NEUROCOGNITIVE AND PSYCHOSOCIAL OUTCOMES IN PATIENTS WITH OBSTRUCTIVE SLEEP APNEA TREATED WITH CONTINUOUS POSITIVE AIRWAY PRESSURE
by
Esther Yuet Ying Lau
Submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy
at
Dalhousie University Halifax, Nova Scotia June 2008
© Copyright by Esther Yuet Ying Lau, 2008 Library and Bibliotheque et 1*1 Archives Canada Archives Canada Published Heritage Direction du Branch Patrimoine de I'edition
395 Wellington Street 395, rue Wellington Ottawa ON K1A0N4 Ottawa ON K1A0N4 Canada Canada
Your file Votre reference ISBN: 978-0-494-43943-2 Our file Notre reference ISBN: 978-0-494-43943-2
NOTICE: AVIS: The author has granted a non L'auteur a accorde une licence non exclusive exclusive license allowing Library permettant a la Bibliotheque et Archives and Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par telecommunication ou par I'lnternet, prefer, telecommunication or on the Internet, distribuer et vendre des theses partout dans loan, distribute and sell theses le monde, a des fins commerciales ou autres, worldwide, for commercial or non sur support microforme, papier, electronique commercial purposes, in microform, et/ou autres formats. paper, electronic and/or any other formats.
The author retains copyright L'auteur conserve la propriete du droit d'auteur ownership and moral rights in et des droits moraux qui protege cette these. this thesis. Neither the thesis Ni la these ni des extraits substantiels de nor substantial extracts from it celle-ci ne doivent etre imprimes ou autrement may be printed or otherwise reproduits sans son autorisation. reproduced without the author's permission.
In compliance with the Canadian Conformement a la loi canadienne Privacy Act some supporting sur la protection de la vie privee, forms may have been removed quelques formulaires secondaires from this thesis. ont ete enleves de cette these.
While these forms may be included Bien que ces formulaires in the document page count, aient inclus dans la pagination, their removal does not represent il n'y aura aucun contenu manquant. any loss of content from the thesis. •*• Canada DALHOUSIE UNIVERSITY
To comply with the Canadian Privacy Act the National Library of Canada has requested that the following pages be removed from this copy of the thesis:
Preliminary Pages Examiners Signature Page (pii) Dalhousie Library Copyright Agreement (piii)
Appendices Copyright Releases (if applicable) lb my husband, %azdin 'Wong for his unfailing Cove and support, and to our Lord, Jesus Christ for Jdis Salvation and for giving me strength. TABLE OF CONTENTS LIST OF TABLES viii LIST OF FIGURES x ABSTRACT xi LIST OF ABBREVIATIONS AND SYMBOLS USED xii ACKNOWLEDGEMENTS xiv CHAPTER ONE: INTRODUCTION 1 OVERVIEW OF OSA 1 Definitions and Prevalence 1 Pathophysiology of OSA 3 Sleep Architecture Associated with OSA 5 Treatments for OSA 5 MEDICAL SEQUELAE OF OSA 6 DAYTIME CONSEQUENCES OF OSA 8 Excessive Daytime Sleepiness and Daily Functioning 8 Mood 12 Quality of Life 17 OSA AND NEUROCOGNITIVE CONSEQUENCES 21 Global Intellectual Functioning 22 Attention and Concentration 23 Motor Functions 24 Memory 26 Executive Function 27 MECHANISMS OF THE NEUROCOGNITIVE DYSFUNCTION 33 NEUROIMAGING FINDINGS ON PATIENTS WITH OSA 43 Prefrontal Cortex (PFC) and Sleep 50 REVERSIBILITY OF DAYTIME DYSFUNCTIONS WITH TREATMENT OF CPAP 52 THEORY OF WORKING MEMORY 61 PURPOSE OF THE PRESENT STUDY 65 CHAPTER TWO: METHODS 68 PARTICIPANTS AND DESIGN 68 OSA Group 68 Controls 69 Design 70 Procedures 71
v MEASURES AND TESTS 72 Screening 72 Hospital Record 74 Clinical and Demographic Variables 74 Sleep Questionnaires (Appendix D) 74 Mood Assessment 76 Functional Outcomes and Quality of Life 77 Polysomnography 79 Neuropsychological Variables 80 Working Memory Components 81
