Embed Size (px)
Citation preview
Sleep and Alzheimer’s disease:
A critical examination of the risk that Sleep Problems or Disorders particularly Obstructive Sleep Apnea
pose towards developing Alzheimer’s disease
by
Omonigho A. Michael Bubu
A dissertation submitted in partial fulfillment of the requirements of the degree of Doctor of Philosophy Public Health
with a concentration in Epidemiology Department of Epidemiology and Biostatistics
College of Public Health University of South Florida
Major Professor: Kevin Kip, Ph.D., FAHA, AAAS Fellow Co-Major Professor: Dave Morgan, Ph.D.
Alfred Mbah, Ph.D. Ricardo Osorio, M.D. MA
Date of Approval: November 16, 2017
Keywords: Mild Cognitive Impairment, Beta Amyloid, Cerebrospinal Fluid, Hippocampal Atrophy
Copyright © 2017, Omonigho A. Michael Bubu
DEDICATION
First, I humbly dedicate this body of work to God Almighty, who graced and enabled me to be
able to do this. Without YOU Lord, I would not have had this opportunity. I am truly thankful for YOUR
infinite mercies and grace.
Second, I dedicate this work to my wife, Queen Ogie Umasabor-Bubu, M.D., M.P.H, CPH, CIC.
Your unwavering love, support and resilience through thick and thin, I sincerely appreciate. Without you,
this could not have materialized.
Third, I dedicate this to my kids, Oruno Darah Bubu, and Oreva Charis Bubu. Thank you for
being the best kids in the world and coping with daddy and mummy, as we joggled being international
students and being parents at the same time. Your smiles and the joy you bring to our lives, make the
challenges we faced worthwhile. You both, as well as mummy, are truly God’s expression of HIS divine
love towards me.
Lastly, I dedicate this work to the memory of parents, Chief Samuel Onogharigho Bubu, and Mrs.
Elizabeth Itete-Bubu. Your labor was not in vain, and even though it saddens my heart that both of you
are not around to see us your children achieve our dreams and goals in life, be rest assured that we would
do you proud. Keep on resting in the bosom of our Lord.
ACKNOWLEDGMENTS
I thank my committee members for their support, encouragement and belief in me. Dr. Kevin
Kip, thank you for accepting to be my major advisor and providing me the opportunity to see this work
through. To me, you are God sent, and I am truly grateful for your guidance, and counsel. Dr. Dave
Morgan, thank you for standing by me in my most trying period. Your belief in me is palpable and I am
extremely fortunate to have such an authority in the field in my committee. Dr. Ricardo Osorio, thank you
for welcoming me with open arms and having confidence in me. Your belief in me emboldens me to aim
higher in the field as a researcher. Your friendship is of inestimable value. Dr. Alfred Mbah, thank you for
being a friend and a brother. Your help and guidance in helping me consolidate my understanding of SAS
and helping me whenever I was stuck, is sincerely appreciated. I must thank Dr. Jim Mortimer and Dr.
Amy Borenstein for your mentorship, counsel, guidance and support in my most trying time. Thank you
for your constant belief in me, and pushing me to know no limitations. Dr. Karen Liller, I am eternally
thankful for your mentorship and grace. You stood by me in my most trying times, patient, insightful and
empathetic. You all have contributed in no small way to any success I have had.
I am thankful for all my critics, for you made me a better person. I am grateful for librarians
Allison Howard and John Orriola’s assistance in developing search strategies as I embarked on my
research. I am also thankful for Pharmacists Kayode and Salewa Ogundipe; Pastor Zak Moussa, Pastor
Mathew Owotoki, Pastors (Mr. & Mrs. Faith), Pastor Isaac Olabisi and Rev. Adetunji for their spiritual
and moral guidance. You all were my helpers of the war and I remain eternally grateful. Thank you to all
my other professors and academic staff for making me the academic that I am today. God bless you all.
i
TABLE OF CONTENTS
List of Tables v
List of Figures vii
Abstract ix
Section 1: Systematic Review: Sleep and Alzheimer’s disease 1 Note to the Reader 1 Abstract 1 Introduction 3 Sleep and Normal Aging 3 Sleep and Demented Older Adults 4 Previous narrative reviews 4
Objective of this systematic review 5 Methods 5
Identification of Eligible studies 6 Electronic search 6 Searching other sources 7 Selection criteria 7
Type of study 7 Sleep abnormalities and/or sleep disorders 8 Cognitive decline and/or Alzheimer’s disease 9 Data extraction 9 Assessment of study quality 10 Risk of bias in individual studies 11 Observational data from studies 11
Results 11 Study selection 11
Study characteristics 12 Participant and setting 13
Sleep abnormalities and/or disorders 13 Cognitive decline and/or Alzheimer’s dementia 14 Study findings 14
A. Sleep depth, cognitive decline and Alzheimer’s disease 14 A.1: Cross-sectional studies 15
A.2: Case-control studies 17 A.3: Cohort studies 19 A.4: Experimental studies 22 A.5: Randomized Controlled Trials 25
B. Sleep wake-cycle/circadian rhythm, Cognitive decline and Alzheimer’s disease 25
ii
B.1: Cross-sectional studies 25 B.2: Case-control studies 27
B.3: Cohort studies 29 B.4: Experimental studies 30 B.5: Randomized Controlled Trials 33
C. Hypoxia, Cognitive decline and Alzheimer’s disease 33 C.1: Cross-sectional studies 33 C.2: Case-control studies 35 C.3: Cohort studies 35 C.4: Experimental studies 36 C.5: Randomized Control Trials 39 D. Insomnia, Cognitive decline and Alzheimer’s disease 39 D.1: Cross-sectional studies 39 D.2: Cohort studies 40 E. Restless Leg Syndrome, Cognitive decline and Alzheimer’s disease 42 E.1: Cross-sectional studies 42 E.2: Case-control studies 42
Discussion 43 Sleep depth, Cognitive decline and Alzheimer’s disease 43
Circadian Rhythm abnormalities, Cognitive decline and Alzheimer’s disease 46 Hypoxia, Cognitive decline and Alzheimer’s disease 48 Insomnia, Cognitive decline and Alzheimer’s disease 50 Restless Leg Syndrome, Cognitive decline and Alzheimer’s disease 50 Summary of evidence 50
Sleep is associated with cognitive decline and/or AD/AD pathology 50 Sleep problems precede the onset of cognitive impairment 52
Sleep problems may lead to increased AD pathology or may have their effects on dementia through other pathways 53
Limitations 54 Conclusion and Future Directions 54 Research agenda and gaps in literature I intend to address 55 References 56 Section 2: Sleep, Cognitive Impairment and Alzheimer’s disease: A Systematic Review and Meta- Analysis 110
Note to the Reader 110 List of Authors 110 Author contributions 111
Abstract 111 Introduction 113 Methods 113
Identification of eligible studies 114 Electronic search 114 Searching other sources 114 Selection criteria 115
Types of studies 115 Sleep problems and/or disorders 115 Alzheimer’s disease/cognitive impairment 116 Participant and setting 116 Data extraction 116 Assessment of study quality 116
iii
Statistical analysis 117 Results 118
Study selection 118 Study characteristics 119
Design and quality 119 Participant and setting 119
Sleep problems and/or disorders 120 Cognitive impairment and/or Alzheimer’s disease 120 Meta-analysis 121
Overall meta-analysis based on risk estimates for the effect of sleep on cognitive impairment and Alzheimer’s disease 121 Sub-group meta-analysis based on risk estimates for the effect of sleep on cognitive impairment, Pre-clinical and ICD9/DSMIV Diagnoses of Alzheimer’s disease 121 Funnel Plot Asymmetry Test and Trim and Fill Analysis Assessment of Publication Bias 122 Sub-group meta-analysis based on risk estimates for the effect of different sleep problems and disorders on cognitive impairment and Alzheimer’s disease 122 Sub-group meta-analysis based on risk estimates for the effect of self-report and actigraphic data of poor sleep on cognitive impairment and/or Alzheimer’s disease 123 Exploring Potential Causes of Heterogeneity 124 Sensitivity Analyses 124
Meta-regression analysis 126 Population Attributable Risk percent 126
Discussion 126 Summary of main results 126 Effect of sleep problems on cognitive impairment or Alzheimer’s disease (meta-analysis) 127 Potentially influencing factors (meta-regression analysis) 128 Study results in comparison with other reviews on the topic 128 Strengths and Limitations of the study 129 Future directions 131 Conclusion 131
References 132 Appendix A Supplemental Tables and Figures: Sleep and AD Meta-analysis 153
Section 3: Obstructive Sleep Apnea is associated with longitudinal increases in brain Aβ42, CSF-TAU, PTAU, and decrease in CSF Aβ42 burden, in elderly Cognitive Normal and MCI Individuals 165
Key Points 165 Abstract 165 Introduction 167 Methods 168
Study Participants 168 OSA diagnosis 169 NL, MCI and AD diagnosis 169 Florbetapir-PET Imaging Acquisition and Interpretation 169 Cerebrospinal Fluid Methods 169
iv
Covariates/Potential Confounders 170 Data Analysis 170
Results 171 Demographic and Clinical Characteristics 171 Baseline AD Biomarker levels by clinical group and OSA status 172 Rate of change in AD Biomarker by OSA status 172
Brain Aβ-42 levels 172 CSF Aβ-42, TAU and PTAU levels 173
Discussions 173 Baseline AD Biomarker levels by clinical group and OSA status 174 Rate of change in AD Biomarker by OSA status 174 Strengths and Limitations 176
Conclusions 176 References 176
Section 4: Obstructive Sleep Apnea: A Distinct Physiological Phenotypic Risk Factor in older adults with Cognitive decline and Alzheimer’s disease 189
Abstract 189 Introduction 190
Age-related and Age-dependent OSA co-morbidities 190 Review Objective 193
Methods 193 Search strategy 193 Selection criteria 194 Reviewing procedure and Data extraction 195 Assessment of study quality 195
Results 196 OSA and Cognition (Cross-sectional studies) 196
Main findings 196 OSA and Cognitive Decline (Longitudinal studies) 199
Main findings 199 Summary, critique and future research directions 200
OSA and Alzheimer’s disease (Cross-sectional studies) 202 Main findings 202
OSA and AD Pathology 203 Main findings 203
Cross-sectional studies 203 Prospective/Longitudinal studies 204 Randomized Controlled Trials 205
Summary, critique and future research directions 206 Discussion 208 Conclusion 211 References 211
Appendix B Oxford University Press License Terms and Conditions 249 Appendix C IRB Letter 252 Appendix D Sleep, Cognitive Decline, and Alzheimer’s disease: A Systematic Review and Meta- Analysis 254
v
LIST OF TABLES
Table 1.1 Literature Search Terms for the Systematic Review of Sleep and Alzheimer’s disease 78
Table 1.2 Scoring System for Quality of Paper of the Systematic Review of Sleep and Alzheimer’s disease 79 Table 1. 3 Quality Assessment Scores for cross-sectional studies 82 Table 1.4 Quality Assessment Scores for case-control studies 83 Table 1. 5 Quality Assessment Scores for cohort studies/Randomized Clinical Trials 84 Table 1.6 Quality Assessment Scores for Experimental Studies 85 Table 1.7 Sleep Depth and Cognitive/Alzheimer’s disease measures: Descriptive Study Characteristics and Main Findings 86 Table 1.8 Sleep-wake cycle abnormalities and Cognitive/Alzheimer’s disease measures: Descriptive Study Characteristics and Main Findings 95 Table 1.9 Sleep Disordered Breathing/Sleep Apnea and Cognitive/Alzheimer’s disease measures: Descriptive Study Characteristics, and Main Findings 101 Table 1.10 Insomnia and Cognitive/Alzheimer’s disease measures: Descriptive Study Characteristics and Main Findings 105 Table 1.11 Restless Leg Syndrome (RLS) and Cognitive/Alzheimer’s disease measures: Descriptive Study Characteristics and Main Findings 106 Table 2.1 Characteristics of included studies. Meta-analysis of sleep, cognitive decline and Alzheimer's disease 137 Table 2. 2 Pooled relative risks from sub-group meta-analysis for the effect of sleep on Cognitive decline and/or Alzheimer's disease 142 Table 2.3 Results of meta-regression analysis examining potential effects of different factors on the natural logarithm of the odds ratio between sleep and cognitive decline and/or Alzheimer's disease 144 Table 1A Main findings from included studies and comments regarding computed conversions of the different indices to a common index: odds ratios and regarding computing a summary effect size of the impact of sleep problems on AD 151 Table 2A Formula used for converting among effect sizes and for calculating the variance of
vi
the effect sizes 157 Table 3A Formula used for computing the variance of a composite or a difference 158
Table 3.1 Descriptive Characteristics of Participants by Obstructive Sleep Apnea Status at Baseline 181 Table 3.2 Covariance Parameter estimates and Between Subject variation in AD Biomarker Deposition 182 Table 4.1 Literature Search Terms for the Systematic Review of Obstructive Sleep Apnea, Cognition and Alzheimer’s disease 233 Table 4.2 Descriptive Study Characteristics and Main Findings: Obstructive Sleep Apnea and Cognition (Cross-sectional studies) 234 Table 4.3 Descriptive Study Characteristics and Main Findings: Obstructive Sleep Apnea and Cognitive Decline (Longitudinal studies) 241 Table 4.4 Descriptive Study Characteristics and Main Findings: Obstructive Sleep Apnea, and Alzheimer’s disease/AD Pathology 243
vii
LIST OF FIGURES Figure 1.1 Flow Diagram of Systematic Review of Sleep and Alzheimer’s disease 107
Figure 2.1 Study retrieval and selection for effects of sleep on cognitive decline and/or Alzheimer’s disease meta-analysis 145 Figure 2.2 Forest plot presenting overall meta-analysis based on risk estimates for the effect of sleep on cognitive decline and/or Alzheimer’s disease 146 Figure 2.3 Forest plot presenting sub-group meta-analysis based on risk estimates for the effect of sleep on Cognitive decline, Pre-clinical and Symptomatic Alzheimer’s disease 147 Figure 2.4 Trim and fill funnel plot for effects of sleep on cognitive decline and/or Alzheimer’s Disease meta-analysis 148 Figure 2.5 Forest plot presenting sub-group meta-analysis based on risk estimates for the effect of different sleep problems and disorders on cognitive decline and/or Alzheimer’s Disease 149 Figure 2.6 Forest plot presenting sub-group meta-analysis based on risk estimates for the effect of self-report versus actigraphic data of poor sleep on cognitive decline and/or Alzheimer’s disease 150 Figure 1A Forest plot presenting sub-group meta-analysis based on continent in which study was published for the effect of sleep problems and disorders on cognitive decline and/or Alzheimer’s disease 159 Figure 2A: Forest plot presenting sub-group meta-analysis based on day versus night time sleepiness for the effect of sleep problems and disorders on cognitive decline and/or Alzheimer’s disease 160 Figure 3A Forest plot presenting sub-group meta-analysis based on study temporality for the effect of sleep problems and disorders on cognitive decline and/or Alzheimer’s Disease 161 Figure 4A Forest plot presenting meta-analysis risk estimates based on omitting one study for the effect of sleep problems and disorders on cognitive decline and/or Alzheimer’s Disease 162 Figure 3.1 Baseline Alzheimer Disease CSF Biomarker burden and Brain Aβ-42 levels in Cognitive Normal subjects by OSA status 184 Figure 3.2 Baseline Alzheimer Disease CSF Biomarker burden and Brain Aβ-42 levels in Mild Cognitive Impairment subjects by OSA status 185
viii
Figure 3.3 Baseline Alzheimer Disease CSF Biomarker burden and Brain Aβ-42 levels in Alzheimer Disease subjects by OSA status 186 Figure 4.1 Study retrieval and selection for Obstructive Sleep Apnea, Cognitive Decline and/or Alzheimer’s disease Review 246
ix
ABSTRACT
This dissertation is a critical examination of the relationship between sleep problems and/or
disorders, particularly Obstructive Sleep Apnea (OSA) and Alzheimer Disease (AD). First, I conducted an
exhaustive systematic review of existing literature, and identified gaps in research that led to specific
research aims. For the first aim, I conducted the first ever-published meta-analysis examining sleep,
cognitive decline and AD, providing an aggregate effect of sleep on AD. Second, focusing on OSA, I
conducted a study examining OSA’s effect on longitudinal changes on AD biomarkers in cognitive
normal, MCI and AD subjects, using data from the Alzheimer Disease Neuroimaging Initiative (ADNI).
Lastly, I conducted a review, integrating over 3 decades of research examining OSA and cognition; OSA
and subsequent cognitive decline; and OSA and AD; with particular focus in appreciating the
heterogeneity of OSA and its outcomes in distinct age groups.
Results and implications from my research indicate that ample evidence exists linking sleep
impairments and circadian regulating mechanisms directly to clinical symptoms in AD. Sleep problems
and/or disorders increases your risk of cognitive decline and AD. OSA is associated with increased AD
biomarker burden over time, and effects longitudinal changes in these biomarkers, such that OSA subjects
progress faster than non-OSA subjects do. OSA may be age-dependent in older adults (60 – 70 years old)
and the elderly (70 years and above) and is associated with neurodegenerative diseases particularly,
cognitive decline and AD. Intermittent hypoxia and sleep fragmentation are two main processes by which
OSA induces neurodegenerative changes. Therefore, clinical interventions aimed at OSA, such as
treatment with CPAP or dental appliances, in cognitive normal and MCI patients, could possibly slow the
progression of cognitive impairment to AD.
1
SECTION 1
SYSTEMATIC REVIEW: SLEEP AND ALZHEIMER’S DISEASE
Note to the Reader:
Portions of this section have been previously published in SLEEP journal, 2017, Jan 1; 40(1). doi:
10.1093/sleep/zsw032 and have been reproduced with permission from Oxford University Press.
ABSTRACT
Background/Objective: Alzheimer’s disease (AD) is a neurodegenerative disorder with expected
incidence of approximately a million people each year, and a total estimated prevalence of 11 to 16
million people by 2050 in the United States. Mounting evidence implicates sleep as one of the risk factors
for AD with an association between sleep disruption, cognitive ability, and neurodegenerative disease.
We systematically reviewed all available studies examining any association and/or relationship and/or
biologic plausibility between sleep and its related problems/disorders and Alzheimer’s disease, cognitive
decline or dementia and evaluated the evidence for a causal association.
Methods: Original published literature assessing any association of sleep and its related
problems/disorders with AD, cognitive decline or dementia was identified by searching four bibliographic
databases, namely PubMed, Embase, Web of Science, and Cochrane library (i.e., the Cochrane Central
Register of Controlled Trials). Searches on all databases were from the onset of the archives until and
including November, 2014. The terms Alzheimer, Mild Cognitive Impairment, Sleep, Sleep Disorders
and Circadian Rhythm were identified as MeSH terms. Other MeSH search headings related to sleep and
various study types including cross-sectional, case control, cohort (retrospective and prospective),
Randomized Clinical Trials (RCTs) were also identified. Articles identified from the search were first
2
screened using titles and abstracts of the publications and eligible articles for this review had to meet
certain inclusion and exclusion criteria.
Results: Seventy-two publications that met the selection and final inclusion criteria were identified by the
literature search: 18 studies and 12 studies used a cross-sectional analysis and case-control design
respectively, to examine the association of sleep alterations/disturbances, circadian rhythm abnormalities,
and/or sleep disorders with AD, cognitive decline and/or dementia; 22 studies were prospective studies
examining the relationship between sleep and cognitive outcomes; 4 RCTs examined cognitive effects of
treating sleep and its related problems/disorders in AD; and 16 experimental studies examined the
associations between sleep and AD pathology and/or AD biomarkers. Altogether, there is substantial
evidence providing support that sleep is associated with cognitive decline and/or AD/AD pathology.
Prospective studies provide support that sleep problems precede the onset of cognitive impairment,
including AD and dementia, and experimental studies provide the most compelling evidence of the
plausibility of causal associations between sleep deprivation and AD pathology and/or AD biomarkers.
Conclusion: There is growing experimental and epidemiological evidence for a reciprocal interaction
between cognitive decline and sleep alteration. Ample evidence exists linking sleep impairments and
circadian regulating mechanisms directly to clinical symptoms in AD. However, long-term longitudinal
studies are needed to determine the relationship of chronic sleep deprivation, circadian rhythm/sleep-
wake abnormalities, sleep-disordered breathing and sleep disorders with cognitive decline.
3
INTRODUCTION
Alzheimer’s disease (AD) is a neurodegenerative disorder with an incidence in the United States
expected by 2050 to be of a million new cases yearly, and a corresponding prevalence of 11 to 16 million
people.1 In 2013, approximately 5.2 million Americans had Alzheimer’s Disease (AD), including an
estimated 5 million age 65 and older.2 AD and other dementias cause significant morbidity and mortality
to those afflicted, with an enormous socio-economic impact: an estimated US$604 billion worldwide,
constituting about 1% of the aggregated global gross domestic product.3 Preventive and/or therapeutic
measures targeted at delaying or curing AD are therefore imperative.4 Advancing age is the most
significant risk factor for AD with most individuals with AD diagnosed at age 65 or older. Various risk
factors for AD including family history,5-7 Apolipoprotein E epsilon4 allele (APOE-ε4 gene),8,9
smoking,10-12 diabetes,13-15 obesity in midlife,16-20 high cholesterol in midlife,17,21 hypertension in
midlife,22-24 physical activity,25-28 education,29,30 head circumference,31,32 head trauma,33,34 brain reserve35
and social and cognitive engagement,36-38 have been identified in epidemiological studies and additional
risk factors are being investigated.39 Mounting evidence implicates sleep as one of these factors.
Sleep and Normal Aging
Structural alterations of sleep, including sleep maintenance difficulty, sleep latency increase,
frequent nighttime and early morning awakenings, occur as part of the normal aging process.40 Various
sleep stages also show some level of alteration with increasing age including a reduction in slow wave
sleep (SWS) and a compensatory increase in sleep stages 1 and 2.40 Rapid-eye movement (REM) sleep
changes are less distinct and seem to appear later with increasing age.41,42 Microstructural alterations such
as decrease in K complexes and sleep spindles are also observed.43 Apart from these structural alterations
of sleep, sleep-wake rhythm disturbances have also been described in the elderly. Greater tendency for
daytime sleep and increased propensity to phase advancing are observed frequently.41 Circadian sleep-
wake rhythm disorders do not result from behavioral factors alone but also from neuro-hormonal
modifications, especially melatonin whose plasma concentration declines with age.44 Psychiatric and
4
somatic pathologies, pharmacological treatments, decrease in physical activity and exposure to light also
play a role in the origin of sleep disturbances in normal aging.40 Sleep disorders such as sleep-related
breathing disorders (SRBD), restless legs syndrome (RLS) and periodic limb movements in sleep (PLMS)
also increase significantly with age.41
Sleep and Demented older adults
Demented older adults exhibit significant sleep disturbance and degenerate sleep patterns,
including shorter sleep duration and fragmented sleep,43,45-47 altered circadian rest/activity patterns,45 and
elevated rates of sleep disordered breathing (SDB).48,49 It is estimated that about 45% of AD patients have
sleep disturbances.50,51 When symptoms appear at an early stage of the disease, they are frequently
associated with a more severe cognitive decline.50,52 Both circadian rhythm disruptions and sleep
disturbances have a major impact on the quality of life of patients and their caregivers.53-56 They have
been conventionally thought of as complications of various psychiatric co-morbidities50,57 secondary to
disease pathology. However, researchers have begun to examine whether sleep and its related problems
contribute to the risk of AD.58 Sources of sleep disturbance in AD are thought to include degeneration of
neural pathways that regulate sleep-wake patterns and sleep architecture.50,57 Sleep disorders may also
exacerbate cognitive symptoms through impairment of sleep-dependent memory consolidation
processes.59,60 Interestingly, circadian and sleep disorders may have a causal role in the pathophysiology
of AD, and could be critical to the goal of preventing AD, because effective interventions exist to
improve sleep.61
Previous narrative reviews
There are few narrative review articles,40,62-64 that have examined this association with focus on
different aspects of sleep disturbances and/or neurologic diseases. Palma et al.,62 focusing on sleep
deprivation, reviewed how sleep loss constitutes a risk factor for various neurologic diseases including
AD. In discussing sleep loss and AD, they concluded that it was necessary to distinguish between the
5
possibility that the risk of AD is influenced by changes in rest activity patterns and sleep-wake cycles,
through their effects on underlying neurodegenerative processes or the alternative explanation that rest
activity and sleep-wake patterns are just markers of a subclinical neurodegenerative process. Slats et al.,63
focusing on the role of hypocretin and melatonin, described the impairments of both sleep regulating
systems and circadian rhythms in AD and their link to clinical symptoms. They further discussed how
sleep regulating systems, especially neurotransmitters such as melatonin and hypocretin, could affect AD
pathophysiology. They concluded that there is substantial evidence of impairments in both sleep and
circadian regulating mechanisms in AD pathophysiology, which may be linked directly to clinical
symptoms in AD patients. Peter-Derex et al.,40 did a thorough clinical review of sleep and Alzheimer’s
disease, observing that micro-architectural sleep alterations, nocturnal sleep fragmentation, decrease in
nocturnal sleep duration, diurnal napping and inversion of the sleep-wake cycle are the main disorders
observed in patients with AD. Further, they noted that there is growing experimental and epidemiological
evidence for a close reciprocal interaction between cognitive decline and sleep alterations.
Objective of this systematic review
To the best of our knowledge, there are no systematic review articles on sleep and AD or
Dementia. In this context, we conducted a broad systematic review of all available studies examining any
association of sleep and its related problems/disorders with Alzheimer’s disease, cognitive decline or
dementia, and evaluated the evidence for a causal association. Recommendations for future directions of
research are also proposed.
METHODS
We conducted this systematic review in accordance with the exemplified step-by-step guide for
systematic reviews and meta-analysis by Pai et al.,65 and for the purpose of reporting, we adhered to the
Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement by
Moher et al.66 We also consulted the Cochrane handbook for systematic reviews of interventions.67
6
Identification of Eligible Studies
Electronic search
An extremely sensitive search strategy was developed and assisted with identification of all
eligible published articles in scientific (medical, psychological and epidemiological) journals for all years
up to November 2014. The electronic search strategy combined terms characterizing AD/AD
pathogenesis, as the outcome variable, Sleep, as the exposure variable, and a third set specified various
study types. All terms within each set were combined with the Boolean operator OR, and then all three
sets were then combined using AND.
Original published literature assessing associations of sleep and its related problems/disorders
with Alzheimer’s disease, cognitive decline or dementia were identified by searching four bibliographic
databases on November 21, 2014, namely:
• Medline (through PubMed; all years 1946 - present)
• Embase (through www.embase.com by Elsevier B.V. 2011; all years from 1947 - present)
• Web of ScienceTM (Thomson Reuters 2014; all years from 1900 – present)
• Cochrane library (i.e., the Cochrane Central Register of Controlled Trials; all years from 1900 -
present).
The terms Alzheimer, Mild Cognitive Impairment, Sleep, Sleep Disorders and Circadian Rhythm were
identified as MeSH terms. Other MeSH search headings related to sleep included sleep loss, sleep
deprivation, sleep apnea, sleep disruption, excessive daytime sleepiness, Obstructive Sleep Apnea (OSA),
Sleep-Disordered Breathing, Rapid Eye Movement Behavioral Disorder, melatonin, and hypocretin.
Search terms related to Alzheimer’s disease and AD pathogenesis included amyloid beta, amyloid
cascade hypothesis, Amyloid Precursor Protein (APP) and APOE-ε4. MeSH terms associated with
various study types included cross-sectional, case control, cohort (retrospective and prospective),
Randomized Clinical Trials (RCTs), literature reviews, systematic reviews, and ecological studies.
7
Articles written in other languages other than English were also considered and screened with the help of
Google translator.
In PubMed, a search using Alzheimer’s disease and AD pathogenesis MeSH terms yielded
136,119 articles; and a search using sleep and its related problems/disorders MeSH terms yielded 185,925
articles. The combination of the Alzheimer’s disease and AD pathogenesis MeSH terms and sleep and its
related problems/disorders MeSH with MeSH terms associated with various study types, yielded a total of
904 publications. The PubMed search was adapted to searching the other databases. Table 1.1 provides a
list of the search terms used. The complete search strategy including the saved search histories in the
various databases is available on request.
Searching other sources
To locate additional relevant studies, the reference lists of articles identified through database
searches and the Bibliographies of narrative review articles were examined. We also hand-searched the
previous year’s issues (November 2013 to November 2014) of high ranking journals that published more
than one relevant identified article during the initial search, including Sleep Medicine Reviews,
Neurology, JAMA (Journal of American Medical Association), Sleep Medicine, American Journal of
Alzheimer’s Disease and other Dementias, Sleep, Journal of Sleep Research, Journal of clinical sleep
medicine, Journal of Alzheimer’s Disease, Brain, and Neurobiology of Aging.
Selection criteria
Type of study
Original articles from observational studies (cross-sectional, case-control and
retrospective/prospective cohort), Randomized Controlled Trials (RCT), and Animal Experimental
studies were included in this systematic review. Articles identified in the search were first screened using
titles and abstracts of the publications. Eligible studies for this review had to meet the following final
study inclusion criteria: (1) They reported a primary or secondary data analysis using an appropriate
8
epidemiologic study design of any association of sleep and its related problems/disorders with
Alzheimer’s disease, cognitive decline or dementia; (2) The diagnosis of Alzheimer's
disease/dementia/cognitive decline was based on a self-reported diagnosis, on meeting standardized
criteria (e.g. International Classification of Disease-9 (ICD -9), Diagnostic Statistical Manual-IV (DSM-
IV)) or objective measures of Alzheimer pathology (e.g., Positron Emission Tomography (PET) scan); (3)
The finding of sleep and its related abnormalities/disorders was based on self-reported diagnosis or
standardized criteria/questionnaires (e.g., PQSI Pittsburg Sleep Quality Index (PQSI),68 Epworth
Sleepiness Scale (ESS)) or objective measures (e.g., wrist actigraphy69 or polysomnography (PSG)), (4)
Inclusion of a comparison group with or without varying levels of sleep disturbances/disorders or
cognitive decline/dementia/AD, whichever is applicable, and (5) The study provided a valid measure of
association between sleep and its related problems/disorders and Alzheimer’s disease, cognitive decline
or dementia with 95% confidence intervals and/or p-values. Exclusion criteria included the following: (1)
case series/reports, systematic and/or literature reviews, abstracts or editorials, (2) articles that did not
examine an association between sleep and its related problems/disorders and Alzheimer’s disease,
cognitive decline or dementia, and (2) any RCT that did not examine the cognitive effects of treating
sleep and its related problems/disorders in AD.
Sleep abnormalities and/or disorders
Sleep abnormality and/or disorder was considered as the risk factor of interest in this systematic
review. The international Classification of Sleep Disorders (ICSD-2)70 aptly describes and classifies sleep
disorders, however, previous studies71,72 examining the association between sleep and workplace injuries
made use of various concepts to define sleep problems in order to fit their study design. Accordingly, in
this systematic review, we classified sleep based on the measured sleep parameter and eligible studies in
this review were grouped on this basis. Sleep parameters including duration, quality, efficiency and/or
disturbance were classified as sleep depth problems; circadian rhythm abnormalities, and excessive
daytime sleepiness (EDS) were classified as sleep-wake cycle abnormalities; the apnea hypopnea index
9
(AHI) that measures hypoxia levels in sleep was used to refer to Sleep Disordered Breathing (SDB),
Sleep Apnea (SA) and/or Obstructive Sleep Apnea Syndrome (OSAS); and lastly other sleep disorders
referred to insomnia and Restless Leg Syndrome (RLS). Studies that examined multiple sleep parameters
e.g., circadian rhythm abnormalities and Insomnia, were reviewed under the various sleep categories,
however, were still counted as one study.
Cognitive decline and/or Alzheimer’s disease Dementia
The outcome of interest in this review was Cognitive Decline and/or Alzheimer’s disease.
Eligible studies were expected to have diagnosed cognitive decline and/or Alzheimer's disease based on a
self-reported diagnosis that met standardized criteria (e.g. International Classification of Disease-9 (ICD -
9), Diagnostic Statistical Manual-IV (DSM-IV)) or on objective measures of Alzheimer pathology (e.g.,
Positron Emission Tomography (PET) scan). AD is the most common form of Dementia accounting for
up to 50-80% of dementia cases,1 as such studies that had dementia as an outcome were also considered.
Studies that solely looked at specific forms of dementia other than AD e.g., vascular, Lewy body,
frontotemporal dementia were excluded.
Data Extraction
An independent assessment of all the titles and abstracts of all eligible study identified by the
search strategy was performed by two authors (OB, SS) using EndNote X6. Potential articles for the study
were retrieved and examined for possible inclusion by the same two authors. Discordances that were
unresolved in unison were referred to a third author (AB). Excel spreadsheet and extraction sheets created
in Access (Microsoft Office version 2010) was then used for the data extraction. This was independently
done by two reviewers (OB, JM), and extracted categories included the study design, study population,
exposure assessment, outcome assessment, statistical methods and tests used, covariates as well as the
main results of the study. Differences in the information extracted were resolved by a third author (MD)
Reviewers were not blinded to the funding, authors, or institutions.
10
Assessment of study quality
Study quality was assessed using an adaptation from previous systematic reviews73,74 and the
modified version of the Newcastle-Ottawa scale (NOS) for quality assessment of the observational
studies,75,76 with addition of new items relevant to this review (Table 1.2). Two review authors (OB, JM)
assessed each included study separately. Any disagreement between the two review authors was settled
by consensus, or where necessary, by a third party (YW). Parameters used for the quality assessment
included well specified hypothesis, study design type, appropriately described sample population, sample
size, definition of sleep and its related abnormalities, definition of Alzheimer's disease/dementia/cognitive
decline, statistical analytical methods, and inclusion or approach used to adjust for potential confounders.
The quality rating scale was adapted to specific study design, and each category had specific maximum
number of awarded stars (Tables 1.3 – 1.6). Quality scores were awarded such that the minimum number
of stars a study can get was 5. For cross-sectional studies, 16 stars was the maximum possible; for case-
control, cohort studies and RCTs 20 stars was the maximum possible, and for animal experimental studies
15 stars was the maximum possible. Quality indicators of most significance for this review (given the
highest weights) included appropriate definition of sleep and its related abnormalities, study population
and adjustment for potential confounders (such as age, years of education, marital status, forced
expiratory volume (FEV) in one second based on spirometry, self-reported history of physician’s
diagnosis of asthma, number of hours of sleep and napping, and use of benzodiazepines, Chronic
Obstructive Pulmonary Disease (COPD), nonfatal coronary heart disease including myocardial infarction,
stroke, the ankle-brachial index, Apolipoprotein E-epsilon4 (APOE-ε4) genotype, Body Mass Index
(BMI) and Depression). Studies were divided into three categories: low quality (< 50% of the maximum
number of stars), medium quality (≈ 55-70% of the maximum number of stars), and high quality (75% or
more of the maximum number of stars) (Tables 1.3 – 1.6).
11
Risk of Bias in Individual Studies
Studies were thoroughly reviewed and the risk of bias within individual studies was assessed by
examining the appropriateness of the study design as it relates to the study hypothesis, how each study
defined and/or ascertained their exposure and/or outcome, and whether appropriate comparison was done
with the unexposed participants. Appropriate identification and/or selection of study participants to
minimize selection bias risk was also important. It was essential that diagnoses were properly assessed
and/or validated, and that studies reported which diagnosis or symptomatology was utilized, and if using
various databases, how overlapping data was handled. Study data were also assessed to determine if
completeness and potential publication/reporting bias could be evaluated.
Observational data from studies
Evidence from the reviewed studies was harmonized and integrated using a narrative synthesis77
and reported data is presented making use of accepted and recognized conventional guidelines78
RESULTS
Study Selection
A total of 2341 publications were identified doing a literature search in PubMed, Embase, Web of
Science, and Cochrane library (i.e., the Cochrane Central Register of Controlled Trials), with each
yielding 904, 1026, 305 and 106 publications respectively (Figure 1.1). One thousand, one hundred and
twenty duplicates were identified and removed of which 90% were from PubMed, and Embase. After
duplicates were removed, 1221 publications were screened for titles resulting in the exclusion of 180
publications because titles did not reflect an examination of an association between sleep and Alzheimer’s
disease. The remaining 1041 publications were screened using the published abstracts resulting in the
exclusion of 840 publications that were removed because they were either case series/reports or were
systematic and/or literature reviews, abstracts or editorials. Two hundred and one full text articles were
assessed for eligibility and 130 publications were further excluded, of which 105 publications identified
12
were excluded because they did not address an association between sleep and its related
problems/disorders and Alzheimer’s disease, cognitive decline or dementia or did not estimated a measure
of association that included 95% confidence intervals or p-values. Twenty-five RCTs were excluded
because they did not examine cognitive effects of treating sleep and its related problems/disorders in AD.
One study was included from a search of the reference lists of articles identified through database
searches. Seventy-two articles met all eligibility criteria.
Study Characteristics (Design and Quality)
Among the 72 studies identified, 18 studies51,79-95 and 12 studies96-107 used a cross-sectional
analysis and case-control design respectively, to examine the association of sleep alterations/disturbances,
circadian rhythm abnormalities, and/or sleep disorders with AD, cognitive decline and/or dementia; 22
studies108-129 were prospective studies examining the relationship between sleep and cognitive outcomes;
4 RCTs130-133 examined cognitive effects of treating sleep and its related problems/disorders in AD; and
16 experimental studies134-149 examined the associations between sleep and AD pathology and/or AD
biomarkers (Table 3). One110 of the prospective studies examining the relationship between sleep and
cognitive outcomes was in Polish; the remaining studies were in English. In total 36 studies were assessed
to be of high quality (16 experimental studies,134-149 3 RCTs,130,132,133 14 prospective studies,109,111,112,114-
116,118,119,121,122,124,126-128 and 3 cross-sectional studies79,83,86); 22 studies were assessed to be of medium
quality (1 RCT,131 6 prospective studies,108,117,120,123,125,129 7 case-control studies,97-99,101,105-107 and 8 cross-
sectional studies51,84,85,88-90,93,94); and 14 studies were assessed to be of low quality (2 prospective
studies,110,150 5 case-control96,100,102-104 and 7 cross-sectional studies80-82,87,91,92,95) (Table 1.3 – 1.6). Type of
study, selection bias related to sampling, measurements of sleep and/or AD solely based on self-reporting,
and insufficient adjustment for core confounders, were the main shortfalls.
13
Participants and setting
The 72 studies included in the systematic review were published between 1998 and 2014. Human
participants of the included studies were middle aged and elderly adults of both sexes with age ranging
from 40 – 91 years cutting across North America (38 studies), Europe (22 studies), Australia (1 study),
Asia (7 studies) and collaborations between continents (5 studies). Participants were drawn from settings
that included communities (32 studies), hospital/patient population (20 studies), institutions e.g., nursing
home residents (4 studies), and brain autopsies (1 study). In total, 50 studies included both sexes, 6
studies were based on females, and three studies were based on males only. Fifteen experimental studies
that made use of AD transgenic mice were also included.
Sleep abnormalities and/or disorders
There was a considerable amount of variation in the methods used by the studies to assess for
sleep problems. Most studies (n=35) utilized objective measures (e.g., wrist actigraphy69 or
polysomnography (PSG), while 11 studies utilized standardized questionnaires including the Pittsburg
Sleep Quality Index (PQSI),68 Epworth Sleepiness Scale (ESS), Berlin questionnaire,151 Behavioral
pathology in AD rating scale, the BEHAVE-AD,152 Women’s Health Initiative Insomnia Rating Scale
(WHIIRS),153 and the International RLS Study Group questionnaire.154,155 Only five studies utilized self-
constructed questionnaires while 4 studies made use of Fluorescence Immuno Assays (FIA) to determine
melatonin/hypocretin concentrations. The experimental studies (n=16) conducted cell culture, transfection
and hypoxic treatment tests or utilized immuno-assays, Enzyme Linked Immuno-sorbent Assay (ELISA),
Western Blot tests to determine melatonin concentrations and/or forced experimental mice into states of
wakefulness. In total, 30 studies (i.e., 10 cross-sectional,51,79,80,82,84,85,88,92-94 5 case-control,96,98-100,105 9
prospective cohort,109-111,116,118,120,122,125,126 1 retrospective cohort,123 4 experimental studies136,138,142 and 1
RCT133) examined sleep depth parameters, 23 studies examined sleep-wake /circadian rhythm
abnormalities (i.e., 5 cross-sectional,87,89,91,95,102 4 case-control,97,101,106,107 6 prospective
cohort,108,118,119,124,127,129 7 experimental studies134,135,137,140,143-145 and 1 RCT131), 16 studies examined
14
hypoxia states (AHI) in sleep apnea/sleep disordered breathing (i.e., 4 cross-sectional,81,83,86,90 1 case-
control,103 3 prospective cohort,125,128,150 1 retrospective cohort,112 2 RCTs,130,132 and 5 experimental
studies139,141,147-149), 6 studies (i.e., 2 cross-sectional87,92 and 4 prospective cohort studies114,115,117,121)
examined sleep as insomnia and 3 studies (i.e., 2 cross-sectional84,92 and 1 case-control104) examined
sleep as RLS.
Cognitive decline and/or Alzheimer’s disease Dementia
There was also great variability in the methods used to diagnose cognitive decline and/or
Alzheimer’s disease Dementia. Most studies made use of self-reported diagnosis that met standardized
criteria (e.g. International Classification of Disease-9 (ICD -9),156 Diagnostic Statistical Manual-IV
Revised Text (DSM-IV TR),157 National Institute of Neurological and Communicative Disorders and
Stroke/Alzheimer’s Disease and Related Disorders Association (NINCS–ARDRA).158 Mild Cognitive
Impairment (MCI) was mostly diagnosed using the Petersen criteria. 159 Cognitive function was assessed
by a battery of neuropsychological tests (mostly with the Mini Mental State Examination (MMSE), 160
California Verbal Learning Test (short form),161 Rey-Osterrieth complex figure,162 Verbal Fluency,
Wechsler Adult Intelligence Score (digit symbol, digit span).163 Emotional state was mostly assessed with
the Geriatric Depression Scale164 and functioning in daily living with Functional Activities
Questionnaire,165 and dementia severity was mostly assessed using the Clinical Dementia Rating Scale
(CDR). Objective measures of Alzheimer pathology (e.g., Positron Emission Tomography (PET) scan for
amyloid load (Aβ), Polymerase Chain reaction (PCR) for APOE genotyping), were also done.
Study findings
A. Sleep depth and cognitive decline and/or Alzheimer’s disease dementia
In total, 30 studies (i.e., 10 cross-sectional,51,79,80,82,84,85,88,92-94 5 case-control,96,98-100,105 9
prospective cohort,109-111,116,118,120,122,125,126 1 retrospective cohort,123 4 experimental studies136,138,142 and 1
RCT133) examined the association between sleep depth (i.e., sleep quality, duration, sleep latency, sleep
15
efficiency and/or disturbance), and cognitive decline and/or Alzheimer’s disease Dementia. Table 1.7
presents the results of the descriptive study characteristics, statistical methods and main findings of the
studies.
A.1 Cross-sectional studies
Nebes et al88 examined the association between sleep quality (i.e., sleep duration, sleep latency
and sleep efficiency) and cognitive performance while controlling for the effects of some common
confounding conditions including anticholinergic drug burden and found that sleep quality was associated
with cognition. The association was not explained by confounding factors such as cerebrovascular
disease, depression, or medication usage. Blackwell et al79 (2006) examined the association between
objectively measured fragmented sleep and decreased cognitive function in a cohort of 2,242 community-
dwelling women 65 years old or older with a baseline Cognitive Abilities Screening Instrument (CASI)
score of >=74 points. The CASI is a composite of the Hasegawa Dementia Screening Scale, the Folstein
Mini-Mental State Examination,160 and the Modified Mini-Mental State Test.166 They found that women
with <70% sleep efficiency had a 61% higher risk of cognitive impairment when compared with women
with sleep efficiency >=70%. Higher sleep latency was also associated with higher risk of cognitive
impairment as was higher Wake after Sleep Onset (WASO). There was no significant relationship for
total sleep time. Saint Martin et al93 examined the association between sleep quality, subjective
cognitive complaints, and neuropsychological performance in a cohort of 272 healthy elderly subjects
(mean age 74.8 ± 1.1 years) recruited from a population-based cross-sectional study on aging and
cardiovascular morbidity, and found that in healthy elderly subjects subjective sleep quality and duration
was not associated with subjective and objective cognitive performances, except for attention. However,
in this study, sleep duration and quality were self-reported and the authors had no information on sleep
structure and sleep fragmentation, two factors implicated in self-reported cognitive deficits.
Guarnieri et al84 described the frequency and characteristics of sleep disturbances in a large
cohort of persons with mild cognitive impairment or dementia and found that over 60% of the participants
16
had one or more sleep disturbances, almost invariably associated with each other but without any evident
and specific pattern of co-occurrence. Persons with AD and those with MCI had the same frequency of
any sleep disorder. This study suffered from considerable selection bias issues as patients were enrolled
consecutively and selected for specific characteristics including the presence and type of sleep disorders.
Moran et al51 using 215 attendees at a memory clinic who were diagnosed with AD found the prevalence
of sleep disturbance in the sample to be 24.5%. Sleep disturbance in AD patients was also associated with
other behavioral symptoms, notably aggressiveness. Pistacchi et al92 enrolled 236 patients (78 men and
158 women) with different subtypes of dementia and investigated the frequency and characteristics of
sleep disorders in the sample. Patients with MCI had a frequency of sleep disturbances of any type equal
to that of patients with AD. Fetveit et al82 aimed to describe the characteristics of 24-hour sleep patterns
in nursing home patients with dementia and to determine whether various degrees of dementia have any
influence on total sleep duration. They found that sleep duration was positively correlated with the
severity of dementia in nursing home patients.
Studies that examine sleep associations with Alzheimer’s disease pathology or Alzheimer’s
disease biomarkers are extremely important to help clarify and consolidate efforts at establishing that AD
pathology might link poor sleep to cognitive outcomes. Since AD biomarkers are evident early in the
Alzheimer’s disease process, such findings might help foster directed preventive efforts at a time when
prevention may still be feasible.167 Spira et al94 examined the association between self-reported sleep
variables and Aβ deposition in 70 community-dwelling older adults who were part of the Baltimore
Longitudinal Study of Aging cohort. Participants underwent [11C]-Pittsburgh compound B PET (PiB
PET) amyloid imaging and completed self-report sleep measures. The investigators found that self-
reported shorter sleep duration and poorer sleep quality both were associated with greater Aβ burden
according to a cortical (i.e. global) measure of PiB uptake, and a measure of uptake in the precuneus – a
region in which Aβ aggregates early in the Alzheimer’s disease course.168 Ju et al85 determined whether
Aβ deposition in preclinical AD, prior to the appearance of cognitive impairment, is associated with
17
changes in quality or quantity of sleep in 142 cognitively normal middle-aged and older adults. They
found that amyloid deposition was associated with poorer sleep quality, specifically poorer sleep
efficiency (% time in bed that was spent asleep), compared to those without amyloid deposition. On the
contrary, sleep quantity (i.e., total sleep time) did not vary between groups. Napping also frequently was
associated with amyloid deposition.
Craig et al80 investigated the monoamine-A-oxidase promoter polymorphism that may impact
sleep/wake control by balancing the availability of serotonin, a precursor of melatonin. The authors found
that the high-activity 4-repeat allele of the monoamine-A-oxidase variable number tandem repeat
promoter polymorphism conferred increased susceptibility to sleep disturbance (p = .008). Nevertheless,
it is worth noting that allelic variations in this gene have been associated with behavioral and personality
traits in other contexts, which may explain this association.169 A quantitative sleep disturbance score was
significantly higher in the patients possessing MAO-A 4-repeat allele genotypes. APOE had no influence
on the development of an altered sleep
A.2 Case-control studies
Hita-Yañez et al98 determined whether changes in polysomnographic (PSG) sleep patterns
accompany subjective sleep complaints in patients with mild cognitive impairment (MCI), whether
meaningful changes in objective sleep physiology are predicted by self-reported sleep measures in MCI
patients, and whether incipient neurodegeneration contributes to exacerbate sleep misperception. Both
PSG and self-reported sleep measures confirmed that sleep is altered in patients with MCI. Whereas
subjective sleep responses predicted fragmentation of slow wave sleep (SWS) in healthy elderly (HE)
individuals, this relationship was not evident in MCI patients. Furthermore, patients with MCI showed
significant discrepancies in the estimation of sleep onset latency when compared with HE subjects. This
study compared PSG sleep data recorded in 1 night with subjective sleep quality over the past few months
to establish relationships between objective and subjective sleep in HE subjects and patients with MCI.
This approach implicitly assumes that PSG data obtained from one single night is representative of a
18
typical night in the past few months. PSG sleep studies were performed without previous adaptation of
participants to the sleep laboratory. As such, authors note that results might have been affected by the
first-night effect (i.e., differences observed on the first PSG sleep recording in comparison with
consecutive ones), which has previously been demonstrated to affect older patients more than younger
ones.170-172
The ApoE4 allele is a known susceptibility gene for AD,9,173-177 especially in homozygous
cases.178 In another study, Hita-Yañez et al99 determined if impaired sleep patterns appear years before
AD diagnosis and whether changes in sleep patterns parallel memory decline as well as its relationship
with the Apolipoprotein E (APOE) e4 genotype in 25 Mild Cognitive Impairment (MCI) patients (7
females, mean age: 70.5 ± 6.8 yr) and 25 healthy old (HO) volunteers (13 females, mean age: 67.1 ± 5.3
yr). The study results showed a significant shortening of rapid eye movement (REM) sleep together with
increased fragmentations of slow-wave sleep in MCI patients relative to healthy elders. Reduction of
REM sleep in MCI patients with APOE e4 was more noticeable than in e4 non-carriers. Changes in sleep
patterns were not correlated with memory performance in MCI patients. Instead, increased REM sleep
was associated with enhanced immediate recall in MCI e4 non-carriers.
Chen et al96 investigated changes in muscular activity, which may indicate a deficiency of
Acetylcholine (ACh), among patients with cognitive impairment, and recruited 9 controls without
dementia, 6 patients with mild cognitive impairment (MCI), and 6 patients with mild Alzheimer's disease
(AD). None of the participants had sleep complaints, and all AD patients were receiving cholinesterase
inhibitors. All participants underwent PSG examination using surface electromyographic (sEMG)
recording of both legs. A one-way analysis of variance (ANOVA) test was used to analyze the differences
in root mean square (rms) amplitude, peak frequency, and mean frequency among the three groups and
the results showed that all rms, mean frequencies, and peak frequencies increased significantly (p<0.05)
in MCI patients. The authors concluded that a deficiency of Ach may result in an increase sEMG activity
in MCI patients, and that because cholinesterase inhibitors are capable of suppressing sEMG activity in
19
AD patients, an increase in sEMG activity is associated with a deficiency of Ach, which could be an early
indicator of dementia. Hot et al100 examined relationships between episodic memory deficits and
electroencephalography [EEG] abnormalities occurring during sleep in patients with early AD.
Polysomnographic sleep acquisition was performed and included continuous recordings of EEG. Results
showed that relative to age-matched controls, AD patients presented faster mean theta frequency in both
REM sleep and slow wave sleep [SWS]. In AD patients, a correlative analysis revealed that faster theta
frequency during SWS was associated with better delayed episodic recall. Authors concluded that the
results suggest that changes in theta rhythm during REM and SWS are a relevant index of brain
impairments in the early stage of AD whereas sleep architecture is not, and that changes in theta activity
during SWS are related to episodic memory performance.
Sanchez-Espinosa et al105 examined associations between changes in physiological sleep and
plasma Aβ concentrations in 21 healthy old (HO) adults and 21 amnestic Mild Cognitive Impairment
(aMCI) subjects. They further assessed whether these two factors were associated with cortical volume
loss in each group. Sleep parameters were assessed with PSG. The investigators found significant
relationships between disrupted slow-wave sleep (SWS) and increased plasma levels of Aβ42 among
aMCI, but not HO subjects. They also found correlations between shortened rapid-eye movement (REM)
sleep in aMCI and thinning of the posterior cingulate, precuneus, and post-central gyrus; whereas higher
levels of Aβ40 and Aβ42 accounted for grey matter (GM) loss of posterior cingulate and entorhinal
cortex, respectively.
A.3 Cohort studies
Keage et al118 investigated the effect of self- reported sleeping behaviors and incidence of
cognitive impairment over two and 10 years in two healthy aged English populations. Napping behavior
was found to be negatively associated with incident cognitive impairment, with those who napped more
being less likely to be cognitively impaired at two years (OR = 0.38, 95% CI 0.22–0.67) and 10 years
(OR = 0.48, 95% CI 0.29–0.78). Short night time sleep duration (<= 6.5 h) increased the risk of incident
20
cognitive impairment over 10 years (OR = 2.02, 95% CI 1.17–3.48). The effect of long sleep duration (>=
9 h) failed to reach significance at both two and 10 years. The study did not account for relationships
between sleep and cognition with medication, cardiovascular factors, and metabolic diseases (other than
BMI). In addition, only cognitively healthy individuals were included at baseline so that the prognostic
effect of these sleep measures in those cognitively impaired or demented at baseline cannot be
extrapolated. Tworoger et al125 examined the association of sleep duration, snoring, and difficulty
sleeping with cognitive function in a cohort of community-dwelling women. Results showed that women
sleeping ≤5 hours/night scored worse than women sleeping 7 hours/night (mean difference on global
score combining all cognitive tests = 0.15 standard units, 95% CI: -0.28, -0.02). Women who regularly
had difficulty falling or staying asleep scored 0.11 units lower on the global score (95% CI: -0.22, 0.01)
compared with those who rarely had difficulty sleeping. These differences were equivalent to the mean
differences in score observed between participants who were 4 to 5 years apart in age. Authors found no
associations with snoring or with any of the sleep variables and cognitive decline over 2 years. In a
unique study, Virta et al126 investigated the association between midlife sleep characteristics (mean age
52 years) and late life cognitive function and dementia 18–26 years later in 2,336 members of the Finnish
Twin cohort who were at least 65 years of age, and found that reports of short (<7 h) and long (>8 h)
sleep duration, poor sleep quality, and use of hypnotic medications for more than 60 days per year were
associated with lower cognitive composite scores in old age; with long sleep being associated with
roughly 1.8 times the odds of developing AD. In this study, questionnaire data from 1981 were used in
the assessment of sleep characteristics, use of hypnotics, and covariates at baseline and between 1999 and
2007, and participants were assigned a linear cognitive score with a maximum score of 51 based on a
telephone interview (mean score 38.3, SD 6.1). Some of the associations differed by participant
characteristics, including Apolipoprotein E (APOE) e4 allele status, sex, and age, suggesting that poor
sleep may differentially affect cognition in different populations. The study revealed an association
between midlife sleep length and cognitive performance and AD based on medical records. With a mean
21
follow-up time of above 20 years, temporality can be assessed between sleep disturbance and cognitive
outcomes, thereby providing evidence of a potential causal link from poor sleep to cognitive decline.
Song et al122 examined the associations between sleep stage distributions and subsequent decline
in cognitive function in community-dwelling men aged 67 or older (n = 2,601) who were free of probable
dementia at the time of sleep evaluation. Follow-up averaged 3.4 years. Sleep stages were identified by
in-home polysomnography at the initial sleep visit. The average percent of time spent in stages of sleep
were as follows: N1, 6.7% ± 4.1%; N2, 62.6% ± 9.5%; N3, 11.3% ± 8.9%; and R, 19.4% ± 6.6%. The
authors found that increased time in Stage N1 and less time in Stage R were associated with worsening
cognitive performance in older men over time. Blackwell et al111 (2014) examined associations of
objectively- and subjectively-measured sleep with subsequent cognitive decline over 3.4 ± 0.5 y in 2,822
cognitively intact community-dwelling older men (mean baseline age 76.0 ± 5.3 y). Diminished
actigraphic sleep efficiency (i.e. spending a lower proportion of time in bed asleep), greater nighttime
wakefulness, greater number of long wake episodes, and poor self-reported sleep quality measured by a
score greater than 5 on the Pittsburgh Sleep Quality Index (PSQI) were all associated with subsequent
cognitive decline. However, there was no association between sleep duration and cognition.
Sterniczuk et al123 examined whether self-reported sleep problems in individuals who were not
demented at baseline, are early independent markers of risk for developing AD or dementia over an
approximately 4-year period. Data on four sleep variables (sleeping problems and fatigue in past 6
months, sleep medication use, and ‘recent trouble sleeping or a change in pattern’) were collected, and the
authors created a composite ‘sleep disturbance index’ from those items. The authors found a 23% greater
odds of dementia or Alzheimer’s disease among participants with higher scores on the sleep disturbance
index compared to those with lower scores after adjusting for age, sex, education, body mass index
(BMI), and cognition. Also, individual sleep items independently predicted AD or dementia and mortality
within ~4 years. Benito-León et al109 using a population-based study of 3,857 people without dementia
aged 65 years and older (NEDICES [Neurological Disorders in Central Spain]) examined whether long
sleep duration was associated with increased risk of dementia mortality. In a Cox model that adjusted for
22
numerous demographic factors and comorbidities, the hazard ratio for dementia mortality in long sleepers
was 1.63 (p = 0.03). Hahn et al116 examined a subjective report of change (reduced duration and/or
depth) in sleep pattern in relation to subsequent risk of incident all-cause dementia and Alzheimer disease
(AD) over a 9-year follow-up. Participants included 214 Swedish adults aged 75 and over who were
dementia-free both at baseline and at first follow-up (3 years later). Results showed that reduced sleep
was associated with a 75% increased all-cause dementia risk (hazard ratio: 1.75; 95% confidence interval:
1.04-2.93; Wald = 4.55, df = 1, p = 0.035) and double the risk of AD (hazard ratio: 2.01; 95% confidence
interval: 1.12-3.61; Wald = 5.47, df = 1, p = 0.019) after adjusting for age, gender, education, lifestyle and
vascular factors. However, results did not hold after adjusting for depressive symptoms, which was
measured with only one self-reported item. Findings are limited by the fact that the study did not include
objective longitudinal data on sleep. Bidzan et al110 found that patients in the preclinical period of
dementia were more likely to experience sleep disturbances and with greater intensity. The study was
conducted in Poland using 150 care centers residents age 55 and older. The presence of dementia was
excluded in the participants. After a follow-up period of 7 years, diagnosis of AD was established in 25
participants, while 111 people formed the control group.
Lim et al120 examined whether better sleep consolidation attenuates the relationship of the APOE
genotype to the risk of incident AD and the burden of AD pathology, in 698 community-dwelling older
adults without dementia (mean age, 81.7 years; 77% women) of the Rush Memory and Aging Project.
Participants were followed for up to 6 years. They found that better sleep consolidation attenuated the
effect of APOE genotype on incident AD and development of neurofibrillary tangle pathology, and
concluded that assessment of sleep consolidation may identify APOE+ individuals at high risk for
incident AD, and interventions to enhance sleep consolidation should be studied as potentially useful
means to reduce the risk of AD and development of neurofibrillary tangles in APOE ε4+ individuals.
A.4 Experimental studies
Mander et al142 determined whether regional brain atrophy, reduced slow wave activity (SWA)
during non–rapid eye movement (NREM) sleep and impaired long-term retention of episodic memories,
23
independently recognized features of aging, are inter-related. The authors found that “age-related medial
prefrontal cortex (mPFC) gray-matter atrophy was associated with reduced NREM SWA in older adults
and statistically mediated the impairment of overnight sleep–dependent memory retention. Moreover, this
memory impairment was further associated with persistent hippocampal activation and reduced task-
related hippocampal-prefrontal cortex functional connectivity, potentially representing impoverished
hippocampal-neocortical memory transformation”. Together, these data support a model in which age-
related mPFC atrophy diminishes SWA, the functional consequence of which is impaired long-term
memory. Such findings suggest that sleep disruption in the elderly, mediated by structural brain changes,
represents a contributing factor to age-related cognitive decline in later life. At a translational level, data
from this study endorses the possibility that improvements of SWA in older adults, through physiological,
behavioral or pharmacological means may represent a treatment target for minimizing the cognitive
decline associated with deficient long-term memory retention in later life.
Studies in AD mouse models support an association between sleep deprivation and AD
pathology. Kang et al138 in their study used in-vivo micro-dialysis in amyloid precursor protein (APP)
transgenic (Tg2576) mice (i.e. mice that overexpress mutant familial AD APP genes) and demonstrated
that both acute and chronic sleep deprivation alters brain interstitial fluid (ISF) Aβ diurnal rhythm. The
study also provided evidence of effects of hypocretin signaling on Aβ metabolism in studies of humans.
