ALZHEIMER’S DISEASE AND MILD COGNITIVE IMPAIRMENT …

The Pennsylvania State University

The Graduate School

College of Medicine

FUNCTIONAL OLFACTORY DEFICITS IN THE OLFACTORY SYSTEM OF

ALZHEIMER’S DISEASE AND MILD COGNITIVE IMPAIRMENT PATIENTS AS A

POTENTIAL DIAGNOSTIC MARKER

A Dissertation in

Neuroscience

by

Megha Vasavada

2014 Megha Vasavada

Submitted in Partial Fulfillment

of the Requirements

for the Degree of

Doctor of Philosophy

August 2014

ii

The dissertation of Megha Vasavada was reviewed and approved* by the following:

Qing X. Yang

Professor of Radiology and Neurosurgery

Dissertation Advisor

Chair of Committee

Paul J. Eslinger

Professor of Neurology

Ralph Norgren

Professor of Neural and Behavioral Sciences

Patricia Grigson

Professor of Neural and Behavioral Sciences

Co-Chair of Neuroscience Graduate Program

Colin Barnstable

Professor and Chair of Neural and Behavioral Sciences

*Signatures are on file in the Graduate School

iii

ABSTRACT

Alzheimer's disease affects 5.4 million individuals in the US, causing debilitating

memory and cognitive impairment. By the time the disease has clinically manifested itself, the

pathology has progressed to the neocortex. Currently, early diagnosis and understanding of

Alzheimer’s pathology on functional deficits is the key for the development of therapy. While

volumetric measurements of the hippocampus provide excellent diagnostic assistance, post-

mortem studies have shown that the earliest pathological markers of Alzheimer’s (amyloid beta

plaques and neurofibrillary tangles) are found first in olfactory areas of the brain. Clinically,

olfaction is affected in the earliest stages of Alzheimer’s disease and mild cognitive impaired

(MCI) patients, a group considered to be at the highest risk for Alzheimer’s. Often olfactory

deficits appear prior to the manifestation of cognitive symptoms.

Magnetic resonance imaging (MRI) provides the ability to noninvasively examine the

functional and structural changes that occur prior to presentation of behavioral symptoms in

individuals with Alzheimer’s disease and MCI. Therefore, in this dissertation, MRI techniques

were utilized to investigate the involvement of the primary olfactory cortex in Alzheimer’s

disease and MCI subjects, and to determine the sensitivity of these techniques as potential

diagnostic markers of disease. The same subjects were used in each analysis, therefore the subject

information, behavioral tests, and data collection is the same for each chapter.

In chapter 2, we used both volumetric and functional MRI (fMRI) measurements to study

the diagnostic potential by investigating the primary olfactory cortex. Behavioral tests, including

the University of Pennsylvania Smell Identification Test (UPSIT), as well as cognitive tests

demonstrated olfactory and memory impairments in both Alzheimer’s and MCI patient groups

(one-way Analysis of Variance (ANOVA), P < 0.0001). The volumetric MRI of the primary

iv

olfactory cortex showed decreased volume in both Alzheimer’s and MCI subjects compared with

age-matched normal controls (one-way ANOVA, P < 0.001). However in terms of both volume

measurement and behavioral performance, MCI values ranged between those of Alzheimer’s and

those of normal controls.

On the other hand, the olfactory fMRI results showed that activation signal change in the

primary olfactory cortex was significantly and nearly equally decreased in the Alzheimer’s and

MCI subjects when compared with normal controls (one-way ANOVA, P < 0.0001). This

suggests that although behavioral and volumetric measurement may be variable in MCI subjects,

their activation signal in the brain is already changing. We also established that combining the

UPSIT score, hippocampal volume, and activation signal change in the primary olfactory cortex

increases the diagnostic specificity and sensitivity of Alzheimer’s and MCI.

In chapter 3, the dominant role of the central olfactory system in Alzheimer’s and MCI

was established. Whether olfactory deficits in Alzheimer’s disease and MCI are more dominantly

due to peripheral or central olfactory system deterioration is unclear. While several studies agree

with central olfactory system deterioration based on observations, olfactory fMRI and

pathological evidence are inconclusive. The olfactory paradigm used in this study had a visual

cue ―Smell?‖ accompanied by either odor presentation or no odor presentation. The presentation

with ―Smell?‖ without congruent odor presentation allowed for analysis of the primary olfactory

cortex with an afferent stimulus that was perceived as equal to the subjects. The visual and motor

systems were not impaired in AD and MCI subjects therefore all no stimulus provided could be

perceived as unequal. We hypothesized that if the dysfunction is outside the brain similar

activation signal change would be observed in all three groups when the visual cue ―Smell?‖ was

presented without congruent odor and group differences would only be found when the visual cue

and odor were presented congruently (this is when stimulus is perceived differently between the

v

groups since both MCI and AD subjects have trouble with olfactory function). Without an

odorant; however, the normal controls exhibited greater activation signal change compared with

both the Alzheimer’s and MCI subjects (one-way ANOVA, P < 0.05). This suggested central

olfactory system dominance; however, our study was not able to disprove concomitant

dysfunction of the peripheral olfactory system in Alzheimer’s and MCI.

In chapter 4, functional connectivity analysis was performed on our olfactory fMRI data

to learn further about the olfactory network. Functional connectivity is defined as the correlation

of interregional neural interactions during particular tasks or from spontaneous activity during

rest. We observed functional connectivity of the piriform was decreased to the striatum, thalamus,

and anterior cingulate cortex for both Alzheimer’s and MCI subjects (ANOVA, P < 0.001). The

Alzheimer’s group trended toward greater disconnection of the olfactory network compared with

MCI subjects, although the difference did not achieve statistical significance. The trend toward

preservation of connectivity in MCI subjects may explain their observed higher behavioral

function.

Therefore, we conclude that the central olfactory system is the dominant system involved

in Alzheimer’s and MCI patients, and is causing olfactory deficits. We demonstrated that fMRI

showed decreased activation in the primary olfactory cortex of MCI subjects, and was in fact

similar to the decreased activation of Alzheimer’s disease subjects. This indicates consistent early

functional changes in the brains of MCI subjects despite variability in their behavioral and

volumetric measurements. fMRI, thus has great potential to be used as an early diagnostic marker

in Alzheimer’s disease and MCI, and may also be used to study the progression of disease.

vi

TABLE OF CONTENTS

List of Figures .......................................................................................................................... ix

List of Tables ........................................................................................................................... x

List of Abbreviations ............................................................................................................... xi

Acknowledgements .................................................................................................................. xiii

Chapter 1 Introduction ............................................................................................................. 1

1.1 Alzheimer’s Disease .................................................................................................. 1

1.2 Mild Cognitive Impairment........................................................................................ 5

1.3 Anatomy of the Olfactory System .............................................................................. 5

1.4 Olfactory Deficits in Alzheimer’s Disease................................................................. 8

1.5 Pathological Changes in Olfactory Areas .................................................................. 11

1.5.1 Neurofibrillary Tangles ...................................................................................... 11

1.5.2 Amyloid Beta Plaques ....................................................................................... 12

1.5.3 Atrophy .............................................................................................................. 12

1.6 Neuroimaging in Alzheimer’s Disease ...................................................................... 13

1.5.1 Volumetric Studies ............................................................................................ 15

1.5.2 Functional MRI Studies ..................................................................................... 16

1.7 Central versus Peripheral Olfactory dysfucntion in Alzheimer’s Disease ................. 18

1.8 Rationale .................................................................................................................... 19

1.9 References .................................................................................................................. 22

Chapter 2 Functional and structural degeneration of the primary olfactory cortex in AD

and MCI ........................................................................................................................... 33

2.1 Abstract .................................................................................................................... 33

2.2 Introduction ................................................................................................................ 34

2.3 Methods ...................................................................................................................... 36

2.3.1 Study Cohort ...................................................................................................... 36

2.3.2 Behavioral Tests ................................................................................................ 38

2.3.3 Olfactory Stimulation Paradigm ........................................................................ 38

2.3.4 Imaging Protocol ................................................................................................ 41

2.3.5 fMRI Data Processing and Analysis .................................................................. 41

2.3.6 Region of Interest Analysis of the Primary Olfactory Cortex and

Hippocampus .................................................................................................... 42

2.4 Results ........................................................................................................................ 44

2.4.1 Demographics and Behavioral Results .............................................................. 44

2.4.2 Aging Effect ....................................................................................................... 46

2.4.3 Olfactory fMRI .................................................................................................. 46

vii

2.4.4 Relations of Brain Volume and Activation Volume in the Primary

Olfactory Cortex and Hippocampus ............................................................... 49

2.4.5 Correlation Between the Behavioral and MRI Results ...................................... 51

2.4.6 Logistic Regression Analysis ............................................................................. 53

2.5 Discussion .................................................................................................................. 55

2.6 References .................................................................................................................. 60

Chapter 3 Functional Connectivity of the Piriform is Disrupted in AD and MCI ................... 65

3.1 Abstract .................................................................................................................... 65

3.2 Introduction ................................................................................................................ 66

3.3 Methods ...................................................................................................................... 68

3.3.1 Study Cohort ...................................................................................................... 68




3.3.5 fMRI Data Processing and Analysis .................................................................. 69

3.3.6 Region of Interest Analysis of the Primary Olfactory Cortex and

Hippocampus .................................................................................................... 70

3.4 Results ........................................................................................................................ 70


3.4.2 Aging Effect ....................................................................................................... 71

3.4.3 Olfactory fMRI .................................................................................................. 71

3.4.4 Correlation Between the Behavioral and MRI Results ...................................... 75

3.4.5 Four Lavender Concentrations ........................................................................... 75

3.5 Discussion .................................................................................................................. 78

3.6 References .................................................................................................................. 86

Chapter 4 Central Olfactory Dysfunction is the Dominant Cause of Olfactory Deficits in

AD and MCI .......................................................................................................... 90

4.1 Abstract .................................................................................................................... 90

4.2 Introduction ................................................................................................................ 91

4.3 Methods ...................................................................................................................... 93

4.3.1 Study Cohort ...................................................................................................... 93




4.3.5 Functional Connectivity Analysis ...................................................................... 94

4.3.6 Statistical Analysis ............................................................................................. 97

4.4 Results ........................................................................................................................ 97


4.4.2 Functional Connectivity of the Piriform ............................................................ 97

4.4.3 Lateralization of Connectivity ........................................................................... 101

viii

4.4.4 Correlation of Functional Connectivity to the University of Pennsylvania

Smell Identification Test and Cognitive Tests .................................................... 105

4.5 Discussion .................................................................................................................. 107

4.6 References .................................................................................................................. 112

Chapter 5 Conclusion ............................................................................................................... 121

5.1 Olfactory System in Alzheimer’s Disease ............................................................... 121

5.2 Olfactory fMRI Paradigm ........................................................................................ 122

5.3 Central Olfactory System Dysfunction Causes Olfactory Symptoms ...................... 123

5.4 Volumetric Measurements ....................................................................................... 124

5.5 Olfactory fMRI ........................................................................................................ 125

5.6 Future Studies .......................................................................................................... 126

5.7 Summary .................................................................................................................. 127

5.8 References ................................................................................................................ 129

ix

LIST OF FIGURES

Figure 1-1. Pathological stages. ............................................................................................... 4

Figure 1-2. Human olfactory system........................................................................................ 7

Figure 2-1. Olfactory fMRI paradigm...................................................................................... 40

Figure 2-2. 3D display of the primary olfactory cortex. .......................................................... 43

Figure 2-3. Olfaction and cognitive tests. ................................................................................ 45

Figure 2-4. Activation in the primary olfactory cortex and hippocampus. .............................. 47

Figure 2-5. Activation volume in AD and MCI. ...................................................................... 48

Figure 2-6 Structural and functional changes .......................................................................... 50

Figure 2-7. Receiver operating characteristic (ROC) curves. .................................................. 54

Figure 3-1. Olfactory activation maps. .................................................................................... 73

Figure 3-2. Activated volume. ................................................................................................. 74

Figure 3-3. Four concentrations ............................................................................................... 77

Figure 3-4. Olfactory fMRI paradigm with and without olfactory stimulation ....................... 81

Figure 3-5. Hemodynamic response function (HRF).. ............................................................. 83

Figure 4-1. Functional connectivity of the piriform. ............................................................... 99

Figure 4-2. Functional connectivity disruption ........................................................................ 100

Figure 4-3. Olfactory network matrix. ..................................................................................... 102

Figure 4-4. Lateralization of olfactory network ....................................................................... 104

Figure 4-5. Correlations between smell and functional connectivity. ..................................... 106

x

LIST OF TABLES

Table 1-1 Clinical stages of Alzheimer’s disease .................................................................... 3

Table 2-1 Demographic and behavioral data of the study cohort ............................................ 37

Table 2-2. Correlations between behavioral and MRI measurements of all subjects.. ............ 52

Table 3-1. Correlations between behavioral and imaging measurements of all subjects.. ...... 76

Table 4-1. Anatomically defined regions of interest. ............................................................... 96

xi

LIST OF ABBREVIATIONS

AD Alzheimer’s Disease

MCI Mild Cognitive Impairment

MRI Magnetic Resonance Imaging

FMRI Functional Magnetic Resonance Imaging

UPSIT University of Pennsylvania Smell Identification Test

HRF Hemodynamic Response Function

POC Primary Olfactory Cortex

ApoE e4 Apolipoprotein E epsilon 4

NFT Neurofibrillary Tangles

PET Positron Emission Tomography

ROI Region of Interest

CDR Clinical Dementia Rating Scale

CN Cognitively Normal Controls

MMSE Mini-Mental State Examination

DRS-2 Dementia Rating Scale-2

CVLT-II California Verbal Learning Test-Second Edition Short Form

BOLD Blood Oxygen Level Dependent

FOV Field of View

TE Echo Time

TR Repetition Time

TA Acquisition Time

FA Flip Angle

IT Inversion Time

xii

SPM Statistical Parametric Mapping

MNI Montreal Neurological Institute

FSLVIEW FMRIB Software Library View

ROC Receiver Operating Characteristic

ANOVA Analysis of Variance

DPARSF Data Processing Assistant for Resting-State Fmri

REST Resting-State Fmri Data Analysis Toolkit

FWE Family Wise Error Corrected

DTI Diffusion Tensor Imaging

xiii

ACKNOWLEDGEMENTS

I would like extend my most sincere gratitude to my advisor, Dr. Qing Yang, along with

my other committee members, Dr. Grigson, Dr. Norgren, and Dr. Eslinger. Your mentorship

during my time at Penn State has largely contributed to my development as a scientist and has

been invaluable. I would also like to thank my colleagues at the NMR Center for providing

guidance, discussion, and friendship.

Last, but not least, I would like to thank my amazing family and friends for their

continued support and love, not only during my time as a graduate student but through all of my

endeavors. I am eternally grateful to my parents and my brother for fully encouraging all of my

dreams. To my mother-in-law, you have and continue to be a wonderful role model for me.

Thank you for that and your endless wisdom. Finally to my husband Rahul, I cannot thank you

enough for being with me every step of the way and for your infinite love, encouragement, and

support.

1

Chapter 1

Introduction

1.1 Alzheimer’s Disease

Alzheimer’s disease (AD) is the most common form of dementia affecting 5.4 million

individuals in the United States alone, making it the nation’s sixth leading cause of death [1].

While other leading causes of death have decreased in prevalence, deaths from Alzheimer’s

disease have actually increased by 68% from 2000 to 2010 [2]. There are several factors

contributing to the increased mortality from AD including increased life expectancy as well as

lack of a cure, preventive precautions, or drug therapies to stop the progression of the disease. In

addition, many drug therapies that are currently available or in clinical testing are most effective

only in the earliest stages of Alzheimer’s. Currently Alzheimer’s disease is diagnosed based on

patients’ cognitive symptoms primarily manifested as memory loss. Alzheimer’s patients can be

placed into three stages; mild/early, moderate, and advanced (Table 1-1). However by the time

most patients are cognitively impaired, the neuropathology of AD has already reached stages

three and four of the six pathological stages of Alzheimer’s disease (Fig. 1-1). In stages three and

four, the characteristic amyloid beta plaques and neurofibrillary tangles of AD have already

reached the neocortex [3]. In contrast, stages one and two are considered preclinical and

symptomatically silent [4], although early pathologic changes do begin in the entorhinal cortex

during stage one and later progress to the neocortical areas, and ultimately encompassing the

majority of the brain [5]. While we know this pathology exists through post-mortem examination

during stage one and two, noninvasive methods are not available to examine the pathology in

potential patients. Thus finding a noninvasive marker for detection during the early stages of AD

2

is critical in allowing earlier possible targeted drug therapy which can prevent progression of the

disease, and in ultimately unlocking a cure.

More recently, the olfactory system has become a focus in Alzheimer’s disease due to the

high correlation between the location of the olfactory processing regions and the pattern of

pathology in stages one and two [3, 6-11]. The olfactory areas greatly overlap with the first areas

showing amyloid beta plaques [12-13] and neurofibrillary tangles [14-16]. Olfaction has also

become a focus in AD patients because these patients display olfactory deficits during the early

stages of the disease [17-19]. Behavioral olfactory testing, neuroimaging, and postmortem

evidence exist for the involvement of the olfactory system in AD [3-19]. While we know that AD

patients have olfactory deficits and post-mortem studies at different stages have shown

development of the pathology first occurs in the regions involved in olfaction, it is unclear if this

deficit can be utilized as an early diagnostic marker. Hence, in this dissertation, we focus on the

involvement of the olfactory system in Alzheimer’s disease at the neuroimaging level in order to

investigate the diagnostic potential of investigating the dysfunction of the olfactory system as an

early marker of AD.

3

Table 1-1. Clinical stages of Alzheimer’s disease [4-5]

Clinical stages Symptoms Pathological stage correlate

Mild/Early stage Learning and memory impairment

language, executive functions,

perception (agnosia), or execution

of movements (apraxia), olfactory

difficulties

Diagnosis mostly occurs here

Stages 1 and 2:

Trans-entorhinal and

entorhinal cortex

Moderate Deterioration progresses and

hinders independence

Long term memory is affected

Reading/writing skills lost

Motor function becomes less

coordinated

Behavioral and neuropsychiatric

deteriorations seen at this stage

Stages 3 and 4:

Severe damage to the Trans-

entorhinal and entorhinal

cortex

Hippocampus

Neocortex

Advanced Dependent on care-giver

Need help with all basic activities

of daily living

Lose ability to communicate

Unable to recognize loved ones

Stages 5 and 6

Severe damage to the

neocortex and hippocampus

Motor and sensory fields

4

Figure 1-1. Pathological stages. Development of neurofibrillary changes in 2,661 nonselected

autopsy cases by Braak and Braak [5]. Stages III and IV are generally when diagnosis of

Alzheimer’s occurs. Figure modified from Braak and Braak 1991 [3].

5

1.2 Mild Cognitive Impairment

Mild cognitive impairment is defined as a greater cognitive decline than expected for the

individual’s age and education level and is prevalent in 3 to 16% of individuals over the age of 65

years [20]. It is considered the transitional stage between normal functioning and Alzheimer’s.

With a 15% annual rate of conversion, MCI patients are at the highest risk for developing AD

[21]. Not all MCI patients, however, will develop AD, some may develop AD and other forms of

dementia (about 11-33% develop dementia in a 2 year period), some will never progress, and

others about 44% diagnosed with MCI at the initial visit will revert back to normal cognitive

function [20]. Similar to AD, there is no cure, preventive precautions, or drug therapies to stop

the progression of the disease. MCI diagnosis is not well defined and cognitive decline as

described under MCI does not generally impede day to day functioning, therefore, many

individuals with cognitive issues above the norm for their age will not be diagnosed. It is also not

understood which MCI patients will move on to develop AD, other forms of dementia, remain

MCI, or revert to normal functioning. This group is the continuum from normal functioning to

dementia and research should be focused on individuals with MCI or likely to develop MCI for

early intervention. Even without therapy, proper diagnosis of MCI is important. It allows the

patient to prepare for the future and participate in clinical trials and research studies.

1.3 Anatomy of the Olfactory System

The olfactory system is responsible for identifying and detecting odorants and it is able to

detect and discriminate between multitudes of molecules. The olfactory system is unique in that is

different from the sensory systems in three fundamental ways [22]. One, unlike other sensory

systems, it connects to the cortex via two pathways: one that travels directly to the cortex without

6

a relay through the thalamus and another indirect pathway that relays through the thalamus.

Second, the neural integration and analysis of olfactory stimuli may only be topographically

organized in the epithelium and olfactory bulb. Beyond the olfactory bulb, there is no evidence of

topographic evidence. Third, olfactory receptors are constantly replaced by mitotic division of the

basal stem cell population. This may be due to the fact that they are the only neurons that are

directly exposed to the environment. The olfactory system includes the olfactory epithelium,

olfactory nerve, olfactory bulb, olfactory tract, olfactory tubercule, olfactory cortex, amygdala,

hippocampus, orbitofrontal cortex, and the thalamic dorsomedial nucleus. These numerous

components can be divided into peripheral (outside the brain) and central olfactory systems (Fig.

1-2). The peripheral portion is involved in the detection of external odorant and includes the

olfactory epithelium and olfactory nerve. The central olfactory system is involved in integrating

and processing the signal and includes the olfactory bulb, olfactory tract, olfactory cortex

(piriform), anterior olfactory nucleus, amygdala, olfactory tubercule, hippocampus, and the

orbitofrontal cortex [23]. Many of these areas such as the amygdala, hippocampus, and the

orbitofrontal cortex are involved in other functions as well.

The olfactory receptors in the olfactory epithelium are the first regions contacted by the

odorant molecules and these receptors project to the mitral cells of the olfactory bulb. The

olfactory bulb is the first central olfactory system structural and also the first processing station in

the olfactory pathway [23]. The axons from these mitral cells then travel to the brain via the

olfactory tract and project primarily to the primary olfactory cortex (POC), located within the

medial temporal lobe. The primary olfactory cortex includes the piriform cortex, entorhinal

cortex, anterior cortical nucleus of the amygdala, and the periamygdaloid cortex. Neurons from

the POC send projections to the dorsomedial nucleus of the thalamus, the nucleus accumbens,

putamen, caudate, and the hippocampus. It is believed that the thalamic connections serve as a

7

conscious mechanism for odor perception, while the amygdala and entorhinal areas (components

of the limbic system) are more involved in the affective components of olfaction. [22-25]

Figure 1-2. Human olfactory system. A) Odorants are transduced at the olfactory epithelium (1).

Different receptor types (three illustrated, 1,000 in mammals) converge via the olfactory nerve

onto common glomeruli at the olfactory bulb (2). From here information is conveyed via the

lateral olfactory tract to the primary olfactory cortex (3). Information is further relayed

throughout the brain, most notably to the orbitofrontal cortex (5) via a direct and indirect route

through the thalamus (4) [5]. B) Schematic representation of the principal human olfactory

pathways [6]. Figure A taken from Sela and Sobel [24] and figure B taken from Tham et al [25].

8

1.4 Olfactory Deficits in Alzheimer’s Disease

The evaluation of olfactory deficits has been completed using several smell tests

including the University of Pennsylvania Smell Identification (UPSIT), Sniffin’ Sticks, and the

Connecticut Chemosensory Clinical Research Center Test. The UPSIT is a 40 question, self-

administered, scratch and sniff smell test with high test-retest reliability [26]. The Sniffin’ Sticks

test is based on administration from a pen-like device and tests detection threshold, odorant

memory, and odor identification [27]. Finally, the Connecticut Chemosensory Clinical Research

Center Test is an identification test composed of seven stimuli [28].

In some cases of olfactory deterioration, the dysfunction is associated with

neurodegenerative diseases such as AD, Parkinson’s disease, multiple sclerosis, and Huntington’s

disease [17-18]. Individuals with these disorders consistently display either deficit in threshold

detection, odorant identification, and/or odorant memory relative to age-matched controls [29-

31]. Specifically, patients with AD show dysfunction in all of the areas listed compared with age-

matched normal controls. In one of the earliest studies examining olfaction in AD, Serby et al

showed that AD subjects performed at a significantly lower level on an identification smell task

compared with patients with alcoholic dementia, non-demented alcoholics, young controls, and

older controls [32]. Since then, several studies have revealed that AD individuals have

significantly decreased olfactory performance, specifically in odor detection and identification

[33-41]. In addition, a more recent meta-analysis of 39 studies from 1970 to 2011 also revealed

that AD patients have greater dysfunction in smell identification and recognition tasks than in

detection tasks [18]. This is because greater cognitive function is needed to perform the

identification task compared with the odorant detection task. These olfactory impairments

displayed in AD individuals correlate with cognitive decline and with the progression of the

disease [19]. Olfactory deficits appear in the early stages of AD and increase in congruence with

9

the rise in cognitive and memory impairments as well as with the severity of dementia [19, 29-30,

42].