STATISTICAL ANALYSES 92 CHAPTER THREE: RESULTS 95 PARTICIPANTS' CHARACTERISTICS 95
PRE- AND POST-TREATMENT COMPARISONS 95
BETWEEN-GROUP COMPARISONS 97 Polysomnography 97 Working Memory Tasks 97 Neuropsychological Tests 103 Sleep Questionnaires 105 Mood Assessment 105 Daily Functioning and Quality of Life 106 REGRESSION ANALYSES 106 Working Memory Tasks 106 Neuropsychological Tests 107 Mood, Daily Functioning, and Quality of Life 108
CORRELATIONAL ANALYSES 109 CHAPTER FOUR: DISCUSSION 111 EFFECTIVENESS OF CPAP TREATMENT 111 PERFORMANCE OF INDIVIDUALS WITH OSA TREATED WITH CPAP ON NEUROCOGNITIVE MEASURES 116 Basic Storage and Rehearsal Components of Working Memory 117 Central Executive Function of Working Memory 117 Neuropsychological Tests 123 PREDICTORS OF RESIDUAL NEUROCOGNITIVE DEFICITS 127
VI PREDICTORS OF PSYCHOSOCIAL OUTCOMES 130 Emotional Functioning 131 Daily Functioning 132 Quality of Life 133 IMPACT OF EXECUTIVE DIFFICULTIES ON PERCEIVED COGNITIVE EFFICIENCY AND QUALITY OF LIFE 134 FACTORS RELATING TO QUALITY OF LIFE 135 OTHER FINDINGS OF INTERESTS 137 Between-Group Comparison on Sleep Related Variables After Treatment 137 Between-Group Comparison on Emotional Functioning After Treatment 138 Between-Group Comparison on Daily Functioning and Quality of Life 138 Undetected OSA in Community Dwelling Individuals 139 STRENGTHS AND LIMITATIONS OF THE CURRENT STUDY 140 FUTURE RESEARCH DIRECTIONS 144 CHAPTER FIVE: CONCLUSIONS 147 REFERENCES 200 APPENDIX A - SCREENING QUESTIONNAIRES 229 APPENDIX B - MEDICAL RECORD SUMMARY & PRE-TREATMENT SLEEP STUDY FORM 246 APPENDIX C - HEALTH INTERVIEW 248 APPENDIX D - SLEEP AND PSYCHOSOCIAL QUESTIONNAIRES 250 APPENDIX E - NEUROPSYCHOLOGICAL TESTS 267 APPENDIX F - WORKING MEMORY SPAN 277
vn LIST OF TABLES
Table 1 Group Demographic Characteristics 150 Table 2 OSA Group Medical History 150 Table 3 Pre- and Post-treatment Comparisons on Respiratory and Hypoxemia Indices, Sleep Questionnaires, Daytime Function, and Quality of Life 151 Table 4 Comparison between the OSA Group Treated with CPAP and the Control Group on Polysomnographic Variables 152 Table 5 Comparison between the OSA Group Treated with CPAP and the Control Group on Verbal Working Memory Tasks 153 Table 6 Comparison between the OSA Group Treated with CPAP and the Control Group on Spatial Working Memory Tasks 154 Table 7 Comparison between the OSA Group Treated with CPAP and the Control Group on the Verbal-Spatial Dual Task 155 Table 8 Comparison between the OSA Group Treated with CPAP and the Control Group on the Working Memory Span Task 155 Table 9 Comparison between the OSA Group Treated with CPAP and the Control Group on Standardized Neuropsychological Tests (Raw Scores) 156 Table 10 Number of Participants and Within-group Percentages of the OSA Group Treated with CPAP and the Control Group with Scores Below Clinical Cut-offs Defined by Norms on Standardized Neuropsychological Tests 158 Table 11 Comparison between the OSA Group Treated with CPAP and the Control Group on Sleep Questionnaires 160 Table 12 Comparison between the OSA Group Treated with CPAP and the Control Group on Mood Questionnaires 161 Table 13 Comparison between the OSA Group Treated with CPAP and the Control Group on Functional Outcome Measures 162 Table 14 Comparison between the OSA Group Treated with CPAP and the Control Group on Quality of Life Measures 163 Table 15a Significant Predictors of Neurocognitive Outcomes Based on Stepwise Regressions in Participants with OSA Treated with CPAP 164