Circadian rhythm mean peak levels between 7 p.m. and 9 p.m. in human CSF Aβ levels were 28% higher
than the mean trough levels, which occurred between 9 a.m. and 11 a.m. A similar rhythm was found in
mice, with increased Aβ levels in brain interstitial fluid (ISF) while awake, compared to the sleep, both in
wild type C57BL6 mice and human APP transgenic mice, Tg2576. After being sleep deprived, mice spent
more time sleeping resulting in reductions in ISF Aβ levels. This suggests an association between
wakefulness, and increased ISF Aβ. Moreover, infusing the ventricle with a dual hypocretin receptor
antagonist resulted in suppressed ISF Aβ levels and the natural diurnal variation of Aβ was abolished. In
addition, hypocretin infusion led to significant decreases in ISF Aβ levels. The amount of ISF Aβ also
24
significantly increased during acute sleep deprivation and during orexin infusion in multiple sub-regions
of the cortex. Remarkably, systemic treatment with infusion of a dual orexin receptor antagonist once
daily for 8 weeks decreased the Aβ plaque formation in aged, sleep deprived, APP transgenic mice. This
finding suggests that Aβ diurnal rhythm is related to sleep/wake regulation and that sleep deprivation
increases Aβ levels and promotes subsequent Aβ plaque formation in APP transgenic mice. Rothman et
al146 compared male AD transgenic mice that were subjected to 6-hours/day sleep for 6-weeks to controls
(>7-hour/day sleep). Biochemical (immunoblot) outcomes were compared to mice undergoing daily cage
transfers (large cage control; LCC) as well as control mice that remained in their home cage (control;
CTL). Mice subjected to 6-hours/day sleep displayed a non-significant increase of ∼80% and a two- fold
increase of both Aβ and P-Tau levels respectively in the cortex (P <0.22). Significant positive correlations
between cortical Aβ and p -Tau levels and circulating corticosterone were seen and possibly indicate a
potential role for glucocorticoids in mediating behavioral and biochemical changes observed after sleep
restriction in a mouse model of AD. Di Meco et al136 evaluated the functional and biological
consequences on 3xTg mice that underwent 4 hours sleep restraint per day for 8 weeks. Starting at the age
of 8 months, mice were randomized to 2 groups, one underwent a 20/ 4-hour light and/or dark cycle for 2
months (4 males and 5 females), the other was kept on a normal light and/or dark cycle, as controls (5
males and 4 females). Compared to controls, behavioral assessment showed that sleep deprived mice had
a significant decline in their learning and memory. Differences in the levels of soluble amyloid-b peptides
between the groups were not observed, however reductions in tau phosphorylation, with a significant
increase in its insoluble fraction was observed. Sleep deprivation also resulted in lower levels of
postsynaptic density protein 95 and increased glial fibrillary acidic protein levels. Finally, although total
levels of the transcription factor cellular response element binding protein were unchanged, its
phosphorylated form was significantly diminished in brains of sleep-deprived mice when compared with
controls. The authors noted that their mice were relatively young in age and that the observed absence of
effect on the Aβ levels provides support for their hypothesis that sleep deprivation is not a simple
biomarker of the underlying neurodegenerative disease, but a direct contributor to its pathogenesis.
25
A.5 Randomized Controlled Trials
Ooms et al133 conducted an RCT to determine the effect of 1 night of total sleep deprivation on
cerebrospinal fluid Aβ42 protein levels in healthy middle-aged men. Participants were cognitively normal
middle-aged men (40-60 years of age) with normal sleep (n = 26) recruited from the local population.
Participants were randomized to 1 night with unrestricted sleep (n = 13) or 1 night of total sleep
deprivation (24 hours of wakefulness) (n = 13). Sleep was monitored using continuous polysomnographic
recording from 3 PM until 10 AM. Cerebrospinal fluid samples were collected using an intrathecal
catheter at defined times to compare cerebral Aβ42 concentrations between evening and morning. A night
of unrestricted sleep led to a 6% decrease in Aβ42 levels of 25.3 pg/mL (95% CI [0.94, 49.6], P = .04),
whereas sleep deprivation counteracted this decrease. A difference of 75.8 pg/mL of Aβ42 was shown
between the unrestricted sleep and sleep deprivation group (95% CI [3.4, 148.4], P = .04) when
accounting for the individual trajectories of Aβ42 over time. Individual evening and morning trajectories
of Aβ42 concentrations varied between the sleep deprivation and unrestricted sleep groups (P = .04) in
contrast to stable Aβ40, tau, and total protein levels.
B. Sleep wake-cycle/circadian rhythm and Cognitive decline and/or Alzheimer’s disease Dementia
Table 1.8 presents the results of the descriptive study characteristics, statistical methods and main
findings of studies that examined the association between sleep-wake /circadian rhythm abnormalities and
cognitive decline and/or AD. In total, there were 23 studies (i.e., 5 cross-sectional,87,89,91,95,102 4 case-
control,97,101,106,107 6 prospective cohort,108,118,119,124,127,129 7 experimental studies134,135,137,140,143-145 and 1
RCT131).
B.1 Cross-sectional studies
Oosterman et al89 examined whether nocturnal sleep and the circadian organization of the sleep-
wake rhythm had predictive value for cognitive functioning in home-dwelling elderly people and found
26
an association between the rest-activity rhythm and cognitive performance independent of age. Rest-
activity was assessed using actigraphy, and mental speed, memory, and executive function were evaluated
using a battery of standardized cognitive tests. The authors noted that majority of the subjects suffered
from at least one cardiovascular risk factor. Merlino et al87 examined the relationship between disturbed
sleep and cognitive impairment in the elderly and found that among sleep disturbances, excessive daytime
sleepiness was independently associated with the presence of dementia. In addition, the frequency of
excessive daytime sleepiness increased progressively across the different categories of cognitive decline,
as measured by means of MMSE and Global Deterioration Scale (GDS)179 scores.
Pat-Horenczyk et al91 examined differences in hour-by-hour sleep/wakefulness profiles between
severely and mild to moderately demented patients and assessed how many elderly patients remained
almost fully asleep or nearly fully awake in each hour of a 24-hour period. They found that patients with
mild to moderate dementia showed a disproportionate amount of wakefulness during the night, in contrast
to patients with severe dementia showing a disproportionate amount of sleepiness during the day. They
also found that with the progression of dementia both the capacity to maintain sleep and the capacity to
maintain wakefulness were impaired, resulting in almost complete fragmentation of sleep/wakefulness
during the night and day. Sleep/wakefulness patterns were recorded using actigraphy and levels of
dementia were assessed by the Mini-Mental State Examination (MMSE). This study suggests that sleep
disturbances parallel the severity of dementia\. However, patients were not evaluated for type of
dementia nor were records available to check whether patients had Alzheimer’s disease or dementia of
other types. Yesavage et al95 examined indices of sleep/wake function collected from individuals with
AD in relation to relevant polymorphisms in circadian clock-related genes in two cohorts of AD
participants. [Cohort 1 (n=124): individuals with probable AD recruited from the Stanford/Veterans
Affairs NIA Alzheimer’s Disease Core Center (n=81) and the Memory Disorders Clinic at the University
of Nice School of Medicine (n=43). Cohort 2 (n=176): individuals with probable AD derived from the
Alzheimer’s Disease Neuroimaging Initiative (ADNI) data set]. Adjusting for multiple tests, none of the
27
candidate gene SNPs were significantly associated with the amount of wake after sleep onset, a marker of
sleep consolidation, and no relationships likely to be of clinical relevance were detected. The authors
therefore concluded that it was unlikely that a relationship with a clinically meaningful correlation exists
between the circadian rhythm-associated SNPs and WASO in individuals with AD.
A pacemaker located in the suprachiasmatic nucleus (SCN) which serves as a master clock180,181
organizes circadian regulation of sleep and wakefulness mechanisms. Melatonin, N-acetyl-5-
methoxytryptamine, is one of the most important timing signals generated by the SCN and is mostly
known for its relation to circadian rhythms and sleep. Naismith et al102 examined whether patients with
mild cognitive impairment (MCI) had significant alterations in the timing of melatonin secretion onset
and amount as well as changes in sleep architecture. They found that patients with MCI had advanced
timing of their melatonin secretion onset relative to controls, but the levels of melatonin secreted did not
differ between groups. The MCI group also had greater wake after sleep onset (WASO) and increased
rapid eye movement sleep latency. There were differential associations between dim light melatonin onset
and cognition between the two groups, with earlier dim light melatonin onset being associated with poorer
memory performance in MCI patients. Participants’ sleep-wake behavior for the 14-nights prior to
commencing the in-laboratory portion of the protocol was assessed using actigraphy and sleep diaries.
Assessment of circadian function (melatonin rhythm timing and levels) and sleep architecture were
examined using melatonin and polysomnographic (PSG) assessments on separate nights.
B.2 Case-control studies
Two studies conducted in the Netherlands investigated the circadian rhythm of hypocretin-1
(Hypocretin-1 and -2 are hypothalamic neuropeptides, also known as orexins, that form an important
regulating system acting on the sleep–wake cycle182,183) in Alzheimer’s disease. Fronczek et al97
assessed whether the neurodegenerative process of AD affects hypothalamic hypocretin/orexin neurons.
Investigators quantified the total number of hypocretin-1 immuno-reactive neurons in postmortem
hypothalami of AD patients (n = 10) and matched controls (n = 10) and measured hypocretin-1
28
concentrations in postmortem ventricular cerebrospinal fluid (CSF) of 24 AD patients and 25 controls.
The numbers and levels of hypocretin-1 immunoreactive neurons and cerebrospinal fluid (CSF)
hypocretin-1 respectively, were significantly lower (i.e., 40% decreased cell number, and 14% lower CSF
hypocretin-1 levels respectively) in AD patients compared to controls. (p = 0.038). These findings
indicate that the hypocretin system is affected in advanced AD. Slats et al107 examined the association
between Hypocretin-1 and Amyloid-β42 CSF levels in 6 patients with AD (59-85 yrs., MMSE 16-26) and
6 healthy volunteers (64-77 yrs.). The authors could show no difference in CSF hypocretin-1 circadian
rhythm between AD and controls. However, lower mean CSF Aβ42 levels were significantly associated
with lower hypocretin-1 levels. The authors postulated that this finding might suggest a relationship
between AD pathology and hypocretin disturbance, which could hold possibilities for treatment of AD
related sleep disorders.107
Schmidt et al106 investigated cerebrospinal fluid (CSF) levels of melanin-concentrating hormone
(MCH) and hypocretin-1 (HCRT-1, orexin-A) in patients with Alzheimer’s disease (AD) and Healthy
subjects (HS) and the potentially causal relationship between these polypeptides and CSF levels of the
AD marker total Tau (T-tau), hyper phosphorylated tau (P-tau) and Ab42, cognitive performance and
behavioral symptoms in AD. The study was carried out in 33 patients with mild to severe AD and 33
subjects without any psychiatric or neurological disorder (HS). A significant main effect of diagnosis
(p=0.01) on MCH levels was found between AD and HS. MCH correlated with T-tau (p=0.01) and P-tau
(p=0.05) in the AD but not in the HS. CSF-MCH correlated negatively with MMSE scores in the AD
(p=0.05) and was increased in more severely affected patients (MMSE≤20) compared to HS (p=0.001)
and BPSD-positive patients compared to HS (p=0.05). For CSF-HCRT- 1, a significant main effect of sex
(p=0.05) with elevated levels in females was found whereas diagnosis and the sex*diagnosis interaction
were not significant. These results should stimulate additional research on the role of the hypothalamus in
AD pathogenesis and development of behavioral disturbances in AD. Future studies should include
repeated MCH/HCRT-1 level measurements in combination with sleep-wake-actigraphy and assessment
of sleep by standardized questionnaires, histochemical determinations of MCH in animal and post-
29
mortem tissues. Liu et al101 analyzed data from 85 patients with AD (mean age, 75 ± 1.1 yr) and 82 age-
matched controls (mean age, 76 ± 1.4 yr), to determine melatonin levels in postmortem cerebrospinal
fluid in relation to aging, AD, and APOE ε4 genotype. They found that melatonin levels in AD patients
expressing APOE e3/4 was significantly higher than that in patients expressing APOE e4/4 (p = 0.02).
CSF melatonin levels were significantly lower and about half of those in control subjects (p=.00126).
B.3 Cohort studies
Walsh et al127 using a study cohort of 1,287 community-dwelling older women (82.8 ± 3.1 y)
who were free of dementia at the baseline visit and were followed for 5 years, found that weaker circadian
activity rhythms (CAR) patterns were associated with lower cognitive function, especially poorer
executive function. Keage et al118 found that daytime sleepiness was significantly associated with
incident cognitive impairment over 10 years (OR = 2.43, 95% CI 1.24–4.75). Anderson et al108 also
examined the association between subjective and objective measures of sleep and wakefulness and other
health parameters in a 421 men and women, aged 87–89. The study was nested in the Newcastle 85+
Study, a population-based longitudinal study of health and ageing in the very old. Sleep-wake parameters
were assessed subjectively with the Pittsburgh Sleep Questionnaire Inventory (PSQI) and the Epworth
Sleepiness Score (ESS), while wrist actigraphy was used as an objective measure. The authors found that
impaired sleep–wake cycles were significantly associated with cognitive impairment, disability,
depression, increased number of falls, body mass index and arthritis but not with any other specific
disease markers or decreased survival. The subjective sleep questionnaires did not differentiate well
between objectively determined disturbance of sleep–wake cycles and those with normal cycles. Tranah
et al124 studied whether measured circadian activity rhythms are prospectively associated with incident
dementia or mild cognitive impairment (MCI) in 1,282 community-dwelling older women. They found
that older women with decreased circadian activity rhythm amplitude and robustness and delayed rhythms
had a higher likelihood of developing dementia or MCI when comparing those in the lowest quartile of
amplitude (OR =1.57, 95% CI, 1.09–2.25) or rhythm robustness (OR=1.57, 95%CI, 1.10–2.26) compared
30
to women in the highest quartile. An increased risk of dementia or MCI (OR=1.83, 95% CI, 1.29–2.61)
also was found for women whose timing of peak activity occurred later in the day (after 3:51PM) when
compared to those with average timing (1:34PM–3:51PM). Lim et al119 tested the hypothesis that sleep
fragmentation is associated with incident Alzheimer’s disease (AD) and the rate of cognitive decline
among 737 older adults without dementia at baseline, and reported that those with significant actigraphic
sleep fragmentation (in the 90th percentile) had a 1.5-fold increased risk of developing AD as compared
with those with low sleep fragmentation (10th percentile). A Cox regression model controlling for age,
sex, and education, also revealed that a higher level of sleep fragmentation was associated with an
increased risk of incident AD (HR = 1.22, 95%CI 1.03-1.44, P = 0.02 per 1 SD increase in sleep
fragmentation). In an earlier prospective study Yesavage et al129 determined if sleep parameters in AD
patients change over time as a function of APOE carrier status in 44 subjects, of which 25 were APOE ε4
carriers and 19 were non-ε4 carriers. Participants had been assessed at least during 2 stages of cognitive
impairment, defined as a decline from one stage of AD to another. They found that among APOE ε4
carriers, wake after sleep onset was associated with lower cognitive function; for non-ε4 subjects,
increases in WASO and declines in total sleep time, sleep efficiency, and the amplitude of the rest/activity
circadian rhythm over time were associated with lower cognitive function.
B.4 Experimental studies
Craig et al134 explored the effects of acute and chronic disruption of circadian rhythms on
memory using a phase shifting schedule that can continually challenge the rats’ circadian system by using
repeated phase shifts and recovery sessions. Twenty-four male Long Evans hooded rats (300–400 g)
obtained from Charles River (Saint-Constant, PQ) were used for all studies. They were divided into three
groups: acutely phase shifted rats (acute; n = 8), chronically phase shifted rats (chronic; n = 8), control
rats for water maze testing (control; n=7) and control rats for fear conditioning (control; n = 14). The
authors demonstrated a significant learning and memory deficit on a spatial version of the water maze
task (used for behavioral testing in the animals) in the chronically phase shifted, but not in the acutely
31
phase shifted animals. Moreover, they did not find any impairment in fear conditioning suggesting that
chronic phase shifting predominantly affects hippocampal memory. They proposed that chronic circadian
disruption may play a role in the development of age-related cognitive deficits and dementia in the
elderly. Jyoti et al137 characterized sleep, electroencephalogram (EEG) disturbances and cognitive
dysfunction in transgenic mice carrying transgenes for amyloid-protein precursor (APPswe) and
presenilin 1 (PSEN1A246E) at 5 (pre-plaque) and 20 months, relative to PSEN1 and wild-type (WT)
mice, using a novel wireless microchip device. The study findings suggested that APP/PSEN1 mice
exhibit abnormalities in activity and sleep architecture preceding amyloid plaque deposition as well as
age-related changes in cortical EEG power. Though not fully recapitulating the profile of AD patients,
this suggests activity and EEG recordings are sensitive and translational biomarkers in murine models.
Several anti-amyloidogenic effects of melatonin have been demonstrated in vitro and in animal
studies. Pappolla et al144 examined the interactive ability of melatonin with Aβ1–40 and Aβ1–42 to
determine whether melatonin inhibits the progressive formation of β-sheets and amyloid fibrils. These
investigators found that melatonin crosses the blood-brain barrier, is relatively devoid of toxicity, and
concluded that it may constitute a potential new therapeutic agent in Alzheimer’s disease. Olcese et al143
examined the potential for long-term melatonin treatment to protect Alzheimer’s transgenic mice against
cognitive impairment and development of β-amyloid (Aβ) neuropathology. Authors administered
melatonin (100 mg/L drinking water) to AD transgenic (Tg) mice from 2–2.5 months of age to their
killing at age 7.5 months. Findings revealed that melatonin suppresses Aβ aggregation in brain
homogenates. Inflammatory cytokines such as tumor necrosis factor (TNF)-α were decreased in
hippocampus (but not plasma) of Tg+ melatonin mice. Finally, the cortical mRNA expression of three
antioxidant enzymes (SOD-1, glutathione peroxidase, and catalase) was significantly reduced to non-Tg
levels by long-term melatonin treatment in Tg mice. Lahiri et al140 examined the effects of dietary
melatonin on brain levels of nitric oxide synthase, synaptic proteins and amyloid beta-peptides (Aβ) in
mice. They found that dietary melatonin supplementation led to a significant reduction in levels of toxic
32
cortical Aβ of both short and long forms that are involved in amyloid deposition and plaque formation in
Alzheimer's diseases. However, melatonin supplements did not significantly change cerebral cortical
levels of nitric oxide synthase or synaptic proteins such as synaptophysin and SNAP-25.
Deregulation of protein kinases/protein phosphatases is responsible for tau hyper
phosphorylation. Among them, glycogen synthase kinase-3 (GSK-3) is a known tau kinase that plays a
crucial role in AD pathology.184 In-vivo and in-vitro tau phosphorylation can be undertaken by GSK-3
and over-expression of GSK-3 leads to tau hyper phosphorylation.185,186 In an Alzheimer’s-affected brain,
GSK-3 is activated in pretangle neurons and accumulates in paired helical filaments.187 Under the above
premise, Deng et al135 carried out an in-vitro study on N2a (Neuro-2a: a fast growing neuroblastoma cell
line) cells that were grown and differentiated in 96-well culture plates at density of 1.5×105 cells in 100
μL. Cells were exposed to various concentrations of wortmannin (an indirect GSK-3 activator) for 2 h at
37 ºC in the presence or absence of a 12-h-preincubation with melatonin 25, 50, and 100 μmol/L or
Vitamin E (VE). The authors explored the underlying mechanism of tau hyper phosphorylation in an
Alzheimer's affected brain and the possible arresting strategies. One of the findings of the study was that
preincubation of the cells with melatonin or vitamin E attenuated differentially wortmannin induced
oxidative stress as well as GSK-3 over-activation and tau hyper phosphorylation. The authors concluded
that wortmannin might be an effective tool for reproducing AD-like tau hyper phosphorylation cell model
and that melatonin/vitamin E can effectively protect the cells from wortmannin-induced impairments.
Contrary to the above findings on the possible therapeutic value of melatonin on cortical Aβ level
reduction, Quinn et al145 examined the effect of chronic melatonin therapy initiated after the appearance
of plaques in Tg2576 mice and failed to demonstrate any effect, as chronic melatonin therapy did not alter
amyloid burden or oxidative damage after 4-months of treatment. Variations in melatonin dose, age of
treatment initiation, and species of Aβ measured could possibly have accounted for the different results.
Quinn et al. noted that one of the prior studies used a lower dose and still found an effect at this dose
could not have accounted for the different result. The age at treatment initiation varied across the studies
and might possibly explain the different results. Another study188 using similar Tg2576 mice initiated
33
treatment at 4 months of age (prior to the appearance of hippocampal and cortical plaques) also found
diminished Aβ and nitrotyrosine at 7–8 months, but no significant difference in histologically detectable
Aβ or brain nitrotyrosine in animals at 15 months of age. The Quinn et al study initiated treatment at 14
months.
B.5 Randomized Controlled Trial
Asayama et al131 conducted a double blind study to examine the effects of melatonin on the
sleep-wake rhythm, cognitive and non-cognitive functions in Alzheimer type of dementia. The subjects
were 9 persons given a placebo (PLA), and 11 given melatonin (3 mg) (MLT). The drugs were given at
8:30pm each day for 4 weeks. Melatonin administration improved sleep time and night activity
significantly, but had no significant effect with regard to improving daytime naps and activity. Melatonin
administration improved cognitive and non-cognitive functions significantly. Melatonin seemed to be
useful for care of the Alzheimer type of dementia patients
C. Hypoxia (Apnea Hypopnea Index) and Cognitive decline and/or Alzheimer’s disease
Sixteen studies (i.e., 4 cross-sectional,81,83,86,90 1 case-control,103 3 prospective cohort,125,128,150 1
retrospective cohort,112 2 RCTs,130,132 and 5 experimental studies139,141,147-149) examined hypoxia states
(AHI) in sleep apnea/sleep disordered breathing and its association with cognitive decline and/or AD.
Table 1.9 presents the results of the descriptive study characteristics, statistical methods and main
findings of the studies.
C.1 Cross-sectional studies
Kadotani et al86 conducted a cross-sectional study to determine whether genetic variation at the
level of APOE is associated with Sleep Disordered-Breathing (SDB) in a probability based sampling of
791 middle-aged adults, mean age of 49 and age range 32-68. They found that independent of age, sex,
body mass index, and ethnicity, participants with APOE epsilon4 had a significantly higher risk of
moderate-to-severe Sleep Disordered-Breathing (apnea-hypopnea index (AHI) ≥15%) (P =.003).
34
Participants with APOE epsilon4 also had a significantly higher mean (SEM) apnea-hypopnea index
(p=.01). These effects increased with the number of APOE epsilon4 alleles carried. In a similar study,
Gottlieb et al83 examined the association of the APOE e4 allele with Obstructive Sleep Apnea/Hypopnea
(OSAH) in a community-dwelling cohort, exploring age dependency of the association. The investigators
conducted a genetic association study, nested within a prospective cohort study of the cardiovascular
consequences of OSAH. APOE e4 allele was associated with increased odds of OSAH (OR 1.41, 95% CI
1.06 to 1.87, p = 0.02) after controlling for age, sex and body mass index (BMI). The effect was greater in
subjects < 75 (OR 1.61, CI 1.02 to 2.54) as in those ≥75 years old (OR 1.32, CI 0.91 to 1.90). Exploratory
analyses revealed that the strongest effect of APOE e4 was in subjects age <65 (OR 3.08, CI 1.43 to
6.64), and was stronger in those with hypertension or cardiovascular disease than in those without. The
authors therefore concluded that the APOE e4 allele is associated with increased risk of OSAH,
particularly in individuals under age 65.83 Osorio et al90 examined the interaction between sleep-
disordered breathing and Apolipoprotein E genotype on cerebrospinal fluid biomarkers for Alzheimer’s
disease in a total of 95 cognitively normal elderly participants who were analyzed for SDB severity.
Among the 95 participants, 25 were considered free of SDB (apnea and/or hypopnea with 4% O2-
desaturation index (AHI4%) < 5) and were included as normal controls, 51 had mild SDB (AHI4% 5-
14.99), and 19 had moderate to severe SDB (AHI4% > 15). Results showed that in ApoE3+ subjects,
significant differences were found between sleep groups for p-tau and t-tau (p=0.017, p = 0.043
respectively), and the AHI4% was significantly correlated with p-tau (p = 0.023), t-tau (p = 0.021), and
Aβ-42 (p = 0.021). In ApoE2+ subjects, the AHI4% was significantly correlated with lower levels of CSF
Aβ-42 (p = 0.004), identical to ApoE4+ subjects where there was also a trend toward lower CSF Aβ-42
levels.
Lastly, Djonlagic et al81 examined the effect of aging on sleep-dependent motor memory
consolidation in patients with and without OSA. Participants included 44 patients (19–68 years) who had
been referred by a physician for a baseline polysomnography (PSG) evaluation and were assigned either
35
to the OSA group (AHI>5/h), or control (Non-OSA) group (AHI<5/h). All subjects performed the
Psychomotor Vigilance Task (PVT) and the Motor Sequence Learning Task (MST) in the evening and
again in the morning after their PSG. Results showed that healthy sleepers displayed a greater degree of
sleep-dependent overnight improvement on the MST, an effect not diminished by increasing age. The
existence of untreated obstructive sleep apnea was associated with an aging-related cognitive deficit,
otherwise not present in individuals without OSA. Other research has linked OSA to a greater risk of
developing dementia. Future studies are therefore necessary to determine if the restriction of memory
consolidation is tied to or has a role to play in the onset of neurodegenerative disease.
C.2 Case-control studies
O’Hara et al103 investigated the relationship between obstructive sleep apnea/hypopnea (OSAH)
and cognition in 36 older adults, i.e., 18 APOE epsilon4 carriers, and 18 non-carriers. These authors
found that higher levels of AHI were associated with lower memory scores in the APOE epsilon4 carriers
only. Minimum oxygen saturation (MinSaO2) did not differ significantly between the groups, but was
lower than 90% in both, indicative of the occurrence of hypoxia. MinSaO2 was not associated with
cognition, although there was a non-significant trend for minSaO2 to negatively impact memory in the
APOE epsilon4 carriers only.
C.3 Cohort studies
Cohen-Zion et al150 examined changes in cognitive function associated with Sleep Disordered
Breathing (SDB) in 46 participants aged 65 and older who completed visits 3 and 4. The original cohort
included community-dwelling elders age 65 and older with high risk for SDB studied from 1981 through
1985 and then followed every 2 years. The study results suggest that declining cognitive function is
associated primarily with increases in daytime sleepiness. Although cognitive decline was also associated
with increases in respiratory disturbance index (RDI: defined as number of apneas plus hypopneas per
hour of sleep), this association did not hold in a more inclusive model which also included variable of
36
SDB, oximetry, sleep and subjective report. The findings could suggest that any relationship between
SDB and cognitive function may be mediated by the effect of SDB on daytime sleepiness. Tworoger et
al125 found no associations with snoring, sleep duration, and difficulty sleeping and cognitive decline over
2 years. Chang et al112 in a retrospective matched-control cohort study estimated and compared the risk
of dementia in Sleep Apnea (SA) and non-SA patients among persons aged 40 and above over a 5-year
follow-up period. Overall findings suggest that SA patients had a 70% greater risk of developing
dementia within 5 years of diagnosis compared to non-SA age- and sex-matched patients, after adjusting
for other risk factors (95% confidence interval (CI) = 1.26-2.31; P < .01).
Yaffe et al128 examined the prospective relationship between sleep disordered breathing (SDB),
and cognitive impairment and investigated potential mechanisms of this association. Using data from 298
women without dementia (mean [SD] age: 82.3 [3.2] years), the authors found that women with SDB had
an 85% increased risk of developing MCI or dementia compared to women without SDB after adjusting
for age, race, body mass index, education, smoking status, presence of diabetes, presence of hypertension,
medication use (antidepressants, benzodiazepines, or non- benzodiazepine anxiolytics) and baseline
cognitive scores (adjusted OR=1.85, 95% CI 1.11-3.08). Measures of disordered breathing i.e., elevated
oxygen desaturation index (≥15 events/hour) and high percentage (>7%) of sleep time in apnea or
hypopnea, were both associated with risk of developing MCI or dementia (adjusted OR=1.71, 95% CI
1.04 − 2.83 and adjusted OR=2.04, 95% CI 1.10 − 3.78, respectively). Sleep fragmentation measures
(arousal index and wake after sleep onset) or sleep duration (total sleep time) were not associated with
risk of cognitive impairment.
C.4 Experimental studies
Shiota et al147 examined the direct effect of intermittent hypoxia (IH) in pathophysiology of AD
in-vivo and in-vitro. In-vivo, 15 male triple transgenic AD mice were exposed to either chronic
intermittent hypoxia (CIH) or normoxia (5% O2 and 21% O2 every 10 min, 8 h/day for 4 weeks). In-
vitro, human neuroblastoma SH-SY5Y cells stably expressing wild-type amyloid-β protein precursor
were exposed to either IH (8 cycles of 1% O2 for 10 min followed by 21% O2 for 20 min) or normoxia.
37
Results showed that CIH significantly increased levels of Aβ42 but not Aβ40 in the brains of mice
without the increase in hypoxia-inducible factor 1, alpha subunit (HIF-1α) expression. Furthermore, CIH
significantly increased intracellular Aβ-42 in the brain cortex. There were no significant changes in
cognitive function. IH significantly increased levels of Aβ42 in the medium of SH-SY5Y cells without
the increase in the HIF-1α expression. CIH directly and selectively increased levels of Aβ42 in the AD
model. Kaushal et al139 hypothesized that acute short-term Intermittent hypoxia (IH) and sleep
fragmentation (SF) (major manifestations of sleep apnea), as well as their combination (IH+SF) may
reveal unique susceptibility in sleep integrity in a murine model of AD. They found that slow wave sleep
(SWS) was significantly decreased, and rapid eye movement (REM) sleep was almost suppressed during
acute exposure to IH alone and IH+SF for 6 h in hApoE4 animals, with milder effects in WT controls.
Reduced delta power during SWS did not show post exposure rebound in hApoE4 unlike WT controls. IH
and IH+SF induced hypothermia was more prominent in hApoE4 than wild type (WT) controls. Mice
subjected to SF also showed sleep deficits but without hypothermia. “hApoE4 mice, unlike WT controls,
exhibited increased sleep propensity, especially following IH and IH+SF, suggesting limited ability for
sleep recovery in hApoE4 mice.” These findings affirm the probable impact of IH and SF in adjusting
sleep architecture and sleep homeostasis including sustenance of body temperature. In addition, the
increased susceptibility and defined recovery capability of hApoE4 mice to sleep apnea suggests that
early identification and treatment of sleep apnea in AD patients may permit interventions to slow AD
progression.
Zhang et al149 investigated the possibility of a direct molecular link between hypoxic insults and
Amyloid Precursor Protein (APP) processing and found that acute hypoxia increased the expression and
the enzymatic activity of β-secretase (BACE1) (Aβ is derived from the β-amyloid precursor protein
(APP) by sequential proteolytic cleavages from β-secretase (BACE1) and presenilin-1 (PS1)/γ-secretase)
by up-regulating the level of BACE1 mRNA, resulting in increases in the APP C-terminal fragment-β
(βCTF) and Aβ. Hypoxia had no effect on the level of PS1, APP, and tumor necrosis factor--converting
38
enzyme (TACE, an enzyme known to cleave APP at the α-secretase cleavage site). Sequence analysis,
mutagenesis, and gel shift studies revealed binding of Hypoxia Inducible Factor-1 (HIF-1) to the BACE1
promoter. Overexpression of HIF-1 α increased BACE1 mRNA and protein level, whereas down-
regulation of HIF-1 α reduced the level of BACE1. Hypoxic treatment failed to further potentiate the
stimulatory effect of HIF-1α overexpression on BACE1 expression, suggesting that hypoxic induction of
BACE1 expression is primarily mediated by HIF-1α. Lastly, the authors observed significant reduction in
BACE1 protein levels in the hippocampus and the cortex of HIF-1α conditional knock-out mice. These
results provide evidence that hypoxia/HIF-1α plays a vital role in balancing the amyloidogenic processing
of APP. It also provides a molecular mechanism to explain increased incidence of AD following cerebral
ischemic and stroke injuries. Following in the same line of thought, Li et al141 investigated molecular
mechanisms underlying hypoxia mediated AD using AD transgenic mice. The authors demonstrated that
repeated hypoxia increased β-amyloid (Aβ) generation and neuritic plaque formation by elevating β-
cleavage of APP in APPswe + PS1A246E transgenic mice. They also found that hypoxia enhanced the
expression of APH-1a, a component of ɣ-secretase complex, which in turn may lead to increase in ɣ-
cleavage activity. Furthermore, these authors demonstrated that repeated hypoxia treatment can activate
macroautophagy, which may contribute to the increases in Aβ production. Taken together, the results
suggest an important role of hypoxia in modulating the APP processing by facilitating both β- and ɣ-
cleavage which may result in a significant increase of Aβ generation. Sun et al148 constructed a series of
deletion luciferase reporter plasmids containing human BACE1 gene promoter regions to investigate
whether BACE1 is one of the downstream target genes of the hypoxia signaling pathway. In the study,
authors found that hypoxia significantly increased BACE1 gene expression, resulting in increased β-
secretase activity and Aβ production. Furthermore, hypoxia treatment markedly increased Aβ deposition
and potentiated the memory deficit in AD transgenic mice. The results demonstrate that hypoxia can
facilitate AD pathogenesis, and they provide a molecular mechanism to link vascular factors to AD.
39
C.5 Randomized Controlled Trials
Ancoli-Israel et al130 examined whether treatment of obstructive sleep apnea (OSA) with
continuous positive airway pressure (CPAP) in patients with Alzheimer's disease (AD) would result in
improved cognitive function. Participants were randomized to either therapeutic CPAP for six weeks or
placebo CPAP for three weeks followed by therapeutic CPAP for three weeks. A comparison of subjects
randomized to 3 weeks of therapeutic versus placebo CPAP suggested no significant improvements in
cognition. A comparison of pre- versus post-treatment neuropsychological test scores after 3 weeks of
therapeutic CPAP in both groups showed a significant improvement in cognition. Similarly, Kushida et
al132 determined the neurocognitive effects of continuous positive airway pressure (CPAP) therapy on
patients with obstructive sleep apnea (OSA) in a 6-month, randomized, double-blind, 2-arm, sham-
controlled, multicenter trial. The results showed that CPAP treatment improved both subjectively and
objectively measured sleepiness, especially in individuals with severe OSA (AHI >30). CPAP use also
resulted in mild, transient improvement in the most sensitive measures of executive and frontal-lobe
function for those with severe disease, which suggests the existence of a complex OSA-neurocognitive
relationship.
D. Insomnia and cognitive decline and/or Alzheimer’s disease
Six studies (i.e., 2 cross-sectional87,92 and 4 prospective cohort studies114,115,117,121) examined the
association between insomnia and cognitive decline and/or AD. Table 1.10 presents the results of the
descriptive study characteristics, statistical methods and main findings of the studies.
D.1 Cross-sectional studies
Pistacchi et al92 in their study found that patients with MCI had a frequency of sleep disturbances
of any type equal to that of patients with AD and presented mostly with insomnia. In the study insomnia
occurred more frequently in AD. In their cross-sectional study examining the association between
multiple sleep parameters including insomnia and cognitive impairment in the elderly, Merlino et al87 did
not find an association between insomnia, the most common sleep complaint in their sample, and the
40
presence of cognitive decline. These studies assessed the presence of sleep disorders by means of
subjective symptoms and not using objective instruments, such as wrist actigraphy or polysomnography.
Polysomnography is essential for a correct diagnosis of some sleep disturbances. In particular, sleep
disordered breathing can be diagnosed only by means of polysomnographic recordings. Because
assessment of sleep parameters was based on a structured interview, obtaining self-reported description of
sleep characteristics from subjects with significant cognitive impairment becomes challenging and
questionable.
D.2 Cohort studies
Examining whether self-reported symptoms of insomnia independently increase risk of cognitive
decline in older adults, Osorio et al121 and Cricco et al114 found that there is a greater risk of Alzheimer's
disease in adults with self-reported insomnia. Osorio et al121 examined the relationship between insomnia
and AD in 346 cognitively normal older adults evaluated during multiple visits over 7.7 years of follow-
up. Insomnia was collected only during the normal cognition stage and screened using items 4, 5, and 6
from the Hamilton Depression Rating Scale (HAM-D). Participants who reported insomnia had an
increased odds of 2.39 of developing AD when compared to the non-insomnia group. (OR = 2.39, 95%
confidence interval (CI) = 1.03–5.55). Depression did not confound the association, and individuals with
insomnia showed a faster progression to dementia (chi-square = 3.94, P = .047). Cricco et al114 found in
their study sample that self-reported symptoms of insomnia independently predicted incident cognitive
decline in older men, but not in women. Compared to those with no insomnia at the third annual follow
up (FU3) those reporting symptoms of insomnia at both baseline and FU3 had an adjusted odds ratio
(AOR) of 1.49 (95% CI = 1.03–2.14) for cognitive decline among non-depressed men. Men with
insomnia at FU3 only were not at increased risk (OR = 1.16, 95% CI = 0.82–1.65). These relationships
were not found in women. Men and women with depressive symptoms at FU3 were at increased risk for
cognitive decline independent of insomnia. Participants included over 6,000 community-dwelling men
and women age 65 and older from the prospective cohort of the four sites of the Established Populations
41
for Epidemiologic Studies of the Elderly (EPESE) in Iceland. The authors noted that sleep complaints
data were only available at 3-year intervals, making misclassification of exposure in some cases more
likely, e.g., some participants classified as experiencing chronic sleep disturbances may have had
transient episodes of sleep difficulty during follow-up but may have had undisturbed sleep over the period
between follow-ups. Likewise, incident cases could have had symptoms for up to 3 years. Thus, the odds
ratios found in this study can be considered conservative estimates of risk.
Using the Maastricht Ageing Study (MAAS) cohort in Netherlands, Jelicic et al117 examined
whether subjective sleep complaints (i.e., extent to which an individual was distressed by difficulties with
falling asleep, waking up too early and by a restless or disturbed sleep) in a population-based sample of
838 middle aged and older adults (≥ 50 years) predicted cognitive decline over a period of 3 years.
Results showed that subjective sleep complaints were negatively associated with cognitive performance at
follow-up (β = -0.05 95% CI: -0.09 to -0.00, p = 0.03). Foley et al115 examined the longitudinal
association between sleep disturbances (insomnia and daytime sleepiness) and incidence of dementia and
cognitive decline in 2346 Japanese-American men aged 71 to 93 in the Honolulu-Asia Aging Study
(HAAS). Sleep disturbances were assessed by a questionnaire that asked whether participants were sleepy
during the day or had insomnia (defined as “usually” having trouble falling asleep or waking up far too
early and not going back to sleep). Individuals were screened for possible dementia using the CASI.
Dementia was diagnosed according to less-restrictive diagnostic criteria developed by Cummings et al.189
for very mild dementia and more-restrictive diagnostic criteria for mild or more severe dementia from the
Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised (DSM-IIIR).190 Results
from the study showed that participants with excessive daytime sleepiness at baseline (8%) were twice as
likely to be diagnosed with incident dementia than were those not reporting daytime sleepiness [adjusted
odds ratio (OR) = 2.19, 95% confidence interval (CI) = 1.37–3.50] and about 40% more likely to have a 9
or more point drop in their CASI score between examinations [OR = 1.44, 95% CI =1.01–2.08]. Insomnia
was not associated with cognitive decline or incidence of dementia. The authors noted that there was lack
42
of information to identify persons with sleep disordered breathing, which has been shown to affect
cognitive performance191,192 and has a 25% prevalence in older persons.193
E. Restless Leg Syndrome and cognitive decline and/or Alzheimer’s disease
Table 1.11 presents the results of the descriptive study characteristics, statistical methods and main
findings of the 3 studies (i.e., 2 cross-sectional84,92 and 1 case-control104) that examined any association
between restless leg syndrome and cognitive decline and/or AD.
E.1 Cross-sectional
Restless legs syndrome (RLS) can inhibit falling asleep or being able to stay asleep. RLS
incidence is higher in women and increases with age.194 Depression, antidepressant drug intake, renal
failure, and peripheral neuropathy, are known risk factors for RLS and also increase with age. Guarnieri
et al84 reported a 6% prevalence of RLS in MCI and Demented (AD/non-AD patients) assessed by the
international RLS study group155,195 questionnaire. However, in cognitively impaired individuals,
assessing RLS by use of a questionnaire might be arduous. Another study196 reported an RLS prevalence
of 24% in 59 demented individuals using specific guidelines of probable RLS that emphasizes behavioral
indicators and supportive features such as dopaminergic responsiveness, patient’s past history, and
presence of periodic limb movements, in cognitively impaired elderly.197 Pistacchi et al92 enrolled 236
patients (78 men and 158 women) with different subtypes of dementia and showed that RLS was one of
the most common symptoms in MCI patients.
E.2 Case-control studies
Given the degree of chronic sleep deprivation that RLS patients’ experience, Pearson et al104
hypothesized that RLS patients would exhibit decreased cognitive performance particularly on the more
pre-frontal cortical functioning (PFC) focused cognitive tasks relative to age matched controls with
relatively less or no impairments on other cognitive tasks. Results showed that RLS patients compared to
controls showed significant (P<0.05) and sizeable (20–40%) deficits on Verbal Fluency tests and the Trail
43
Making tests. The authors concluded that RLS patients have statistically significant specific cognitive
deficits and show cognitive deficits similar to those seen with 36 h total sleep deprivation.
DISCUSSION
The purpose of this systematic review was to examine the association of sleep and its related
problems/disorders with Alzheimer’s disease, cognitive decline or dementia and to evaluate the evidence
for a causal association. We systematically and critically synthesized and evaluated data from 72 original
publications. Study quality was assessed using an adaptation from previous systematic reviews73,74 and the
modified version of the Newcastle-Ottawa scale (NOS) for quality assessment of the observational
studies,75,76 with addition of new items relevant to this review. Appropriateness of study design, study
populations, study samples, size of the study sample, exposure and outcome measurement, statistical
analytic design, adjustment of potential sources of confounding, variations in study populations, and
numerous sources of bias were considered. Epidemiological studies can assess risk and contribute to
understanding of possible causation but possess built-in limitations related to their design. Consistency of
results, dose-response relationships and temporality play a crucial role in determining causation; it is
however necessary to consider high quality studies when accepting or rejecting causal associations
between exposure and outcome. Out of the 72 studies reviewed, only 14 studies were assessed to be of
low quality, while 58 studies were of medium to high quality, signifying that the publications reviewed in
this systematic review were mostly satisfactory in their design.
In this comprehensive review of seventy-two studies, there is strong evidence that sleep and its
related abnormalities/disorders are associated with cognitive decline and/or Alzheimer’s disease
independent of other factors. Summarizations of sectional findings already discussed in the sections above
are provided below.
A. Sleep depth and Cognitive decline and/or Alzheimer’s disease
44
The vast majority of the cross-sectional studies reviewed demonstrated an association between
sleep depth and cognitive decline and/or AD and/or Dementia. The association between sleep quality and
cognition was not explained by confound factors such as cerebrovascular disease, depression, or
medication usage.88 Objectively measured disturbed sleep was consistently related to poorer cognition,
whereas total sleep time was not. Sleep duration during the 24-h day was positively correlated with the
severity of dementia in nursing home patients,82 however, in a community-dwelling of women 65 years
old or older, there was no significant relationship for total sleep time with poorer cognition. These
findings may suggest that it is disturbance of sleep rather than quantity that affects cognition in demented
patients, while both sleep disturbance and duration affect cognition in the healthy elderly.
It is important to note that one cross-sectional study93 reported no significant differences between
good sleepers (GS) (PSQI <= 5) and poor sleepers (PSQI >= 5) in any of the objective cognitive function
tests except for the Trail Making Test A (TMA-A), processing speed being longer in the PS group (p <
0.001), and as such concluded that in healthy elderly subjects subjective sleep quality and duration was
not associated with subjective and objective cognitive performances, except for attention. However, in
this study, sleep duration and quality were self-reported and the authors had no information on sleep
structure and sleep fragmentation, two factors implicated in self-reported cognitive deficits.
Cross-sectional studies examining the association between sleep depth and AD pathology also
demonstrate an association. Both Spira et al94 and Ju et al85 showed that amyloid deposition in the
preclinical stage of AD appears to be associated with worse sleep quality. Spira et al also demonstrated an
association with sleep duration, however, Ju et al did not. The contrasted findings could possibly be due
to the variations in Aβ42 burden assessment. In Ju et al Aβ42 was measured using the Alzheimer Disease
Research Center (ADRC) Biomarker Core using enzyme linked immunosorbent assay while Spira et al
measured β-amyloid burden, using carbon 11–labeled Pittsburgh compound B positron emission
tomography distribution volume ratios (DVRs).
45
Results from reviewed case-control studies examining sleep depth and cognitive decline and/or
AD suggest that sleep disruptions are evident years before diagnosis of AD and that sleep is significantly
impaired in patients with MCI at both the objective and subjective level.98,99 This may have implications
for sleep being used as a surrogate marker of preclinical Alzheimer disease and for early detection of
dementia and/or therapeutic management of sleep complaints in MCI patients.
Case-control studies reviewed show an association between sleep depth and AD pathology
demonstrating that increased plasma Aβ42 levels are significantly associated with fragmented SWS in
amnestic MCI subjects,105 thereby suggesting that sleep disruptions may signal Aβ burden in persons at
increased risk for AD, and that changes in theta rhythm during REM and SWS are a relevant index of
brain impairments in the early stage of AD, and that a deficiency of Ach may result in an increase of
sEMG activity in MCI patients. Since cholinesterase inhibitors are capable of suppressing sEMG activity
in AD patients, it is possible that an increase in sEMG activity is associated with a deficiency of Ach,
which could be an early indicator of dementia
All cohort studies (prospective and retrospective) that evaluated the relationship between sleep
depth parameters and cognitive decline and/or AD reveal that sleep depth parameters increase the risk of
cognitive decline and/or developing AD. Both subjective and objective sleep assessments yield similar
outcomes of a relationship between sleep depth parameters and cognitive decline and/or AD independent
of other factors including depression/depressive symptoms.109,111,122 These findings suggest that sleep
depth parameters such as daytime napping, and night-time sleep duration, may be modifiable behaviors
open to intervention strategies, or, clinical indicators of future decline in older individuals. However, after
adjusting for depressive symptoms, Hahn et al116 could no longer show that self-reported sleep problems
may increase the risk for dementia. However, the authors measured depressive symptoms with only one
self-reported item. They also noted that they did not include longitudinal data on sleep. Research with
longitudinal sleep data is definitely preferable. Also, the relatively small sample size (n=214) may have
biased the results toward a Type II error, particularly in the more refined analyses.
46
The only cohort study reviewed,120 examining the relationship between sleep depth parameters
and AD pathology show that better sleep consolidation attenuates the effect of APOE genotype on
incident AD and development of neurofibrillary tangle pathology. Assessment of sleep consolidation may
identify APOE+ individuals at high risk for incident AD, and interventions to enhance sleep consolidation
should be studied as potentially useful means to reduce the risk of AD and development of neurofibrillary
tangles in APOE ε4+ individuals.
Experimental studies and RCTs evaluated the relationship between sleep depth parameters and
AD pathology and all demonstrate that sleep deprivation, or prolonged wakefulness, interferes with a
physiological morning decrease in Aβ42 and that chronic sleep deprivation increases cerebral Aβ42
levels, which elevates the risk of Alzheimer disease. Interestingly, Kang et al138 showed that orexin may
play a role in the pathogenesis of Alzheimer’s disease and Rothman et al146 demonstrated significant
positive correlations between cortical Aβ and p -Tau levels and circulating corticosterone. The latter
finding might suggest a probable role for glucocorticoids in mediating behavioral and biochemical
changes observed after sleep restriction in a mouse model of AD.
B. Circadian Rhythm/sleep-wake abnormalities and cognitive decline and/or Alzheimer’s
disease
All cross-sectional studies that evaluated any association between circadian Rhythm/sleep-wake
abnormalities and cognitive decline and/or Alzheimer’s disease demonstrated that an association between
the rest-activity rhythm and cognitive performance was present in elderly people. They also showed that
with the progression of dementia, excessive daytime sleepiness, the capacity to maintain sleep and the
capacity to maintain wakefulness are impaired, and result in complete fragmentation of sleep/wakefulness
during the night and day.87,89,91 However, it is unlikely that a relationship with a clinically meaningful
correlation exists between the circadian rhythm-associated SNPs and WASO in individuals with AD.95
All case-control studies that examined any association between circadian Rhythm/sleep-wake
abnormalities and cognitive decline and/or Alzheimer’s disease showed that circadian misalignment and
47
sleep disruption was evident in patients with MCI, and was consistent with changes observed in
Alzheimer’s disease. Such findings could be a marker for disease trajectory, and may even be implicated
in disease pathogenesis.
Case-control studies that examined any association between circadian rhythm/sleep-wake
abnormalities and AD pathology demonstrated that the hypocretin system was affected in advanced AD,
and that CSF melatonin levels was lower in AD patients than in old control subjects. CSF melatonin
levels from APOE-e3/4 genotype patients were significantly higher than those from the APOE-e4/4
genotype, suggesting a relationship between melatonin levels and signs and symptoms of AD.97,101,107 An
interesting study by Schmidt et al106 reported that cerebrospinal fluid melanin concentrating hormone
(CSF-MCH) correlated with T-tau (p=0.01) and P-tau (p=0.05) in the AD and that CSF-MCH correlated
negatively with MMSE scores in the AD (p=.001). These results add to preclinical studies and may allow
further research on the role of the hypothalamus in pathology and occurrence of behavioral disturbances
in AD. The authors noted that future investigations, including repeated MCH/HCRT-1 level
measurements in combination with sleep-wake-actigraphy and assessment of sleep by questionnaires,
histochemical determinations of MCH in animal and post-mortem tissues are warranted to understand the
hypothalamic impact in AD in greater detail.
All prospective cohort studies in this systematic review that examined the relationship between
circadian rhythm/sleep-wake abnormalities and cognitive decline and/or AD revealed associations
between the two, with older, healthy individuals with decreased circadian activity rhythm amplitude and
robustness, and delayed rhythms demonstrating increased risks of developing dementia and MCI.
Yesavage et al129 also showed that sleep-wake APOE status is associated with the progression of
sleep/wake disturbances in AD
Most experimental studies demonstrated anti-amyloidogenic effects of melatonin either in-vitro
and/or in-vivo in animal studies. These results suggest that melatonin can provide a combination of
48
antioxidant and anti-amyloidogenic features that can be explored either as a preventive or therapeutic
treatment for AD or as a model for development of anti-amyloidogenic indole analogs.
It is important to note that Quinn et al145 examined the effect of chronic melatonin therapy
initiated after the appearance of plaques in Tg2576 mice and failed to demonstrate any effect, as chronic
melatonin therapy did not alter amyloid burden or oxidative damage after 4-months of treatment.
Variations in melatonin dose, age of treatment initiation, and species of Aβ measured could possibly have
accounted for the different results. Quinn et al. noted that one of the prior studies used a lower dose and
still found an effect at this dose could not have accounted for the different result. The age at treatment
initiation varied across the studies and might possibly explain the different results. Another study188 using
similar Tg2576 mice initiated treatment at 4 months of age (prior to the appearance of hippocampal and
cortical plaques) also found diminished Aβ and nitrotyrosine at 7–8 months, but no significant difference
in histologically detectable Aβ or brain nitrotyrosine in animals at 15 months of age. The Quinn et al
study initiated treatment at 14 months.
Melatonin administration improved cognitive and non-cognitive functions significantly, and
seemed to be useful for care of the Alzheimer type of dementia patients as shown in the RCT conducted
by Asayama et al.131
C. Hypoxia (Apnea Hypopnea Index) and Cognitive decline and/or Alzheimer’s disease
Cross-sectional studies in our review that examined the association between hypoxemia and
cognitive decline and/or AD all showed that an association exists between SDB/OSA and CSF
Alzheimer’s disease-biomarkers in cognitively normal elderly individuals, thereby suggesting that
existing therapies for SDB such as continuous positive airway pressure could delay the onset to MCI or
dementia in normal elderly individuals. Cross-sectional studies in our review also showed that the APOE
£4 allele was associated with increased risk of SDB/OSAH, particularly in individuals under age 65.83,86
49
The only case-control study that evaluated the association between hypoxemia and cognitive
decline and/or AD also demonstrated that nocturnal sleep apnea/hypopnea was associated with lower
memory performance in APOE e4 carriers.103
Most cohort studies (retrospective and prospective) evaluated in our study revealed that
OSA/SDB may be a risk factor for dementia, with some suggesting that this relationship could be gender-
dependent, age-dependent, and time-dependent.112,128,150 In contrast, Tworoger et al125 in their study
found no associations between snoring or any of the sleep variables examined (average sleep duration,
snoring frequency, and difficulty sleeping over the previous 4 weeks) and cognitive decline over 2 years.
All 5 experimental studies that examined an association between hypoxemia and AD pathology
demonstrated an association between hypoxia and AD biomarkers. Two studies148,149 found that hypoxia
significantly increased BACE1 gene expression, resulting in increased β-secretase activity and Aβ
production. Overexpression of HIF-1 α increased BACE1 mRNA and protein level, whereas down-
regulation of HIF-1 α reduced the level of BACE1. Hypoxic treatment failed to further potentiate the
stimulatory effect of HIF-1α overexpression on BACE1 expression, suggesting that hypoxic induction of
BACE1 expression is primarily mediated by HIF-1α. Furthermore, hypoxia treatment markedly increased
Aβ deposition and potentiated the memory deficit in AD transgenic mice. These results provide evidence
that hypoxia/HIF-1α plays a vital role in balancing the amyloidogenic processing of APP, that hypoxia
can facilitate AD pathogenesis, and they provide a molecular mechanism to link vascular factors to AD.
Two other studies141,147 demonstrated that repeated hypoxia increased β-amyloid (Aβ) generation
and neuritic plaque formation by elevating β-cleavage of AD transgenic mice. They also found that
hypoxia enhanced the expression of APH-1a, a component of ɣ-secretase complex, which in turn may
lead to increase in ɣ-cleavage activity. Furthermore, these authors demonstrated that repeated hypoxia
treatment can activate macroautophagy, which may contribute to the increases in Aβ production. Taken
together, the results suggest an important role of hypoxia in modulating the APP processing by
facilitating both β- and ɣ-cleavage which may result in a significant increase of Aβ generation
50
Both RCTs130,132 that evaluated the cognitive effects of CPAP treatment demonstrated that CPAP
improved both subjectively and objectively measured sleepiness, especially in individuals with severe
OSA (AHI >30). CPAP use resulted in mild, transient improvement in the most sensitive measures of
executive and frontal-lobe function for those with severe disease, which suggests the existence of a
complex OSA-neurocognitive relationship.
D. Insomnia and Cognitive decline and/or Alzheimer’s disease
All the 4 prospective studies reviewed that examined the association between insomnia
parameters and cognitive decline and/or AD revealed that there is a greater risk of cognitive decline
and/or Alzheimer's disease in adults with insomnia and/or insomnia like symptoms. The only study that
did not find an association was cross-sectional and assessment of sleep parameters was based on a
structured interview, obtaining self-reported description of sleep characteristics from subjects with
significant cognitive impairment.
E. Restless Leg Syndrome and Cognitive decline and/or Alzheimer’s disease
All 3 studies (2 cross-sectional and 1 case-control) that evaluated RLS and cognitive decline
and/or AD associations concluded that RLS patients have statistically significant specific cognitive
deficits. Pearson et al104 showed that RLS patients compared to controls showed significant (P<0.05) and
sizeable (20–40%) deficits on Verbal Fluency tests and the Trail Making tests.
Summary of evidence
1. Sleep is associated with cognitive decline and/or AD/AD pathology
Taken together, the cross-sectional studies reviewed in this study generally provide support for an
association of sleep and poor cognitive outcomes, including AD and dementia. They show that
individuals with AD , dementia or MCI are more likely to have several indicators of sleep problems,
including sleep depth problems such poor sleep quality, sleep efficiency, sleep latency and sleep
51
deprivation; sleep-wake/circadian rhythm abnormalities, excessive daytime sleepiness and OSA/SDB.
However a significant limitation of cross-sectional studies is that causal inferences cannot be made
because the temporal sequence of sleep and circadian disruption, neurodegeneration, and cognitive
decline cannot be assessed. In addition, depression or its symptoms may be important mediators or
confounders of the association between sleep variables and dementia, and should be considered in these
studies.116 However, not all studies clearly addressed the role of depression with certain studies having a
history of depression as an exclusion criteria without considering current depression which might well be
important when one is examining factors in relation to circadian rhythm. Some studies51,87,93 assessed the
presence of sleep disorders by means of subjective symptoms and not using objective instruments, such as
polysomnography, thereby limiting the credibility of information obtained from subjects with an
important cognitive impairment. Polysomnography is essential for a correct diagnosis of some sleep
disturbances. In particular, sleep disordered breathing can be diagnosed only by means of
polysomnographic recordings. Other studies made use of tools that provided limited information on sleep
maintenance, and no detail of sleep times or other sleep parameters and others had very small sample size.
Furthermore, the association between sleep and Alzheimer’s disease pathology, measured by
Alzheimer’s disease biomarkers such as Aβ and Phosphorylated Tau accumulations, has not been
extensively investigated in humans’ in-vivo. The above study findings linking sleep to AD pathology
definitely complement those linking poor sleep to cognitive decline and dementia, however, because of
the cross-sectional nature of the studies, we cannot establish temporality and consequently causality, as
we are unable to infer whether this pathology might cause or result from poor sleep. As such, Aβ
deposition may be a cause and/or occur as a complication during the progression of AD, and/or sleep
disturbance and Aβ plaques may have a shared cause. It is therefore evident that prospective
observational studies are needed to better understand the temporal order of sleep disturbance and AD
pathogenesis and/or biomarkers. This will aid efforts to better evaluate potential opportunities for targeted
and effective prevention.
52
Altogether, it can also be seen that studies suggest that sleep variables interact with the APOE
genotype to affect Alzheimer’s disease-related outcomes. Notably, an association between sleep
disturbance especially Sleep Disordered Breathing (SDB) and CSF Alzheimer’s disease-biomarkers in
cognitively normal elderly individuals is evident. Current treatments for SDB such as continuous positive
airway pressure (CPAP) could interrupt the onset of mild cognitive impairment (MCI) or dementia in
normal elderly individuals.
2. Sleep problems precede the onset of cognitive impairment
Taken together, these prospective studies provide support that sleep problems precede the onset
of cognitive impairment, including AD and dementia. Some of the findings suggest interactions by age,
and/or sex.112,114,126 These findings require replication, and if these associations are causal, prevention
efforts may need to be targeted differently at specific ages or in different populations. It is however
important to note that longitudinal studies of dementia are prone to miss both prevalent and incident cases
because full neurological evaluations are not conducted on all participants, especially when selected
participants refuse further neuropsychological testing and examination or die over the course of follow-
up.198-200 An important limitation of most of the prospective studies is the length of the follow up. Only 4
studies had greater than 5 years follow up. Although longitudinal studies can assess temporality, AD is a
lifelong disease process and preclinical AD is very much present prior to the onset of symptomatic AD,
therefore, reverse causation might have actually been the case in these studies since a <5 year follow-up
period may not be sufficient and thus may not reflect the true prospective nature of AD biomarkers and/or
cognitive decline, as certain participants may already have had the disease but were asymptomatic. Also
self-reported data on sleep duration may be subject to bias, including the existence of a change in the
typical sleep duration anticipating the onset of dementia. On the other hand, objective sleep measures are
particularly useful in studies of sleep, aging, and cognition since self-report sleep measures do not
necessarily correlate with objectively measured sleep, and discrepancies between these modes of
assessment may be more pronounced among older adults with cognitive impairment.108,201
53
3. Sleep problems may lead to increased AD pathology or may have their effects on dementia
through other pathways
Altogether, it can be seen that experimental studies provide the most compelling evidence of the
plausibility of causal associations between sleep deprivation and AD pathology and/or AD biomarkers,
but typically are performed in animals. The power of experimental mice models to mirror the extremely
complex physiological and biochemical processes in humans is a matter of dispute because of inherent
translational limitations of these models imposed by variations in size and physiology, as well as
differences in the homology of targets between mice and humans. However, from the above study
findings it is clear that alteration of the sleep-wake cycle and circadian rhythm regulation could be
important players in the onset and development of sporadic AD. Therefore, amending and/or
improvement of sleep-wake cycle abnormalities could be a viable therapeutic strategy for AD at risk
individuals. Both animal experimental studies and RCTs also demonstrate that hypoxemia might
contribute to cognitive dysfunction in severe cases, however, in mildly affected patients,192,202 the
interacting effect of cognitive reserve,203 intelligence level,204 and depression205,206 needs to be
considered. We also see from the above study findings that disrupted melatonin levels could be a
surrogate origin of sleep pathology, as decline in melatonin levels are known to occur with aging and this
decline is enhanced with AD, even in preclinical stages. Studies also show that melatonin’s cognitive
benefits could involve its anti-Aβ aggregation, anti-inflammatory, and/or antioxidant properties. Findings
provide reinforcement for long-term melatonin therapy as a primary or complementary strategy for
abating the progression of Alzheimer disease. Also as a result of the fact that the age at treatment
initiation of melatonin might play a role in its ability to suppress Aβ levels, the implication for human
studies could be that melatonin is probably unlikely to have a role in the treatment of established cerebral
amyloidosis characteristic of AD. If this is found to be true, then clinical trials of melatonin will have to
focus on prevention rather than treatment of established cerebral amyloidosis.