Previous studies have also revealed olfactory dysfunction in those at risk for developing

AD [24] such as patients with mild cognitive impairment (MCI), which is considered to be the

transitional stage between normal aging and AD [21]. With a 15% annual rate of conversion,

MCI patients are at the highest risk for developing AD [21]. In a longitudinal study, Devanand et

al evaluated 90 MCI patients using the UPSIT and concluded that olfactory deficits in MCI

patients predicted AD at follow-up [43]. 19 out of 47 MCI subjects with low olfactory

performance developed AD compared with zero out of 30 MCI subjects with high olfactory

performance. Wilson et al also found that below-average (score of 8 out of 12, 25th percentile)

olfactory performance in older individuals without cognitive impairments predicted subsequent

development of MCI with risk increased by 50% compared with subjects with above-average

(score of 11 out of 12, 75th percentile) olfactory performance [44]. These studies show the

promising potential of olfactory deficits as diagnostic markers and as identifiers of disease

progression; however, it should be noted that while these studies lay the foundation, there is still a

lack of understanding in who will develop MCI and who will develop AD. Devanand et al

showed that 40% of the MCI subjects with olfactory impairment developed AD at follow-up (on

average 20 months after initial visit); however this study is limited in that many patients with

MCI will develop AD after longer intervals and the MCI subjects in the study ranged from 6

months to 10 years of cognitive symptoms [43]. The number of years of cognitive deficits was

not controlled and it still remains elusive as to which MCI patients will develop AD and at what

time interval. Similarly with Wilson et al’s study, 30% of the subjects (80 years) who tested as

cognitively normal at baseline developed MCI within 5 years of evaluation. Risk increased by

50% for those with olfactory performance below-average but the subjects scoring between the

10

25th and 75

th percentile were not included [44]. In general this age group has a higher overall risk

of developing olfactory and memory loss [1]. The limitation of this study is that it was

administered to a population that has a 33% prevalence of AD. A younger cohort will not

progress through symptoms as quickly and this study repeated in a younger cohort would provide

greater information for conversion from normal functioning to MCI.

Olfactory dysfunction is also found in individuals at risk for developing Alzheimer’s but

who do not present with cognitive symptoms, specifically Apolipoprotein E epsilon 4 (ApoE e4)

carriers. ApoE e4 is the largest known genetic risk factor for late-onset AD. Cognitively normal

subjects with the presence of the ApoE e4 gene had significantly higher olfactory dysfunction

than subjects without the presence of ApoE e4 (P = 0.006) [45]. MCI subjects with the ApoE e4

allele were not able to detect as many odors as MCI subjects without ApoE e4, suggesting that

olfactory impairment in MCI subjects may be a marker for AD and ApoE e4 may be involved in

olfactory identification dysfunction [46]. These behavioral studies in AD, MCI, normal controls,

and several other groups provide a cogent argument for the predictive power of olfactory

symptoms in the development of cognitive impairment symptoms. The results from these

investigations suggest an early change in the brain is the cause of the olfactory symptoms

associated with Alzheimer’s.

The above studies provide a cogent argument for the involvement of olfactory deficits in

AD and in MCI patients and for the ability of olfactory tests to provide diagnostic ability. We also

know symptomatic changes occur post dysfunctions in the brain; therefore, in vivo brain studies

have a great potential in identifying degeneration or disturbances in the olfactory regions prior to

symptom formation.

11

1.5 Pathological Changes in Olfactory Areas

Neuropathological postmortem studies provide evidence for the behavioral olfactory

deficits displayed by AD patients [6-8]. The characteristic markers of AD including amyloid beta

plaques, neurofibrillary tangles (NFTs), and atrophy have been observed in olfactory related

structures such as the olfactory bulb and tract, nasal epithelium, anterior olfactory nucleus,

olfactory cortex, entorhinal cortex, amygdala and periamygdaloid cortex, hippocampus, and

ventral striatum [3, 6-11]. Lesions have typically been found in these olfactory-related regions,

while visual, auditory, motor, and other sensory areas remain fairly intact [6, 47-48].

1.5.1 Neurofibrillary Tangles

In the work done by Braak and Braak, six stages of the disease progression can be

identified with respect to the location of the neurons bearing NFTs and the severity of clinical

manifestations (transentorhinal stages I-II: clinically silent cases; limbic stages III-IV: incipient

Alzheimer's disease; neocortical stages V-VI: fully developed Alzheimer's disease) [49-50].

Figure 1-1 shows the development of NFT distribution at different stages based on 2,661 non-

selected autopsy cases [5]. In 110 autopsy studies, Christen-Zaech et al found AD-type

degenerative changes in the olfactory bulb, tract and anterior olfactory nucleus in a high

percentage of the AD cases [14]. These changes were not only found in the severe cases but also

in the early and moderate cases (clinical staging) suggesting early involvement of the olfactory

structures [15-16].

12

1.5.2 Amyloid Beta Plaques

Positron emission tomography (PET), allowing for in vivo studies of plaque distribution,

has demonstrated that plaques are distributed in the caudate, amygdala, hippocampus, insula, and

retrospenial cingulate [12]. Rowe et al showed that AD subjects as well as amnestic MCI subjects

have considerable plaque burden in important regions of the olfactory network, including the

gyrus rectus, lateral temporal lobe, thalamus, putamen, cingulate, and orbitofrontal cortex [13].

PET studies offer in vivo evidence; however, the procedure involving injection of radioactive

substance is very invasive. Another study also reported increased plaques in the caudate,

putamen, and anterior/posterior cingulate in early AD subjects (clinical staging- symptomatic)

compared with normal controls [51].

1.5.3 Atrophy

Not only are NFTs and plaques observed in olfactory-related regions, but these areas also

show atrophy. Cell death was reported in the olfactory epithelium, olfactory bulb and anterior

olfactory nucleus [11, 52]. Axonal loss was also observed in the peripheral and central regions of

the olfactory tract of AD patients at autopsy [53].

This pattern of pathological findings in AD and MCI patients strongly indicate that

olfactory structures and pathways are severely affected in AD and MCI patients and are possibly

a substrate for initial involvement of neuropathological processes, deserving more extensive and

focused scientific investigation.

13

1.6 Neuroimaging in Alzheimer’s disease

Neuroimaging allows for in vivo measurements of the changes occurring in the brains of

AD and MCI patients and the changes that occur during the progression of AD and MCI from the

earliest stages. Clinically, neuroimaging, specifically MRI is used to rule out other causes of

cognitive dysfunction. Neuroimaging studies including structural MRI, functional MRI (fMRI),

and PET techniques have; however, provided an immense amount of information supporting the

role of olfaction in AD and MCI and have shown the ability to be used as more than in just

differential diagnosis. These methodologies have all demonstrated that olfactory changes occur at

the earliest stages. Of these many techniques, volumetric measurements have been most widely

used to examine the olfactory processing areas.

In this dissertation MRI, specifically structural and functional measurements were

utilized. MRI is a medical technique that’s allows for the investigation of the anatomy and

function of the body by using strong magnetic fields and radiowaves to form images. MRI

involves the imaging of protons, which are abundant in tissue. When placed in the magnetic field,

the protons align either with or against the direction of the field. When radiofrequency energy at

the appropriate frequency is applied to the protons, the ones aligned with the field absorb the

energy and reverse directions [54]. The protons will then release the energy and ―relax‖ back to

their original alignment at a rate determined by the T1 and T2 relaxation times which depend on

the physical and chemical characteristics of the tissue. The released energy is used to map the

spatially localized signal intensities which are represented on the image as points of relative

darkness or brightness. The signal intensities depend on several factors including the strength of

the magnetic field, pulse sequence, and tissue characteristics. Anatomical images can be T1-

weighted and T2-weighted [54]. The weighting depends on the pulse sequence, repetition time

(interval between repetitions of the pulse sequence), and echo time (interval between

14

radiofrequency excitation and the measurement of the resonance signal) used [88]. Tissues with

large amounts of water appear dark in T1-weighted images and bright in T2-weighted. Structural

MRI is used frequently for patients diagnosed with MCI and AD, mainly for the purpose of

excluding other neurological diseases [55].

fMRI is a well-established method for the delineation of the different regions of the brain

by studying the level of activation change in response to specific experimental conditions. fMRI

mapping is the production of activation maps that show average level of engagement of different

regions of the brain during a task or as a response to a stimulus [56]. Comparing these maps

between conditions or between groups allows evaluation of the different responses. fMRI uses

echo planar sequences that are sensitive to changes in blood oxygen level dependent (BOLD)

signal. This signal indirectly reflects neuronal activity by corresponding to the concentration of

deoxyhemoglobin. The magnetic resonance signal is derived from exiting hydrogen nuclei with a

radiofrequency pulse and detecting the radio waves that are emitted as the hydrogen nuclei return

to a lower-energy state. Deoxyhemoglobin and oxyhemoglobin have different magnetic

properties where deoxyhemoglobin is paramagnetic and it makes the local magnetic field over a

microscopic domain inhomogenous [56]. In order to estimate the BOLD signal during a task

paradigm, Statistical Parametric Mapping software (SPM) utilizes the general linear model

employing a hypothesized neural model convolved with a canonical hemodynamic response

function (HRF), considered to be the responses of the system to a brief period of neural

stimulation [57-58]. When we specify the onset vectors and durations, SPM will convolve them

with the canonical HRF.

fMRI data can also be analyzed to reveal how neural systems interact with one another

when performing specific tasks or when responding to a stimulus. The relationship of these

different regions and neural systems is generally described as ―functional connectivity‖. As used

15

in this dissertation, this term will mean the temporal correlations of multiple spatially-distinct

brain regions that are engaged simultaneously during a task. Functional connectivity is the

observed correlations and does not comment on the mediation or direction of the correlations. In

this dissertation the task is an olfactory fMRI paradigm which will be described in depth in

chapter 2.

MRI, both structural and functional, has not been officially used for diagnosis of AD or

MCI. Structural MRI as stated above has been utilized to rule out other neurological diseases that

may cause the symptoms. In general both have been used for research purposes and both have

become extremely popular techniques for the study of MCI and AD as well as other diseases.

1.6.1 Volumetric Studies

Structural MRI studies provide a measure of the atrophy in the brain that result from

dendritic dearborization, loss of synapses, neuronal cell loss and degeneration, and axonal loss.

These structural changes can be measured at the level of whole brain, gray matter, white matter,

or region of interest. An overall brain study of white matter reported less white matter in the

corpus callosum, right superior parietal lobe, cingulum, frontal, temporal, and occipital lobes of

AD patients compared with normal controls [59]. A more recent study reported that the white

matter atrophy in AD patients centers on the lateral temporal and parietal regions including the

cingulum and posterior corpus callosum [60]. Gray matter atrophy is also found in the frontal

cortex and the cerebellum of AD patients [61]. With overall loss of white and gray matter in the

brain there is also an enlargement of the ventricles in AD patients [62].

Several other areas of the brain have been studied that are related to memory function.

The entorhinal cortex and the hippocampus are atrophied in both AD and MCI subjects [63-66].

16

Studies have also demonstrated that degeneration of the hippocampus, white matter, and medial

temporal lobe correlate with declining cognitive performance and memory function [61, 67].

Volumetric changes are not only present in AD and MCI subjects, but they also correlate to

olfactory dysfunction in these patients. Hippocampal and parahippocampal gyrus volumes

positively correlated with odor identification test performance (n = 571, r = 0.16, P < 0.001),

suggesting a relationship between regional brain atrophy and olfactory performance [68].

While most of the MRI literature focuses on the hippocampus, few studies have revealed

atrophy in olfactory related structures. The amygdala was found to be significantly smaller in AD

patients compared with controls [69], and in MCI, less gray matter was found in the right

amygdala [70]. Reduced gray matter density in the olfactory bulb and tract was also seen in AD

and MCI subjects [61, 71]. A more recent study using a surface-based anatomical mesh modeling

technique and region of interest (ROI) analysis reported up to a 12-15% loss in the left and right

olfactory/orbitofrontal cortex of MCI and AD subjects [72]. While MRI studies demonstrate

atrophy in the olfactory bulb and tract, it should be noted that these measurements are subject to

artifact given the difficulty of accurately imaging the ventral portion of the brain due to its

proximity to the sinuses. To our knowledge volumetric studies of the primary olfactory cortex

have not been done. Since the primary olfactory cortex is an early site of pathology formation and

it can be more accurately imaged, it is integral to study this region in AD and MCI patients.

1.6.2 Functional MRI Studies

Functional MRI studies are on the forefront of improving identification of MCI and AD

because functional brain activation changes occur prior to volumetric and behavioral changes.

The medial temporal lobe has been studied extensively using fMRI due to its involvement in

episodic and visual memory, as well as being an early site of AD pathology [12-14, 49-50, 60,

17

62-67]. The earliest fMRI studies investigated brain activity during cognitive tasks. These studies

have led to widely ranging results from decreased to increased activation in the temporal and

frontal cortex in early AD. The increased activation is suggested to be a compensatory

mechanism [73]. The deficits in cognition and memory can be masked by the compensatory

mechanism in standard neuropsychological tests [74-76].

Functional deficits of the central olfactory structures have been detected using effective

olfactory fMRI paradigms in AD subjects [77-78]. Wang et al reported a decrease in activation in

the primary olfactory cortex of AD subjects compared with normal controls during presentation

of olfactory stimuli [77]. Li et al reported disruption of odor quality coding in the piriform cortex

of AD subjects [78]. Olfactory fMRI studies have focused on AD; however, very few functional

studies have focused on MCI patients. In a PET study, Cross and colleagues reported a positive

correlation between white matter integrity in the olfactory tract and metabolic activity in the

olfactory processing structures in MCI patients [79]. Finally in an electroencephalogram study,

Morgan and Murphy used olfactory event related potentials to demonstrate functional decline in

individuals at risk for Alzheimer’s disease at much earlier stages, while Peters reported AD and

MCI patients had no olfactory event-related potential compared with normal controls [80-81].

These studies are limited, however, in their spatial accuracy. Electroencephalogram measures

signals on the surface and has lower spatial resolution. These studies suggest olfaction could

significantly aid in pre-clinical AD detection but cannot provide information on location of

disruption.

More recently with the advancements in neuroimaging, functional connectivity has

emerged as a powerful method to study brain network changes in various disease states and task

conditions. Defined as the temporal correlation of blood oxygen level dependent fluctuations in

anatomically distinct brain regions [82], functional connectivity allows for inference of brain

18

networks and their temporal dynamics during a range of mental states. Resting-state fMRI studies

have shown a breakdown of the default mode network in AD and MCI patients [83-85]. The

default mode network is anatomically defined as including the posterior cingulate cortex, the

tempero-parietal junction, the precuneus, the medial prefrontal cortex, and part of the medial

temporal cortex. It is characterized by higher activity during periods of rest and is conversely

suppressed while engaged in task performance [82, 86-87]. Resting-state fMRI studies have

shown decreased hippocampal functional connectivity in AD subjects [88], further supporting the

disruption of brain networks in AD. Currently resting-state functional connectivity studies are not

specific to AD and MCI. Many other diseases show disconnection of the default mode network

such as schizophrenia, anxiety, autism spectrum disorders, and depression [89]. More information

may be provided regarding early changes in pre-clinical AD and MCI patients by studying more

specific networks such as the olfactory network which to our knowledge has not been

investigated.

1.7 Central versus Peripheral Olfactory Dysfunction in Alzheimer’s Disease

Behavioral, pathologic, volumetric, and functional studies have provided strong support

for the involvement of the olfactory system in Alzheimer’s and MCI patients. However it remains

unclear whether the cause of olfactory dysfunction in AD and MCI is related to the peripheral or

central olfactory system. While only a few studies have focused on this question, some behavioral

and pathological studies have leaned toward the cause as a central olfactory problem in AD.

Performance on odorant identification tests is worse in AD and MCI subjects compared with

performance on threshold detection tests, indicating central olfactory dysfunction [33, 37]. A

separate autopsy study also reported less severe pathology in the peripheral olfactory areas

compared with the central olfactory areas of AD patients [90]. Another study also suggested

19

olfactory impairments associated with AD are likely due to damage in the central olfactory

pathways based on neuropathological changes in the olfactory epithelium and central olfactory

pathways [16]. Nonetheless post-mortem studies are inconclusive in their findings because they

cannot definitively rule out peripheral olfactory dysfunction as dominant in AD and MCI. More

recently, functional deficits of the central olfactory structures in AD have been detected using

olfactory fMRI [77-78]. However, these studies require peripheral afferent information and

therefore still cannot definitively rule out peripheral olfactory dominance. Thus the central- or

peripheral-dominant olfactory impairment in AD and MCI remains a conundrum.

1.8 Rationale

Olfactory impairment is present at the earliest stages of Alzheimer’s disease and has even

been demonstrated in patients with MCI. Specifically, AD and MCI patients present with deficits

in odor detection, odor identification, and odor memory and these symptoms often precede

cognitive impairments [17-18, 21, 30, 33, 35, 39-45]. These results are based on research studies

rather than clinical observations. Currently olfactory testing is not done in the clinic unless

individuals specifically notice olfactory deficits and complain of these symptoms. The general

population is not aware of changing olfactory sense and therefore these changes go unnoticed.

Pathologic changes found in the brain support these behavioral symptoms [6-11, 47-50]. Braak

and Braak confirmed that early AD pathology overlaps with olfactory processing areas. Autopsy

and PET studies have reported amyloid plaques and neurofibrillary tangles in the olfactory bulb,

olfactory tract, nasal epithelium, and piriform/ primary olfactory cortex at the early stage of

Alzheimer’s disease [3, 5, 12-13, 51]. Neuroimaging studies also report atrophy of olfactory

related structures such as the amygdala, and olfactory bulb and tract, hippocampus, and

orbitofrontal cortex [69-72]. It should be stated; neuroimaging of the olfactory bulb and tract is

20

unreliable due to artifact from the sinuses. Behavioral symptoms and macrostructural changes are

present in the earliest stages of AD and MCI. Prior to these visible changes, however,

microstructural changes occur which have been far less investigated. Of the few studies, fMRI

and electrophysiology techniques have demonstrated decreased activation and event-related

potentials respectively in AD subjects during olfactory stimulation [77-78, 80-81]. More

investigation of pre-clinical AD and those at risk for development of AD is needed. Olfaction is

affected at the behavioral, marcostructural, and microstructural levels at the earliest stages of AD,

providing a unique system to study Alzheimer’s disease.

While previous studies have provided substantial knowledge of olfactory deficits, many

points still remain unknown. Can fMRI show earlier and more drastic changes in MCI patients

than behavioral and volumetric measurements? Are olfactory deficits in AD and MCI patients

due to dysfunction of the central olfactory system? Is the dysfunction of the central olfactory

system able to be monitored to follow the progression of MCI to AD? And lastly, do olfactory

fMRI and the study of the central system have the potential to predict which MCI patients will

develop AD?

In this dissertation, I further investigate the role of the olfactory system and utilize

structural and olfactory functional MRI to find a potential early diagnostic marker for AD and to

answer many of the questions stated above. As far as we know, no study has investigated

volumetric changes of the primary olfactory cortex or examined the activation signal change in

the primary olfactory cortex during an olfactory paradigm in MCI and AD subjects. In the second

chapter, therefore, I investigate volume and activation changes of the primary olfactory cortex

and demonstrate the use of olfactory fMRI in conjunction with olfactory testing in increasing the

specificity and sensitivity of disease diagnosis. In the following chapter, we bring forth further

findings that support olfactory dysfunction is due to a central olfactory system disruption which

21

has not been demonstrated using an fMRI study. Finally in the last study we utilize a relatively

novel technique, functional connectivity, to report disruption of the olfactory network in AD and

MCI subjects. To our knowledge, functional connectivity of the central olfactory system has not

been examined in MCI and AD subjects. This dissertation aims to show for the first time that 1)

olfactory fMRI picks up earlier and more drastic changes in MCI subjects than behavioral

olfactory and cognitive testing and volumetric decreases; 2) central olfactory dysfunction is the

reason for olfactory deficits in AD and MCI subjects utilizing our novel olfactory paradigm, and

3) disruption of the olfactory network in AD and MCI subjects and the preservation of this

network in MCI subjects compared with AD subjects may explain higher performance on

behavioral tests. Overall this dissertation will demonstrate the great potential of olfactory fMRI in

allowing for early diagnosis of pre-clinical AD and MCI patients and in monitoring of disease

progression. Ultimately, earlier identification of AD can allow for future treatment with targeted

drug therapy that can potentially modify the disease before it has manifested clinically.

22

1.9 References

[1] Alzheimer's Association, Thies W, Bleiler L. 2011 Alzheimer's disease facts and figures.

Alzheimers Dement. 2011 Mar;7(2):208-44. doi: 10.1016/j.jalz.2011.02.004.

[2] Tejada-Vera B. Mortality from Alzheimer's disease in the United States: data for 2000 and

2010. NCHS Data Brief. 2013 Mar;(116):1-8.

[3] Braak H, Braak E. Neuropathological stageing of Alzheimer-related changes. Acta

Neuropathol. 1991;82(4):239-59.

[4] Grober E, Dickson D, Sliwinski MJ, Buschke H, Katz M, Crystal H, Lipton RB. Memory

and mental status correlates of modified Braak staging. Neurobiol Aging. 1999 Nov-

Dec;20(6):573-9.

[5] Braak H, Braak E. Frequency of stages of Alzheimer-related lesions in different age

categories. Neurobiol Aging. 1997 Jul-Aug;18(4):351-7.

[6] Mann DMA, Esiri MM. The site of the earliest lesions of Alzheimer’s disease. N Engl J

Med 1988; 318:789-90.

[7] Ohm TG, Braak H. Olfactory bulb changes in Alzheimer'sdisease. Acta Neuropathol 1987;

73: 365-9.

[8] Price JL, Davis PB, Morris JC, White DL. The distribution of tangles, plaques and related

immunohistochemical markers in healthy aging and Alzheimer's disease. Neurobiol. Aging

1991; 12: 295-312.

[9] Attems J, Jellinger KA, Olfactory tau pathology in Alzheimer’s disease and mild cognitive

impairment. Clin. Neuropathol 2006; 25: 265-71.

[10] Esiri MM, Wilcock PK. The olfactory bulb in Alzheimer’s disease. J Neurol Neurosurg

Psychiat 1984; 47:56-60.

23

[11] Talamo BR, Rudel R, Kosik KS, Lee VM, Neff S, Adelman L, Kauer JS. Pathological

changes in olfactory neurons in patients with Alzheimer's disease. Nature. 1989 Feb

23;337(6209):736-9

[12] Frisoni GB, Lorenzi M, Caroli A, Kemppainen N, Någren K, Rinne JO. In vivo mapping of

amyloid toxicity in Alzheimer disease. Neurology. 2009 Apr 28;72(17):1504-11. doi:

10.1212/WNL.0b013e3181a2e896.

[13] Rowe CC, Ng S, Ackermann U, Gong SJ, Pike K, Savage G, Cowie TF, Dickinson KL,

Maruff P, Darby D, Smith C, Woodward M, Merory J, Tochon-Danguy H, O'Keefe G,

Klunk WE, Mathis CA, Price JC, Masters CL, Villemagne VL. Imaging beta-amyloid

burden in aging and dementia. Neurology. 2007 May 15;68(20):1718-25.

[14] Christen-Zaech S, Kraftsik R, Pillevuit O, Kiraly M, Martins R, Khalili K, Miklossy J. Early

olfactory involvement in Alzheimer's disease. Can J Neurol Sci. 2003 Feb;30(1):20-5.


immunohistochemical markers in healthy aging and Alzheimer's disease. Neurobiol Aging.

1991 Jul-Aug;12(4):305-312.

[16] Arnold SE, Hyman BT, Flory J, Damasio AR, Van Hoesen GW. The topographical and

neuroanatomical distribution of neurofibrillary tangles and neuritic plaques in the cerebral

cortex of patients with Alzheimer's disease. Cereb Cortex. 1991 Jan-Feb;1(1):103-16.

[17] Barresi M, Ciurleo R, Giacoppo S, Foti Cuzzola V, Celi D, Bramanti P, Marino S.

Evaluation of olfactory dysfunction in neurodegenerative diseases. J Neurol Sci. 2012 Dec

15;323(1-2):16-24. doi: 10.1016/j.jns.2012.08.028.