Vlll Table 15b Stepwise Regression Model of the Verbal 2-back Task in Participants with OSA Treated with CPAP 165 Table 15c Stepwise Regression Model of the Spatial 2-back Task in Participants with OSA Treated with CPAP 166 Table 15d Stepwise Regression Model of the Working Memory Span - Sentence Verification Proportion in Participants with OSA Treated with CPAP 167 Table 15e Stepwise Regression Model of Digit Symbol in Participants with OSA Treated with CPAP 168 Table 15f Stepwise Regression Model of Wisconsin Card Sorting Test in Participants with OSA Treated with CPAP 169 Table 15g Stepwise Regression Model of Grooved Pegboard Test in Participants with OSA Treated with CPAP 170 Table 16a Significant Predictors of Psychosocial Outcomes Based on Stepwise Regressions in Participants with OSA Treated with CPAP 171 Table 16b Stepwise Regression Model of the Beck Depression Inventory in Participants with OSA Treated with CPAP 172 Table 16c Stepwise Regression Model of the Profile of Mood States in Participants with OSA Treated with CPAP 173 Table 16d Stepwise Regression Model of the Functional Outcomes of Sleep Questionnaire in Participants with OSA Treated with CPAP 177 Table 16e Stepwise Regression Model of the Cognitive Failures Questionnaire in Participants with OSA Treated with CPAP 180 Table 16f Stepwise Regression Model of the Quebec Sleep Questionnaire in Participants with OSA Treated with CPAP 181 Table 16g Stepwise Regression Model of the Quality of Life - Visual Analogue Scale in Participants with OSA Treated with CPAP 184 Table 17 Correlations of Quebec Sleep Questionnaire with Functional Outcomes of Sleep Questionnaire, Beck Depression Inventory, and Profile of Mood States 185
IX LIST OF FIGURES
Figure 1 Prefrontal Model of OSA Deficits 186 Figure 2 Working Memory Model 187 Figure 3 Sequence of Events in the Verbal Memory Scanning Task 188 Figure 4 Sequence of Events in the Spatial Memory Scanning Task 189 Figure 5 Schematic of the Tracking Component of the Verbal-Spatial Dual Task 190 Figure 6 Sequence of Events in the Verbal 2-Back Task 191 Figure 7 Sequence of Events in the Spatial 2-Back Task 192 Figure 8 The Number of Participants at Each Stage of the Protocol 193 Figure 9a Percentages of Participants with OSA Pre- and Post-Treatment and of Healthy Controls Scoring in the Clinically-Significant Range for Respiratory Disturbance Index (RDI) 194 Figure 9b Percentages of Participants with OSA Pre- and Post-Treatment, and of Healthy Controls Scoring in the Clinically-Significant Range on the Epworth Sleepiness Scale (ESS) 195 Figure 9c Percentages of Participants with OSA Pre- and Post-Treatment, and of Healthy Controls Scoring in the Clinically-Significant Range on the Pittsburg Sleep Quality Index (PSQI-Global Score) 196 Figure 9d Percentages of Participants with OSA Pre- and Post-Treatment, and of Healthy Controls Scoring in the Clinically-Significant Range on the Functional Outcomes of Sleep Questionnaire (FOSQ-Total Score) 197 Figure 10 Interaction Effects between Task Condition and Group on Spatial N-Back Task: the OSA Group Deteriorated Disproportionately in Accuracies in the 2-back Condition as Compared to the Healthy Control Group 198 Figure 11 Interaction Effects between Task Condition and Group on Working Memory Span: the OSA Group Deteriorated Disproportionately on the Measure of the Proportion of Sentences Verified in the Dual Condition as Compared to the Healthy Control Group 199
x ABSTRACT
Obstructive sleep apnea (OSA) is the most common sleep disorder and is characterized by nighttime disrupted breathing and hypoxemia, daytime sleepiness, and changes in cognition and mood. The treatment of choice is continuous positive airway pressure (CPAP), and treated individuals often report better sleep and reduced fatigue, yet commonly have persistent cognitive dysfunction and reduced performance at work and daily activities. We investigated the neurocognitive and psychosocial outcomes of patients with OSA treated with CPAP. Thirty-seven individuals with moderate to severe OSA and compliant on CPAP treatment for at least three months were studied with working memory tasks, neuropsychological