54
LIMITATIONS
Several methodological concerns must be considered when interpreting the findings of this
systematic review. First, as with any systematic review, one limitation relates to the potential bias
introduced in the paper selection process. To minimize this bias, we adhered strictly to the guidelines and
recommendations in the Cochrane handbook67 whenever possible. We also ensured that studies were not
restricted by language to guard against language bias. Secondly, the possible bias introduced by the
individual studies that were included is a further limitation. Various study designs were rated low for
assigning the grade of evidence.207 In addition to study design, conclusions made from the poor quality
studies could introduce possible bias. We included both high- and low-quality studies in the
summarization of findings because both mostly came to the same conclusion, therefore possibly
indicating that the study quality might not have influenced the overall conclusion. Thirdly, there was
considerable variation between the studies in the assessment methods used to verify sleep problems and
cognitive decline and/or AD/Dementia, suggesting heterogeneity that could possibly raise some questions
about the comparability of results across studies. Our research question focused on a broad concept of
sleep problems based on having any sleep problem symptoms. Thus, we assumed a common underlying
concept by making general conclusions regarding sleep problems and cognitive decline and/or AD.
Fourthly, as in any systematic review, publication bias may have affected the representativeness of the
included studies by over reporting significant findings. Non assessment of publication bias with a funnel
plot is a limitation. Lastly the possibility of a “file drawer” effect exists because exclusion and inclusion
criteria are always subjective.
CONCLUSION AND FUTURE DIRECTIONS
In summary, there is substantial evidence providing support for sleep and its related abnormalities
as a potential cause of cognitive decline, dementia, and Alzheimer’s disease pathology. Ample evidence
linking sleep impairments and circadian regulating mechanisms directly to clinical symptoms in AD
55
patients exist. Future research is needed to make clear the contribution of the different genetic,
neurodegenerative and environmental factors to sleep impairment. There is also growing experimental
and epidemiological evidence for close reciprocal interaction between cognitive decline and sleep
alteration, however, long-term longitudinal studies are required to better understand the relationship
between chronic sleep deprivation, circadian rhythm/sleep-wake abnormalities, sleep-disordered
breathing, sleep disorders like OSAS and cognitive decline. Future studies need to be able to define causal
pathways more definitively so we can be clear for example, whether neurodegeneration is leading to both
sleep changes and memory changes, or whether aging is contributing to OSA and/or sleep architecture
changes that in turn affect memory, and/or whether the effect of aging on sleep architecture predisposes to
both apnea and memory impairment. Thus, further research that examines these basic mechanisms as well
as employing clinical/translational approaches are needed. More research with objective sleep measures,
Alzheimer’s disease biomarkers, and cognitive measures are also needed. There is also significant
evidence from in vitro and animal studies for an important role of both melatonin and hypocretin in
modulating AD pathophysiology. Although the exact underlying mechanisms are unknown as of yet,
future studies examining the role of lack of sleep and circadian rhythm deterioration as causative factors
in the onset of AD, especially human studies, are needed, as their findings can be promising for the
development of new prevention and/or treatment opportunities.
RESEARCH AGENDA AND GAPS IN THE LITEARTURE I INTEND TO ADDRESS
The need for a meta-analysis to quantify the effect of sleep problems/disorders on cognitive
impairment and AD. This would allow for efficient integration of existing information from various
studies enabling us to establish whether the literature is consistent and can be generalized across
populations, settings, and/or whether findings vary significantly by particular subsets.
AIM I: To conduct a meta-analysis to quantify the aggregate effect of sleep problems and/or
disorders on cognitive decline and Alzheimer's disease.
56
The need to be able to define causal pathways more definitively so we can be clear for example,
whether neurodegeneration is leading to both sleep changes and memory changes, or whether aging is
contributing to OSA and/or sleep architecture changes that in turn affect memory, and/or whether the
effect of aging on sleep architecture predisposes to both apnea and memory impairment. One way to be
able to do this is to conduct more research examining relationships between OSA (preferably with
objective sleep measures) and Alzheimer’s disease biomarkers, and/or cognitive measures. This led to the
following study aims:
AIM IIa: To examine whether the presence of Obstructive Sleep Disordered Breathing is
associated with CSF P-Tau, Hippocampal atrophy and β-amyloid burden in older adults with
MCI and AD.
AIM IIb: To examine the effect of Obstructive Sleep Apnea on longitudinal changes of
Alzheimer's disease biomarkers in cognitive normal, mild cognitive impairment and Alzheimer's
disease patient
The need for clarity in understanding and appreciating the heterogeneity of OSA and its outcomes
in distinct age groups especially as it relates to cognitive function, subsequent cognitive decline and AD.
One way to do this is to integrate decades of research examining OSA and cognition; OSA and
subsequent cognitive decline; and OSA and AD; with particular focus in appreciating the heterogeneity of
OSA and its outcomes in distinct age groups.
AIM III: To conduct a Review titled "Obstructive Sleep Apnea: A Distinct Physiological
Phenotypic Risk Factor in older adults with Cognitive decline and Alzheimer’s disease"
References
1. Thies W, Bleiler L. 2013 Alzheimer's disease facts and figures. Alzheimer's & dementia : the
journal of the Alzheimer's Association. 2013;9(2):208-245.
57
2. Hebert LE, Beckett LA, Scherr PA, Evans DA. Annual incidence of Alzheimer disease in the
United States projected to the years 2000 through 2050. Alzheimer disease and associated
disorders. 2001;15(4):169-173.
3. Wimo A, Jonsson L, Bond J, Prince M, Winblad B. The worldwide economic impact of dementia
2010. Alzheimer's & dementia : the journal of the Alzheimer's Association. 2013;9(1):1-11 e13.
4. Brookmeyer R, Johnson E, Ziegler-Graham K, Arrighi HM. Forecasting the global burden of
Alzheimer's disease. Alzheimer's & dementia : the journal of the Alzheimer's Association.
2007;3(3):186-191.
5. Fratiglioni L, Ahlbom A, Viitanen M, Winblad B. Risk factors for late-onset Alzheimer's disease:
a population-based, case-control study. Annals of neurology. 1993;33(3):258-266.
6. Green RC, Cupples LA, Go R, et al. Risk of dementia among white and African American
relatives of patients with Alzheimer disease. JAMA : the journal of the American Medical
Association. 2002;287(3):329-336.
7. Mayeux R, Sano M, Chen J, Tatemichi T, Stern Y. Risk of dementia in first-degree relatives of
patients with Alzheimer's disease and related disorders. Archives of neurology. 1991;48(3):269-
273.
8. Farrer LA, Cupples LA, Haines JL, et al. Effects of age, sex, and ethnicity on the association
between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and
Alzheimer Disease Meta Analysis Consortium. JAMA : the journal of the American Medical
Association. 1997;278(16):1349-1356.
9. Saunders AM, Strittmatter WJ, Schmechel D, et al. Association of apolipoprotein E allele epsilon
4 with late-onset familial and sporadic Alzheimer's disease. Neurology. 1993;43(8):1467-1472.
10. Anstey KJ, von Sanden C, Salim A, O'Kearney R. Smoking as a risk factor for dementia and
cognitive decline: a meta-analysis of prospective studies. American journal of epidemiology.
2007;166(4):367-378.
58
11. Pendlebury ST, Rothwell PM. Prevalence, incidence, and factors associated with pre-stroke and
post-stroke dementia: a systematic review and meta-analysis. Lancet neurology.
2009;8(11):1006-1018.
12. Rusanen M, Kivipelto M, Quesenberry CP, Jr., Zhou J, Whitmer RA. Heavy smoking in midlife
and long-term risk of Alzheimer disease and vascular dementia. Archives of internal medicine.
2011;171(4):333-339.
13. Ohara T, Doi Y, Ninomiya T, et al. Glucose tolerance status and risk of dementia in the
community: the Hisayama study. Neurology. 2011;77(12):1126-1134.
14. Reitz C, Brayne C, Mayeux R. Epidemiology of Alzheimer disease. Nature reviews. Neurology.
2011;7(3):137-152.
15. Wu W, Brickman AM, Luchsinger J, et al. The brain in the age of old: the hippocampal formation
is targeted differentially by diseases of late life. Annals of neurology. 2008;64(6):698-706.
16. Fitzpatrick AL, Kuller LH, Lopez OL, et al. Midlife and late-life obesity and the risk of dementia:
cardiovascular health study. Archives of neurology. 2009;66(3):336-342.
17. Kivipelto M, Ngandu T, Fratiglioni L, et al. Obesity and vascular risk factors at midlife and the
risk of dementia and Alzheimer disease. Archives of neurology. 2005;62(10):1556-1560.
18. Raji CA, Ho AJ, Parikshak NN, et al. Brain structure and obesity. Human brain mapping.
2010;31(3):353-364.
19. Whitmer RA, Gustafson DR, Barrett-Connor E, Haan MN, Gunderson EP, Yaffe K. Central
obesity and increased risk of dementia more than three decades later. Neurology.
2008;71(14):1057-1064.
20. Xu WL, Atti AR, Gatz M, Pedersen NL, Johansson B, Fratiglioni L. Midlife overweight and
obesity increase late-life dementia risk: a population-based twin study. Neurology.
2011;76(18):1568-1574.
59
21. Solomon A, Kivipelto M, Wolozin B, Zhou J, Whitmer RA. Midlife serum cholesterol and
increased risk of Alzheimer's and vascular dementia three decades later. Dementia and geriatric
cognitive disorders. 2009;28(1):75-80.
22. Debette S, Seshadri S, Beiser A, et al. Midlife vascular risk factor exposure accelerates structural
brain aging and cognitive decline. Neurology. 2011;77(5):461-468.
23. Launer LJ, Ross GW, Petrovitch H, et al. Midlife blood pressure and dementia: the Honolulu-
Asia aging study. Neurobiol Aging. 2000;21(1):49-55.
24. Ninomiya T, Ohara T, Hirakawa Y, et al. Midlife and late-life blood pressure and dementia in
Japanese elderly: the Hisayama study. Hypertension. 2011;58(1):22-28.
25. Larson EB, Wang L, Bowen JD, et al. Exercise is associated with reduced risk for incident
dementia among persons 65 years of age and older. Annals of internal medicine. 2006;144(2):73-
81.
26. Laurin D, Verreault R, Lindsay J, MacPherson K, Rockwood K. Physical activity and risk of
cognitive impairment and dementia in elderly persons. Archives of neurology. 2001;58(3):498-
504.
27. Mortimer JA. Physical activity and cognitive function in Alzheimer disease. JAMA : the journal
of the American Medical Association. 2009;301(3):272-273; author reply 273-274.
28. Willis BL, Gao A, Leonard D, Defina LF, Berry JD. Midlife fitness and the development of
chronic conditions in later life. Archives of internal medicine. 2012;172(17):1333-1340.
29. Evans DA, Bennett DA, Wilson RS, et al. Incidence of Alzheimer disease in a biracial urban
community: relation to apolipoprotein E allele status. Archives of neurology. 2003;60(2):185-189.
30. Stern Y, Gurland B, Tatemichi TK, Tang MX, Wilder D, Mayeux R. Influence of education and
occupation on the incidence of Alzheimer's disease. JAMA : the journal of the American Medical
Association. 1994;271(13):1004-1010.
31. Graves AB, Mortimer JA, Larson EB, Wenzlow A, Bowen JD, McCormick WC. Head
circumference as a measure of cognitive reserve. Association with severity of impairment in
60
Alzheimer's disease. The British journal of psychiatry : the journal of mental science.
1996;169(1):86-92.
32. Mortimer JA, Snowdon DA, Markesbery WR. Small head circumference is associated with less
education in persons at risk for Alzheimer disease in later life. Alzheimer disease and associated
disorders. 2008;22(3):249-254.
33. Mortimer JA, French LR, Hutton JT, Schuman LM. Head injury as a risk factor for Alzheimer's
disease. Neurology. 1985;35(2):264-267.
34. Mortimer JA, van Duijn CM, Chandra V, et al. Head trauma as a risk factor for Alzheimer's
disease: a collaborative re-analysis of case-control studies. EURODEM Risk Factors Research
Group. Int J Epidemiol. 1991;20 Suppl 2:S28-35.
35. Mortimer JA. Brain reserve and the clinical expression of Alzheimer's disease. Geriatrics.
1997;52 Suppl 2:S50-53.
36. Hall CB, Lipton RB, Sliwinski M, Katz MJ, Derby CA, Verghese J. Cognitive activities delay
onset of memory decline in persons who develop dementia. Neurology. 2009;73(5):356-361.
37. Wilson RS, Bennett DA, Bienias JL, et al. Cognitive activity and incident AD in a population-
based sample of older persons. Neurology. 2002;59(12):1910-1914.
38. Wilson RS, Mendes De Leon CF, Barnes LL, et al. Participation in cognitively stimulating
activities and risk of incident Alzheimer disease. JAMA : the journal of the American Medical
Association. 2002;287(6):742-748.
39. Ballard C, Gauthier S, Corbett A, Brayne C, Aarsland D, Jones E. Alzheimer's disease. Lancet.
2011;377(9770):1019-1031.
40. Peter-Derex L, Yammine P, Bastuji H, Croisile B. Sleep and Alzheimer's disease. Sleep medicine
reviews. 2014.
41. Cooke JR, Ancoli-Israel S. Normal and abnormal sleep in the elderly. Handbook of clinical
neurology. 2011;98:653-665.
61
42. Van Cauter E, Leproult R, Plat L. Age-related changes in slow wave sleep and REM sleep and
relationship with growth hormone and cortisol levels in healthy men. JAMA : the journal of the
American Medical Association. 2000;284(7):861-868.
43. Bliwise DL. Sleep in normal aging and dementia. Sleep. 1993;16(1):40-81.
44. Touitou Y. Human aging and melatonin. Clinical relevance. Experimental gerontology.
2001;36(7):1083-1100.
45. Prinz PN, Peskind ER, Vitaliano PP, et al. Changes in the sleep and waking EEGs of
nondemented and demented elderly subjects. Journal of the American Geriatrics Society.
1982;30(2):86-93.
46. Prinz PN, Vitaliano PP, Vitiello MV, et al. Sleep, EEG and mental function changes in senile
dementia of the Alzheimer's type. Neurobiology of aging. 1982;3(4):361-370.
47. Vitiello MV, Prinz PN, Williams DE, Frommlet MS, Ries RK. Sleep disturbances in patients with
mild-stage Alzheimer's disease. Journal of gerontology. 1990;45(4):M131-138.
48. Hoch CC, Reynolds CF, 3rd, Kupfer DJ, Houck PR, Berman SR, Stack JA. Sleep-disordered
breathing in normal and pathologic aging. The Journal of clinical psychiatry. 1986;47(10):499-
503.
49. Mant A, Saunders NA, Eyland AE, Pond CD, Chancellor AH, Webster IW. Sleep-related
respiratory disturbance and dementia in elderly females. Journal of gerontology.
1988;43(5):M140-144.
50. McCurry SM, Reynolds CF, Ancoli-Israel S, Teri L, Vitiello MV. Treatment of sleep disturbance
in Alzheimer's disease. Sleep medicine reviews. 2000;4(6):603-628.
51. Moran M, Lynch CA, Walsh C, Coen R, Coakley D, Lawlor BA. Sleep disturbance in mild to
moderate Alzheimer's disease. Sleep medicine. 2005;6(4):347-352.
52. Vitiello MV, Prinz PN. Alzheimer's disease. Sleep and sleep/wake patterns. Clinics in geriatric
medicine. 1989;5(2):289-299.
62
53. Bianchetti A, Scuratti A, Zanetti O, et al. Predictors of mortality and institutionalization in
Alzheimer disease patients 1 year after discharge from an Alzheimer dementia unit. Dementia
(Basel, Switzerland). 1995;6(2):108-112.
54. Pollak CP, Perlick D. Sleep problems and institutionalization of the elderly. Journal of geriatric
psychiatry and neurology. 1991;4(4):204-210.
55. Pollak CP, Perlick D, Linsner JP, Wenston J, Hsieh F. Sleep problems in the community elderly
as predictors of death and nursing home placement. Journal of community health.
1990;15(2):123-135.
56. Singer C, Tractenberg RE, Kaye J, et al. A multicenter, placebo-controlled trial of melatonin for
sleep disturbance in Alzheimer's disease. Sleep. 2003;26(7):893-901.
57. Meeks TW, Ropacki SA, Jeste DV. The neurobiology of neuropsychiatric syndromes in
dementia. Current opinion in psychiatry. 2006;19(6):581-586.
58. Ju YE, Lucey BP, Holtzman DM. Sleep and Alzheimer disease pathology-a bidirectional
relationship. Nature reviews. Neurology. 2013.
59. Rauchs G, Schabus M, Parapatics S, et al. Is there a link between sleep changes and memory in
Alzheimer's disease? Neuroreport. 2008;19(11):1159-1162.
60. Rauchs G HC, Bertran F, Desgranges B, Eustache F. . Sleep and episodic memory: a review of
the literature in young healthy subjects and potential links between sleep changes and memory
impairment observed during aging and Alzheimer’s disease. . Rev Neurol (Paris) 2010;.
2010;166:(873e):81.
61. Bloom HG, Ahmed I, Alessi CA, et al. Evidence-based recommendations for the assessment and
management of sleep disorders in older persons. Journal of the American Geriatrics Society.
2009;57(5):761-789.
62. Palma JA, Urrestarazu E, Iriarte J. Sleep loss as risk factor for neurologic disorders: a review.
Sleep medicine. 2013;14(3):229-236.
63
63. Slats D, Claassen JA, Verbeek MM, Overeem S. Reciprocal interactions between sleep, circadian
rhythms and Alzheimer's disease: focus on the role of hypocretin and melatonin. Ageing research
reviews. 2013;12(1):188-200.
64. Spira AP, Chen-Edinboro LP, Wu MN, Yaffe K. Impact of sleep on the risk of cognitive decline
and dementia. Current opinion in psychiatry. 2014;27(6):478-483.
65. Pai M, McCulloch M, Gorman JD, et al. Systematic reviews and meta-analyses: an illustrated,
step-by-step guide. The National medical journal of India. 2004;17(2):86-95.
66. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and
meta-analyses: the PRISMA statement. BMJ (Clinical research ed.). 2009;339:b2535.
67. Higgins JPT GSe. Cochrane handbook for systematic reviews of interventions. Version 5.1.0
[updated March 2011]. . http://www.cochranehandbook.org [accessed 11.23.14]. 2011.
68. Buysse DJ, Reynolds CF, 3rd, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality
Index: a new instrument for psychiatric practice and research. Psychiatry research.
1989;28(2):193-213.
69. Ancoli-Israel S, Cole R, Alessi C, Chambers M, Moorcroft W, Pollak CP. The role of actigraphy
in the study of sleep and circadian rhythms. Sleep. 2003;26(3):342-392.
70. Medicine. AAoS. The international classification of sleep disorders, revised: diagnostic and
coding manual. Chicago, USA: . American Academy of Sleep Medicine, ISBN 0-9657220-1-5;
2001. 2001.
71. Kling RN, McLeod CB, Koehoorn M. Sleep problems and workplace injuries in Canada. Sleep.
2010;33(5):611-618.
72. Uehli K, Mehta AJ, Miedinger D, et al. Sleep problems and work injuries: a systematic review
and meta-analysis. Sleep medicine reviews. 2014;18(1):61-73.
73. Bhui K, Stansfeld S, Hull S, Priebe S, Mole F, Feder G. Ethnic variations in pathways to and use
of specialist mental health services in the UK. Systematic review. The British journal of
psychiatry : the journal of mental science. 2003;182:105-116.
64
74. Dinos S, Khoshaba B, Ashby D, et al. A systematic review of chronic fatigue, its syndromes and
ethnicity: prevalence, severity, co-morbidity and coping. Int J Epidemiol. 2009;38(6):1554-1570.
75. Wells G SB, O’Connell D, Peterson J,Welch V, Losos M, et al. . The Newcastle-Ottawa scale
(NOS) for assessing the quality of nonrandomised studies in metaanalyses,.
http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp/; 2011 [accessed 11.25.14].
2011.
76. D. P. Personal communication 2011. 2011.
77. Popay J RH, Sowden A et al. Developing guidance on the conduct of narrative synthesis in
systematic reviews. . J Epidemiol Community Health 2005;59:A7.
78. Moher D, Liberati A, Tetzlaff J, Altman DG, Grp P. Preferred Reporting Items for Systematic
Reviews and Meta-Analyses: The PRISMA Statement. Journal of Clinical Epidemiology.
2009;62(10):1006-1012.
79. Blackwell T, Yaffe K, Ancoli-Israel S, et al. Poor sleep is associated with impaired cognitive
function in older women: the study of osteoporotic fractures. The journals of gerontology. Series
A, Biological sciences and medical sciences. 2006;61(4):405-410.
80. Craig D, Hart DJ, Passmore AP. Genetically increased risk of sleep disruption in Alzheimer's
disease. Sleep. 2006;29(8):1003-1007.
81. Djonlagic I, Guo M, Matteis P, Carusona A, Stickgold R, Malhotra A. Untreated sleep-disordered
breathing: links to aging-related decline in sleep-dependent memory consolidation. PloS one.
2014;9(1):e85918.
82. Fetveit A, Bjorvatn B. Sleep duration during the 24-hour day is associated with the severity of
dementia in nursing home patients. International journal of geriatric psychiatry.
2006;21(10):945-950.
83. Gottlieb DJ, DeStefano AL, Foley DJ, et al. APOE epsilon4 is associated with obstructive sleep
apnea/hypopnea: the Sleep Heart Health Study. Neurology. 2004;63(4):664-668.
65
84. Guarnieri B, Adorni F, Musicco M, et al. Prevalence of Sleep Disturbances in Mild Cognitive
Impairment and Dementing Disorders: A Multicenter Italian Clinical Cross-Sectional Study on
431 Patients. Dementia and geriatric cognitive disorders. 2012;33(1):50-58.
85. Ju YE, McLeland JS, Toedebusch CD, et al. Sleep quality and preclinical Alzheimer disease.
JAMA neurology. 2013;70(5):587-593.
86. Kadotani H, Kadotani T, Young T, et al. Association between apolipoprotein E epsilon4 and
sleep-disordered breathing in adults. JAMA : the journal of the American Medical Association.
2001;285(22):2888-2890.
87. Merlino G, Piani A, Gigli GL, et al. Daytime sleepiness is associated with dementia and cognitive
decline in older Italian adults: A population-based study. Sleep medicine. 2010;11(4):372-377.
88. Nebes RD, Buysse DJ, Halligan EM, Houck PR, Monk TH. Self-reported sleep quality predicts
poor cognitive performance in healthy older adults. The journals of gerontology. Series B,
Psychological sciences and social sciences. 2009;64(2):180-187.
89. Oosterman JM, van Someren EJ, Vogels RL, Van Harten B, Scherder EJ. Fragmentation of the
rest-activity rhythm correlates with age-related cognitive deficits. Journal of sleep research.
2009;18(1):129-135.
90. Osorio RS, Ayappa I, Mantua J, et al. The interaction between sleep-disordered breathing and
apolipoprotein E genotype on cerebrospinal fluid biomarkers for Alzheimer's disease in
cognitively normal elderly individuals. Neurobiol Aging. 2014;35(6):1318-1324.
91. Pat-Horenczyk R, Klauber MR, Shochat T, Ancoli-Israel S. Hourly profiles of sleep and
wakefulness in severely versus mild-moderately demented nursing home patients. Aging (Milan,
Italy). 1998;10(4):308-315.
92. Pistacchi M, Gioulis M, Contin F, Sanson F, Marsala SZ. Sleep disturbance and cognitive
disorder: epidemiological analysis in a cohort of 263 patients. Neurological sciences : official
journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology.
2014.
66
93. Saint Martin M, Sforza E, Barthelemy JC, Thomas-Anterion C, Roche F. Does subjective sleep
affect cognitive function in healthy elderly subjects? The Proof cohort. Sleep medicine.
2012;13(9):1146-1152.
94. Spira AP, Gamaldo AA, An Y, et al. Self-reported sleep and beta-amyloid deposition in
community-dwelling older adults. JAMA neurology. 2013;70(12):1537-1543.
95. Yesavage JA, Noda A, Hernandez B, et al. Circadian clock gene polymorphisms and sleep-wake
disturbance in Alzheimer disease. The American journal of geriatric psychiatry : official journal
of the American Association for Geriatric Psychiatry. 2011;19(7):635-643.
96. Chen PC, Wu D, Chen CC, Chi NF, Kang JH, Hu CJ. Rapid eye movement sleep atonia in
patients with cognitive impairment. Journal of the neurological sciences. 2011;305(1-2):34-37.
97. Fronczek R, van Geest S, Frolich M, et al. Hypocretin (orexin) loss in Alzheimer's disease.
Neurobiology of aging. 2012;33(8):1642-1650.
98. Hita-Yanez E, Atienza M, Cantero JL. Polysomnographic and subjective sleep markers of mild
cognitive impairment. Sleep. 2013;36(9):1327-1334.
99. Hita-Yanez E, Atienza M, Gil-Neciga E, Cantero JL. Disturbed sleep patterns in elders with mild
cognitive impairment: the role of memory decline and ApoE epsilon4 genotype. Current
Alzheimer research. 2012;9(3):290-297.
100. Hot P, Rauchs G, Bertran F, et al. Changes in sleep theta rhythm are related to episodic memory
impairment in early Alzheimer's disease. Biological psychology. 2011;87(3):334-339.
101. Liu RY, Zhou JN, van Heerikhuize J, Hofman MA, Swaab DF. Decreased melatonin levels in
postmortem cerebrospinal fluid in relation to aging, Alzheimer's disease, and apolipoprotein E-
epsilon4/4 genotype. The Journal of clinical endocrinology and metabolism. 1999;84(1):323-327.
102. Naismith SL, Hickie IB, Terpening Z, et al. Circadian Misalignment and Sleep Disruption in Mild
Cognitive Impairment. Journal of Alzheimers Disease. 2014;38(4):857-866.
103. O'Hara R, Schroder CM, Kraemer HC, et al. Nocturnal sleep apnea/hypopnea is associated with
lower memory performance in APOE epsilon4 carriers. Neurology. 2005;65(4):642-644.
67
104. Pearson VE, Allen RP, Dean T, Gamaldo CE, Lesage SR, Earley CJ. Cognitive deficits
associated with restless legs syndrome (RLS). Sleep medicine. 2006;7(1):25-30.
105. Sanchez-Espinosa MP, Atienza M, Cantero JL. Sleep deficits in mild cognitive impairment are
related to increased levels of plasma amyloid-beta and cortical thinning. NeuroImage.
2014;98:395-404.
106. Schmidt FM, Kratzsch J, Gertz HJ, et al. Cerebrospinal fluid melanin-concentrating hormone
(MCH) and hypocretin-1 (HCRT-1, orexin-A) in Alzheimer's disease. PloS one.
2013;8(5):e63136.
107. Slats D, Claassen JA, Lammers GJ, Melis RJ, Verbeek MM, Overeem S. Association between
hypocretin-1 and amyloid-beta42 cerebrospinal fluid levels in Alzheimer's disease and healthy
controls. Current Alzheimer research. 2012;9(10):1119-1125.
108. Anderson KN, Catt M, Collerton J, et al. Assessment of sleep and circadian rhythm disorders in
the very old: the Newcastle 85+ Cohort Study. Age and ageing. 2014;43(1):57-63.
109. Benito-Leon J, Louis ED, Villarejo-Galende A, Romero JP, Bermejo-Pareja F. Long sleep
duration in elders without dementia increases risk of dementia mortality (NEDICES). Neurology.
2014;83(17):1530-1537.
110. Bidzan L, Grabowski J, Dutczak B, Bidzan M. [Sleep disorders in the preclinical period of the
Alzheimer's disease]. Psychiatria polska. 2011;45(6):851-860.
111. Blackwell T, Yaffe K, Laffan A, et al. Associations of objectively and subjectively measured
sleep quality with subsequent cognitive decline in older community-dwelling men: the MrOS
sleep study. Sleep. 2014;37(4):655-663.
112. Chang WP, Liu ME, Chang WC, et al. Sleep apnea and the risk of dementia: a population-based
5-year follow-up study in Taiwan. PloS one. 2013;8(10):e78655.
113. Cohen-Zion M SC, Marler, Shochat T, Kripke DF, Ancoli-Israel S. . Changes in cognitive
function associated with sleep disordered breathing in older people. . J Am Geriatr Soc. 2001;.
2001;49((12):):1622-1627.
68
114. Cricco M, Simonsick EM, Foley DJ. The impact of insomnia on cognitive functioning in older
adults. Journal of the American Geriatrics Society. 2001;49(9):1185-1189.
115. Foley D, Monjan A, Masaki K, et al. Daytime sleepiness is associated with 3-year incident
dementia and cognitive decline in older Japanese-American men. Journal of the American
Geriatrics Society. 2001;49(12):1628-1632.
116. Hahn EA, Wang HX, Andel R, Fratiglioni L. A change in sleep pattern may predict Alzheimer
disease. The American journal of geriatric psychiatry : official journal of the American
Association for Geriatric Psychiatry. 2014;22(11):1262-1271.
117. Jelicic M, Bosma H, Ponds RW, Van Boxtel MP, Houx PJ, Jolles J. Subjective sleep problems in
later life as predictors of cognitive decline. Report from the Maastricht Ageing Study (MAAS).
International journal of geriatric psychiatry. 2002;17(1):73-77.
118. Keage HA, Banks S, Yang KL, Morgan K, Brayne C, Matthews FE. What sleep characteristics
predict cognitive decline in the elderly? Sleep medicine. 2012;13(7):886-892.
119. Lim AS, Kowgier M, Yu L, Buchman AS, Bennett DA. Sleep Fragmentation and the Risk of
Incident Alzheimer's Disease and Cognitive Decline in Older Persons. Sleep. 2013;36(7):1027-
1032.
120. Lim AS, Yu L, Kowgier M, Schneider JA, Buchman AS, Bennett DA. Modification of the
Relationship of the Apolipoprotein E epsilon4 Allele to the Risk of Alzheimer Disease and
Neurofibrillary Tangle Density by Sleep. JAMA neurology. 2013;70(12):1544-1551.
121. Osorio RS, Pirraglia E, Aguera-Ortiz LF, et al. Greater risk of Alzheimer's disease in older adults
with insomnia. Journal of the American Geriatrics Society. 2011;59(3):559-562.
122. Song Y, Blackwell T, Yaffe K, Ancoli-Israel S, Redline S, Stone KL. Relationships Between
Sleep Stages and Changes in Cognitive Function in Older Men: The MrOS Sleep Study. Sleep.
2014.
123. Sterniczuk R, Theou O, Rusak B, Rockwood K. Sleep disturbance is associated with incident
dementia and mortality. Current Alzheimer research. 2013;10(7):767-775.
69
124. Tranah GJ, Blackwell T, Stone KL, et al. Circadian activity rhythms and risk of incident dementia
and mild cognitive impairment in older women. Annals of neurology. 2011;70(5):722-732.
125. Tworoger SS, Lee S, Schernhammer ES, Grodstein F. The association of self-reported sleep
duration, difficulty sleeping, and snoring with cognitive function in older women. Alzheimer
disease and associated disorders. 2006;20(1):41-48.
126. Virta JJ, Heikkila K, Perola M, et al. Midlife sleep characteristics associated with late life
cognitive function. Sleep. 2013;36(10):1533-1541, 1541A.
127. Walsh CM, Blackwell T, Tranah GJ, et al. Weaker Circadian Activity Rhythms are Associated
with Poorer Executive Function in Older Women. Sleep. 2014.
128. Yaffe K, Laffan AM, Harrison SL, et al. Sleep-disordered breathing, hypoxia, and risk of mild
cognitive impairment and dementia in older women. Jama. 2011;306(6):613-619.
129. Yesavage JA, Friedman L, Kraemer H, et al. Sleep/wake disruption in Alzheimer's disease:
APOE status and longitudinal course. Journal of geriatric psychiatry and neurology.
2004;17(1):20-24.
130. Ancoli-Israel S, Palmer BW, Cooke JR, et al. Cognitive effects of treating obstructive sleep apnea
in Alzheimer's disease: a randomized controlled study. Journal of the American Geriatrics
Society. 2008;56(11):2076-2081.
131. Asayama K, Yamadera H, Ito T, Suzuki H, Kudo Y, Endo S. Double blind study of melatonin
effects on the sleep-wake rhythm, cognitive and non-cognitive functions in Alzheimer type
dementia. Journal of Nippon Medical School = Nippon Ika Daigaku zasshi. 2003;70(4):334-341.
132. Kushida CA, Nichols DA, Holmes TH, et al. Effects of continuous positive airway pressure on
neurocognitive function in obstructive sleep apnea patients: The Apnea Positive Pressure Long-
term Efficacy Study (APPLES). Sleep. 2012;35(12):1593-1602.
133. Ooms S, Overeem S, Besse K, Rikkert MO, Verbeek M, Claassen JA. Effect of 1 night of total
sleep deprivation on cerebrospinal fluid beta-amyloid 42 in healthy middle-aged men: a
randomized clinical trial. JAMA neurology. 2014;71(8):971-977.
70
134. Craig LA, McDonald RJ. Chronic disruption of circadian rhythms impairs hippocampal memory
in the rat. Brain research bulletin. 2008;76(1-2):141-151.
135. Deng YQ, Xu GG, Duan P, Zhang Q, Wang JZ. Effects of melatonin on wortmannin-induced tau
hyperphosphorylation. Acta pharmacologica Sinica. 2005;26(5):519-526.
136. Di Meco A, Joshi YB, Pratico D. Sleep deprivation impairs memory, tau metabolism, and
synaptic integrity of a mouse model of Alzheimer's disease with plaques and tangles.
Neurobiology of aging. 2014;35(8):1813-1820.
137. Jyoti A, Plano A, Riedel G, Platt B. EEG, activity, and sleep architecture in a transgenic
AbetaPPswe/PSEN1A246E Alzheimer's disease mouse. Journal of Alzheimer's disease : JAD.
2010;22(3):873-887.
138. Kang JE, Lim MM, Bateman RJ, et al. Amyloid-beta dynamics are regulated by orexin and the
sleep-wake cycle. Science (New York, N.Y.). 2009;326(5955):1005-1007.
139. Kaushal N, Ramesh V, Gozal D. Human apolipoprotein E4 targeted replacement in mice reveals
increased susceptibility to sleep disruption and intermittent hypoxia. American journal of
physiology. Regulatory, integrative and comparative physiology. 2012;303(1):R19-29.
140. Lahiri DK, Chen D, Ge YW, Bondy SC, Sharman EH. Dietary supplementation with melatonin
reduces levels of amyloid beta-peptides in the murine cerebral cortex. Journal of pineal research.
2004;36(4):224-231.
141. Li L, Zhang X, Yang D, Luo G, Chen S, Le W. Hypoxia increases Abeta generation by altering
beta- and gamma-cleavage of APP. Neurobiology of aging. 2009;30(7):1091-1098.
142. Mander BA, Rao V, Lu B, et al. Prefrontal atrophy, disrupted NREM slow waves and impaired
hippocampal-dependent memory in aging. Nature neuroscience. 2013;16(3):357-364.
143. Olcese JM, Cao C, Mori T, et al. Protection against cognitive deficits and markers of
neurodegeneration by long-term oral administration of melatonin in a transgenic model of
Alzheimer disease. Journal of pineal research. 2009;47(1):82-96.
71
144. Pappolla M, Bozner P, Soto C, et al. Inhibition of Alzheimer beta-fibrillogenesis by melatonin.
The Journal of biological chemistry. 1998;273(13):7185-7188.
145. Quinn J, Kulhanek D, Nowlin J, et al. Chronic melatonin therapy fails to alter amyloid burden or
oxidative damage in old Tg2576 mice: implications for clinical trials. Brain research.
2005;1037(1-2):209-213.
146. Rothman SM, Herdener N, Frankola KA, Mughal MR, Mattson MP. Chronic mild sleep
restriction accentuates contextual memory impairments, and accumulations of cortical Abeta and
pTau in a mouse model of Alzheimer's disease. Brain research. 2013;1529:200-208.
147. Shiota S, Takekawa H, Matsumoto SE, et al. Chronic intermittent hypoxia/reoxygenation
facilitate amyloid-beta generation in mice. Journal of Alzheimer's disease : JAD. 2013;37(2):325-
333.
148. Sun X, He G, Qing H, et al. Hypoxia facilitates Alzheimer's disease pathogenesis by up-
regulating BACE1 gene expression. Proceedings of the National Academy of Sciences of the
United States of America. 2006;103(49):18727-18732.
149. Zhang X, Zhou K, Wang R, et al. Hypoxia-inducible factor 1alpha (HIF-1alpha)-mediated
hypoxia increases BACE1 expression and beta-amyloid generation. The Journal of biological
chemistry. 2007;282(15):10873-10880.
150. Cohen-Zion M, Stepnowsky C, Marler, Shochat T, Kripke DF, Ancoli-Israel S. Changes in
cognitive function associated with sleep disordered breathing in older people. Journal of the
American Geriatrics Society. 2001;49(12):1622-1627.
151. Netzer NC, Stoohs RA, Netzer CM, Clark K, Strohl KP. Using the Berlin Questionnaire to
identify patients at risk for the sleep apnea syndrome. Annals of internal medicine.
1999;131(7):485-491.
152. Reisberg B, Borenstein J, Salob SP, Ferris SH, Franssen E, Georgotas A. Behavioral symptoms in
Alzheimer's disease: phenomenology and treatment. The Journal of clinical psychiatry. 1987;48
Suppl:9-15.
72
153. Levine DW, Kripke DF, Kaplan RM, et al. Reliability and validity of the Women's Health
Initiative Insomnia Rating Scale. Psychological assessment. 2003;15(2):137-148.
154. Allen RP PD, Hening WA, Trenkwalder C, Walters AS, Montplaisi J, . Restless Legs Syndrome
Diagnosis and Epidemiology Workshop at the National Institutes of Health, International Restless
Legs Syndrome Study Group: Restless legs syndrome: diagnostic criteria, special considerations,
and epidemiology. A report from the restless legs syndrome diagnosis and epidemiology
workshop at the National Institutes of Health. Sleep medicine. 2003;4:101–119.
155. Rothdach AJ, Trenkwalder C, Haberstock J, Keil U, Berger K. Prevalence and risk factors of RLS
in an elderly population: the MEMO study. Memory and Morbidity in Augsburg Elderly.
Neurology. 2000;54(5):1064-1068.
156. WHO. World Health Organization International statistical classification of diseases and related
health problems. http://www.who.int/classification/icd/en.
157. APA. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders.
4th ed. Text revised. . Washington, DC: American Psychiatric Association; 2000. 2000.
158. McKhann G DD, Folstein M, Katzman R, Price D, Stadlan EM Clinical diagnosis of
Alzheimer’s disease: report of the NINCDS-ADRDA work group under the auspices of
Department of Health and Human Services Task Force on Alzheimer’s Disease. . Neurology 34:.
1984;34:939-944.
159. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive
impairment: clinical characterization and outcome. Archives of neurology. 1999;56(3):303-308.
160. Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical method for grading the
cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189-198.
161. Delis DC, Kramer JH, Kaplan E, &, Ober BA. California Verbal Learning Test: Second Edition.
San Antonio, TX Psychological Corporation. 2000.
162. Shin MS, Park SY, Park SR, Seol SH, Kwon JS. Clinical and empirical applications of the Rey-
Osterrieth Complex Figure Test. Nature protocols. 2006;1(2):892-899.
73
163. Wechsler D. The Measurement of Adult Intelligence. Baltimore (MD). Williams & Witkins. 1939:
p. 229.
164. Yesavage JA, Brink TL, Rose TL, et al. Development and validation of a geriatric depression
screening scale: a preliminary report. J Psychiatr Res. 1982;17(1):37-49.
165. Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities
of daily living. The Gerontologist. 1969;9(3):179-186.
166. Teng EL, Chui HC. The Modified Mini-Mental State (3MS) examination. The Journal of clinical
psychiatry. 1987;48(8):314-318.
167. Jack CR, Jr., Knopman DS, Jagust WJ, et al. Hypothetical model of dynamic biomarkers of the
Alzheimer's pathological cascade. Lancet neurology. 2010;9(1):119-128.
168. Mintun MA, Larossa GN, Sheline YI, et al. [11C]PIB in a nondemented population: potential
antecedent marker of Alzheimer disease. Neurology. 2006;67(3):446-452.
169. Meyer-Lindenberg A, Buckholtz JW, Kolachana B, et al. Neural mechanisms of genetic risk for
impulsivity and violence in humans. Proceedings of the National Academy of Sciences of the
United States of America. 2006;103(16):6269-6274.
170. Agnew HW, Jr., Webb WB, Williams RL. The first night effect: an EEG study of sleep.
Psychophysiology. 1966;2(3):263-266.
171. Browman CP, Cartwright RD. The first-night effect on sleep and dreams. Biological psychiatry.
1980;15(5):809-812.
172. Webb WB, Campbell SS. The first night effect revisited with age as a variable. Waking and
sleeping. 1979;3(4):319-324.
173. Locke PA, Conneally PM, Tanzi RE, Gusella JF, Haines JL. Apolipoprotein E4 allele and
Alzheimer disease: examination of allelic association and effect on age at onset in both early- and
late-onset cases. Genetic epidemiology. 1995;12(1):83-92.
174. Peacock ML, Fink JK. ApoE epsilon 4 allelic association with Alzheimer's disease: independent
confirmation using denaturing gradient gel electrophoresis. Neurology. 1994;44(2):339-341.
74
175. Poirier J. Apolipoprotein E in animal models of CNS injury and in Alzheimer's disease. Trends in
neurosciences. 1994;17(12):525-530.
176. Schmechel DE, Saunders AM, Strittmatter WJ, et al. Increased amyloid beta-peptide deposition
in cerebral cortex as a consequence of apolipoprotein E genotype in late-onset Alzheimer disease.
Proceedings of the National Academy of Sciences of the United States of America.
1993;90(20):9649-9653.
177. Strittmatter WJ, Saunders AM, Schmechel D, et al. Apolipoprotein E: high-avidity binding to
beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease.
Proceedings of the National Academy of Sciences of the United States of America.
1993;90(5):1977-1981.
178. Henderson AS, Easteal S, Jorm AF, et al. Apolipoprotein E allele epsilon 4, dementia, and
cognitive decline in a population sample. Lancet. 1995;346(8987):1387-1390.
179. Reisberg B, Ferris SH, de Leon MJ, Crook T. The Global Deterioration Scale for assessment of
primary degenerative dementia. The American journal of psychiatry. 1982;139(9):1136-1139.
180. Jin X, Shearman LP, Weaver DR, Zylka MJ, de Vries GJ, Reppert SM. A molecular mechanism
regulating rhythmic output from the suprachiasmatic circadian clock. Cell. 1999;96(1):57-68.
181. Reppert SM, Weaver DR. Coordination of circadian timing in mammals. Nature.
2002;418(6901):935-941.
182. Peyron C, Tighe DK, van den Pol AN, et al. Neurons containing hypocretin (orexin) project to
multiple neuronal systems. The Journal of neuroscience : the official journal of the Society for
Neuroscience. 1998;18(23):9996-10015.
183. Saper CB, Fuller PM, Pedersen NP, Lu J, Scammell TE. Sleep state switching. Neuron.
2010;68(6):1023-1042.
184. Bhat RV, Budd Haeberlein SL, Avila J. Glycogen synthase kinase 3: a drug target for CNS
therapies. Journal of neurochemistry. 2004;89(6):1313-1317.
75
185. Bhat RV, Budd SL. GSK3beta signalling: casting a wide net in Alzheimer's disease. Neuro-
Signals. 2002;11(5):251-261.
186. Li X, Bijur GN, Jope RS. Glycogen synthase kinase-3beta, mood stabilizers, and neuroprotection.
Bipolar disorders. 2002;4(2):137-144.
187. Pei JJ BE, Braak H, Grundke-Iqbal I, Iqbal K, Winblad B, et al. . Distribution of active glycogen
synthase kinase 3beta (GSK- 3beta) in brains staged for Alzheimer disease neurofibrillary
changes. . J Neuropathol Exp Neurol 1999; . 1999;58::1010–1019.
188. Matsubara E, Bryant-Thomas T, Pacheco Quinto J, et al. Melatonin increases survival and
inhibits oxidative and amyloid pathology in a transgenic model of Alzheimer's disease. Journal of
neurochemistry. 2003;85(5):1101-1108.
189. Cummings JL BD. Dementia: A Clinical Approach, 2nd Ed. Boston: Butterworth, . 1992:pp 1–
17.
190. Association. AP. Diagnostic and Statistical Manual of Mental Disorders, 3rd Ed., Revised.
Washington, DC: . American Psychiatric Association,. 1987.
191. Bliwise DL. Is sleep apnea a cause of reversible dementia in old age? Journal of the American
Geriatrics Society. 1996;44(11):1407-1409.
192. Redline S, Strauss ME, Adams N, et al. Neuropsychological function in mild sleep-disordered
breathing. Sleep. 1997;20(2):160-167.
193. Ancoli-Israel S, Kripke DF, Klauber MR, Mason WJ, Fell R, Kaplan O. Sleep-disordered
breathing in community-dwelling elderly. Sleep. 1991;14(6):486-495.
194. Tison F, Crochard A, Leger D, Bouee S, Lainey E, El Hasnaoui A. Epidemiology of restless legs
syndrome in French adults: a nationwide survey: the INSTANT Study. Neurology.
2005;65(2):239-246.
195. Boeve BF, Molano JR, Ferman TJ, et al. Validation of the Mayo Sleep Questionnaire to screen
for REM sleep behavior disorder in an aging and dementia cohort. Sleep medicine.
2011;12(5):445-453.
76
196. Rose KM, Beck C, Tsai PF, et al. Sleep disturbances and nocturnal agitation behaviors in older
adults with dementia. Sleep. 2011;34(6):779-786.
197. Allen RP, Kushida CA, Atkinson MJ. Factor analysis of the International Restless Legs
Syndrome Study Group's scale for restless legs severity. Sleep medicine. 2003;4(2):133-135.
198. Beckett LA. Community-based studies of Alzheimer's disease: statistical challenges in design and
analysis. Statistics in medicine. 2000;19(11-12):1469-1480.
199. Clayton D SD, Dunn G et al. . Analysis of longitudinal binary data from multi-phase sampling. . J
R Stat Soc Series B 1998;. 1998;60::71–87.
200. Gao S, Hui SL. Estimating the incidence of dementia from two-phase sampling with non-
ignorable missing data. Statistics in medicine. 2000;19(11-12):1545-1554.
201. Van Den Berg JF, Van Rooij FJ, Vos H, et al. Disagreement between subjective and actigraphic
measures of sleep duration in a population-based study of elderly persons. Journal of sleep
research. 2008;17(3):295-302.
202. Lim DC, Veasey SC. Neural injury in sleep apnea. Curr Neurol Neurosci Rep. 2010;10(1):47-52.
203. McCoy JG, Strecker RE. The cognitive cost of sleep lost. Neurobiology of learning and memory.
2011;96(4):564-582.
204. Quan SF, Wright R, Baldwin CM, et al. Obstructive sleep apnea-hypopnea and neurocognitive
functioning in the Sleep Heart Health Study. Sleep medicine. 2006;7(6):498-507.
205. Cross RL, Kumar R, Macey PM, et al. Neural alterations and depressive symptoms in obstructive
sleep apnea patients. Sleep. 2008;31(8):1103-1109.
206. Sachs-Ericsson N, Joiner T, Plant EA, Blazer DG. The influence of depression on cognitive
decline in community-dwelling elderly persons. The American journal of geriatric psychiatry :
official journal of the American Association for Geriatric Psychiatry. 2005;13(5):402-408.
77
207. Atkins D, Best D, Briss PA, et al. Grading quality of evidence and strength of recommendations.
BMJ (Clinical research ed.). 2004;328(7454):1490.
78
Table 1.1 Literature Search Terms for the Systematic Review of Sleep and Alzheimer’s disease
Sleep Related Alzheimer's Disease Related Alzheimer's disease pathology related Study Types
Sleep (MeSH) Alzheimer's disease Dementia Amyloid beta Cross sectional
Sleep loss Cognitive Decline Amyloid beta deposition/generation Prevalence studies
Sleep deprivation Cognitive Impairment Alzheimer's disease pathogenesis Experimental studies
Sleep disruption Mild Cognitive Impairment (MeSH) Amyloid Precursor Protein Randomized Control Trials
Sleep apnea MCI Amyloid cascade hypothesis Incidence studies
Obstructive sleep apnea aMCI Inflammation Epidemiologic studies
Sleep Disordered Breathing Cognition P-Tau Case control
Excessive Daytime sleepiness Alzheimer (MeSH) Phosphorylated Tau Prospective Cohort
Sleep Disorders (MeSH) Hippocampal atrophy Retrospective cohort
Rapid Eye Movement Behavioral Disorder Apolipoprotein E Historical cohort
NREM Slow wave sleep APOE4 Closed cohort
Hypoxia Case-Comparison
Sleep Dependent Memory Nested case control
Sleep quality Case-Referent Study
Sleep function Matched case control
Wakefulness Cochrane database of systematic reviews
Circadian Rhythm (MeSH) Meta-analysis
Melatonin Synthesis
Hypocretin Research integration
79
Table 1.2: Scoring system for quality of paper of the Systematic Review of Sleep and Alzheimer's disease
· Hypotheses (maximum possible=1 star)
o The hypothesis/research question of the study not clearly described
o The hypothesis/research question of the study clearly described ¨
· Sampling (maximum possible= 2 stars)a
o Sample source and size
Ø Routine data=0
Ø Sample size: less than 50 per comparison groups
Ø Sample size: more than 50 per comparison groups ¨
o Type of sampling
Ø Consecutive patients or opportunistic
Ø Random representative of populations studied ¨
· Type of Study (maximum possible=3 stars)
o Ecological Study
o Case-control/Cross-sectional ¨
o Retrospective/Prospective Cohort ¨¨
o Randomized Control Trial/Experimental ¨¨¨
· Selection (Cohort studies) (maximum possible=4 stars)
o Representativeness of the exposed cohort
Ø Truly representative of the average distribution of sleep problems in the community ¨
Ø Somewhat representative of the average distribution of sleep problems in the community ¨
Ø Selected group of users e.g., nurses, volunteers
Ø No description of the derivation of the cohort
o Selection of the non-exposed cohort
Ø Drawn from the same community as the exposed cohort ¨
Ø Drawn from a different source
Ø No description of the derivation of the non-exposed cohort
o Length of follow up of cohorts to determine outcome
Ø Follow up period greater than 5 years ¨
Ø Follow up period less than 5 years
o Adequacy of follow up of cohorts (Likelihood to introduce bias)
Ø Complete follow up- all subjects accounted for ¨
Ø Follow up rate ≥ 50% and description of loss to follow up subjects
Ø Follow up rate < 50% and no description of loss to follow up subjects
Ø Follow up rate not mentioned
Table continues
80
Table 1.2 continued
· Selection (Case-control studies) (maximum possible=4 stars)
o Is the case definition adequate?
Ø Yes, with independent validation ¨
Ø Yes, e.g., record linkage or based on self-reports ¨
Ø No description
o Representativeness of the cases
Ø Consecutive or obviously representative series of cases ¨
Ø Potential for selection biases or not stated
o Selection of controls
Ø Community controls ¨
Ø Hospital controls
Ø No description of source
o Non response rate
Ø Same rate for both groups ¨
Ø Non respondents described
Ø Rate different and no designation
· Selection (Experimental studies) (maximum possible=4 stars)
o Appropriate description of experimental mice/population
Ø Yes, with transgenic status described ¨
Ø No description
o Appropriate description of experimental procedures
Ø Yes, with application of standard protocols ¨
Ø Standardized protocols not adhered to.
o Controls or comparison group
Ø Comparison group present ¨
Ø No controls used
o Comparability of experimental conditions
Ø Mice/population in both groups exposed to similar experimental conditions except for exposure/intervention ¨
Ø Experimental conditions not described
Ø Conditions different and no designation
Table continues
81
Table 1.2 continued
· Methodology (maximum possible=5 stars)
o Definition/ Operationalization of Alzheimer's Disease/Dementia/Cognitive Decline(2)b
Ø No standardized measures (e.g. self-reported diagnosis)
Ø Standardized Operationalized criteria (e.g. ICD-9, ICD-10, DSM-IV)* ¨
Ø Objective measures (e.g., PET scan, MRI) * ¨¨
o Definition/Operationalization of Sleep and its related abnormalities/disorders
Ø No standardized measures (e.g. self-reported diagnosis) =0
Ø Standardized Operationalized criteria/questionnaire (e.g. ICD-9, ICD-10, DSM-IV, PSQI, ESS)^ ¨
Ø Objective measures (e.g., Actigraph, Polysomnography)^ ¨¨
o Study Population (Not applicable to Animal Experimental studies)
Ø No appropriate description of study population
Ø In-patient non-population based
Ø Study population is population based ¨
· Statistical analyses and results (maximum possible=5 stars)
o Statistical analysis
Ø Statistical methods not described =0
Ø Statistical methods clearly described ¨
o Statistical outcomes
Ø Statistical outcome not clearly reported
Ø Statistical outcomes clearly reported (e.g. OR, RR, CIs) ¨
o Other explanatory variables in the analysis (e.g. confounders) (4)c
Ø No explanatory variables involved
Ø Age and/or gender controlled for ¨
Ø Age, gender, SES and risk factors (sleep related and AD related covariates) ¨¨
o Results
Ø Research questions/hypotheses not clearly answered/tested
Ø Research questions/hypotheses clearly answered/tested ¨
a: Sample size values. The sample size value of 50 participants per group has been chosen as these permits reasonable estimates of rates of CF. Not Applicable to Animal Experimental studies
b: ICD-9/10 International Classification of Disease version 9/10; DSM-IV Diagnostic Statistical Manual-IV; PQSI Pittsburg Sleep Quality Index; ESS Epworth Sleepiness Scale c: Adjustment for confounding variables. From an overview of studies on sleep and Alzheimer's Disease; Age, years of education, marital status, forced expiratory volume in one second based on spirometry, self-reported history of physician’s diagnosis of asthma, number of hours of sleep and napping, and use of benzodiazepines. Chronic obstructive pulmonary disease, Nonfatal coronary heart disease including myocardial infarction, Stroke, The ankle-brachial index calculated as the mean of the posterior tibial pressure on each side divided by the mean of the two right brachial pressures. Apolipoprotein E genotype, BMI and Depression were all-important covariates to adjust for. Not Applicable to Animal Experimental studies
* Must be independent blind assessment or made use of record linkage to receive a star
^ Must be use of secure record (eg surgical records) or made use of structured interview where blind to case/control status to receive a star
82
Table 1.3 : Quality Assessment of Cross-sectional studies
First Author, Country, Year Published
(Reference)
Hypotheses
(maximum
possible=1 star)
Sampling
(maximum
possible=2 stars)
Type of Study
(maximum
possible=3
stars)
Methodology
(maximum
possible=5 stars)
Statistical
Analysis
(maximum
possible=5 stars)
Total (maximum
possible=16
stars)
Quality Rating
Blackwell, United States, 2006 () ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 13 High
Nebes, United States, 2009 () ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 10 Medium
Saint Martin, France, 2012 () ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 10 Medium
Guarnieri, Italy, 2012 () ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 9 Medium
Moran, Ireland, 2005 () ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 9 Medium
Pistacchi, Italy, 2014 () ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 8 Low
Fetveit, Norway, 2006 () ♦
♦ ♦ ♦ ♦ ♦ ♦ 7 Low
Ju, United States, 2013 () ♦
♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 9 Medium
Spira, United States, 2013 () ♦
♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 10 Medium
Craig, Northern Ireland, UK, 2006 () ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 8 Low
Oosterman, Netherlands, 2009 () ♦
♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 9 Medium
Merlino, Italy, 2010 () ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 8 Low
Pat-Horenczyk, United States, 1998 () ♦
♦ ♦ ♦ ♦ ♦ ♦ 7 Low
Yesavage, United States & France, 2011 () ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 8 Low
Gottlieb, United States, 2004 () ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 13 High
Kadotani, United States, 2001 () ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 12 High
Osorio, United States, 2014 () ♦
♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 10 Medium
Djonlagic, United States, 2014 () ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ 8 Low
83
Table 1.4: Quality Assessment Scores of Case-control studies
First Author, Country, Year
Published (Reference)
Hypotheses
(maximum
possible=1 star)
Sampling
(maximum
possible=2 stars)
Type of Study
(maximum
possible=3
stars)
Selection (Case-
control study)
(maximum
possible=4 stars)
Methodology
(maximum
possible=5
stars)
Statistical
Analysis
(maximum
possible=5
stars)
Total
(maximum
possible=20
stars)
Quality Rating
Chen, Taiwan, 2011 () ♦
♦ ♦♦♦ ♦♦♦ ♦♦ 10 Low
Hita-Yañez, Spain, 2013 () ♦
♦ ♦♦♦ ♦♦♦ ♦♦♦ 11 Medium
Hita-Yañez, Spain, 2012 () ♦
♦ ♦♦♦ ♦♦♦♦ ♦♦♦ 12 Medium
Hot, France, 2011 () ♦
♦ ♦♦ ♦♦♦ ♦♦ 9 Low
Sanchez-Espinosa, Spain, 2014 () ♦
♦ ♦♦♦♦ ♦♦♦♦ ♦♦♦ 13 Medium
Naismith, Australia, 2014 () ♦
♦ ♦♦♦ ♦♦♦ ♦♦ 10 Low
Fronczek, Netherland, 2012 () ♦
♦ ♦♦♦ ♦♦♦♦ ♦♦ 11 Medium
Slats, Netherlands, 2012 () ♦
♦ ♦♦ ♦♦♦♦ ♦♦♦♦ 12 Medium
Schmidt, Germany, 2013 () ♦
♦ ♦♦ ♦♦♦♦ ♦♦♦ 11 Medium
Liu, Netherlands and China, 1999 () ♦ ♦ ♦ ♦♦ ♦♦♦♦ ♦♦♦ 12 Medium
O’Hara, United States, 2005 () ♦
♦ ♦♦ ♦♦♦♦ ♦♦ 10 Low
Pearson, United States, 2006 () ♦ ♦ ♦♦ ♦♦♦ ♦♦♦ 10 Low
84
Table 1.5: Quality Assessment Scores of Cohort studies/Randomized Clinical Trials
First Author, Country, Year
Published (Reference)
Hypotheses
(maximum
possible=1 star)
Sampling
(maximum
possible=2
stars)
Type of
Study
(maximum
possible=3
stars)
Selection
(Cohort studies)
(maximum
possible=4 stars)
Methodology
(maximum
possible=5 stars)
Statistical
Analysis
(maximum
possible=5
stars)
Total
(maximum
possible=20
stars)
Quality
Rating
Keage, Australia, & United Kingdom
2012 () ♦ ♦♦ ♦♦ ♦♦♦♦ ♦♦♦ ♦♦♦♦ 16 High
Tworoger, United States, 2006 () ♦ ♦♦ ♦♦ ♦♦ ♦♦♦ ♦♦♦♦ 14 Medium
Virta, Finland, 2013 () ♦ ♦♦ ♦♦ ♦♦♦♦ ♦♦♦ ♦♦♦ 15 High
Song, United States, 2014 () ♦ ♦♦ ♦♦ ♦♦♦ ♦♦♦ ♦♦♦♦ 15 High
Blackwell, United States, 2014 () ♦ ♦♦ ♦♦ ♦♦♦ ♦♦♦ ♦♦♦♦♦ 16 High
Sterniczuk, Canada, 2013 () ♦ ♦♦ ♦♦ ♦♦♦ ♦♦ ♦♦♦♦ 14 Medium
Benito-León, Spain, 2014 () ♦ ♦♦ ♦♦ ♦♦♦♦ ♦♦ ♦♦♦♦♦ 16 High
Hahn, United States, 2013 () ♦ ♦ ♦♦ ♦♦♦♦ ♦♦ ♦♦♦♦♦ 15 High
Bidzan, Poland, 2011 () ♦
♦♦ ♦♦ ♦♦ ♦♦ 9 Low
Lim, Canada, 2013 () ♦ ♦ ♦♦ ♦♦♦ ♦♦♦ ♦♦♦ 13 Medium
Ooms, Netherlands, 2014 () ♦ ♦ ♦♦♦ ♦♦ ♦♦♦♦ ♦♦♦♦ 15 High
Walsh, United States 2014 () ♦ ♦♦ ♦♦ ♦♦♦ ♦♦♦ ♦♦♦♦ 15 High
Anderson, United Kingdom, 2014 () ♦ ♦♦ ♦♦ ♦♦ ♦♦♦ ♦♦♦ 13 Medium
Tranah, United States, 2011 () ♦ ♦♦ ♦♦ ♦♦♦ ♦♦♦ ♦♦♦♦♦ 16 High
Lim, United States, 2013 () ♦ ♦♦ ♦♦ ♦♦ ♦♦♦ ♦♦♦♦♦ 15 High
Yesavage, United States, 2004 () ♦
♦♦ ♦♦ ♦♦♦♦ ♦♦ 11 Medium
Asayama, Japan, 2003 () ♦ ♦ ♦♦♦ ♦♦ ♦♦♦ ♦♦♦♦ 14 Medium
Cohen-Zion, United States, 2001 () ♦
♦♦ ♦♦ ♦♦♦ ♦♦♦ 10 Low
Chang, Taiwan, 2013 () ♦ ♦♦ ♦♦ ♦♦♦ ♦♦♦ ♦♦♦♦ 15 High
85
Yaffe, United States, 2011 () ♦ ♦ ♦♦ ♦♦ ♦♦♦♦ ♦♦♦♦♦ 15 High
Kushida, United States, 2012 () ♦ ♦ ♦♦♦ ♦♦ ♦♦♦♦ ♦♦♦♦ 15 High
Ancoli-Israel, United States, 2008 () ♦ ♦ ♦♦♦ ♦♦ ♦♦♦♦ ♦♦♦♦ 15 High
Cricco, Iceland, 2001 () ♦ ♦♦ ♦♦ ♦♦ ♦♦♦ ♦♦♦♦♦ 15 High
Jelicic, The Netherlands, 2002 () ♦ ♦♦ ♦♦ ♦♦ ♦♦♦ ♦♦♦ 13 Medium
Foley, United States, 2001 () ♦ ♦♦ ♦♦ ♦♦ ♦♦♦ ♦♦♦♦♦ 15 High
Osorio, United States, 2011 () ♦ ♦♦ ♦♦ ♦♦♦ ♦♦♦ ♦♦♦♦ 15 High
Table 1.6: Quality Assessment Scores of Experimental studies
First Author, Country, Year Published
(Reference)
Hypotheses
(maximum
possible=1
star)
Type of Study
(maximum
possible=3 stars)
Selection
(Experimental
studies)
(maximum
possible=4 stars)
Methodology
(maximum
possible=4
stars)
Statistical
Analysis
(maximum
possible=3
stars)
Total
(maximum
possible=15
stars)
Quality Rating
Rothman, United States, 2013 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Di Meco, United States, 2014 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Kang, United States, 2009 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Mander, United States, 2013 ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦♦♦♦ ♦ ♦ 13 High
Craig, Canada, 2008 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Jyoti, Scotland, UK, 2010 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Quin, United States, 2005 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
DENG. China, 2005 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Papolla, United States, 1998 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Olcese, United States, 2009 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Lahiri, United States, 2004 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
86
Shiota, Japan, 2013 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Kaushal, United States, 2012 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Sun, Canada, China, United States, 2006
() ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Li, China, 2009 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Zhang, China, 2007 () ♦ ♦ ♦ ♦ ♦♦♦♦ ♦♦♦♦ ♦ 13 High
Table 1.7. Sleep Depth and Cognitive/Alzheimer's Disease Measures: Descriptive Study Characteristics, and Main Findings
First Author,
Country, Year
Published (Reference)
Study Design Sample size N Study Population Sleep Depth
Assessment
Cognitive/AD Measures
Assessment Conclusion
Blackwell, United States, 2006 ()
Cross sectional N=2,242 Community-dwelling women 65 years old or older
Actigraphy
Mini-Mental State Examination (MMSE) and the Trail Making B Test (Trails B)
Objectively measured disturbed sleep was consistently related to poorer cognition, whereas total sleep time was not.
