.1342

64
Roher, A. E., Garami, A., Tyas, S. L., Maarouf, C. L., Kokjohn, T. A., Belohlavek, M., . . .
Emmerling, M. R. (2011). Transcranial Doppler ultrasound blood flow velocity and
pulsatility index as systemic indicators for Alzheimer's disease. Alzheimer's & Dementia,
7(4), 445-455.
Silvestrini, M., Pasqualetti, P., Baruffaldi, R., Bartolini, M., Handouk, Y., Matteis, M., . . .
Vernieri, F. (2006). Cerebrovascular Reactivity and Cognitive Decline in Patients With
Alzheimer Disease. Stroke, 37(4), 1010.
Skoog, I. (2003). Highs and lows of blood pressure: a cause of Alzheimer's disease? Neurology,
2(6), 334.
Snyder, H. M., Corriveau, R. A., Craft, S., Faber, J. E., Greenberg, S. M., Knopman, D., . . .
Carrillo, M. C. (2015). Vascular contributions to cognitive impairment and dementia
including Alzheimer's disease. Alzheimers Dement, 11(6), 710-717.
doi:10.1016/j.jalz.2014.10.008
Sousa, I., Vilela, P., & Figueiredo, P. (2014). Reproducibility of hypocapnic cerebrovascular
reactivity measurements using BOLD fMRI in combination with a paced deep breathing
task. NeuroImage(1095-9572 (Electronic)).
Sun, Y.-w., Qin, L.-d., Zhou, Y., Xu, Q., Qian, L.-j., Tao, J., & Xu, J.-r. (2011). Abnormal
functional connectivity in patients with vascular cognitive impairment, no dementia: A
resting-state functional magnetic resonance imaging study. 223(2), 388-394.
Suri, S., Mackay, C. E., Kelly, M. E., Germuska, M., Tunbridge, E. M., Frisoni, G. B., . . .
Filippini, N. (2015). Reduced cerebrovascular reactivity in young adults carrying the
APOE ε4 allele. Alzheimer's & Dementia, 11, 648-657.

65
Tai, L. M., Ghura, S., Koster, K. P., Liakaite, V., Maienschein-Cline, M., Kanabar, P., . . . LaDu,
M. J. (2015). APOE-modulated Abeta-induced neuroinflammation in Alzheimer's
disease: current landscape, novel data, and future perspective. J Neurochem, 133(4), 465-
488. doi:10.1111/jnc.13072
Tai, L. M., Thomas, R., Marottoli, F. M., Koster, K. P., Kanekiyo, T., Morris, A. W., & Bu, G.
(2016). The role of APOE in cerebrovascular dysfunction. Acta Neuropathol, 131(5),
709-723.
Thomas, T., Thomas, G., McLendon, C., Sutton, T., & Mullan, M. (1996). β-Amyloid-mediated
vasoactivity and vascular endothelial damage. 380, 168.
Tombaugh, T. N., Kozak, J., & Rees, L. (1999). Normative Data Stratified by Age and Education
for Two Measures of Verbal Fluency: FAS and Animal Naming. Archives of Clinical
Neuropsychology, 14(2), 167-177.
Toth, P., Tarantini, S., Csiszar, A., & Ungvari, Z. (2017). Functional vascular contributions to
cognitive impairment and dementia: mechanisms and consequences of cerebral
autoregulatory dysfunction, endothelial impairment, and neurovascular uncoupling in
aging. American Journal of Physiology - Heart and Circulatory Physiology, 312(1), H1.
Toth, P., Tucsek, Z., Sosnowska, D., Gautam, T., Mitschelen, M., Tarantini, S., . . . Ungvari, Z.
(2013). Age-related autoregulatory dysfunction and cerebromicrovascular injury in mice
with angiotensin II-induced hypertension. J Cereb Blood Flow Metab, 33.
doi:10.1038/jcbfm.2013.143
Tuladhar, A. M., van Dijk, E., Zwiers, M. P., van Norden, A. G. W., de Laat, K. F., Shumskaya,
E., . . . de Leeuw, F.-E. (2016). Structural network connectivity and cognition in cerebral
small vessel disease. Human Brain Mapping, 37(1), 300-310.