testing, overnight polysomnographic sleep study (PSG), and self-reported measures of sleep quality, daytime sleepiness, mood, functional outcomes, and quality of life (QoL). Their performance and ratings were compared to 27 age- and education-matched healthy controls. After CPAP treatment, the OSA group showed improved sleep quality and did not differ from controls on measures of the PSG, except for a slightly higher RDI for the control group. In terms of neurocognitive function, treated individuals with OSA performed at a comparable level to controls on basic working memory storage functions but still showed a significant reduction on tests of working memory requiring the central executive. The OSA group also performed worse on neuropsychological measures of complex attention, executive function, and psychomotor speed. With regard to psychosocial functioning, patients show significant improvements on their functional outcomes and quality of life after treatment, and were comparable to controls on mood, functional outcomes, and QoL, except with lower activity level. Sleepiness was found to be a significant predictor of mood, affective states, functional outcomes, and QoL, while subjective sleep quality predicted functional outcomes and QoL. In conclusion, while CPAP improved nighttime sleep, daytime sleepiness, neurocognitive as well as psychosocial outcomes, there were persistent neurocognitive deficits and residual functional difficulties. These results highlight the importance of assessment of neurocognitive and psychosocial outcomes in diagnostic protocols of OSA, and call for interventions targeting these enduring difficulties.
xi LIST OF ABBREVIATIONS AND SYMBOLS USED
AHI Apnea-hypopnea index AHTI Apnea-hypopnea time index ApoE Apolipoprotein E BDI Beck Depression Inventory BMI Body mass index Cho Choline CO Carbon monoxide CPAP Continuous positive airway pressure CT Computed Tomography DSM-IV Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition) EEG Electroencephalogram ERP Event-related brain potential ESS Epworth Sleepiness Scale fMRI Functional Magnetic Resonance Imaging FOSQ Functional Outcomes of Sleep Questionnaire GSI General Severity Index 'H-MRS Proton Magnetic Resonance Spectroscopy MOS Medical Outcomes Survey MMPI Minnesota Multiphasic Personality Inventory MRI Magnetic Resonance Imaging MRS Magnetic Resonance Spectroscopy MSLT Multiple Sleep Latency Test NAA N-acetylaspartate NREM Non-rapid eye movement OSA Obstructive sleep apnea-hypopnea syndrome PASAT Paced Auditory Serial Addition Test PET Positron Emission Tomography PFC Prefrontal cortex POMS Profile of Mood States
Xll PSG Polysomnography PVT Psychomotor Vigilance Test QOL Quality of life Visual Analogue Scale QSQ Quebec Sleep Questionnaire REM Rapid eye movement RT Reaction time RDI Respiratory disturbance index
Sa02 Saturation of arterial oxygen SAQLI Calgary Sleep Apnea Quality of Life Index SF-36 Short Form 36 SCL-90 Symptoms Checklist-90 SD Standard deviation SDS Zung Self-Rating Depression Scale SFQ-V2 Sexual Function Questionnaire Version 2 SPECT Single Photon Emission Computed Tomography
sPo2 Saturation of peripheral oxygen SWS Slow wave sleep TST Total sleep time UPPP Uvulopalatopharyngoplasty VBR Ventricle-to-brain ratio WAIS-R Wechsler Adult Intelligence Scale-Revised WCST Wisconsin Card Sorting Test WMS-R Wechsler Memory Scale-Revised
% Percent
Xlll ACKNOWLEDGEMENTS
I could not thank my supervisor, Dr. Gail Eskes enough for her guidance, advice, thoughtful comments on drafts of the manuscript, and most importantly, her relentless support and trust in me. I would like to acknowledge my PhD Committee members, Dr. Ben Rusak, Dr. Ray Klein, and Dr. Penny Corkum for all the invaluable advice and comments they provided from the inception to the completion of this study.