Nebes, United States, 2009 ()
Cross-sectional N=157 Participants were 65–80 years old community volunteers
Pittsburgh Sleep Quality Index
[PSQI]
Repeatable Battery for the Assessment of Neuropsychological Status (RBANS). Test of Nonverbal Intelligence (TONI))
Sleep problems may contribute to performance variability between elderly individuals but only in certain cognitive domains.
87
Guarnieri, Italy, 2012 () Cross-sectional N= 431
Participants were patients enrolled consecutively in 10 Italian neurological centers: 204 had Alzheimer’s disease (AD), 138 mild cognitive impairment (MCI), 43 vascular dementia (VaD), 25 frontotemporal dementia and 21 Lewy body dementia (LBD) or Parkinson’s disease dementia (PDD).
PSQI, Berlin questionnaire and the International
Restless Leg Syndrome (RLS)
Study Group questionnaire
DSM-IV-TR criteria. AD and Vascular Dementia (VaD) were diagnosed respectively according to NINCS-ARDRA and NINDS-AIREN criteria. LBD and PDD, frontotemporal dementia (FTD) and MCI were diagnosed according to specific consensus guidelines
A careful clinical evaluation of sleep disorders should be performed routinely in the clinical setting of persons with cognitive decline. Instrumental supports should be used only in selected patients
Pistacchi, Italy, 2014 () Cross-sectional N= 236
236 patients (78 men and 158 women) were enrolled with different subtypes of dementia: (AD), (VaD), mixed dementia (MCI), dementia with Lewy bodies (DLB), (PDD), and frontotemporal lobar degeneration (FTLD)
PSQI, Berlin questionnaire and the International
Restless Leg Syndrome (RLS)
Study Group questionnaire
DSM-IV-TR criteria . AD and VaD were diagnosed respectively according to NINCS-ARDRA and NINDS-AIREN criteria . LBD and PDD, frontotemporal dementia (FTD) and MCI were diagnosed according to specific consensus guidelines .
Findings demonstrate that sleep disturbance was related to dementia
Fetveit, Norway, 2006 ()
Cross-sectional N=23 Nursing home residents with varying degree of dementia
Actigraphy
Grades of dementia were assessed by Mini-Mental Status Examination Score (MMSE) and Clinical Dementia Rating (CDR).
Sleep duration during the 24-h day was positively correlated with the severity of dementia in nursing home patients.
88
Ju, United States, 2013 ()
Cross-sectional N=142
Longitudinal studies of memory and aging at the Washington University Knight Alzheimer Disease Research Center (ADRC). The Adult Children Study, in which all were age 45-75 years at baseline and 50% have a parental history of sAD.
Actigraphy
Aβ42 was measured by the Alzheimer's Disease Research Center (ADRC) Biomarker Core using enzyme linked immunosorbant assay (INNOTEST; Innogenetics, Ghent, Belgium).
Amyloid deposition in the preclinical stage of AD appears to be associated with worse sleep quality, but not with changes in sleep quantity.
Spira, United States, 2013 ()
Cross-sectional N= 70
Participants were adults (mean age, 76 [range, 53-91] years) from the neuroimaging sub-study of the Baltimore Longitudinal Study of Aging, a normative aging study.
Self-reported sleep variables: The 5-item Women’s
Health Initiative Insomnia Rating Scale (WHIIRS)
β-Amyloid burden, measured by carbon 11–labeled Pittsburgh compound B positron emission tomography distribution volume ratios (DVRs).
Among community-dwelling older adults, reports of shorter sleep duration and poorer sleep quality are associated with greater Aβ burden.
Craig, Northern Ireland, UK, 2006 ()
Cross-sectional N=426
Subjects were identified from outpatient memory-clinic records based at the Belfast City Hospital Trust, Mater Infirmorum, and Holywell Hospital. Patients fulfilled the criteria for probable AD.
A validated neuropsychological
assessment tool, the
Neuropsychiatric Inventory with
Caregiver Distress (NPI-D), was
employed to record sleep disturbances during the course of the dementing
illness.
Outpatient memory-clinic records. DNA extracted from peripheral blood leukocytes was analyzed for the APOE and MAO-A 30 bp VNTR polymorphisms
Sleep disturbance in AD is common and distressing and is associated with genetic variation at MAO-A.
89
Hita-Yañez, Spain, 2012 ()
Case-control N=50
Twenty-five MCI patients (7 females, mean age: 70.5 ± 6.8 yr) and 25 healthy old (HO) volunteers (13 females, mean age: 67.1 ± 5.3 yr)
Polysomnography (PSG)
ApoE genotype was determined by conventional PCR (polymerase chain reaction) methods. The diagnosis of MCI was based on consensus criteria. Global cognitive status was assessed with the Mini Mental State Examination (MMSE).
Sleep disruptions are evident years before diagnosis of AD, which may have implications for early detection of dementia and/or therapeutic management of sleep complaints in MCI patients.
Chen, Taiwan, 2011 () Case-control N=21
Authors recruited 9 controls without dementia, 6 patients with mild cognitive impairment (MCI), and 6 patients with mild Alzheimer's disease (AD). None of the participants had sleep complaints, and all AD patients were receiving cholinesterase inhibitors.
Polysomnography (PSG)
MMSE. NINCDS-ADRDA. All patients with AD were taking acetyl cholinesterase inhibitors (AChEI)
A deficiency of Ach may result in an increase of sEMG activity in MCI patients.
Hita-Yañez, Spain, 2013 ()
Case-control N=50
Twenty-five patients with MCI (7 females, mean age: 70.5 ± 6.8 y) and 25 HE subjects (13 females, mean age: 67.1 ± 5.3 y) were enrolled in the study,
Overnight PSG recordings and
self-reported sleep measures were
obtained
The diagnosis of MCI was based on consensus criteria. Global cognitive status was assessed with the Mini Mental State Examination (MMSE).
Sleep is significantly impaired in patients with mild cognitive impairment at both the objective and subjective level, which may be used as a surrogate marker of preclinical Alzheimer disease.
90
Hot, France, 2011 () Case-control N=28
Fourteen un-medicated AD patients (seven men and seven women; mean age (±SD): 76.7 ± 3.8 years) participated in this study
Polysomnography (PSG)
Episodic memory was assessed with an original task derived from Grober and Buschke’s procedure (1987).This task consisted in learning 15 words, presented by series of three on separate cards
Changes in theta rhythm during REM and SWS are a relevant index of brain impairments in the early stage of AD, and are related to episodic memory performance.
Sanchez-Espinosa, Spain, 2014 ()
Case-control N=21
Participants were primarily recruited from older people's associations, normal community health screening, and hospital outpatient services.
Polysomnography (PSG)
Neuropsychological testing; diagnostic criteria of aMCI proposed by Petersen et al. Plasma Aβ levels. Cerebral MRI acquisition, image preprocessing and cortical thickness estimation was done for each participant
Increased plasma Aβ42 levels are significantly associated with fragmented SWS in aMCI subjects, suggesting that sleep disruptions may signal Aβ burden in persons at increased risk for AD. Reduced REM sleep and plasma Aβ levels in aMCI subjects were significantly related to thinning of cortical regions targeted by AD neuropathology.
Keage, Australia, & United Kingdom 2012 ()
Prospective cohort N= 2012
The Medical Research
Council Cognitive Function
and Ageing Study (CFAS), a longitudinal multi-center population-based study of ageing and dementia (www.cfas.ac.uk). . Follow-up
time: 10-years
Self-reported measures .
MMSE
Daytime napping, night-time sleep duration, and excessive daytime sleepiness may be modifiable behaviors open to intervention strategies, or, clinical indicators of future decline in older individuals.
91
Tworoger, United States, 2006 ()
Prospective cohort N= 1844
Participants (all women) were members of the Nurses’
Health Study cohort aged 70 to 81 years at initial cognitive interview in 2000. Follow-up
period: 2years
Self-reported measures .
Women completed six tests of cognitive function encompassing general cognition, verbal memory, category fluency, and attention. Tests were repeated 2 years later.
Diminished cognition is a risk factor for dementia. However, the lack of an association with prospective cognitive decline warrants further investigation
Virta, Finland, 2013 () Prospective cohort N= 2,336
The subjects of the current study were members of the older Finnish Twin Cohort, consisting of same-sex twin pairs born in Finland prior to 1958 with both co-twins alive in 1967. Follow-up time: 18-
26years
Self-reported measures .
Between 1999 and 2007, participants were assigned a linear cognitive score with a maximum score of 51 based on a telephone interview (mean score 38.3, Standard Deviation (SD) 6.1).
This is the first study indicating that midlife sleep length, sleep quality, and use of hypnotics are associated with late life cognitive function. .
Song, United States, 2014 ()
Prospective cohort N = 2,601
The Osteoporotic Fractures in Men Study (MrOS). Participants were age 65 and older and free of probable dementia at sleep visit. Follow-up averaged 3.4 years
Sleep stages were identified by in-
home polysomnography at the initial sleep visit (2003-2005).
Cognitive outcomes were assessed with the Trail Making Test Part B and Modified Mini-Mental State Examination (3MS) at sleep visit and two follow-up
Increased time in Stage N1 and less time in Stage R are associated with worsening cognitive performance in older men over time
Blackwell, United States, 2014 ()
Prospective Cohort N=2822
The Osteoporotic Fractures
in Men Study (MrOS). Participants were age 65 and older (mean age 76.0 ± 5.3 y)and free of probable dementia at sleep visit. Follow-up averaged 3.4 ± 0.5
y.
Actigraphy
Clinically significant cognitive decline: five-point decline on the Modified Mini-Mental State examination (3MS), change score for the Trails B test time in the worse decile.
Among older community-dwelling men, reduced sleep efficiency, greater nighttime wakefulness, greater number of long wake episodes, and poor self-reported sleep quality were associated with subsequent cognitive decline.
92
Sterniczuk, Canada, 2013 ()
Retrospective Cohort
N= 28,697
The Survey of Health,
Ageing and Retirement in
Europe (SHARE) (N = 30,038). Follow-up period:
4.3years
Self-reported sleep measures
Alzheimer’s disease/Dementia was reported by the participants themselves and/or by a proxy respondent .Cognition was assessed using performance-based cognitive tests
These findings indicate that sleep disturbance may exist prior to the manifestation of other typical symptoms observed in AD (e.g., memory loss).
Benito-León, Spain, 2014 ()
Prospective cohort N= 3,857
Neurological Disorders in
Central Spain (NEDICES) Study, a longitudinal, population-based survey followed for a median of 12.5
years
Self-reported sleep measures
ICD-9 codes for deaths that occurred before 1999 and ICD-10 codes for deaths occurring thereafter.
Self-reported long sleep duration was associated with 58% increased risk of dementia-specific mortality in this cohort of elders without dementia.
Hahn, United States, 2013 ()
Prospective cohort N= 214
The population-based
Kungsholmen Project, a longitudinal study of aging and dementia among adults aged 75 years and older residing in Kungsholmen district, Stockholm, Sweden. Follow-
up time: Up to 9-years
Comprehensive Psychopathological
Rating Scale (CPRS)
DSM-III-Revised and AD diagnosis was made using the NINCDS-ADRDA criteria
Self-reported sleep problems may increase the risk for dementia, and depressive symptoms may explain this relationship.
Bidzan, Poland, 2011 () Prospective cohort N=150
Care centers residents age 55 and older, who had had the presence of dementia excluded (n = 2910).. Follow-up time:
7-years
The Neuropsychiatric
Inventory-Nursing Home (NPI-NH)
scale and the Association for
Methodology and Documentation in
Psychiatry (AMDP) scale.
Alzheimer’s disease Assessment Scale Cognitive (ADAS-cog) scale. Dementia in the Alzheimer’s disease was diagnosed based on the NINCDS-ADRDA criteria.
The patients in the preclinical period of dementia experienced sleep disturbances more frequently and with greater intensity, which in combination with other factors may have some prognostic value.
93
Lim, Canada, 2013 () Prospective cohort N= 698
Participants were community-dwelling older adults without dementia (mean age, 81.7 years; 77% women) in the Rush Memory and Aging
Project. follow-up period of
up to 6 years.
Actigraphy
Neuropsychometric tests. NINCDS-ADRDA criteria. Autopsies were performed on 201 participants who died, and β-amyloid (Aβ) and neurofibrillary tangles were identified by immunohistochemistry and quantified. Peripheral blood lymphocytes were used for DNA extraction, and APOE genotype was determined
Better sleep consolidation attenuates the effect of APOE genotype on incident AD and development of neurofibrillary tangle pathology.
Rothman, United States, 2013 ()
Experimental Not Applicable
Male 3xTgAD mice were subjected to sleep restriction(SR) for 6h/day for 6 weeks using the modified multiple platform technique, and behavioral(Morris water maze, fear conditioning, open field)
Mice were tested in the Morris water
maze (MWM)
Hippocampal and cortical Aβ and P-Tau concentrations were obtained
Significant positive correlations between cortical Aβ and p -Tau levels and circulating corticosterone indicate a potential role for GCs in mediating behavioral and biochemical changes observed after sleep restriction in a mouse model of AD
Di Meco, United States, 2014 ()
Experimental Not Applicable
A total of 18 mice were kept in a pathogen-free environment, on a 12-hour light and/or dark cycle and had access to food and water ad libitum. Starting at the age of 8 months, mice were randomized to 2 groups, one underwent a 20/ 4-hour light and/or dark cycle for 2 months (4 males and 5 females), the other was kept on a normal light and/or dark cycle, as controls (5 males and 4 females).
The locomotor cage test was used to record the sleep-wake cycle of the animals. The fear conditioning tests. The Morris water-
maze test
Mouse brain homogenates were extracted in radio-immuno-precipitation assay (RIPA) buffer and were prepared for immunohistochemistry. Cd-5 kinase (Cdk-5) activity was carried out.
Authors conclude that alteration of the sleep-wake cycle and circadian rhythm regulation could be important players in the onset and development of sporadic AD. Correction of sleep-wake cycle aberrations could be a viable therapeutic strategy for individuals bearing this risk factor.
94
Kang, United States, 2009 ()
Experimental Not Applicable Both wild-type mice and human APP transgenic (Tg2576) mice
Mice were forced into wakefulness for 6 hours at the beginning of the second 12-hour
light period when they would
naturally be asleep.
Authors monitored hippocampal Aβ levels using in vivo micro dialysis in both wild-type mice and human APP transgenic (Tg2576) mice, which express a mutated form of human amyloid precursor protein (APP). ISF Aβ was assessed in Tg2576 mice at 3 months of age, several months earlier than Aβ deposition begins.
Sleep-wake cycle and orexin may play a role in the pathogenesis of Alzheimer’s disease.
Mander, United States, 2013
Experimental N=33
A group of cognitively normal older adults (n=18 mean age 20.4 ± 2.1) and a group of healthy young adults (n=15 mean age 72.1 ± 6.6)
Measured with Polysomnogram (PSG), starting at their habitual bed time to minimize the influence of
age-related circadian
differences.
Approximately 2 h post-awakening, participants performed an event-related fMRI scanning session while performing a long-delay (10 h) recognition test
Together, these data support a model in which age-related mPFC atrophy diminishes SWA, the functional consequence of which is impaired long-term memory.
Ooms, Netherlands, 2014 ()
Randomized Controlled Trial
N=26
The Alzheimer,
Wakefulness, and Amyloid
Kinetics (AWAKE) study at the Radboud Alzheimer Center, a randomized clinical trial that took place between
June 1, 2012, and October 1,
2012.
Sleep was monitored using
continuous polysomnographic recording from 3 PM until 10 AM.
Cerebrospinal fluid, Aβ42, P-tau, and T-tau were deter-mined using the xMAP-based Innobia assay (Innogenetics) and CSF Aβ40 was measured using an enzyme-linked immuno-sorbent assay
Sleep deprivation, or prolonged wakefulness, interferes with a physiological morning decrease in Aβ42. Chronic sleep deprivation increases cerebral Aβ42 levels, which elevates the risk of Alzheimer disease.
95
Table 1.8. Sleep-wake cycle abnormalities and Cognitive/Alzheimer's Disease Measures: Descriptive Study Characteristics, and Main
Findings
First Author,
Country, Year
Published
(Reference)
Study Design Sample
size N Study Population
Sleep-wake cycle
Assessment
Cognitive/AD Measures
Assessment Conclusion
Oosterman, Netherlands, 2009 ()
Cross-sectional (Correlational)
N= 144
Study was part of a larger study on the effects of cardiovascular risk factors, and participants (90 males and 54 females, aged 69.5 ± 8.5 years (mean ± SD), range 50–91) were recruited among people visiting the outpatient clinic (of cardiology, internal medicine, or neurology) of the Saint Lucas Andreas Hospital in Amsterdam, the Netherlands.
Wrist Actigraphy
Mental speed: Stroop test (Stroop, 1935). The Word (W) and Color (C) cards of the Stroop test Memory: 15-words test. The Dutch version (Saan and Deelman, 1986) of the Auditory Verbal Learning Test (Rey, 1964) .Digit span forward (Wechsler, 1987).
An association between the rest-activity rhythm and cognitive performance is present in elderly people.
Merlino, Italy, 2010 ()
Cross-sectional N=750
Subjects aged 65 years or older, living independently or in an institution, who resided in the seventh district of Udine, Italy were recruited.
Sleep disorders were investigated with a battery of standardized questions and questionnaires
The diagnosis of dementia was made according to the DSM-IV-TR criteria . AD and VaD were diagnosed respectively according to NINCS-ARDRA and NINDS-AIREN criteria . LBD and PDD, frontotemporal dementia (FTD) and MCI were diagnosed according to specific consensus guidelines .
Excessive daytime sleepiness was significantly related to dementia. Insomnia, was not associated with the presence of cognitive decline.
Pat-Horenczyk, United States, 1998 ()
Cross-sectional N= 67
Nursing home residents (mean age = 85.7 years). There were 46 severely demented patients, and 21 mild-moderately demented patients
Actigraphy Levels of dementia were assessed by Mini-Mental State Examination (MMSE)
With the progression of dementia, both the capacity to maintain sleep and the capacity to maintain wakefulness are impaired, and result in complete fragmentation of sleep/wakefulness during the night and day.
Yesavage, United States & France, 2011 ()
Cross-sectional N=300
Two cohorts of AD participants. Cohort 1 (n=124): individuals with probable AD recruited from the Stanford/Veterans Affairs NIA Alzheimer’s Disease Core Center (n=81) and the Memory Disorders Clinic at the University of Nice School of Medicine (n=43). Cohort 2 (n=176): individuals with probable AD derived from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) data set.
Wrist actigraphy data for seven days in Cohort 1 and by the Neuropsychiatric Inventory (NPI-Q) for Cohort 2.
Diagnosis of probable AD by NINCDS-ADRDA criteria based on relevant neurological, medical, neuroimaging, and neuropsychological assessments.
It is unlikely that a relationship with a clinically meaningful correlation exists between the circadian rhythm-associated SNPs and WASO in individuals with AD.
96
Naismith, Australia, 2014 ()
Case-control N=58
Participants (30 patients with MCI and 28 age-matched controls) were recruited from a specialist ‘Healthy Brain Ageing’ Clinic, at the Brain & Mind Research Institute, Sydney, Australia.
Actigraphy. Polysomnographic (PSG) assessments, on separate nights. Salivary melatonin was assessed to determine dim light melatonin onset (DLMO)
A clinical diagnosis of MCI was obtained using Petersen’s criteria
Circadian misalignment and sleep disruption is evident in patients with MCI, and is consistent with changes observed in Alzheimer’s disease. Such findings could be a marker for disease trajectory, and may even be implicated in disease pathogenesis.
Fronczek, Netherland, 2012 ()
Case-control N=49
Ten patients with AD (Braak 5 or 6) and 10 non-demented controls (Braak 0 or 1) were matched for sex, age, postmortem delay (PMD), and fixation time (hypothalamus group).
Hypocretin-1 immunocytochemistry and neuron quantification and measurement in ventricular CSF was done.
Clinical diagnosis and Neuropathological examination according to van de Nes et al. 1993 and Braak et al. 1998
Hypocretin system is affected in advanced AD. This is reflected in a 40% decreased cell number, and 14% lower CSF hypocretin-1 levels.
Slats, Netherlands, 2012 ()
Case-control N=12 Six patients with AD (59-85 yrs, MMSE 16-26) and six healthy volunteers (64-77 yrs).
CSF hypocretin-1 was measured using a commercially available RIA kit (Phoenix Pharmaceuticals, Belmont, CA).
CSF Aβ42 was determined using the xMAP-based Innobia assay (Innogenetics NV, Ghent, Belgium) CSF Aβ40 using ELISA (the Genetics Company, Schlieren, Switzerland).Total protein level in CSF was analyzed using the Lowry method.
Mean Aβ42 levels and mean hypocretin-1 levels and amplitude may suggest a relationship between AD pathology and hypocretin disturbance, which could hold possibilities for treatment of AD related sleep disorders.
Schmidt, Germany, 2013 ()
Case-control N=66
33 patients with mild to severe AD and 33 subjects without any psychiatric or neurological disorder (HS) that were consecutively recruited to participate in the study.
Melanin-concentrating hormone (MCH) and hypocretin-1 (HCRT-1, orexin-A) were measured using a fluorescence immunoassay (FIA)
Consortium to Establish a Registry for Alzheimer’s Disease (CERAD), Wechsler Memory Scale-Revision (WMS-R), clock drawing test, and Trail Making Test (TMT), a MRI head-scan, genotyping of ApoE4 and determination of Ab42, T-tau and P-tau in the CSF.
This report on MCH in patients with AD adds to preclinical studies and may allow further research on the role of the hypothalamus in pathology and occurrence of behavioral disturbances in AD.
97
Liu, Netherlands and China, 1999 ()
Case-control N=167
Autopsies were performed within the framework of The Netherlands Brain Bank. Ventricular postmortem CSF was obtained at autopsy, 1–12 h after death, from 85 patients with AD (mean age, 75 ± 1.1 yr) and 82 age-matched controls (mean age, 76 ± 1.4 yr) without a primary neurological or psychiatric disease.
Melatonin in postmortem CSF was measured by a direct RIA
ApoE genotyping. The genotype of each extracted DNA sample was determined by polymerase chain reaction (PCR)
CSF melatonin levels fromApoE-€3/4 genotype patients were significantly higher thanthose from the ApoE-€4/4 genotype, suggesting a relationship between melatonin levels and signs and symptoms of AD. CSF melatonin levels was lower in AD patients than in old control subjects.
Walsh, United States 2014 ()
Prospective Cohort
N=1287
1,287 community-dwelling older women (82.8 ± 3.1 y) participating in an ongoing prospective study who were free of dementia at the baseline visit. Follow-up
time: 5 years
Actigraphy
Modified Mini-Mental Status Examination (3MS), California Verbal Learning Task (CVLT), digit span, Trail Making Test B (Trails B), categorical fluency, and letter fluency.
Weaker Circadian Activity Rhythm patterns are associated with worse cognitive function, especially executive function, in older women without dementia.
Keage, Australia, & United Kingdom 2012 ()
Prospective cohort
N= 2012
The Medical Research Council
Cognitive Function and Ageing Study (CFAS), a longitudinal multi-center population-based study of ageing and dementia (www.cfas.ac.uk). Follow-up
time: 10-years
Self-reported sleep measures
MMSE
Excessive daytime sleepiness may be modifiable behaviors open to intervention strategies, or, clinical indicators of future decline in older individuals.
Anderson, United Kingdom, 2014 ()
Prospective Cohort
N= 421
The study was nested in the Newcastle
85+ Study, a population-based longitudinal study of health and ageing in the very old. Follow-up time: 2 years
Pittsburgh Sleep Questionnaire Inventory (PSQI) and the Epworth Sleepiness Score (ESS) as subjective measures of sleep and wake.
Cognition assessed by the Mini-Mental State Examination (MMSE)
Abnormal sleep–wake patterns are associated with cognitive impairment
98
Tranah, United States, 2011 ()
Prospective Cohort
N=1282
Women were participants of the Study of
Osteoporotic Fractures (SOF), a longitudinal epidemiologic study of 10,366 community-dwelling women age 65 years or older. Follow-up time: 4.9
years
Circadian activity rhythms were measured with the Sleep-Watch-O (Sleep-Watch-OVR, Ambulatory Monitoring, Inc., Ardsley, NY) actigraph.
Modified Mini-Mental State Examination (3MS); California Verbal Learning Test (CVLT) delayed recall; Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE); self-reported previous dementia diagnosis
Older, healthy women with decreased circadian activity rhythm amplitude and robustness, and delayed rhythms have increased odds of developing dementia and MCI
Lim, United States, 2013 ()
Prospective Cohort
N=737
Rush Memory and Aging Project (MAP)which is an ongoing community-based cohort study of aging which began in 1997. Follow-up period was up to 6
years (mean 3.3 years).
Actigraphy Battery of Neuropsychological Tests
Sleep fragmentation in older adults is associated with incident AD and the rate of cognitive decline.
Yesavage, United States, 2004 ()
Prospective Cohort
N=44
To be included in the current analysis, the patient must have been assessed during at least 2 stages of cognitive impairment, defined as a decline from one stage of AD to another.
Actigraphy
APOE genotypes were then determined. Diagnosis of probable AD by NINCDS-ADRDA criteria and a score of 15 or higher at entry on the Mini-Mental State Exam (MMSE). .
APOE status is associated with the progression of sleep/wake disturbances in AD
Craig, Canada, 2008 ()
Experimental Not
Applicable
Twenty-four male Long Evans hooded rats (300–400 g) obtained from Charles River (Saint-Constant, PQ) were used for all studies. They were divided into three groups: acutely phase shifted rats (acute; n = 8), chronically phase shifted rats (chronic; n = 8), control rats for water maze testing (control; n=7) and control rats for fear conditioning (control; n = 14).
In order to create a state of circadian disruption a two-stage phase shifting session which lasted 16 days was used.
Rats underwent Behavioral Testing on a 6 day version of the water maze task, followed by 3 days of testing on a fear conditioning task. They then underwent a Rapid Acquisition Task training in three stages.
Authors propose that chronic circadian disruption may play a role in the development of age-related cognitive deficits and dementia in the elderly
99
Jyoti, Scotland, UK, 2010 ()
Experimental Not
Applicable
APP/PSEN1 and PSEN1 Transgenic mice. Transgenic mice were singly housed and assigned to the following groups, depending upon age (5 or 20 months) and genotype: 1) APP/PSEN1 [5 months: n = 8; 20 months: n = 6]; 2) PSEN1 [5 months: n = 8; 20 months: n = 11]; 3) WT [5 months: n = 9; 20 months: n = 8].
Circadian activity was recorded in PhenoTyper home cages (Noldus, The Netherlands) through video observation techniques and XY coordinates recorded over 7 consecutive days.
Authors characterized sleep, electroencephalogram (EEG) disturbances i.e., endophenotypes of Alzheimer’s disease (AD) patients alongside cognitive dysfunction in transgenic mice carrying transgenes for amyloid-protein precursor (APPswe) and presenilin 1 (PSEN1A246E) at 5 (pre-plaque) and 20 months, relative to PSEN1 and wild-type (WT) mice, using a novel wireless microchip device.
APP/PSEN1 mice exhibit abnormalities in activity and sleep architecture preceding amyloid plaque deposition as well as age-related changes in cortical EEG power.
Quinn, United States, 2005 ()
Experimental Not
Applicable Transgenic mice that were bred from a breeding pair of Tg2576 mice
Plasma melatonin
Soluble and formic acid extractable beta amyloid determination was done and Aβ burden in hippocampus and cerebral cortex was quantified with NIH image.
Melatonin failed to produce anti-amyloid or antioxidant effects when initiated after the age of amyloid plaque deposition. These findings diminish the possibility that melatonin will be useful for the treatment of established Alzheimer’s disease.
DENG. China, 2005 ()
Experimental Not
Applicable
N2a cells were grown and differentiated in 96-well culture plates at density of 1.5×105 cells in 100 μL. Cells were exposed to various concentrations of wortmannin for 2 h at 37 ºC in the presence or absence of a 12-h-preincubation with melatonin 25, 50, and 100 μmol/L or Vitamin E (VE). Dimethylsulfoxide (Me2SO, 0.01%), in which wortmannin, melatonin and VE were dissolved, served as a vehicle control.
Western blot, P-labeling and the detection of malondialdehyde level and superoxide dismutase activity
MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-tetrazolium bromide), crystal violet assay, phase-contrast, dead end colorimetric apoptosis detection system (TUNEL) and electron microscopy were used to detect cell viability, morphology and apoptosis
Wortmannin is an effective tool for reproducing Alzheimer-like tau hyperphosphorylation cell model and melatonin/vitamin E can effectively protect the cells from wortmannin-induced impairments
100
Papolla, United States, 1998 ()
Experimental Not
Applicable
Aliquots of Aβ1–40 and Aβ1–42 at a concentration of 250 μm in 5 mm Tris-HC1, pH 7.4, were incubated at room temperature
Melatonin or the melatonin analog 5-hydroxy-N-acetyl-tryptamine (NAT) (Sigma) orN-t-butyl-α-phenylnitrone (PBN) (Sigma),
Peptides Aβ1–40 and Aβ1–42 were synthesized in the W. M. Keck Foundation (Yale University, CT), and their purity was evaluated by amino acid sequence and laser desorption mass spectrometry.
Melatonin can provide a combination of antioxidant and anti-amyloidogenic features that can be explored either as a preventive or therapeutic treatment for AD or as a model for development of anti-amyloidogenic indole analogs.
Olcese, United States, 2009 ()
Experimental Not
Applicable APP + PS1 double transgenic (Tg) mice
Authors administered melatonin (100 mg/L drinking water) to APP + PS1 double transgenic (Tg) mice from 2–2.5 months of age to their killing at age 7.5 months.
Mice were initially genotyped at the time of weaning and then had confirmatory genotyping within several months thereafter, with further confirmation via plasma Ab40 expression.
Results are consistent with a melatonin-facilitated removal of Aβ from the brain. Inflammatory cytokines such as tumor necrosis factor (TNF)-α were decreased in hippocampus (but not plasma) of Tg+ melatonin mice.
Lahiri, United States, 2004 ()
Experimental Not
Applicable
Male B6C3F1 mice, a hybrid between C57BL/6 and C3H from Harlan Labs (Indianapolis, IN, USA), aged 6 months (young group), 12 months (middle-aged group) and 27 months (old group), were housed two or three per cage and were maintained on a 12 hr light/dark cycle in a temperature controlled (22 ± 1°C) room.
An ELISA procedure was used for the quantitative measurement of melatonin in serum
A sensitive enzyme-linked immunosorbent assay (ELISA) test for measuring levels of Aβ-40 was performed in the murine brain homogenates. ELISA test was also performed using the IBL assay reagents for the quantitative determination of Aβ-42 in the murine brain homogenates.
Melatonin supplementation may retard neurodegenerative changes associated with brain aging. Depletion of melatonin in the brain of aging mice may in part account for this adverse change.
Asayama, Japan, 2003 ()
Randomized Controlled Trial
N=20
The subjects were 9 persons given a placebo (PLA), and 11 given melatonin (3 mg)(MLT). The drugs were given at 20: 30 each day for 4 weeks.
Actigraph data from 18 patients (PLA8, MLT10)
Diagnosis of AD via CT or MRI and a clinical diagnosis based on DSM IV criteria and the NINCDS-ADRDA criteria, Dementia severity was assessed by the Clinical Dementia Rating Scale (CDR), Cognitive funcion was assessed by the Alzheimer's Disease Assessment Scale (ADAS) the MMSE
Melatonin administration improved sleep time and night activity significantly, but no significant effect with regard to improving daytime naps and activity. Melatonin administration improved cognitive and non-cognitive functions significantly.
101
Table 1.9. Sleep Disordered Breathing/Sleep Apnea and Cognitive/Alzheimer's Disease Measures: Descriptive Study
Characteristics, Statistical Methods and Main Findings
First Author, Country, Year
Published (Reference) Study Design
Sample
size N Study Population Hypoxia Assessment
Cognitive/AD Measures
Assessment Conclusion
Gottlieb, United States, 2004 ()
Cross sectional N=1,775 .
Participants were aged 40 to 100 years from the Sleep
Heart Health Study (SHHS) a longitudinal cohort study of the relation of sleep-disordered breathing to CVD.
Obstructive Sleep Apnea/Hypopnea (OSAH) was defined as an Apnea Hypopnea Index (AHI) ≥ 15.
APOE was genotyped
The APOE £4 allele is associated with increased risk of OSAH, particularly in individuals under age 65.
Kadotani, United States, 2001 ()
Cross-sectional N=791 Probability based sampling of 791 middle-aged adults, mean age of 49 and age range 32-68
Nocturnal Polysomnography with Sleep Disordered Breathing defined as AHI ≥ 15
APOE was genotyped
A significant proportion of Sleep Disordered Breathing is associated with APOE £4 allele in the general population
Osorio, United States, 2014 () Cross-sectional N=95
Ninety-five cognitively normal elderly participants were recruited at NYU Center for Brain Health ongoing between 2009 and 2013.
Among the 95 participants, 25 were considered free of SDB (AHI4% < 5) and were included as normal controls, 51 had mild SDB (AHI4% 5-14.99), and 19 had moderate to severe SDB (AHI4% > 15).
CSF measures of phosphorylated-tau (p-tau), total-tau (t-tau), and amyloid beta 42 (Ab-42), as well as ApoE allele status
There is an association between SDB and CSF Alzheimer’s disease-biomarkers in cognitively normal elderly individuals..
Djonlagic, United States, 2014 ()
Cross-sectional N=44
Participants (19–68 years) who had been referred by a physician for a baseline polysomnography (PSG) evaluation.
Based on their PSG, patients were assigned either to the OSA group (AHI>5/h), or control (Non-OSA) group (AHI<5/h).
Psychomotor Vigilance Task (PVT) and the Motor Sequence Learning Task (MST)
The presence of untreated obstructive sleep apnea is associated with an aging-related cognitive deficit, otherwise not present in individuals without OSA.
102
O’Hara, United States, 2005 ()
Case-control N=36
Twelve men and 24 women participated. All were over age 60, with a mean age of 70.6 (SD = 8.1), a mean of 16.1 (SD = 2.4) years of education, and a mean Mini-Mental State Examination (MMSE) of 28.7 (SD = 1.2; range 27 to 30). Eighteen had an APOE €4 allele, and 18 were non-carriers
The apnea-hypopnea index (AHI), i.e., the average number of apneas or hypopneas per hour of estimated sleep measured OSAH severity in each subject.
APOE genotyping. MMSE; the Rey Auditory Verbal Learning Test of immediate verbal recall (RAVLT1); short-term free recall (RAVLT6) and 30-minute delayed free-recall (RAVLT Delayed); the Stroop Color and Word (SCW) measure of attention and cognitive flexibility; and the Symbol-Digit Modalities Test (SDMT) of information processing speed.
Nocturnal sleep apnea/hypopnea is associated with lower memory performance in APOE €4 carriers
Cohen-Zion, United States, 2001 ()
Prospective cohort
N=46
Community-dwelling people age 65 and older with high risk for SDB were originally studied from 1981 through 1985 and then followed every 2 years. Follow-up period:
2years
Subjects’ sleep was recorded using the Modified Respitrace/Medilog system, which measured thoracic and abdominal respiration, tibialis electromyogram (EMG), and wrist activity (to discriminate sleep/wake states).
MMSE
The results of this study could theoretically suggest that any relationship between SDB and cognitive function may be mediated by the effect of SDB on daytime sleepiness.
Tworoger, United States, 2006 ()
Prospective cohort
N = 1844
Participants were women aged 70 to 81 years at initial cognitive interview in 2000 and members of the Nurses’
Health Study cohort. Follow-
up period: 2years
Authors collected information about average sleep duration and snoring frequency on questionnaires in 1986 and 2000;
Women completed six tests of cognitive function encompassing general cognition, verbal memory, category fluency, and attention. Tests were repeated 2 years later.
Associations between sleep patterns and initial cognitive function may be clinically relevant given that diminished cognition is a risk factor for dementia. However, the lack of an association with prospective cognitive decline was seen
103
Chang, Taiwan, 2013 () Retrospective cohort
N=8484
Longitudinal Health
Insurance Database 2005 (LHID2005) in Taiwan, 1414 patients with Sleep Apnea (SA) aged 40 years Follow-
up period: 5years
Diagnoses of SA [ICD-9-Codes 327.23, 780.51, 780.53, 780.57] between January 2003 and December 2005. Clinically assessed including only cases in which patients obtained ≥2 SA diagnoses in outpatient visits or ≥1 inpatient service.
ICD-9 code
SA may be a gender-dependent, age-dependent, and time-dependent risk factor for dementia
Yaffe, United States, 2011 () Prospective cohort
N=298 Study of Osteoporotic
Fractures. Follow-up
period: 2.3 years
Sleep disordered breathing was defined as an apnea-hypopnea index of 15 or more events per hour of sleep.
Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition) criteria. MCI was diagnosed using the modified criteria by Peterson et al
Among older women, those with sleep disordered breathing, compared with those without SDB, were associated with an increased risk of developing cognitive impairment
Shiota, Japan, 2013 () Experimental Not Applicable
In vivo, 15 male triple transgenic AD mice were exposed to either Chronic Intermittent Hypoxia (CIH) or normoxia (5% O2 and 21% O2 every 10 min, 8 h/day for 4 weeks). In vitro, human neuroblastoma SH-SY5Y cells stably expressing wild-type amyloid-β protein precursor were exposed to either IH (8 cycles of 1% O2 for 10 min followed by 21% O2 for 20 min) or normoxia
CIH was applied (n = 9) by exposing to alternating 5% O2 and 21% O2 every 10 min for 8 h per day during daytime for 8 weeks in a chamber (370×260×250 mm, 26 L, Sibata Scientific Technology Ltd, Tokyo, Japan). Control mice (n = 6) were kept under normoxia and touched by human hands once a day to balance out their stress through direct human contact.
Amyloid-β (Aβ) profile, cognitive brain function, and brain pathology were evaluated.
Results suggest that Obstructive Sleep Apnea (OSA) would aggravate AD. Early detection and intervention of OSA in AD may help to alleviate the progression of the disease.
Kaushal, United States, 2012 ()
Experimental Not Applicable
Adult male human ApoE4-targeted replacement mice [B6.129P2-Apoetm3(APOE*4)MaeN8] (hApoE4) and C57BL/6NTac (wild-type, WT) mice,
HYPOXIA MEASURE O2 concentration was continuously measured by an O2 analyzer and was changed from 1:00 PM until 7:00 PM by a computerized system controlling the gas valve outlets.
SLEEP INTEGRITY
MEASURE Polysomnography
The increased susceptibility and limited recovery ability of hApoE4 mice to sleep apnea suggests that early recognition and treatment of the latter in AD patients may restrict the progression and clinical manifestations of this frequent neurodegenerative disorder
104
Sun, Canada, China, United States, 2006 ()
Experimental Not Applicable
APP23 Transgenic mice
APP23 mice (8 months of age, n = 20, 50% female) were assigned randomly to hypoxia and control groups. The hypoxia group was treated in a hypoxia chamber at 8% O2 for 16 h/day for 1 month.
The water maze test. Neuritic Plaque Staining was done.
The results demonstrate that hypoxia can facilitate AD pathogenesis, and they provide a molecular mechanism to link vascular factors to AD
Li, China, 2009 () Experimental Not Applicable
Transgenic mice (The Jackson Lab, No. 003378, APPSwe + PS1A246E)
The mice were treated with hypoxia once a day and repeated for 60 days. Mice were assigned randomly to hypoxia and control groups (9 months old female, 10 mice in each group).
Neuritic plaques staining was performed.
Taken together, the results suggest an important role of hypoxia in modulating the APP processing by facilitating both β- and ɣ-cleavage which may result in a significant increase of Aβ generation.
Zhang, China, 2007 () Experimental Not Applicable
Mouse neuroblastoma N2a cells stably expressing human APP695 (N2a-APP) were maintained in cell culture medium containing 50% DME high glucose medium
Cell culture, Transfection and Hypoxic treatment was done.
Immunoblotting and Antibodies, Aβ ELISA Assay, In Vitro β-Secretase Activity Assay, Pulse-chase experiments, Quantitative Real-time Polymerase Chain Reaction (RT-PCR), HIF-1α RNA Interference, Tissue Isolation from HIF-1α Conditional Knock-out Mice were also carried out on the mice models
The results demonstrate an important role for hypoxia/HIF-1α in modulating the amyloidogenic processing of APP and provide a molecular mechanism for increased incidence of AD following cerebral ischemic and stroke injuries
Kushida, United States, 2012 ()
Randomized Controlled Trial
N=1,098
The Apnea Positive Pressure
Long-term Efficacy Study (APPLES) was a 6-month, randomized, double-blind, 2-arm, sham-controlled, multicenter trial conducted at 5 U.S. University, hospital, or private practices.
Diagnosis of Obstructive Sleep Apnea (OSA) with an apnea-hypopnea index (AHI) ≥ 10 and age ≥ 18 years
Three neurocognitive variables, each representing a neurocognitive domain: Pathfinder Number Test-Total Time (attention and psychomotor function [A/P]), Buschke Selective Reminding Test-Sum Recall (learning and memory [L/M]), and Sustained Working Memory Test-Overall Mid-Day Score (executive and frontal-lobe function [E/F])
CPAP treatment improved both subjectively and objectively measured sleepiness, especially in individuals with severe OSA (AHI >30).
105
Ancoli-Israel, United States, 2008 ()
Randomized Controlled Trial
N=52
Participants were men and women with mild-moderate AD and OSA recruited from the University of California San Diego (UCSD) Alzheimer's Disease Research Center.
Diagnosis of Obstructive Sleep Apnea (OSA) with an apnea-hypopnea index (AHI) ≥ 10.CPAP unit pressure was set at 8 cm H2O pressure while the system mask's pressure varied from 0.5 cm H2O at end-expiration to 0.0 cm H2O during inspiration
NINCS-ARDRA and a CT or MRI of the brain consistent with AD done within 24 months Mini Mental Status Examination (MMSE) score greater than 17.
OSA may aggravate cognitive dysfunction in dementia and thus may be a reversible cause of cognitive loss in AD patients.
Table 1.10. Insomnia and Cognitive/Alzheimer's Disease Measures: Descriptive Study Characteristics, and Main Findings
First Author,
Country, Year
Published
(Reference)
Study Design Sample size
N Study Population Insomnia Assessment Cognitive/AD Measures Assessment Conclusion
Merlino, Italy, 2010 ()
Cross-sectional N=750
Subjects aged 65 years or older, living independently or in an institution, who resided in the seventh district of Udine, Italy were recruited.
Insomnia and other sleep abnormalities were investigated with a battery of standardized questions and questionnaires
DSM-IV-TR criteria . AD and VaD were diagnosed respectively according to NINCS-ARDRA and NINDS-AIREN criteria . LBD and PDD, frontotemporal dementia (FTD) and MCI were diagnosed according to specific consensus guidelines .
Insomnia was not associated with the presence of cognitive decline. As opposed to insomnia, excessive daytime sleepiness was significantly related to dementia.
Cricco, Iceland, 2001 ()
Prospective cohort N=6444
The four sites of the Established Populations
for Epidemiologic Studies of the Elderly (EPESE). Follow-up period: 3 years
Insomnia was defined as self-report of trouble falling asleep or waking up too early most of the time.
Cognitive decline was defined as two or more errors on the Short Portable Mental Status Questionnaire (SPMSQ) at FU3.
Chronic insomnia independently predicts incident cognitive decline in older men.
Jelicic, The Netherlands, 2002 ()
Prospective Cohort
N=838
Maastricht Ageing Study
(MAAS), a prospective cohort study of cognitive ageing based in the Netherlands. Follow-up
period: 3 years
Self-reported sleep measures
Cognitive performance at follow-up, measured with the Mini Mental Status Examination (MMSE)
Subjective sleep complaints predict cognitive decline in middle aged and older adults
106
Foley, United States, 2001 ()
Prospective Cohort
N=2346
Japanese-American men age 71 to 93 of The
Honolulu-Asia Aging
Study (HAAS)
Sleep disturbances were assessed by questionnaire.
Dementia was diagnosed according to less-restrictive diagnostic criteria developed by Cummings et al. for very mild dementia and more-restrictive diagnostic criteria for mild or more severe dementia from the Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Re-vised (DSM-IIIR).
Daytime sleepiness in older adults may be an early indicator of decline in cognitive functioning and onset of dementia.
Osorio, United States, 2011 ()
Prospective Cohort
N=346
Brain aging studies at New York University’s Alzheimer’s Disease Research Center. Follow-
up period: over 7.7 years.
Insomnia was collected only during the normal cognition stage and screened using items 4, 5, and 6 from the Hamilton Depression Rating Scale (HAM-D)
Mini-Mental State examination There is a greater risk of Alzheimer's disease in adults with Insomnia
Table 1.11 Restless Less Syndrome (RLS) and Cognitive/Alzheimer's Disease Measures: Descriptive Study Characteristics, and Main
Findings
First Author,
Country, Year
Published
(Reference)
Study Design Sample size
N Study Population RLS Assessment Cognitive/AD Measures Assessment Conclusion
Pearson, United States, 2006 ()
Case-control N=31
Sixteen patients off RLS treatment for at least 2 weeks and 15 age- and gender-matched control subjects
RLS patients were required to have daily symptoms and to have periodic leg movements in sleep (PLMS) of greater than 20/h on a screening nocturnal Polysomnogram (PSG).
Cognitive Tests consisted of the following: the Stroop color-word test, Trail Making Test (A and B), the Porteus Maze Test (Vineland Revision), the Colored Progressive Matrices Test, and two verbal fluency tests (words beginning with a fixed first letter and words in categories).
RLS patients have statistically significant specific cognitive deficits and show cognitive deficits similar to those seen with 36 h total sleep deprivation.
107
Figure 1.1 Flow Diagram of Systematic Review of Sleep and Alzheimer’s disease
Studies identified through database searching (n = 2341)
(PubMed=904, Embase=1026, Web of Science=305, Cochrane library=106)
Studies after duplicates removed (n = 1221)
Studies screened for titles (n = 1221)
Studies excluded (n = 180)
Full-text articles assessed for eligibility
(n = 201)
Full-text articles excluded, with reasons (n = 130)
- Review article (n = 7) - No risk estimate (n =
17) - Sleep not assessed (n =
18) - Relationship not
studied (n = 53) - Not peer reviewed (n =
10) - Alzheimer not
examined (n = 25)
Studies included in systematic review
(n = 72)
Article identified from reference search
(n = 1)
Studies screened for abstracts
(n = 1041)
Studies excluded (n = 840)
Duplicates removed (n = 1120)
108
SECTION 2
SLEEP, COGNITIVE IMPAIRMENT AND ALZHEIMER’S DISEASE: A SYSTEMATIC REVIEW
AND META-ANALYSIS
Note to the Reader:
This entire chapter has been previously published in SLEEP journal, 2017, Jan 1; 40(1). doi:
10.1093/sleep/zsw032 and have been reproduced with permission from Oxford University Press.
List of Authors: Omonigho M Bubu MD, MPH1*, Michael Brannick PhD2, James Mortimer PhD1, Ogie
Umasabor-Bubu MD MPH1, Yuri V. Sebastião MPH1, Yi Wen MS3 , Skai Schwartz PhD1, Amy R.
Borenstein PhD1, Yougui Wu PhD1, David Morgan PhD4,5, William M. Anderson PhD6
1. University of South Florida, College of Public Health, Department of Epidemiology &
Biostatistics, 13201 Bruce B. Downs Avenue MDC56, Tampa, FL 33612
2. University of South Florida, College of Arts and Sciences, Psychology Department, 4202 East
Fowler Ave, PCD4118G, Tampa, FL 33620
3. University of South Florida, College of Engineering, Department of Chemical and Biomedical
Engineering, 4202 East Fowler Ave, PCD4118G, Tampa, FL 33620
4. University of South Florida, Morsani College of Medicine, Molecular Pharmacology and
Physiology, 12901 Bruce B. Downs Avenue MDC 8, Tampa, FL 33612
5. Byrd Alzheimer Institute, 4001 E. Fletcher Ave, Tampa, FL 33613
6. University of South Florida, Morsani College of Medicine, Sleep Medicine, Internal
Medicine,12901 Bruce B. Downs Avenue MDC 8, Tampa, FL 33612
109
Author contributions:
Bubu OM designed and performed the study and wrote all sections of the manuscript under the
supervision of committee members, Mortimer J, Morgan D, Anderson W, Borenstein AR, Wu Y and
Schwartz S. Brannick M supervised the statistical analysis. Umasabor-Bubu OQ, Sebastiao Y, and Wen Y
assisted in writing, editing and proofreading the article.
ABSTRACT
Study Objectives: Mounting evidence implicates disturbed sleep or lack of sleep as one of the risk
factors for Alzheimer’s disease (AD) but the extent of the risk is uncertain. We conducted a broad
systematic review and meta-analysis to quantify the effect of sleep problems/disorders on cognitive
impairment and AD.
Methods: Original published literature assessing any association of sleep problems or disorders with
cognitive impairment or AD was identified by searching PubMed, Embase, Web of Science, and the
Cochrane library. Effect estimates of individual studies were pooled and relative risks (RR) and 95%
confidence intervals (CI) were calculated using random effects models. We also estimated the population
attributable risk (PAR).
Results: Twenty-seven observational studies (n = 69,216 participants) that provided 52 RR estimates
were included in the meta-analysis. Individuals with sleep problems had a 1.55 (95% CI: 1.25-1.93), 1.65
(95% CI: 1.45-1.86) and 3.78 (95% CI: 2.27-6.30) times higher risk for AD, cognitive impairment and
preclinical AD than individuals without sleep problems respectively. The overall meta-analysis revealed
that individuals with sleep problems had a 1.68 (95% CI: 1.51-1.87) times higher risk for the combined
outcome of cognitive impairment and/or AD. Approximately 15% of AD in the population may be
attributed to sleep problems.
Conclusion: This meta-analysis confirmed the association between sleep and cognitive impairment or
AD and, for the first time, consolidated the evidence to provide an “average” magnitude of effect. As
110
sleep problems are of a growing concern in the population, these findings are of interest for potential
prevention of AD.
Statement of Significance: This study systematically and efficiently integrated existing information from
various studies, and for the first time, consolidated the evidence of the association between sleep and
Alzheimer’s disease (AD) and/or cognitive impairment by providing an “average” magnitude of the
effect. Subgroup meta-analysis of various aspects of sleep problems examined in this analysis each
significantly increased the risk for AD. The public health relevance of these findings is buttressed by the
population attributable risk percent, suggesting that approximately 15% of AD may be prevented should
interventions be implemented to reduce sleep problems and disorders. These novel findings are of
pertinent interest to further understand the etiology of AD and are likely to be of interest to researchers,
clinicians’ and public health policy makers.
111
INTRODUCTION
Alzheimer’s disease (AD) is a neurodegenerative disorder afflicting approximately five million
older Americans, with a projected annual incidence of one million cases in the United States by 2050.1, 2
AD and other dementias are associated with significant morbidity and mortality, with worldwide costs
estimated at US$604 billion in 2010.3 Preventive and/or therapeutic measures targeted at delaying or
curing AD are therefore imperative to mitigate this societal burden.
Mounting evidence implicates disturbed sleep or lack of sleep as a risk factor for AD. Demented
older adults exhibit significant sleep disturbance and unhealthy sleep patterns, including shorter sleep
duration and fragmented sleep,4-7 altered circadian rest/activity patterns,5 and elevated rates of sleep
disordered breathing (SDB).8, 9 It is estimated that about 45% of AD patients have sleep disturbances.10, 11
However, the ultimate etiologic role of sleep in AD pathogenesis remains unknown. Sources of sleep
disturbances in AD are thought to include degeneration of neural pathways that regulate sleep-wake
patterns and sleep architecture.10, 12 Sleep disorders also may exacerbate cognitive symptoms through
impairment of sleep-dependent memory consolidation processes.13, 14 It is possible that circadian rhythms
and sleep disorders have a causal role in the pathophysiology of AD. If this is true, they could be critical
to the goal of preventing this disease, because effective interventions exist to improve sleep.15 Narrative
review articles examining this association16-19 reflect the opinion and consensus among specialists that
there is growing experimental and epidemiologic evidence for a close reciprocal association of cognitive
impairment and AD with sleep problems and disorders. However, to date, no review has quantified the
effect of having sleep problems and disorders on cognitive impairment and AD. In this context, we
conducted a broad systematic review and meta-analysis to quantify the effect between sleep problems and
disorders on cognitive impairment and AD.
METHODS
112
We conducted this review in accordance with the step-by-step guide for systematic reviews and
meta-analysis by Pai et al.,20 and adhered to the preferred reporting items for systematic reviews and
meta-analyses (PRISMA) statement by Moher et al.,21 and the Cochrane handbook.22
Identification of eligible studies
Electronic search
A broad search strategy was employed to identify all eligible published articles in scientific
(medical, psychological and epidemiological) journals up to November 2014. This strategy combined
terms characterizing clinical AD or AD pathology as the outcome variable, sleep, as the exposure
variable, and a third set of terms specifying specific study types including cross-sectional, case control,
cohort (retrospective and prospective), Randomized Clinical Trials (RCTs), literature reviews, systematic
reviews, and ecological studies. All terms within each set were combined with the Boolean operator OR,
and the three sets were combined using AND. Articles were identified by searching four bibliographic
databases on November 21, 2014:
• Medline (through PubMed; all years 1946 - present)
• Embase (through www.embase.com by Elsevier B.V. 2011; all years from 1947 - present)
• Web of ScienceTM (Thomson Reuters 2014; all years from 1900 – present)
• Cochrane library (i.e., the Cochrane Central Register of Controlled Trials; all years from 1900 -
present).
Articles written in other languages other than English were also considered and screened with the help of
Google translator. The PubMed search (see supplemental Table 2A in appendix) was adapted to
searching the other databases.
Searching other sources
To locate additional relevant studies, the reference lists of articles identified through database
searches and the bibliographies of narrative review articles were examined. We also hand-searched the
previous year’s issues (November 2013 to November 2014) of high ranking journals that published more
113
than one relevant identified article during the initial search, for example Sleep Medicine Reviews,
Neurology, JAMA (journal of American medical association), Sleep Medicine, Sleep, Journal of Sleep
Research and Journal of Clinical Sleep Medicine.
Selection criteria
Type of studies
Original articles from observational studies (cross-sectional, case-control and
retrospective/prospective cohort) and randomized controlled trials (RCT) were included in this review.
Eligible studies for the meta-analysis were required to: i) have reported a primary or secondary data
analysis using any of the above mentioned study design of any association of sleep problems or disorders
with cognitive impairment or AD; ii) have an explicit measure of cognitive impairment or AD (e.g.
performance on individual cognitive tests, international classification of disease-9 (ICD -9), diagnostic
statistical manual-IV text revision (DSM-IV TR)) or objective measures of Alzheimer pathology (e.g.,
positron emission tomography (PET) scan); iii) have an explicit or obvious measure of sleep problems or
disorders (e.g., Pittsburgh sleep quality index (PQSI)), Epworth sleepiness scale (ESS)) or objective
measures (e.g., wrist actigraphy or polysomnography (PSG)); iv) study a comparison group without sleep
problems or disorders in cohort studies; and v) provide sufficient data to quantify the measure of
association of sleep problems or disorders with cognitive impairment or AD. The following studies were
excluded: i) case series, case reports, narrative reviews, abstracts or editorials, and ii) any RCT that did
not examine the cognitive effects of treating sleep and its related problems or disorders in AD.
Sleep problems and/or disorders
Sleep problems and/or disorders were considered as the risk factor of interest in this systematic
review. All sleep disorders described in the international classification of sleep disorders (ICSD-2)23 were
considered. Previous studies24, 25 employed various concepts to define sleep problems in order to fit their
114
study design. Accordingly, for sub-group analysis, we grouped eligible studies in this review measuring
sleep using parameters such as quality, efficiency and/or disturbance as sleep quality.
Alzheimer’s disease/cognitive impairment
The outcomes of interest in this review were cognitive impairment or AD. AD is the most
common form of dementia accounting for up to 50-80% of dementia cases.2 Studies that solely looked at
specific forms of dementia other than AD, e.g., vascular, Lewy body, or frontotemporal dementia, were
excluded.
Participants and setting
Participants of the included studies were middle aged and elderly adults of both sexes with age
ranging from 40 – 91 years. Studies were considered from multiple continents and no geographic
restrictions were imposed.
Data Extraction
An independent assessment of all titles and abstracts of eligible studies identified by the search
strategy was performed by two authors (OB, OU) using EndNote X6. Potential articles for the study were
retrieved and examined for possible inclusion by the same two authors. Discordant decisions regarding
inclusion were resolved by two other authors (YS and YW). Data extraction was performed by two
authors (OB, OU). Extracted fields included study design, study population, exposure and outcome
assessment, statistical methods and tests used, covariates, and the main results of the study. Differences in
the information extracted were resolved by two other authors (YS and YW). Reviewers were not blinded
to the funding, authors, or institutions.