[18] Rahayel S, Frasnelli J, Joubert S. The effect of Alzheimer's disease and Parkinson's disease

on olfaction: a meta-analysis. Behav Brain Res. 2012 May 16;231(1):60-74. doi:

10.1016/j.bbr.2012.02.047.

24

[19] Murphy C, Gilmore MM, Seery CS, Salmon DP, Lasker BR. Olfactory thresholds are

associated with degree of dementia in Alzheimer's disease. Neurobiol Aging. 1990 Jul-

Aug;11(4):465-9.

[20] Ritchie K. Mild cognitive impairment: an epidemiological perspective. Dialogues Clin

Neurosci 2004; 6: 401–08.

[21] Petersen RC, Smith GE, Waring SC, Ivnik RJ, Kokmen E, Tangelos EG. Aging, memory,

and mild cognitive impairment. Int Psychogeriatr 1997; 9:65-9.

[22] Paxianos G, Mai JK. The human Nervous System. Elsevier accademic press, 3rd Edition.

2012 Ch. 34

[23] Parent A, Carpenter MB. Carpenter's human neuroanatomy. Baltimore: Williams & Wilkins,

9th Edition. 1996 Ch. 18

[24] Sela L, Sobel N. Human olfaction: a constant state of change-blindness. Exp Brain Res.

2010 Aug;205(1):13-29. doi: 10.1007/s00221-010-2348-6.

[25] Tham WW, Stevenson RJ, Miller LA. The functional role of the medio dorsal thalamic

nucleus in olfaction. Brain Res Rev. 2009 Dec 11;62(1):109-26. doi:

10.1016/j.brainresrev.2009.09.007.

[26] Doty RL, Shaman P, Dann M. Development of the University of Pennsylvania Smell

Identification Test: a standardized microencapsulated test of olfactory function. Physiol

Behav. 1984 Mar;32(3):489-502

[27] Hummel T, Sekinger B, Wolf SR, Pauli E, Kobal G. 'Sniffin' sticks': olfactory performance

assessed by the combined testing of odor identification, odor discrimination and olfactory

threshold. Chem Senses. 1997 Feb;22(1):39-52.

[28] Cain WS, Gent JF, Goodspeed RB, Leonard G. Evaluation of olfactory dysfunction in the

Connecticut Chemosensory Clinical Research Center. Laryngoscope. 1988 Jan;98(1):83-8.

25

[29] Waldton S. Clinical observations of impaired cranial nerve function in senile dementia. Acta

Psychiatr Scand. 1974;50(5):539-47.

[30] Ferreyra-Moyano H, Barragan E. The olfactory system and Alzheimer's disease. Int J

Neurosci. 1989 Dec;49(3-4):157-97.

[31] Doty RL, Stern MB, Pfeiffer C, Gollomp SM, Hurtig HI. Bilateral olfactory dysfunction in

early stage treated and untreated idiopathic Parkinson's disease. J Neurol Neurosurg

Psychiatry. 1992 Feb;55(2):138-42.

[32] Serby M. Olfaction and Alzheimer's disease. Prog Neuropsychopharmacol Biol Psychiatry.

1986;10(3-5):579-86.

[33] Serby M, Larson P, Kalkstein D. The nature and course of olfactory deficits in Alzheimer's

disease. Am J Psychiatry. 1991 Mar;148(3):357-60.

[34] Warner MD, Peabody CA, Flattery JJ, Tinklenberg JR. Olfactory deficits and Alzheimer's

disease. Biol Psychiatry. 1986 Jan;21(1):116-8.

[35] Moberg PJ, Pearlson GD, Speedie LJ, Lipsey JR, Strauss ME, Folstein SE. Olfactory

recognition: differential impairments in early and late Huntington's and Alzheimer's

diseases. J Clin Exp Neuropsychol. 1987 Dec;9(6):650-64.

[36] Rezek DL. Olfactory deficits as a neurologic sign in dementia of the Alzheimer type. Arch

Neurol. 1987 Oct;44(10):1030-2.

[37] Koss E, Weiffenbach JM, Haxby JV, Friedland RP. Olfactory detection and identification

performance are dissociated in early Alzheimer's disease. Neurology. 1988 Aug;38(8):1228-

32.

[38] Kesslak JP, Cotman CW, Chui HC, Van Den Noort S, Fang H, Pfeffer R, et al. Olfactory

tests as possible probes for detecting and monitoring Alzheimer’s disease. Neurobiol Aging

1988; 9:399-403.

26

[39] Morgan CD, Nordin S, Murphy C. Odor identification as an early marker for Alzheimer’s

disease: impact of lexical functioning and detection sensitivity. J Clin Exper Neuropsychol

1995; 15:793-803.

[40] Nordin S, Murphy C: Impaired sensory and cognitive olfactory function in questionable

Alzheimer’s disease. Neuropsychology 1996; 10:113-9.

[41] Djordjevic J, Jones-Gotman M, De Sousa K, Chertkow H. Olfaction in patients with mild

cognitive impairment and Alzheimer's disease. Neurobiol Aging. 2008 May;29(5):693-706.

[42] Lehrner J, Pusswald G, Gleiss A, Auff E, Dal-Bianco P. Odor identification and self-

reported olfactory functioning in patients with subtypes of mild cognitive impairment. Clin

Neuropsychol. 2009 Jul;23(5):818-30. doi: 10.1080/13854040802585030. Epub 2009 Feb

11.

[43] Devanand DP, Michaels-Marston KS, Liu X, Pelton GH, Padilla M, Marder K, et al.

Olfactory deficits in patients with mild cognitive impairment predict Alzheimer's disease at

follow-up. Am J Psychiatry 2000; 157:1399-405.

[44] Wilson RS, Schneider JA, Arnold SE, Tang Y, Boyle PA, Bennett DA. Olfactory

identification and incidence of mild cognitive impairment in older age. Arch Gen Psychiatry.

2007 Jul;64(7):802-8.

[45] Murphy C, Bacon AW, Bondi MW, Salmon DP. Apolipoprotein E status is associated with

odor identification deficits in nondemented older persons. Ann N Y Acad Sci. 1998 Nov

30;855:744-50.

[46] Wang QS, Tian L, Huang YL, Qin S, He LQ, Zhou JN. Olfactory identification and

apolipoprotein E epsilon 4 allele in mild cognitive impairment. Brain Res. 2002 Sep

27;951(1):77-81.

27

[47] Pearson RC, Esiri MM, Hiorns RW, Wilcock GK, Powell TP. Anatomical correlates of the

distribution of the pathological changes in the neocortex in Alzheimer disease. Proc Natl

Acad Sci U S A. 1985 Jul;82(13):4531-4.

[48] Harrison PJ. Pathogenesis of Alzheimer's disease--beyond the cholinergic hypothesis:

discussion paper. J R Soc Med. 1986 Jun;79(6):347-52.

[49] Braak H, Braak E. Morphological criteria for the recognition of Alzheimer's disease and the

distribution pattern of cortical changes related to this disorder. Neurobiol Aging. 1994 May-

Jun;15(3):355-6; discussion 379-80.

[50] Hyman BT. The neuropathological diagnosis of Alzheimer's disease: clinical-pathological

studies. Neurobiol Aging. 1997 Jul-Aug;18(4 Suppl):S27-32.

[51] Edison P, Archer HA, Hinz R, Hammers A, Pavese N, Tai YF, Hotton G, Cutler D, Fox N,

Kennedy A, Rossor M, Brooks DJ. Amyloid, hypometabolism, and cognition in Alzheimer

disease: an [11C]PIB and [18F]FDG PET study. Neurology. 2007 Feb 13;68(7):501-8.

[52] Arnold SE, Smutzer GS, Trojanowski JQ, Moberg PJ. Cellular and molecular

neuropathology of the olfactory epithelium and central olfactory pathways in Alzheimer's

disease and schizophrenia. Ann N Y Acad Sci. 1998 Nov 30;855:762-75.

[53] Armstrong RA, Syed AB, Smith CU. Density and cross-sectional areas of axons in the

olfactory tract in control subjects and Alzheimer's disease: an image analysis study. Neurol

Sci. 2008 Feb;29(1):23-7. doi: 10.1007/s10072-008-0854-0. Epub 2008 Apr 1.

[54] Bottomley PA, Hardy CJ, Argersinger RE, Allen-Moore G. A review of 1H nuclear

magnetic resonance relaxation in pathology: are T1 and T2 diagnostic? Med Phys

1987;14:1-37

[55] Edelman RR, Warach S. Magnetic resonance imaging (2) N Engl J Med. 1993 Mar

18;328(11):785-91.

28

[56] Rogers BP, Morgan VL, Newton AT, Gore JC. Assessing functional connectivity in the

human brain by fMRI. Magn Reson Imaging. 2007 Dec;25(10):1347-57. Epub 2007 May

11.

[57] Friston KJ, Fletcher P, Josephs O, Holmes A, Rugg MD, Turner R. Event-related fMRI:

characterizing differential responses. Neuroimage. 1998 Jan;7(1):30-40.

[58] Calhoun VD, Stevens MC, Pearlson GD, Kiehl KA. fMRI analysis with the general linear

model: removal of latency-induced amplitude bias by incorporation of hemodynamic

derivative terms. Neuroimage. 2004 May;22(1):252-7.

[59] Canu E, Agosta F, Spinelli EG, Magnani G, Marcone A, Scola E, Falautano M, Comi G,

Falini A, Filippi M. White matter microstructural damage in Alzheimer's disease at different

ages of onset. Neurobiol Aging. 2013 Oct;34(10):2331-40. doi:

10.1016/j.neurobiolaging.2013.03.026. Epub 2013 Apr 24.

[60] Migliaccio R, Agosta F, Possin KL, Rabinovici GD, Miller BL, Gorno-Tempini ML. White

matter atrophy in Alzheimer's disease variants. Alzheimers Dement. 2012 Oct;8(5

Suppl):S78-87.e1-2. doi: 10.1016/j.jalz.2012.04.010.

[61] Ashford JW, Salehi A, Furst A, Bayley P, Frisoni GB, Jack CR Jr, Sabri O, Adamson MM,

Coburn KL, Olichney J, Schuff N, Spielman D, Edland SD, Black S, Rosen A, Kennedy D,

Weiner M, Perry G. Imaging the Alzheimer brain. J Alzheimers Dis. 2011;26 Suppl 3:1-27.

doi: 10.3233/JAD-2011-0073.

[62] Ott BR, Cohen RA, Gongvatana A, Okonkwo OC, Johanson CE, Stopa EG, Donahue JE,

Silverberg GD, Alzheimer's Disease Neuroimaging Initiative. Brain ventricular volume and

cerebrospinal fluid biomarkers of Alzheimer's disease. J Alzheimers Dis. 2010;20(2):647-57.

doi: 10.3233/JAD-2010-1406.

29

[63] Kesslak JP, Nalcioglu O, Cotman CW. Quantification of magnetic resonance scans for

hippocampal and parahippocampal atrophy in Alzheimer’s disease. Neurology 1991; 41:51-

4.

[64] Jack CR, Petersen RC, O’Brien PC, Tangalos EG. MR-based hippocampal volumetry in the

diagnosis of Alzheimer’s disease. Neurology 1992; 42: 183-8.

[65] Convit A, de Leon MJ, Golomb J, George AE, Tarshish CY, Bobinski M, et al.

Hippocampal atrophy in early Alzheimer’s disease: anatomic specificity and validation.

Psychiatr Q 1993; 64:371-87.

[66] Foundas AL, Leonard CM, Mahoney M, Agee OF, Heilman KM. Atrophy of the

hippocampus, parietal cortex, and insula in Alzheimer’s disease: a volumetric magnetic

resonance imaging study. Neurol Neuropsychol Behav Neurol 1997; 10:81-9.

[67] Carmichael O, Schwarz C, Drucker D, Fletcher E, Harvey D, Beckett L, Jack CR Jr, Weiner

M, DeCarli C; Alzheimer's Disease Neuroimaging Initiative. Longitudinal changes in white

matter disease and cognition in the first year of the Alzheimer disease neuroimaging

initiative. Arch Neurol. 2010 Nov;67(11):1370-8. doi: 10.1001/archneurol.2010.284.

[68] Devanand DP, Tabert MH, Cuasay K, Manly JJ, Schupf N, Brickman AM, Andrews H,

Brown TR, DeCarli C, Mayeux R. Olfactory identification deficits and MCI in a multi-

ethnic elderly community sample. Neurobiol Aging. 2010 Sep;31(9):1593-600. doi:

10.1016/j.neurobiolaging.2008.09.008. Epub 2008 Oct 28.

[69] Cavedo E, Boccardi M, Ganzola R, Canu E, Beltramello A, Caltagirone C, Thompson PM,

Frisoni GB. Local amygdala structural differences with 3T MRI in patients with Alzheimer

disease. Neurology. 2011 Feb 22;76(8):727-33. doi: 10.1212/WNL.0b013e31820d62d9.

[70] Zheng D, Sun H, Dong X, Liu B, Xu Y, Chen S, Song L, Zhang H, Wang X. Executive

dysfunction and gray matter atrophy in amnestic mild cognitive impairment. Neurobiol

30

Aging. 2014 Mar;35(3):548-55. doi: 10.1016/j.neurobiolaging.2013.09.007. Epub 2013 Oct

9.

[71] Thomann PA, Dos Santos V, Seidl U, Toro P, Essig M, Schröder J. MRI-derived atrophy of

the olfactory bulb and tract in mild cognitive impairment and Alzheimer's disease. J

Alzheimers Dis. 2009; 17:213-21

[72] Prestia A, Baglieri A, Pievani M, Bonetti M, Rasser PE, Thompson PM, et al. The in vivo

topography of cortical changes in healthy aging and prodromal Alzheimer's disease. Suppl

Clin Neurophysiol. 2013; 62:67-80

[73] Johnson KA, Fox NC, Sperling RA, Klunk WE. Brain imaging in Alzheimer disease. Cold

Spring Harb Perspect Med. 2012 2(4):a006213. doi: 10.1101/cshperspect.a006213.

[74] Johnson SC, Saykin AJ, Baxter LC, Flashman LA, Santulli RB, McAllister TW, et al. The

relationship between fMRI activation and cerebral atrophy: comparison of normal aging and

Alzheimer disease. Neuroimage 2000; 11:179-87.

[75] Buchsbaum MS, Kesslak JP, Lynch G, Chui H. Temporal and hippocampal metabolic rate

during an olfactory memory task assessed by positron emission tomography in patients with

dementia of the Alzheimer type and controls: Preliminary studies. Arch Gen Psychiat 1991;

48:840-7.

[76] Raichle ME, Fiez JA, Videen TO, MacLeod AM, Pardo JV, Fox PT, et al. Practice-related

changes in human brain functional anatomy during nonmotor learning. Cereb Cortex 1994;

4:8-26.

[77] Wang J, Eslinger PJ, Doty RL, Zimmerman EK, Grunfeld R, Sun X, et al. Olfactory deficits

detected by fMRI in early Alzheimer’s disease. Brain Research 2010; 1357:184-94.

[78] Li W, Howard JD, Gottfried JA. Disruption of odour quality coding in piriform cortex

mediates olfactory deficits in Alzheimer's disease. Brain. 2010; 133:2714-26

31

[79] Cross DJ, Anzai Y, Petrie EC, Martin N, Richards TL, Maravilla KR, Peskind ER,

Minoshima S. Loss of olfactory tract integrity affects cortical metabolism in the brain and

olfactory regions in aging and mild cognitive impairment. J Nucl Med. 2013

Aug;54(8):1278-84. doi: 10.2967/jnumed.112.116558.

[80] Morgan CD, Murphy C. Individuals at risk for Alzheimer's disease show differential

patterns of ERP brain activation during odor identification. Behav Brain Funct. 2012 Jul

31;8:37. doi: 10.1186/1744-9081-8-37.

[81] Peters JM, Hummel T, Kratzsch T, Lötsch J, Skarke C, Frölich L. Olfactory function in mild

cognitive impairment and Alzheimer's disease: an investigation using psychophysical and

electrophysiological techniques. Am J Psychiatry. 2003 Nov; 160(11):1995-2002.

[82] Fox MD, Raichle ME. Spontaneous fluctuations in brain activity observedwith functional

magnetic resonance imaging. Nature reviews Neuroscience. 2007: 8(9):700–11.

[83] Greicius MD, Srivastava G, Reiss AL, Menon V. 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. 2004:

101(13):4637–42.

[84] Wang K1, Liang M, Wang L, Tian L, Zhang X, Li K, Jiang T. Altered functional

connectivity in early Alzheimer's disease: a resting-state fMRI study. Hum Brain Mapp.

2007: 28(10):967-78.

[85] Rombouts SARB, Barkhof F, Goekoop R, Stam CJ, Scheltens P. Altered resting state

networks in Mild Cognitive Impairment and mild Alzheimer’s Disease: an fMRI study.

Human Brain Mapping. 2005: 26:231-239.

[86] Raichle ME, Snyder AZ. A default mode of brain function: a brief history of anevolving

idea. NeuroImage. 2007: 37(4):1083–90.

32

[87] Buckner RL, Andrews-Hanna JR, Schacter DL. The brain’s default network: anatomy,

function, and relevance to disease. Annals of the New York Academy of Sciences. 2008:

1124:1–38.

[88] Allen, G., et al., Reduced hippocampal functional connectivity in Alzheimer disease. Arch

Neurol, 2007. 64(10): p. 1482-7.

[89] Broyd SJ, Demanuele C, Debener S, Helps SK, James CJ, Sonuga-Barke EJ. Default-mode

brain dysfunction in mental disorders: a systematic review. Neurosci Biobehav Rev. 2009

Mar;33(3):279-96. doi: 10.1016/j.neubiorev.2008.09.002. Epub 2008 Sep 9.

[90] Davies DC, Brooks JW, Lewis DA. Axonal loss from the olfactory tracts in Alzheimer's

disease. Neurobiol Aging 1993; 14:353-7.

33

Chapter 2

Functional and structural degeneration of the primary olfactory cortex in AD and MCI

2.1 Abstract

Background: Olfactory deficits are prevalent in patients with Alzheimer’s disease (AD) and mild

cognitive impairment (MCI) and often precede cognitive and memory problems. This coincides

with the early site of AD pathology in the central olfactory structures. Therefore, the primary

olfactory cortex (POC) was examined using noninvasive neuroimaging techniques.

Methods: Olfactory structural and functional magnetic resonance imaging (MRI/fMRI) and

olfactory and cognitive tests were administered to 27 cognitively normal controls, 21 MCI, and

15 AD subjects. Total structural volumes and fMRI activation volumes of the POC and

hippocampus were measured.

Results: Prominent atrophy in the POC and hippocampus was found in MCI and AD subjects and

correlated closely with the behavioral measures (P < 0.05). Activation in the POC and

hippocampus showed a more drastic decline in the MCI group than the behavioral (cognitive and

olfactory) tests or volumetric results. While behavioral and volumetric results declined

continuously from normal controls to MCI to AD, olfactory fMRI results in the POC and

hippocampus were similar in the MCI and the AD groups.

Conclusion: Olfactory fMRI detected earlier functional changes in the MCI group than behavioral

and volumetric measurement. Therefore, olfactory fMRI has potential to aid early diagnosis of

AD and MCI.

34

2.2 Introduction

Standardized behavioral tests have demonstrated that olfactory deficits begin in the early

stages of Alzheimer’s disease (AD) [1-3]. These include odor threshold detection, identification,

and memory deficits [4-9]. Longitudinal studies have indicated that disease progression is

significantly correlated with olfactory impairment [2, 3], even after aging effects have been taken

into account [10]. Olfactory dysfunction has also been found in individuals with mild cognitive

impairment (MCI) [11]. This is significant, because MCI individuals convert to AD at an annual

rate of 15% [12]. Therefore, olfactory deficits have the potential to be a robust biomarker of early

stage AD and preclinical AD [13].

Postmortem studies have provided a neuropathological basis for the observed olfactory

deficits in AD patients [13, 14]. Classic AD pathology (amyloid beta plaques and neurofibrillary

tangles) has been shown to be distributed preferentially in olfactory-related structures when

compared with visual, auditory, and somatosensory brain areas [14-19]. Olfactory structures

include the olfactory bulb and tract, anterior olfactory nucleus, piriform cortex, entorhinal cortex,

amygdala and periamygdaloid cortex. Furthermore, this AD-related pathological pattern has been

shown to be present in the earliest stages of disease [20].

Clinical diagnosis of AD requires an extensive evaluation: interview and medical history,

brain imaging, blood chemistries, clinical exam, and neuropsychological evaluation of multiple

cognitive and behavioral domains. Clinically, neuroimaging serves primarily to rule out other

diseases. However, neuroimaging studies have shown potential for more than just differential

diagnosis. Meta analyses of more than 50 voxel-based morphometry studies have confirmed

significant atrophy in the medial temporal lobe in MCI and AD [21-22]. This technique has

detected hippocampal atrophy in AD patients with a mean volume loss between 20% and 52%

35

[23-29]. In contrast, there are limited studies on atrophy of the central olfactory structures [30-

31], specifically, there are no studies investigating the atrophy of the primary olfactory cortex and

the relationship between the atrophy and olfactory function. Due to the course of the pathology

in AD, we hypothesize that it is likely a more pervasive atrophy exists in the central olfactory

structures in early and preclinical AD.

Establishing a specific relationship between the pathological changes in a given brain

structure in vivo and the corresponding functional decline is of great importance for developing

imaging biomarkers. Previously functional deficits of the central olfactory structures have been

detected using olfactory functional magnetic resonance imaging (fMRI) in AD patients [32-33].

As the volumetric studies, studies of activation signal change are also limited and have mostly

focused on AD patients rather than MCI patients. The olfactory deficit in early AD provides a

unique opportunity to investigate such specific brain structure-to-function relationships in vivo.

Thus, we hypothesize that 1) atrophy exists in the POC of MCI and AD, which is correlated to the

atrophy of the hippocampus; and 2) olfactory fMRI activation is correlated to the POC atrophy

and is more sensitive to earlier changes. Testing our hypotheses, we conducted concurrent

measurements of olfactory fMRI and volume of the POC and hippocampus and determined the

relationship between these two measurements. We correlated these results with the behavioral

measurements, and established a direct linkage between clinical presentations and

neurobiological measures of pathology (local atrophy) at the early sites (POC and hippocampus)

of AD degeneration. We further demonstrated that combined measurements of olfactory fMRI

and atrophy of the hippocampus can yield a more sensitive and specific marker for classifying

MCI and AD than volumetric measurements alone.

36

2.3 Methods

2.3.1 Study Cohort (Used in chapters 3 and 4)

Sixty-three subjects were enrolled in this study: 15 AD (Clinical Dementia Rating Scale

(CDR) of 0.05 or 1), 21 MCI (CDR of 0.5), and 27 normal controls (CN), (Table 2-1). No

significant age, gender, or education differences existed. Pennsylvania State University College

of Medicine Institutional Review Board approved the study and subjects provided written consent

prior to participation. Subjects were screened for other neurologic and psychiatric conditions;

including checking for complications specific to olfactory dysfunction (e.g., head trauma, viral

infection, allergies) and for contraindications to MRI (e.g., not-MRI-safe metal implants). AD

and MCI subjects were clinically diagnosed by a board certified neurologist in accordance with

NINCDS-ADRDA criteria [34] and Peterson criteria [35], respectively. 14 AD subjects and 12

MCI subjects were being treated with a cholinesterase inhibitor and/or memantine.

37

Table 2-1. Demographic and behavioral data of the study cohort

CN (n = 27) MCI (n = 21) AD (n = 15)

Male/Female

Age (year)

Educational level (year)

UPSIT

MMSE

CVLT-II

DRS-2

12/15

69.5 ± 10.4

16.0 ± 1.7

34.0 ± 4.2

28.5 ± 1.5

62.6 ± 13.1

13.3 ± 1.6

10/11

73.2 ± 9.0

14.6 ± 2.9

24.2 ± 8.6*

26.5 ± 1.9

47.3 ± 12.7*

9.6 ± 3.2*

5/10

71.9 ± 11.9

14.3 ± 3.0

15.5 ± 8.4*,†

18.9 ± 5.4*,†

21.3 ± 14.1*,†

3.9 ± 2.5*,†

Abbreviations: CN, cognitively normal controls; MCI, mild cognitive impaired; AD,

Alzheimer’s disease; UPSIT, University of Pennsylvania Smell Identification Test; MMSE, Mini-

Mental State Examination; CVLT-II, California Verbal Learning Test- Short Form Version 2;

DRS-2, Dementia Rating Scale 2.