I would like to express my gratitude to the investigation team members, Dr. Margaret Rajda, Ms. Kathy Spurr, and especially Dr. Deborah Morrison for her assistance in reviewing sleep studies and in creating a research-friendly environment, without which this study would not have been possible. My thanks also go to Dr. Tarvinder Kukreja who assisted in patient recruitment. I am grateful for the service of the sleep technologists, Ms. Tracey Kent, Ms. Carrie Edward-Young, and Ms. Patricia Verboom in conducting and scoring the sleep studies. I also appreciate the assistance of the secretaries of the QEII Sleep Disorders Laboratory, Ms. Lois Whynot, Ms. Evie Clarke, and Ms. Janice Lively.
I am thankful for the dedication and hard work of the research assistants for this study, Ms. Cristine Pensa, Ms. Ella Laur, Ms. Kerri Jones, and Ms. Luiza Radu. I am indebted to Dr. Beverly Butler for her most needed assistance in conducting neuropsychological testing for some of the participants of this study. The technical support of Dr. John Christie was appreciated in the programming of the experimental tasks of this study. My thanks also go to Ms. Margie Clow Bohan of the Dalhousie Writing Centre for her editorial comments of the manuscript.
This study would still have been just a dream without the funding support of a project grant of the Nova Scotia Health Research Foundation, and a scholarship of the Sir Edward Youde Memorial Fund.
Last, but not least I really appreciate the time and effort of all the volunteering participants, from which I have learned so much.
xiv 1
CHAPTER ONE: INTRODUCTION
In this study, I aimed to investigate the neurocognitive function and psychosocial outcomes of individuals with obstructive sleep apnea/hypopnea syndrome (OSA for short) treated with continuous positive airway pressure (CPAP) for at least three months. My major hypothesis was that executive function of treated OSA patients could continue to be impaired, while their basic attentional and memory functions would be restored by treatment. I also reported their psychosocial outcomes, namely mood, daily functioning, and quality of life.
I proposed that (a) treated OSA patients will have normal basic maintenance and rehearsal function of working memory; (b) in contrast, OSA patients will perform not as well as normal controls on tasks that demand the involvement of the central executive; (c) the performance on tests of executive function will correlate with perceived cognitive efficiency and quality of life; and (d) performance on executive tests will correlate with hypoxemia indices.
Overview of OSA
Definitions and Prevalence
OSA is characterized by repeated episodes of upper airway obstruction during sleep, resulting in excessive night-time snoring, periodic breathing interruptions
(apneas/hypopneas), and intermittent blood gas abnormalities (hypoxemia and hypercapnia). OSA is one of the most common sleep disorders, affecting at least 2% of females and 4% of males in the middle-aged population (Partinen & Hublin, 2005).