Assessment of study quality
Study quality was assessed using an adaptation from previous systematic reviews26, 27 and the
modified version of the Newcastle-Ottawa scale (NOS) for quality assessment of the observational
studies,28, 29 with addition of new items relevant to this review. Parameters used for the quality assessment
115
included well-specified hypothesis, study design type, appropriately described sample, sample size,
definition of sleep and its related abnormalities, definition of cognitive impairment or AD, statistical
analytic methods, and inclusion of or approach used to adjust for potential confounders. For each
assessment category, a star rating system was used with specific maximum number of stars awarded (see
supplemental table 2). Studies were divided into three categories: low quality (< 50% of the maximum
number of stars), medium quality (≈ 55-70% of the maximum number of stars), and high quality (70% or
more of the maximum number of stars).
Statistical analysis
This systematic review consisted of both binary and continuous outcomes. Effect sizes from the
reviewed studies included odds ratios (OR), hazard ratios (HR), Pearson’s correlation coefficient (r), beta
estimates (β) and standardized mean differences (d). To proceed with the meta-analysis, we converted the
different indices to a common index: odds ratios (meta-analyses were computed using log (OR) but were
converted back to odds for presentation) (see Table 2B in supplemental material). Conversions
included converting from d to the log(OR) (Log(OR) = dπ
√� , where π is the mathematical constant
(approximately 3.14159)), converting from r to d (d = d��
√���� ), and converting from β to OR (exp(β) (see
Table 2C in supplemental material).30 The question of whether or not it is appropriate to combine effect
sizes from studies using varying metrics is under debate. We considered the fact that studies included in
the review and subsequent analyses had to meet certain inclusion and exclusion criteria and were judged
to be comparable in relevant ways. Since our aim was to compute a summary effect size of the impact of
sleep problems on AD, we accounted for multiple outcome measures on the same subjects by computing
the mean of the effect sizes for each study and used the average effect size as the unit of analysis (see
Table 2D in supplemental material). For example, we pooled the effect estimates in a single study that
had separate outcome measures using the mini mental state examination (MMSE) and the trail making B
test (Trails B) on the same subjects.31 This influenced the calculation of the variance of the sum and mean
effects of the correlated variables (see Table 2D in supplemental material). For our analysis, we worked
116
with a plausible range of correlations ranging from 0.5-1.0 and performed sensitivity analyses to find the
most appropriate standard error range. Our method could only have overestimated the variance and
underestimated the precision of our summary effect. Individual studies that examined the effect of sleep
on AD on different populations were treated as independent.32-38 Re-analyses of reported effect estimates
was conducted where appropriate in order to maintain consistency of definitions. For example, effect
estimates for sleep duration of ≤ 5 h, 6 h, and for <= 6.5 h were pooled to a single estimate for sleep
duration of <= 6.5 h.32, 38, 39 Furthermore, we converted the risk estimate for better sleep consolidation to a
risk estimate for poor sleep consolidation.40 Participants in the reviewed studies were between the ages of
40 -91. Because the prevalence of AD in this population and in the general population is just under 5%,2
no relevant artificial bias was introduced by pooling effects from cohort studies with effects from case-
control and cross-sectional studies. Therefore, we will refer to the pooled effect estimates as relative
risks, understanding that these are estimated in some studies. Effect estimates of all individual studies
were pooled and relative risks (RR) and 95% confidence intervals (CI) were calculated using random
effects models. Heterogeneity of results across studies was assessed using chi-square test and quantified
using I-square statistics.41 Subgroup analyses were conducted to explore potential causes of heterogeneity.
We anticipated heterogeneity and as such restricted subgroup analyses to situations where four or more
studies were available per subgroup, because heterogeneity is better assessed with multiple studies.41
Multivariate meta-regression analysis was also performed to examine the effect of possible influential
factors (i.e., sex, mean age, sample size and study quality). We assessed for publication bias using a
funnel plot asymmetry test and trim-and-fill. Additionally, the population attributable risk percent (PAR
%), which estimates the percentage of the outcome (AD) that would be prevented in the population if the
exposure (sleep problems or disorders) was eliminated, was calculated. All statistical analysis were
conducted using metaphor in R42 and SAS® 9.4.43
RESULTS
Study Selection
117
Figure 1 shows the study retrieval and selection process for the effects of sleep on cognitive
impairment and/or AD. Of the 2,341 publications identified initially from the literature search, 27 studies
were included in the meta-analysis (Figure 1). A total of 1,120 duplicates were removed. The remaining
1,221 publications were screened for titles resulting in the exclusion of 180 publications. After this
exclusion, 1,041 publications were screened using published abstracts, resulting in the exclusion of 840
publications that were either case series/reports, literature reviews, abstracts or editorials. Two hundred
and one full text articles were assessed for eligibility and 130 publications (105 observational studies and
25 RCTs) were further excluded because they were either review articles, did not provide any measure of
association, were not peer reviewed or because sleep and AD were not assessed. One study was included
from a search of the reference lists of articles identified through database searches. The remaining 72
studies were selected for the systematic review. For the meta-analysis, 45 publications were further
excluded because they lacked sufficient information needed for a meta-analysis, were duplicate studies
presenting similar or exact information from same study population or had an experimental study design,
leaving 27 studies with 52 effect estimates.
Study characteristics
Design and Quality
Among the 27 studies identified for the meta-analysis, there were eight cross-sectional31, 37, 44-49
one case-control.50 two retrospective cohort33, 51 and sixteen prospective cohort,32, 34, 36, 38-40, 52-61 (Table 1).
Of these, 15 were considered to be of high quality, ten were of medium quality and two were of low
quality. Selection bias related to sampling, measurements of sleep and/or AD solely based on self-report,
and insufficient adjustment for core confounders were the main limitations. All of the included articles in
the meta-analysis were written in English and published in peer-reviewed journals.
Participants and setting
118
The 27 studies included in the meta-analysis were published between 2001 and 2014 and included
a total of 69,216 participants from four continents (Table 1). Pooled relative risks included 25 estimates
from studies conducted in North America, 14 estimates from Europe, one from Australia, and four from
Asia (Table 2). Twenty-one studies included both sexes, of which three had separate analysis for men and
women, five were limited to women, and one was limited to men (Table 1).
Sleep problems/disorders
Sleep problems were based on a physician diagnosis (e.g. ICD-9 codes and apnea hypopnea index
(AHI) scores for sleep apnea), or were assessed using self-reported questionnaires or interviews (Table
1). There was considerable variability in the methods used to define and validate sleep problems. Nine
studies utilized objective measures (e.g., wrist actigraphy or PSG), while seven studies utilized
standardized questionnaires such as the PQSI or ESS. Five studies utilized self-constructed questionnaires
and one study made use of fluorescence immune-assay (FIA) to determine melatonin/hypocretin
concentrations.50 Most studies reported risk estimates for individual symptoms. Multiple symptoms were
also assessed in some studies. Sleep parameters assessed included (but were not limited to) sleep
duration, sleep quality, daytime sleepiness, sleep latency, sleep efficiency, wake after sleep onset, sleep
fragmentation, circadian rhythm abnormalities, obstructive sleep apnea, and insomnia. The interval over
which sleep problems was assessed extended from about a week to a median of 12.5 years.32
Cognitive impairment and/or Alzheimer’s disease
There also was significant variability in the methods used to diagnose cognitive impairment or
AD. Most studies used self-reported diagnoses with information collected to determine whether they
likely met standardized criteria (e.g. (DSM-IV TR) (Table 2.1). Cognitive function was assessed by
various global cognitive function and neuropsychological tests. Individuals were regarded as cognitively
normal with Mini-Mental State Examination (MMSE) scores between 27 and 30, and a Clinical Dementia
Rating of 0. For "Mild Cognitive Impairment," MMSE scores ranged from 22-27 according to a DSM-III-
119
R definition (ICD-10: 23-28). Studies utilizing objective measures of Alzheimer pathology (e.g., PET
scan for amyloid load (Aβ)) to examine associations between sleep and preclinical AD were also
considered because we intended to also describe possible potential mechanisms showing that sleep is
possibly associated with changes in AD biomarkers predictive of individuals who develop AD in
subsequent sub-group analyses..
Meta-analysis
Overall meta-analysis based on risk estimates for the effect of sleep on cognitive impairment and
Alzheimer’s disease
Sleep problems including short and long sleep duration, poor sleep quality, circadian rhythm
abnormality, insomnia and obstructive sleep apnea, were associated with a significantly increased relative
risk (RR) for the combined outcome of cognitive impairment and AD of 1.68 (95% CI: 1.51-1.87) (Fig.
2.2). The relative risks ranged from 1.0734 to 4.2549 with two low outliers; 0.56,61 and 0.76,32 and five
high outliers 5.10,50 5.13, and 5.85,59 6.08,33 and 8.17.37 Removing these 7 outliers from the analysis had
no significant effect on the pooled RR (RR: 1.62, 95% CI: 1.47-1.79). There was significant heterogeneity
among the effect sizes (I-square 64.50%, p < 0.001). The random-effects variance component was .05
(95% CI: .023-.195). A prediction interval that considers both sampling error and random-effects
variance showed relative risk values ranging from 1.05 to 2.50. Removal of the 7 outliers reduced
heterogeneity but it remained statistically significant (I-square 54.36%, p < 0.001), indicating that these
outliers were not the sole source of heterogeneity.
Sub-group meta-analysis based on risk estimates for the effect of sleep on cognitive impairment, Pre-
clinical and ICD9/DSMIV Diagnoses of Alzheimer’s disease
Separate analyses of the studies in terms of the outcome measures were also conducted. Our
outcome measures included the use of cognitive tests by studies examining sleep and the risks of
cognitive impairment; the use of AD biomarkers or abnormal proteins by studies examining sleep and the
120
risk of preclinical AD; and the use of ICD9/DSMIV diagnoses of AD by studies examining the risk of
sleep and symptomatic AD respectively. Sleep problems including short and long sleep duration, poor
sleep quality, circadian rhythm abnormality, insomnia and obstructive sleep apnea, were associated with
significantly increased relative risks of cognitive impairment (RR: 1.64, 95% CI: 1.45-1.87) ), preclinical
AD (RR: 3.78, 95% CI: 2.27-6.30) and ICD9/DSMIV diagnoses of AD (RR: 1.55, 95% CI: 1.25-1.93)
respectively (Fig. 2.3) These results highlight potential mechanistic relationships suggesting that sleep is
not only associated with cognitive impairment, or symptomatic AD but is also associated with changes in
AD biomarkers predictive of persons that ultimately develop AD.
Funnel Plot Asymmetry Test and Trim and Fill Analysis Assessment of Publication Bias
Regarding possible publication bias, a funnel plot asymmetry test was significant (z = 5.5109, p <
.0001),62 indicating that some form of publication bias exist. Furthermore, a trim-and-fill analysis63
(which aims both to identify and correct for funnel plot asymmetry arising from publication bias by
essentially ‘trimming’ or removing the smaller studies causing funnel plot asymmetry, and then using the
trimmed funnel plot to estimate the true ‘center’ of the funnel, while replacing the omitted studies and
their missing ‘counterparts’ around the center (filling)) imputed 17 unpublished studies and confirmed
that in smaller studies, weaker effects were less often published (Fig. 2.4). The trim and fill method does
not require any assumptions regarding mechanisms leading to bias in publication, rather it provides an
estimate of the number of missing studies, and calculates an estimated effect ‘adjusted’ for the publication
bias (based on the filled studies).
Sub-group meta-analysis based on risk estimates for the effect of different sleep problems and
disorders on cognitive impairment and Alzheimer’s disease
Analyzing various sleep problems and disorders (Fig. 2.5), the highest relative risk for AD and/or
cognitive impairment was noted for obstructive sleep apnea (OSA) (RR: 2.37, 95% CI: 1.82-3.08). The
relative risk for cognitive impairment and/or AD associated with sleep quantity was 1.86 (95% CI: 1.34-
121
2.57), while sleep quality was 1.62 (95% CI: 1.34-1.97). Insomnia (RR: 1.38 95% CI: 1.13-1.67) and
circadian rhythm abnormalities (RR: 1.38, 95% CI: 1.18-1.61) were associated with the lowest relative
risks. After excluding the outliers, the relative risk for cognitive impairment or AD associated with OSA,
sleep quantity and sleep quality remained statistically significant (RR: 2.30, 95% CI: 1.59-2.41; RR:
1.72, 95% CI: 1.24-2.61 and RR: 1.51, 95% CI: 1.244-1.81 respectively).
Subgroup analyses (Table 2.2) investigating various sleep parameters revealed that poor sleep
efficiency was associated with the highest risk for cognitive impairment or AD, followed by sleep latency
and wake after sleep onset (WASO). Nighttime problems were associated with the strongest measure of
effect for cognitive impairment or AD, with smaller effects seen for night- and daytime problems. When
examined separately, men and women with sleep problems including short and long sleep duration, poor
sleep quality, circadian rhythm abnormality, insomnia and obstructive sleep apnea, each had similar
relative risks. The risk for cognitive impairment or AD due to sleep duration was higher in long sleepers
(> 8 h/day) when compared to normal sleepers (7 – 8 h/day), than in short sleepers (≤ 7 h/day). Studies
with ≥ 5 years of follow-up reported higher relative risks due to sleep problems including short and long
sleep duration, poor sleep quality, circadian rhythm abnormality, insomnia and obstructive sleep apnea,
than those with < 5 years of follow-up. Studies from North America and Europe reported similar relative
risks due to sleep problems. Prospective studies showed slightly lower relative risks than non-prospective
studies, perhaps indicating a problem with temporality. Finally, the estimated risk for cognitive
impairment or AD due to sleep problems was lower in the high-quality studies than in the low-quality
studies.
Sub-group meta-analysis based on risk estimates for the effect of self-report and actigraphic data of
poor sleep on cognitive impairment and/or Alzheimer’s disease
Furthermore, we analyzed studies that generated data on total sleep times (measured
actigraphically) and examined these effects separately in relation to cognition from more generally
defined poor sleep or self-reported sleep durations (Fig. 2.5). The relative risk for cognitive impairment
122
or AD associated with self-reported poor sleep was 1.49 (95% CI: 1.26-1.76), while actigraphic report
was 1.80 (95% CI: 1.56-2.09). These findings somewhat corroborate effects of self-reported sleep
durations in relation to cognition examined in terms of actigraphic data.
Exploring Potential Causes of Heterogeneity
The overall meta-analysis included a wide array of studies including different measures with
varying neurobiological substrates (breathing control in sleep, circadian rhythmicity, sleep durations,
perceived sleep depth) as well as varying outcome measures of AD, including ICD9/DSM diagnoses,
cognitive tests and abnormal proteins; and this could possibly be a potential source of heterogeneity.
Therefore, we explored this and other potential causes for heterogeneity and conducted several subgroup
meta-analyses (Figures 2.4-2.6 and Table 2.2) (also see Figures 2A-2C in supplemental material). The
differences seen across the subgroups failed to account for significant amounts of heterogeneity except for
the varying outcome measures of AD and sample size. However, tendencies could be observed. As
sample size increased, the effect size decreased. This agreed with the funnel plot. The varying outcome
measures of AD accounted for 12.78% heterogeneity. Several of the other subgroup analyses were close
to being statistically significant; however, there were only 52 effect sizes in the analysis, so power may
have been an issue for some sub-analyses.
Sensitivity Analyses
Sensitivity analysis incorporating only studies that examined the relationship between sleep and
cognitive impairment and/or symptomatic AD for the overall meta-analysis yielded a RR of 1.62 (95%
CI: 1.45-1.80), suggesting that exclusion of studies that examined sleep associations with preclinical AD
could reduce the overall effect size, however, the change in effect size was not statistically significant.
Similarly, removing effect estimates of outliers from the analysis had no significant effect on the pooled
RR (RR: 1.62, 95% CI: 1.47-1.79). Furthermore, exclusion of studies examining sleep associations with
preclinical AD for the sub-group meta-analysis based on risk estimates for the effect of different sleep
123
problems and disorders on cognitive impairment and Alzheimer’s disease produced similar relative risks
for the relationships between OSA, sleep quantity, sleep quality, insomnia and cognitive impairment
and/or AD.
Sensitivity analyses incorporating the 17 studies imputed by trim-and-fill showed an estimated
relative risk of 1.42 (95% CI: 1.28-1.58), suggesting that bias could reduce the size of the effect, but is
unlikely to explain it entirely. It should also be noted that the small sample, large effect size studies were
mostly based on results of studies in which correlations were converted to odds ratios, so there may be a
methodological explanation of funnel asymmetry other than bias. Altogether, the results support an
association between sleep problems and cognitive impairment or AD. Despite possible publication bias
and evidence of heterogeneity in the effect sizes, the overall association between sleep problems and
cognitive impairment or AD remained positive in all sensitivity analyses.
Furthermore, we clearly understand that testing for heterogeneity may not explain the variability
in the data because its basis may not simply be empirical but more fundamentally conceptual, therefore
we conducted several sensitivity analyses on both our exposure and outcome. Sensitivity analyses
examining the effect of omitting one study revealed no effect as the estimated relative risk remained
stable (See Fig. 2D in supplemental material). Sensitivity analyses were also conducted to see if all of
the subcategories of exposure can be combined into one exposure. We examined the effect of the
exposure on cognitive impairment and AD initially combined as presented in the overall meta-analysis
(Fig. 2.2) and while excluding each subcategory (Sleep Quality, Sleep Quantity, Obstructive Sleep
Apnea, Circadian Rhythm Abnormalities and Insomnia) from the exposure one at a time. No significant
change in the effect size was seen. We also conducted separate analyses of studies in terms of the
outcome measure since different outcome measures of cognitive impairment and AD was used. We
conducted analyses with the combined outcome of cognitive impairment and AD and examined the
change in the estimate while excluding each subcategory (cognitive tests, ICD-9/DSM IV and abnormal
124
proteins). No statistically significant difference was observed in the change in effect estimate. (Type I
error rate (α=0.05)).
Meta-regression analysis
A multivariate meta-regression analysis (Table 2.3) was conducted to examine the potential
influence of different factors on the natural logarithm of the odds ratio of sleep problems with cognitive
impairment or AD. None of the studied factors was statistically significant except for sample size (p =
.01), with larger samples showing smaller effects.
Population attributable risk percent
The PAR% suggested that approximately 15% of AD may be attributed to sleep problems, using
a conservative estimate of 30% as the prevalence of sleep problems in the populations of the included
studies and the pooled relative risk (RR: 1.55). By comparison, the prevalence of sleep problems in AD
ranges between 25% – 50%.11
DISCUSSION
Summary of the main results
In the present meta-analysis comprising 27 studies, individuals with sleep problems including
short and long sleep duration, poor sleep quality, circadian rhythm abnormality, insomnia and obstructive
sleep apnea, had a 1.68 times higher risk for the combined outcome of cognitive decline and AD
compared to individuals without sleep problems. Individuals with sleep problems including short and
long sleep duration, poor sleep quality, circadian rhythm abnormality, insomnia and obstructive sleep
apnea, had a 1.55, 1.65 and 3.78 times higher risk for AD, cognitive impairment and preclinical AD than
individuals without sleep problems respectively. A subgroup meta-analysis of classified sleep disorders
revealed a relative risk of 2.37 for obstructive sleep apnea, 1.86 for sleep duration problems, 1.62 for
sleep quality problems, and 1.38 for both insomnia and circadian rhythm abnormalities. Various aspects
of sleep problems examined in this analysis each significantly increased the risk for cognitive impairment
125
or AD. Approximately 15% of cognitive impairment or AD may be attributed to sleep problems including
short and long sleep duration, poor sleep quality, circadian rhythm abnormality, insomnia and obstructive
sleep apnea.
Effect of sleep problems on cognitive impairment or Alzheimer’s disease (meta-analysis)
Altogether, sleep problems including short and long sleep duration, poor sleep quality, circadian
rhythm abnormality, insomnia and obstructive sleep apnea, significantly increased cognitive impairment
or AD risk. Our findings show a RR increase seemingly inverse to diagnostic confidence (e.g., 1.55, 1.65
and 3.78 for AD, MCI and preclinical AD). These findings may reflect the use of more objective
investigative or diagnostic measures to identify individuals with preclinical AD in our study, however it
highlights potential mechanistic relationships suggesting that sleep is not only associated with cognitive
impairment, or symptomatic AD but is also associated with changes in AD biomarkers predictive of
persons that ultimately develop AD. Although there was heterogeneity in effect sizes, the variables
included in the meta-analysis failed to explain significant amounts of this variability. OSA appeared to be
a strong risk factor for AD. This may be explained by the fact that hypoxia plays a vital role and may
contribute to increases in Aβ production64, 65 thereby facilitating AD pathogenesis.66 Insomnia and
circadian rhythm abnormalities were each associated with an intermediate risk for cognitive impairment
or AD. Sleep quantity and quality seemed more important than insomnia and circadian activity rhythm as
a risk factor for cognitive impairment or AD. This could be because individuals with insomnia and/or
circadian rhythm abnormalities tend to have poor sleep quantity and quality, and as such the effect seen
could possibly reflect other underlying sleep problems such that individuals in the sleep quantity or
quality groups may also have associated insomnia and/or circadian rhythm abnormalities. Both long sleep
(> 8 h/day) and short sleep duration (< 7 h/day) increased the risk for cognitive impairment or AD when
compared to normal sleepers (7 – 8 h/day), with a higher risk seen in long sleepers. Although certain
prospective studies have shown associations between long sleep and cognitive impairment,67, 68 it is highly
plausible that this association is as a result of a general decline with old age and maybe reflective of other
126
disease conditions such as depression or other neurological disorders. AD patients, like in Parkinson’s
disease, may also exhibit a more bimodal distribution of sleep patterns (i.e. short or long) possibly
reflective of some pre-existing genetic predisposition that develops at disease inception.
Potentially influencing factors (meta-regression analysis)
The heterogeneity seen across studies could not be explained by our subgroup meta-analyses,
except for sample size and the different AD measures. It is likely that much of the heterogeneity can be
explained by the conceptual differences in exposure and/or outcome assessment. We explored this further
by conducting several sensitivity analyses on both our exposure and outcome, each initially combined as
presented in the overall meta-analysis and also with exclusion of each subcategory one at a time. No
significant change in the effect size was seen. Sensitivity analyses examining the effect of omitting one
study revealed no effect as the estimated relative risk remained stable. Several of the subgroup analyses
were close to being statistically significant. However, since there were only 52 effect sizes in the analysis,
power may have been an issue. Pooled relative risks from sub-group meta-analyses for the effect of sleep
problems on cognitive impairment or AD revealed consistent moderate to strong relative risks and these
relationships could possibly contribute to a causal interpretation of the findings. No RCT was included,
suggesting that additional work examining the effect of sleep therapy on cognitive impairment or AD
would be useful.
Among the factors tested only sample size significantly influenced the effect size between sleep
problems and AD such that larger sample size studies tended to result in smaller risk estimates for
cognitive impairment or AD while smaller sample size studies tended to result in larger risk estimates for
cognitive impairment or AD (Table 3). Overall, the results indicate that the meta-analyzed estimate
reflects an appropriate estimate of the general association. The need for proper multivariate modeling
approaches is underscored by the fact that the heterogeneity due to adjustment was close to statistical
significance in some instances, although power was limited.
Study results in comparison with other reviews on the topic
127
The results of this systematic review and meta-analysis are largely consistent with previous
reviews. Our study is the first systematic review and meta-analysis to quantify an “average” magnitude of
the association between sleep problems and cognitive impairment or AD. Previous narrative review
articles that have examined this association did so with a focus on different aspects of sleep
disturbances/disorders16-19. Palma et al.,16 focusing on sleep deprivation, reviewed how sleep loss
constitutes a risk factor for various neurologic diseases including AD. Slats et al.,18 focusing on the role of
hypocretin and melatonin, concluded that there is substantial evidence for impairments in both sleep and
circadian regulating mechanisms resulting from AD pathology, which may be linked to clinical
symptoms. Peter-Derex et al.,17 performed a thorough clinical review of sleep and AD, observing that
micro-architectural sleep alterations, nocturnal sleep fragmentation, decreased sleep duration, diurnal
napping and inversion of the sleep-wake cycle are the main disorders observed in patients with AD. Spira
et al.,19 concluded that poor sleep is a risk factor for cognitive impairment and AD and that healthy sleep
appears to play an important role in maintaining brain health with age, and could play a key role in AD
prevention. Various reviews69, 70 on the effect of OSA on AD agree on a bidirectional relationship
between these entities.
Strengths and Limitations of the study
Our systematic review allows for a systematic, scientific, constructive and yet replicable review
of the association between sleep and AD with a pre-specified hypothesis and specific inclusion and
exclusion criteria. We were able to systematically and efficiently integrate existing information from
various studies enabling us to establish whether the literature is consistent and can be generalized across
populations, settings, and/or whether findings vary significantly by particular subsets. The meta-analyses
in particular increased the power and precision of estimates of the risk of cognitive impairment or AD
from sleep problems. The explicit methods employed in our meta-analysis limit bias and, hopefully,
improved reliability and accuracy of our conclusions. Lastly, the generalizability of our findings can be
128
established as this review included a diversity of studies across various continents, providing an
interpretive context not available in any one study.
Several methodological concerns must be considered when interpreting the findings of this meta-
analysis. First, one potential limitation relates to the possible bias introduced in the paper selection
process. To minimize this bias, we adhered strictly to the guidelines and recommendations in the
Cochrane handbook 22 whenever possible. We also ensured that studies were not restricted by language to
guard against language bias. Second, the possible bias introduced by the individual studies that were
included is a further limitation. Various study designs were rated low for assigning the grade of
evidence71. With a scarcity of controlled studies, this analysis relied mostly on observational studies.
However, the subgroup meta-analysis revealed no significant difference between the prospective and non-
prospective studies, suggesting that our average estimate adequately reflects the general effect. In addition
to study design, conclusions made from the poor quality studies could introduce possible bias. We
included high, medium and low-quality studies in the summarization of findings. Our meta-regression
analysis indicated that study quality might not have influenced the overall conclusion. Third, there was
considerable variability between the studies in the assessment methods used to verify sleep problems and
AD or cognitive impairment, revealing moderate heterogeneity that could possibly raise questions about
the comparability of results across studies. Our research question focused on a broad concept of sleep
problems based on having any symptoms. Thus, we assumed a common underlying concept by making
general conclusions regarding sleep problems and cognitive impairment or AD. We based our assumption
on findings from reviews that consistently found associations of various sleep problems or disorders such
as sleep deprivation/loss,16, 19 sleep-wake abnormalities,17, 18 or sleep apnea,69, 70 with cognitive
impairment or AD. Moreover, our sensitivity analyses on the exposure and outcome measures which
involved examining the effect of sleep problems on cognitive impairment and AD with the exposure or
outcome measures combined and separated into subcategories did not reveal any statistical significant
change in the estimate. The persistence of heterogeneity in the subgroup meta-analysis of sleep
129
parameters and various sleep problems and disorders suggests within-study variation, possibly due to lack
of objective assessment, standardization or success in controlling for psychiatric and somatic co-
morbidities. However, the prediction interval that considered both sampling error and study heterogeneity
suggested that although the association may vary across studies, it appears to be positive in at least 97.5
percent of them. Finally, although funnel plot asymmetry was consistent with publication bias, trim-and-
fill results were also positive; suggesting that our overall result might be an overestimate, but that the
association does not appear to be explained by biased reporting of studies.
Future directions
Existing reviews note growing experimental and epidemiologic evidence for close reciprocal
associations between cognitive impairment and sleep alteration. However, longer-term longitudinal
studies are required to better understand whether sleep problems precede the onset of cognitive
impairment or AD by many years. Future studies need to be able to define causal pathways more
definitively so we can test whether neurodegeneration is leading to both sleep changes and memory
changes, or whether aging is contributing to OSA and/or sleep architecture changes that in turn affect
memory, or whether the effect of aging on sleep architecture predisposes to both apnea and memory
impairment. More research with objective sleep measures, AD biomarkers, and cognitive measures also
are needed. To facilitate appropriate comparisons among studies, it would be expedient to have greater
standardization of sleep and measures of cognitive impairment or AD. Future research is needed to make
clear the contribution of different genetic, neurodegenerative and environmental factors to sleep
impairment, as their findings can be promising for the development of new prevention and/or treatment
opportunities.
Conclusion
This meta-analysis confirmed the association between sleep and cognitive impairmentor AD and,
for the first time, consolidated the evidence by providing an “average” magnitude of the effect (RR=1.55,
95% CI: 1.25-1.93). The public health relevance of this finding is buttressed by the PAR%, suggesting
130
that approximately 15% of cognitive impairment or AD might be prevented if effective interventions
could be implemented to reduce sleep disorders.
References
1. Hebert LE, Beckett LA, Scherr PA, Evans DA. Annual incidence of Alzheimer disease in the
United States projected to the years 2000 through 2050. Alzheimer disease and associated disorders
2001;15:169-73.
2. Thies W, Bleiler L. 2013 Alzheimer's disease facts and figures. Alzheimer's & dementia : the
journal of the Alzheimer's Association 2013;9:208-45.
3. Wimo A, Jonsson L, Bond J, Prince M, Winblad B. The worldwide economic impact of dementia
2010. Alzheimer's & dementia : the journal of the Alzheimer's Association 2013;9:1-11 e3.
4. Bliwise DL. Sleep in normal aging and dementia. Sleep 1993;16:40-81.
5. Prinz PN, Peskind ER, Vitaliano PP, et al. Changes in the sleep and waking EEGs of
nondemented and demented elderly subjects. Journal of the American Geriatrics Society 1982;30:86-93.
6. Prinz PN, Vitaliano PP, Vitiello MV, et al. Sleep, EEG and mental function changes in senile
dementia of the Alzheimer's type. Neurobiol Aging 1982;3:361-70.
7. Vitiello MV, Prinz PN, Williams DE, Frommlet MS, Ries RK. Sleep disturbances in patients with
mild-stage Alzheimer's disease. Journal of gerontology 1990;45:M131-8.
8. Hoch CC, Reynolds CF, 3rd, Kupfer DJ, Houck PR, Berman SR, Stack JA. Sleep-disordered
breathing in normal and pathologic aging. The Journal of clinical psychiatry 1986;47:499-503.
9. Mant A, Saunders NA, Eyland AE, Pond CD, Chancellor AH, Webster IW. Sleep-related
respiratory disturbance and dementia in elderly females. Journal of gerontology 1988;43:M140-4.
10. McCurry SM, Reynolds CF, Ancoli-Israel S, Teri L, Vitiello MV. Treatment of sleep disturbance
in Alzheimer's disease. Sleep medicine reviews 2000;4:603-28.
11. Moran M, Lynch CA, Walsh C, Coen R, Coakley D, Lawlor BA. Sleep disturbance in mild to
moderate Alzheimer's disease. Sleep Med 2005;6:347-52.
131
12. Meeks TW, Ropacki SA, Jeste DV. The neurobiology of neuropsychiatric syndromes in
dementia. Current opinion in psychiatry 2006;19:581-6.
13. Rauchs G, Schabus M, Parapatics S, et al. Is there a link between sleep changes and memory in
Alzheimer's disease? Neuroreport 2008;19:1159-62.
14. Rauchs G HC, Bertran F, Desgranges B, Eustache F. . Sleep and episodic memory: a review of
the literature in young healthy subjects and potential links between sleep changes and memory
impairment observed during aging and Alzheimer’s disease. . Rev Neurol (Paris) 2010; 2010;166::81.
15. Bloom HG, Ahmed I, Alessi CA, et al. Evidence-based recommendations for the assessment and
management of sleep disorders in older persons. Journal of the American Geriatrics Society 2009;57:761-
89.
16. Palma JA, Urrestarazu E, Iriarte J. Sleep loss as risk factor for neurologic disorders: a review.
Sleep medicine 2013;14:229-36.
17. Peter-Derex L, Yammine P, Bastuji H, Croisile B. Sleep and Alzheimer's disease. Sleep medicine
reviews 2014.
18. Slats D, Claassen JA, Verbeek MM, Overeem S. Reciprocal interactions between sleep, circadian
rhythms and Alzheimer's disease: focus on the role of hypocretin and melatonin. Ageing research reviews
2013;12:188-200.
19. Spira AP, Chen-Edinboro LP, Wu MN, Yaffe K. Impact of sleep on the risk of cognitive decline
and dementia. Current opinion in psychiatry 2014;27:478-83.
20. Pai M, McCulloch M, Gorman JD, et al. Systematic reviews and meta-analyses: an illustrated,
step-by-step guide. The National medical journal of India 2004;17:86-95.
21. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and
meta-analyses: the PRISMA statement. BMJ (Clinical research ed.) 2009;339:b2535.
22. Higgins JPT GSe. Cochrane handbook for systematic reviews of interventions. Version 5.1.0
[updated March 2011]. . http://www.cochranehandbook.org [accessed 11.23.14]. 2011.
132
23. Medicine. AAoS. The international classification of sleep disorders, revised: diagnostic and
coding manual. Chicago, USA: . American Academy of Sleep Medicine, ISBN 0-9657220-1-5; 2001.
2001.
24. Kling RN, McLeod CB, Koehoorn M. Sleep problems and workplace injuries in Canada. Sleep
2010;33:611-8.
25. Uehli K, Mehta AJ, Miedinger D, et al. Sleep problems and work injuries: a systematic review
and meta-analysis. Sleep medicine reviews 2014;18:61-73.
26. Bhui K, Stansfeld S, Hull S, Priebe S, Mole F, Feder G. Ethnic variations in pathways to and use
of specialist mental health services in the UK. Systematic review. The British journal of psychiatry : the
journal of mental science 2003;182:105-16.
27. Dinos S, Khoshaba B, Ashby D, et al. A systematic review of chronic fatigue, its syndromes and
ethnicity: prevalence, severity, co-morbidity and coping. Int J Epidemiol 2009;38:1554-70.
28. Wells G SB, O’Connell D, Peterson J,Welch V, Losos M, et al. . The Newcastle-Ottawa scale
(NOS) for assessing the quality of nonrandomised studies in metaanalyses,.
http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp/; 2011 [accessed 11.25.14]. 2011.
29. D. P. Personal communication 2011. 2011.
30. Borenstein Michael HLV HJPT, Rothstein H.R. Introduction to Meta-Analysis. Chichester, U.K.:
John Wiley & Sons, 2009. , © 2009:45-9.
31. Blackwell T, Yaffe K, Ancoli-Israel S, et al. Poor sleep is associated with impaired cognitive
function in older women: the study of osteoporotic fractures. The journals of gerontology. Series A,
Biological sciences and medical sciences 2006;61:405-10.
32. Benito-Leon J, Bermejo-Pareja F, Vega S, Louis ED. Total daily sleep duration and the risk of
dementia: a prospective population-based study. Eur J Neurol 2009;16:990-7.
33. Chang WP, Liu ME, Chang WC, et al. Sleep apnea and the risk of dementia: a population-based
5-year follow-up study in taiwan. PLoS One 2013;8:e78655.
133
34. Cricco M, Simonsick EM, Foley DJ. The impact of insomnia on cognitive functioning in older
adults. Journal of the American Geriatrics Society 2001;49:1185-9.
35. Gottlieb DJ, DeStefano AL, Foley DJ, et al. APOE epsilon4 is associated with obstructive sleep
apnea/hypopnea: the Sleep Heart Health Study. Neurology 2004;63:664-8.
36. Lim AS, Kowgier M, Yu L, Buchman AS, Bennett DA. Sleep Fragmentation and the Risk of
Incident Alzheimer's Disease and Cognitive Decline in Older Persons. Sleep 2013;36:1027-32.
37. Osorio RS, Ayappa I, Mantua J, et al. The interaction between sleep-disordered breathing and
apolipoprotein E genotype on cerebrospinal fluid biomarkers for Alzheimer's disease in cognitively
normal elderly individuals. Neurobiol Aging 2014;35:1318-24.
38. Tworoger SS, Lee S, Schernhammer ES, Grodstein F. The association of self-reported sleep
duration, difficulty sleeping, and snoring with cognitive function in older women. Alzheimer disease and
associated disorders 2006;20:41-8.
39. Keage HA, Banks S, Yang KL, Morgan K, Brayne C, Matthews FE. What sleep characteristics
predict cognitive decline in the elderly? Sleep medicine 2012;13:886-92.
40. Lim AS, Yu L, Kowgier M, Schneider JA, Buchman AS, Bennett DA. Modification of the
Relationship of the Apolipoprotein E epsilon4 Allele to the Risk of Alzheimer Disease and
Neurofibrillary Tangle Density by Sleep. JAMA neurology 2013;70:1544-51.
41. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses.
BMJ (Clinical research ed.) 2003;327:557-60.
42. Viechtbauer W. Conducting meta-analyses in R with the metafor package. . Journal of Statistical
Software, (2010). ;36:1–48. .
43. Inc. SI. SAS® 9.4 System Guide to Software Updates Cary NC, USA. SAS Institute Inc. 2013.
44. Ju YE, McLeland JS, Toedebusch CD, et al. Sleep quality and preclinical Alzheimer disease.
JAMA neurology 2013;70:587-93.
45. Merlino G, Piani A, Gigli GL, et al. Daytime sleepiness is associated with dementia and cognitive
decline in older Italian adults: A population-based study. Sleep medicine 2010;11:372-7.
134
46. Nebes RD, Buysse DJ, Halligan EM, Houck PR, Monk TH. Self-reported sleep quality predicts
poor cognitive performance in healthy older adults. The journals of gerontology. Series B, Psychological
sciences and social sciences 2009;64:180-7.
47. Oosterman JM, van Someren EJ, Vogels RL, Van Harten B, Scherder EJ. Fragmentation of the
rest-activity rhythm correlates with age-related cognitive deficits. Journal of sleep research 2009;18:129-
35.
48. Saint Martin M, Sforza E, Barthelemy JC, Thomas-Anterion C, Roche F. Does subjective sleep
affect cognitive function in healthy elderly subjects? The Proof cohort. Sleep medicine 2012;13:1146-52.
49. Spira AP, Gamaldo AA, An Y, et al. Self-reported sleep and beta-amyloid deposition in
community-dwelling older adults. JAMA neurology 2013;70:1537-43.
50. Schmidt FM, Kratzsch J, Gertz HJ, et al. Cerebrospinal fluid melanin-concentrating hormone
(MCH) and hypocretin-1 (HCRT-1, orexin-A) in Alzheimer's disease. PloS one 2013;8:e63136.
51. Sterniczuk R, Theou O, Rusak B, Rockwood K. Sleep disturbance is associated with incident
dementia and mortality. Current Alzheimer research 2013;10:767-75.
52. Blackwell T, Yaffe K, Laffan A, et al. Associations of objectively and subjectively measured
sleep quality with subsequent cognitive decline in older community-dwelling men: the MrOS sleep study.
Sleep 2014;37:655-63.
53. Grandner MA, Drummond SP. Who are the long sleepers? Towards an understanding of the
mortality relationship. Sleep medicine reviews 2007;11:341-60.
54. Foley D, Monjan A, Masaki K, et al. Daytime sleepiness is associated with 3-year incident
dementia and cognitive decline in older Japanese-American men. Journal of the American Geriatrics
Society 2001;49:1628-32.
55. Hahn EA, Wang HX, Andel R, Fratiglioni L. A Change in Sleep Pattern May Predict Alzheimer
Disease. The American journal of geriatric psychiatry : official journal of the American Association for
Geriatric Psychiatry 2013.
135
56. Jelicic M, Bosma H, Ponds RW, Van Boxtel MP, Houx PJ, Jolles J. Subjective sleep problems in
later life as predictors of cognitive decline. Report from the Maastricht Ageing Study (MAAS).
International journal of geriatric psychiatry 2002;17:73-7.
57. Osorio RS, Pirraglia E, Aguera-Ortiz LF, et al. Greater risk of Alzheimer's disease in older adults
with insomnia. Journal of the American Geriatrics Society 2011;59:559-62.
58. Tranah GJ, Blackwell T, Stone KL, et al. Circadian activity rhythms and risk of incident dementia
and mild cognitive impairment in older women. Annals of neurology 2011;70:722-32.
59. Virta JJ, Heikkila K, Perola M, et al. Midlife sleep characteristics associated with late life
cognitive function. Sleep 2013;36:1533-41, 41A.
60. Walsh CM, Blackwell T, Tranah GJ, et al. Weaker circadian activity rhythms are associated with
poorer executive function in older women. Sleep 2014;37:2009-16.
61. Yaffe K, Laffan AM, Harrison SL, et al. Sleep-disordered breathing, hypoxia, and risk of mild
cognitive impairment and dementia in older women. JAMA : the journal of the American Medical
Association 2011;306:613-9.
62. Viechtbauer W. Conducting meta-analyses in R with the metafor package. Journal of Statistical
Software 2010;36:1-48.
63. Duval S, Tweedie R. Trim and fill: A simple funnel-plot-based method of testing and adjusting
for publication bias in meta-analysis. Biometrics 2000;56:455-63.
64. Li L, Zhang X, Yang D, Luo G, Chen S, Le W. Hypoxia increases Abeta generation by altering
beta- and gamma-cleavage of APP. Neurobiology of aging 2009;30:1091-8.
65. Zhang X, Zhou K, Wang R, et al. Hypoxia-inducible factor 1alpha (HIF-1alpha)-mediated
hypoxia increases BACE1 expression and beta-amyloid generation. The Journal of biological chemistry
2007;282:10873-80.
66. Sun X, He G, Qing H, et al. Hypoxia facilitates Alzheimer's disease pathogenesis by up-
regulating BACE1 gene expression. Proceedings of the National Academy of Sciences of the United
States of America 2006;103:18727-32.
136
67. Benito-Leon J, Louis ED, Bermejo-Pareja F. Cognitive decline in short and long sleepers: a
prospective population-based study (NEDICES). J Psychiatr Res 2013;47:1998-2003.
68. Johar H, Kawan R, Emeny RT, Ladwig KH. Impaired Sleep Predicts Cognitive Decline in Old
People: Findings from the Prospective KORA Age Study. Sleep 2016;39:217-26.
69. Gagnon K, Baril AA, Gagnon JF, et al. Cognitive impairment in obstructive sleep apnea.
Pathologie-biologie 2014.
70. Kinugawa K, Nguyen-Michel VH, Mariani J. [Obstructive sleep apnea syndrome: a cause of
cognitive disorders in the elderly?]. La Revue de medecine interne / fondee ... par la Societe nationale
francaise de medecine interne 2014;35:664-9.
71. Atkins D, Best D, Briss PA, et al. Grading quality of evidence and strength of recommendations.
BMJ (Clinical research ed.) 2004;328:1490.
137
Table 2.1. Characteristics of included studies. Meta-analysis of sleep, cognitive decline and Alzheimer's disease/cognitive decline
Authors, year Study design
(study quality)
Sample N
(location)
Study population, mean age/age
range in years (follow-up time,
years)
Sleep assessment Cognitive/AD
measures
Impact of sleep problems/disorders
on AD
Benito-León et al.
201432
Prospective cohort
(high)
3,857
(Spain)
Community-dwelling elderly, 65
(median,12.5)
Daily total sleep duration
(self-reported)
Death
certificates,
ICD-9
Long sleepers (≥ 9h), aHR=1.63 [1.04-
2.56]* Short sleepers (≤ 5h), aHR=0.76
[0.33-1.74]
Blackwell et al.
200631
Cross sectional
(high)
2,242
(U.S.)
Community-dwelling women, 83.5
(N/A)
Poor sleep (wrist
actigraphy)
MMSE, Trails
B
aOR=1.48 [1.44–1.52]*ˠ
Blackwell et al.
201451
Prospective cohort
(high)
2822
(U.S.)
Community-dwelling men, 76, (3.4 ±
0.5 y)
Sleep quality (wrist
actigraphy, PSQI, ESS).
MMSE, Trails
B
aOR=1.44 [1.38, 1.51]*ˠ
Chang et al.
201333
Retrospective cohort
(high)
8484
(Taiwan)
Sleep apnea patients. 40+ (5) Sleep apnea (ICD-9) ICD-9 aHR=1.70 [1.26-2.31]*;
Females, aHR=2.38 [1.51–3.74]*;
Males, 50-59 years, aHR=6.08 [3.15-
11.70]*; 2.5 to 5 y of follow-up,
aHR=2.31 [2.07-2.58]*ˠ
Cohen-Zion et al.
200152
Prospective cohort
(Low)
46
(U.S.)
Community-dwelling, 80 (2) Wrist actigraphy MMSE EDS, aOR=2.26 [1.21-4.20]*;
SA, aOR=4.04 [2.40-6.68]*
Cricco et al.
200134
Prospective cohort
(high)
6,444
(Iceland)
Community-dwelling elderly, 72 (3) Insomnia (self-reported) SPMSQ Male non-depressed, aOR=1.33 [1.25-
1.42]*; Male depressed, aOR=1.89
[1.60-2.23]*; Female non-depressed,
aOR=1.07 [1.03-1.11]*;
Female depressed, aOR=1.23 [1.16-
1.31]*
138
Foley et al.
200153
Prospective cohort
(high)
2,346
(U.S.)
Community-dwelling Japanese-
American men, 76.6 (3)
Sleep disturbance (self-
reported)
CASI, DSM-
IIIR.
aOR=1.44 [1.34-1.54]*
Hahn et al.
201354
Prospective cohort
(high)
214
(U.S.)
Community-dwelling elderly, 83.4
(9)
Poor sleep (CPRS) DSM-IIIR
NINCDS-
ADRDA
aOR=1.78 [1.50-2.11]*ˠ
Jelicic et al.
200255
Prospective cohort
(medium)
838
(Netherlands)
Community-dwelling elderly, 63.3 (3) Poor sleep (self-reported) MMSE aOR=1.46 [1.38-1.54]*
Keage et al.
201239
Prospective cohort
(high)
2,012
(Australia,
U.K.)
Community-dwelling elderly, 64-94
(10)
EDS (self-report) MMSE aOR=2.09 [1.75-2.50]*ˠ
Lim et al.
2013a40
Prospective cohort
(medium)
698
(Canada)
Community-dwelling elderly, 83.3 (6) Sleep consolidation
(actigraphy)
NINCDS-
ADRDA
aHR=1.49 [1.39-1.60]*
Lim et al.
2013b36
Prospective cohort
(high)
737
(Canada)
Community-dwelling elderly, 81.6 (6) Sleep fragmentation
(actigraphy)
Cognitive
performance
aHR =1.22 [1.03-1.44]*;
Male, aHR=1.33 [0.90-1.96]*;
Female, aHR=1.29 [0.99-1.69]*
Merlino et al.
201044
Cross-sectional
(low)
750
(Italy)
Institutionalized and community-
dwelling elderly, 76.7 (NA)
Insomnia (self-report) DSM-IV-TR,
NINCS-
ARDRA,
NINDS-
AIREN
cOR=1.76 [1.02–3.11]*
Nebes et al.
200945
Cross-sectional
(medium)
157
(U.S.)
Community volunteers, 72.2 (NA) Sleep quality (PSQI) RBANS
TONI
aOR=2.21 [1.88-2.59]*
139
Ooms et al.
201461
Randomized controlled
trial (high)
26
(Netherlands)
Middle-aged men, 49.9 (1 night) Sleep deprivation (PSG) CSF Aβ42,
Aβ40, P-tau,
T-tau
aOR=5.06 [1.48-17.28]*
Oosterman et al.
200946
Cross-sectional
(medium)
144
(Netherlands)
Cardiology/internal medicine
outpatients, 69.5 (NA)
Sleep fragmentation
(actigraphy)
Cognitive
performance
aOR=1.84 [1.60-2.12]*
Osorio et al.
201156
Prospective cohort
(high)
346
(U.S.)
Brain aging study participants, 24-96
(7.7)
Insomnia (HAM-D) MMSE Insomnia, aOR=2.39 [1.03–5.55]*;
Insomnia w/o depression, aOR=3.32
[1.33–8.28]*
Osorio et al.
201437
Cross-sectional
(medium)
95
(U.S.)
Brain aging study participants, 67.7
(NA)
SDB (AHI ≥ 15). CSF P-tau, T-
tau, Aβ42
aOR=4.46 [2.35-8.48]*
Saint Martin et al.
201247
Cross-sectional
(medium)
272
(France)
Community-dwelling elderly, 74.8
(NA)
Sleep quality (PSQI)
MMSE,
GBSRT,
BVRT, Trail
A & B,
WAIS-III,
Stroop test,
AFCFT
aOR=1.65 [1.46-1.86]*
Schmidt et al.
201349
Case-control
(medium)
66
(Germany)
Consecutive AD patients and controls,
62.9 (NA)
Melanin-concentrating
hormone (MCH) and
hypocretin-1 (HCRT-1,
CERAD,
WMS-R, Trail
B, MRI, CSF
aOR=5.1 [1.98-13.12]*
140
orexin-A) (fluorescence
immunoassay (FIA))
Aβ42, Aβ40,
P-tau, T-tau
Spira et al.
201348
Cross-sectional
(medium)
70
(U.S.)
Normative aging study participants,
76.4 (NA)
WHIIRS β-amyloid Lower sleep quality, aOR=3.38 [2.05–
5.58]*; Insomnia, aOR=1.87 [1.19–
2.94]*
Sterniczuk et al.
201350
Retrospective cohort
(medium)
28,697
(Canada)
Community-dwelling middle-aged
and elderly, 64.6 (4.3)
Sleep disturbance Index
(SDI) (self-report)
AD (self-
reported),
aOR=1.18 [1.17-1.19]*
Tranah et al.
201157
Prospective cohort
(high)
1,282
(U.S.)
Community-dwelling women, 82.7
(4.9)
CAR (actigraphy). Modified
MMSE CVLT
IQCODE,
dementia
(self-reported)
aOR=1.66 [1.55-1.77]*
Tworoger et al.
200638
Prospective cohort
(medium)
1,844
(U.S.)
Community-dwelling women, 74.2 (2) Sleep duration (self-
reported)
Cognitive
performance
≤ 6 h sleep, aOR=2.35 [1.90-2.91]*;
Difficulty initiating sleep, aOR=2.21
[1.84-2.65]*;
≥ 8 h sleep, a0R=1.31 [0.76-2.29]
Virta et al.
201358
Prospective cohort
(high)
2,336
(Finland)
Elderly twins, 65+ (18-26) Poor sleep (self-reported) Linear
cognitive
score (self-
report)
≤ 7 h sleep, aOR=4.15 [3.12–5.54]*;
Hypnotic use, aOR=1.23 [0.73–2.08]
141
Walsh et al.
201459
Prospective cohort
(high)
287
(U.S.)
Community-dwelling women, 82.8 (5) CAR (actigraphy) Modified
MMSE
CVLT, Trails
B
aOR=3.43 [1.74–6.75]*
Yaffe et al.
201160
Prospective cohort
(high)
298
(U. S.)
Community-dwelling women, 82.3
(2.3)
SDB (AHI ≥ 15). DSM-IV Sleep-wake problems, aOR=1.09
[0.90–1.32]ˠ; SDB , aOR=1.87 [1.61–
2.18]*ˠ; Sleep
duration, aOR=0.56 [0.46-0.67]*ˠ
*: significant, ˠ: multiple outcome measure on same subject, so mean of the outcome computed. Abbreviations; a: adjusted, Aβ40/42: amyloid beta-40/42, AD: Alzheimer's disease, AFCFT: alphabetic fluency and category fluency
tasks, AHI ≥ 15: apnea hypopnea index of 15 or more events per hour of sleep, APOE: apolipoprotein epsilon4, BVRT: Benton visual retention test, c: crude, CAR: circadian activity rhythms, CASI: cognitive abilities screening
instrument, CERAD: consortium to establish a registry for Alzheimer’s disease, CPRS: comprehensive psychopathological rating scale, CSF: cerebrospinal fluid, CVLT: California verbal learning test , DSM-IIIR/IV-TR:
diagnostic and statistical manual of mental disorders, third edition/fourth edition, text revised, EDS: Excessive daytime sleepiness, ESS: Epworth sleepiness scale, GBSRT: Grober and Buschké selective reminding test, HAM-D:
Hamilton Depression Rating Scale , HR: hazard ratio, ICD-9/10: international classification of diseases ninth/tenth edition AD criteria, IQCODE: informant questionnaire on cognitive decline in the elderly, , MMSE: mini mental
state examination, MRI: magnetic resonance imaging, N: number of participants, NA: not applicable, N/A: not available, NINCDS-ADRDA: national institute of neurological and communicative disorders and stroke and the
Alzheimer's disease and related disorders association, NINDS-AIREN: national institute of neurological disorders and stroke and association internationale pour la recherché et l'enseignement en neurosciences vascular dementia
criteria, OR: odds ratio, PSG: polysomnography, PSQI: Pittsburgh sleep quality index, P-tau: phosphorylated tau, RBANS: repeatable battery for the assessment of neuropsychological status, RR: relative risk, SDB: sleep
disordered breathing, SPMSQ: short portable mental status questionnaire , TONI: test of nonverbal intelligence, Trails B: trail making b test, T-tau: total-tau, WAIS-III: Wechsler adult intelligence scale third version, WHIIRS:
women’s health initiative insomnia rating scale, WMS-R: Wechsler memory scale-revision
142
Table 2.2: Pooled relative risks from sub-group meta-analysis for the effect of sleep on Alzheimer's disease/cognitive decline
Subgroup N Relative risk (95% CI)
Sleep parameter
Sleep Efficiency 7 1.98 (1.11, 3.52)
Sleep Latency 4 1.33 (1.13, 1.57)
Wake After Sleep Onset 7 1.26 (1.19, 1.33)
Daytime vs. nighttime sleepiness
Daytime Sleepiness 7 1.49 (1.17, 1.90)
Nighttime problems 31 1.7 (1.48, 1.94)
Night & Daytime problems 6 1.45 (1.17, 1.78)
Gender
Female 7 1.56 (1.24, 1.96)
Male 13 1.49 (1.23, 1.79)
Both male & female 24 1.74 (1.49, 2.03)
Sleep duration
Long sleepers > 8 hrs/day 6 2.05 (1.24, 3.38)
Short sleepers ≤ 7 hrs/day 6 1.63 (1.13, 2.34)
Follow-up time
< 5 years 18 1.4 (1.23, 1.59)
≥ 5 years 18 1.76 1.46, 2.112
Classified sleep disorders
Insomnia 11 1.5 (1.25, 1.80)
Obstructive Sleep Apnea Syndrome 11 1.96 (1.59, 2.41)
Circadian-Rhythm Sleep Disorders 8 1.38 (1.09, 1.75)
143
Sleep Depth (Not classified) 18 1.62 (1.38, 1.91)
Continent
North America 25 1.53 (1.35, 1.74)
Europe 14 1.54 (1.27, 1.86)
Australia 1 2.09 (1.16, 3.78)
Asia 4 2.23 (1.60, 3.11)
Temporality
Prospective studies 36 1.57 (1.41, 1.77)
Non-prospective studies 8 1.70 (1.44, 2.02)
Study quality
High quality 31 1.58 (1.40, 1.79)
Medium quality 10 1.67 (1.32, 2.10)
Low quality 3 2.17 (1.38, 3.41)
Abbreviations N: Number of relative risk estimates, CI: confidence interval
144
Table 2.3: Results of meta-regression analysis examining potential effects of different factors on the natural logarithm of the odds ratio between sleep and Alzheimer's disease/cognitive decline
Meta-regression variables
Risk for Alzheimer's disease
Univariable analysis Multivariable analysis
Coef. p-Value Coef. p-Value
Proportion of females (in %) -0.0004 0.8647 0.0007 0.7224
Mean age (in years) -0.0022 0.7898 -0.0147 0.2287
Sample size (in number of people) -0.000 0.0413 -0.000 0.0175
Quality rating (in %) 0.1043 0.3046 0.0198 0.8721
145
Figure 2.1: Study retrieval and selection for effects of sleep on Alzheimer’s disease/cognitive decline meta-analysis.
Studies identified through database searching (n = 2341)
(PubMed=904, Embase=1026, Web of Science=305, Cochrane library=106)
Studies after duplicates removed (n = 1221)
Studies screened for titles (n = 1221)
Studies excluded (n = 180)
Full-text articles assessed for eligibility (n = 201)
Full-text articles excluded, with reasons (n = 130)
- Review article (n = 7) - No risk estimate (n = 17) - Sleep not assessed (n = 18) - Relationship not studied (n =
53) - Not peer reviewed (n = 10) - Alzheimer not examined (n
= 25)
Studies included in systematic review
(n = 72)
Studies included in meta-analysis (n = 27)
Article identified from reference search (n = 1)
Studies screened for abstracts
(n = 1041)
Studies excluded (n = 840)
Duplicates removed (n = 1120)
Studies excluded from analysis, with reasons (n = 45)
- Study design (n = 16) - Duplicate studies (n =
12) - Insufficient
information for a meta-analysis (n = 17)
146
Figure 2.2 Study retrieval and selection for effects of sleep on cognitive decline and/or Alzheimer’s disease meta-analysis
Abbreviations: EDS, excessive daytime sleepiness; RR, relative risk; SA, sleep apnea; SDB, sleep disordered breathing; TST, total sleep time; WASO, wake after sleep onset; w/D, with depression; w/o D, without depression; w/o, without
147
Figure 2.3: Forest plot presenting sub-group meta-analysis based on risk estimates for the effect of sleep on Cognitive decline, Pre-clinical and Symptomatic Alzheimer’s disease
Abbreviations: EDS, excessive daytime sleepiness; RR, relative risk; SA, sleep apnea; SDB, sleep disordered breathing; TST, total sleep time; WASO, wake after sleep onset; w/D, with depression; w/o D, without depression; w/o, without
148
Figure 2.4*: Trim and fill funnel plot for effects of sleep on Alzheimer’s disease/cognitive decline meta- analysis
*Filled dots represent published studies. Empty dots represent unpublished and/or imputed studies
Observed Outcome
Sta
nd
ard
Err
or
0.7
92
0.5
94
0.3
96
0.1
98
0.0
00
-1.00 0.00 1.00 2.00
149
Figure 2.5 Forest plot presenting sub-group meta-analysis based on risk estimates for the effect of different sleep problems and disorders on cognitive decline and/or Alzheimer’s disease
Abbreviations: EDS, excessive daytime sleepiness; RR, relative risk; SA, sleep apnea; SDB, sleep disordered breathing; TST, total sleep time; WASO, wake after sleep onset; w/D, with depression; w/o D, without depression; w/o, without
150
Figure 2.6: Forest plot presenting sub-group meta-analysis based on risk estimates for the effect of self-report versus actigraphic data of poor sleep on cognitive decline and/or Alzheimer’s disease
Abbreviations: EDS, excessive daytime sleepiness; RR, relative risk; SA, sleep apnea; SDB, sleep disordered breathing; TST, total sleep time; WASO, wake after sleep onset; w/D, with depression; w/o D, without depression; w/o, without
151
APPENDIX A
SUPPLEMENTAL TABLES AND FIGURES: SLEEP AND AD META-ANALYSIS
Table 1A: Main findings from included studies and comments regarding computed conversions of the different indices to a common index: odds ratios and regarding
computing a summary effect size of the impact of sleep problems on AD
Authors, year, reference
number, (country) Study Main Findings Comments for conversion
Impact of sleep problems/disorders on
Alzheimer's disease
Benito-León et al. 2014,32 (Spain)
Long sleepers (≥ 9h) aHR=1.63 [1.04-2.56] Short sleepers (≤ 5h) aHR=0.76 [0.33-1.74]
No conversions and/or mean computation done
Long sleepers (≥ 9h), aHR=1.63 [1.04-2.56]* Short sleepers (≤ 5h), aHR=0.76 [0.33-1.74]
Blackwell et al. 2006,31 (United States)
Sleep efficiency (MMSE) aOR=1.61 [1.20–2.16], Sleep efficiency (Trails B) aOR=1.96 [1.43–2.67], Sleep latency (MMSE) aOR=1.23 [1.13–1.33]; Sleep latency (Trails B) aOR=1.13 [1.04–1.24], WASO (MMSE) aOR=1.15 [1.06–1.23]; Sleep latency (Trails B) aOR=1.24 [1.15–1.34]. Napping (MMSE) aOR=1.42 [1.05–1.93]; Napping (Trails B) aOR=1.74 [1.26–2.40].