66
van Beek, A. H. E. A., Claassen, J. A. H. R., Rikkert, M. G. M. O., & Jansen, R. W. M. M.
(2008). Cerebral Autoregulation: An Overview of Current Concepts and Methodology
with Special Focus on the Elderly. Journal of Cerebral Blood Flow and Metabolism, 28.
doi:10.1038/jcbfm.2008.13
Wang, Z. (2014). Characterizing early Alzheimer's disease and disease progression using
hippocampal volume and arterial spin labeling perfusion MRI. Journal of Alzheimer's
Disease, 42, S495-S502.
Weller, R. O., Boche, D., & Nicoll, J. A. R. (2009). Microvasculature changes and cerebral
amyloid angiopathy in Alzheimer's disease and their potential impact on therapy. Acta
Neuropathologica, 118, 87-102.
Weller, R. O., Subash, M., Preston, S. D., Mazanti, I., & Carare, R. O. (2008). Perivascular
drainage of amyloid-ß peptides from the brain and its failure in cerebral amyloid
angiopathy and Alzheimer's disease. Brain Pathology, 18, 253-266.
Wierenga, C. E., Clark, L. R., Dev, S. I., Shin, D. D., Jurick, S. M., Rissman, R. A., . . . Bondi,
M. W. (2013). Interaction of age and APOE genotype on cerebral blood flow at rest.
Journal of Alzheimers Disease, 34.
Wierenga, C. E., Dev, S. I., Shin, D. D., Clark, L. R., Bangen, K. J., Jak, A. J., . . . Bondi, M. W.
(2012). Effect of mild cognitive impairment and APOE genotype on resting cerebral
blood flow and its association with cognition. Journal of Cerebral Blood Flow &
Metabolism, 32(8), 1589-1599. doi:10.1038/jcbfm.2012.58
Wu, Z., Guo, H., Chow, N., Sallstrom, J., Bell, R. D., Deane, R., . . . Zlokovic, B. V. (2005).
Role of the MEOX2 homeobox gene in neurovascular dysfunction in Alzheimer disease.

67
Nat Med, 11(9), 959-965.
doi:http://www.nature.com/nm/journal/v11/n9/suppinfo/nm1287_S1.html
Yew, B., & Nation, D. A. (2017). Cerebrovascular resistance and cerebral amyloidosis: effects
on cognitive decline, cortical atrophy, and progression to dementia. Brain, 140, 1987-
2001.
Zazulia, A. R., Videen, T. O., Morris, J. C., & Powers, W. J. (2010). Autoregulation of cerebral
blood flow to changes in arterial pressure in mild Alzheimer's disease. Journal of
Cerebral Blood Flow and Metabolism, 30(11), 1883-1889.
doi:http://dx.doi.org/10.1038/jcbfm.2010.135
Zlokovic, B. V. (2011). Neurovascular pathways to neurodegeneration in Alzheimer’s disease
and other disorders. Nat Rev Neurosci, 12.

Abstract (if available)

Abstract Cerebrovascular contributions to cognitive decline have been increasingly recognized. Evaluation of dynamic cerebrovascular function, however, has been restricted to animal paradigms, invasive manipulations, and/or methods offering suboptimal measurement of microvascular function. Breath holding and hyperventilation elicit alterations in end-tidal carbon dioxide (etCO₂) and associated hemodynamic variables. We therefore developed and validated a neuroimaging paradigm incorporating breathing manipulations for assessment of dynamic cerebrovascular function. Community-dwelling younger (n = 27) and older (n = 40) adults performed guided breath holds and hyperventilatory paced breathing while undergoing pseudo-continuous arterial spin labelling (pCASL) MRI. Continuous blood pressure and capnography were simultaneously collected and synchronized with pCASL-derived cerebral blood flow (CBF). Cerebrovascular resistance index (CVRi) was calculated by dividing mean arterial blood pressure by CBF. Cerebrovascular reactivity (CO₂R) was calculated by dividing change in CBF by change in etCO₂. Older adults also completed cognitive testing, T2 FLAIR imaging (from which white matter hyperintensity [WMH] burden was determined), and APOE genotyping. Across age groups, breath holding led to increased etCO₂, CBF, and CO₂R, and decreased CVRi. Paced breathing led to decreased etCO₂, CBF, and CO₂R, and increased CVRi. Older adults exhibited diminished hemodynamic response to both breathing manipulations. Attenuated response was associated with poorer cognitive performance and APOE ε4. Lower resting-state CO₂R, CVRi, and CBF were associated with poorer cognitive performance, APOE ε4, and greater WMH severity. Findings validate our paradigm as an effective measure of hemodynamic changes associated with dementia risk factors in non-clinical populations. This may facilitate detection of cerebrovascular changes characterizing the earliest stages of pathological aging.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Indexing cerebrovascular resistance in cognitive decline and Alzheimer's disease

PDF

The role of blood pressure variability in cognitive decline, cerebrovascular disease and Alzheimer’s disease

PDF

A virtual reality exergaming system to enhance brain health in older adults at risk for Alzheimer’s disease

PDF

Cross-sectional association of blood pressure, antihypertensive medications, MRI volumetric measures and cognitive function scores in an aging population

Asset Metadata

Creator Yew, Belinda (author)

Core Title Contributions of dynamic cerebrovascular function to cognitive decline and dementia: development and validation of a novel neuroimaging approach

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Publication Date 07/12/2020

Defense Date 06/09/2020

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag aging,blood pressure,cerebral blood flow,cerebrovascular reactivity,cerebrovascular resistance,cognitive decline,dementia,OAI-PMH Harvest

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Nation, Daniel (committee chair), Beam, Christopher (committee member), Han, Duke (committee member), Marmarelis, Vasilis (committee member), Mather, Mara (committee member)

Creator Email belinday@usc.edu,bxy2000@tc.columbia.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-326788

Unique identifier UC11663457

Identifier etd-YewBelinda-8657.pdf (filename),usctheses-c89-326788 (legacy record id)

Legacy Identifier etd-YewBelinda-8657.pdf

Dmrecord 326788

Document Type Dissertation

Rights Yew, Belinda

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