Note: Mean ± standard deviation is reported.

* P <0.05, Analysis of variance (ANOVA) when compared with CN.

† P <0.05, ANOVA when compared with MCI.

38

2.3.2 Behavioral Tests (Used in chapters 3 and 4)

All participants were administered the University of Pennsylvania Smell Identification

Test (UPSIT, Sensonics, Inc., Haddon Heights, NJ, USA) to assess their smell identification

function, and clinical neurocognitive examinations, which included the Mini-Mental State

Examination (MMSE), the Mattis Dementia Rating Scale-2 (DRS-2), and the California Verbal

Learning Test-Second Edition Short Form (CVLT-II). The MMSE is a 30-point test of general

cognitive ability used commonly in clinical practice. The DRS-2 is a more detailed measure of

general cognitive ability with age-corrected scaled sub-scores in five areas: attention,

inhibition/perseveration, construction, conceptualization, and memory. The CVLT-II provides a

comprehensive assessment of verbal learning and memory. We also conducted a medical history

evaluation for all participants.

2.3.3 Olfactory Stimulation Paradigm (Used in chapters 3 and 4)

The odor stimulation paradigm was executed using a programmable olfactometer

(Emerging Tech Trans, LLC, Hershey, PA, USA) to deliver odorants to subject’s nostrils

accurately without any optical, acoustic, thermal, or tactile cues to the subject. The olfactometer

delivered 6 L/min of constant airflow at room temperature bilaterally to the subjects' nostrils. The

odor stimulation paradigm and MRI image acquisition were synchronized using optical triggers

from the MR scanner.

The stimulus was lavender oil (Givaudan Flavors Corporation, East Hanover, NJ, USA)

diluted in 1,2-propanediol (Sigma, St. Louis, MO, USA). Lavender is an effective, pleasant, and

familiar olfactory stimulant with minimal propensity to stimulate the trigeminal system [36]. Four

concentrations of lavender were used (Fig. 2-1) based upon a previous study on young controls

39

[37]. The odor was presented for 6 s separated by 30 s of odorless air. The presentation order was

from weakest to strongest concentration with three presentations of each concentration before

moving onto the subsequent higher concentration. This was done in order to offset the habituation

effect [37]. The olfactory fMRI paradigm also included a visual component and a motor response.

The visual component included the words ―Rest‖ and ―Smell?‖. When the word ―Smell?‖

appeared on the screen the subject was asked to respond ―yes‖ or ―no‖ depending on whether they

smelled the lavender odor or not using the button presses in each hand. When ―Rest‖ was

displayed on the screen, the subject was asked to just rest and continue paying attention to the

screen. The word ―Smell?‖ was always displayed for 6 s and was paired with either constant

odorless air or with lavender odor while the word ―Rest‖ was displayed for 12 s and paired with

only odorless air. Periods with ―Rest‖ and odorless air were used as the baseline condition. The

odorless air was kept constant throughout the olfactory paradigm so the subject could not detect

changes in airflow when odor was delivered. Respiration patterns during the execution of fMRI

paradigm were monitored and recorded via a chest belt. Respiration was monitored to confirm the

subject was awake throughout the paradigm.

40

Figure 2-1. The olfactory fMRI paradigm. Four concentrations of lavender were presented. Each

concentration was presented three times before the next higher concentration was presented in a

stepwise fashion. Odor presentation started with the weakest concentration and ended with the

strong concentration. The visual cue was a display of the words ―Smell?‖ and ―Rest‖ on an LCD

screen. When ―Smell?‖ appeared on the screen the subject provided responses using a button

press device in each hand, left hand if no smell and right hand if they smelled the stimulus.

41

2.3.4 Imaging Protocol (Used in chapters 3 and 4)

The imaging data scans were performed on a 3.0 T MRI system (Magnetom Trio,

Siemens Medical Solutions, Erlangen, Germany) with an 8 channel head coil. fMRI was utilized

to study the blood oxygen level dependent (BOLD) signal change during odor stimulation. A

BOLD signal sensitive T2*-weighted echo planar imaging sequence was used to acquire

functional data with slices = 34, slice thickness = 4mm, field of view (FOV) = 230 x 230,

acquisition matrix = 80 x 80, echo time (TE) = 30 ms, repetition time (TR) = 2000 ms, flip angle

(FA) = 90º, acceleration factor = 2, and acquisition time (TA) = 7 min 56 s with 234 repetitions.

T1-weighted images with 1 mm isotropic resolution were acquired with MPRAGE method for

structural assessment of the POC and hippocampus: TE = 2.98 ms, TR = 2300 ms, inversion time

(IT) = 900 ms, FA = 9º, FOV = 256 mm x 256 mm x 160 mm, acquisition matrix = 256 x 256 x

160, acceleration factor = 2, and TA = 6 min 21 s.

2.3.5 fMRI Data Processing and Analysis (Used in chapters 3)

Statistical Parametric Mapping (SPM8, Wellcome Trust Centre for Neuroimaging,

University College London, UK) was used to analyze all imaging data. The first 10 images were

discarded to remove initial transit signal fluctuations. The following standardized procedure was

used to preprocess the fMRI data: 1) spatial realignment within the session to remove any minor

head movements (movement < 2 mm, rotation < 1º); 2) co-registration with high-resolution

anatomical image; 3) normalization to the Montreal Neurological Institute (MNI) brain template

[38] in a spatial resolution of 2 mm x 2 mm x 2 mm; and 4) smoothing with an 8 mm x 8 mm x 8

mm (full width at half maximum) Gaussian smoothing kernel. A statistical parametric map was

generated at the individual level by fitting the stimulation paradigm to the functional data with a

default 128-s high pass filter, convolved with the canonical hemodynamic response function

42

(uncorrected, P < 0.001, extent threshold = 6). Olfactory activation maps at the group level were

generated using one-sample t-test (uncorrected, P < 0.001, extent threshold = 10). In this chapter,

we focused on the condition with ―Smell?‖ plus lavender odor. In chapter 3, we investigate the

condition with ―Smell?‖ plus odorless air. For several subjects, paradigm-correlated minor

movements were corrected by incorporating movement parameters as covariates in the paradigm

estimation step.

2.3.6 Region of Interest Analysis of the Primary Olfactory Cortex and Hippocampus (Used

in chapters 3)

FMRIB Software Library View (FSLview, Analysis Group, FMRIB, Oxford, UK) was

used to perform the bilateral manual segmentation of the hippocampus and POC on T1-weighted

images. The POC included the anterior olfactory nucleus, olfactory tubercule, piriform cortex,

anterior portion of the periamygdaloid cortex and amygdala, and anterior perforated substance

(Fig. 2-2) [32]. The hippocampus included the hippocampal formation, dentate gyrus, subiculum,

parasubiculum, and presubiculum. Segmentation of the ROIs was performed by two investigators

and reviewed by a neuro-radiologist (all blind to the subject’s group assignment). Each ROI

volume was corrected using the subject’s own intracranial volume. Once the bilateral volume was

calculated, the average was used to analyze the volumetric data. The ROIs were normalized and

then overlaid onto the fMRI maps to calculate the activated volumes per subject. These data were

analyzed using GraphPad Prism 6 (GraphPad Software San Diego, CA), IBM SPSS Statistics

software was used to perform the logistic regression analysis and create receiver operating

characteristic (ROC) curves, and MedCalc was used to compare the ROC curves.

http://www.fmrib.ox.ac.uk/analysis

43

Figure 2-2. 3D display of the primary olfactory cortex. The POC includes the anterior olfactory

nucleus, olfactory tubercule, piriform cortex, anterior portion of the periamygdaloid cortex and

amygdala, and anterior perforated substance.

44

2.4 Results

2.4.1 Demographics and Behavioral Results

Table 2-1 provides a summary of the demographic information and cognitive/behavioral

test results of the three groups. The behavioral tests (MMSE, CVLT-II, DRS-2, and UPSIT)

showed significant differences between the three groups (one-way analysis of variance

(ANOVA), P < 0.0001). The CN group attained significantly higher scores on the UPSIT, CVLT-

II, and DRS-2 than AD and MCI groups and the MCI group had higher scores than AD. The CN

and MCI groups also had significantly higher MMSE performance than the AD group. However,

the MCI group exhibited a large variation in neurocognitive and olfactory performances,

overlapping with scores from the CN and AD groups. Olfactory scores and cognitive tests

(CVLT-II: P < 0.0001, r = 0.66; DRS-2: P < 0.0001, r = 0.73; MMSE: P < 0.0001, r = 0.70) were

positively correlated, suggesting a strong association between olfactory and cognitive functions

(Fig. 2-3). Olfactory scores and cognitive tests were also correlated when only the patient

population was investigated (CVLT-II: P = 0.0054, r = 0.45; DRS-2: P = 0.0018, r = 0.50;

MMSE: P = 0.0003, r = 0.56).

45

Figure 2-3. Olfaction and cognitive tests. University of Pennsylvania Smell Identification Test

(UPSIT) scores correlated with the California Verbal Learning Test-II (CVLT-II) for all subjects

(A) and only patient population (B), Mini-Mental State Examination (MMSE) for all subjects (C)

and only patient population (D), and the Dementia Rating Scale-2 (DRS-2) for all subjects (E)

and only patient population (F).

46

2.4.2 Aging Effect

When examining all subjects, the UPSIT score (r = 0.33, P = 0.0094), hippocampal

volume (r = 0.42, P = 0.0006), and activation volume in the POC (r = 0.36, P = 0.0035) showed

strong negative correlations with age. The CN group alone exhibited significant aging effects in

the UPSIT score (r = 0.64, P = 0.0003), hippocampal volume (r = 0.58, P = 0.0014), and fMRI

activation volumes in the POC (r = 0.45, P = 0.019) and hippocampus (r = 0.40, P = 0.036). The

MCI group also showed significant age effects in the UPSIT scores (r = 0.45, P = 0.041),

hippocampal volume (r = 0.65, P = 0.0015), and fMRI activation in the POC (r = 0.49, P =

0.025). However, the data for the AD subjects showed no significant aging effects, indicating that

the predominant influence on these measurements were due to the disease. The measurements

that showed age effects were corrected for subsequent analyses.

2.4.3 Olfactory fMRI

Figure 2-4 illustrates the olfactory activation maps within the segmented hippocampus

and POC from the three cohorts (one-sample t-test, P < 0.001). In the CN group, strong activation

was observed in both ROIs. MCI and AD groups yielded much less activation in the two ROIs.

The fMRI data was quantified in terms of number of activated voxels in each subject’s segmented

ROIs (Fig. 2-5). The activated volume in the POC (one-way ANOVA, P < 0.0001) and

hippocampus (one-way ANOVA, P = 0.0064) showed significant differences among the groups.

A multiple comparisons test revealed that both the MCI and AD groups had more than 50% less

activation volume than the CN group. Although the patient groups differed in severity of

cognitive decline, both presented nearly the same level of reductions in fMRI activation volume

in these brain structures. In all subjects, a positive correlation was observed between activation

volume within the POC and hippocampus (r = 0.68, P < 0.0001). In the patient groups, a positive

47

correlation was detected between activation volume in the POC and hippocampus (r = 0.49, P <

0.005).

Figure 2-4. Activation in the primary olfactory cortex and hippocampus. Activation maps (hot

color, one sample t-tests, P < 0.001, uncorrected with extent threshold = 6) during odor

presentation. The color scale indicates the significance of activation. The underlay image for each

group is the mean T1-weighted image (Montreal Neuroimaging Institute (MNI) space, Z = -29 to

-17) of the subjects within the cohort. The average segmented primary olfactory cortex (blue)

and the hippocampus (cyan) ROIs from the cognitively normal controls (CN) are indicated. The

CN group had significantly greater activation in both ROIs compared with the mild cognitively

impaired and Alzheimer’s disease groups.

48

Figure 2-5. Activation volume in AD and MCI (mean ± standard error). Activation volume

(voxels activated) decreased by greater than 50 percent for both the Alzheimer’s and mild

cognitive impairment patients in the primary olfactory cortex (A) and hippocampus (B).

Notes: * P ≤ 0.05,

*** P ≤ 0.001,

49

2.4.4 Relations of Brain Volume and Activation Volume in the Primary Olfactory Cortex

and Hippocampus

We quantitatively analyzed the relationship of olfactory activation change and the local

structural change (atrophy) to determine if the activation decreases were due to the volume of the

ROIs. The ratio of the fMRI activation volume and the brain volume of the structure (i.e., percent

volume activated) was utilized to examine this relationship. As shown in figure 2-6A and 2-6B,

MCI and AD subjects had more than a two-fold reduction in percent volume activated compared

with the CN group (30% volume activated for the hippocampus and 55% volume activated for the

POC) in both the hippocampus and the POC. Most interestingly, the MCI group showed nearly

the same level of deficit in percent volume activated as the AD group, with each group having

approximately 11% volume activated in the hippocampus (Fig. 2-6A) and 26% volume activated

in the POC (Fig. 2-6B). The percent volume activated was reduced in both groups by up to 55.4%

for the hippocampus and by up to 52.2% for the POC.

Figure 2-6 (C and D) shows the comparisons in POC and hippocampal volumes between

the groups. Significant brain atrophy was observed in the hippocampus (one-way ANOVA, P <

0.0001) and POC (one-way ANOVA, P < 0.001). A multiple comparison test demonstrated that

both MCI and AD subjects had significantly smaller hippocampal (P < 0.0001) and POC (P =

0.001) volumes than the CN group. While the comparison showed greater reduced volumes of

both ROIs in the AD subjects than the MCI subjects, the difference did not reach significance.

The volumes of the hippocampus and the POC were positively correlated among the three groups

(r = 0.55, P < 0.0001). Specifically, the volumes of the ROIs were positively correlated when

examining just the patient population (r = 0.38, P < 0.05).

50

Figure 2-6. Structural and functional changes. Percent volume activated and volumes of the

hippocampus and primary olfactory cortex (POC) (mean ± standard error). The percent volume

activated in the hippocampus (A) and POC (B) in mild cognitive impaired (MCI) and

Alzheimer’s disease (AD) subjects was decreased by more than 50 percent than that of the

cognitively normal controls (CN). The hippocampus (C) and POC (D) were significantly smaller

in volume in both MCI and AD subjects compared with CN.

Notes: * P ≤ 0.05,

** P ≤ 0.01,

*** P ≤ 0.001,

**** P ≤ 0.0001.

51

The volumes of the ROIs were positively correlated with the activation volume in the

POC (r = 0.35, P = 0.0053), and in the hippocampus (r = 0.40, P = 0.0013), respectively.

However, the decrease in activation volume in the patient groups was greater than the volume

changes.

2.4.5 Correlation Between the Behavioral and MRI Results

All significant correlations between the behavioral tests and MRI results are listed in

Table 2-2. Greater brain volumes, activation volume, and percent activation volumes in the ROIs

were correlated with higher cognitive and olfactory scores. However, the volume measurements

had a greater correlation with the behavioral tests than the activation volume and percent

activation volumes.

52

Table 2-2. Correlations between behavioral and MRI measurements of all subjects.

UPSIT CVLT-II MMSE DRS-2

Hippocampal Volume P < 0.0001

r = 0.55

P = 0.0006

r = 0.42

P < 0.0001

r = 0.47

P < 0.0001

r = 0.55

POC Volume P < 0.0001

r = 0.55

P = 0.006

r = 0.34

P = 0.001

r = 0.40

P = 0.004

r = 0.36

Hippocampal Activation

Volume

P = 0.02

r = 0.29

NS P = 0.02

r = 0.30

P = 0.007

r = 0.34

POC Activation Volume P = 0.0002

r = 0.45

P = 0.002

r = 0.39

P = 0.001

r = 0.40

P = 0.0008

r = 0.41

Hippocampal Percent

Volume Activated

NS NS P = 0.04

r = 0.26

P = 0.03

r = 0.28

POC Percent Volume

Activated

P = 0.002

r = 0.39

P = 0.02

r = 0.29

P = 0.005

r = 0.35

P = 0.007

r = 0.34

Abbreviations: POC, primary olfactory cortex; UPSIT, University of Pennsylvania Smell

Identification Test; CVLT-II, California Verbal Learning Test- Short Form Version 2; not

significant, NS; MMSE, Mini-Mental State Examination; DRS-2, Dementia Rating Scale 2.

53

2.4.6 Logistic Regression Analysis

Logistic regression analysis of all groups showed a composite measure of MRI volume,

fMRI activated volume, CVLT-II, and UPSIT scores was 90.5% accurate for group prediction.

The CN group was predicted with 96.3% accuracy (one subject was incorrectly predicted to be

MCI). The MCI group was predicted with 85.7% accuracy (two subject were incorrectly

predicted to be CN and one was predicted to be AD). And lastly, the AD group was predicted

with 86.7% accuracy (two subjects were incorrectly predicted to be MCI). When looking at just

the CN and MCI groups, a composite measure based on activated volume in the POC,

hippocampal volume, and UPSIT scores was 93.8% accurate. The CN group was predicted with

92.6% accurancy and the MCI group was predicted with 95.2% accuracy. Adding the UPSIT

scores and/or olfactory fMRI improved specificity and sensitivity for distinguishing AD/MCI

from CN than using the hippocampal volume data alone (Fig. 2-7). Adding UPSIT score and

POC activation volume to the volume of hippocampus yielded a sensitivity of 0.952, a specificity

of 0.926, and the area under the ROC curve was 0.972 for the classification of MCI (Fig. 2-7A).

ROC curve comparison showed a significant difference when UPSIT and POC activation volume

was added to the hippocampal volume (P ≤ 0.05). Adding UPSIT scores to the hippocampal

volume improved the sensitivity and specificity to 100% and the area under the ROC curve was

1.0 for the classification of AD (Fig. 2-7B). This curve also showed a statistically significant

increase compared to just hippocampal volume (P ≤ 0.05). Olfactory fMRI and UPSIT data

increased the specificity and sensitivity of MCI and AD diagnosis in our sample.

54

Figure 2-7. Receiver operating characteristic (ROC) curves. University of Pennsylvania Smell

Identification Test (UPSIT) and/or olfactory fMRI improve the specificity and sensitivity of

classifying Alzheimer’s disease (AD) and mild cognitive impaired (MCI) from cognitively

normal controls (CN). Distinguishing MCI from CN was improved to 0.972 when including

volume of hippocampus, primary olfactory cortex (POC) activation volume, and UPSIT score as

classifiers (A) and distinguishing AD from CN was improved to 1 by using UPSIT score together

with the volume of the hippocampus as classifiers (B).

55

2.5 Discussion

In our study, the UPSIT confirmed clinical manifestations of olfactory deficits in AD and

MCI, which is consistent with literature [1-10]. We identified a strong positive correlation

between the UPSIT and cognitive test scores (r ≥ 0.67, r-squared ≥ 45%, P ≤ 0.0001); suggesting

cognitive decline is associated with olfactory deficits. When investigating just the patient

population the positive correlation remained (r ≥ 0.45, r-squared ≥ 20%, P ≤ 0.005). The normal

controls alone did not show a significant correlation between olfactory and cognitive

performance. This was expected due to the fact that the cognitive battery used in this study is

better at finding cognitive abnormality than it is at staging cognitively functioning normal

individuals. The age effects observed in the CN and MCI groups in olfactory, volumetric, and

activation results were completely absent in the AD group. Evidently, the functional and

pathological abnormalities present in AD totally deviated from the normal aging behaviors in all

aspects of our assessments. The presence of age dependencies and a general trend toward AD-

related behavior shown in the measurements of the MCI group concords with their at-risk and in-

transitional-state status.

Utilizing concurrent volumetric and functional MRI measurements, we demonstrated that

the POC is degenerating in AD and that its volume decrease is comparable to the hippocampus,

which is considered to be the gold standard. Of note, our hippocampal volume results in the

patient groups agree with previous literature [23-29]. In this study, we found a significant positive

correlation between the POC and hippocampal volumes (r = 0.55, r-squared = 30%, P ≤ 0.0001)

and both ROIs were significantly smaller in MCI and AD subjects compared with CN subjects.

These findings provide the first in vivo evidence of the involvement of primary central olfactory

structures beyond olfactory bulb and tract in AD pathology. We also observed a positive

correlation between atrophy in the POC and UPSIT scores (r = 0.55, r-squared = 30%, P ≤

56

0.0001), establishing a specific relationship between olfactory behavioral deficits and

pathological changes in a key brain structure for olfaction. Such a relationship has been

previously hypothesized based on postmortem pathological observations that greater AD

pathology was found in central olfactory regions compared with other sensory systems but has

not been tested with in vivo studies.

The olfactory deficits in AD and MCI create a unique opportunity for fMRI to directly

address the functional consequences of neuropathological changes (i.e., cell death/atrophy) in the

involved brain regions. As our results demonstrated, the volumes of the POC and hippocampus

positively correlated with the olfactory activated volume within these structures. Furthermore, the

activated volumes in these structures in AD and MCI subjects were reduced to less than one-half

of the CN sample. Most interestingly, unlike the marked descending trend in brain volume from

CN, to MCI and to AD, the decline of activated volume by the MCI group in the two structures

reached nearly the same level as that of the AD subjects. This result suggests that the olfactory

function in the POC at the MCI stage has deteriorated to a similar level as that in AD, preceding

volumetric changes. This provides an explanation for why activation in the POC was less

correlated to behavioral performance compared to the POC volume. While the behavioral

performance showed a gradual decline, the fMRI results showed a more drastic decline in the

MCI group. From these results, we can draw two important conclusions: 1) the decrease in

activated volume in the MCI and AD groups was not exclusively due to decrease in the respective

volumes of the ROIs. It has been long postulated that reduction of fMRI activation in AD could

be proportional to the brain volume atrophy in the specific brain regions. Previous fMRI

experiments using cognitive paradigms have been inconclusive in this regard, partially because of

the compensatory mechanism of the brain to recruit other regions as resources for cognitive tasks

and the associated confounding variables (education levels, intelligence, etc.) [39-41]. Our

57

olfactory fMRI paradigm, however, involved an odor perception task with minimal cognitive

demand. In this case, compensatory mechanisms are less likely. The task only involved

identifying if an odor was present or not, the subject did not have to identify the odorant,

therefore, very little cognitive effort was needed in this paradigm. 2) From a clinical perspective,

these results suggest that olfactory fMRI activation volume in the POC could potentially offer a

more sensitive functional imaging marker for the early detection of AD than volumetric

measurements. Note that the difference in POC olfactory fMRI activation was much greater than

that seen in POC volume between the groups; the POC activation showed a more drastic drop for

both the MCI and AD subjects while the volume in MCI and AD subjects was more gradual. The

results from our ROC analysis demonstrated that adding olfactory data (fMRI and UPSIT) to

hippocampal volume measurement significantly increased diagnostic sensitivity and specificity

for classifications of both MCI and AD (P ≤ 0.05). In this regard, the fact that olfactory deficits

occur at the early stages of the disease, as memory and cognitive impairments begin to emerge,

immediately offers two advantages. First, it offers a technical advantage of reducing the

difficulties in data collection and interpretation associated with advanced memory and cognitive

deficits in later stages of AD. Second, it offers a simple effective experimental paradigm with

minimal cognitive confounds and associated variability, which increases the feasibility for

utilizing fMRI to study preclinical and early AD. The combination of olfactory fMRI,

hippocampal volume measurement, and UPSIT results may provide a way to predict status of

disease onset or a tool to monitor the disease progression. Further longitudinal studies in the MCI

population will elucidate the ability to predict which MCI patients will develop AD and when.

The degeneration of the brain tissue in AD is likely to begin decades before onset of

clinical symptoms. This may be associated with a gradual decline in memory and cognition over a

long pre-clinical period. These deficits can be masked by the compensatory mechanism in

58

standard neuropsychological tests [39-41]. However, the deficits in olfactory function are

unlikely to be offset by a compensatory mechanism because our olfactory task is passive with

minimal motivational/cognitive confounds. For example, as seen in Table 2-1, UPSIT (involves

cognitive processes) scores followed a similar trend of gradual decline from CN to MCI to AD as

the cognitive data. This phenomenon has been attributed to compensatory mechanisms at the MCI

stage and their failure when the disease progresses to AD. Conversely, the functional activation

volume in the MCI group had reached the same level as the AD group. Thus, olfactory fMRI tests

can be more sensitive in the early detection of AD.