We accounted for multiple outcome measures on the same subjects by computing the mean of the effect sizes for each study and used the average effect size as the unit of analysis
aOR=1.44 [1.07–1.94]*ˠ
Blackwell et al. 2014,51 (United States)
Sleep efficiency (Trails B) aOR=1.53 [1.07, 2.18]; WASO aOR=1.47 [1.09, 1.98]; LWEP aOR=1.38 [1.02, 1.86]. 3MS:LWEP aOR=1.36 [1.09, 1.71]
We accounted for multiple exposure measures on the same subjects by computing the mean of the effect sizes for each study and used the average effect size as the unit of analysis
aOR=1.48 [1.17, 1.88]*ˠ;
Chang et al. 2013,33 Taiwan
SA aHR=1.70 [1.26-2.31] Females aHR=2.38 [1.51–3.74] Males 50-59 years aHR= 6.08 [1.96-18.90], Females ≥ 70 years aHR=3.20 [1.71–6.00]. First 2.5 years of follow-up aHR: =2.04 [1.35-3.07]. 5 years of follow-up aHR=1.70 [1.26-2.31]
No conversions and/or mean computation done for separate populations and measures. Follow-up time measures were on the same subjects as such we computed the mean of the effect sizes and used the average effect size as the unit of analysis
aHR=1.70 [1.25-2.31]*; Females, aHR=2.38 [1.51–3.74]*; Males, 50-59 years, aHR=6.08 [1.96-18.90]*; 2.5 to 5 y of follow-up, aHR=2.31 [1.45-3.69]*ˠ
152
Cohen-Zion et al. 2001,52 United States
Increased RDI (t=-2.17, 42 df, r=-0.32, P=.036) Increased EDS (t=-3.32, 42 df, r=-0.46, P=.002) EDS (t39 = -2.45, r=-0.37, P= .019)-Adjusted model RDI (t39=- 0.66, r=-0.11, P=.515)-Adjusted model
Converted from r to d (d = d2r/√(1-r^2 ) ) and from d to the log(OR) (Log(OR) = dp/√3 and mean computation done for separate populations
SA, aOR=2.26 [0.75-6.80]; EDS, aOR=4.04 [1.48-11.06]*
Cricco et al. 2001,34 Iceland
Non-depressed men chronic insomnia aOR=1.49 [1.03–2.14] Non-depressed men incident insomnia aOR=1.16 [0.82–1.65] Depressed men incident insomnia aOR=1.60 [0.87-2.95] Depressed men chronic insomnia aOR=2.18 [1.30-3.67] Non-depressed women chronic insomnia aOR=1.18 [0.92–1.52] Non-depressed women incident insomnia aOR=0.95 [0.71–1.26] Depressed women incident insomnia aOR=1.10 [0.75-1.61] Depressed women chronic insomnia aOR=1.36 [1.01-1.84]
We accounted for multiple exposure measures on the same subjects by computing the mean of the effect sizes for each study and used the average effect size as the unit of analysis for each population
Male non-depressed, aOR=1.33 [0.93-1.90]ˠ; Male depressed, aOR=1.89 [1.07-3.34]*ˠ; Female non-depressed, aOR=1.07 [1.03-1.11]*ˠ; Female depressed, aOR=1.23 [0.87-1.73]ˠ;
Foley et al. 2001,53 United States
Excessive daytime sleepiness DSM-IIIR aOR=2.19 [1.37–3.50] Excessive daytime sleepiness CASI aOR=1.44 [1.01–2.08]. Insomnia DSM-IIIR aOR=0.99 [0.70-1.41] Insomnia CASI aOR=1.14 [0.91-1.43]
We accounted for multiple exposure measures on the same subjects by computing the mean of the effect sizes for each exposure category and used the average effect size as the unit of analysis
EDS: aOR=1.82 [1.19-2.79]* Insomnia: aOR=1.07 [0.81-1.42]
Hahn et al. 2013,54 United States
Reduced sleep (Dementia)a aHR=1.75 [1.04-2.93] Reduced sleep (AD)a aHR=2.01 [1.12-3.61] Reduced sleep (Dementia)b aHR=1.50 [0.87-2.59] Reduced sleep (AD)b aHR=1.69 [0.91-3.14] Reduced sleep (Dementia)c aHR=1.73 [1.02-2.91] Reduced sleep (AD)c aHR=2.02 [1.11-3.65] Reduced sleep (Dementia)d aHR=1.68 [0.98-2.86] Reduced sleep (AD)d aHR=1.86 [1.00-3.70]
We accounted for multiple outcome measures on the same subjects by computing the mean of the effect sizes for each study and used the average effect size as the unit of analysis
aOR=1.78 [0.86-3.70]ˠ
Jelicic et al. 2002,55 The Netherlands
Overall sleep problems continuous scores β=-0.05 (-0.09,-0.00) Overall sleep problems dichotomized scores β=-0.38 (-0.71,-0.05)
Converted from β to OR aOR=1.46 [1.05-2.03]*
153
Ju et al. 2013, United States
Sleep efficiency OR=5.60 [0.97- 32.5] No conversions and/or mean computation done
OR=5.60 [0.97- 32.5]
Keage et al. 2012,39 Australia, & United Kingdom
Napping (2years) aOR=0.38 [0.22–0.67]. Napping (10years) aOR=0.48 [0.29–0.78]. Sleep duration (<= 6.5 h 10years) aOR=2.02 [1.17–3.48] Sleep duration (>= 9 h 10years) aOR=1.27 [0.64-2.48] Daytime sleepiness (10 years) aOR=2.43 [1.24–4.75].
We accounted for multiple exposure measures on the same subjects by computing the mean of the effect sizes for each exposure category and used the average effect size as the unit of analysis
Sleep Quality: aOR=1.10 [0.58-2.07]ˠ <= 6.5h sleep: aOR=2.02 [1.17-3.48]*ˠ; >= 9h sleep: aOR=1.27 [0.65-2.48]ˠ;
Lim et al. 2013a,40 Canada Better Sleep consolidation aHR 0.67 [0.46-0.97] We converted the risk estimate for better sleep consolidation to a risk estimate for poor sleep consolidation
aHR=1.49 [1.02-2.17]*
Lim et al. 2013b,36 Canada Sleep fragmentation aHR =1.22 [1.03-1.44] Male aHR=1.33 [0.90-1.96] Female aHR=1.29 [0.99-1.69]
No conversions and/or mean computation done
aHR =1.22 [1.03-1.44]*; Male, aHR=1.33 [0.90-1.96]; Female, aHR=1.29 [0.98-1.69]
Merlino et al. 2010,44 Italy Excessive daytime sleepiness OR=1.76 (1.02–3.11) No conversions and/or mean computation done
cOR=1.76 [1.02–3.11]*
Nebes et al. 2009,45 United States
Sleep latency correlated the RBANS (r = −.241, p = .002) Sleep latency correlated the TONI (r = −.258, p = .001). Sleep efficiency correlated with the RBANS (r = .185, p = .020), Sleep efficiency correlated with the TONI (r = .197, p = .013), Sleep efficiency correlated with the N-Back (r = .211, p = .008).
Converted from r to d (d = d2r/√(1-r^2 ) ) and from d to the log(OR) (Log(OR) = dp/√3 and computed the mean of the effect sizes for each study and used the average effect size as the unit of analysis
aOR=2.21 [1.26-3.87]*
Oosterman et al. 2009,46 Netherlands
Fragmentation of the sleep-wake rhythm (Mental: r =-0.16, Memory: - 0.19, andExecutive: - 0.16 respectively)
Converted from r to d (d = d2r/√(1-r^2 ) ) and from d to the log(OR) (Log(OR) = dp/√3 and computed the mean of the effect sizes for each study and used the average effect size as the unit of analysis
aOR=1.84 [1.09-3.11]*
Osorio et al. 2011,56 United States
Insomnia aOR=2.39 [1.03–5.55]. Insomnia without depression aOR=3.32 [1.33–8.28].
No conversions and/or mean computation done
Insomnia, aOR=2.39 [1.03–5.55]*; Insomnia w/o depression, aOR=3.32 [1.33–8.28]*
154
Osorio et al. 2014,37 United States
In ApoE3+ subjects, significant differences were found between sleep groups for p-tau (F[df2] = 4.3, p =0.017), and t-tau (F[df2] = 3.3, p = 0.043). Additionally, among ApoE3+ subjects, the apnea and/or hypopnea with 4% O2-desaturation index was positively correlated with p-tau (r = 0.30, p = 0.023), t-tau (r = 0.31, p = 0.021), and Aβ-42 (r = 0.31, p = 0.021). In ApoE2+ subjects, the apnea and/or hypopnea with 4% O2-desaturation index was correlated with lower levels of CSF Aβ-42 (r = −0.71, p = 0.004), similarly to ApoE4+ subjects where there was also a trend toward lower CSF Aβ-42 levels.
Converted from r to d (d = d2r/√(1-r^2 ) ) and from d to the log(OR) (Log(OR) = dp/√3 and computed the mean of the effect sizes for each study population and used the average effect size as the unit of analysis
ApoE3+ subjects: aOR=3.20 [1.18-8.66]* ApoE2+ subjects: aOR= 8.17 [1.96-34.07]*
Saint Martin et al. 2012,47 France
PSQI correlated with the MMSE (r = -0.14, p = .03) Sleep Quality correlated with Delayed free recall (r = -0.14, p < .05)
Converted from r to d (d = d2r/√(1-r^2 ) ) and from d to the log(OR) (Log(OR) = dp/√3 and computed the mean of the effect sizes for each study and used the average effect size as the unit of analysis
aOR=1.65 [1.02-2.68]*
Schmidt et al. 2013,49 Germany
A significant main effect of diagnosis (F(1,62) = 8.490, p<0.01) on MCH levels was found between AD (93.76 ± 13.47 pg/mL) and HS (84.65 ± 11.40 pg/mL). MCH correlated with T-tau (r = 0.47; p<0.01) and P-tau (r = 0.404; p,0.05) in the AD but not in the HS. CSF-MCH correlated negatively with MMSE scores in the AD (r = -0.362, p<0.05) and was increased in more severely affected patients (MMSE ≤ 20) compared to HS (p<0.001) and Behavioral and Psychological Symptoms of Dementia (BPSD) positive patients compared to HS (p<0.05). In CSF-HCRT- 1, a significant main effect of sex (F(1,31) = 4.400, p<0.05) with elevated levels in females (90.93 ± 17.37 pg/mL vs. 82.73 ± 15.39 pg/mL) was found whereas diagnosis and the sex*diagnosis interaction were not significant.
Converted from r to d (d = d2r/√(1-r^2 ) ) and from d to the log(OR) (Log(OR) = dp/√3 and computed the mean of the effect sizes for each study and used the average effect size as the unit of analysis
aOR=5.1 [1.31-19.89]*
Spira et al. 2013,48 United States
Shorter sleep duration Mean cortical DVR (B = 0.08 [95% CI, 0.03-0.14]; r =.38; P = .005) Mean precuneus DVR (B = 0.11 [0.03-0.18]; r =.36; P = .007). Lower sleep quality Mean cortical DVR (B = 0.04 [95% CI, -0.01-0.09]; r =.19; P = .13) Mean Precuneus DVR (B = 0.08 [0.01-0.15]; r=.29; P = .03). WHIIRS, Women’s Health Initiative Insomnia Rating Scale Mean cortical DVR (B = 0.01 [95% CI, -0.004-0.02]; r =.16; P=.23) Mean Precuneus DVR(B = 0.01 [95% CI, -0.004-0.02] r=.18; P=.16).
Converted from β to OR Converted from r to d (d = d2r/√(1-r^2 ) ) and from d to the log(OR) (Log(OR) = dp/√3 and accounted for multiple exposure measures on the same subjects by computing the mean of the effect sizes for each study and used the average effect size as the unit of analysis
Sleep duration, aOR= 4.25 [1.52-11.88]*ˠ; Lower sleep quality, aOR=2.51 [0.96–6.56]ˠ; Insomnia, aOR=1.87 [0.73–4.80]ˠ
155
Sterniczuk et al. 2013,50 Canada
SDI aOR=1.23 [1.11-1.36] SDI aOR=1.12 [1.00-1.25]
We accounted for multiple outcome measures on the same subjects by computing the mean of the effect sizes for each study and used the average effect size as the unit of analysis
aOR=1.18 [1.06-1.31]*ˠ;
Tranah et al. 2011,57 United States
Decreased activity rhythms aOR=1.57 [1.09–2.25] Rhythm robustness aOR=1.57 [1.10–2.26] Peak Activity timing aOR=1.83 [1.29–2.61]
We accounted for multiple outcome measures on the same subjects by computing the mean of the effect sizes for each study and used the average effect size as the unit of analysis
aOR=1.66 [1.16-1.31]*ˠ;
Tworoger et al 2006,38 United States
<=5hrs of sleep aOR=2.19 (1.10-4.39) 6hrs of sleep aOR=1.01 (0.61-1.67) 8hrs of sleep aOR=1.16 (0.74-1.82) >=9hrs of sleep aOR=1.46 (0.77-2.76) TICS aOR=2.51 (1.38, 4.54) Women who regularly had difficulty falling or staying asleep TICS (aOR: 1.85, 95% CI: 1.04, 3.30) global score (aOR: 2.57, 95% CI: 1.39, 4.74) TICS (Telephone Interview of Cognitive Status)
We accounted for multiple exposure measures on the same subjects by computing the mean of the effect sizes for each exposure category and used the average effect size as the unit of analysis
≤ 6 h sleep, aOR=1.68 [0.92-3.08]ˠ; Difficulty initiating sleep, aOR=2.21 [1.22-4.01]*ˠ; ≥ 8 h sleep, a0R=1.30 [0.71-2.39]ˠ;
Virta et al 2013,58 Finland,
Sleep Length (N=2328) Sleep Length (< 7 h/day) (β = -0.84 (-1.51,-0.17), P = 0.014 Sleep Length (< 7 h/day) (β = -0.79 (-1.44,-0.14), P = 0.019 and Sleep Length (> 8 h/day) (β = -1.66 (-2.37,-0.95) P < 0.001 Sleep Length (> 8 h/day) (β = -1.61 (-2.33,-0.91) P < 0.001. Poor sleep quality (β = -1.00 (-1.77,-0.23) P = 0.011. Poor sleep quality (β = -0.89 (-1.69,-0.09) P = 0.028 The use of hypnotics ≥ 60 days per year (β = -1.92 (-3.11,-0.73), P = 0.002). The use of hypnotics ≥ 60 days per year (β = -1.58 (-2.75,-0.41), P = 0.008). Model 1 adjusted for age, sex, education, ApoE status, and follow-up. Model 2 adjusted for life satisfaction, obesity, hypertension, physical inactivity, heavy drinking, and binge drinking in addition to the factors adjusted for in Model 1
We accounted for multiple exposure measures on the same subjects by computing the mean of the effect sizes for each exposure category and used the average effect size as the unit of analysis
≤ 7 h sleep, aOR=2.26 [1.17–4.38]*ˠ; >=8 h sleep, aOR=5.13 [2.51-10.49]*ˠ; Sleep quality, aOR=3.72 [2.45-5.65]*ˠ; Hypnotic use, aOR=5.85 [1.80–19.03]*ˠ
156
Walsh et al 2014,59 United States
Women lowest vs. highest quartile CAR amplitude (group difference (d) = 30.42 sec, d = -1.01 words respectively, P < 0.05). Categorical fluency mesor (d = -0.86 words, P < 0.05). Women Categorical fluency later acrophase categorical fluency (d = -0.69 words, P < 0.05). Controlling for baseline Mini-Mental State Examination and sleep factors had negligible effects on the results.
Converted from d to the log(OR) aOR=3.43 [1.08–10.85]*
Yaffe et al. 2011,60 United States
SDB (Women) aOR=1.85 [1.11-3.08]. AHI (≥15 events/hour) aOR=1.71 [1.04 − 2.83] >7% of sleep time in apnea/hypopnea aOR=2.04 [1.10 − 3.78] respectively). Sleep Fragmentation Mild aOR=0.54 [0.29-0.98] Sleep Fragmentation High aOR=0.58 [0.32-1.07] WASO Mild aOR=1.17 [0.63-2.19] WASO High aOR=1.79 [0.97-3.29] Sleep Duration Mild aOR=0.58 [0.31-1.09] Sleep Duration High aOR=0.83 [0.46-1.51]
We accounted for multiple exposure measures on the same subjects by computing the mean of the effect sizes for each exposure category and used the average effect size as the unit of analysis
Sleep Fragmentation, aOR=0.56 [0.31-1.02]ˠ WASO, aOR=1.48 [0.80–2.74]ˠ; SDB , aOR=1.87 [1.08–23.23]*ˠ; TST, aOR=0.71 [0.37-1.35]ˠ
*: significant, ˠ: multiple exposure/outcome measure on same subject, so mean of the outcome computed. Abbreviations; a: adjusted, AD: Alzheimer's disease, APOE: apolipoprotein epsilon4, c: crude, CAR: circadian activity rhythms, CASI: cognitive abilities screening instrument, MMSE: mini mental state examination, OR: odds ratio, PSG: polysomnography, PSQI: Pittsburgh sleep quality index, RR: relative risk, SDB: sleep disordered breathing, Trails B: trail making b test, TST: total sleep time, WASO: wake after sleep onset,
157
Table 2A: Formula used for converting among effect sizes and for calculating the variance of the
effect sizes
Converting Among Effect
Sizes*
Formula Variance Formula**
Converting from d to the log odds ratio
Log(OR) = dπ
√� , where π is
the mathematical constant (approximately 3.14159)
The variance of Log(OR) would then be
��(��) = �π
�3
Converting from r to d d = d��
√���� The variance of d computed in this way (converted from r) is
� = 4�(1 − �� )�
Converting from d to r r = �
√���� , where a is a
correction factor for cases where n1 ≠ n2,
a = (��� ��)�
����
The correction factor (a) depends on the ratio of n1 to n2, rather than the absolute values of these numbers. Therefore, if n1 and n2 are not known precisely, use n1= n2, which will yield a=4.
The variance of r computed in this way (converted from d) is
� = �� �(�� + �)�
Converting from the log odds ratio to d
d = Log(OR)* √�π
� = ��(��) ∗ √�π
converting from β to OR exp(β) N/A
*Effect sizes from the reviewed studies included odds ratios (OR), hazard ratios (HR), Pearson’s correlation coefficient (r), beta estimates (β) and standardized mean differences (d). To proceed with the meta-analysis, we converted the different indices to a common index: odds ratios (meta-analyses were computed using Log(OR) but were converted back to OR for presentation). ** Reference #30 in manuscript: Borenstein Michael HLV, Higgins J. P. T., Rothstein H.R. Introduction to Meta-Analysis. : John Wiley & Sons, Ltd.; © 2009
158
Table 3A: Formula used for computing the variance of a composite or a difference
Computing the variance of a composite or
a difference*
Formula**
The variance of the sum of two correlated variables
If we know that the variance of Y1 is V1 and the variance of Y2 is V2, then Var (!� + !�) = � + � + 2�#�#� Where r is the correlation coefficient that describes the extent to which Y1 and Y2 co-vary. If Y1 and Y2 are inextricably linked (so that a change in one determines completely the change in the other), then r =1, and the variance of the sum is roughly twice the sum of the variances. At the other extreme, if Y1 and Y2 are unrelated, then r=0 and the variance is just the sum of the individual variances.
The variance of the mean of two correlated variables
Var $�� (!� + !�)% = &�
�'� Var (!� + !�) =
�( )� + � + 2�#�#� *
The variance of the mean of several correlated variables
var& �+ ∑ !-+-.� '= & �
+'� var)∑ !-+-.� �*=
& �+'� /∑ -�
+-.� + ∑ /�-0 1- #02- 30 2
The variance of the difference between two correlated variables
If we know that the variance of Y1 is V1 and the variance of Y2 is V2, then Var (!� − !�) = � + � − 2�#�#�
*Since our aim was to compute a summary effect size of the impact of sleep problems on cognitive decline and Alzheimer’s disease, we accounted for multiple exposure or outcome measures on the same subjects by computing the mean of the effect sizes for each study and used the average effect size as the unit of analysis. This influenced the calculation of the variance of the sum and mean effects of the correlated variables. For our analysis, we worked with a plausible range of correlations ranging from 0.5-1.0 and performed sensitivity analyses to find the most appropriate standard error range. ** Reference #30 in manuscript: Borenstein Michael HLV, Higgins J. P. T., Rothstein H.R. Introduction to Meta-Analysis. : John Wiley & Sons, Ltd.; © 2009
159
Figure 1A Forest plot presenting sub-group meta-analysis based on continent in which study was published for the effect of sleep problems and disorders on cognitive decline and/or Alzheimer’s disease
160
Figure 2A: Forest plot presenting sub-group meta-analysis based on day versus night time sleepiness for the effect of sleep problems and disorders on cognitive decline and/or Alzheimer’s disease
161
Figure 3A Forest plot presenting sub-group meta-analysis based on study temporality for the effect of sleep problems and disorders on cognitive decline and/or Alzheimer Disease
162
Figure 4A Forest plot presenting meta-analysis risk estimates based on omitting one study for the effect of sleep problems and disorders on cognitive decline and/or Alzheimer Disease
163
SECTION 3
OBSTRUCTIVE SLEEP APNEA IS ASSOCIATED WITH LONGITUDINAL INCREASES IN BRAIN
FLORBETAPIR PET IMAGING, CSF TAU, PTAU, AND DECREASE IN CSF AΒ42 BURDEN, IN
ELDERLY COGNITIVE NORMAL AND MCI INDIVIDUALS.
KEY POINTS
Question: Does Obstructive Sleep Apnea (OSA) affect longitudinal changes in brain amyloid deposition
and cerebrospinal fluid (CSF) Aβ42, t-tau and p-tau181 in cognitive normal (NL), mild cognitive
impairment (MCI) and Alzheimer’s Disease (AD) older adults.
Findings: OSA accelerates increases in brain amyloid deposition, CSF TAU and PTAU and decreases in
CSF biomarkers burden, over time, both in elderly NL and MCI individuals.
Meaning: Clinical interventions aimed at OSA, such as treatment with CPAP or dental appliances, in NL
and MCI patients, could possibly slow the progression of cognitive impairment to AD.
ABSTRACT
Importance: Recent studies demonstrate that OSA is associated with AD biomarkers. To evaluate
evidence for a causal association between OSA and AD, demonstrating AD-specific neuropathology
changes resulting from increased disease burden in OSA patients is vital.
Objective: To determine the effect of OSA on longitudinal changes in brain amyloid deposition and CSF
Aβ42, t-tau and p-tau181 in NL, MCI and AD older adults.
Design: Longitudinal study. Mean follow up time was 2.52 ±0.51 years.
164
Setting: Data used for this study were obtained from the Alzheimer’s disease Neuroimaging Initiative
(ADNI) database (adni.loni.usc.edu) on December 7, 2016.
Participants: 516 NL, 798 MCI and 325 AD subjects.
Exposure: Participant self-reported physician diagnosis of OSA.
Main Outcomes and Measures: Rate of change in brain Aβ42, CSF TAU, PTAU, and CSF Aβ42
burden over time. Multi-level mixed effects linear regression models with randomly varying intercepts
and slopes were constructed to test whether the rate of change in biomarker data differed between subjects
with and without OSA. The final models were adjusted for age, sex, BMI, CPAP use, APOE e4 status,
and history of chronic medical conditions.
Results: The self-reported prevalence of OSA in the 3 study cohorts was as follows: NL (6%), MCI
(13%), AD (7%). Across all groups, mean ages of OSA+ and OSA- were 72.3 ±7.1 and 73.9 ±7.3
respectively. Females were 49% in the NL group, 40% in the MCI group, and 37% in the AD group. In
NL and MCI groups, the annual rate of AD biomarker burden change over time, indicated that OSA+
subjects experienced faster increase in brain Aβ-42 (B = - .06, 95% CI, -.09, -.04 for both), CSF TAU (B
= -2.89, 95% CI, -3.51 to -2.29 and B = -1.89, 95% CI, -2.91, -.87, respectively), CSF PTAU (B = -1.21,
95% CI, -1.71, -.74 and, B = -1.48, 95% CI, -2.05, -.94, respectively), and a faster decrease in CSF Aβ-42
(B = 3.93, 95% CI, 3.56, 4.31 and B = 2.69, 95% CI, 2.02, 3.36, respectively); compared to OSA-
subjects, over the follow-up period.
Conclusion and Relevance: OSA appears to accelerate increases in brain amyloid deposition, CSF TAU
and PTAU and decreases in CSF biomarkers burden, over time, both in elderly Cognitive Normal and
MCI individuals. Clinical interventions aimed at OSA, such as treatment with CPAP or dental appliances,
in cognitive normal and MCI patients, could possibly slow the progression of cognitive impairment to AD
165
INTRODUCTION
Obstructive Sleep Apnea (OSA) and Alzheimer’s disease (AD) are both common chronic disease
conditions in older adults. OSA affects between 19% and 57% of individuals aged over 65.1-4 By 2050,
the incidence of AD in the U.S is expected to approximate a million people each year, with a total
estimated prevalence of 11 to 16 million people.5 Both OSA and AD cause significant morbidity and
mortality to those afflicted,6-8 and have very high socio-economic burden worldwide.5,8-16 It is therefore
pertinent that preventive and/or treatment efforts targeting both disease conditions are consolidated to
create a better quality of life in older adults.
OSA is characterized by intermittent partial or complete breathing cessation during sleep,
occurring from narrowing of various upper airway sites.1 OSA related hypoxemia and sleep fragmentation
have been implicated as possible links between OSA and AD, with recent studies demonstrating
associations between OSA and AD biomarkers in cognitive normal (NL) and MCI older adults.17-21
Clearly distinguishing whether OSA individuals with normal cognition or MCI are at heightened risk to
develop AD is critical to our ultimate mission of preventing AD.
AD biomarkers serve as in-vivo indicators of particular AD pathologies.22-24 They can be used in
the clinical setting to predict future clinical course of AD. 25,26 PET amyloid imaging of significant Aβ
load and significant levels of cerebrospinal fluid (CSF) Aβ42 are robust predictors of the development of
AD in at risk populations’ i.e. cognitive normal to amnestic mild cognitive impairment (aMCI) and MCI
to AD.25,26 Aβ deposition commences many years prior to AD symptoms’ onset27,28 while MRI brain
imaging show abnormalities later in the disease trajectory.29 This suggests a time-dependent risk of
developing AD from amnestic MCI. Significant increases of CSF P-tau have been demonstrated in AD
patients compared with controls.30 P-tau in CSF is a measurable core marker of AD.31 Since cross-
sectional studies demonstrate that OSA is associated with AD biomarkers, we further examined OSA’s
effect on longitudinal changes of these biomarkers, especially since they have been shown to predict time-
to-progression from MCI to AD.32 Demonstrating associations between OSA and AD-specific
166
neuropathology changes, progression or worsening consolidates the evidence for a potential causal
relationship between OSA and AD.
METHODS
Data used in the preparation of this article were obtained from the Alzheimer’s Disease
Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). The ADNI was launched in 2003 as a
public-private partnership, led by Principal Investigator Michael W. Weiner, MD. The primary goal of
ADNI has been to test whether serial magnetic resonance imaging (MRI), positron emission tomography
(PET), other biological markers, and clinical and neuropsychological assessment can be combined to
measure the progression of mild cognitive impairment (MCI) and early Alzheimer’s disease (AD). Thus
far, ADNI has recruited over 2,000 adults aged 55 to 90 years, consisting of cognitively normal (NL)
older individuals, people with early or late MCI, and people with early AD-dementia. Follow-up is at 6
month intervals with duration ranging from 2 to 3 years as specified in the protocols for ADNI-1, ADNI-
2, and ADNI-GO.
Study Participants
Participant data used for this study were based on medical history from baseline and follow-up
visits obtained from downloaded ADNI data on December 7, 2016. Study participants included 1,639
subjects (516 Cognitive Normal (NL), 798 MCI and 325 AD). Subjects missing important study covariates
were excluded in each of the study groups. Subjects having co-morbid sleep disorders, body mass index
(BMI) change greater than 5 between any follow-up visits, were further excluded, since disturbed sleep,
and BMI are known independent risk factor for cognitive decline and AD, and being overweight is
associated with SDB and AD.33 Other exclusions included subjects who indicated they have had previous
OSA surgery, and those who had a reversible diagnosis (i.e. they had an MCI or AD diagnosis at any time
point but an NL diagnosis or MCI diagnosis at their last visit, thereby allowing us to control for unspecified
diagnoses.
167
OSA diagnosis
Presence or absence of OSA was self-reported (variable name: MHDESC). Patients with reported
“sleep apnea” or “OSA” were labeled OSA+ and the remaining participants were considered OSA−. To
ensure that patients were allocated into the correct groups, three physicians (R.O, S.A and O.B.) reviewed
medical history descriptions in the ADNI download, for OSA+ and OSA-.
NL, MCI and AD Diagnosis
ADNI criteria for subject classification is described elsewhere.34 In summary, NL subjects had no
memory complaints while MCI and AD subjects had memory complaints. NL and MCI scored between
24 – 30 on the Mini-Mental State Examination (MMSE) while AD subjects scored between 20 – 26. NL,
MCI and AD subjects had a Clinical Dementia Rating (CDR) score of zero, 0.5 with a mandatory
requirement of the memory box score being 0.5 or greater, and 0.5 or 1, respectively. In addition, MCI
subjects had to have largely intact general cognition and functional performance, and could not qualify for
dementia diagnosis.35 The diagnosis of AD was made using established clinical criteria.36 Patient
diagnosis was recorded at 6-month intervals for 24 months. Subjects were classified as converters if they
converted to AD between 12 months and 24 months and as stable if they did not convert by 24 months.
Florbetapir-PET Imaging Acquisition and Interpretation
ADNI Florbetapir summary data are updated regularly, and uploaded to LONI by the University
of California at Berkeley group.37,38 More information on the methods can be found at:
https://adni.loni.usc.edu/wp-content/uploads/2010/05/ADNI2_PET_Tech_Manual_0142011.pdf,
http://adni.loni.usc.edu/updated-florbetapir-av-45-pet-analysis-results/.
Cerebrospinal Fluid Methods (CSF)
CSF bio-specimen data collection are explained in detail at http://adni.loni.usc.edu/data-
samples/biospecimen-data/ and (http://www.adni-info.org/index.php).
168
Covariates/potential confounders
Covariates/potential confounders in our study include age, sex, educational status, smoking
status, body mass index (BMI), APOE4 status, hypertension, diabetes and history of cardiovascular
disease.
Data analysis
Given that ADNI data is unbalanced with unequal numbers of measurement for each study
participant, we used multi-level mixed effects linear regression models with normal errors39-41 for our
longitudinal analyses to examine the relationship between OSA and CSF P-tau, and Aβ-42 volumes, as
they provide a flexible and valuable tool for analyzing such unbalanced longitudinal data. Our method
allowed for incorporation of all the available information in the data, and possibly reduced or even
eliminated any bias resulting from an analysis confined to the complete cases.42
Specifically, to determine the specific type of modeling, we examined profile plots for the
biomarker data levels over time, by OSA status (=1 if OSA, =0 if non-OSA) using both the original data
and within-subject residuals. We also examined trajectory plots obtained by subtracting the baseline
measurement from the original measurements (i.e. dij = Yij − Yi1), and examined plots of mean and
variance of the biomarker data at each time point by OSA status. Further, using the slopes as summary
statistics, we conducted different formal tests to compare the two OSA statuses. Based on the above, our
assumption was that the rate of change of biomarker data is approximately linear over time, so we
proposed a parametric model with time as a continuous variable. Using PROC MIXED in SAS software;
we fit the models with randomly varying intercepts and slopes and allowed them to depend on exposure
group (i.e., OSA status). Based on the findings from above, we fit a random coefficients model for the
relationship between OSA status and time using unstructured covariance model. For finding a
parsimonious covariance structure, we fit as many different covariance models and conducted a likelihood
ratio test of each nested model using information criteria for non-nested models. Finally, using the
preferred covariance structure obtained above, we used PROC MIXED to construct a test of whether the
169
rate of change in biomarker data differed between the two OSA groups. The final models were adjusted
for age, sex, BMI, CPAP use, APOE e4 status, and history of chronic medical conditions.
These analyses allowed us to examine whether there was significant variation between OSA and
non-OSA subjects in mean AD biomarker level at baseline (intercept), as well as whether significant
variation in the change in AD biomarker level over time (slope) occurred. Furthermore, the covariance
between the baseline AD biomarker level (intercept) and AD biomarker change over time indicated
whether OSA+ or OSA- subjects had experienced a faster increase or decrease in AD biomarker level
over time (significant slope). It also allowed for assessment of significant differences in the rate of change
in AD biomarker level between OSA+ and OSA- patient groups over time. Statistical analyses was
performed using SAS software (version 9.4; SAS Institute Inc., Cary, NC). For sensitivity analyses, we
attempted using the dichotomized brain Aβ level outcome (Aβ+ vs. Aβ-) and examined the rate of change
in AD biomarker level in these populations (SDB+/ Aβ+ vs. SDB-/ Aβ+; SDB+/ Aβ- vs. SDB-/ Aβ-; and
SDB+/ Aβ+ vs. SDB+/ Aβ-; SDB-/ Aβ - vs. SDB-/ Aβ+). However, power issues limited our analyses;
moreover, convergence issues did not permit any estimation with these categorical variables.
RESULTS
Demographic and Clinical Characteristics
Table 3.1 shows the demographic and clinical characteristics of study participants. Overall, the
mean ages of OSA+ and OSA- were 72.3 ±7.1 and 73.9 ±7.3 respectively. Forty-nine percent were female
in the NL group, while in the MCI and AD groups, females consisted of 40% and 37% of the participants
respectively. In the NL group, 6% were OSA+, while 13% of the MCI group and 7% of the AD group
were OSA+. Participants differed markedly in APOE4 status with 28%, 51% and 66% being APOE
positive in the NL, MCI and AD groups respectively. Participants were similar in age, educational status,
BMI, and chronic medical conditions across groups (i.e. NL, MCI and AD), however, within group
170
differences (i.e. OSA+ vs. OSA-) existed on these characteristics. Mean follow up time for all participants
was 2.52 ±0.51 years.
Baseline AD Biomarker levels by clinical group and OSA status
Table 3.1 and Figure 3.1 show baseline CSF Aβ-42, TAU, PTAU and Brain Florbetapir PET
imaging levels by NL, MCI and AD group and OSA status without controlling for confounders. There
were significant differences in CSF Aβ-42 levels for both MCI (P =0.04), and AD (P <0.01) groups
respectively, with OSA+ participants having significantly higher levels at baseline. No significant
difference was seen in NL group by OSA status. Significant difference in TAU levels was seen for the
MCI patients (P =0.02), with OSA+ individuals having significantly lower levels. No significant
difference was seen in both the NL and AD groups. For PTAU levels, no significant difference was seen
across all groups. For Brain Florbetapir PET imaging levels, at baseline, there was significant differences
between OSA groups for both NL and MCI participants (P =0.02 for all), with OSA+ participants having
significantly lower brain Florbetapir levels. No difference was seen in AD group. For the MCI group, the
inclusion of an interaction term OSA*APOE4 was also significant (P =0.01).
Rate of Change in AD Biomarker by OSA status
Table 3.2 reports the covariance parameter estimates and between subject (i.e. OSA+ vs. OSA-)
variation in AD biomarker levels over time.
Brain Aβ-42 levels
In NL participants, there was significant variation by OSA status in the annual change in brain
Florbetapir PET volumes over time (slope) (mean SUVR; B = 0.06, p < .0001). The covariance between
the baseline brain Florbetapir level and the annual brain Florbetapir volume change over time was -.06,
indicating that OSA+ subjects experienced faster increase in brain Aβ-42 deposition over time (p < .0001)
when compared to OSA- subjects. In MCI participants, there was also significant variation by OSA status
in the annual change in brain Florbetapir volumes over time (slope) (mean SUVR; B = 0.08, p < .0001).
171
The covariance between the baseline brain Aβ-42 level and the annual brain Aβ-42 volume change over
time was -.006, indicating that OSA+ subjects experienced faster increase in brain Aβ-42 deposition over
time (p < .0001) when compared to OSA- subjects. For the MCI group, the inclusion of an interaction
term OSA*APOE4 was also significant (P =0.01) with APOE4+/OSA+ individuals showing a faster
increase in brain Aβ-42 compared to APOE4-/OSA- patients. No significant variation in the change in
brain Florbetapir PET volumes by OSA status over time was seen for the AD group.
CSF Aβ-42, TAU and PTAU levels
For the NL and MCI groups, there was significant variation in the annual change in CSF Aβ-42
by OSA status (mean SUVR; B = -2.708, mean SUVR; B = -2.264, p < .0001 for all respectively) over
time. Significant differences in the annual change in TAU and PTAU levels over time (slope) were also
seen in both groups. Covariance parameters between the baseline CSF Aβ-42, TAU and PTAU volume
change over time indicated that OSA+ subjects experienced a faster decrease in CSF Aβ-42 and increases
in TAU and PTAU volumes over time (p < .0001 for all) compared to OSA- participants, in both the CN
and MCI groups (Table 3.2). No significant variations in the change in CSF Aβ-42, TAU and PTAU over
time was seen for the AD group.
DISCUSSION
Major findings from this study include the following:
i. At baseline, OSA+ patients had significantly higher CSF Aβ-42 in the MCI group; and lower
brain Florbetapir PET in NL and MCI groups; and lower TAU levels in the MCI group,
compared to OSA- patients.
ii. Longitudinally, there was significant differences in the annual rate of change in brain Florbetapir
PET, CSF Aβ-42, TAU and PTAU levels over the follow up period for the NL and MCI group
iii. OSA+ subjects experienced significantly faster increase in brain Aβ-42, TAU and PTAU levels
and a faster decrease in CSF Aβ-42 over the follow-up period in both the NL and MCI groups.
172
iv. For the MCI group, APOE4+/OSA+ individuals showed a faster increase in brain Aβ-42
compared to APOE4-/OSA- patients
Baseline AD Biomarker levels by clinical group and OSA status
At baseline, the data showed that OSA+ patients had significantly higher CSF Aβ-42 and lower
brain Aβ-42 in NL and MCI groups, and lower TAU levels in the MCI group, compared to OSA- patients.
Theoretically, this may be attributed to slow wave sleep (SWS) occurring during non-rapid eye movement
(NREM) sleep, a known Aβ modulator.43-45 Sleep is involved in the clearance of Aβ,45 and fragmented
SWS limits brain Aβ clearance, thereby leaving higher levels that reflect in the CSF.46,47 At cross-section,
disrupted SWS has been shown to be associated with higher CSF Aβ-42 in middle-aged48,49 and older
adults.50 It is also important to note that CSF tau protein pathology and CSF Aβ-42 burden may not be
specific to AD in these participants.51 The effect of normal aging in NL subjects on CSF turnover and
brain Aβ-42 clearance is still not fully understood.52,53 Furthermore, cross-sectional data show little
evidence for an association between age and CSF Aβ-42.54,55 It is also noteworthy that the functional
significance of high CSF Aβ42 levels is not fully known.
Rate of Change in AD Biomarker by OSA status
The major objective of this study was to examine the effect of OSA on longitudinal changes in
brain amyloid deposition, and AD CSF biomarkers. There were significant differences in the annual rate
of change in brain Florbetapir PET, CSF Aβ-42, TAU and PTAU levels over the follow-up period for the
NL and MCI group, with OSA+ subjects experiencing significantly faster increase in brain Florbetapir
PET, TAU and PTAU levels and a faster decrease in CSF Aβ-42. The direction of these longitudinal
changes are in cognizance with emanating evidence-linking disturbed sleep, OSA with AD pathogenesis.
Self-reported sleep duration and quality has been shown to be associated with higher brain Aβ-42
burden.19 MCI individuals with higher apnea-hypopnea index and oxygen desaturation index
demonstrated higher brain Aβ-42 burden globally and regionally in the precuneus.20 A recent study by
Yun et al56 examined whether OSA increased brain amyloid burden relative to controls in middle-aged
173
participants from the Korean Genome and Epidemiology Study. After adjusting for potential confounders,
OSA patients had a higher brain amyloid burden in the right posterior cingulate gyrus and right temporal
cortex, relative to controls, suggesting that OSA may have been responsible for these findings. Bu et al.17
examined whether hypoxia indices in OSA was associated with serum amyloid (Aβ40, Aβ42), total tau
and phosphorylated tau 181 (P-tau 181) comparing their levels with controls. Results from this study
demonstrated significantly higher levels of serum Aβ40, Aβ42, total Aβ and P-tau 181 in OSA patients
compared to control. Hypoxia indices including apnea-hypopnea index, the oxygen desaturation index,
and the mean and lowest oxyhemoglobin saturations correlated positively with Aβ40, Aβ42, and total Aβ
in OSA patients, suggesting that hypoxia possibly effects AD pathogenesis. In a population of MCI
individuals, Liguori et al57 compared CSF β-amyloid42, tau proteins, and lactate levels in OSA versus
CPAP treated OSA and controls. Findings from the study demonstrated lower CSF β-amyloid42, higher
lactate levels and t-tau/ Aβ42 ratio compared to controls and CPAP treated OSA patients. Controls and
CPAP treated OSA MCI patients had similar AD biomarker levels. These findings suggest that OSA may
effect early AD biomarker changes that may be susceptible to CPAP treatment.
In our study, among MCI patients, APOE4+/OSA+ individuals showed a faster increase in brain
Aβ-42 compared to APOE4-/OSA- patients, suggesting that the APOE4 allele may be a moderator of the
association between OSA and AD. In a population of cognitive normal (NL) older adults, Osorio et al18
demonstrated an association between intermittent hypoxia and increases in CSF T-Tau, P-Tau and Aβ42
in ApoE3+, suggesting that hypoxia may be responsible for changes in the AD biomarkers in this
population. In ApoE4+ subjects, OSA severity tended towards lower CSF Aβ42 levels. There was no
difference in the rate of biomarker change in the AD group by OSA status. This may be because in AD
patients, risk profile is possibly linear with brain atrophy until it reaches a threshold thereby exhibiting a
ceiling effect at higher levels of brain Aβ.32 OSA’s effect as it relates to AD biomarker changes may
therefore be attenuated or have reached its maximum.
174
Strengths and Limitations
Our study possesses several strengths including a well-defined cohort, and objective assessment
of CSF P-Tau, Hippocampal atrophy and β-amyloid burden, which allowed for a high degree of certainty
regarding measurement of our outcome.58 Furthermore, our statistical analytic methods were robust with
respect to unbalanced number of observations per subject over time. However, we were limited by our
measurement of OSA since this was self-reported. It is known that self-reported sleep measures can be
impacted by diminished cognition59 and in certain situations might not be correlated with objective
measurements using polysomnography or wrist actigraphy.60 Moreover any positive findings from our
study will need to be replicated using objective measures of sleep function. The ADNI racial demographic
is largely white; as such this limits the generalizability of our findings to other racial populations.
CONCLUSIONS
OSA appears to accelerate increases in brain amyloid deposition, CSF TAU and PTAU and
decreases in CSF biomarkers burden, over time, both in elderly Cognitive Normal and MCI individuals.
Sleep fragmentation and/or intermittent hypoxia from OSA are likely candidate mechanisms. Thus,
clinical interventions aimed at OSA, such as treatment with CPAP or dental appliances, in cognitive
normal and MCI patients, could possibly mitigate or slow the progression of cognitive impairment to AD.
Further research examining mechanisms underlying these observed effects are needed.
References
1. Aloia MS, Arnedt JT, Davis JD, Riggs RL, Byrd D. Neuropsychological sequelae of obstructive
sleep apnea-hypopnea syndrome: a critical review. Journal of the International
Neuropsychological Society : JINS. 2004;10(5):772-785.
2. Castronovo V, Canessa N, Strambi LF, et al. Brain activation changes before and after PAP
treatment in obstructive sleep apnea. Sleep. 2009;32(9):1161-1172.
175
3. Ju G, Yoon IY, Lee SD, Kim TH, Choe JY, Kim KW. Effects of sleep apnea syndrome on
delayed memory and executive function in elderly adults. Journal of the American Geriatrics
Society. 2012;60(6):1099-1103.
4. Young T, Peppard PE, Gottlieb DJ. Epidemiology of obstructive sleep apnea: a population health
perspective. American journal of respiratory and critical care medicine. 2002;165(9):1217-1239.
5. 2015 Alzheimer's disease facts and figures. Alzheimer's & dementia : the journal of the
Alzheimer's Association. 2015;11(3):332-384.
6. Hebert LE, Beckett LA, Scherr PA, Evans DA. Annual incidence of Alzheimer disease in the
United States projected to the years 2000 through 2050. Alzheimer Dis Assoc Disord.
2001;15(4):169-173.
7. Hebert LE, Weuve J, Scherr PA, Evans DA. Alzheimer disease in the United States (2010-2050)
estimated using the 2010 census. Neurology. 2013;80(19):1778-1783.
8. Heinzer R, Marti-Soler H, Haba-Rubio J. Prevalence of sleep apnoea syndrome in the middle to
old age general population. The Lancet Respiratory medicine. 2016;4(2):e5-6.
9. Chiatti C, Furneri G, Rimland JM, et al. The economic impact of moderate stage Alzheimer's
disease in Italy: evidence from the UP-TECH randomized trial. International psychogeriatrics /
IPA. 2015;27(9):1563-1572.
10. Choo WY, Low WY, Karina R, Poi PJ, Ebenezer E, Prince MJ. Social support and burden among
caregivers of patients with dementia in Malaysia. Asia-Pacific journal of public health.
2003;15(1):23-29.
11. Frahm-Falkenberg S, Ibsen R, Kjellberg J, Jennum P. Health, social and economic consequences
of dementias: a comparative national cohort study. Eur J Neurol. 2016;23(9):1400-1407.
12. Gerves C, Bellanger MM, Ankri J. Economic analysis of the intangible impacts of informal care
for people with Alzheimer's disease and other mental disorders. Value in health : the journal of
the International Society for Pharmacoeconomics and Outcomes Research. 2013;16(5):745-754.
176
13. Gibson GJ. Obstructive sleep apnoea syndrome: underestimated and undertreated. British medical
bulletin. 2004;72:49-65.
14. Gray A, Fenn P. Alzheimer's disease: the burden of the illness in England. Health trends.
1993;25(1):31-37.
15. Hossain JL, Shapiro CM. The prevalence, cost implications, and management of sleep disorders:
an overview. Sleep Breath. 2002;6(2):85-102.
16. Lopez-Bastida J, Serrano-Aguilar P, Perestelo-Perez L, Oliva-Moreno J. Social-economic costs
and quality of life of Alzheimer disease in the Canary Islands, Spain. Neurology.
2006;67(12):2186-2191.
17. Bu XL, Liu YH, Wang QH, et al. Serum amyloid-beta levels are increased in patients with
obstructive sleep apnea syndrome. Scientific reports. 2015;5:13917.
18. Osorio RS, Ayappa I, Mantua J, et al. The interaction between sleep-disordered breathing and
apolipoprotein E genotype on cerebrospinal fluid biomarkers for Alzheimer's disease in
cognitively normal elderly individuals. Neurobiol Aging. 2014;35(6):1318-1324.
19. Spira AP, Gamaldo AA, An Y, et al. Self-reported Sleep and beta-Amyloid Deposition in
Community-Dwelling Older Adults. JAMA Neurol. 2013.
20. Spira AP, Yager C, Brandt J, et al. Objectively Measured Sleep and beta-amyloid Burden in
Older Adults: A Pilot Study. SAGE open medicine. 2014;2.
21. Yun CH, Lee HY, Lee SK, et al. Amyloid Burden in Obstructive Sleep Apnea. Journal of
Alzheimer's disease : JAD. 2017.
22. Hampel H, Burger K, Teipel SJ, Bokde AL, Zetterberg H, Blennow K. Core candidate
neurochemical and imaging biomarkers of Alzheimer's disease. Alzheimers Dement.
2008;4(1):38-48.
23. Hampel H, Teipel SJ, Fuchsberger T, et al. Value of CSF beta-amyloid1-42 and tau as predictors
of Alzheimer's disease in patients with mild cognitive impairment. Mol Psychiatry.
2004;9(7):705-710.
177
24. Shaw LM, Vanderstichele H, Knapik-Czajka M, et al. Cerebrospinal fluid biomarker signature in
Alzheimer's disease neuroimaging initiative subjects. Annals of neurology. 2009;65(4):403-413.
25. Brys M, Pirraglia E, Rich K, et al. Prediction and longitudinal study of CSF biomarkers in mild
cognitive impairment. Neurobiol Aging. 2009;30(5):682-690.
26. Waragai M, Okamura N, Furukawa K, et al. Comparison study of amyloid PET and voxel-based
morphometry analysis in mild cognitive impairment and Alzheimer's disease. J Neurol Sci.
2009;285(1-2):100-108.
27. Bourgeat P, Chetelat G, Villemagne VL, et al. Beta-amyloid burden in the temporal neocortex is
related to hippocampal atrophy in elderly subjects without dementia. Neurology. 2010;74(2):121-
127.
28. Bouwman FH, Schoonenboom NS, Verwey NA, et al. CSF biomarker levels in early and late
onset Alzheimer's disease. Neurobiol Aging. 2009;30(12):1895-1901.
29. Carlson NE, Moore MM, Dame A, et al. Trajectories of brain loss in aging and the development
of cognitive impairment. Neurology. 2008;70(11):828-833.
30. Buerger K HH. Evaluation of phosporylated tau protein as a core biomarker of Alzheimer’s
disease. . In: Iqbal I, Winblad B, editors Proceedings of the 9th International Conference on
Alzheimer’s Disease and Related Disorders Philadelphia, Pennsylvania: Alzheimer’s
Association; 2004.p. 46–58.
31. Frank RA, Galasko D, Hampel H, et al. Biological markers for therapeutic trials in Alzheimer's
disease. Proceedings of the biological markers working group; NIA initiative on neuroimaging in
Alzheimer's disease. Neurobiol Aging. 2003;24(4):521-536.
32. Jack CR, Jr., Wiste HJ, Vemuri P, et al. Brain beta-amyloid measures and magnetic resonance
imaging atrophy both predict time-to-progression from mild cognitive impairment to Alzheimer's
disease. Brain : a journal of neurology. 2010;133(11):3336-3348.
33. Fitzpatrick AL, Kuller LH, Lopez OL, et al. Midlife and late-life obesity and the risk of dementia:
cardiovascular health study. Arch Neurol. 2009;66(3):336-342.
178
34. Petersen RC, Aisen PS, Beckett LA, et al. Alzheimer's Disease Neuroimaging Initiative (ADNI):
clinical characterization. Neurology. 2010;74(3):201-209.
35. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive
impairment: clinical characterization and outcome. Arch Neurol. 1999;56(3):303-308.
36. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of
Alzheimer's disease: report of the NINCDS-ADRDA Work Group under the auspices of
Department of Health and Human Services Task Force on Alzheimer's Disease. Neurology.
1984;34(7):939-944.
37. Landau SM, Breault C, Joshi AD, et al. Amyloid-beta imaging with Pittsburgh compound B and
florbetapir: comparing radiotracers and quantification methods. Journal of nuclear medicine :
official publication, Society of Nuclear Medicine. 2013;54(1):70-77.
38. Landau SM, Lu M, Joshi AD, et al. Comparing positron emission tomography imaging and
cerebrospinal fluid measurements of beta-amyloid. Annals of neurology. 2013;74(6):826-836.
39. Harville DA. Maximum Likelihood Approaches to Variance Component Estimation and to
Related Problems. . Journal of the American Statistical Association. 1977;72((358):):p. 339-340.
40. Jennrich RI, Schluchter MD. Unbalanced repeated-measures models with structured covariance
matrices. Biometrics. 1986;42(4):805-820.
41. Laird NM, Ware JH. Random-effects models for longitudinal data. Biometrics. 1982;38(4):963-
974.
42. Ten Have TR, Kunselman AR, Pulkstenis EP, Landis JR. Mixed effects logistic regression
models for longitudinal binary response data with informative drop-out. Biometrics.
1998;54(1):367-383.
43. Cirrito JR, Yamada KA, Finn MB, et al. Synaptic activity regulates interstitial fluid amyloid-beta
levels in vivo. Neuron. 2005;48(6):913-922.
44. Nir Y, Staba RJ, Andrillon T, et al. Regional slow waves and spindles in human sleep. Neuron.
2011;70(1):153-169.
179
45. Xie L, Kang H, Xu Q, et al. Sleep drives metabolite clearance from the adult brain. Science (New
York, NY). 2013;342(6156):373-377.
46. Hosselet J, Ayappa I, Norman RG, Krieger AC, Rapoport DM. Classification of sleep-disordered
breathing. American journal of respiratory and critical care medicine. 2001;163(2):398-405.
47. Hosselet JJ, Norman RG, Ayappa I, Rapoport DM. Detection of flow limitation with a nasal
cannula/pressure transducer system. American journal of respiratory and critical care medicine.
1998;157(5 Pt 1):1461-1467.
48. Ju YE, Finn MB, Sutphen CL, et al. Obstructive sleep apnea decreases central nervous system-
derived proteins in the cerebrospinal fluid. Annals of neurology. 2016;80(1):154-159.
49. Ju YS, Ooms SJ, Sutphen C, et al. Slow wave sleep disruption increases cerebrospinal fluid
amyloid-beta levels. Brain : a journal of neurology. 2017;140(8):2104-2111.
50. Varga AW, Wohlleber ME, Gimenez S, et al. Reduced Slow-Wave Sleep Is Associated with High
Cerebrospinal Fluid Abeta42 Levels in Cognitively Normal Elderly. Sleep. 2016;39(11):2041-
2048.
51. de Leon MJ, DeSanti S, Zinkowski R, et al. MRI and CSF studies in the early diagnosis of
Alzheimer's disease. J Intern Med. 2004;256(3):205-223.
52. Bading JR, Yamada S, Mackic JB, et al. Brain clearance of Alzheimer's amyloid-beta40 in the
squirrel monkey: a SPECT study in a primate model of cerebral amyloid angiopathy. J Drug
Target. 2002;10(4):359-368.
53. Silverberg GD, Levinthal E, Sullivan EV, et al. Assessment of low-flow CSF drainage as a
treatment for AD: results of a randomized pilot study. Neurology. 2002;59(8):1139-1145.
54. Andreasen N, Minthon L, Davidsson P, et al. Evaluation of CSF-tau and CSF-Abeta42 as
diagnostic markers for Alzheimer disease in clinical practice. Arch Neurol. 2001;58(3):373-379.
55. Hulstaert F, Blennow K, Ivanoiu A, et al. Improved discrimination of AD patients using beta-
amyloid(1-42) and tau levels in CSF. Neurology. 1999;52(8):1555-1562.
180
56. Yun CH, Lee HY, Lee SK, et al. Amyloid Burden in Obstructive Sleep Apnea. Journal of
Alzheimer's disease : JAD. 2017;59(1):21-29.
57. Liguori C, Mercuri NB, Izzi F, et al. Obstructive Sleep Apnea is Associated With Early but
Possibly Modifiable Alzheimer's Disease Biomarkers Changes. Sleep. 2017;40(5).
58. Johnson KA, Fox NC, Sperling RA, Klunk WE. Brain imaging in Alzheimer disease. Cold Spring
Harb Perspect Med. 2012;2(4):a006213.
59. Van Den Berg JF, Van Rooij FJ, Vos H, et al. Disagreement between subjective and actigraphic
measures of sleep duration in a population-based study of elderly persons. J Sleep Res.
2008;17(3):295-302.
60. Unruh ML, Redline S, An MW, et al. Subjective and objective sleep quality and aging in the
sleep heart health study. J Am Geriatr Soc. 2008;56(7):1218-1227.
Probably need to add Figure Legends here.
181
Table 3.1: Descriptive Characteristics of Participants by Obstructive Sleep Apnea Status at Baseline
Cognitive Normal
Characteristics All OSA- OSA+
Number of participants (%) 516 (100) 487 (94) 29 (6) Female gender, number (%) 253 (49) 241 (49) 9 (32) Age, years, median (interquartile range) 74 (71, 78) 71 (70, 78) 71 (70, 76)
APOE positive, number (%) 145 (28) 140 (29) 5 (17) Education, years, median (interquartile range) 16 (14, 18) 16 (14, 18) 16 (15, 18) BMI (kg/m2) 27.2 ± 4.8 27.1 ± 4.7 29.8 ± 5.9 Hypertension, number (%) 251 (49) 235 (48) 16 (55) Diabetes, number (%) 45 (9) 40 (8) 5 (17) Thyroid Disease, number (%) 112 (22) 109 (22) 3 (10)
Respiratory Disease, number (%) 123 (24) 100 (21) 23 (79)
CSF-ABETA pg/ml median (interquartile range) 210 (155,
241) 209 (155,
241) 226 (199, 259)
TAU pg/ml median (interquartile range) 59 (45, 84) 59 (45 83) 56 (48, 84) PTAU pg/ml median (interquartile range) 27 (20, 40) 27 (20, 42) 27 (20, 32) Aβ, median (interquartile range) 1.1 (1.0, 1.2) 1.1 (1.0, 1.2) 1.0 (1.0, 1.0)
Mild Cognitive Impairment
Characteristics All OSA- OSA+
Number of participants (%) 798 (100) 695 (87) 103 (13) Female gender, number (%) 319 (40) 297 (43) 25 (25) Age, years, median (interquartile range) 74 (68, 79) 71 (70, 78) 71 (70, 76) APOE positive, number (%) 410 (51) 368 (53) 42 (41)
Education, years, median (interquartile range) 16 (14, 18) 16 (14, 18) 16 (14, 18) BMI (kg/m2) 26.9 ± 4.6 26.5 ± 4.4 29.4 ± 5.3 Hypertension, number (%) 395 (49) 337 (48) 58 (56) Diabetes, number (%) 75 (9) 57 (8) 18 (17) Thyroid Disease, number (%) 156 (20) 133 (19) 23 (22) Respiratory Disease, number (%) 195 (24) 133 (19) 62 (60)
CSF-ABETA pg/ml median (interquartile range) 153 (130,
209) 150 (128,
206) 169 (139, 214)
TAU pg/ml median (interquartile range) 80 (54, 116) 81 (54, 122) 67 (51, 97) PTAU pg/ml median (interquartile range) 36 (23, 51) 37 (23, 52) 32 (21, 44)
Aβ, median (interquartile range) 1.2 (1.0, 1.4) 1.2 (1.0, 1.4) 1.1 (1.0, 1.3)
Alzheimer's Disease
Characteristics All OSA- OSA+
Number of participants (%) 325 (100) 303 (93) 22 (7) Female gender, number (%) 119 (37) 113 (37) 6 (27)
Age, years, median (interquartile range) 76 (71, 80) 76 (71, 80) 71 (64, 76)
182
APOE positive, number (%) 216 (66) 198 (65) 18 (82) Education, years, median (interquartile range) 16 (14, 18) 16 (14, 18) 16 (14, 18) BMI (kg/m2) 25.9 ± 4.6 25.6 ± 4.3 29.0 ± 6.8 Hypertension, number (%) 165 (51) 149 (49) 16 (73) Diabetes, number (%) 32 (10) 30 (10) 2 (9)
Thyroid Disease, number (%) 65 (20) 61 (20) 4 (18) Respiratory Disease, number (%) 67 (21) 52 (17) 15 (68)
CSF-ABETA pg/ml median (interquartile range) 132 (116,
151) 115 (96,
182) 115 (79, 151)
TAU pg/ml median (interquartile range) 80 (54, 116) 67 (51, 97) 81 (54, 122) PTAU pg/ml median (interquartile range) 42 (33, 61) 41 (33, 61) 54 (35, 66)
Aβ, median (interquartile range) 1.4 (1.3, 1.5) 1.4 (1.3, 1.5) 1.4 (1.3, 1.6) Abbreviation: Aβ: amyloid beta, APOE: Apolipoprotein epsilon, BMI: body mass index, CSF: cerebrospinal fluid, TAU: tau protein, PTAU: phosphorylated tau
Table 3.2 Covariance Parameter estimates and Between Subject variation in AD Biomarker Deposition
Parameters Estimate 95% CI P-value
OSA+ vs. OSA- (Cognitive Normal Patients)
β-Amyloid Burden over time (slope) 0.06 .02, .11 <.0001
β-Amyloid Burden over time (covariance) -0.06 -.09, -.04 <.0001
CSF Aβ-42 volume over time (slope) -2.71 -3.11, -2.35 <.0001
CSF Aβ-42 volume over time (covariance) 3.93 3.56 - 4.31 <.0001
CSF TAU volume over time (slope) 3.68 3.31 - 4.07 <.0001
CSF TAU volume over time (covariance) -2.89 -3.51, -2.29 <.0001
CSF PTAU volume over time (slope) 1.22 1.02 - 1.42 <.0001
CSF PTAU volume over time (covariance) -1.21 -1.71, - .74 <.0001
OSA+ vs. OSA- (Mild Cognitive Impairment Patients)
β-Amyloid Burden over time (slope) 0.08 .05, .12 <.0001
β-Amyloid Burden over time (covariance) -0.06 -.09, -.04 <.0001
CSF Aβ-42 volume over time (slope) -2.62 -3.23, -2.03 <.0001
CSF Aβ-42 volume over time (covariance) 2.69 2.02, 3.36 <.0001
CSF TAU volume over time (slope) 2.21 1.58, 2.86 <.0001
CSF TAU volume over time (covariance) -1.89 -2.91, -.87 <.0001
CSF PTAU volume over time (slope) 1.74 1.22, 2.27 <.0001
CSF PTAU volume over time (covariance) -1.48 -2.05, -.94 <.0001
OSA+ vs. OSA- (Alzheimer's disease Patients)
β-Amyloid Burden over time (slope) 0.07 -1.19, 1.33 0.33
β-Amyloid Burden over time (covariance) -0.29 -2.07, 1.49 0.31
183
CSF Aβ-42 volume over time (slope) -1.11 -3,31, 1.09 0.53
CSF Aβ-42 volume over time (covariance) -1.14 -3.38, 1.63 0.56
CSF TAU volume over time (slope) 0.26 -1.02, 1,28 0.47
CSF TAU volume over time (covariance) -0.15 -1.94, 1.64 0.47
CSF PTAU volume over time (slope) 0.94 0.23, 1.65 0.11
CSF PTAU volume over time (covariance) -0.16 -1.66, 1.34 0.11 Abbreviation: Aβ: amyloid beta, APOE: Apolipoprotein epsilon, BMI: body mass index, CSF: cerebrospinal fluid, TAU: tau protein, PTAU: phosphorylated tau
Models were adjusted for age, sex, BMI, CPAP use, APOE e4 status, and history of chronic medical
conditions.