Higher estimates have been reported by other studies and the prevalence is even higher in 2
the elderly population (Ancoli-Israel et al, 1999; Hiestand et al, 2006). It is estimated that more than 80% of OS A in the general population remains undiagnosed based on the
Wisconsin Sleep Cohort Study (Young et al, 1997b), and that 25% of middle-aged men and 10% of middle-aged women may actually have the illness (Young et al, 2002). In the Sleep Heart Health Study (Gottlieb et al, 1999), 22% of 1,824 participants were reported to have a respiratory disturbance index (RDI) of more than 15 events, which is the conventional cut-off for moderate OSA. In the 2005 National Sleep Foundation
(NSF) annual Sleep in America poll (Hiestand et al.), one in four American adults (31% of men and 21% of women) were identified as high risk for OSA as measured by the
Berlin Questionnaire, and the figures increase to 37% for men and 29% for women in the
50-64 age group. The Cleveland Family Study reported incidence rates of sleep apnea in follow-ups of 285 individuals without sleep apnea at baseline (Tishler et al, 2003). It was found that per year, 7% developed OSA with RDI more than 5 (mild), and 2% showed increases in their RDI to more than 15.
Brief definitions are provided as follows and more detailed technical definitions are given in the Methods section. An apnea refers to a cessation of airflow, and a hypopnea is a partial cessation or a reduction in airflow. Severity of OSA is usually measured by the apnea-hypopnea index (AHI), which is the number of apneas and hypopneas per hour of sleep. Respiratory Disturbance Index (RDI) is often used synonymously with AHI, but other times includes respiratory effort related arousals
(RERAs) in addition to apneas and hypopneas (Kushida et al, 2005). In research, AHI and RDI are used interchangeably to refer to the number of apneas and hypopneas per hour of sleep, without specification of whether RERAs are included. In this study, 3
RERAs were not included in the calculation of RDI. Hypoxemia refers to blood oxygen desaturation, and hypercapnia refers to an abnormally high level of carbon dioxide in the blood.
Measures of OS A severity are usually based on the frequency of respiratory events at night (RDI or AHI), as well as the level of oxygen desaturation during sleep as measured by arterial oxygen saturation (SaCh) or peripheral oxygen saturation measured by pulse oximetry (Sp02). Ye et al. (2005) argued that the apnea-hypopnea time index
(AHTI), which measures the total time that apneas and hypopneas happen per hour is a better indicator of OS A severity, based on the higher correlations with sleepiness and other clinical symptoms of OSA. However, AHTI has not been widely adopted in clinical practice or research. It should also be noted that these indices are a proxy for breathing disruption and may not consistently reflect severity of hypoxemia (Sateia,
2003).
Pathophysiology of OSA
Predisposing pathophysiological factors of OSA include pharyngeal narrowing secondary to lymphatic tissue hypertrophy or mandibular abnormalities, obesity, hypothyroidism, brainstem dysfunction, and excessive alcohol or sedative hypnotic ingestion (Hudgel, 1989). In response to hypoxemia and hypercapnia developing during apneic episodes, patients hyperventilate immediately post-apnea during which snoring is especially loud. As a result of this hyperventilation, blood C02 is reduced to a level that may cause further hypoventilation and eventually apnea. Snoring is usually not as loud at this stage. Due to this hypocapnia effect, respiratory effort in the early phase of apnea may be minimal. As hypoxia and hypercapnia become more prominent in apneic 4
episodes, inspiratory phasic activity of the chest wall muscles, primarily the diaphragm, are stimulated to progressively increase, leading to vigorous respiratory efforts and restless sleep. The respiratory effort is the distinctive feature of OSA as opposed to central sleep apnea, which is mainly caused by ventilator control instability. The apnea is usually broken with a loud snort, sometimes accompanied by mumbling. The whole cycle can repeat up to two times in a minute with as few as two to three breaths between apneas. These apneas cause arterial oxygen desaturation, sometimes to a reduction of more than 50%. The increased respiratory effort is associated with arousals to re establish airway patency, leading to fragmented sleep.