Several limitations in our study should be addressed. Our current study was limited by

the small sample size of the patient groups and our MCI group consisted of patients who will

convert to AD and those who will not. We also did not control for the MCI subjects who were

being treated with a cholinesterase inhibitor and/or memantine. A comparison between MCI on

cholinesterase inhibitor and/or memantine and MCI who were not being treated with medication

did however show that there were no significant differences in cognitive and olfactory

performance, hippocampal and POC volume, or hippocampal and POC activation. To validate

our findings future investigations should include longitudinal studies using the gold standard of

post mortem findings to confirm the diagnosis with larger MCI and AD cohorts. In addition, our

current analysis was focused only on two ROIs. Further analyses will include broader brain areas

using functional network analysis in order to understand the relationship between olfactory

deficits and cognitive impairments. Manual segmentation of the ROIs was carried out in this

study, which is time consuming and impractical for future clinical trials and applications. This

highlights the need to develop more efficient processing tools for future studies. Finally, fMRI

studies are expensive so it may not seem feasible for all patients to have an MRI. It is important

59

to note, however, that almost all MCI patients already have an MRI done for differential

diagnosis.

In summary, our study provided in vivo volumetric MRI data showing that the POC is

involved in MCI and AD, and likely provides the basis for the olfactory deficits in these patients.

We demonstrated that not only was the volume of the POC decreased in a similar fashion to that

of the hippocampus in both MCI and AD patients, but also that olfactory fMRI showed greater

differences in functional activation than the morphological or behavioral differences between the

groups. Of great importance, our study also showed that olfactory fMRI could be used in

conjunction with volume measurement of hippocampus and the UPSIT to increase the diagnostic

sensitivity and specificity of at risk patients. It also provides the potential to be utilized to study

disease progression and with longitudinal studies it provides the potential to help identify which

MCI patients will develop AD.

60

2.6 References

[1] Ferreyra-Moyano H. The olfactory system and Alzheimer’s disease. Int J Neurosci 1989;

49:157-97.

[2] Knupfer L, Spiegel R. Differences in olfactory test performance between normal aged,

Alzheimer and vascular type dementia individuals. Int J Geriat Psychiat 1986; 1:3-14.


associated with degree of dementia in Alzheimer’s disease. Neurobiol Aging 1990; 11:465-

9.

[4] Warner MD, Peabody CA, Flattery JJ, Tinklenberg JR. Olfactory deficits and Alzheimer’s

disease. Bio Psychiat 1986; 21:116-8.


performance are dissociated in early Alzheimer’s disease. Neurology 1988; 38:1228-32.



1988; 9:399-403.

[7] Serby M, Larson P, Kalkstein DS. The nature and course of olfactory deficits in

Alzheimer’s disease. Am J Psychiatry 1991; 148:357-60.



1995; 15:793-803.

[9] Doty RL, Reyes PF, Gregor T. Presence of both odor identification and detection deficits in

Alzheimer’s disease. Brain Research Bulletin 1987; 18:597-600.

[10] Murphy C, Nordin S, Acosta L: Odor learning, recall, and recognition memory in young and

elderly adults. Neuropsychology 1997; 11:126-37.

61








[14] Pearson RCA, Esiri MM, Hiorns RW, Wilcock GK, Powell TPS. Anatomical correlates of

the distribution of the pathological changes in Alzheimer’s disease. Proc Nat Acad Sci

(USA) 1985; 82:4531-4.


Med 1988; 318:789-90.


73: 365-9.



1991; 12: 295-312.




Psychiat 1984; 47:56-60.


Neuropathol 1991; 82: 239-59

62

[21] Yang J, Pan P, Song W, Huang R, Li J, Chen K, et al. Voxelwise meta-analysis of gray

matter anomalies in AD and MCI using anatomic likelihood estimation. J Neurol Sci. 2012;

15:21-9

[22] Ferreira LK, Diniz BS, Forlenza OV, Busatto GF, Zanetti MV. Neurostructural predictors of

Alzheimer's disease: a meta-analysis of VBM studies. Neurobiol Aging. 2011; 32:1733-41

[23] Mega MS, Thompson PM, Toga AW, Cummings JL. Brain mapping in dementia: In:

Mazziotta JC, Toga AW, Frackowiak RSJ, editors. Brain mapping: the disorders. San Diego:

Academic Press; 2000, p. 218-48.

[24] Jack CR, Petersen RC, O’Brien PC, Tangalos EG. MR-based hippocampal volumetry in the

diagnosis of Alzheimer’s disease. Neurology 1992; 42: 183-8.

[25] Kesslak JP, Nalcioglu O, Cotman CW. Quantification of magnetic resonance scans for

hippocampal and parahippocampal atrophy in Alzheimer’s disease. Neurology 1991; 41:51-

4.

[26] Killiany RJ, Moss MB, Albert MS, Sandor T, Tieman J, Jolesz F. Temporal lobe regions on

magnetic resonance imaging identify patients with early Alzheimer’s disease. Arch Neurol

1993; 50: 949-54.

[27] Convit A, de Leon MJ, Golomb J, George AE, Tarshish CY, Bobinski M, et al.

Hippocampal atrophy in early Alzheimer’s disease: anatomic specificity and validation.

Psychiatr Q 1993; 64:371-87.

[28] Foundas AL, Leonard CM, Mahoney M, Agee OF, Heilman KM. Atrophy of the

hippocampus, parietal cortex, and insula in Alzheimer’s disease: a volumetric magnetic

resonance imaging study. Neurol Neuropsychol Behav Neurol 1997; 10:81-9.

[29] Jack CR, Petersen RC, Xu YC, Waring SC, O’Brien PC, Tangalos EG, et al. Medial

temporal atrophy on MRI in normal aging and very mild Alzheimer’s disease. Neurology

1997; 49:786-94.

63

[30] Thomann PA, Dos Santos V, Seidl U, Toro P, Essig M, Schröder J. MRI-derived atrophy of

the olfactory bulb and tract in mild cognitive impairment and Alzheimer's disease. J

Alzheimers Dis. 2009; 17:213-21

[31] Prestia A, Baglieri A, Pievani M, Bonetti M, Rasser PE, Thompson PM, et al. The in vivo

topography of cortical changes in healthy aging and prodromal Alzheimer's disease. Suppl

Clin Neurophysiol. 2013; 62:67-80





[34] McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clnical diagnosis

of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of

Department of Health and Human Services Task Force on Alzheimer’s disease. Neurology

1984; 34:939-44.

[35] Peterson RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive

impairment: clinical characterization and outcome. Archives of Neurology 1999; 56:303-8.

[36] Allen W. Studies on the level of anesthesia for the olfactory and trigeminal respiratory

reflexes in dogs and rabbits. Am J Physiol 1936; 115:579-87.

[37] Karunanayaka P, Eslinger PJ, Wang JL, Weitekamp CW, Molitoris S, Gates KM, Molenaar

PC, Yang QX. Networks involved in olfaction and their dynamics using independent

component analysis and unified structural equation modeling. Hum Brain Mapp 2013; Epub

ahead of print

[38] Collins DL, Zijdenbos AP, Kollokian V, Sled JG, Kabani NJ, Holmes CJ, et al. Design and

construction of a realistic digital brain phantom. IEEE Trans Med Imaging 1998; 17:463-8.

64

[39] Johnson SC, Saykin AJ, Baxter LC, Flashman LA, Santulli RB, McAllister TW, et al. The

relationship between fMRI activation and cerebral atrophy: comparison of normal aging and

Alzheimer disease. Neuroimage 2000; 11:179-87.

[40] Buchsbaum MS, Kesslak JP, Lynch G, Chui H. Temporal and hippocampal metabolic rate

during an olfactory memory task assessed by positron emission tomography in patients with

dementia of the Alzheimer type and controls: Preliminary studies. Arch Gen Psychiat 1991;

48:840-7.

[41] Raichle ME, Fiez JA, Videen TO, MacLeod AM, Pardo JV, Fox PT, et al. Practice-related

changes in human brain functional anatomy during nonmotor learning. Cereb Cortex 1994;

4:8-26.

65

Chapter 3

Central Olfactory Dysfunction is the Dominant Cause of Olfactory Deficits in AD and MCI

3.1 Abstract

Introduction: Olfactory deficits are present in Alzheimer’s disease (AD) and mild cognitively

impaired (MCI) patients. However, whether these deficits are due to dysfunction of the central or

peripheral olfactory system is unknown. In this study we further investigate the central olfactory

system to elucidate whether it is the dominant system causing olfactory deficits in AD and MCI

patients.

Methods: The same subjects and data acquired in Chapter 2 will be used here. Twenty-seven

cognitively normal controls (CN), 21 MCI, and 15 AD subjects completed structural and

olfactory functional MRI (fMRI) studies. The olfactory fMRI consisted of lavender odorant and

clean air presentation with a visual stimulus.

Results: The CN subjects had greater activated volume of the hippocampus and primary olfactory

cortex during both the odor and no odor presentations conditions than either the MCI or AD

subjects (P < 0.05). Significant differences were not observed between the odor and no odor

conditions for each group. Both conditions correlated with the cognitive and olfactory tests.

Conclusion: The no odor condition elicited the same functional response as the odor condition for

each of the three groups showing dysfunction of the central olfactory system. This suggests that

the central olfactory system is the dominant system causing the olfactory dysfunction present in

AD and MCI patients.

66

3.2 Introduction

Patients with Alzheimer’s disease (AD) and mild cognitive impairment (MCI), a

population of high-at-risk for AD, present with olfactory deficits in threshold detection, odor

memory, and odor identification. Several studies have reported lower scores on olfactory tests

such as the University of Pennsylvania Smell Identification Test (UPSIT) and the Sniffin’ Sticks

in AD and MCI subjects compared with age-matched normal controls [1-11]. The MCI patients

show less olfactory dysfunction compared with the AD patients. Longitudinal studies have also

shown that olfactory symptoms in AD correlate to cognitive decline and the progression of the

disease. It has been hypothesized that the olfactory deficits observed behaviorally in AD are

caused by the pathological changes (plaques, neurofiblirary tangles, and atrophy) in the central

nervous system. Indeed, AD pathologies have been demonstrated in the central olfactory system

[12-18]. Plaques and neurofiblirary tangles can be found in the olfactory bulb, anterior olfactory

nucleus, piriform cortex, and olfactory epithelium at the earliest stages of AD. The pathology in

these regions increases as the disease progresses and the symptoms worsen.

While it is known that olfactory deficits are prevalent in AD and in MCI, it is not known

whether the olfactory deficits derive from central or peripheral pathology. The olfactory system

can be divided into central and peripheral components. The peripheral olfactory system includes

the olfactory epithelium and olfactory nerve; and the central olfactory system includes the

olfactory bulb, olfactory tract, anterior olfactory nucleus, piriform cortex, amygdala, olfactory

tubercule, hippocampus, and the orbitofrontal cortex [19]. The peripheral olfactory system is

involved in the initial detection of odorants while the central part is involved in integrating and

processing the signal. The olfactory receptors in the olfactory epithelium project to the mitral

cells of the olfactory bulb via the olfactory nerve. The axons from the mitral cells travel to the

brain via the olfactory tract and project primarily to the piriform (primary olfactory cortex),

67

olfactory tubercle, amygdala, and entorhinal cortex. The primary olfactory cortex (POC) includes

the piriform cortex, entorhinal cortex, anterior cortical nucleus of the amygdala, and the

periamygdaloid cortex. Neurons from here send projections to the dorsomedial nucleus of the

thalamus, the basal forebrain, the limbic system, and the hippocampus [7-20].

Very few studies have discussed the role of the central olfactory system in AD and MCI

versus the role of the peripheral olfactory system; however, behavioral and pathological studies

have leaned towards olfaction being a central rather than peripheral problem in AD patients.

Based on higher performance on odor identification tests compared to threshold detection tests in

AD and MCI subjects, few studies have concluded a central olfactory problem where processing

and integration of the odorant is the issue [21-22]. The strongest evidence is provided by a few

post-mortem studies that also indicate central olfactory dysfunction as the cause of the olfactory

deficits. An autopsy study reported less severe pathology in the peripheral olfactory areas

compared with the central olfactory areas suggesting a central problem [23]. Ter Laak et al also

concluded identification is processed in the central olfactory structures based on neuron loss in

the anterior olfactory nucleus [24]. Another study also suggested olfactory impairments

associated with AD are likely due to damage in the central olfactory pathways based on

neuropathological changes in the olfactory epithelium and central olfactory pathways [20]. While

these studies support central system dominance, post-mortem studies overall are inconclusive in

their findings because they cannot definitively rule out a dominant peripheral olfactory

contribution in AD and MCI. Several other autopsy studies have reported pathology in and

degeneration of peripheral olfactory regions including axons of the olfactory tract and olfactory

epithelium [14-15, 17, 25]. Current literature focuses on behavioral and pathological distribution

to provide evidence for the involvement of the central olfactory system and also more recently

functional deficits of the central olfactory structures in AD have been detected using olfactory

68

functional magnetic resonance imaging (fMRI) [26-27]. fMRI studies require peripheral afferent

information and therefore, still cannot definitively rule out peripheral olfactory dominance. Thus,

the central- or peripheral-dominant olfactory problem in AD and MCI remains open.

In our study, we further investigate this uncertainty using an olfactory functional

magnetic resonance imaging (fMRI) study. The paradigm includes a visual cue (―SMELL?‖)

accompanied by either odor presentation or no odor presentation in order to elucidate the

dominance of the central olfactory system. The visual stimulus is used to enhance the olfactory

activation and examine the central olfactory regions when an olfactory stimulus is not provided

[28]. We specifically investigated the POC and the hippocampus (role in processing and

perception of olfactory information). We hypothesize that the olfactory deficits present in AD and

MCI patients are due to central olfactory system dysfunction.

3.3 Methods

3.3.1 Study Cohort

The same subjects outlined in section 2.3.1 were used for this investigation. Sixty-three

subjects were enrolled in this study, including 15 AD subjects (Clinical Dementia Rating Scale

(CDR) of 0.5 or 1), 21 MCI subjects (CDR of 0.5), and 27 age-matched CN, (Table 2-1).

3.3.2 Behavioral Tests

All behavioral tests were discussed previously in section 2.3.2. All participants were

administered the University of Pennsylvania Smell Identification Test (UPSIT) to assess their

smell identification function, and clinical neurocognitive examinations, which included the Mini-

69

Mental State Examination (MMSE), the Mattis Dementia Rating Scale-2 (DRS-2, and the

California Verbal Learning Test-Second Edition Short Form (CVLT-II).

3.3.3 Olfactory Stimulation Paradigm

Olfactory stimulation paradigm and odorant used were previously discussed in section

2.3.3. The odor stimulation paradigm was executed using a programmable olfactometer.

Lavender oil (Givaudan Flavors Corporation, East Hanover, NJ, USA) diluted in 1,2-propanediol

(Sigma, St. Louis, MO, USA) was used as the olfactory stimulant. Please refer to section 2.3.3

and figure 2-1 for details regarding the olfactory paradigm.

3.3.4 Imaging Protocol

The imaging protocol is discussed in detail in section 2.3.4. Scans were performed on a

3.0 T MRI system (Magnetom Trio, Siemens Medical Solutions, Erlangen, Germany) with an 8

channel head coil. fMRI was utilized to study the blood oxygen level dependent (BOLD) signal

change responding to the odor stimulation in the POC and the hippocampus. A BOLD signal

sensitive T2*-weighted echo planar imaging sequence was used to acquire functional data.

3.3.5 fMRI Data Processing and Analysis

Statistical Parametric Mapping (SPM8, Wellcome Trust Centre for Neuroimaging,

University College London, UK) was used to analyze all imaging data. The details of the

standardized procedure are found in section 2.3.5.

70

3.3.6 Region of Interest Analysis of the Primary Olfactory Cortex and Hippocampus

This procedure is the same as described in section 2.3.6. FMRIB Software Library View

(FSLview) was used to perform the bilateral manual segmentation of the hippocampus and POC

on T1-weighted images from each subject (Fig. 2-2) [26]. The volumes of the ROIs from each

subject were corrected using the intracranial volume of the subject. Once the bilateral volume was

calculated it was used to analyze the volumetric data. The ROIs were normalized and then

overlaid onto the fMRI maps to calculate the activated volumes within the ROIs during both odor

(visual cue ―Smell?‖ and lavender odor) and no odor (visual cue ―Smell?‖ and fresh air)

conditions. Each of the four lavender concentrations was also examined to investigate habitation

effects. The UPSIT scores and fMRI data were corrected for age effects. This data was analyzed

using GraphPad Prism 6 (GraphPad Software San Diego, CA).

3.4 Results


Table 2-1 shows the demographic information and cognitive/behavioral test results of the

three subject cohorts. As stated in section 2.4.1, the behavioral tests (MMSE, CVLT-II, DRS-2,

and UPSIT) showed significant differences between the three groups (one-way analysis of

variance (ANOVA) analysis, P < 0.0001). Multiple comparisons tests showed that the CN group

had higher scores, indicating overall greater neurocognitive and smell identification functions

(UPSIT, CVLT-II, and DRS-2). However, the MCI group exhibited a large variation in all

behavioral tests, overlapping with scores from CN and AD groups.

71

3.4.2 Aging Effect

Aging effects were seen in olfactory scores and fMRI measurements. These effects are

discussed in detail in section 2.4.2. Thus, the measurements that showed age effects were

corrected for subsequent analyses.

3.4.3 Olfactory fMRI

Figure 3-1 shows the olfactory activation maps within the segmented POC and

hippocampus from the three study groups for both the odor and no odor conditions (one sample t-

test, P < 0.001, extent threshold = 6). In the CN group, strong activation was observed in the two

ROIs; however, both MCI and AD groups yielded much less activation in both ROIs. This was

true for both conditions, although a more drastic decrease in activation was observed in the MCI

group for the odor condition while a decreasing stepwise trend was present in the no odor

condition.

The data from the olfactory fMRI paradigm was quantified in terms of activated voxels in

the segmented ROI for each subject (Fig. 3-2). During odor presentation (―Smell?‖ and lavender),

the activated volume in the POC (one-way ANOVA, P = 0.0002) and hippocampus (one-way

ANOVA, P = 0.01) showed significant group differences as seen in Chapter 2. A multiple

comparisons test revealed that both patient groups had greater than 50% less activation volume

when compared to the CN group with both MCI and AD groups presenting nearly the same level

of reductions in fMRI activation volume in the POC and hippocampus. Similarly during the no

odor (―Smell?‖ and fresh air) condition the activated volume in the POC (one-way ANOVA, P =

0.0007) and hippocampus (one-way ANOVA, P = 0.03) showed significant differences among

the three groups. The data here was more stepwise showing CN had significantly greater

72

activation than both AD and MCI, and MCI tended to have greater activation than the AD group;

however the difference between MCI and AD subjects did not reach significance. Significant

differences between the no odor and odor conditions were not found in any of the three groups,

showing that both the odor and no odor conditions produced similar results.

73

Figure 3-1. Olfactory activation maps. Activation maps (one sample t-tests, P < 0.001,

uncorrected with extent threshold = 6) for both odor (A) and no odor (B). Activation is shown

only in the average primary olfactory cortex (POC) and the hippocampus from the cognitively

normal controls (CN). The color scale indicates the significance of activation. The underlay

image for each group is the mean T1-weighted image (Montreal Neuroimaging Institute (MNI)

space, Z = -28 to -14) of the subjects within the cohort. The CN group had significantly greater

activation in both ROIs compared with the mild cognitively impaired and Alzheimer’s disease

groups in both the odor and no odor conditions.

74

Figure 3-2. Activated volume. Activated volume in the primary olfactory cortex (POC) and

hippocampus (mean ± standard error) during odor and no odor conditions. The activated volume

in the POC (A) and hippocampus (B) in mild cognitive impaired (MCI) and Alzheimer’s disease

(AD) subjects was decreased by more than 50 percent than that of the cognitively normal controls

(CN) during odor presentation. The no odor conditions showed a decrease in activation in a more

stepwise fashion.

Notes: * P ≤ 0.05, ANOVA when compared with CN- All Odors

+ P <0.05, ANOVA when compared with CN- No Odor

75

3.4.4 Correlation Between the Behavioral and MRI Results

Positive correlations were demonstrated between the behavioral tests and activation in the

POC during both the odor and no odor conditions; greater functional activation was correlated

with higher cognitive and olfactory scores (Table 3-1). A positive correlation was also observed

for the hippocampus during odor and no odor conditions; however, the activation in the POC for

both conditions demonstrated a greater correlation to the behavioral tests than the activation in the

hippocampus.

3.4.5 Four Lavender Concentrations

Analysis of activation within the POC during presentation of each concentration showed

an overall stabilization through the paradigm (Fig. 3-3). All four concentrations (―Smell?‖ and

lavender) produced a similar amount of activation as one another. In the POC, the CN subjects

(approximately 300 voxels activated for each concentration) had the highest activation while the

MCI and AD (both groups had approximately 100 activated voxels for each concentration)

subjects looked similar for the all four concentrations. However, during the -3.5 concentration the

AD subjects had greater activation than the other 3 higher concentrations, but this did not reach

significance. Similar results were found in the hippocampus for the CN subjects (130 to 150

activated voxels for each concentration). The pattern of activation was stable during presentation

of all concentrations. This was also seen for the MCI subjects (30 to 50 activated voxels for each

concentration). While the AD group showed a decreasing trend as the concentration of the

lavender odor increased, the change did not reach significance. The hippocampus, overall, had

less activation compared with the POC for each concentration of lavender.

76

Abbreviations: POC, primary olfactory cortex; UPSIT, University of Pennsylvania Smell

Identification Test; CVLT-II, California Verbal Learning Test- Short Form Version 2; not

significant, NS; MMSE, Mini-Mental State Examination; DRS-2, Dementia Rating Scale 2.

Table 3-1. Correlations between behavioral and imaging measurements of all subjects.

POC- Odor

Condition

POC- No

Odor

Condition

Hippocampus-

Odor Condition

Hippocampus-

No Odor

Condition

CVLT-II P = 0.0002

r = 0.39

P < 0.0001

r = 0.49

NS NS

MMSE P = 0.001

r = 0.40

P < 0.0001

r = 0.54

P = 0.02

r = 0.30

NS

DRS-2 P = 0.0008

r = 0.41

P < 0.0001

r = 0.54

P = 0.007

r = 0.34

P = 0.02

r = 0.30

UPSIT P = 0.0002

r = 0.45

P < 0.0001

r = 0.48

P = 0.02

r = 0.30

P = 0.04

r = 0.26

77

Figure 3-3. Four concentrations. Activated volume during each concentration presented in the

POC (A) and hippocampus (B). Both ROIs showed stabilized activation during the whole

paradigm. No significant differences between the four concentrations were found. Cognitively

normal controls had greater activation for all concentrations compared with MCI and AD

subjects, while MCI and AD subjects had similar activated volume in both the POC and the

hippocampus.

78

3.5 Discussion

In this study, we established the dominant role of the central olfactory system in

Alzheimer’s and MCI. The use of a visual cue without any odor allowed for analysis of the POC

with an afferent stimulus that was perceived as equal for all of the subjects. Thus without an

odorant, activation in the POC was expected to have been equal between the three groups;

however, the normal controls had greater activation signal change compared with both the

Alzheimer’s and MCI subjects. In fact, the results of the odor and no odor conditions for each

group showed no significant differences.

The olfactory system is unique in that it is the only sensory system with a direct

connection to the cortex without a relay through the thalamus. An indirect path going through the

thalamus also exists and this path is believed to serve as a conscious mechanism for odor

perception. The amygdala and entorhinal areas (components of the limbic system) on the other

hand are more involved in the affective components of olfaction [7-20]. Olfactory dysfunction is

observed in several neurodegenerative diseases and specifically in AD and MCI patients.

It has been hypothesized that the olfactory dysfunction observed in AD and MCI subjects

is a central problem rather than a peripherally dominant problem. This has been reported by

studies showing greater deficits in odor identification tests than odor detection tests. Identification

of an odorant takes memory and one of the key symptoms in AD is memory deficit, suggesting a

possible relationship between olfaction and memory. Also, studies investigating AD pathology

find plaques, neurofiblirary tangles, and atrophy in the central olfactory regions. However, these

studies have also shown the same pathology in the peripheral olfactory system. Therefore, while

it is suggested that olfactory deficits are due to degeneration of the central olfactory system, it is

still uncertain that the olfactory dysfunction is not peripherally dominant [21-22]. In our study,

79

we investigated activation specifically in the POC and in the hippocampus to demonstrate the

dominance of the central olfactory system in olfactory deficits in AD and MCI patients during an

fMRI task. The paradigm involved presentation of a visual cue that was accompanied with either

an odor or with no odor (fresh air) (Fig. 3-4). The visual cue with olfactory stimulation allowed

investigation of the differences between the three groups when an afferent olfactory stimulus was

present. This is important because prior studies and UPSIT scores demonstrate olfactory deficits

in AD and MCI patients; therefore the odor condition provides a stimulus that is registered

unequally by the three groups (the CN had no issues detecting the odorant while MCI and AD

subjects had some difficulty). As expected, AD and MCI subjects had significantly decreased

activated volumes in the POC and hippocampus than the CN. The condition with the visual cue

and no olfactory stimulation allowed investigation of activation patterns in the three groups when

information to all subjects was equal. We tested their ability to see the words ―Rest‖ and

―Smell?‖ on the screen and confirmed that the AD and MCI subjects had no visual impairment.