184
Figure 3.1: Baseline Alzheimer Disease Biomarker burden and Brain
Aβ-42 levels in Cognitive Normal subjects by OSA status
185
Figure 3.2: Baseline Alzheimer Disease Biomarker burden and Brain
Aβ-42 levels in Mild Cognitive Impairment subjects by OSA status
186
Figure 3.3: Baseline Alzheimer Disease Biomarker burden and Brain
Aβ-42 levels in Alzheimer disease subjects by OSA status
187
SECTION 4
Obstructive Sleep Apnea: A Distinct Physiological Phenotypic Risk Factor in older adults with Cognitive
decline and Alzheimer’s disease
ABSTRACT
Introduction: Studies suggest that Obstructive Sleep Apnea (OSA) in the elderly may result in varying
functional outcomes relative to OSA in the middle-aged. Therefore, understanding and appreciating the
heterogeneity of OSA and its outcomes in distinct age groups especially at it relates to cognition,
subsequent cognitive decline and Alzheimer disease (AD), is critical in mitigating the deleterious effects
of OSA.
Methods: In this review, we integrated over 3 decades of research examining OSA and cognition; OSA
and subsequent cognitive decline; and OSA and AD, with particular focus in appreciating the
heterogeneity of OSA and its outcomes in distinct age groups. A systematic literature search of
bibliographic databases including PubMed/Medline, Embase, Psych INFO and Cochrane library for
clinical trials, was conducted to identify all eligible studies (published from their onset up until August
31, 2017) that examined associations between OSA and cognitive function, OSA and subsequent
cognitive decline, and OSA and AD
Results: Thirty-four studies examining the association between OSA and cognition, 7 studies examining
the association between OSA and subsequent cognitive decline, and 15 studies examining the association
of OSA and AD or AD pathology were identified by the literature search after applying specific inclusion
and exclusion criteria. The data suggests that OSA is associated with greater cognitive deficits in middle-
188
age adults than in older adults and the elderly; greater risk of subsequent cognitive decline in middle-aged
adults than in older adults and elderly; and in older adults with MCI rather than in healthy older adults,
OSA may be the critical correlate of cognition.
Conclusion: OSA may be age-dependent in older adults (60 – 70 years old) and the elderly (70 years and
above) and is associated with neurodegenerative diseases particularly, cognitive decline and AD. In the
middle-aged (30 – 60 years old), the data suggests that OSA may be age-related and presents with a
distinct phenotype, with cardiovascular, and possibly metabolic effects. Intermittent hypoxia and sleep
fragmentation are two main processes by which OSA induces neurodegenerative changes.
INTRODUCTION
Cognitive decline in older adults and Alzheimer’s disease (AD) both have debilitating effects on
individuals affected, with significant socioeconomic implications.1-4 It is well known that advanced age is
the most notable risk factor for cognitive decline and AD. With projected substantial increase in the
ageing population, it is pertinent that efforts directed at comprehending risk factors of cognitive decline
and AD be consolidated. There is increasing evidence in recent years linking cognitive decline and AD to
various sleep problems and disorders.5-8 Notably, the link between cognitive impairment and Obstructive
Sleep Apnea (OSA) is well-established,9-12 and recent evidence suggest that a longitudinal risk between
OSA, cognitive decline and AD exist.13-15 It is therefore vital to clearly understand the distinct and unique
relationship OSA has with cognitive decline and AD in the elderly as this will help inform prevention and
treatment strategies of both OSA and AD.
Age-related and Age-dependent OSA co-morbidities
Obstructive sleep apnea (OSA) is characterized by intermittent hypoxemia, sleep fragmentation
and intrathoracic pressure changes.16,17 OSA is an issue of public health significance, because it is highly
prevalent,18-21 results in serious morbidity and significant mortality, 22-26 and has a high socio-economic
189
impact.27,28 Studies suggest that OSA in the elderly may result in varying functional outcomes relative to
OSA in the middle-aged. Therefore, understanding and appreciating the heterogeneity of OSA and its
outcomes in distinct age groups is critical in mitigating the deleterious effects of OSA.
Bliwise et al.29 in their study using the Bay Area Sleep Cohort described two peak incidences of
OSA, one being age-dependent and the other being age-related. The concept describes OSA as an age-
dependent condition in the elderly, relative to OSA being an age-related condition in the middle-aged.
OSA affects 3–7% of the middle-aged population and becomes more prevalent with increasing age.20,30-37
In middle-aged individuals, Bliwise noted that OSA is thought to show an age-related occurrence
conferring a specific period of susceptibility,38,39 in association with disparate morbidities. Age-related
diseases and disorders are not particularly related to the process of aging, occurring at a distinctive
age/age-group and then lessening in frequency with increasing age (e.g. multiple sclerosis and
amyotrophic lateral sclerosis).29
Conversely, in the elderly, the prevalence of OSA ranges from 30-80%,32,40-46 contingent on the
definition of OSA. The significance of these high rates in the elderly is ambiguous; while certain studies
suggest reduced mortality with increasing age,47 others suggest otherwise.48 However, the pathogenesis of
OSA somewhat appears to involve normal aging with exponential increases in mortality and morbidity
with advancing age.49,50 This physiologic susceptibility with increased incidence with chronological age
suggests that OSA is age-dependent in the elderly. Mechanisms underlying how aging increases the risk
of OSA are not completely understood. The effects of advancing age are generally considered to function
with other physiologic systems experiencing aging.51,52 Notable factors in OSA pathogenesis during
aging include an extremely collapsible airway, limited upper airway dilator muscle activity, diminished
respiratory arousal threshold, and a precarious ventilatory control system.35,53,54 Furthermore, age-
dependent changes in lung function as reported in cross-sectional studies,55-57 are associated with
increments in OSA frequency.
190
Much focus on OSA’s age-dependence in the elderly has been in cardiovascular outcomes
relative to cognitive outcomes. However, the various co-morbidities associated with increasing age in
OSA increases susceptibility to cognitive decline.58,59 Cardiovascular outcomes of OSA such as
hypertension, coronary heart disease, and congestive heart failure; 25,60,61 stroke; 62-64 multiple
inflammatory and metabolic effects including diabetes; 65-68 are associated with increased severity of
OSA, and may be more likely to occur in young and middle-aged (30 – 70 years), rather than older
populations.69-73 In young and middle-aged adults, substantive evidence exists establishing relationships
between OSA and cognitive function, including areas such as memory, motor, vigilance and executive
functions.74-76 Studies examining cardiovascular outcomes of OSA in the elderly show varying findings.
Some studies77-79 report increased mortality and incidence of stroke in older adults while other
studies47,48,80 report decreased mortality with age in OSA patients compared to the general population. In
the elderly, multiple studies demonstrate associations between OSA and cognitive decline.15,81-83 We
demonstrated that in the elderly, OSA patients had an earlier age of cognitive decline to MCI and to AD
than non-OSA controls,84 and showed that individuals with OSA were more than two times more likely to
have AD compared to non-OSA individuals in a recent meta-analysis.5 A meta-analysis of cross-sectional
studies also concluded OSA-AD associations, such that AD patients were at an increased risk of
presenting with OSA than cognitively normal individuals.85 Of particular concern is that longitudinal
studies also provide evidence of an association between OSA and cognitive decline. A 5-year prospective
study of older women (mean ±SD age: 82.3 ±3.2 years) found elevated oxygen desaturation index (≥15
events/ hr) to be associated with increased risk of developing mild cognitive impairment (MCI) or
dementia.15 In general, the literature regarding OSA, cognitive function, cognitive decline and AD have a
relationship with advancing age, suggesting that age may be an effect modifier of these association and
that OSA may present as a distinct physiologic phenotypic risk factor in older adults with cognitive
decline and AD.
191
Review Objective
Recent narrative reviews on OSA, Cognitive Decline and AD described the cognitive profiles
found in association with OSA in children and adults (young, middle-aged and elderly adults);9,86
explored mechanisms, neurobiology and treatment for geriatric psychiatry audience; and discussed
probable explanatory mechanisms linking OSA, depression and cognitive dysfunction. 87-89 Other
narrative discussions focused on the probable explanatory mechanisms linking OSA to dementia.82
Previous systematic and meta-reviews focused on how OSA affects specific neurocognitive domains,
producing inconsistent74,90 and sometimes non-conclusive findings.12,91 The only meta-review focusing on
older adults reported a small association between OSA and cognitive dysfunction and suggested that some
individuals may be more at risk of adverse cognitive effects.92
In this review, we integrate over 3 decades of research examining OSA and cognition; OSA and
subsequent cognitive decline; and OSA and AD; with particular focus in appreciating the heterogeneity of
OSA and its outcomes in distinct age groups. We note that the data suggests that OSA may be age-
dependent in older adults (60 – 70 years old) and the elderly (70 years and above) and is associated with
neurodegenerative diseases particularly, cognitive decline and AD. In the middle-aged (30 – 60 years
old), the data suggests that OSA may be age-related and presents with a distinct phenotype, with
cardiovascular, and possibly metabolic effects. We also explore and describe possible mechanisms linking
OSA as a precipitator of AD pathogenesis especially in the elderly, by systematically reviewing clinical
and epidemiological evidence of such an association. Where findings are discrepant, we focus on
methodological differences among studies.
METHODS
Search strategy
A systematic literature search of bibliographic databases including PubMed/Medline, Embase,
Psych INFO and Cochrane library for clinical trials, was conducted to identify all eligible studies
192
(published from their onset up until August 31, 2017) that examined associations between OSA and
cognitive function, OSA and subsequent cognitive decline, and OSA and AD. Our search strategy utilized
the combination of terms characterizing cognitive function, cognitive impairment, AD or AD pathology
as the dependent variables, OSA as the independent variable and a third set of terms specifying specific
study types including clinical and epidemiological studies. For the dependent variables, search terms
included cognition, cognitive function, executive function, neuropsychological function, cognitive
decline, mild cognitive impairment, Alzheimer’s disease, AD biomarkers and amyloid beta. For the
independent variable, search terms included sleep apnea; sleep disordered breathing and obstructive sleep
apnea. All terms were combined with the Boolean Logic operator OR, within each set, and the three sets
were combined using AND. Furthermore, we performed a manual search of references of included
articles to identify relevant references not identified by the automated search. Indexing and management
of the references was done using EndNote X7 (Thompson Reuters, New York, NY). This review was
conducted adhering to the preferred reporting items for systematic reviews and meta-analyses (PRISMA)
statement by Moher et al.93
Selection criteria
Eligible studies for this review had to meet the following selection criteria: 1) be original research
investigations examining associations between OSA and cognition, OSA and cognitive decline, and OSA
and AD; 2) conducted in humans; 3) include both healthy controls and OSA patients and conducted
between group comparisons. Studies without controls that conducted within group comparisons based on
OSA severity were also considered; 4) conducted in adults (young, middle-aged and older); 5) studies
examining cognition or cognitive decline as an outcome must have utilized objective neuropsychological
cognitive tests; 6) studies examining AD or AD pathology as an outcome must have utilized objective
measures; and 7) diagnosis of OSA must have been based on polysomnography. Excluded studies include
case reports, case series, narrative or systematic reviews, abstracts or editorials. Studies conducted in
OSA patients that did not include relevant cognitive parameters (i.e. executive, motor, verbal, attention,
193
memory) and those that examined the effects of CPAP but did not include an examination of OSA vs.
control at baseline, were also excluded.
Reviewing Procedure and Data Extraction
Bibliographic data base searches were conducted first in May 2017. Another search was
conducted in August 2017 to identify manuscripts published between May and August 2017. Independent
examination of all titles and abstracts of identified eligible studies by the search strategy was performed
by two authors (OB and XX) using EndNote X7. Selected articles at this stage were further assessed by
retrieving and reading the manuscript’s full text. Identified articles for the review were assessed for
possible inclusion by the same two authors. Where there were discordant decisions regarding inclusion, a
resolution was reached by two other authors (XX and XX). Two authors (OB and XX) performed data
extraction for each reference. Extracted fields included authors, year of publication, study design, study
population, age, exposure and outcome assessment, statistical analytic methods used, covariates, and the
main findings of the study. Two other authors (XX and XX) resolved discrepancy in the information
extracted. Reviewers were not blinded to the funding, authors, or institutions. Figure 1 shows a summary
of the study selection and retrieval process.
Assessment of Study Quality
We assessed the quality of included studies in this review using an adaptation from the modified
version of the Newcastle-Ottawa scale (NOS) for quality assessment of the observational studies,94 with
addition of new items relevant to this review. Following adaptations from previous reviews, parameters
used for the quality assessment included well-specified hypothesis, study design type, appropriately
described sample, sample size, assessment and definition of OSA, cognitive impairment or AD, statistical
analytic methods used, and approach used to adjust for potential confounders. We utilized a star rating
system with increasing number representing increasing quality. Tables 1-3.
194
RESULTS
OSA and Cognition (Cross-sectional studies)
Main findings
Table 4.1 summarizes the findings from 34 studies that examined the association of OSA and
cognitive function at a single time point. Most studies that were conducted in the middle-aged adults
consistently linked OSA and deficits in attention,95-99 memory,100-102 executive function,99,103,104 motor
function,99 reaction time,105 psychomotor vigilance106 and information processing speed.107 Explanations
and plausible mechanisms responsible for these findings in the middle-aged include daytime sleepiness or
drowsiness from fragmented sleep because of frequent apneic episodes,108-110 and hypoxia.111,112
Specifically, deficits of attention and memory may be due to fragmented sleep and excessive daytime
sleepiness,102,113 while motor function, executive function, reaction time and vigilance may be related to
the severity of hypoxemia.114-116 For example, studies where middle-aged adults with OSA who
complained of excessive daytime somnolence were compared to healthy controls, consistently showed
scores in memory and attention that were lower than normal.99,100,102 Furthermore, correlation analysis
revealed that daytime somnolence correlated with attention while nocturnal hypoxemia correlated with
executive function and visual-constructive abilities.99
Studies in the middle-aged that examined the effect of CPAP treatment on cognitive impairment
after both a short and long-term treatment period revealed conflicting results. Studies included in our
review administered CPAP and assessed for its effect after 3-months,96,97 6 months117 and 12 months.98
After 3-months of CPAP treatment, Canessa et al.96 observed significant improvements involving
memory, attention, and executive functioning in OSA patients, while Saunamaki et al.117 did not show
improvements in OSA patients’ visuospatial organizational skills or their mental set-shifting performance,
after a 6 months CPAP treatment. Variations in methods of assessing cognitive decline, and the possible
presence of confounding variables, such as rapid eye movement (REM) and slow-wave sleep (SWS)
occurring after first night of CPAP use could be responsible for the discrepant findings.
195
Castronova et al.96 employed imaging techniques using diffused tensor imaging (DTI) to examine
changes in white matter integrity and cognition following CPAP treatment in severe OSA patients. Pre-
treatment OSA patients showed impairments in most cognitive areas, and diffuse reduction of white
matter integrity. Limited changes of white matter were seen after 3 months of CPAP. However, the
authors report an almost complete reversal of white matter abnormalities over the course of 12 months
CPAP treatment. In addition, significant improvements involving memory, attention, and executive-
functioning paralleled white matter changes after treatment. Other studies using imaging techniques
showed regions of diminished grey matter volume in the left hippocampus, frontal structures of the brain
and within more lateral temporal areas in patients with OSA after CPAP treatment.97,118,119 These findings
suggest that cognitive impairment seen in middle-aged patients with moderate to severe OSA is
associated with damage to brain tissue in areas involved in several cognitive tasks. In addition, nocturnal
intermittent hypoxia and fragmented sleep could be culprits in cognitive and brain structural deficits in
OSA.
Studies that restricted their populations to older adults or whose mean age occurred in that range
(i.e., ages 61 – 70) generally show weaker results with cognition where associations were identified.
Otherwise the plurality of findings in studies where potential confounders were accounted for were null
findings. First, consider evidence for hypoxia indices and executive, memory and attention. Boland et
al.120 in their study examined the effect of Respiratory Disturbance Index (RDI) values on attention,
executive function and memory in 1700 subjects free of clinically diagnosed SDB who underwent at-
home polysomnography (PSG) as part of the Sleep Heart Health Study (SHHS). This study controlled for
age, education, occupation, field center, diabetes, hypertension, body-mass index, use of CNS
medications, and alcohol drinking status, and found no association between the RDI and any of the three
cognitive function measures assessed. Foley et al.121 assessed the association between sleep-disordered
breathing and cognitive functioning in an elderly cohort of Japanese-American men and found no
association between sleep-disordered breathing and memory function, concentration, and attention.
Phillips et al.122 and Sforza et al.123 also had null findings between AHI findings and global cognition,
196
attention, executive function, memory, language and motor functions. Dlugaj et al.124 examined the
association between Mild Cognitive Impairment (MCI) and MCI sub-types and SDB severity and found
no association. Except for Sforza and Dlugaj, the other studies made use of a healthy population and this
may have contributed to this finding.
On the contrary, other studies conducted in older adults showed associations between OSA and
delayed verbal memory,125,126 non-verbal IQ memory delayed recall,127 executive function, and
attention128,129 and decreased vigilance.130 As earlier specified most of these studies were only able to
demonstrate relatively weak associations. For example, Yesavage et al.,129 Terpening et al.,131 and Berry
et al.,127 demonstrated effects between hypoxia measures and cognitive measures that were minimally
significant. Perhaps OSA in older adults with MCI rather than in healthy older adults is the critical
correlate of cognition in older adults. Kim et al.132 specifically examined the effect of SDB in older adults
with MCI on language test performance. Individuals with MCI have an increased risk of developing AD.
Although the study found no differences between MCI and control groups on AHI and hypoxia measures,
higher AHI was associated with language test performance among individuals with MCI. These findings
possibly reflect frontal-subcortical pathology in patients with MCI suggesting that susceptibility to a
specific brain damage associated with OSA could increase the risk for AD.
Examination of studies conducted in the elderly (i.e. ages > 70 years) are also contradictory;
however, more recent studies tend to show associations between hypoxemia and executive function or
global cognition. Blackwell et al130 examined SDB and cognition in elderly men and adjusted for age,
race, education, BMI, lifestyle, comorbidities and medication use. Findings from the study showed
significant associations between SDB and cognition. Spira et al.133 conducted a similar study in elderly
women and determined whether the apolipoprotein E (APOE) e4 allele modifies the association. Results
showed that increased AHI was associated with decreased cognition and APOE4 increased this risk by
five times.
197
OSA and Cognitive Decline (Longitudinal studies)
Main findings
Table 4.2 summarizes the only seven prospective epidemiological studies that have attempted to
answer this critical question. Six of the studies were in older adults and the elderly while one study
included a mixture of middle-aged, older adults and the elderly.
In older adults and elderly, OSA at baseline may be a weak predictor of cognitive decline. Only
two studies that included older adults and the elderly reported significant cognitive associations with OSA
that were mostly modest. Blackwell et al. conducted a population-based study in a setting that included
six centers in the United States. They included community-dwelling older men without cognitive
impairment at baseline, and followed them for approximately 3 years. Results from the study showed a
modest association between nocturnal hypoxemia and subsequent decline in cognition. Martin et al
utilized a subset of individuals enrolled in the PROgnostic indicator OF cardiovascular and
cerebrovascular events, study (PROOF). They assessed whether cognitive function changes occurred in
untreated elderly SBD patients and without dementia. After an 8-year follow-up, SDB showed a
minimally significant association with decline in the attentional domain and this was more evident in the
patients with an AHI > 30. There were no significant changes in the executive and memory functions over
the follow-up period. In this study chronic hypoxemia accounted for between 4–7% of variance in the
decline in attention. In contrast, Cohen-Zion et al in their study utilizing community-dwelling elderly
individuals with high risk for SDB found no association of hypoxia indices with cognitive decline after
controlling for oximetry, sleep and subjective report. However, a significant association existed between
increases in daytime sleepiness and declining cognitive function. In addition, Lutsey et al utilizing a
subset of individuals from The Atherosclerosis Risk in Communities (ARIC) study, who participated in
the Sleep Heart Health Study (SHHS), examined the effect of OSA and cognitive decline over a 15-year
follow-up. Overall, this study found no evidence that OSA severity, degree of nocturnal hypoxemia, sleep
fragmentation, or habitual sleep duration were associated with subsequent cognitive decline. The
relationship between midlife OSA and later life cognition was also null.
198
Studies examining the relationship between OSA and Dementia outcomes tend to be more
consistent in their findings. Yaffe et al in their prospective study of SDB and cognition in elderly women
without dementia at baseline who were a substudy of the Study of Osteoporotic Fractures (SOF) followed
for approximately 5 years, found that elderly women had an 85% increased risk of developing
MCI/Dementia. In another study, Yaffe and her colleagues conducted a retrospective cohort study using
medical record data, and examined the relationship between a diagnosis of sleep disturbance and
dementia in elderly veterans. Sleep disturbance was significantly associated with an increased risk of
dementia and specifically, OSA demonstrated significant associations with increased risk of AD, Vascular
dementia and other dementias combined (frontotemporal dementia, senile dementia, or dementia not
otherwise specified). Furthermore, Chang et al in their retrospective matched-control cohort study that
utilized data from Taiwan’s Health Insurance Database estimated dementia risk in OSA versus non-OSA
patients in individuals 40 years and older, followed for 5 years. Results from the study showed a 70%
increased risk of developing dementia among OSA compared to non-OSA individuals. This study also
demonstrated gender-dependent, age-dependent and time-dependent associations of OSA and dementia.
OSA females were more like to develop dementia; OSA males aged 50-59 (middle-aged) had an
increased risk for developing dementia, while OSA females aged ≥ 70 years (elderly) had a greater risk of
developing dementia, and OSA patients were more likely to develop dementia in the first 2.5 years of
follow-up.
Summary, critique and future research directions
In middle-aged adults, cross-sectional studies mostly demonstrate that OSA is often associated
with cognitive decline. Longitudinal studies demonstrating that OSA can precede cognitive decline in
middle-aged adults are rare, however, the one study that examined this relationship, suggests so.
Nocturnal intermittent hypoxia and fragmented sleep could be culprits in cognitive and brain structural
deficits in middle-aged OSA patients. Studies utilizing brain-imaging techniques suggest that cognitive
impairment seen in middle-aged patients with moderate to severe OSA is associated with damage to brain
tissue in areas involved in several cognitive tasks including grey matter volume in the left hippocampus,
199
frontal structures of the brain and lateral temporal areas. CPAP treatment seems to be beneficial however
over a long-term as one study reported an almost complete reversal of white matter abnormalities over the
course of 12 months CPAP treatment.
In older adults and elderly, cross-sectional studies have not consistently demonstrated OSA-
Cognition associations and studies that identified such only revealed relatively weak associations. This
distinction and pattern in OSA-Cognition associations in middle-aged, older adults and elderly provides
the foundation for envisioning age-related modification of OSA-cognition relationships. Most of the
studies that examined this relationship in healthy community older adults revealed null associations. On
the contrary, studies examining this association in older adult OSA patients with MCI consistently
showed significant associations of OSA with cognition, suggesting that perhaps OSA in older adults with
MCI rather than in healthy older adults is the critical correlate of cognition in older adults. Longitudinal
studies in older adults and elderly, suggest that OSA at baseline may be a weak predictor of cognitive
decline. However, longitudinal studies examining the relationship between OSA and Dementia outcomes
produced consistent significant findings. This evidence that OSA can possibly precede MCI/AD and that
perhaps OSA in older adults with MCI rather than in healthy older adults is the critical correlate of
cognitive decline in older adults, may indicate that the presence and deposition of the AD biomarker, beta
amyloid, may be a moderator in these relationships. We consider this theme in the remaining sections.
Methodological differences existed among studies reviewed and can be rightly viewed as
limitations in the field of OSA and AD research. Issues related to the single assessment of OSA in
longitudinal studies, absent or incomplete CPAP intervention information during follow-up, and the
possibility that the etiological timeframe relevant for the association between OSA and AD could be
outside the examined period, variability in ways in which cognition was assessed, and issues relating to
selection bias, are all opportunities for future improvements. Many studies utilized clinic patients. It is
clear that the likelihood of clinic attendance in such participants is associated with disturbed sleep.
Therefore, when analyses are conducted on only such participants, biased results are likely the outcome.
In addition, not all studies accounted for the possible role of depression and its symptoms, which may be
200
important mediator or confounder of the association between disturbed sleep and cognition.134,135
Therefore, new research in the field should endeavor to separate causality relating to OSA, depression and
cognition.
Despite methodological strengths such as use of PSG sleep measures, and certain long follow-
ups, these were not all present in all studies, therefore limiting the strength of causal inferences that can
be made. Lastly, future studies should examine whether these associations are causal, focusing on the
mechanisms responsible for the somewhat different OSA effects seen at different ages or in different
populations.
OSA and Alzheimer’s disease (Cross-sectional studies)
Main findings
Studies examining any association between OSA and a diagnosis of AD were conducted in older
adults and the elderly since most AD diagnoses occur after age 65 (Tables 4.3 and 4.4). The scarcity of
epidemiological studies examining OSA and AD co-morbid associations was noteworthy. The five
studies included were conducted some 2 to 3 decades ago and had varying findings (Table 4.3). Two
studies136,137 demonstrated a significant association for OSA in AD patients relative to controls while
three of the studies had null associations. However, a recent meta-analysis of these studies concluded that
the aggregate odds ratio for OSA in AD vs. healthy control was 5.05 and homogeneous.85 Given the
cross-sectional nature of these analyses, the data cannot be interpreted as suggesting direction of causality
or temporality between OSA and AD. However, it does suggest the possibility of a reciprocal relationship
between these two disorders. OSA possibly facilitates progressive central nervous alterations and
excessive daytime sleepiness in patients and this further increases the risk of cognitive decline or AD
progression in older adults.86
201
OSA and AD Pathology
Main findings
Cross-sectional studies
Observational studies of OSA and AD pathology are vital in establishing links between OSA and
AD. If the relationship between OSA and AD is causal then demonstrating AD-specific neuropathology
resulting from increased disease burden in OSA patients can be vital. Cross-sectional studies examining
this association were conducted mostly in older adults and the elderly. We identified two studies
conducted in middle-aged participants (Table 4.4). A recent study by Yun et al138 examined whether OSA
increased brain amyloid burden relative to controls in middle-aged participants from the Korean Genome
and Epidemiology Study. After adjusting for potential confounders, OSA patients had a higher brain
amyloid burden in the right posterior cingulate gyrus and right temporal cortex, relative to controls,
suggesting that OSA may have been responsible for these findings. In this study, examination of regional
differences in cortical thickness were not significant. Bu et al.139 examined whether hypoxia indices in
OSA was associated with serum amyloid (Aβ40, Aβ42), total tau and phosphorylated tau 181 (P-tau 181)
comparing their levels with controls. Results from this study demonstrated significantly higher levels of
serum Aβ40, Aβ42, total Aβ and P-tau 181 in OSA patients compared to control. Hypoxia indices
including apnea-hypopnea index, the oxygen desaturation index, and the mean and lowest oxyhemoglobin
saturations correlated positively with Aβ40, Aβ42, and total Aβ in OSA patients, suggesting that hypoxia
possibly effects AD pathogenesis.
In older adults, all cross-sectional studies examined demonstrated associations between OSA and
AD pathology (Table 3). Osorio et al140 examined whether an association existed between SDB severity,
cerebrospinal fluid (CSF) Alzheimer's disease (AD) biomarkers, and the APOE alleles in cognitively
normal older adults. Findings from the study demonstrated an association between SDB and CSF
202
AD- biomarkers. They also found an association between intermittent hypoxia and increases in CSF T-
Tau, P-Tau and Aβ42 in ApoE3+, suggesting that hypoxia may be responsible for changes in the AD
biomarkers in this population of cognitive normal older adults. In ApoE4+ subjects, the study did not
detect significant SDB group differences for Aβ42, T-Tau or P-Tau, though SDB severity tended towards
lower CSF Aβ42 levels. However, participants were at varying stages of preclinical disease amongst the
ApoE4+ subjects. A recent study by Liguori et al141 compared CSF β-amyloid42, tau proteins, and lactate
levels in OSA versus CPAP treated OSA and controls. All participants had MCI. Findings from the study
demonstrated lower CSF β-amyloid42, higher lactate levels and t-tau/ Aβ42 ratio compared to controls
and CPAP treated OSA patients. Controls and CPAP treated OSA MCI patients had similar AD
biomarker levels. These findings suggest that OSA may effect early AD biomarker changes that may be
susceptible to CPAP treatment. Furthermore, in the elderly MCI patients, Spira et al showed that greater
SDB severity was associated with greater brain amyloid burden
Prospective/Longitudinal studies
Lutsey et al142 examined whether diagnosed OSA in the middle-aged was associated with adverse
morphological brain changes 15 years later. They utilized participants from the Atherosclerosis Risk in
Communities Study (ARIC). After accounting for body mass index in a series of multivariate models,
OSA at mid-life was not associated with cerebral markers of dementia or White Matter Hyper intensity
brain volume or regional brain volume changes. This study’s strength is its length of follow-up, since
dementia is known to have a long preclinical duration and middle-aged individuals may represent the
etiological relevant phase, similar to other dementia risk factors including diabetes,143-145 hypertension,146-
149 and smoking.145,150 Also, use of community based participants rather than small sample sized clinic
participants is also a plus. However, various methodological concerns could have been responsible for
their findings. Selection bias issues could have occurred from loss to follow-up as a third of participants
did not attend follow-up for neurocognitive assessments. The study had relatively few diagnosed severe
OSA patients, necessitating lumping of moderate and severe OSA patients together, which could have
203
attenuated any association that could potentially have occurred in severe OSA patients. The study also
did not account for CPAP use or treatment of OSA during the follow-up period. In contrast, Osorio et al14
using Alzheimer’s Disease Neuroimaging Initiative (ADNI) data examined whether SDB was associated
with an earlier onset age at MCI or AD. The study also determined whether CPAP treatment delayed MCI
or AD onset. Findings showed that in all population subsets examined, SDB patients had an earlier onset
age to MCI or AD-dementia. CPAP use also delayed the age of MCI onset. This study utilized a well-
defined longitudinal cohort; however, study limitations include the self-reported SDB and CPAP status.
Randomized Controlled Trials
The most effective treatment for OSA is continuous positive airway pressure (CPAP). With
CPAP treatment, OSA patients without dementia show an increase in slow wave sleep (SWS) and rapid
eye movement (REM) sleep.151-154 However, what is the effect of CPAP on sleep parameters and
cognition in AD patients with OSA? All four RCTs identified in this review included elderly participants
(mean age > 70) and reported significant improvements in deep sleep,155 mood,156 cognition,156,157 and
daytime sleepiness158 in OSA patients with AD (Table 4.4). For example, in a randomized placebo-
controlled trial, Cooke et al examined the effect of CPAP on sleep parameters in AD patients with OSA.
The researchers administered 3 weeks of CPAP treatment and compared the outcome with 3 weeks
placebo CPAP in patients with AD and OSA. Results from the study showed significant improvements in
SWS after one night with the improved effect extending for three weeks. Chong et al also examined the
effect of CPAP treatment on daytime sleepiness and found a significant reduction in patients with mild-
moderate AD with SDB after CPAP treatment. Examining the effects of CPAP on cognition in AD
patients, Ancoli-Israel et al157 compared CPAP treatment for three weeks versus placebo CPAP for three
weeks and found significant cognitive improvements. In addition, post-hoc analyses showed
improvements in episodic verbal learning and memory, executive functioning such as cognitive
flexibility, and mental processing speed.
204
A limitation identified in these RCTs include the issue of power to detect meaningful changes
across treatment arms. Some studies were underpowered and as such could not make definitive statements
specific to improvements within certain cognitive constructs. Other limitations include examination of
sleep parameters post-hoc while the study was powered for changes in cognition, inability to make causal
inferences due to non-random group assignment (continued use versus discontinuation of CPAP), limited
validation of ESS in elderly patients with AD, and issues with generalizability. Despite these limitations,
there is sufficient evidence to conclude that CPAP treatment may be effective in improving cognition in
OSA patients with AD.
Summary, critique and future research directions
In older adults and the elderly, cross-sectional studies suggest the possibility of a reciprocal relationship
between OSA and AD. The aggregate odds ratio in older adults, for OSA in AD vs. healthy control was
5.05 and homogeneous as demonstrated in a recent meta-analysis. OSA patients have an increased risk of
cognitive decline as shown in the above in the OSA and Cognition section, and cognitive decline
increases the risk for AD. Methodological concerns relating to these studies include the fact that
participants were mostly probable AD, because of the clinical and cognitive presentations of the enrolled
patients and the lack of reported neuroimaging or biomarkers data. Participants were also significantly
more educated which could have been responsible for the null findings seen in some studies. Higher
educational level and cognitive reserve are known protective risk factors for OSA induced cognitive
decline.86,159 Other studies failed to control for demographic and other possible confounders of the
association between OSA and AD. For example, Smallwood et al.160 failed to account for gender
differences in their sample. Women were significantly higher in their sample, which could have been
responsible for their null findings since studies suggest that OSA is more prevalent in men and increases
with age.161
In middle-aged, older adults and elderly individuals, cross-sectional data consistently suggests
that there is an association between OSA and preclinical-AD in the cognitively normal. Studies
consistently demonstrated OSA associations with AD biomarkers including higher levels of serum Aβ40,
205
Aβ42, total Aβ and P-tau 181 in OSA patients compared to control, with APOE possibly acting as an
effect modifier. Although determination of the causal relationships and directionality of these associations
is limited in these studies, it is possible that OSA increases the risk for preclinical AD or vice versa.
Prospective studies examining whether OSA accelerates amyloid deposition and effects regional
brain morphological changes that contribute to the development or progression of Alzheimer’s disease are
sparse in all age groups. The two prospective studies examined had contradictory findings. Both studies
also had their strengths and weaknesses (see section on OSA and AD Pathology/Prospective studies).
Therefore, future research is needed in the field to address the temporal nature of OSA-AD relationship
and the possible mechanisms underlying the relationship.
Evidence from RCTs provide an insight into the possibility of causal associations between OSA
and AD and are somewhat compelling. All RCTs were conducted in the elderly, and showed that CPAP
treatment not only improved sleep parameters in AD patients with OSA, such as improved SWS, and
excessive daytime sleepiness, it also increased cognitive function including areas such as episodic verbal
learning and memory, executive functioning such as cognitive flexibility, and mental processing speed.
Such compelling is the evidence such that effects were seen with a single night of CPAP treatment. These
findings clearly provides evidence that AD patients (particularly mild to moderate AD) with OSA can
benefit from CPAP treatment. It is also important to note that studies have shown that management of AD
with donepezil improved various hypoxia and sleep duration measures, suggesting that the cholinesterase
inhibitor improves subjective measures of OSA independently, 162,163 therefore highlighting the potential
reciprocal relationship between sleep and AD.
Future RCTs need to include dose-response studies stratified not only to the mild, moderate, and
severe categories of OSA, but also to include categories addressing duration of disease, extent of
intermittent hypoxemia extent, fragmented sleep severity, and presence of comorbidities. Issues with
design and sample size of double-blinded, placebo controlled clinical trials addressing the effect of CPAP
exist and need to be improved. Furthermore, as Pan et al164 noted in their review, non-inferiority trials
utilizing sleep apnea dental devices or other non-PAP therapeutics will also be beneficial.
206
DISCUSSION
Altogether, over three decades of research has investigated OSA-cognition, OSA-cognitive decline
and OSA-AD associations in the middle-aged, older adults and elderly. The data suggests the following:
- OSA is associated with greater cognitive deficits in middle-age adults than in older adults and the
elderly;
- OSA is associated with greater risk of subsequent cognitive decline in middle-aged adults than in
older adults and elderly;
- OSA in older adults with MCI rather than in healthy older adults may be the critical correlate of
cognition in older adults, therefore suggesting that presence or absence of amyloid burden might
act as a moderator in these relationships. This could be responsible for findings showing
increased amyloid deposition in older MCI patients with higher AHI but not in cognitive normal
controls165
- There is a link between OSA and AD biomarkers of neurodegeneration (e.g. Aβ40, Aβ42, total
Aβ and P-tau 181) and subsequent cognitive decline, in the middle-aged, older-adults and even in
elderly cognitive normal individuals.
We believe that this distinct pattern observed in OSA-Cognition and OSA-AD biomarker associations in
middle-aged, older adults and elderly provides the foundation for envisioning age-related modification of
OSA-cognition relationships.
A pertinent question arises from the findings: why are OSA-cognitive associations particularly
pronounced in the middle-aged and less so in older adults or the elderly? Several studies suggest that the
link between sleep and cognition weakens with increasing age because the aging brain is unable to
adequately and efficiently facilitate specific sleep supported cognitive processes.166-171 If this is true, then
it could have been responsible for the null or weaker results with cognition where associations were
identified in older adults and the elderly in this review. It also implies that improving duration and quality
of sleep in older adults and the elderly may not improve cognitive dysfunction because of diminished
neural plasticity, increased neuronal loss and atrophy, neuroendocrine changes, and nocturnal hypoxia
207
that occurs as one ages.172 These neurobiological changes seen in older adults and the elderly may also
compromise memory consolidation processes despite possible preservation of slow wave sleep amplitude
and density. Scullin and Bliwise in their review make the case that functional weakening of the brain in
their support of sleep-specific cognitive processes as we age, that hippocampal—neocortical
consolidation will not occur regardless of slow wave sleep quantity, if the hippocampus, neocortex, or
hippocampal—neocortical connections are greatly impaired in aging.170,173
Another pertinent question from the above is what mechanisms underlie the subsequent cognitive
decline in the middle-aged, even when no OSA-cognition associations are seen in older adults and the
elderly? Scullin and Bliwise highlight the possibility of disturbed sleep causing changes in sleep
modulated cognitive functions across the lifespan such that there is weakening of and/or compensation
attempts incited at sleep-cognition links. They note that studies demonstrate increases in insulin, cortisol,
blood pressure, proinflammatory cytokines and sympathetic tones, during acute or chronic sleep
restriction states,174,175 all of which possibly hasten cognitive aging.176 In addition, experimental studies
show that restriction of sleep in young animals can increase beta amyloid burden,177,178 and cause protein
misfolding,179 therefore compromising consolidation of memory.180,181 The prolonged accumulation over
time of beta amyloid and protein misfolding in the young and middle-aged further results in irreversible
neurodegeneration,182 thereby suggesting that sleep disturbances could possibly underlie subsequent
cognitive decline, even when no sleep-cognition associations are expressed in older adults and the
elderly.172
In middle-aged OSA individuals with cognitive deficits and particularly in OSA elderly AD
patients, randomized placebo controlled clinical trials provide promising findings of improvement of
cognitive deficits that range from attention and vigilance to memory and executive function. Although
CPAP improved hypoxia and other sleep measures relative to pre-CPAP indices, treated OSA patients
demonstrate comparable level of performance in certain areas such as working memory storage functions,
but still show deficits in other areas such as working memory.157,183 Furthermore, intricate
neuropsychological measures of attention, psychomotor speed and executive function tend to improve
208
only minimally since intermittent hypoxia upregulates death of neurons and hypertrophy of astrocytes in
the hippocampus and parietal cortex.184-186 The literature suggests that deprived sleep negatively affects
the prefrontal cortex (PFC) which largely controls executive function.187,188 Maillet and Rajah189 in their
review found that task performance matched between age groups yielded negative associations between
regional PFC and activity in older adults, however, the reverse was the case between volume and right
lateral PFC, when older adults performed worse than younger adults did. In addition, several studies
showed that in healthy older adults, PFC activity is positively related to grey matter volume in the medial
temporal lobe, but the reverse was observed in MCI and AD patients. In older adults, prefrontal sleep
spindles that facilitates hippocampal episodic learning are markedly reduced, and the increase in disturbed
sleep further compromises cognition.190,191 Therefore, OSA worsens metabolic injury192 that is particularly
resplendent in middle-aged, and exacerbates neuronal injury and facilitates memory and cognitive
impairment,193-197 that is particularly resplendent in older adults and the elderly.
Recent exciting trends in the literature point to a link between OSA and AD biomarkers of
neurodegeneration and subsequent cognitive decline, in the middle-aged, older-adults and even in elderly
cognitive normal individuals. This finding across our entire examined age spectrum poses a thought
provoking question as to why this is so. The data suggests that intermittent hypoxia and sleep
fragmentation are two main processes by which OSA induces neurodegenerative changes. Chronic
intermittent hypoxia, hypercapnia and hypertension in OSA can induce neuronal damage, including
axons,198 white matter,199 and reduced Diffusion tensor imaging based mean diffusivity in multiple brain
regions.200 Multiple studies also show grey matter loss in OSA patients compared to controls.119,201,202 As
a result of the pathophysiological effect of hypoxia precipitating hypertension,203-206 hypoperfusion,207-209
impaired glucose metabolism,210-212 and adverse cardiovascular,213,214 and metabolic consequences,215,216
beta amyloid deposition,217-220 and tau hyperphosphyrylation,212,221-224 ultimately these effects lead to
cognitive decline and progress to AD. In addition, inflammation,210,221-224 and oxidative stress,225,226
upregulate neurocognitive impairment in OSA and sustained OSA-cognitive dysfunction overlay with
that seen in AD-associated cognitive decline.82
209
CONCLUSION
OSA may be age-dependent in older adults (60 – 70 years old) and the elderly (70 years and
above) and is associated with neurodegenerative diseases particularly, cognitive decline and AD. In the
middle-aged (30 – 60 years old), the data suggests that OSA may be age-related and presents with a
distinct phenotype, with cardiovascular, and possibly metabolic effects. Intermittent hypoxia and sleep
fragmentation are two main processes by which OSA induces neurodegenerative changes.
References
1. Harada CN, Natelson Love MC, Triebel KL. Normal cognitive aging. Clinics in geriatric
medicine. 2013;29(4):737-752.
2. Hebert LE, Beckett LA, Scherr PA, Evans DA. Annual incidence of Alzheimer disease in the
United States projected to the years 2000 through 2050. Alzheimer disease and associated
disorders. 2001;15(4):169-173.
3. Hebert LE, Weuve J, Scherr PA, Evans DA. Alzheimer disease in the United States (2010-2050)
estimated using the 2010 census. Neurology. 2013;80(19):1778-1783.
4. Thies W, Bleiler L. 2013 Alzheimer's disease facts and figures. Alzheimer's & dementia : the
journal of the Alzheimer's Association. 2013;9(2):208-245.
5. Bubu OM, Brannick M, Mortimer J, et al. Sleep, Cognitive impairment, and Alzheimer's disease:
A Systematic Review and Meta-Analysis. Sleep. 2017;40(1).
6. McKinnon A, Terpening Z, Hickie IB, et al. Prevalence and predictors of poor sleep quality in
mild cognitive impairment. Journal of geriatric psychiatry and neurology. 2014;27(3):204-211.
7. Naismith SL, Hickie IB, Terpening Z, et al. Circadian misalignment and sleep disruption in mild
cognitive impairment. Journal of Alzheimer's disease : JAD. 2014;38(4):857-866.
210
8. Naismith SL, Rogers NL, Lewis SJ, et al. Sleep disturbance in mild cognitive impairment:
differential effects of current and remitted depression. Acta neuropsychiatrica. 2011;23(4):167-
172.
9. Gagnon K, Baril AA, Gagnon JF, et al. Cognitive impairment in obstructive sleep apnea.
Pathologie-biologie. 2014;62(5):233-240.
10. Naismith S, Winter V, Gotsopoulos H, Hickie I, Cistulli P. Neurobehavioral functioning in
obstructive sleep apnea: differential effects of sleep quality, hypoxemia and subjective sleepiness.
Journal of clinical and experimental neuropsychology. 2004;26(1):43-54.
11. Stranks EK, Crowe SF. The Cognitive Effects of Obstructive Sleep Apnea: An Updated Meta-
analysis. Archives of clinical neuropsychology : the official journal of the National Academy of
Neuropsychologists. 2016;31(2):186-193.
12. Vaessen TJ, Overeem S, Sitskoorn MM. Cognitive complaints in obstructive sleep apnea. Sleep
medicine reviews. 2015;19:51-58.
13. Chang WP, Liu ME, Chang WC, et al. Sleep apnea and the risk of dementia: a population-based
5-year follow-up study in taiwan. PLoS One. 2013;8(10):e78655.
14. Osorio RS, Gumb T, Pirraglia E, et al. Sleep-disordered breathing advances cognitive decline in
the elderly. Neurology. 2015;84(19):1964-1971.
15. Yaffe K, Laffan AM, Harrison SL, et al. Sleep-disordered breathing, hypoxia, and risk of mild
cognitive impairment and dementia in older women. JAMA : the journal of the American Medical
Association. 2011;306(6):613-619.
16. Epstein LJ, Kristo D, Strollo PJ, Jr., et al. Clinical guideline for the evaluation, management and
long-term care of obstructive sleep apnea in adults. Journal of clinical sleep medicine : JCSM :
official publication of the American Academy of Sleep Medicine. 2009;5(3):263-276.
17. Sateia MJ. International classification of sleep disorders-third edition: highlights and
modifications. Chest. 2014;146(5):1387-1394.
211
18. Arnardottir ES, Bjornsdottir E, Olafsdottir KA, Benediktsdottir B, Gislason T. Obstructive sleep
apnoea in the general population: highly prevalent but minimal symptoms. The European
respiratory journal. 2016;47(1):194-202.
19. Heinzer R, Vat S, Marques-Vidal P, et al. Prevalence of sleep-disordered breathing in the general
population: the HypnoLaus study. The Lancet Respiratory medicine. 2015;3(4):310-318.
20. Peppard PE, Young T, Barnet JH, Palta M, Hagen EW, Hla KM. Increased prevalence of sleep-
disordered breathing in adults. American journal of epidemiology. 2013;177(9):1006-1014.
21. Senaratna CV, Perret JL, Lodge CJ, et al. Prevalence of obstructive sleep apnea in the general
population: A systematic review. Sleep medicine reviews. 2016.
22. Bolona E, Hahn PY, Afessa B. Intensive care unit and hospital mortality in patients with
obstructive sleep apnea. Journal of critical care. 2015;30(1):178-180.
23. Gami AS, Olson EJ, Shen WK, et al. Obstructive sleep apnea and the risk of sudden cardiac
death: a longitudinal study of 10,701 adults. Journal of the American College of Cardiology.
2013;62(7):610-616.
24. Ge X, Han F, Huang Y, et al. Is obstructive sleep apnea associated with cardiovascular and all-
cause mortality? PLoS One. 2013;8(7):e69432.
25. Wang X, Ouyang Y, Wang Z, Zhao G, Liu L, Bi Y. Obstructive sleep apnea and risk of
cardiovascular disease and all-cause mortality: a meta-analysis of prospective cohort studies.
International journal of cardiology. 2013;169(3):207-214.
26. Won CH, Chun HJ, Chandra SM, Sarinas PS, Chitkara RK, Heidenreich PA. Severe obstructive
sleep apnea increases mortality in patients with ischemic heart disease and myocardial injury.
Sleep & breathing = Schlaf & Atmung. 2013;17(1):85-91.
27. Sullivan. F. Hidden health crisis costing America billions. Underdiagnosing and undertreating
obstructive sleep apnea draining healthcare system. . Darien, IL: American Academy of Sleep
Medicine, 2016.Available from: http://www.aasmnet.org/sleep-apnea-economic-impact.aspx.
212
28. Watson NF. Health Care Savings: The Economic Value of Diagnostic and Therapeutic Care for
Obstructive Sleep Apnea. Journal of clinical sleep medicine : JCSM : official publication of the
American Academy of Sleep Medicine. 2016;12(8):1075-1077.
29. Bliwise DL. EPIDEMIOLOGY OF AGE-DEPENDENCE IN SLEEP DISORDERED
BREATHING (SDB) IN OLD AGE: THE BAY AREA SLEEP COHORT (BASC). Sleep
medicine clinics. 2009;4(1):57-64.
30. Ancoli-Israel S, Klauber MR, Butters N, Parker L, Kripke DF. Dementia in institutionalized
elderly: relation to sleep apnea. Journal of the American Geriatrics Society. 1991;39(3):258-263.
31. Ancoli-Israel S, Klauber MR, Stepnowsky C, Estline E, Chinn A, Fell R. Sleep-disordered
breathing in African-American elderly. American journal of respiratory and critical care
medicine. 1995;152(6 Pt 1):1946-1949.
32. Ancoli-Israel S, Kripke DF, Klauber MR, Mason WJ, Fell R, Kaplan O. Sleep-disordered
breathing in community-dwelling elderly. Sleep. 1991;14(6):486-495.
33. Ancoli-Israel S, Kripke DF, Klauber MR, et al. Natural history of sleep disordered breathing in
community dwelling elderly. Sleep. 1993;16(8 Suppl):S25-29.
34. Duran J, Esnaola S, Rubio R, Iztueta A. Obstructive sleep apnea-hypopnea and related clinical
features in a population-based sample of subjects aged 30 to 70 yr. American journal of
respiratory and critical care medicine. 2001;163(3 Pt 1):685-689.
35. Malhotra A, Huang Y, Fogel R, et al. Aging influences on pharyngeal anatomy and physiology:
the predisposition to pharyngeal collapse. The American journal of medicine.
2006;119(1):72.e79-14.
36. Mehra R, Stone KL, Blackwell T, et al. Prevalence and correlates of sleep-disordered breathing in
older men: osteoporotic fractures in men sleep study. Journal of the American Geriatrics Society.
2007;55(9):1356-1364.
213
37. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered
breathing among middle-aged adults. The New England journal of medicine. 1993;328(17):1230-
1235.
38. Bliwise DNaIK, MH.; Roth, T.; Dement, WC., editors. . Principles and Practice of Sleep
Medicine. 4. Philadelphia: Saunders/Elsevier; 2005. . 2005:p. 24-38.
39. Young T. Age dependence of sleep disordered breathing. In: Kuna, ST.; Suratt, PM.; Remmers,
JE.,editors. . Sleep and Respiration in Aging Adults Elsevier; New York: 1991. 1991:p. 161-170.
40. Ancoli-Israel S, Kripke DF, Mason W, Messin S. Sleep apnea and nocturnal myoclonus in a
senior population. Sleep. 1981;4(4):349-358.
41. Bixler EO, Vgontzas AN, Ten Have T, Tyson K, Kales A. Effects of age on sleep apnea in men:
I. Prevalence and severity. American journal of respiratory and critical care medicine.
1998;157(1):144-148.
42. Ocasio-Tascon ME, Alicea-Colon E, Torres-Palacios A, Rodriguez-Cintron W. The veteran
population: one at high risk for sleep-disordered breathing. Sleep & breathing = Schlaf &
Atmung. 2006;10(2):70-75.
43. Phoha RL, Dickel MJ, Mosko SS. Preliminary longitudinal assessment of sleep in the elderly.
Sleep. 1990;13(5):425-429.
44. Stoohs RA, Gingold J, Cohrs S, Harter R, Finlayson E, Guilleminault C. Sleep-disordered
breathing and systemic hypertension in the older male. Journal of the American Geriatrics
Society. 1996;44(11):1295-1300.
45. Tishler PV, Larkin EK, Schluchter MD, Redline S. Incidence of sleep-disordered breathing in an
urban adult population: the relative importance of risk factors in the development of sleep-
disordered breathing. JAMA : the journal of the American Medical Association.
2003;289(17):2230-2237.
214
46. Zamarron C, Gude F, Otero Y, Alvarez JM, Golpe A, Rodriguez JR. Prevalence of sleep
disordered breathing and sleep apnea in 50- to 70-year-old individuals. A survey. Respiration;
international review of thoracic diseases. 1999;66(4):317-322.
47. Lavie P, Lavie L, Herer P. All-cause mortality in males with sleep apnoea syndrome: declining
mortality rates with age. The European respiratory journal. 2005;25(3):514-520.
48. Punjabi NM, Caffo BS, Goodwin JL, et al. Sleep-disordered breathing and mortality: a
prospective cohort study. PLoS medicine. 2009;6(8):e1000132.
49. Janssens JP, Pautex S, Hilleret H, Michel JP. Sleep disordered breathing in the elderly. Aging
(Milan, Italy). 2000;12(6):417-429.
50. Edwards BA WA, Sands SA, Owens RL, Eckert DJ, White DP, Malhotra A. . Obstructive sleep
apnea in older adults is a distinctly different physiological phenotype. . Sleep. 2014(37):1227–
1236. .
51. Dement WM, LE.; Bliwise, DL. . Physiological markers of aging: human sleep pattern changes.
In: Reff, ME.; Schneider, EL., editors. Biological Markers of Aging Washington, DC: US
Department of Health and Human Services; . Apr. 1982:p. 177-187.(NIH Publication 182-2221).
52. Kannel WH, H. . Vital capacity as a biomarker of aging. In: Reff, ME.; Schneider, EL., editors.
Biological Markers of Aging Washington, DC: US Department of Health and Human Services;
Apr. 1982: p. 145-160.(NIH Publication 182-2221).
53. Eckert DJ WD, Jordan AS, Malhotra A, Wellman A. . Defining phenotypic causes of obstructive
sleep apnea. Identification of novel therapeutic targets. . American journal of respiratory and
critical care medicine. 2013(188):996–1004. .
54. Wellman A ED, Jordan AS, Edwards BA, Passaglia CL, Jackson AC, Gautam S, Owens RL,
Malhotra A, White DP. . A method for measuring and modeling the physiological traits causing
obstructive sleep apnea. . J Appl Physiol 2011(110):1627–1637. .
215
55. Dhar S, Shastri SR, Lenora RA. Aging and the respiratory system. The Medical clinics of North
America. 1976;60(6):1121-1139.
56. Morris JF, Koski A, Johnson LC. Spirometric standards for healthy nonsmoking adults. The
American review of respiratory disease. 1971;103(1):57-67.
57. Smith WD, Cunningham DA, Patterson DH, Rechnitzer PA, Koval JJ. Forced expiratory volume,
height, and demispan in Canadian men and women aged 55-86. Journal of gerontology.
1992;47(2):M40-44.
58. Biessels GJ, Staekenborg S, Brunner E, Brayne C, Scheltens P. Risk of dementia in diabetes
mellitus: a systematic review. Lancet neurology. 2006;5(1):64-74.
59. Justin BN, Turek M, Hakim AM. Heart disease as a risk factor for dementia. Clinical
epidemiology. 2013;5:135-145.
60. Hla KM, Skatrud JB, Finn L, Palta M, Young T. The effect of correction of sleep-disordered
breathing on BP in untreated hypertension. Chest. 2002;122(4):1125-1132.
61. Hla KM, Young T, Finn L, Peppard PE, Szklo-Coxe M, Stubbs M. Longitudinal association of
sleep-disordered breathing and nondipping of nocturnal blood pressure in the Wisconsin Sleep
Cohort Study. Sleep. 2008;31(6):795-800.
62. Barone DA, Krieger AC. Stroke and obstructive sleep apnea: a review. Current atherosclerosis
reports. 2013;15(7):334.
63. Li M, Hou WS, Zhang XW, Tang ZY. Obstructive sleep apnea and risk of stroke: a meta-analysis
of prospective studies. International journal of cardiology. 2014;172(2):466-469.
64. Lipford MC, Flemming KD, Calvin AD, et al. Associations between Cardioembolic Stroke and
Obstructive Sleep Apnea. Sleep. 2015;38(11):1699-1705.
65. Einhorn D, Stewart DA, Erman MK, Gordon N, Philis-Tsimikas A, Casal E. Prevalence of sleep
apnea in a population of adults with type 2 diabetes mellitus. Endocrine practice : official journal
of the American College of Endocrinology and the American Association of Clinical
Endocrinologists. 2007;13(4):355-362.
216
66. Kent BD, McNicholas WT, Ryan S. Insulin resistance, glucose intolerance and diabetes mellitus
in obstructive sleep apnoea. Journal of thoracic disease. 2015;7(8):1343-1357.
67. Meslier N, Gagnadoux F, Giraud P, et al. Impaired glucose-insulin metabolism in males with
obstructive sleep apnoea syndrome. The European respiratory journal. 2003;22(1):156-160.
68. Wang X, Bi Y, Zhang Q, Pan F. Obstructive sleep apnoea and the risk of type 2 diabetes: a meta-
analysis of prospective cohort studies. Respirology (Carlton, Vic). 2013;18(1):140-146.
69. Haas DC, Foster GL, Nieto FJ, et al. Age-dependent associations between sleep-disordered
breathing and hypertension: importance of discriminating between systolic/diastolic hypertension
and isolated systolic hypertension in the Sleep Heart Health Study. Circulation. 2005;111(5):614-
621.
70. Newman AB, Nieto FJ, Guidry U, et al. Relation of sleep-disordered breathing to cardiovascular
disease risk factors: the Sleep Heart Health Study. American journal of epidemiology.
2001;154(1):50-59.
71. Nieto FJ, Herrington DM, Redline S, Benjamin EJ, Robbins JA. Sleep apnea and markers of
vascular endothelial function in a large community sample of older adults. American journal of
respiratory and critical care medicine. 2004;169(3):354-360.
72. Nieto FJ, Young TB, Lind BK, et al. Association of sleep-disordered breathing, sleep apnea, and
hypertension in a large community-based study. Sleep Heart Health Study. JAMA : the journal of
the American Medical Association. 2000;283(14):1829-1836.
73. Shahar E, Whitney CW, Redline S, et al. Sleep-disordered breathing and cardiovascular disease:
cross-sectional results of the Sleep Heart Health Study. American journal of respiratory and
critical care medicine. 2001;163(1):19-25.
74. Beebe DW, Groesz L, Wells C, Nichols A, McGee K. The neuropsychological effects of
obstructive sleep apnea: a meta-analysis of norm-referenced and case-controlled data. Sleep.
2003;26(3):298-307.
217
75. Bucks RS, Olaithe M, Eastwood P. Neurocognitive function in obstructive sleep apnoea: a meta-
review. Respirology (Carlton, Vic). 2013;18(1):61-70.
76. Wallace A, Bucks RS. Memory and obstructive sleep apnea: a meta-analysis. Sleep.
2013;36(2):203-220.
77. Martinez-Garcia MA, Campos-Rodriguez F, Soler-Cataluna JJ, Catalan-Serra P, Roman-Sanchez
P, Montserrat JM. Increased incidence of nonfatal cardiovascular events in stroke patients with
sleep apnoea: effect of CPAP treatment. The European respiratory journal. 2012;39(4):906-912.
78. Martinez-Garcia MA, Soler-Cataluna JJ, Ejarque-Martinez L, et al. Continuous positive airway
pressure treatment reduces mortality in patients with ischemic stroke and obstructive sleep apnea:
a 5-year follow-up study. American journal of respiratory and critical care medicine.
2009;180(1):36-41.
79. Munoz R, Duran-Cantolla J, Martinez-Vila E, et al. Severe sleep apnea and risk of ischemic
stroke in the elderly. Stroke. 2006;37(9):2317-2321.
80. Gottlieb DJ, Yenokyan G, Newman AB, et al. Prospective study of obstructive sleep apnea and
incident coronary heart disease and heart failure: the sleep heart health study. Circulation.
2010;122(4):352-360.
81. Arli B, Bilen S, Titiz AP, et al. Comparison of Cognitive Functions Between Obstructive Sleep
Apnea Syndrome and Simple Snoring Patients: OSAS May Be a Modifiable Risk Factor for
Cognitive Decline. Applied neuropsychology Adult. 2015;22(4):282-286.
82. Daulatzai MA. Evidence of neurodegeneration in obstructive sleep apnea: Relationship between
obstructive sleep apnea and cognitive dysfunction in the elderly. Journal of neuroscience
research. 2015;93(12):1778-1794.
83. Lutsey PL, Bengtson LG, Punjabi NM, et al. Obstructive Sleep Apnea and 15-Year Cognitive
Decline: The Atherosclerosis Risk in Communities (ARIC) Study. Sleep. 2016;39(2):309-316.
84. Osorio RS, Pirraglia E, Aguera-Ortiz LF, et al. Greater risk of Alzheimer's disease in older adults
with insomnia. Journal of the American Geriatrics Society. 2011;59(3):559-562.
218
85. Emamian F, Khazaie H, Tahmasian M, et al. The Association Between Obstructive Sleep Apnea
and Alzheimer's Disease: A Meta-Analysis Perspective. Frontiers in aging neuroscience.