Physiological damages of OSA-associated sleep fragmentation, hypoxemia, and arterial oxygen desaturation have been reported extensively in the literature (Beebe,
2005). Beebe and Gozal (2002) suggested that hypoxia and changes in intra- and extra cellular pH resulting from hypoxia and hypercapnia may create a suboptimal environment for restorative cellular processes involving mitochondrial integrity, protein synthesis, and gene regulation. Rapid cycling of hypoxia and re-oxygenation may induce the production of oxygen free radicals (Douglas & Polo, 1994). The resulting oxidative stress has been suggested to be involved in the pathogenesis of cardiovascular diseases in patients with OSA (El Solh et ah, 2006; Lavie, 2003). Even more pervasive damage can be incurred as there is evidence that antioxidant mechanisms like antioxidant enzymes and vitamins in patients with OSA are adversely altered (Barcelo et al, 2006), rendering the patients vulnerable to the detrimental effects of oxidative stress such as neuronal apoptosis. This effect is not explained by obesity as the BMI of patients and controls were comparable in this study. Nevertheless, Phillips and Grunstein (2006) warned 5
against a simplistic model that portrays a direct cause-effect relationship between hypoxemia in OSA and cardiovascular endpoints. Evidence from the literature is inconclusive and the authors proposed large, multicentre, controlled intervention studies with factorial design to clarify the mechanisms behind these comorbidities. OSA-related neuronal, neurochemical, and neuroanatomical damage and the more specific role of the frontal regions of the cerebrum are discussed in later sections.
Sleep Architecture Associated with OSA
Untreated patients with OSA are also shown to differ from normal subjects in their sleep architecture (Bardwell et al, 2000). They have shorter sleep latency (ranging from 4-24 min), probably due to the accumulated sleep debt. Their rapid eye movement
(REM) latency (ranging from 63-175 min) is longer and REM% (ranging from 6-20%) is less than normal subjects, which may be a consequence of frequent sleep disruption by respiratory events. Patients with OSA also have decreased slow wave sleep (SWS%) relative to people with normal sleep. A recent study reported that patients with OSA
recalled violent or highly-anxious dreams, which were absent in the control group
(Carrasco et al., 2006).
Treatments for OSA
Several kinds of treatment are available for patients with OSA, including weight
loss for obese patients, nasal continuous positive airway pressure (CPAP), surgery
(uvulopalatopharyngoplasty (UPPP) or mandible advancement), and pharmacologic
treatment. Among them, CPAP is considered the treatment of choice for moderate to
severe OSA (Aloia et al, 2003; Bedard et al, 1993; He et al, 1988). CPAP consists of
pressurized air, individually titrated and delivered through a nasal mask to prevent 6
breathing disturbances. CPAP works as a pressure splint that forces the soft palate
anteriorly and hypopharyngeal wall laterally and anteriorly to increase pharyngeal patency (Hudgel, 1989). CPAP is found to be effective in treating obstructive respiratory
events, thereby improving oxygen saturation and reducing sleep fragmentation (Grunstein,
2005; Patel et al, 2003). CPAP also increases REM sleep amount and density (defined
as the percentage of 3-s mini-epochs of REM sleep containing at least one REM), and
eliminates violent/aggressive content in dreams (Carrasco et al, 2006). CPAP has also been shown to partially reverse the decreased antioxidant capacity in OS A (Barcelo et al,
2006), structural and functional cardiac morphological alterations (Shivalkar et al, 2006),
as well as nocturnal and diurnal hypertension (Becker et al, 2003; Faccenda et al, 2001).
Two weeks of CPAP therapy were also found to result in significant reduction in both
daytime and nighttime blood pressure while oxygen therapy and sham CPAP did not
improve blood pressure (Norman et al, 2006). Reversibility of daytime difficulties in
response to CPAP treatment will be discussed in more detail below.
Medical Sequelae of OSA
OSA has invariably been associated with various medical conditions, including
hypertension (Kuniyoshi & Somers, 2006; Nieto et al, 2000; Peppard et al, 2000;
Sanner & Tepel, 2006), cardiovascular diseases (Kaneko et al, 2003; Peker et al, 2002;
Shivalkar et al, 2006), glucose intolerance (Babu et al, 2005), atherosclerosis
(Kobayashi et al, 2006), and cerebrovascular pathologies (Svatikova et al, 2005; Wieber,
2005; Yaggi et al, 2005). Patients with untreated OSA showed increased risk of fatal
and nonfatal cardiovascular events, with odd ratios of 2.87 and 3.17 respectively as 7
compared with healthy individuals (Marin et al, 2005). In patients with OS A, low socioeconomic status was found to be a risk factor amongst the more traditional ones like hypertension and hyperlipidemia for cardiovascular disease (Tarasiuk et al, 2006).