We also tested the subjects to confirm they understood the task which was to respond ―yes‖ with

their right hand if ―Smell?‖ was paired with lavender odorant and ―no’ with their left hand if

―Smell?‖ was paired with no odorant. All subjects understood the task and were able to hold the

button presses in each hand and were able to correctly respond to our test. Therefore, the visual

stimulation and motor response were both confirmed as nonissues for the MCI and AD groups.

Based on the above, we concluded that all stimulation during the no odor condition was equal. In

this condition, if the olfactory dysfunction was a peripheral dominant problem, significant group

differences should not have been present in activation volume in the hippocampus and POC.

However, group differences were observed during the no odor condition and therefore the data

suggests the involvement of the central olfactory system in AD and MCI. Specifically, for each

group significant differences between the odor condition and the no odor condition were not

observed. The no odor condition also showed significant group differences with the AD and MCI

80

groups having less activation than the CN subjects; however, a more drastic decrease in activation

volume in the MCI group was observed during odor presentation. This suggests central olfactory

dysfunction as the dominant issue in AD and MCI patients. This involvement of the central

olfactory system was also observed in medial temporal lobe resection patients who showed

similar olfactory deficits to AD and MCI subjects. While these patients showed near normal odor

detection abilities, they had trouble with discrimination and identification of odorants [29-30].

81

Figure 3-4. Olfactory fMRI paradigm with and without olfactory stimulation. ―Rest‖ or ―Smell?‖

were always displayed on the screen and air was consistently flowing to the subject’s nose.

―Rest‖ was always paired with no odor/fresh air and ―Smell?‖ was paired with either no

odor/fresh air or lavender odorant. ―Smell?‖ was the cue to provide an answer to whether the

subject could smell lavender or not. When ―Smell?‖ was paired with no odor/fresh air all

components to the subject were concluded to be the same. When ―Smell?‖ was paired with odor

stimuli to the subjects were not perceived as equal due to the olfactory deficits in the MCI and

AD groups.

82

The paradigm utilized in this dissertation is not a simple olfactory paradigm but involved

olfactory, visual and motor components. The combination of visual and olfactory stimuli

enhances the signal change in the central olfactory regions [27]. This study showed that the odor

and no odor conditions showed similar activation volume in the hippocampus and POC and both

conditions showed group differences. While it may seem from these results that the olfactory

component is unnecessary, previously in our lab, we confirmed that the olfactory component is

indeed necessary to activate the POC (Fig. 3-5). We investigated the paradigm presented in this

study (4-concentration) and a no odor paradigm (only changes were that lavender was never

introduced—the only condition was ―Smell?‖ plus no odor and responses were not collected) in

young controls. The no odor paradigm showed minimal signal percent change (Fig 3-5A and B).

Whereas the four concentration paradigm showed a very nice hemodynamic response function

(HRF) for both conditions (―Smell?‖ + odor and ―Smell?‖ + no odor). While the question of

―Smell?‖ combined with odor provided a greater number of activated voxels in this study, the

olfactory component was needed to prompt activation within the POC.

83

Figure 3-5. Hemodynamic response function (HRF). (A) and (B) show the HRF during two

olfactory fMRI paradigms in young controls: the 4-concentration paradigm (shown in red is the

same paradigm used in this dissertation) and a similar paradigm but without an olfactory stimulus

(shown in black). HRF for the 4-concentration paradigm when ―Smell?‖ is paired with an odor

(A) and the HRF when ―Smell?‖ is paired with no odor (B) are similar. The no odor paradigm

curves are the same in (A) and (B).

84

The activated volume in the POC and hippocampus correlated positively with the

cognitive and olfactory tests. The no odor condition showed significant positive correlations with

all cognitive tests and the UPSIT, further suggesting the dominant role of the central olfactory

system.

We also investigated the activation volume during presentation of the four different

concentrations to investigate whether activation volume would increase as the concentration

increased. All three groups showed stabilized activation volume during each increasing

concentration. The different concentrations did not show significant differences. This was true for

both the POC and the hippocampus. Similar activation for all concentrations for each group

suggests that our paradigm does not allow investigation of differences in concentrations or that an

overall habituation effect is occurring where even though lavender concentration is increasing the

affect size is unchanging. The former is more likely since the paradigm asked only for the subject

to detect the odorant presence and it should be noted that Sobel et al reported higher activation in

the anterior medial thalamus and the inferior frontal gyrus for higher concentrations compared to

lower concentrations of the same odorant [31]. Other regions of the brain should be analyzed to

detect activation changes to the four concentrations. The CN group had higher activation in all

four concentrations of lavender while the MCI and AD subjects showed similar activation volume

for each of the four concentrations. Interestingly, neither MCI nor AD group showed an

increasing trend with increasing lavender concentration. The AD group actually showed a

decreasing trend. In the AD group this may be due to the fact they tended to give up more quickly

because they had greater difficulty with detecting the odorant.

Several limitations in our study should be addressed. As addressed in Chapter 2, our

current study included a small sample size for patient groups and our MCI group consisted of

patients who will and will not develop AD. In addition, our current analysis was focused only on

85

two ROIs. Further analyses will include broader brain areas using functional network analysis in

order to understand the relationship between olfactory deficits and cognitive impairments.

Finally, while the results in this study demonstrated in vivo measurement of the involvement of

the central olfactory system and suggested olfactory deficits in MCI and AD are dominantly due

to central olfactory system dysfunction; they cannot conclude that the peripheral system is not

involved.

The innovative paradigm utilized in this study allows examination of the central olfactory

problem in AD and MCI. In summary, the results of our olfactory fMRI study demonstrated that

the central olfactory system dysfunction is the dominant reason for the olfactory deficit present in

AD and MCI subjects.

86

3.6 References

[1] Ferreyra-Moyano H (1989) The olfactory system and Alzheimer’s disease. Int J Neurosci

49:157-97.

[2] Knupfer L, Spiegel R (1986) Differences in olfactory test performance between normal

aged, Alzheimer and vascular type dementia individuals. Int J Geriat Psychiat 1:3-14.

[3] Murphy C, Gilmore MM, Seery CS, Salmon DP, Lasker BR (1190) Olfactory thresholds are

associated with degree of dementia in Alzheimer’s disease. Neurobiol Aging 11:465-9.

[4] Warner MD, Peabody CA, Flattery JJ, Tinklenberg JR (1986) Olfactory deficits and

Alzheimer’s disease. Bio Psychiat 21:116-8.





1988; 9:399-403.





1995; 15:793-803.

[9] Doty RL, Reyes PF, Gregor T. Presence of both odor identification and detection deficits in

Alzheimer’s disease. Brain Research Bulletin 1987; 18:597-600.

[10] Murphy C, Nordin S, Acosta L: Odor learning, recall, and recognition memory in young and

elderly adults. Neuropsychology 1997; 11:126-37.

87



[12] Pearson RCA, Esiri MM, Hiorns RW, Wilcock GK, Powell TPS. Anatomical correlates of

the distribution of the pathological changes in Alzheimer’s disease. Proc Nat Acad Sci

(USA) 1985; 82:4531-4.


Med 1988; 318:789-90.


73: 365-9.



1991; 12: 295-312.




Psychiat 1984; 47:56-60.

[18] Braak H, Braak E. Neuropathological staging of Alzheimer-related changes. Acta

Neuropathol 1991; 82: 239-59

[19] Parent A, Carpenter MB. Carpenter's human neuroanatomy. Baltimore: Williams & Wilkins,

9th Edition. 1996 Ch. 18



disease and schizophrenia. Ann N Y Acad Sci. 1998 Nov 30;855:762-75.: olfactory

impairments associated with AD are likely due to damage in the central olfactory pathways

88



32.


disease. Am J Psychiatry 1991; 148:357-60.



[24] ter Laak HJ, Renkawek K, van Workum FP. The olfactory bulb in Alzheimer disease: a

morphologic study of neuron loss, tangles, and senile plaques in relation to olfaction.

Alzheimer Dis Assoc Disord 1994; 8:38-48.

[25] Armstrong RA, Syed AB, Smith CU. Density and cross-sectional areas of axons in the

olfactory tract in control subjects and Alzheimer's disease: an image analysis study. Neurol

Sci. 2008 Feb;29(1):23-7. doi: 10.1007/s10072-008-0854-0. Epub 2008 Apr 1.




mediates olfactory deficits in Alzheimer's disease. Brain. 2010; 133:2714-26.

[28] Gottfried JA, Dolan RJ. The nose smells what the eye sees: crossmodal visual facilitation of

human olfactory perception. Neuron 2003; 39:375-86.

[29] Eichenbaum H, Morton TH, Potter H, Corkin S. Selective olfactory deficits in case H.M.

Brain 1983; 106:459-72.

[30] Eskenazi B, Cain WS, Novelly RA, Friend KB. Olfactory functioning in temporal

lobectomy patients. Neuropsychologia 1983; 21:365-74.

89

[31] Sobel N, Prabhakaran V, Hartley CA, Desmond JE, Glover GH, Sullivan EV, Gabrieli JD.

Blind smell: brain activation induced by an undetected air-borne chemical. Brain. 1999

Feb;122 ( Pt 2):209-17.

90

Chapter 4

Functional Connectivity of the Piriform is Disrupted in AD and MCI

4.1 Abstract

Background: Alzheimer’s disease (AD) and mild cognitive impairment (MCI) patients present

with olfactory deficits and these symptoms often precede cognitive and memory problems.

Olfactory dysfunction coincides with the early site of AD pathology in the central olfactory

structures. Resting-state imaging studies have shown the breakdown of networks in the brain;

therefore, we examined the olfactory network in these two patient populations during an olfactory

paradigm. We hypothesized a break down in the olfactory network in AD and MCI subjects.

Methods: Olfactory functional magnetic resonance imaging (fMRI) and olfactory and cognitive

tests were administered to 27 cognitively normal controls, 21 MCI, and 15 AD subjects (same

sample as used chapter 2 and 3). Analysis of functional connectivity of the piriform was

performed on the olfactory fMRI paradigm discussed in Chapter 2.

Results: The piriform showed decreased connectivity in the MCI and AD groups compared with

the controls (ANOVA, P < 0.001). There was decreased connectivity to the caudate, putamen,

hippocampus, nucleus accumbens, and orbitofrontal cortex for both the left and right piriform;

however, the left piriform showed greater functional connectivity disruption than the right

piriform.

Conclusions: The piriform has decreased functional connectivity in both the MCI and AD

subjects. While the difference between the MCI and AD subjects did not reach significance, a

decreasing trend was observed. Lateralization was also present in the degeneration of the

network. Our results suggest disconnection of the olfactory network in AD and MCI subjects.

91

4.2 Introduction

Alzheimer’s disease (AD), the most common form of dementia, is believed to be caused

by the disconnection of the brain networks [1]. Pathologically AD is characterized by amyloid

beta plaques, neurofibrillary tangles (NFTs), and atrophy of the medial temporal lobe. Amyloid

beta plaques and NFTs first form in the medial temporal lobe and progressively spread

throughout the temporal, frontal, and parietal lobes [2]. Such pathological changes may be

responsible for the disconnection of brain networks associated with AD [3-5]. Currently, AD can

only be definitively diagnosed by post-mortem examination; however, olfactory function tests,

including odor threshold detection, identification, and odor memory tests, are being examined as

promising early diagnostic tools [6-11]. Significant olfactory deficits have been reported in both

patients with early AD and mild cognitive impairment (MCI) when compared with cognitively

normal, age-matched controls (CN) [12-17]. Longitudinal studies have also indicated that disease

progression is significantly correlated with olfactory impairment [15-16], even after age effects

have been considered [12].

Traditional neuroimaging primarily serves for differential diagnosis. However, with

recent advances in neuroimaging methodology, techniques are being investigated for their

diagnostic ability, specifically functional magnetic resonance imaging (fMRI). Recently,

functional connectivity has emerged as a powerful method to study brain network changes in

various disease states and task conditions. Defined as the temporal correlation of blood oxygen

level dependent (BOLD) fluctuations in anatomically distinct brain regions [18], functional

connectivity allows for inference of brain networks and their temporal dynamics during a range of

mental states. Supporting the disconnection of brain networks hypothesis, resting-state fMRI

studies have shown a breakdown of the default mode network in AD and MCI patients [19-21].

The default mode network, characterized by higher activity during periods of rest; becoming

92

suppressed while engaged in task performance, is anatomically defined as including the posterior

cingulate cortex, the tempero-parietal junction, the precuneus, the medial prefrontal cortex, and

part of the medial temporal cortex [18, 22-23]. Resting state fMRI studies have also shown

decreased hippocampal functional connectivity to the whole brain in AD subjects [24] further

supporting the disruption of brain networks in AD. Limited studies of functional connectivity

during task paradigms exist. Dennis et al utilized a memory encoding task in young adults either

carrying or not carrying apolipoprotein E epsilon4 allele (ApoE e4) to show that those carrying

this allele had increased functional activations in the medial temporal lobe and functional

connectivity of the medial temporal lobe to areas implicated in memory processing/encoding

[25]. Individuals carrying the ApoE e4 are at a higher risk for developing AD [26-27]. Another

study using an olfactory recognition memory task to investigate neuronal networks in non-

demented ApoE e4 carriers and non-carriers showed differential functional connectivity between

the ApoE e4 carries (at risk for AD) and the non-carriers [28]. Both of these studies, however,

used memory tasks and did not specifically study the olfactory system.

Given the olfactory deficits frequently associated with AD and MCI patients, olfactory

fMRI studies are being performed in an attempt to determine the root of this olfactory

dysfunction. These studies, combining intranasal olfactory stimulation with simultaneous fMRI

acquisition, have found functional deficits in the central olfactory structures in patients with AD

[29-31]. In the previous study (Chapter 2), we showed similarly decreased functional activity in

the primary olfactory cortex (POC) of both AD and MCI subjects during an olfactory stimulation

paradigm. Specific to the MCI group, functional activation showed more significant decreases

than all other measurements (cognitive tests, olfactory identification tests, and volumetric

measurements). This suggests that olfactory fMRI may best capture early changes associated with

AD progression that are not yet apparent using traditional cognitive and smell tests.

93

Therefore, in this study we investigated functional connectivity of the piriform cortex

during the olfactory paradigm described in chapter 2, in order to further investigate the central

olfactory system. We hypothesized that 1) functional connectivity of the piriform will be

decreased in AD and MCI subjects compared with cognitively normal controls, 2) functional

connectivity of the piriform during the paradigm will positively correlate to cognitive and

olfactory tests, and 3) higher functional connectivity will be observed in the MCI group compared

with the AD group offering a potential explanation for the greater behavioral performance seen in

the MCI group.

4.3 Methods

4.3.1 Study Cohort

The study cohort is described in chapter 2.

4.3.2 Behavioral Tests

The behavioral tests are discussed in detail in section 2.3.2, including the University of

Pennsylvania Smell Identification Test (UPSIT) and the three clinical neurocognitive

examinations: the Mini-Mental State Examination (MMSE), Mattis Dementia Rating Scale-2

(DRS-2), California Verbal Learning Test-Second Edition Short Form (CVLT-II).

4.3.3 Olfactory Stimulation Paradigm

The olfactory paradigm and lavender odorant used were previously discussed in section

2.3.3. Figure 2-1 shows the olfactory paradigm.

94

4.3.4 Imaging Protocol

Details regarding the data acquisition are described in chapter 2.

4.3.5 Functional Connectivity Analysis

The preprocessing of fMRI images was conducted using the Data Processing Assistant

for Resting-State fMRI Advanced (DPARSFA) [32], based on Statistical Parametric Mapping

(SPM8) and Resting-State fMRI Data Analysis Toolkit (REST). Preprocessing followed a

standard data processing pipeline. After removing the first 4 images from each subject to exclude

instable signal, remaining 230 images were realigned, co-registered high resolution T1 image to

functional images then segment T1 image , normalized to the Montreal Neurological Institute

(MNI) brain template [33], smoothed with a 8x8x8 mm FWHM Gaussian smoothing kernel,

detrended, temporally band-pass filtered to 0.01–0.08 Hz, and the nuisance covariates (including

six head motion parameters, global signal, white matter signal, and CSF signal) were regressed

out.

Functional connectivity analysis was performed using a hypothesis-driven, seed-based

approach with 29 anatomically defined regions of interest (ROIs) (Table 4-1) each modeled as a 6

mm radius sphere. Neuroanalytica software (Brain Image Analysis, LLC) was utilized to segment

the hippocampus, amygdala, nucleus accumbens, caudate, putamen, and thalamus. Manual

segmentation using FMRIB Software Library View (FSLview, Analysis Group, FMRIB, Oxford,

UK) was performed for the POC on each CN subjects’ T1-weighted images. The POC included

the anterior olfactory nucleus, olfactory tubercule, piriform cortex, anterior portion of the

periamygdaloid cortex and amygdala, and anterior perforated substance. The average region from

the 27 controls was used as the center coordinate in further analysis. The hippocampus, caudate,

http://www.fmrib.ox.ac.uk/analysis

95

and putamen were divided in head and tail, and the POC was divided into three components:

piriform, anterior olfactory nucleus, and entorhinal cortex. The anterior cingulate cortex, posterior

cingulate cortex, and orbitofrontal cortex seed locations were derived from the literature of

previous olfactory studies [34-36]. All ROIs were selected based on their known role in olfactory

processing [37-39]. First-level analysis extracted the average BOLD time course from each seed

and computed the Pearson's correlation coefficients between this time course and the BOLD time

course of every other voxel. Correlation coefficients were converted to z-scores using Fisher's z-

transform to improve normality.

96

Table 4-1. Anatomically defined Regions of Interest.

Abbreviations: POC, primary olfactory cortex.

Note: Each region of interest was modeled as a 6 mm radius sphere. Coordinates indicate center

of sphere in Montreal Neurological Institute (MNI) space.

Region of Interest Left Right

X Y Z X Y Z

Piriform -25 0 -20 25 0 -20

POC Entorhinal Cortex -24 1 -30 24 1 -30

Anterior olfactory nucleus -16 11 -18 16 11 -18

Amygdala -25 -4 -21 25 -3 -22

Hippocampus Head -29 -15 -24 26 -12 -25

Hippocampus Tail -24 -38 -5 25 -36 -4

Anterior cingulate cortex -5 46 4 8 46 4

Orbitofrontal cortex -36 32 -18 36 24 -18

Thalamus -12 -18 5 14 -18 5

Nucleus accumbens -10 9 -9 9 10 -8

Caudate Head -12 15 3 13 17 -2

Caudate Tail -17 -7 -2 19 -2 20

Putamen Head -21 9 -5 23 9 -4

Putamen Tail -28 -6 3 30 -6 3

X Y Z

Posterior cingulate cortex -3 -48 27

97

4.3.6 Statistical Analysis

SPM8 was used to perform second level analysis (one-sample t-tests and analysis of

variance (ANOVA)) on functional connectivity maps generated from each subjects’ bilateral

piriform seeds. A 29x29 matrix of r correlation coefficients was generated by averaging all

subjects ROI-ROI connectivity values in each group, therefore producing 3 matrices. One-way

ANOVA was performed on the matrices from the three groups to show ROI pairs significantly

changing with disease progression. The results of the ANOVA were arranged to clearly show

inter/intra hemispheric connectivity differences between these groups. SPM8 was used to perform

multiple regression analysis in order to show correlations between each groups average

connectivity values and the cognitive tasks (CVLT-II, UPSIT, and DRS-2). r correlation

coefficients from the bilateral piriform seeds to the other 28 ROIs were averaged for each subject

to get an overall correlation coefficient of the olfactory network. These values were used in the

correlation analysis with the UPSIT, CVLT-II, and DRS (GraphPad Prism 6- GraphPad Software

San Diego, CA).

4.4 Results


Table 2-1 provides a summary of the demographic information and cognitive/behavioral

test results of the three groups. These results are discussed in chapter 2.

4.4.2 Functional Connectivity of the Piriform

98

Figure 4-1 shows the z-score functional connectivity maps to the whole brain for each of

the three groups’ left and right piriform ROIs (one-sample t-test; extent threshold = 6; CN: P <

0.001, Family Wise Error corrected (FWE); MCI and AD: P < 0.05, FWE). A stepwise decrease

was observed from CN to MCI to AD subjects in functional connectivity for the right and left

piriform ROIs. This stepwise decrease was also observed in the bilateral anterior olfactory

nucleus and the entorhinal portions of the POC (not shown). The z-score functional connectivity

maps for the left piriform were entered into a one-way ANOVA analysis (P < 0.001) to determine

areas with significantly altered functional connectivity between the three groups (Fig. 4-2A).

Significant group differences were observed in the caudate, putamen, thalamus, and anterior

cingulate cortex. Attempting to further determine when in the disease progression these changes

were occurring, paired t-tests were performed on the z-score functional connectivity maps

between the CN-AD group (Fig. 4-2B) and the CN-MCI (Fig. 4-2C) (two-sample t-test, extent

threshold = 6, P < 0.001). Significant differences between the MCI and AD groups were not

observed. ANOVA of the z-score functional connectivity maps for the right piriform did not

show significant differences between the groups.

99

Figure 4-1. Functional connectivity of the piriform. Activation maps (one sample t-tests; CN: P <

0.001 FWE, MCI and AD: P < 0.05, FWE; extent threshold = 6) are represented for each group.

The underlay is a normalized T1-weighted image in Montreal Neurological Institute space. The

left piriform (A) and the right piriform (B) show decreased functional connectivity of the piriform

to several brain regions including the hippocampus, caudate, putamen, nucleus accumbens,

orbitofrontal cortex, anterior cingulate cortex, cerebellum, and visual cortex.

100

Figure 4-2. Functional connectivity disruption. The underlay is a normalized T1-weighted image

in Montreal Neurological Institute space. A) Significant group differences seen in the anterior

cingulate cortex, putamen, caudate, and thalamus (one-way ANOVA, extent threshold = 6, P <

0.001). B) Normal controls have greater functional connectivity to the anterior cingulate cortex,

putamen, caudate, and thalamus than Alzheimer’s group (two-sample t-test, extent threshold = 6,

P < 0.001). C) Normal controls have greater functional connectivity to the anterior cingulate

cortex, putamen, and thalamus than the mild cognitive impaired subjects (two-sample t-test,

extent threshold = 6, P < 0.001).

101

4.4.3 Lateralization of Connectivity

To assess significant group differences in connectivity between each of the 29 regions

known to be involved in olfactory processing [33-35, 37-39], a one-way ANOVA was performed

using each ROI-ROI correlation coefficient. The results of this ANOVA were binarized with blue

squares indicating significantly different ROI-ROI group differences, and red indicating non-

significant differences (Fig. 4-3, ANOVA, and P < 0.05). In order to see left to right, left to left,

right to right, and right to left hemispheric connectivity differences correction was not performed.

Left to left intrahemispheric connectivity showed the greatest number of group differences with

38 left to left connections out of the possible 98 showing significant group differences. Right to

right intrahemispheric connectivity was minimally affected. Only 5 connections of the 98 total

right to right connections showed significant group differences. These included anterior olfactory

nucleus to the thalamus, caudate tail, and the anterior cingulate cortex, the tail of the

hippocampus to the caudate tail, and the anterior cingulate cortex to the thalamus. A lateralization

of the atrophy in functional connectivity can clearly be observed with the left to left

intrahemispheric connectivity showing the greatest number of changes while the right to right

intrahemispheric connectivity was minimally affected (Fig. 4-3). The matrix also shows overall

more significant left inter- and intra-hemispheric connectivity changes than right inter- and intra-

hemispheric connectivity. ROI seed-based analysis shows that regions in the left hemisphere have

decreased functional connectivity in the MCI and AD groups.