2016;8:78.
86. Rosenzweig I, Glasser M, Polsek D, Leschziner GD, Williams SC, Morrell MJ. Sleep apnoea and
the brain: a complex relationship. The Lancet Respiratory medicine. 2015;3(5):404-414.
87. Davies CR, Harrington JJ. Impact of Obstructive Sleep Apnea on Neurocognitive Function and
Impact of Continuous Positive Air Pressure. Sleep medicine clinics. 2016;11(3):287-298.
88. Kerner NA, Roose SP. Obstructive Sleep Apnea is Linked to Depression and Cognitive
Impairment: Evidence and Potential Mechanisms. The American journal of geriatric psychiatry :
official journal of the American Association for Geriatric Psychiatry. 2016;24(6):496-508.
89. Shastri A, Bangar S, Holmes J. Obstructive sleep apnoea and dementia: is there a link?
International journal of geriatric psychiatry. 2016;31(4):400-405.
90. Aloia MS, Arnedt JT, Davis JD, Riggs RL, Byrd D. Neuropsychological sequelae of obstructive
sleep apnea-hypopnea syndrome: a critical review. Journal of the International
Neuropsychological Society : JINS. 2004;10(5):772-785.
91. Kilpinen R, Saunamaki T, Jehkonen M. Information processing speed in obstructive sleep apnea
syndrome: a review. Acta neurologica Scandinavica. 2014;129(4):209-218.
92. Cross N, Lampit A, Pye J, Grunstein RR, Marshall N, Naismith SL. Is Obstructive Sleep Apnoea
Related to Neuropsychological Function in Healthy Older Adults? A Systematic Review and
Meta-Analysis. Neuropsychology review. 2017.
93. Moher D, Liberati A, Tetzlaff J, Altman DG, Grp P. Preferred Reporting Items for Systematic
Reviews and Meta-Analyses: The PRISMA Statement. Journal of Clinical Epidemiology.
2009;62(10):1006-1012.
94. Wells G SB, O’Connell D, Peterson J,Welch V, Losos M, et al. . The Newcastle-Ottawa scale
(NOS) for assessing the quality of nonrandomised studies in metaanalyses,.
http://wwwohrica/programs/clinical_epidemiology/oxfordasp/; 2011 [accessed 112514]. 2011.
219
95. Bawden FC, Oliveira CA, Caramelli P. Impact of obstructive sleep apnea on cognitive
performance. Arquivos de neuro-psiquiatria. 2011;69(4):585-589.
96. Canessa N, Castronovo V, Cappa SF, et al. Obstructive sleep apnea: brain structural changes and
neurocognitive function before and after treatment. American journal of respiratory and critical
care medicine. 2011;183(10):1419-1426.
97. Castronovo V, Canessa N, Strambi LF, et al. Brain activation changes before and after PAP
treatment in obstructive sleep apnea. Sleep. 2009;32(9):1161-1172.
98. Castronovo V, Scifo P, Castellano A, et al. White matter integrity in obstructive sleep apnea
before and after treatment. Sleep. 2014;37(9):1465-1475.
99. Ferini-Strambi L, Baietto C, Di Gioia MR, et al. Cognitive dysfunction in patients with
obstructive sleep apnea (OSA): partial reversibility after continuous positive airway pressure
(CPAP). Brain research bulletin. 2003;61(1):87-92.
100. Kloepfer C, Riemann D, Nofzinger EA, et al. Memory before and after sleep in patients with
moderate obstructive sleep apnea. Journal of clinical sleep medicine : JCSM : official publication
of the American Academy of Sleep Medicine. 2009;5(6):540-548.
101. Naegele B, Launois SH, Mazza S, Feuerstein C, Pepin JL, Levy P. Which memory processes are
affected in patients with obstructive sleep apnea? An evaluation of 3 types of memory. Sleep.
2006;29(4):533-544.
102. Naegele B, Thouvard V, Pepin JL, et al. Deficits of cognitive executive functions in patients with
sleep apnea syndrome. Sleep. 1995;18(1):43-52.
103. Hrubos-Strom H, Nordhus IH, Einvik G, et al. Obstructive sleep apnea, verbal memory, and
executive function in a community-based high-risk population identified by the Berlin
Questionnaire Akershus Sleep Apnea Project. Sleep & breathing = Schlaf & Atmung.
2012;16(1):223-231.
104. Saunamaki T, Himanen SL, Polo O, Jehkonen M. Executive dysfunction in patients with
obstructive sleep apnea syndrome. European neurology. 2009;62(4):237-242.
220
105. Alchanatis M, Zias N, Deligiorgis N, et al. Comparison of cognitive performance among different
age groups in patients with obstructive sleep apnea. Sleep & breathing = Schlaf & Atmung.
2008;12(1):17-24.
106. Kim H, Dinges DF, Young T. Sleep-disordered breathing and psychomotor vigilance in a
community-based sample. Sleep. 2007;30(10):1309-1316.
107. Verstraeten E, Cluydts R, Pevernagie D, Hoffmann G. Executive function in sleep apnea:
controlling for attentional capacity in assessing executive attention. Sleep. 2004;27(4):685-693.
108. Cheshire K, Engleman H, Deary I, Shapiro C, Douglas NJ. Factors impairing daytime
performance in patients with sleep apnea/hypopnea syndrome. Archives of internal medicine.
1992;152(3):538-541.
109. Telakivi T, Kajaste S, Partinen M, Koskenvuo M, Salmi T, Kaprio J. Cognitive function in
middle-aged snorers and controls: role of excessive daytime somnolence and sleep-related
hypoxic events. Sleep. 1988;11(5):454-462.
110. Valencia-Flores M, Bliwise DL, Guilleminault C, Cilveti R, Clerk A. Cognitive function in
patients with sleep apnea after acute nocturnal nasal continuous positive airway pressure (CPAP)
treatment: sleepiness and hypoxemia effects. Journal of clinical and experimental
neuropsychology. 1996;18(2):197-210.
111. Findley LJ, Barth JT, Powers DC, Wilhoit SC, Boyd DG, Suratt PM. Cognitive impairment in
patients with obstructive sleep apnea and associated hypoxemia. Chest. 1986;90(5):686-690.
112. Greenberg GD, Watson RK, Deptula D. Neuropsychological dysfunction in sleep apnea. Sleep.
1987;10(3):254-262.
113. Roehrs T, Merrion M, Pedrosi B, Stepanski E, Zorick F, Roth T. Neuropsychological function in
obstructive sleep apnea syndrome (OSAS) compared to chronic obstructive pulmonary disease
(COPD). Sleep. 1995;18(5):382-388.
114. Bedard MA, Montplaisir J, Richer F, Malo J. Nocturnal hypoxemia as a determinant of vigilance
impairment in sleep apnea syndrome. Chest. 1991;100(2):367-370.
221
115. Bedard MA, Montplaisir J, Richer F, Rouleau I, Malo J. Obstructive sleep apnea syndrome:
pathogenesis of neuropsychological deficits. Journal of clinical and experimental
neuropsychology. 1991;13(6):950-964.
116. Bliwise DL. Sleep apnea and cognitive function: where do we stand now? Sleep. 1993;16(8
Suppl):S72-73.
117. Saunamaki T, Himanen SL, Polo O, Jehkonen M. Executive dysfunction and learning effect after
continuous positive airway pressure treatment in patients with obstructive sleep apnea syndrome.
European neurology. 2010;63(4):215-220.
118. Salorio CF, White DA, Piccirillo J, Duntley SP, Uhles ML. Learning, memory, and executive
control in individuals with obstructive sleep apnea syndrome. Journal of clinical and
experimental neuropsychology. 2002;24(1):93-100.
119. Torelli F, Moscufo N, Garreffa G, et al. Cognitive profile and brain morphological changes in
obstructive sleep apnea. Neuroimage. 2011;54(2):787-793.
120. Boland LL, Shahar E, Iber C, Knopman DS, Kuo TF, Nieto FJ. Measures of cognitive function in
persons with varying degrees of sleep-disordered breathing: the Sleep Heart Health Study.
Journal of sleep research. 2002;11(3):265-272.
121. Foley DJ, Masaki K, White L, Larkin EK, Monjan A, Redline S. Sleep-disordered breathing and
cognitive impairment in elderly Japanese-American men. Sleep. 2003;26(5):596-599.
122. Phillips BA, Berry DT, Schmitt FA, Magan LK, Gerhardstein DC, Cook YR. Sleep-disordered
breathing in the healthy elderly. Clinically significant? Chest. 1992;101(2):345-349.
123. Sforza E, Roche F, Thomas-Anterion C, et al. Cognitive function and sleep related breathing
disorders in a healthy elderly population: the SYNAPSE study. Sleep. 2010;33(4):515-521.
124. Dlugaj M, Weinreich G, Weimar C, et al. Sleep-disordered breathing, sleep quality, and mild
cognitive impairment in the general population. Journal of Alzheimer's disease : JAD.
2014;41(2):479-497.
222
125. Aloia MS, Ilniczky N, Di Dio P, Perlis ML, Greenblatt DW, Giles DE. Neuropsychological
changes and treatment compliance in older adults with sleep apnea. Journal of psychosomatic
research. 2003;54(1):71-76.
126. Ju G, Yoon IY, Lee SD, Kim TH, Choe JY, Kim KW. Effects of sleep apnea syndrome on
delayed memory and executive function in elderly adults. Journal of the American Geriatrics
Society. 2012;60(6):1099-1103.
127. Berry DT, Phillips BA, Cook YR, et al. Geriatric sleep apnea syndrome: a preliminary
description. Journal of gerontology. 1990;45(5):M169-174.
128. Hayward L, Mant A, Eyland A, et al. Sleep disordered breathing and cognitive function in a
retirement village population. Age and ageing. 1992;21(2):121-128.
129. Yesavage J, Bliwise D, Guilleminault C, Carskadon M, Dement W. Preliminary communication:
intellectual deficit and sleep-related respiratory disturbance in the elderly. Sleep. 1985;8(1):30-33.
130. Blackwell T, Yaffe K, Ancoli-Israel S, et al. Associations between sleep architecture and sleep-
disordered breathing and cognition in older community-dwelling men: the Osteoporotic Fractures
in Men Sleep Study. Journal of the American Geriatrics Society. 2011;59(12):2217-2225.
131. Terpening Z, Lewis SJ, Yee BJ, Grunstein RR, Hickie IB, Naismith SL. Association between
Sleep-Disordered Breathing and Neuropsychological Performance in Older Adults with Mild
Cognitive Impairment. Journal of Alzheimer's disease : JAD. 2015;46(1):157-165.
132. Kim SJ, Lee JH, Lee DY, Jhoo JH, Woo JI. Neurocognitive dysfunction associated with sleep
quality and sleep apnea in patients with mild cognitive impairment. The American journal of
geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry.
2011;19(4):374-381.
133. Spira AP, Blackwell T, Stone KL, et al. Sleep-disordered breathing and cognition in older
women. Journal of the American Geriatrics Society. 2008;56(1):45-50.
223
134. Hahn EA, Wang HX, Andel R, Fratiglioni L. A change in sleep pattern may predict Alzheimer
disease. The American journal of geriatric psychiatry : official journal of the American
Association for Geriatric Psychiatry. 2014;22(11):1262-1271.
135. Lichtenberg PA, Ross T, Millis SR, Manning CA. The relationship between depression and
cognition in older adults: a cross-validation study. The journals of gerontology Series B,
Psychological sciences and social sciences. 1995;50(1):P25-p32.
136. Hoch CC, Reynolds CF, 3rd, Kupfer DJ, Houck PR, Berman SR, Stack JA. Sleep-disordered
breathing in normal and pathologic aging. The Journal of clinical psychiatry. 1986;47(10):499-
503.
137. Reynolds CF, 3rd, Kupfer DJ, Taska LS, et al. Sleep apnea in Alzheimer's dementia: correlation
with mental deterioration. The Journal of clinical psychiatry. 1985;46(7):257-261.
138. Yun CH, Lee HY, Lee SK, et al. Amyloid Burden in Obstructive Sleep Apnea. Journal of
Alzheimer's disease : JAD. 2017.
139. Bu XL, Liu YH, Wang QH, et al. Serum amyloid-beta levels are increased in patients with
obstructive sleep apnea syndrome. Scientific reports. 2015;5:13917.
140. Osorio RS, Ayappa I, Mantua J, et al. The interaction between sleep-disordered breathing and
apolipoprotein E genotype on cerebrospinal fluid biomarkers for Alzheimer's disease in
cognitively normal elderly individuals. Neurobiol Aging. 2014;35(6):1318-1324.
141. Liguori C, Mercuri NB, Izzi F, et al. Obstructive Sleep Apnea is Associated With Early but
Possibly Modifiable Alzheimer's Disease Biomarkers Changes. Sleep. 2017;40(5).
142. Lutsey PL, Norby FL, Gottesman RF, et al. Sleep Apnea, Sleep Duration and Brain MRI Markers
of Cerebral Vascular Disease and Alzheimer's Disease: The Atherosclerosis Risk in Communities
Study (ARIC). PLoS One. 2016;11(7):e0158758.
143. Elias PK, Elias MF, D'Agostino RB, et al. NIDDM and blood pressure as risk factors for poor
cognitive performance. The Framingham Study. Diabetes care. 1997;20(9):1388-1395.
224
144. Gregg EW, Yaffe K, Cauley JA, et al. Is diabetes associated with cognitive impairment and
cognitive decline among older women? Study of Osteoporotic Fractures Research Group.
Archives of internal medicine. 2000;160(2):174-180.
145. Rawlings AM, Sharrett AR, Schneider AL, et al. Diabetes in midlife and cognitive change over
20 years: a cohort study. Annals of internal medicine. 2014;161(11):785-793.
146. Gottesman RF, Schneider AL, Albert M, et al. Midlife hypertension and 20-year cognitive
change: the atherosclerosis risk in communities neurocognitive study. JAMA neurology.
2014;71(10):1218-1227.
147. Kantarci K, Weigand SD, Przybelski SA, et al. Risk of dementia in MCI: combined effect of
cerebrovascular disease, volumetric MRI, and 1H MRS. Neurology. 2009;72(17):1519-1525.
148. Pathan SS, Gottesman RF, Mosley TH, Knopman DS, Sharrett AR, Alonso A. Association of
lung function with cognitive decline and dementia: the Atherosclerosis Risk in Communities
(ARIC) Study. Eur J Neurol. 2011;18(6):888-898.
149. Staessen JA, Richart T, Birkenhager WH. Less atherosclerosis and lower blood pressure for a
meaningful life perspective with more brain. Hypertension. 2007;49(3):389-400.
150. Alonso A, Mosley TH, Jr., Gottesman RF, Catellier D, Sharrett AR, Coresh J. Risk of dementia
hospitalisation associated with cardiovascular risk factors in midlife and older age: the
Atherosclerosis Risk in Communities (ARIC) study. Journal of neurology, neurosurgery, and
psychiatry. 2009;80(11):1194-1201.
151. Aldrich M, Eiser A, Lee M, Shipley JE. Effects of continuous positive airway pressure on phasic
events of REM sleep in patients with obstructive sleep apnea. Sleep. 1989;12(5):413-419.
152. Issa FG, Sullivan CE. The immediate effects of nasal continuous positive airway pressure
treatment on sleep pattern in patients with obstructive sleep apnea syndrome.
Electroencephalography and clinical neurophysiology. 1986;63(1):10-17.
225
153. Loredo JS, Ancoli-Israel S, Kim EJ, Lim WJ, Dimsdale JE. Effect of continuous positive airway
pressure versus supplemental oxygen on sleep quality in obstructive sleep apnea: a placebo-
CPAP-controlled study. Sleep. 2006;29(4):564-571.
154. Verma A, Radtke RA, VanLandingham KE, King JH, Husain AM. Slow wave sleep rebound and
REM rebound following the first night of treatment with CPAP for sleep apnea: correlation with
subjective improvement in sleep quality. Sleep medicine. 2001;2(3):215-223.
155. Cooke JR, Ancoli-Israel S, Liu L, et al. Continuous positive airway pressure deepens sleep in
patients with Alzheimer's disease and obstructive sleep apnea. Sleep medicine. 2009;10(10):1101-
1106.
156. Cooke JR, Ayalon L, Palmer BW, et al. Sustained use of CPAP slows deterioration of cognition,
sleep, and mood in patients with Alzheimer's disease and obstructive sleep apnea: a preliminary
study. Journal of clinical sleep medicine : JCSM : official publication of the American Academy
of Sleep Medicine. 2009;5(4):305-309.
157. Ancoli-Israel S, Palmer BW, Cooke JR, et al. Cognitive effects of treating obstructive sleep apnea
in Alzheimer's disease: a randomized controlled study. Journal of the American Geriatrics
Society. 2008;56(11):2076-2081.
158. Chong MS, Ayalon L, Marler M, et al. Continuous positive airway pressure reduces subjective
daytime sleepiness in patients with mild to moderate Alzheimer's disease with sleep disordered
breathing. Journal of the American Geriatrics Society. 2006;54(5):777-781.
159. Harris P, Fernandez Suarez, M., Surace, E. I., Chrem Mendez, P., Martin, M.E., Clarens, M.F.,et
al. Cognitive reserve and Ab1-42 in mild cognitive impairment (Argentina-Alzheimer’sdisease
neuroimaging initiative). Neuropsychiatr DisTreat 2015(11):2599–2604.
160. Smallwood RG, Vitiello MV, Giblin EC, Prinz PN. Sleep apnea: relationship to age, sex, and
Alzheimer's dementia. Sleep. 1983;6(1):16-22.
161. Franklin KA, Lindberg E. Obstructive sleep apnea is a common disorder in the population-a
review on the epidemiology of sleep apnea. Journal of thoracic disease. 2015;7(8):1311-1322.
226
162. Moraes W, Poyares D, Sukys-Claudino L, Guilleminault C, Tufik S. Donepezil improves
obstructive sleep apnea in Alzheimer disease: a double-blind, placebo-controlled study. Chest.
2008;133(3):677-683.
163. Sukys-Claudino L, Moraes W, Guilleminault C, Tufik S, Poyares D. Beneficial effect of
donepezil on obstructive sleep apnea: a double-blind, placebo-controlled clinical trial. Sleep
medicine. 2012;13(3):290-296.
164. Pan W, Kastin AJ. Can sleep apnea cause Alzheimer's disease? Neuroscience and biobehavioral
reviews. 2014;47:656-669.
165. Spira AP, Yager C, Brandt J, et al. Objectively Measured Sleep and beta-amyloid Burden in
Older Adults: A Pilot Study. SAGE open medicine. 2014;2.
166. Edinger JD, Glenn DM, Bastian LA, Marsh GR. Slow-wave sleep and waking cognitive
performance II: Findings among middle-aged adults with and without insomnia complaints.
Physiology & behavior. 2000;70(1-2):127-134.
167. Kronholm E. Sleep in cognitive life-time trajectory. Sleep medicine. 2012;13(7):777-778.
168. McCrae CS, Vatthauer KE, Dzierzewski JM, Marsiske M. Habitual Sleep, Reasoning, and
Processing Speed in Older Adults with Sleep Complaints. Cognitive therapy and research.
2012;36(2):156-164.
169. Pace-Schott EF, Spencer RM. Age-related changes in the cognitive function of sleep. Progress in
brain research. 2011;191:75-89.
170. Pace-Schott EF, Spencer RM. Age-related changes in consolidation of perceptual and muscle-
based learning of motor skills. Frontiers in aging neuroscience. 2013;5:83.
171. Scullin MK. Sleep, memory, and aging: the link between slow-wave sleep and episodic memory
changes from younger to older adults. Psychology and aging. 2013;28(1):105-114.
172. Scullin MK, Bliwise DL. Sleep, cognition, and normal aging: integrating a half century of
multidisciplinary research. Perspectives on psychological science : a journal of the Association
for Psychological Science. 2015;10(1):97-137.
227
173. Grady C. The cognitive neuroscience of ageing. Nature reviews Neuroscience. 2012;13(7):491-
505.
174. McEwen BS. Sleep deprivation as a neurobiologic and physiologic stressor: Allostasis and
allostatic load. Metabolism: clinical and experimental. 2006;55(10 Suppl 2):S20-23.
175. Vgontzas AN, Zoumakis E, Bixler EO, et al. Adverse effects of modest sleep restriction on
sleepiness, performance, and inflammatory cytokines. The Journal of clinical endocrinology and
metabolism. 2004;89(5):2119-2126.
176. McEwen BS, Sapolsky RM. Stress and cognitive function. Current opinion in neurobiology.
1995;5(2):205-216.
177. Kang JE, Lim MM, Bateman RJ, et al. Amyloid-beta dynamics are regulated by orexin and the
sleep-wake cycle. Science (New York, NY). 2009;326(5955):1005-1007.
178. Xie L, Kang H, Xu Q, et al. Sleep drives metabolite clearance from the adult brain. Science (New
York, NY). 2013;342(6156):373-377.
179. Naidoo N, Ferber M, Master M, Zhu Y, Pack AI. Aging impairs the unfolded protein response to
sleep deprivation and leads to proapoptotic signaling. The Journal of neuroscience : the official
journal of the Society for Neuroscience. 2008;28(26):6539-6548.
180. Borlikova GG, Trejo M, Mably AJ, et al. Alzheimer brain-derived amyloid beta-protein impairs
synaptic remodeling and memory consolidation. Neurobiol Aging. 2013;34(5):1315-1327.
181. Freir DB, Fedriani R, Scully D, et al. Abeta oligomers inhibit synapse remodelling necessary for
memory consolidation. Neurobiol Aging. 2011;32(12):2211-2218.
182. Hita-Yanez E, Atienza M, Gil-Neciga E, Cantero JL. Disturbed sleep patterns in elders with mild
cognitive impairment: the role of memory decline and ApoE epsilon4 genotype. Current
Alzheimer research. 2012;9(3):290-297.
183. Lau EY, Eskes GA, Morrison DL, Rajda M, Spurr KF. Executive function in patients with
obstructive sleep apnea treated with continuous positive airway pressure. Journal of the
International Neuropsychological Society : JINS. 2010;16(6):1077-1088.
228
184. Aviles-Reyes RX, Angelo MF, Villarreal A, Rios H, Lazarowski A, Ramos AJ. Intermittent
hypoxia during sleep induces reactive gliosis and limited neuronal death in rats: implications for
sleep apnea. Journal of neurochemistry. 2010;112(4):854-869.
185. Daulatzai MA. Death by a thousand cuts in Alzheimer's disease: hypoxia--the prodrome.
Neurotoxicity research. 2013;24(2):216-243.
186. Daulatzai MA. Neurotoxic saboteurs: straws that break the hippo's (hippocampus) back drive
cognitive impairment and Alzheimer's Disease. Neurotoxicity research. 2013;24(3):407-459.
187. Drummond SP, Brown GG, Stricker JL, Buxton RB, Wong EC, Gillin JC. Sleep deprivation-
induced reduction in cortical functional response to serial subtraction. Neuroreport.
1999;10(18):3745-3748.
188. Nilsson JP, Soderstrom M, Karlsson AU, et al. Less effective executive functioning after one
night's sleep deprivation. Journal of sleep research. 2005;14(1):1-6.
189. Maillet D, Rajah MN. Association between prefrontal activity and volume change in prefrontal
and medial temporal lobes in aging and dementia: a review. Ageing research reviews.
2013;12(2):479-489.
190. Fandakova Y, Lindenberger U, Shing YL. Deficits in process-specific prefrontal and
hippocampal activations contribute to adult age differences in episodic memory interference.
Cerebral cortex (New York, NY : 1991). 2014;24(7):1832-1844.
191. Mander BA, Rao V, Lu B, et al. Impaired prefrontal sleep spindle regulation of hippocampal-
dependent learning in older adults. Cerebral cortex (New York, NY : 1991). 2014;24(12):3301-
3309.
192. Drager LF, Jun JC, Polotsky VY. Metabolic consequences of intermittent hypoxia: relevance to
obstructive sleep apnea. Best practice & research Clinical endocrinology & metabolism.
2010;24(5):843-851.
229
193. Ayalon L, Ancoli-Israel S, Aka AA, McKenna BS, Drummond SP. Relationship between
obstructive sleep apnea severity and brain activation during a sustained attention task. Sleep.
2009;32(3):373-381.
194. Beebe DW, Gozal D. Obstructive sleep apnea and the prefrontal cortex: towards a comprehensive
model linking nocturnal upper airway obstruction to daytime cognitive and behavioral deficits.
Journal of sleep research. 2002;11(1):1-16.
195. Jackson ML, Howard ME, Barnes M. Cognition and daytime functioning in sleep-related
breathing disorders. Progress in brain research. 2011;190:53-68.
196. Mander BA, Rao V, Lu B, et al. Prefrontal atrophy, disrupted NREM slow waves and impaired
hippocampal-dependent memory in aging. Nature neuroscience. 2013;16(3):357-364.
197. Sforza E, Roche F. Sleep apnea syndrome and cognition. Frontiers in neurology. 2012;3:87.
198. Ludemann P, Dziewas R, Soros P, Happe S, Frese A. Axonal polyneuropathy in obstructive sleep
apnoea. Journal of neurology, neurosurgery, and psychiatry. 2001;70(5):685-687.
199. Kiernan TE, Capampangan DJ, Hickey MG, Pearce LA, Aguilar MI. Sleep apnea and white
matter disease in hypertensive patients: a case series. The neurologist. 2011;17(5):289-291.
200. Kumar R, Chavez AS, Macey PM, Woo MA, Yan-Go FL, Harper RM. Altered global and
regional brain mean diffusivity in patients with obstructive sleep apnea. Journal of neuroscience
research. 2012;90(10):2043-2052.
201. Macey KE, Macey PM, Woo MA, et al. Inspiratory loading elicits aberrant fMRI signal changes
in obstructive sleep apnea. Respiratory physiology & neurobiology. 2006;151(1):44-60.
202. Macey PM, Kumar R, Woo MA, Valladares EM, Yan-Go FL, Harper RM. Brain structural
changes in obstructive sleep apnea. Sleep. 2008;31(7):967-977.
203. Morrell MJ, Finn L, Kim H, Peppard PE, Badr MS, Young T. Sleep fragmentation, awake blood
pressure, and sleep-disordered breathing in a population-based study. American journal of
respiratory and critical care medicine. 2000;162(6):2091-2096.
230
204. Peppard PE, Young T, Palta M, Skatrud J. Prospective study of the association between sleep-
disordered breathing and hypertension. The New England journal of medicine.
2000;342(19):1378-1384.
205. Somers VK, White DP, Amin R, et al. Sleep apnea and cardiovascular disease: an American
Heart Association/american College Of Cardiology Foundation Scientific Statement from the
American Heart Association Council for High Blood Pressure Research Professional Education
Committee, Council on Clinical Cardiology, Stroke Council, and Council On Cardiovascular
Nursing. In collaboration with the National Heart, Lung, and Blood Institute National Center on
Sleep Disorders Research (National Institutes of Health). Circulation. 2008;118(10):1080-1111.
206. Young T, Shahar E, Nieto FJ, et al. Predictors of sleep-disordered breathing in community-
dwelling adults: the Sleep Heart Health Study. Archives of internal medicine. 2002;162(8):893-
900.
207. Corfield DR, Meadows GE. Control of cerebral blood flow during sleep and the effects of
hypoxia. Advances in experimental medicine and biology. 2006;588:65-73.
208. Foster GE, Hanly PJ, Ostrowski M, Poulin MJ. Effects of continuous positive airway pressure on
cerebral vascular response to hypoxia in patients with obstructive sleep apnea. American journal
of respiratory and critical care medicine. 2007;175(7):720-725.
209. Furtner M, Staudacher M, Frauscher B, et al. Cerebral vasoreactivity decreases overnight in
severe obstructive sleep apnea syndrome: a study of cerebral hemodynamics. Sleep medicine.
2009;10(8):875-881.
210. Pallayova M, Steele KE, Magnuson TH, et al. Sleep apnea predicts distinct alterations in glucose
homeostasis and biomarkers in obese adults with normal and impaired glucose metabolism.
Cardiovascular diabetology. 2010;9:83.
211. Spiegel K, Knutson K, Leproult R, Tasali E, Van Cauter E. Sleep loss: a novel risk factor for
insulin resistance and Type 2 diabetes. Journal of applied physiology (Bethesda, Md : 1985).
2005;99(5):2008-2019.
231
212. Zizi F, Jean-Louis G, Brown CD, Ogedegbe G, Boutin-Foster C, McFarlane SI. Sleep duration
and the risk of diabetes mellitus: epidemiologic evidence and pathophysiologic insights. Current
diabetes reports. 2010;10(1):43-47.
213. Dobrowolski P, Klisiewicz A, Florczak E, et al. Independent association of obstructive sleep
apnea with left ventricular geometry and systolic function in resistant hypertension: the RESIST-
POL study. Sleep medicine. 2014;15(11):1302-1308.
214. Narkiewicz K, Somers VK. The sympathetic nervous system and obstructive sleep apnea:
implications for hypertension. Journal of hypertension. 1997;15(12 Pt 2):1613-1619.
215. Lurie A. Metabolic disorders associated with obstructive sleep apnea in adults. Advances in
cardiology. 2011;46:67-138.
216. Seetho IW, Wilding JP. Sleep-disordered breathing, type 2 diabetes and the metabolic syndrome.
Chronic respiratory disease. 2014;11(4):257-275.
217. Guglielmotto M, Aragno M, Autelli R, et al. The up-regulation of BACE1 mediated by hypoxia
and ischemic injury: role of oxidative stress and HIF1alpha. Journal of neurochemistry.
2009;108(4):1045-1056.
218. Guglielmotto M, Tamagno E, Danni O. Oxidative stress and hypoxia contribute to Alzheimer's
disease pathogenesis: two sides of the same coin. TheScientificWorldJournal. 2009;9:781-791.
219. Ogunshola OO, Antoniou X. Contribution of hypoxia to Alzheimer's disease: is HIF-1alpha a
mediator of neurodegeneration? Cellular and molecular life sciences : CMLS. 2009;66(22):3555-
3563.
220. Zhang X, Zhou K, Wang R, et al. Hypoxia-inducible factor 1alpha (HIF-1alpha)-mediated
hypoxia increases BACE1 expression and beta-amyloid generation. The Journal of biological
chemistry. 2007;282(15):10873-10880.
221. Daulatzai MA. Dysfunctional nucleus tractus solitarius: its crucial role in promoting
neuropathogenetic cascade of Alzheimer's dementia--a novel hypothesis. Neurochemical
research. 2012;37(4):846-868.
232
222. Daulatzai MA. Quintessential risk factors: their role in promoting cognitive dysfunction and
Alzheimer's disease. Neurochemical research. 2012;37(12):2627-2658.
223. Daulatzai MA. "Boomerang Neuropathology" of Late-Onset Alzheimer's Disease is Shrouded in
Harmful "BDDS": Breathing, Diet, Drinking, and Sleep During Aging. Neurotoxicity research.
2015;28(1):55-93.
224. Daulatzai MA. Olfactory dysfunction: its early temporal relationship and neural correlates in the
pathogenesis of Alzheimer's disease. Journal of neural transmission (Vienna, Austria : 1996).
2015;122(10):1475-1497.
225. Schild L, Reiser G. Oxidative stress is involved in the permeabilization of the inner membrane of
brain mitochondria exposed to hypoxia/reoxygenation and low micromolar Ca2+. The FEBS
journal. 2005;272(14):3593-3601.
226. Zhu Y, Fenik P, Zhan G, et al. Selective loss of catecholaminergic wake active neurons in a
murine sleep apnea model. The Journal of neuroscience : the official journal of the Society for
Neuroscience. 2007;27(37):10060-10071.
233
Table 4.1: Literature Search Terms for the Systematic Review of Obstructive Sleep Apnea, Cognition and Alzheimer’s disease Obstructive Sleep Apnea
Related
Cognition and Alzheimer's
Disease Related
Alzheimer's disease
pathology related
Sleep apnea Alzheimer's disease Dementia Amyloid beta OR Obstructive Sleep Apnea OR Cognitive Decline OR Amyloid beta
deposition/generation OR Obstructive Sleep Apnea Syndrome
OR Cognitive Impairment OR Alzheimer's disease pathogenesis
OR Sleep Disordered Breathing OR Mild Cognitive Impairment (MeSH)
OR Amyloid Precursor Protein
OR OSA OR MCI OR Amyloid cascade hypothesis
OR OSAH OR aMCI OR Inflammation OR Obstructive Sleep Apnea Hypopnea Syndrome
OR Cognition OR P-Tau
OR Respiratory Disturbance Index
OR Alzheimer (MeSH) OR Phosphorylated Tau
OR Apnea Hypopnea Index OR Cognitive function OR Hippocampal atrophy OR Hypoxia OR Neuropsychological
function OR Apolipoprotein E
OR Sleep Apnea Syndromes OR Executive function OR APOE4 OR Sleep Apnea, Obstructive
OR CSF-Tau
OR Sleep Apnea, Central
OR CSF-PTau OR Anoxia
234
Table 4.2: Descriptive Study Characteristics and Main Findings: Obstructive Sleep Apnea, Cognition, Cognitive Decline and Alzheimer’s disease (Cross-sectional studies)
Authors,
Year
Published
Study Design,
Setting (Study
Quality)
Subjects
Cognitive Domains Adjusted
Variables Major Findings Controls OSA+
Gender SDB Severity N Age N Age
Middle Aged (30-60)
Alchantis et al., 2008
Cross-sectional, sleep clinic
41 49 (33-63) 58 49 (32-
65) N/A
AHI: range 31-137 OSAS;
Range=1-7 NC
Reaction time Age, AHI,
SaO2min/mean, T90, BMI
OSA individual’s ≥ 50 had decreased reaction times. This effect was not seen in those <50.
Bawden et al., 2011
Cross-sectional 20 27.6 (8.7) 17 41.4
(13.0) N/A
AHI: 5-15/>15-30/
>30
Global function, attention, memory, executive function
Age, education OSA individuals had significantly impaired cognitive performance
Canessa et al., 2011
Controlled clinical trial
15 42.15 (6.64)
17 44 (7.63) All men AHI: ≥ 30
Memory, executive functions, attention,
constructional abilities, abstract
reasoning, vigilance, ESS
Age, education
▪ At baseline OSA assoc. with
↓neurocognitive function
▪OSA group had improved
neurocognitive function after 3 months
of CPAP ▪OSA assoc. with structural deficits
Castronovo et al., 2009
Controlled clinical trial
14 42.15 (6.64)
14 43.93 (7.78)
All men AHI > 30 Working memory, Brain activation
(fMRI) Age, education
▪OSA associated with over recruitment of
brain regions in working-memory task
▪Decreases in activation after
treatment
235
Castronovo et al., 2014
Prospective clinical study
15 42.15 (6.64)
13 43.23 (7.63)
All men AHI ≥ 30
Global function, memory, attention, vigilance, abstract reasoning, visuo-
spatial, verbal
Age, education
▪After 12 months of CPAP treatment, almost a complete reversal of white
matter abnormalities was seen ▪Significant
improvements in neuropsychological
function
Ferini-Strambi et al.,
2003
Case-control, sleep clinic
23 55.8 (5.4) 23 56.6 (6.1) 40M6F AHI:
Controls: <5, OSA: ≥5-40
Processing speed, language, executive
function, motor learning
Age, education
▪ 15 days of CPAP treatment returned
visuospatial and motor skills to normal ▪CPAP
did not improve executive function/
constructional abilities
Hrubos-Strom et al.,
2012
Cross-sectional, community
N/A N/A 290 48.2
(11.2) 162M 129F
AHI:<15/ ≥15 Memory, Executive
function Age, sex, education
▪Average oxygen saturation was the
indicator of obstructive sleep apnea severity most strongly assoc.
with cognitive function
Kim et al., 2007
Cross-sectional, community
N/A N/A 611 Range: 35-74
346 M 265 W
AHI: median=2.8; range=0-121
Vigilance
BMI, caffeine/alcohol
consumption, smoking status,
stimulants before testing
▪SDB related with diminished vigilant
attention performance
Kloepfer et al., 2009
Cross-sectional 20 47.4(5.6) 15 46.4 (5.9) 22M 13F
Clinical diagnosis:
AHI >5 and questionnaire
Memory, global Age, sex, IQ
▪Moderate OSA assoc. with impairment of
procedural and verbal declarative memory
Mathieu et al., 2008
Cross-sectional, sleep clinic
Younger: 12
Older: 18
Younger: 38.9 (2.2)
Older: 62.5 (1.8)
Younger: 14
Older: 14
Younger: 37.7 (2.0)
Older: 62.3 (2.0)
26 M 2 W
AHI: 51(4) OSAS
younger; 43(4) OSAS
older
Attention, Executive Function, Memory
Age, education
▪No Group-by-Age interaction for
cognitive performance. ▪Main effects for both
Group and Age
236
Naegele et al., 1995
Cross-sectional 17 49 (3) 17 49 (3) All men RDI > 10
Attentional capacity, memory efficiency, long-term memory, frontal lobe function
Age, education, verbal IQ
▪Memory deficits were associated with the
number of apneas and hypopneas per hour of sleep ▪Typical frontal-
lobe related abnormalities were
associated with levels of nocturnal hypoxemia
Naegele et al., 2006
Cross-sectional 95 47.4 95 48 144M46F RDI ≥ 10 Memory Age, education
▪OSA associated with impaired episodic memory, overall
procedural memory performance, specific
working memory capabilities ▪No impairment of maintenance,
recognition, or forgetfulness
Nikodemova et al., 2013
Cross-sectional, community
AHI <5 AHI 5-14 AHI ≥
15 Total
1,081 M F
AHI: <5/5-14/≥ 15
Memory, Executive function
Age, sex, education, BMI
▪No significant association between SDB and cognitive
performance in APOE4- ▪AHI ≥15, APOE4+ associated
with ↓cognitive function
n: 1,148 age: 52.1
(9.9)
n: 399 age: 56.4
(9.6)
n: 298 age: 56.6
n: 1,845 age: 53.9
(10.1) age range:
30-81
Quan et al., 2006
Cross-sectional, community
74 57.4(9.2) 67 59.4 (9.2) 84M 57F
AHI: ≤5
Attention and vigilance, processing
speed, executive function, motor
learning, visuospatial ability, memory
Age, sex, education
▪No significant impact of OSAH and
neuropsychological function ▪↑hypoxemia associated with ↓motor and processing speed
237
Salorio et al., 2002
Cross sectional 24 44.2(8.5) 28 44.0 (7.9) 34M 18F
Hypoxic episodes: 5-
20/21-35/>35 episodes per
hour
Long-term Learning and Memory,
Executive control Sex, education, IQ
▪OSA individuals exhibited poorer recall, less efficient semantic clustering, poorer use
of semantic cues ▪Other than letter
fluency, deficits were not observed in general
executive control
Saunamaki, Himanen, et
al., 2009
Controlled clinical trial
15 44
Range: 30-63
15 50
Range: 37-59
All men AHI:≤5/> 10 WAIS-R Age, education
▪OSAS associated with mild visually based
cognitive dysfunction and reduced amount of
sleep in the right hemisphere even after
CPAP
Saunamaki et al., 2010
Controlled clinical trial
17 44Range:
30-63 20
50 Range: 37-65
All men AHI:≤5/> 10
WAIS-R; Short term memory, working memory, verbal
fluency, visuomotor tracking, visuospatial
organization
Age, education, IQ
▪OSAS showed poorer performance on
executive tests ▪OSAS did not show any improvement on
executive or visuospatial function even after long-term
CPAP treatment
Saunamaki, Jehkonen et
al., 2009
Controlled clinical trial
20 43
Range: 29-63
40 47
Range: 28-65
All men AHI:≤5/> 10
WAIS-R; Verbally-based cognitive
abilities, visually-based cognitive
abilities,
Age, education
▪OSAS had lower set-shifting and
analysis/synthesis performance than controls ▪Based on
normative data, most had normal, while others had more serious deficits
238
Sharma et al., 2010
Cross-sectional 25 45.6 (6.2) 50 43 (7.5) 63M 12F
AHI > 30
Alertness, working memory, response inhibition, problem solving, executive
function
Age, sex, education
▪OSA subjects had significantly impaired performance on tests on alertness, working
memory, response inhibition, problem solving, executive
function ▪Significance disappeared after
adjusting for delayed information processing
Older Adults (61-70)
Aloia et al., 2003
Cross-sectional, clinic
Noncompliant Compliant
N/A
RDI: 51(20) OSAS
compliant; 46(22) OSAS noncompliant
Attention, Executive Function,
Construction, Motor Speed, Memory,
Language
Age, education, sleep apnea
severity
▪↑RDI related to↓ verbal delayed recall
memory ▪↑sleep fragmentation
and hypoxemia associated with↓ verbal delayed recall memory
▪Cognitive benefits with CPAP compliance
6 64.8 (2.6) 6 64.8 (6.4)
Berry et al., 1990
Cross-sectional, sleep clinic
12 68.4 (2.2) 8 68.6(4.8) Men only AHI: 28(12)
OSAS; 3(3)NC
Global, IQ, Memory ▪OSAS: ↓nonverbal IQ and nonverbal memory
delayed recall
Boland et al., 2002
Cross-sectional, community
N/A N/A 1700 62 (range
52-75) 837M923W
RDI median range: .4-23
Attention, Executive Function, Memory
age, education, occupation, field center, diabetes,
hypertension, body-mass index,
CNS meds, alcohol use
▪No relationships
Dlugaj et al., 2014
Cross- sectional, population-based
AHI <15 AHI 15-29 AHI ≥
30 Total
919M 874F
AHI; <15/15-29/ ≥ 30
Memory, executive function
Age, sex, education
▪SDB not associated with MCI or MCI
subtypes (amnestic and non-amnestic)
n: 1323 age:
62.94 (7.23)
n: 327 age: 65.79
(7.41)
n: 143 age:
62.17 (7.31)
n: 1793 age: 63.79
(7.45)
239
Foley et al., 2003
Cross-sectional, community
N/A N/A 718 Range: 79-97
Men only 71% had
AHI≥5; 19% had AHI ≥ 30
Global, Attention, Executive Function,
Construction, Memory, Language
Age, education, marital status
▪No relationships
Ju et al., 2012 Case-control, sleep
clinic 21 68.7 (5.5) 42 68(4.4)
37M 26F
AHI: Controls :< 15. OSA:
<15.3*
General cognitive/intellectual
ability, executive function
Age, sex, education, BMI
▪Significant findings with delayed recall and
executive function
Kim et al., 2011
Cross-sectional, clinic
N/A N/A 30 67.4 (3.8) 42M 18W
AHI: 13(12) MCI;15(14)
NC
Executive function, Language, Memory,
Visuospatial construction
▪↑AHI associated with
language with MCI
Lutsey et al., 2016
Cross-sectional 521 60.7 (5.1) 445 62(4.9) 435M 531W
AHI (<5/5- 14.9/15-
29.9/≥30) Global
Age, sex, field center, education,
alcohol intake, smoking, physical activity, APOE4, BMI, CRP, CVD
comorbidities
▪OSA category and additional indices of
sleep were not associated with change
in any cognitive test
Phillips et al., 1992
Cross-sectional, community
N/A N/A 92 64.2 (8.6) 44M 48W
AHI: 3(4)
Global, IQ, Attention, Executive Function, Memory, Language,
Motor
None ▪No relationship
between AHI and cognition
Sforza et al., 2010
Cross-sectional, community
382
Combined exposure
and controls: 68(1.8)
445
Combined exposure
and controls: 68(1.8)
343M 484W
AHI: 20(15); 53% had AHI≥15
Global, Attention, Executive Function, Memory, Language
gender, BMI, diabetic status, hypertension,
education level, anxiety and
depression scores, ESS, blood
pressure, self-reported sleep
time
▪No relationships
Terpening et al., 2015
Case-control(convenience
sample), community
N/A N/A 40 63.5 (8.9) 19M21F Processing speed ▪Findings significant**
Yesavage et al., 1985
Cross-sectional, sleep clinic
N/A N/A 41 69.5 (6.5) Men only RDI: 26(30);
73% had RDI>5
Attention, Executive Function, Memory
Age, depression, education, typical sleepiness/fatigue
▪RDI associated with attention and executive
function
Elderly (>70)
240
Blackwell et al., 2011
Cross-sectional, population-based
N/A N/A 2,909 76 (6) Only men AHI: <5/5-
14/≥ 30 Global, Executive
Function
Age, race, clinic, BMI, IADL, CVD
comorbidities, antidepressant
use, benzodiazepine use, depressive
symptoms, education, alcohol
use, smoking, physical activity,
self-reported health
▪↓ time in REM, ↑ time in Stage 1 sleep, and ↑nocturnal hypoxemia
are associated with poorer cognition
Hayward et al., 1992
Cross-sectional, community
N/A N/A 96 78 (3.9) 21 M 75 W
RDI: 6(6) Attention, Executive Function, Memory, Language, Motor
Age, education
▪RDI not associated with memory, verbal,
or motor factors ▪RDI associated with
cerebral efficiency factor
Spira et al., 2008
Cross-sectional, community
391 82.7 (3.3) 57 83.6 (4.3) Women
only AHI: <30/≥
30 Global, Executive
Function
Age, Education, SSRI (BMI and
functional impairment were found to not have
a significant impact on results)
▪↑AHI&↑hypoxemia& ↑central apnea associated with
cognition ▪APOE4+ associated
with 5x risk of cognitive impairment
Abbreviations: a, adjusted; Aβ40/42, amyloid beta-40/42; AD, Alzheimer’s disease; AFCFT, alphabetic fluency and category fluency tasks; AHI ≥ 15, apnea hypopnea index of 15 or more events per hour of sleep; APOE, apolipoprotein epsilon4; BVRT, Benton visual retention test; c, crude; CAR, circadian activity rhythms; CASI, cognitive abilities screening instrument; CERAD, consortium to establish a registry for Alzheimer’s disease; CPRS, comprehensive psychopathological rating scale; CSF, cerebrospinal fluid; CVLT, California verbal learning test; DSM-IIIR/IV-TR, diagnostic and statistical manual of mental disorders; third edition/fourth edition, text revised; EDS, Excessive daytime sleepiness; ESS, Epworth sleepiness scale; GBSRT, Grober and Buschke selective reminding test; HAM-D, Hamilton Depression Rating Scale; HR, hazard ratio; ICD-9/10, international classification of diseases ninth/tenth edition AD criteria; IQCODE, informant questionnaire on cognitive decline in the elderly; MMSE, mini mental state examination; MRI, magnetic resonance imaging; N, number of participants; NA, not applicable; N/A, not available; NINCDS-ADRDA, national institute of neurological and communicative disorders and stroke and the Alzheimer’s disease and related disorders association; NINDS-AIREN, national institute of neurological disorders and stroke and association internationale pour la recherche et l’enseignement en neurosciences vascular dementia criteria; OR, odds ratio; PSG, polysomnography; PSQI, Pittsburgh sleep quality index; P-tau, phosphorylated tau; RBANS, repeatable battery for the assessment of neuropsychological status; RR, relative risk; SDB, sleep disordered breathing; SPMSQ, short portable mental status questionnaire; TONI, test of nonverbal intelligence; Trails B, trail making b test; TST, total sleep time; T-tau, total-tau; WAIS-III, Wechsler adult intelligence scale third version; WASO, wake after sleep onset; WHIIRS, women’s health initiative insomnia rating scale; WMS-R, Wechsler memory scale-revision; w/o, without.
241
Table 4.3: Descriptive Study Characteristics and Main Findings: Obstructive Sleep Apnea, Cognition, Cognitive Decline and Alzheimer’s disease (Longitudinal studies – older adults mainly)
Authors,
Year
Published
Study
design,
setting
(study
quality)
Subjects
Cognitive
Domains Adjusted Variables Major Findings Controls OSA+
Gender SDB
Severity N Age N Age
Blackwell et al., 2015
Longitudinal, community
based 1,504
75.8 (5.3)
1,132 76.4 (5.2)
Men only
ODI: <15, ≥15
AHI: <15, ≥15)
Global
Age, site, race, BMI, education, depressive
symptoms, CVD comorbidities, Parkinson's
disease, IADL, benzodiazepine use,
antidepressant use, self-reported health, physical activity, alcohol intake,
smoking
▪No significant association between AHI and cognitive decline in older community-
dwelling men ▪Modest association of
nocturnal hypoxemia with global cognitive decline
Chang et al., 2013
Longitudinal, community
based 7,070
55.5 (4.78)
1,414 55.5
(4.93)
M: 5,034
F: 3,450
Not specified; clinical
diagnosis (according to
the guidelines of the AASM)
Dementia diagnosis
Age, sex, CVD comorbidities, urbanization
level, income
▪ Hazard ratio of dementia is 1.44 times greater for patients
with SA than for the comparison group
▪SA is an age, time, and gender dependent risk factor for
dementia
Cohen-Zion et al., 2001
Longitudinal, Community
N/A N/A 46 80(3) 7M
39 W RDI: 10(9) Global Age, education
▪↑EDS associated with global cognitive decline over 3 years
follow-up ▪Hypoxemia and RDI not associated with cognition
Lutsey et al., 2016
Longitudinal 521 60.7 (5.1)
445 62(4.9) 435M 531W
AHI: <5/5- 14.9/ 15-29.9/ ≥30
Global
Age, sex, field center, education, alcohol intake, smoking, physical activity, APOE4, BMI, CRP, CVD
comorbidities
▪OSA category and additional indices of sleep were not
associated with change in any cognitive test
N/A N/A 599 67.0 (1) Global
242
Martin et al., 2015
Longitudinal, population
based
224M 335W
AHI: <15/ 15-30/ >30)
Age, sex, education, follow-up length, BMI, ESS, CVD
comorbidities, anxiety, depression
▪Abnormal breathing events associated slightly with decline
in attentional domain ▪No association with changes in executive and memory function
Yaffe et al., 2011
Longitudinal, Community
193 82.1(3.2) 105 82.6(3.1) Women
only AHI: ≥15
Global, Attention, Executive Function, Memory
Age, race, BMI, education, smoking, diabetes,
hypertension, antidepressant use, benzodiazepine use,
non-diazepam anxiolytics use
▪SDB: ↑hypoxemia had ↑risk of developing MCI or dementia over 5 year follow-up▪Sleep
fragmentation and duration not associated with cognition
Yaffe et al., 2015
Longitudinal ###### 68.5 13,480 66.9 Men Only
Not specified; clinical
diagnosis
Dementia (classified using
International
Classification of
Disease, Ninth Revision codes)
Age, CVD comorbidities, obesity, depression, income,
education
▪Those with a sleep disturbance had a 27% increased risk for
dementia (HR: 1.27)
Abbreviations: a, adjusted; Aβ40/42, amyloid beta-40/42; AD, Alzheimer’s disease; AFCFT, alphabetic fluency and category fluency tasks; AHI ≥ 15, apnea hypopnea index of 15 or more events per hour of sleep; APOE, apolipoprotein epsilon4; BVRT, Benton visual retention test; c, crude; CAR, circadian activity rhythms; CASI, cognitive abilities screening instrument; CERAD, consortium to establish a registry for Alzheimer’s disease; CPRS, comprehensive psychopathological rating scale; CSF, cerebrospinal fluid; CVLT, California verbal learning test; DSM-IIIR/IV-TR, diagnostic and statistical manual of mental disorders; third edition/fourth edition, text revised; EDS, Excessive daytime sleepiness; ESS, Epworth sleepiness scale; GBSRT, Grober and Buschke selective reminding test; HAM-D, Hamilton Depression Rating Scale; HR, hazard ratio; ICD-9/10, international classification of diseases ninth/tenth edition AD criteria; IQCODE, informant questionnaire on cognitive decline in the elderly; MMSE, mini mental state examination; MRI, magnetic resonance imaging; N, number of participants; NA, not applicable; N/A, not available; NINCDS-ADRDA, national institute of neurological and communicative disorders and stroke and the Alzheimer’s disease and related disorders association; NINDS-AIREN, national institute of neurological disorders and stroke and association internationale pour la recherche et l’enseignement en neurosciences vascular dementia criteria; OR, odds ratio; PSG, polysomnography; PSQI, Pittsburgh sleep quality index; P-tau, phosphorylated tau; RBANS, repeatable battery for the assessment of neuropsychological status; RR, relative risk; SDB, sleep disordered breathing; SPMSQ, short portable mental status questionnaire; TONI, test of nonverbal intelligence; Trails B, trail making b test; TST, total sleep time; T-tau, total-tau; WAIS-III, Wechsler adult intelligence scale third version; WASO, wake after sleep onset; WHIIRS, women’s health initiative insomnia rating scale; WMS-R, Wechsler memory scale-revision; w/o, without.
243
Table 4.4 Descriptive Study Characteristics and Main Findings: Obstructive Sleep Apnea, and Alzheimer’s disease/AD Pathology
Authors, Year
Published
Study
Design
Subjects
OSA
assessment
Alzheimer's disease
assessment Adjusted Variables Major findings
N
Age
(Mean ±
SD)
Gender
OSA and AD
Hoch et al., 1986
Cross-sectional
80 71.5 (8.1) 33 M 57 F
AHI DSM-III None Significant association
Hoch et al., 1989
Cross-sectional
27 74.5 (5.1) 7 M 20 F
AHI NICNDS-ADRDA,
DSM-III N/A
No association between OSA and dementia
Reynolds et al., 1985
Cross-sectional
61 69.7 (6.8) 19 M 42 F
AI, AHI
DSM 3, Hamilton rating, Folstein
score, and a modified Hachinski
Ischemia score
Gender Significant association between
sleep apnea and dementia in women
Reynolds et al.,1987
Cross-sectional
30 73.3 (9.1) 3 M 12 F
24 Chanel polygraphs
DSM 3, Hamilton rating, Folstein
score, and a modified Hachinski
Ischemia score
N/A No association between OSA
and dementia
Smallwood et al., 1983
Cross-sectional
55 Range: 23-
81 years 45 M 10 F
AHI DSM 3, neurological
examination Age, sex
No relationship between
dementia and apnea severity
OSA and AD pathology
Bu et al., 2015 Cross-
sectional 94
43.62 (9.78)
67M 27F
AHI, ODI, MSaO2, LSaO2
Amyloid beta levels Age, sex Significant association between
hypoxia and amyloid levels
Ligouri et al., 2017
Cross-sectional
50 66.96 (7.98)
33M 17F
AHI
CSI classified by subjective memory decline, cognitive
performance
Age, education Significant association between OSA and CSF AD biomarkers
244
Lutsey et al., 2017
Prospective 312 61.7 (5.0) 145M 167F
AHI, SHHS Sleep Habits
Questionnaire
Neurocognitive exam, brain MRI
age, sex, field center, education, physical
activity, ethanol intake, smoking status, leusire time physical activity,
and APOE e4 risk allele, BMI
No relationship between mid-life OSA and dementia
Osorio et al., 2014
Cross-sectional
95 67.6 37M58F AHI Neuropsychological
test battery
Age, BMI, CSF internal batch, time interval
between sleep study and lumbar puncture,ApoE4
status
Significant association between SDB and AD CSF biomarkers
Osorio et al., 2015
Prospective 2,285 Self reported Self-report; diagnosis by
clinician
APOE e4 status, sex, education, BMI,
depression, cardiovascular disease, hypertension, diabetes,
and age
Significant association between SDB and earlier age at
cognitive decline
Yun et al., 2017
Cross-sectional
38 56.7 (4.0) M: 18 F: 20
AHI Neuropsychological
test battery
Age, sex, education, APOE genotype, status
of sleep duration, hypertension, diabetes,
BMI, exercise, depressive mood,
smoking, and alcohol drinking
Significant association between OSA and amyloid deposition
Random Control Trials
Ancoli-Israel et al., 2008
RCT 52 78.2 (7.2) 39M 13F
Rechtschaffen and Kales
criteria
Nueropsychological test battery
None CPAP improved some cognitive functioning
Chong et al., 2006
RCT 39 78.0 (7.04) 29M 10F
RDI NINCDS-ADRDA
criteria None
CPAP reduces sleepiness in those with AD and OSA
Cooke et al., 2009a
RCT 52 77.8 (7.3) 39M 13F
Rechtschaffen and Kales
criteria
NINCDS-ADRDA criteria, MMSE
None After one night of tCPAP use: deeper sleep, affects for three
weeks
Cooke et al., 2009b
RCT 10 75.7 (5.9) 7M 3F
AHI, PSQI, ESS, FOSQ
Neuropsychological test battery
None Sustained CPAP use associated
with less cognitive decline
245
Spira et al., 2014
RCT 13 71.6 (7.8) 7M 6F
AHI, ODI Neuropsychological
tests, GDS, CDR None
SDB severity associated with amyloid deposition
Abbreviations: a, adjusted; Aβ40/42, amyloid beta-40/42; AD, Alzheimer’s disease; AFCFT, alphabetic fluency and category fluency tasks; AHI ≥ 15, apnea hypopnea index of 15 or more events per hour of sleep; APOE, apolipoprotein epsilon4; BVRT, Benton visual retention test; c, crude; CAR, circadian activity rhythms; CASI, cognitive abilities screening instrument; CERAD, consortium to establish a registry for Alzheimer’s disease; CPRS, comprehensive psychopathological rating scale; CSF, cerebrospinal fluid; CVLT, California verbal learning test; DSM-IIIR/IV-TR, diagnostic and statistical manual of mental disorders; third edition/fourth edition, text revised; EDS, Excessive daytime sleepiness; ESS, Epworth sleepiness scale; GBSRT, Grober and Buschke selective reminding test; HAM-D, Hamilton Depression Rating Scale; HR, hazard ratio; ICD-9/10, international classification of diseases ninth/tenth edition AD criteria; IQCODE, informant questionnaire on cognitive decline in the elderly; MMSE, mini mental state examination; MRI, magnetic resonance imaging; N, number of participants; NA, not applicable; N/A, not available; NINCDS-ADRDA, national institute of neurological and communicative disorders and stroke and the Alzheimer’s disease and related disorders association; NINDS-AIREN, national institute of neurological disorders and stroke and association internationale pour la recherche et l’enseignement en neurosciences vascular dementia criteria; OR, odds ratio; PSG, polysomnography; PSQI, Pittsburgh sleep quality index; P-tau, phosphorylated tau; RBANS, repeatable battery for the assessment of neuropsychological status; RR, relative risk; SDB, sleep disordered breathing; SPMSQ, short portable mental status questionnaire; TONI, test of nonverbal intelligence; Trails B, trail making b test; TST, total sleep time; T-tau, total-tau; WAIS-III, Wechsler adult intelligence scale third version; WASO, wake after sleep onset; WHIIRS, women’s health initiative insomnia rating scale; WMS-R, Wechsler memory scale-revision; w/o, without.
246
Figure 4.1: Study retrieval and selection for Obstructive Sleep Apnea, Cognitive Decline and/or Alzheimer’s disease Review
Studies screened for titles (n = 1,688)
Studies excluded (n =784)
Full-text articles assessed for eligibility
(n =193)
Full-text articles excluded, with reasons (n = 141)
- Review article (n = 15) - OSA not assessed (n =
32 ) - Relationship not
studied (n = 40) - Not peer reviewed (n =
15 ) - Alzheimer not
examined (n = 24) - Duplicate studies (n =
15) Studies included in
systematic review (n = 57)
Article identified from reference
search (n = 5)
Studies screened for abstracts
(n = 904)
Studies excluded (n = 711)
Duplicates removed (n =1029)
Studies identified through database searching (n = 2717)
(PubMed=1,112, Embase = 1,164, Web of Science = 285, Psych Info = 103, Cochrane library = 53
Studies after duplicates removed (n = 1,688)
Ide
nti
fica
tio
n
Scr
ee
nin
g
Elig
ibilit
y
Incl
ud
ed
248
249
250
APPENDIX C
IRB LETTER
10/20/2017 Omonigho Bubu, MD, MPH Epidemiology and Biostatistics 4202 East Fowler Ave. Tampa, FL 33612 RE: Not Human Subjects Research Determination IRB#: Pro00032613 Title: Obstructive Sleep Apnea and the risk of Alzheimer’s disease Dear Dr. Bubu: The Institutional Review Board (IRB) has reviewed your application. The activities presented in the application involve methods of program evaluation, quality improvement, and/or needs analysis. While potentially informative to others outside of the university community, study results would not appear to contribute to generalizable knowledge. As such, the activities do not meet the definition of human subject research under USF IRB policy, and USF IRB approval and oversight are therefore not required. While not requiring USF IRB approval and oversight, your study activities should be conducted in a manner that is consistent with the ethical principles of your profession. If the scope of your project changes in the future, please contact the IRB for further guidance. If you will be obtaining consent to conduct your study activities, please remove any references to "research" and do not include the assigned Protocol Number or USF IRB contact information. If your study activities involve collection or use of health information, please note that there may be requirements under the HIPAA Privacy Rule that apply. For further information, please
251
contact a HIPAA Program administrator at (813) 974-5638. Sincerely, E. Verena Jorgensen, M.D., Chairperson USF Institutional Review Board
252
APPENDIX D
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