102

Figure 4-3. Olfactory network matrix. Correlation matrix for the 29 regions known to be involved

in olfactory processing (one-way ANOVA results of the three groups, P < 0.05). The results of

this ANOVA were binarized with blue squares indicating significantly different ROI-ROI group

differences, and red indicating non-significant differences. Left to left intrahemispheric

connectivity shows greater areas of significant difference than the right to right intrahemispheric

connections. More left sided regions show significant group differences in functional connectivity

both intra- and interhemispherically.

103

In order to further investigate the hemispheric differences in functional connectivity, we

examined the cognitively normal controls. Figure 4-1 indicated piriform is functionally connected

to the bilateral nucleus accumbens, caudate, putamen, amygdala, anterior cingulate cortex,

posterior cingulate cortex, orbitofrontal cortex, insula, dorsolateral prefrontal cortex,

hippocampus, thalamus, cerebellum, and visual cortex during the olfactory paradigm.

Quantification of the data by averaging the correlation values for each subject showed that in CN,

greater functional connectivity was observed when the seed was in the left hemisphere (Fig. 4-4B,

one-sample paired t-test, P = 0.005). Lateralization in functional connectivity was present in the

control group. This lateralization was not observed in the MCI or the AD subjects. The quantified

functional connectivity data also shows significant disruption of the olfactory network for the left

piriform (Fig, 4-4A, ANOVA, P = 0.006).

104

Figure 4-4. Lateralization of olfactory network (mean ± standard error). A) shows significant

group differences for functional connectivity of the left piriform (ANOVA, P = 0.006). Multiple

comparisons test reported decrease in functional connectivity for the left piriform only in both

mild cognitive impairment and Alzheimer’s subjects compared with the normal controls. B)

shows that the left piriform has greater functional connectivity to the anatomically defined

regions than the right piriform only in the control group (paired t-test, P = 0.005).

Notes: *Significant compared with CN Left Piriform

* P < 0.05

** P < 0.01

105

4.4.4 Correlation of Functional Connectivity to the University of Pennsylvania Smell

Identification Test and Cognitive Tests

Significant correlation between the UPSIT and functional connectivity of the left piriform

(Fig. 4-5A left) was observed in the bilateral caudate, putamen, insula, thalamus, anterior

cingulate cortex, posterior cingulate cortex, cerebellum, and visual cortex (multiple regression

analysis, age used as covariate, P < 0.001). Correlation between the UPSIT and functional

connectivity of the right piriform (Fig. 4-5B left) was observed in the bilateral caudate only

(multiple regression analysis, age used as covariate, P < 0.001). The CVLT-II and DRS-2 showed

significantly positive correlation with the functional connectivity of the left piriform only

(multiple regression analysis, P < 0.001). Left and right connectivity results and the UPSIT

showed significant age affects; therefore, prior to correlation studies of the quantified data, the

data were age corrected. The quantified functional connectivity of the left piriform (Fig. 4-5A

right) shows a more significant correlation with the UPSIT (P < 0.0001) than the right piriform

(Fig. 4-5B right, P = 0.0063). The DRS-2 and CVLT-II positively correlated with only the z-

score functional connectivity map of the left piriform (P < 0.005). The patient group (MCI and

AD together) was also examined independently from the controls and functional connectivity of

the left piriform showed significant positive correlation to the UPSIT (multiple regression

analysis, age used as covariate, P < 0.001). Lastly, the AD group alone also showed positive

correlation between the UPSIT and the functional connectivity of the left piriform (P < 0.05).

MCI and CN groups alone did not show significant correlations with the UPSIT.

106

Figure 4-5. Correlations between smell and functional connectivity. The underlay is a normalized

T1-weighted image in Montreal Neurological Institute space. A and B left side show SPM8

generated maps (multiple regression analysis, P < 0.001) and for all subjects for the left (A) and

right (B) piriform. The quantified data (right side) shows more significant correlation for the left

piriform (A) than the right piriform (B).

107

4.5 Discussion

To our knowledge, this is the first study utilizing an olfactory paradigm to investigate the

functional connectivity of the piriform cortex in AD and MCI patients. Overall, the results

showed decreased functional connectivity of the piriform cortex in both patient groups compared

with age-matched normal controls. This disruption was lateralized to the left side. The data also

revealed correlation between behavioral testing and the functional connectivity of the piriform

cortex.

In the cognitively normal controls, high connectivity was observed between the primary

and secondary olfactory regions including the amygdala, hippocampus, orbitofrontal cortex,

anterior cingulate cortex, caudate, nucleus accumbens, and thalamus. These results are consistent

with resting-state studies showing high functional connectivity between the olfactory and limbic

networks [20, 40]. These limbic areas, as part of the reward pathway, may be responding to the

pleasant, rewarding nature of the lavender odorant. Behavioral and anatomical ties between

olfaction and the limbic system exist. Olfactory dysfunction affects 19% of the population and

profoundly effects quality of life and enjoyment in these patients [41]. Previous studies show that

subjects with olfactory deficits report greater problems with social and family life, employment,

and housework [42-43]. This finding is supported by a study that showed a correlation between

anosmia and loss of gray matter in the nucleus accumbens, medial prefrontal cortex including the

middle and anterior cingulate cortices, and the dorsolateral prefrontal cortex [44]. Most of these

areas mentioned above are a part of the limbic loop and are considered with the exception of the

DLPFC and nucleus accumbens to be secondary olfactory areas. There is also high functional

connectivity between the olfactory and limbic networks, as shown by several resting-state

imaging studies [21, 40]. Levy et al also found activation within the cingulate cortex and in

several areas of the limbic system when using an olfactory stimulus [45]. Other studies showed

108

increased activation in the amygdala and left orbitofrontal cortex to an aversive odorant [46]

while pleasant odorants localized to the right OFC and piriform cortex [47-48]. Royet and group

showed that while emotionally valenced visual and olfactory stimuli increased regional cerebral

blood flow in the OFC, the temporal pole, and superior frontal gyrus in the left hemisphere, only

olfactory stimuli increased bilateral rCBF in the amygdala [49]. Our findings in the control

subjects coincide with the above studies showing a clear relationship between the limbic and

olfactory areas. Functional connectivity to the cerebellum, visual cortex, and motor regions was

also observed in the normal controls. Connectivity with the cerebellum during olfactory

paradigms has been reported in several studies [50-52] and it is suggested this may have to do

with the role of the cerebellum in a feedback mechanism to regulate sniff volume [50]. The visual

cortex and motor cortex connectivity observed in the controls is likely due to the visual and motor

component of the olfactory paradigm. The CN group had an easier time with the olfactory

paradigm task and this can be explained by increased ability to process all three components of

the paradigm.

The robust olfactory network seen in the control group was disrupted in both the MCI

and AD subjects in a step-wise trend when the seed was in the left piriform. The MCI subjects

showed greater connectivity than the AD subjects; however, this difference did not reach

significance. This is in contrast to our general linear model results from the previous study which

showed activation in MCI subjects was similar to AD subjects. The trend towards greater

functional connectivity in the MCI group may explain their higher level of behavioral functioning

on the cognitive and olfactory tests compared with the AD group.

In fact, higher UPSIT scores correlated to greater functional connectivity of the bilateral

piriform; however, only the left piriform correlated significantly to the DRS-2 and CVLT-II. The

patient population, when examined without the scores from the controls, showed that higher

109

functional connectivity of the piriform predicted higher UPSIT scores. These correlations provide

a potential explanation for the higher performance by the MCI compared with the AD group on

the cognitive and smell identification tests.

Cognitive and olfactory deficits are not the only symptoms present in AD patients; AD

also includes behavioral and psychiatric changes [53-60]. AD and MCI patients tend to have a

higher incidence of depression, specifically increased levels of apathy and anhedonia. These

conditions, under-treated in AD, exacerbate the symptoms and decrease quality of life for the

patient and caregiver [60]. Our results may provide a potential explanation for the high

comorbidity of apathy and anhedonia found in these patients. The striatal (caudate, putamen, and

nucleus accumbens) areas, the amgydala, which are involved in the processing of emotion and

reward are also involved in the processing of higher order olfaction [37-39]. The nucleus

accumbens specifically has been suggested in olfactory reward processing [61]. It is suggested,

because the olfactory system is the only sense that bypasses the thalamus, that emotional

evaluation occurs prior to and without cognitive processing of olfactory information [62]. In an

anatomically based study, Bitter et al reported significant gray matter decrease in the nucleus

accumbens in patients with anosmia compared to non-anosmic healthy controls [44]. While

connectivity differences amongst the groups were not present between the piriform and the other

primary olfactory regions (amygdala and other POC regions), group differences were present

between the piriform and the higher order olfactory regions (thalamus, striatum, and

hippocampus). Therefore, our results show a disconnection of the piriform to the areas involved

in processing emotion and reward in AD and MCI subjects. This suggests symptoms of apathy

and anhedonia may be, at least partially, due to this disconnection of the limbic system from the

piriform which is consistent with other studies that have shown an association between

depression and olfaction dysfunction [63-73]. In fact, several studies have reported olfactory

110

deficits in individuals diagnosed with depression [67-73]. Olfactory bulb volumes (both right and

left) are decreased in depressed patients relative to age-matched controls [74]. Anti-depressants

resolve this volume loss of the olfactory bulb in depressed individuals as well as reverse

hippocampal loss, which is also seen in depressed patients. Previous studies have shown that

lower olfactory performance is associated with poorer quality of life [75] and with increased

symptoms of depression and apathy [76-77]. fMRI studies have found that depressed patients also

have altered activation of the OFC and the amygdala, which may explain the dysfunction in

processing olfactory properties.

Asymmetry in the olfactory system has been observed; however, the question of

lateralization still remains unanswered [78]. In this study lateralization of the functional

connectivity of the piriform was observed. The functional connectivity matrix of the 29 regions

showed that more left sided regions had significant group differences in functional connectivity,

both intra- and interhemispherically, than the right-sided regions suggesting lateralization in

atrophy of functional connectivity of the olfactory network. The control group showed significant

differences between the left and right hemisphere when the seed was placed in the piriform

suggesting lateralization exists in the olfactory network. The left piriform showed greater

connectivity in the CN than the right piriform. Functional connectivity differences amongst the

three groups were also more significant when the seed was in the left piriform. This suggests that

the olfactory network is lateralized and shows greater degeneration of the olfactory network in the

left hemisphere in AD and MCI subjects during a simple olfactory task.

Several limitations in our study should be addressed. As discussed in chapter 2, the

results in this dissertation are limited by the small sample size of the patient groups. The AD

group consisted of both early onset and late onset patients. Differences in the antero-medial

temporal network and dorsolateral prefrontal cortex network between early onset and late onset

111

AD have been previously reported [79]. Subjects from the MCI group will also not necessarily

convert to AD, limiting our ability to make inferences from their data as a preclinical AD

population. In this analysis, functional connectivity was calculated based on correlation

throughout the entire paradigm. We were not able to focus just on the periods when odor was

presented because each presentation lasted only 6 s (total odor presentation time of 72 s). A

longer period of presentation; however, would be influenced by habituation effects. An analysis

to compare the ―Smell?‖ paired with odor and ―Smell?‖ paired with no odor should be performed

in the future. This may provide differences between the conditions that the general linear model

analysis performed in chapter 3 was not able to show. Future functional connectivity studies

should be completed with a simpler pure olfactory paradigm. In order to validate our findings

future investigations should include longitudinal studies using the gold standard of post mortem

findings to confirm the diagnosis with larger MCI and AD cohorts.

In summary, the results showed that the functional connectivity of the piriform was

decreased in both MCI and AD subjects compared with CN specifically to second order olfactory

processing regions. The decrease was in a stepwise trend from CN to MCI to AD, which provides

a potential explanation for the greater performance on the behavioral tests by the MCI subjects

compared with the AD subjects. This trend is more significant for the left piriform, therefore

displaying lateralization of atrophy of the olfactory network.

112

4.6 References

[1] Delbeuck X, Van der Linden M, Collette F. Alzheimer’s disease as a disconnection

syndrome? Neuropsychol Rev 2003; 13:79–92 [PMID: 12887040].


Neuropathol 1991; 82: 239-59

[3] Sheline YI, Raichle ME, Snyder AZ, Morris JC, Head D, Wang S, Mintun MA. Amyloid

plaques disrupt resting state default mode network connectivity in cognitively normal

elderly. Biol Psychiatry, 2010. 67(6): 584-7.

[4] Adriaanse SM, Sanz-Arigita EJ, Binnewijzend MA, Ossenkoppele R, Tolboom N, van

Assema DM, et al. Amyloid and its association with default network integrity in

Alzheimer's disease. Hum Brain Mapp. 2014 Mar;35(3):779-91. doi: 10.1002/hbm.22213.

Epub 2012 Dec 14.

[5] Buckner RL, Snyder AZ, Shannon BJ, LaRossa G, Sachs R, Fotenos AF, et al., Molecular,

structural, and functional characterization of Alzheimer's disease: evidence for a

relationship between default activity, amyloid, and memory. J Neurosci, 2005. 25(34): p.

7709-17.

[6] Warner MD, Peabody CA, Flattery JJ, Tinklenberg JR. Olfactory deficits and Alzheimer’s

disease. Bio Psychiat 1986; 21:116-8.




tests as possible probes for detecting and monitoring Alzheimer’s disease. Neurobiol

Aging 1988; 9:399-403.

113





1995; 15:793-803.

[11] Doty RL, Reyes PF, Gregor T. Presence of both odor identification and detection deficits

in Alzheimer’s disease. Brain Research Bulletin 1987; 18:597-600.

[12] Murphy C, Nordin S, Acosta L: Odor learning, recall, and recognition memory in young

and elderly adults. Neuropsychology 1997; 11:126-37.




[14] Ferreyra-Moyano H. The olfactory system and Alzheimer’s disease. Int J Neurosci 1989;

49:157-97.

[15] Knupfer L, Spiegel R. Differences in olfactory test performance between normal aged,

Alzheimer and vascular type dementia individuals. Int J Geriat Psychiat 1986; 1:3-14.


associated with degree of dementia in Alzheimer’s disease. Neurobiol Aging 1990; 11:465-

9.




magnetic resonance imaging. Nature reviews Neuroscience2007;8(9):700–11.

[19] Greicius MD, Srivastava G, Reiss AL, Menon V. Default-mode network activity

distinguishes Alzheimer’s disease from healthy aging: evidence from functional MRI.

114

Proceedings of the National Academy of Sciences of the United States of America

2004;101(13):4637–42.

[20] Wang K, Liang M, Wang L, Tian L, Zhang X, Li K, et al., Altered functional connectivity

in early Alzheimer's disease: a resting-state fMRI study. Hum Brain Mapp, 2007. 28(10):

967-78.



Human Brain Mapping 2005; 26:231-239.

[22] Raichle ME, Snyder AZ. A default mode of brain function: a brief history of anevolving

idea. NeuroImage 2007;37(4):1083–90.

[23] Buckner RL, Andrews-Hanna JR, Schacter DL. The brain’s default network:anatomy,

function, and relevance to disease. Annals of the New York Academyof Sciences

2008;1124:1–38.

[24] Allen G, Barnard H, McColl R, Hester AL, Fields JA, Weiner MF, et al., Reduced

hippocampal functional connectivity in Alzheimer disease. Arch Neurol, 2007.

64(10):1482-7.

[25] Dennis NA, Browndyke JN, Stokes J, Need A, Burke JR, Welsh-Bohmer KA, Cabeza R.

Temporal lobe functional activity and connectivity in young adult APOE varepsilon4

carriers. Alzheimers Dement. 2010 Jul;6(4):303-11. doi: 10.1016/j.jalz.2009.07.003. Epub

2009 Sep 9.

[26] Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, Roses

AD, Haines JL, Pericak-Vance MA. Gene dose of apolipoprotein E type 4 allele and the

risk of Alzheimer’s disease in late onset families. Science. 1993; 261:921–923. [PubMed:

8346443]

115

[27] Saunders AM, Schmader K, Breitner JC, Benson MD, Brown WT, Goldfarb L.

Apolipoprotein E epsilon 4 allele distributions in late-onset Alzheimer’s disease and in

other amyloid-forming diseases. Lancet. 1993; 342:710–711. [PubMed: 8103823]

[28] Haase L, Wang M, Green E, Murphy C. Functional connectivity during recognition

memory in individuals genetically at risk for Alzheimer's disease. Hum Brain Mapp. 2013

Mar;34(3):530-42. doi: 10.1002/hbm.21451. Epub 2011 Nov 18.

[29] Wang J, Eslinger PJ, Doty RL, Zimmerman EK, Grunfeld R, Sun X, et al. Olfactory

deficits detected by fMRI in early Alzheimer’s disease. Brain Research 2010; 1357:184-94.



[31] Howard JD, Plailly J, Grueschow M, Haynes JD, Gottfried JA. Odor quality coding and

categorization in human posterior piriform cortex. Nat Neurosci. 2009 Jul;12(7):932-8. doi:

10.1038/nn.2324. Epub 2009 May 31.

[32] Chao-Gan Y, Yu-Feng Z. DPARSF: A MATLAB Toolbox for "Pipeline" Data Analysis of

Resting-State fMRI. Front Syst Neurosci. 2010 May 14;4:13. doi:

10.3389/fnsys.2010.00013. eCollection 2010.

[33] Collins DL, Zijdenbos AP, Kollokian V, Sled JG, Kabani NJ, Holmes CJ, et al. Design and

construction of a realistic digital brain phantom. IEEE Trans Med Imaging 1998; 17:463-8.

[34] Coricelli G, Nagel R. Neural correlates of depth of strategic reasoning in medial prefrontal

cortex. Proc Natl Acad Sci U S A. 2009 Jun 9;106(23):9163-8. doi:

10.1073/pnas.0807721106. Epub 2009 May 22.

[35] Premkumar P. Are you being rejected or excluded? Insights from neuroimaging studies

using different rejection paradigms. Clin Psychopharmacol Neurosci. 2012 Dec;10(3):144-

54. doi: 10.9758/cpn.2012.10.3.144. Epub 2012 Dec 20.

116

[36] Benjamin C, Lieberman DA, Chang M, Ofen N, Whitfield-Gabrieli S, Gabrieli JD, et al.

The influence of rest period instructions on the default mode network. Front Hum Neurosci.

2010 Dec 1;4:218. doi: 10.3389/fnhum.2010.00218. eCollection 2010.

[37] Westermann B, Wattendorf E, Schwerdtfeger U, Husner A, Fuhr P, Gratzl O, et al.

Functional imaging of the cerebral olfactory system in patients with Parkinson's disease. J

Neurol Neurosurg Psychiatry. 2008 Jan;79(1):19-24. Epub 2007 May 22.

[38] Murphy C, Cerf-Ducastel B, Calhoun-Haney R, Gilbert PE, Ferdon S. ERP, fMRI and

functional connectivity studies of brain response to odor in normal aging and Alzheimer's

disease. Chem Senses. 2005 Jan;30 Suppl 1:i170-1.

[39] Hummel T, Fliessbach K, Abele M, Okulla T, Reden J, Reichmann H, et al. Olfactory

FMRI in patients with Parkinson's disease. Front Integr Neurosci. 2010 Oct 28;4:125. doi:

10.3389/fnint.2010.00125. eCollection 2010.

[40] Roy AK, Shehzad Z, Margulies DS, Ckare Kelly AM, Uddin LQ, Gotimer K, et al.

Functional connectivity of the human amygdala using resting state fMRI. Neuroimage

2009; 45:614-626.

[41] Nordin S, Bramerson A. Complaints of olfactory disorders: epidemiology, assessment and

clinical implications. Curr Opin Allergy Clin Immunol. 2008; 8:10–15.

[42] Bramerson A, Nordin S, Bende M. Clinical experience with patients with olfactory

complaints, and their quality of life. Acta Oto-Laryngol. 2007; 127:167–174.

[43] Miwa T, Furukawa M, Tsukatani T, Costanzo RM, DiNardo LJ, Reiter ER. Impact of

olfactory impairment on quality of life and disability. Arch Otolaryngol Head Neck Surg

2001; 127:497–503

[44] Bitter T, Gudziol H, Burmeister HP, Mentzel HJ, Guntinas-Lichius O, Gaser C. Anosmia

leads to a loss of gray matter in cortical brain areas. Chem Senses. 2010 Jun;35(5):407-15.

doi: 10.1093/chemse/bjq028. Epub 2010 Mar 15.

117

[45] Levy LM, Henkin RI, Hutter A, Lin CS, Martins D, Schellinger DJ. Functional MRI of

human olfaction. Comput Assist Tomogr. 1997; 21:849-56.

[46] Zald DH, Pardo JV. Emotion, olfaction, and the human amygdala: amygdala activation

during aversive olfactory stimulation. Proc Natl Acad Sci USA 1997; 94:4119-24.

[47] Zatorre RJ, Jones-Gotman M, Evans AC, Meyer E. Functional localization and

lateralization of human olfactory cortex. Nature 1992; 360:339-40.

[48] Rolls ET, Kringelbach ML, de Araujo IE. Different representations of pleasant and

unpleasant odours in the human brain. Eur. J. Neurosci 2003; 18:695–703.

[49] Royet JP, Zald D, Versace R, Costes N, Lavenne F, Koenig O, Gervais R. Emotional

Responses to Pleasant and Unpleasant Olfactory, Visual, and Auditory Stimuli: a Positron

Emission Tomography Study. J Neurosci 2000; 20:7752-9.

[50] Sobel N, Prabhakaran V, Hartley CA, Desmond JE, Zhao Z, Glover GH, et al. Odorant-

induced and sniff-induced activation in the cerebellum of the human. J Neurosci. 1998 Nov

1;18(21):8990-9001.

[51] Savic I. Processing of odorous signals in humans. Brain Res Bull. 2001 Feb;54(3):307-12.

[52] Small DM, Jones-Gotman M, Zatorre RJ, Petrides M, Evans AC. Flavor processing: more

than the sum of its parts. Neuroreport. 1997 Dec 22;8(18):3913-7.

[53] Mitchell R, Herrmann N, Lanctot KL. The Role of Dopamine in Symptoms and Treatment

of Apathy in Alzheimer’s Disease. CNS Neuroscience & Therapeutics 2010; 1–17.

[54] Benoit M, Koulibaly PM, Migneco O, Darcourt J, Pringuey DJ, Robert PH. Brain perfusion

in Alzheimer’s disease with and without apathy: A SPECT study with statistical parametric

mapping analysis. Psychiatry Res 2002; 114:103–111.

[55] Benoit M, Clairet S, Koulibaly PM, Darcourt J, Robert PH. Brain perfusion correlates of

the apathy inventory dimensions of Alzheimer’s disease. Int J Geriatr Psychiatry 2004;

19:864–869.

118

[56] Lopez OL, Zivkovic S, Smith G, Becker JT, Meltzer CC, DeKosky ST. Psychiatric

symptoms associated with cortical-subcortical dysfunction in Alzheimer’s disease. J

Neuropsychiatry Clin Neurosci 2001; 13:56–60.

[57] Robert PH, Darcourt G, Koulibaly MP, et al. Lack of initiative and interest in Alzheimer’s

disease: A single photon emission computed tomography study. Eur J Neurol 2006;

13:729–735.

[58] Porta-Etessam J, Tobaruela-González JL, Rabes-Berendes C. Depression in Patients with

Moderate Alzheimer Disease: A Prospective Observational Cohort Study.Alzheimer Dis

Assoc Disord. 2011 [Epub ahead of print].

[59] Mega MS, Cummings JL, Fiorello T, Gornbein J. The spectrum of behavioral changes in

Alzheimer’s disease. Neurology 1996; 46:130–135.

[60] Benoit M, Abdrieu S, Lechowski L, Gillette-Guyonnet S, Robort PH, Vellas B. Apathy and

depression in AD are associated with functional deficit and psychotropic prescription. Int J

Geriatr Psychiatry 2008;23:409–414.

[61] Gottfried JA, O’Doherty J, Dolan RJ. Appetitive and aversive olfactory learning in humans

studied using event-related functional magnetic resonance imaging. J Neurosci. 2002;

22:10829–10837.

[62] Cleland TA, Linster C. Central olfactory structures. In: Doty R, editor. Handbook of

olfaction and gustation. NewYork:MarcelDekker. 2003; 165–181.

[63] Haber, S. N., Kunishio, K., Mizobuchi, M. & Lynd-Balta, E. The orbital and medial

prefrontal circuit through the primate basal ganglia. J. Neurosci 1995; 15:4851–4867.

[64] Grapsa E, Samouilidou E, Pandelias K, Pipili C, Papaioannou N, Mpakirzi T, et al.

Correlation of depressive symptoms and olfactory dysfunction in patients on hemodialysis.

Hippokratia 2010; 3:189-192.

119

[65] Norrholm SD, Ouimet CC. Altered dendritic spine density in animal models of depression

and in response to antidepressant treatment. Synapse 2001; 42:151-63.

[66] Song C, Leonard BE. The olfactory bulbectomised rat as a model of depression. Neurosci

Biobehav Rev 29: 2005; 627-647.

[67] Strous RD, Shoenfeld Y. To smell the immune system: olfaction, autoimmunity and brain

involvement. Autoimmun Rev 2006; 6:54-60.

[68] Clepse M, Gossler A, Reich K, Kornhuber J, Thuerauf N. The relation between depression,

anhedonia and olfactory hedonic estimates—A pilot study in major depression.

Neuroscience Letters 2010; 471:139–143.

[69] Amsterdam JD, Settle RG, Doty RL, Abelman E, Winokur A. Taste and smell perception in

depression. Biol Psychiatry. 1987 Dec;22(12):1481-5.

[70] Gross-Isseroff R, Luca-Haimovici K, Sasson Y, Kindler S, Kotler M, Zohar J. Olfactory

sensitivity in major depressive disorder and obsessive compulsive disorder. Biol

Psychiatry. 1994 May 15;35(10):798-802.

[71] Lombion-Pouthier S, Vandel P, Nezelof S, Haffen E, Millot JL. Odor perception in patients

with mood disorders. J Affect Disord. 2006 Feb;90(2-3):187-91. Epub 2005 Dec 27.

[72] Pause BM, Miranda A, Göder R, Aldenhoff JB, Ferstl R. Reduced olfactory performance in

patients with major depression. J Psychiatr Res. 2001 Sep-Oct;35(5):271-7.

[73] Thomas HJ, Fries W, Distel H. Evaluation of olfactory stimuli by depressed patients.

Nervenarzt. 2002 Jan;73(1):71-7.

[74] Negoias S, Croy I, Gerber J, Puschmann S, Petrowski K, Joraschky P, Hummel T. Reduced

olfactory bulb volume and olfactory sensitivity in patients with acute major depression.

Neuroscience 2010; 169:415–421.

[75] Hummel T, Nordin S. Olfactory disorders and their consequences for quality of life—a

review. Acta Otolaryngol 2005; 125:116–121.

120

[76] Deems DA, Doty RL, Settle RG, Moore-Gillon V, Shaman P, Mester AF et al. Smell and

taste disorders: a study of 750 patients from the University of Pennsylvania Smell and

Taste Center. Arch Otorhinolaryngol Head Neck Surg 1991; 117:519–528.

[77] Seo HS, Jeon KJ, Hummel T, Min BC. Influences of olfactory impairment on depression,

cognitive performance, and quality of life in Korean elderly. Eur Arch Otorhinolaryngol

2009; 266:1739–1745.

[78] Brand G, Millot JL, Henquell D. Complexity of olfactory lateralization processes revealed

by functional imaging: a review. Neurosci Biobehav Rev. 2001 Mar;25(2):159-66.

[79] Gour N, Felician O, Didic M, Koric L, Gueriot C, Chanoine V, et al. Functional

connectivity changes differ in early and late-onset alzheimer's disease. 67. Hum Brain

Mapp. 2013 Oct 5. doi: 10.1002/hbm.22379. [Epub ahead of print]

121

Chapter 5

Conclusion

Alzheimer’s disease (AD) is known to the public as the disease of memory impairment

and throughout its history, diagnosis has focused on changes in cognition and memory

impairment. The pathology of Alzheimer’s disease originates in the medial temporal lobe and

progresses outward toward the entorhinal cortex before encompassing the entire brain [1-4]. At

the time of cognitive impairment, amyloid plaques and neurofibrillary tangles have already

developed in the neocortex and the disease is irreversible [1]. Drugs at this stage offer little to no

ameliorative effect. Therefore, the research in this dissertation utilized the early involvement of

the olfactory system in AD and neuroimaging techniques in order to test whether olfactory testing

has great potential as a diagnostic marker in preclinical AD and mild cognitive impairment

(MCI).

5.1 Olfactory System in Alzheimer’s Disease

In the past 35 years, olfaction has become a focus in AD research due to its potential in

early diagnosis and ability to monitor the progression of the disease. The olfactory areas,

primarily located in the medial temporal lobe, are the first regions affected by AD pathology as

confirmed by post-mortem studies [5-11]. Early AD patients and even mild cognitively impaired

(MCI) patients, individuals at risk for developing AD, show significant olfactory deficits when

compared with normal age-matched controls [12-20]. MCI subjects are at the highest risk for

developing AD and are considered to be the intermediate stage between normal cognitive

function and AD [17]. More recently it has been suggested that olfactory dysfunction is the

earliest symptom present in AD and MCI patients, further solidifying olfactory dysfunction as a

detection marker for AD and MCI [19]. Smell identification tests have confirmed olfactory

122

dysfunction in both AD and MCI patients; however, MCI subjects still display a wide range in

performance on these tests, making diagnosis at the MCI stage difficult. Behavioral changes are

generally preceded by changes in the brain at the anatomical and functional level. Therefore,

neuroimaging of the olfactory system in AD and MCI patients provides an opportunity to study

the earliest behavioral symptoms and first sites of pathological change in the disease.

MRI is an immensely powerful tool which allows for in vivo measurements of the brain.

Currently, different applications of MRI allow for measurement of the anatomy, volume, signal

change, fiber tracking, and perfusion changes. MRI is not only helpful in the early diagnosis of

AD; it also has the power to follow the progression of the disease noninvasively. In this set of

studies, volumetric and functional MRI (fMRI) was utilized in order to study the primary

olfactory cortex (POC) in AD and MCI subjects as a potential diagnostic marker of preclinical

AD.

5.2 Olfactory fMRI Paradigm

The olfactory fMRI paradigm used in this dissertation is not a simple olfactory paradigm.

It included several components such as olfactory, visual, and motor. The olfactory component

was presentation of four concentrations of lavender odorant. The four concentrations presented in

successive odor were used based on previous olfactory studies within the NMR lab [21-22]. The

increasing concentrations allowed the offset of habituation effects. Visual presentation of ―Rest‖

and ―Smell?‖ was used to enhance the activation signal change. The olfactory regions are on the

medial ventral surface of the brain and can be difficult to study; therefore, combing visual and

olfactory allowed a consistent signal change. The motor component was added in to confirm that

the subject was paying attention and understood the task. The task was to provide a response, yes

or no using button presses in each hand, when the word ―Smell?‖ appeared on the screen. In

123

Chapter 3 we showed that ―Smell?‖ paired with odor and ―Smell?‖ paired with no odor provided

similar activation volumes in the POC and in the hippocampus for all three groups. Activation

during the no odor condition may mean that the olfactory component is unnecessary and that this

paradigm is testing another process. In a previous study; however, we were able to confirm the

olfactory component is needed. Figure 3-5 shows the results from this study where this 4-

concentration paradigm and another similar paradigm without introduction of odor were tested on

young controls. When no olfactory cue was used, the average hemodynamic response (HRF) was

small for the ―Smell?‖ paired with no odor condition; however, in the 4-concentration paradigm

which did have olfactory stimulation, the ―Smell?‖ paired with no odor showed a nice HRF which

was similar to the ―Smell?‖ paired with odor condition. Therefore, an olfactory stimulus is needed

and this paradigm does indeed involve olfactory function. This paradigm also does involve higher

processing such as attention. The subject needs to pay attention to the screen and report if they

detect an odorant. While the anterior cingulate, which is involved in attention, was not analyzed

consistent activation was observed in the cognitively normal controls. Overall, this novel

paradigm allowed for consistent activation of the primary olfactory cortex, and therefore the

ability to investigate the olfactory system in AD and MCI subjects. In the future a simpler

olfactory paradigm should be tried for the functional connectivity analysis.

5.3 Central Olfactory System Dysfunction Causes Olfactory Symptoms

We established that central olfactory system dysfunction is the dominant cause of the

olfactory deficits observed in AD and MCI subjects using fMRI. The olfactory system is divided

into the peripheral and central olfactory systems. While the peripheral system is important for

detecting an odor, the central system is important in higher processing of the odor. Whether the

olfactory deficits in Alzheimer’s are due to peripherally dominant impairment or centrally

124

dominant impairment has thus far remained unclear. Many studies based on behavioral and

pathological observations have agreed upon centrally dominant impairment [11, 23-25].

However, postmortem and fMRI studies provide inconclusive support for the central dominance

theory [9, 26-27]. Therefore, in our study we used an olfactory paradigm which involved

presentation of a visual cue that was accompanied with either an odor or with fresh air. The visual

cue without the olfactory stimulus allowed analysis of the POC when afferent stimulus to the

subjects was perceived as equal. Activation in the POC was expected to have been equal amongst

the three groups; however, the normal controls had greater activation than both the MCI and AD

subjects. In a previous study, as stated above, we also confirmed that olfactory input was indeed

needed to prompt activation within the POC utilizing a paradigm with just ―Smell?‖ paired with

no odor. Although the data cannot fully exclude the peripheral system, our results suggest that

central damage is the dominant cause of olfactory dysfunction in AD and MCI patients.

5.3 Volumetric Measurements

Volumetric measurements of the brain also supported the role of the central olfactory

system in causing olfactory deficits. The bilateral POC was decreased in size in both MCI and

AD groups when compared with the CN, suggesting atrophy. These data also correlated with the

findings of decreased bilateral hippocampal volume in AD and MCI patients, indicating the POC

is degenerating in parallel with the hippocampus which is the gold standard for AD. These

findings provide evidence of atrophy in a central olfactory region for the first time. Decreased

POC volume also correlated with lower cognitive and UPSIT scores, providing an anatomical

basis for the observed olfactory dysfunction. Nevertheless, it should be noted that there was a

wide range in volumetric change for the MCI group, as was also seen in their UPSIT and

cognitive test scores.

125

5.4 Olfactory fMRI

Olfactory fMRI was utilized to study the functional changes occurring in the POC. MCI

and AD subjects showed greater than 50% decrease in activation compared with the CN group in

both the bilateral POC and hippocampus during odor presentation. Activation changes were not

due simply to the volume decreases of the two regions. The MCI and AD subjects had near equal

activation in both the bilateral POC and hippocampus, indicating activation changes in MCI

patients occur before formal diagnosis of AD. Although MCI subjects showed similar signal

change in the POC and hippocampus to AD subjects, they performed better on cognitive and

smell tests than did AD patients. One possibility for this finding may be that they retain

functional connectivity of different regions in the brain as opposed to those with AD who have

connectivity impairments. Functional connectivity is defined as the correlation of interregional

neural interactions during particular tasks or from spontaneous activity during rest [28].

Currently, resting–state fMRI is being extensively studied and data shows decreased

default mode connectivity in AD subjects [29-31]. Therefore, we chose to study the olfactory

network further by utilizing functional connectivity analysis of the POC, specifically of the

piriform during the olfactory paradigm. MCI and AD subjects showed disconnection of the

olfactory network particularly to the second order olfactory regions such as the striatum. The

second order olfactory regions are involved in emotion and reward processing, suggesting a

relationship between olfactory disconnection with the limbic system and the high comorbidity of

apathy and anhedonia in AD patients. The AD group trended toward greater disconnection of the

olfactory network compared with MCI subjects, although the difference did not achieve statistical

significance. This preservation of connectivity in MCI subjects may explain their higher

performance on behavioral tests. In fact, a strong positive correlation between UPSIT scores and

functional connectivity of the piriform was observed in this patient group (multiple regression

126

analysis, age used as covariate, P < 0.001). Overall, the atrophy of the olfactory network was

more significant when the seed was in the left piriform, indicating atrophy of connectivity is

lateralized. It was also observed that greater connectivity existed in the normal controls when the

seed was in the left piriform compared with when the seed was in the right piriform. These results

indicate lateralization of the olfactory network during a simple olfactory paradigm.

5.6 Future Studies

Future studies should focus on longitudinal analysis of the MCI subjects. Follow-up

studies will closely monitor disease progression. The MCI subjects can then be separated into

converters and non-converters and group analysis at the general linear model, functional

connectivity, volumetric, and behavioral level can be applied. This analysis will inform us of

which changes in the baseline visit can predict convertors. As more data is collected the AD

subjects should be separated into early onset and late onset AD groups as differences in the

antero-medial temporal network and dorsolateral prefrontal cortex network between these two

groups has been previously reported [32].

Along with fMRI, diffusion tensor imaging (DTI) was also acquired. DTI data allows

analyses of the white matter connections by capturing the microstructural architecture of tissue by

measuring the diffusion of water. White matter tracts in almost the entire brain are affected in AD

[33]. MCI subjects have white matter disruption in the commissural and limbic tracts which are

involved in olfactory function [33]. These white matter abnormalities correlate with degeneration

of cognitive function. Preliminary analysis of this data has been completed; however, further

analysis will allow investigation of the olfactory anatomical network in AD and MCI subjects.

As a part of the data collection blood samples have also been obtained. These samples

should be analyzed for genetic markers such as ApoE e4. Not only has the ApoE e4 been

determined to increase risk of the disease, it has been deemed the most prevalent genetic risk

127

factor of the disease. Recent MRI and cognitive studies have also shown that carriers have an

accelerated age-related decrease in both local and regional interconnectivity, as well as increased

loss of mean cortical thickness, and that they were shown to have significant negative correlations

of age and episodic memory performance [34]. Interestingly, the ApoE e4 gene in general has

also been found to be expressed in olfactory brain regions and evidence has supported an overall

olfaction-cognition-ApoE e4 relationship [35]. It has also been found to be strongly pervasive in

the olfactory epithelium of AD patients and has also been clearly associated with olfactory

deficits [36]. Furthermore, ApoE e4 carriers have been shown to have impaired odor

identification and inferior odor threshold sensitivity [36]. Taken together, these research findings

provide an excellent rationale for the continued investigation of the relationship between the

ApoE e4, olfaction, and olfactory fMRI in AD and MCI patients.

These future studies have the potential to further understand the olfactory dysfunction in

AD and MCI. This dissertation shows the potential for olfactory fMRI and olfactory testing as

diagnostic marker and as a way to study the progression of the disease. The addition of genetic

data, DTI analysis, and longitudinal research will allow for more accurate and concrete diagnosis.

5.7 Summary

We have demonstrated that the volume of the POC decreased similarly to that of the

hippocampus in both MCI and AD patients. Cognitive, smell identification, and volumetric

measurements of MCI subjects encompassed a wide range of scores, with some scoring in the

range of normal controls and others in the range of AD subjects. However, significantly

decreased functional activation of MCI subjects compared with normal controls was

demonstrated on fMRI compared with the less discernible morphological or behavioral

differences between MCI and other groups. In summary, MCI subjects functionally resembled

128

AD, but behaviorally, MCI subjects significantly outperformed AD subjects possibly due to

greater functional connectivity. Of great importance, the results showed that olfactory fMRI could

be used in conjunction with the UPSIT and volume measurement of the hippocampus to increase

the diagnostic sensitivity and specificity of at-risk patients. This clearly demonstrates the

potential of olfactory testing and olfactory fMRI in the diagnosis of AD and MCI. Olfactory

fMRI is expensive and may not seem feasible to be used in diagnosis. Nevertheless, most patients

diagnosed with MCI or early AD have MRI studies performed. Therefore addition of an olfactory

fMRI study is very possible.

Olfactory analysis at the behavioral, anatomical, functional, and network level suggest

the involvement of the central olfactory system in AD and MCI. Olfactory fMRI can be used not

only to increase diagnostic specificity and sensitivity in AD but also to track and study the

progression of MCI to AD. Ultimately, further study of olfactory fMRI offers great potential in

early diagnosis of AD and in testing effectiveness or response of patients to future drug therapy.

129

5.8 References


Neuropathol. 1991;82(4):239-59.

[2] Braak H, Braak E. Frequency of stages of Alzheimer-related lesions in different age

categories. Neurobiol Aging. 1997 Jul-Aug;18(4):351-7.

[3] Braak H, Braak E. Morphological criteria for the recognition of Alzheimer's disease and the

distribution pattern of cortical changes related to this disorder. Neurobiol Aging. 1994 May-

Jun;15(3):355-6; discussion 379-80.

[4] Hyman BT. The neuropathological diagnosis of Alzheimer's disease: clinical-pathological

studies. Neurobiol Aging. 1997 Jul-Aug;18(4 Suppl):S27-32.

[5] Talamo BR, Rudel R, Kosik KS, Lee VM, Neff S, Adelman L, Kauer JS. Pathological

changes in olfactory neurons in patients with Alzheimer's disease. Nature. 1989 Feb

23;337(6209):736-9.

[6] Pearson RC, Esiri MM, Hiorns RW, Wilcock GK, Powell TP. Anatomical correlates of the

distribution of the pathological changes in the neocortex in Alzheimer disease. Proc Natl

Acad Sci U S A. 1985 Jul;82(13):4531-4.

[7] Harrison PJ. Pathogenesis of Alzheimer's disease--beyond the cholinergic hypothesis:

discussion paper. J R Soc Med. 1986 Jun;79(6):347-52.

[8] Christen-Zaech S, Kraftsik R, Pillevuit O, Kiraly M, Martins R, Khalili K, Miklossy J. Early

olfactory involvement in Alzheimer's disease. Can J Neurol Sci. 2003 Feb;30(1):20-5.


immunohistochemical markers in healthy aging and Alzheimer's disease. Neurobiol Aging.

1991 Jul-Aug;12(4):295-312.

130

[10] Arnold SE, Hyman BT, Flory J, Damasio AR, Van Hoesen GW. The topographical and

neuroanatomical distribution of neurofibrillary tangles and neuritic plaques in the cerebral

cortex of patients with Alzheimer's disease. Cereb Cortex. 1991 Jan-Feb;1(1):103-16.



disease and schizophrenia. Ann N Y Acad Sci. 1998 Nov 30;855:762-75.

[12] Waldton S. Clinical observations of impaired cranial nerve function in senile dementia. Acta

Psychiatr Scand. 1974;50(5):539-47.

[13] Ferreyra-Moyano H, Barragan E. The olfactory system and Alzheimer's disease. Int J

Neurosci. 1989 Dec;49(3-4):157-97.


associated with degree of dementia in Alzheimer's disease. Neurobiol Aging. 1990 Jul-

Aug;11(4):465-9.

[15] Djordjevic J, Jones-Gotman M, De Sousa K, Chertkow H. Olfaction in patients with mild

cognitive impairment and Alzheimer's disease. Neurobiol Aging. 2008 May;29(5):693-706.

[16] Lehrner J, Pusswald G, Gleiss A, Auff E, Dal-Bianco P. Odor identification and self-

reported olfactory functioning in patients with subtypes of mild cognitive impairment. Clin

Neuropsychol. 2009 Jul;23(5):818-30. doi: 10.1080/13854040802585030. Epub 2009 Feb

11.






131

[19] Wilson RS, Schneider JA, Arnold SE, Tang Y, Boyle PA, Bennett DA. Olfactory

identification and incidence of mild cognitive impairment in older age. Arch Gen Psychiatry.

2007 Jul;64(7):802-8.





[22] Karunanayaka P, Eslinger PJ, Wang JL, Weitekamp CW, Molitoris S, Gates KM, Molenaar

PC, Yang QX. Networks involved in olfaction and their dynamics using independent

component analysis and unified structural equation modeling. Hum Brain Mapp 2013; Epub

ahead of print



32.


disease. Am J Psychiatry 1991; 148:357-60.

[25] ter Laak HJ, Renkawek K, van Workum FP. The olfactory bulb in Alzheimer disease: a

morphologic study of neuron loss, tangles, and senile plaques in relation to olfaction.

Alzheimer Dis Assoc Disord 1994; 8:38-48.




73: 365-9.


magnetic resonance imaging. Nature reviews Neuroscience2007;8(9):700–11.

132

[29] Greicius MD, Srivastava G, Reiss AL, Menon V. Default-mode network

activitydistinguishes Alzheimer’s disease from healthy aging: evidence from functionalMRI.

Proceedings of the National Academy of Sciences of the United States ofAmerica

2004;101(13):4637–42.

[30] Wang K, Liang M, Wang L, Tian L, Zhang X, Li K, et al., Altered functional connectivity in

early Alzheimer's disease: a resting-state fMRI study. Hum Brain Mapp, 2007. 28(10): 967-

78.



Human Brain Mapping 2005; 26:231-239.

[32] Gour N, Felician O, Didic M, Koric L, Gueriot C, Chanoine V, et al. Functional connectivity

changes differ in early and late-onset alzheimer's disease. 67. Hum Brain Mapp. 2013 Oct 5.

doi: 10.1002/hbm.22379. [Epub ahead of print]

[33] Huang H, Fan X, Weiner M, Martin-Cook K, Xiao G, Davis J, Devous M, Rosenberg R,

Diaz-Arrastia R. Distinctive disruption patterns of white matter tracts in Alzheimer's disease

with full diffusion tensor characterization. Neurobiol Aging. 2012 Sep;33(9):2029-45. doi:

10.1016/j.neurobiolaging.2011.06.027. Epub 2011 Aug 27.

[34] Brown JA, Terashima KH, Burggren AC, Ercoli LM, Miller KJ, Small GW, Bookheimer

SY. Brain network local interconnectivity loss in aging APOE-4 allele carriers. Proc Natl

Acad Sci U S A. 2011 Dec 20;108(51):20760-5. doi: 10.1073/pnas.1109038108. Epub 2011

Nov 21.

[35] Finkel D, Reynolds CA, Larsson M, Gatz M, Pedersen NL. Both odor identification and

ApoE-ε4 contribute to normative cognitive aging. Psychol Aging. 2011 Dec;26(4):872-83.

doi: 10.1037/a0023371. Epub 2011 Apr 25.

133

[36] Ruan Y, Zheng XY, Zhang HL, Zhu W, Zhu J. Olfactory dysfunctions in neurodegenerative

disorders. J Neurosci Res. 2012 Sep;90(9):1693-700. doi: 10.1002/jnr.23054. Epub 2012 Jun

5.

VITA

Megha Vasavada

Education

Pennsylvania State University 2009- 2014

Neuroscience Doctorate Program

Brandeis University 2003-2007

Bachelor of Science/Art in Neuroscience, Biology, & Psychology

Honors and Awards

ISMRM Merit Award Magna Cum Laude 2014

Graduate Alumni Endowed Scholarship 2013

Class of 1971 Scholarship 2013

Association for Chemical Sciences Housing Award 2013

Placement on Brandeis University Dean’s List 2006-2007

Publications

Vasavada MM, Wang J, Eslinger PJ, Gill DJ, Karunanayaka P, Yang QX. Functional and

structural degeneration of the primary olfactory cortex in AD and MCI, in preparation

Vasavada MM, McHugh, R, Zhang H, Wang J, Eslinger PJ, Gill DJ, Karunanayaka P, Yang QX.

Functional connectivity of the piriform is disrupted in AD and MCI, in preparation

Vasavada MM, Wang J, Eslinger PJ, Gill DJ, Karunanayaka P, Yang QX. Central olfactory

dysfunction is the dominant cause of olfactory deficits in AD and MCI, in preparation

Vasavada MM, Gill DJ, Wang J, Eslinger PJ, Yang QX. Olfactory dysfunction in AD and MCI,

in preparation

Select Presentations

Vasavada MM, Wang J, Karunanayaka P, Yang QX. Functional Connectivity of the Primary

Olfactory Cortex is decreased in Alzheimer’s Disease. International Society for Magnetic

Resonance in Medicine 2014 (talk)

Vasavada MM, Wang J, Sun X, Eslinger PJ, Karunanayaka P, Yang QX. fMRI of the Primary

Olfactory Cortex in Alzheimer’s Disease and Mild Cognitively Impaired Patients. Human Brain

Mapping 2013 (Poster)

Vasavada MM, Wang J, Sun X, Weitekamp C, Karunanayaka P, Ryan S, Yang QX. Primary

olfactory cortex is affected in Alzheimer’s disease and mild cognitively impaired patients: A

neuroimaging study. The Association for Chemoreception Sciences 2013 (poster)

Patel M, Lemieux S, Wilson S, Corwin R, Hayes J, Stitt J, Engels A, Wang J, Vesek J, Yang QX.

Food craving studied by combined visual and olfactory stimulation. The Association for

Chemoreception Sciences 2012 (poster)

Documents

ALZHEIMER’S DISEASE AND MILD COGNITIVE IMPAIRMENT …