There is also growing evidence that there is a strong association between OSA and the risk of stroke and stroke mortality, independent of a broad range of other cardiovascular risk factors (Arias et al, 2006; Yaggi & Mohsenin, 2004). A recent study provided evidence for a potential mechanism of increased risk of stroke in patients with OSA.
Foster and colleagues (2007) found lower cerebral blood flow response to hypoxia in patients with OSA than in controls. The cerebral blood flow response was moderately correlated with AHI and nocturnal oxyhemoglobin saturation. They also showed that the cerebral blood flow response was normalized in patients after continuous positive airway pressure therapy for four to six weeks. A potential link between OSA and dementia has also been suggested (Bliwise, 2002), as Apolipoprotein E (ApoE) has been shown to be a common marker for Alzheimer's disease (Strittmatter et al, 1993), vascular dementia
(Marin et al, 1998; Myers et al, 1996), as well as OSA (Kadotani et al, 2001). The
Wisconsin Sleep Cohort Study showed that participants with ApoE genotype epsilon 4
(28% of the cohort) have significantly higher mean AHI than participants without the genotype (6.5 vs. 4.8) (Kadotani et al). There are also a disproportionately higher number of participants with ApoE genotype epsilon 4 in subjects with moderate to severe
OSA (AHI > 15) (12%) than in participants with an AHI < 15 (7%), taking into account the effects of age, sex, BMI, and ethnicity. A more recent population-based study also showed that ApoE genotype epsilon 4 was associated with an increased odds ratio for
OSA (Gottlieb et al, 2004). It has also been reported that untreated patients with an 8
apnea index higher than 20 have an 8-year mortality rate of 31%, as compared to 4% in patients with an apnea index lower than 20 (He et al., 1988).
Daytime Consequences of OSA
Engleman and Douglas (2004) commented that patients may find the daytime consequences OSA more important than the nocturnal events, which on the contrary usually receive more, if not all of clinicians' attention and drive the diagnostic decisions.
During the daytime, individuals with OSA experience excessive sleepiness and fatigue, decreased cognitive function and personality changes, resulting in significant negative consequences in work and driving performance as well as lowered quality of life
(Guilleminault & Bassiri, 2005; Mulgrew et al., 2007; Weaver & George, 2005). Two thirds of new patients report lowered work efficiency and difficulties in performing new tasks (Ulfberg et al., 1996). Memory problems and concentration difficulties were also reported in two thirds and three quarters of patients with OSA respectively (Flemons &
Reimer, 1998). About two thirds of patients have problems with social and interpersonal function (Kales et al., 1985), and about one third to one half of patients manifested significant psychiatric symptoms of depression or anxiety (Cheshire et ah, 1992; Douglas,
1998; Millman et al, 1989).
Excessive Daytime Sleepiness and Daily Functioning
The most obvious consequence and manifestation of untreated OSA is probably subjective sleepiness and high propensity to fall asleep during the daytime (Engleman &
Douglas, 2004). To date, the most common means of measuring sleepiness objectively are the Multiple Sleep Latency Test (Richardson et al., 1978), Maintenance of 9
Wakefulness Test (Mitler et al, 1982), and subjectively the Epworth Sleepiness Scale
(Johns, 1991) and the Stanford Sleepiness Scale (Hoddes et al, 1973). Engleman and
Douglas reviewed 29 studies that measured sleepiness using subjective or objective means. They concluded that at least moderate impairments in terms of excessive daytime sleepiness are indicated in patients with OSA. Accumulating evidence suggests that the main causes of daytime sleepiness in patients with OSA are sleep fragmentation and sleep architecture disruptions (Nowak et al, 2006).
Some studies have investigated the relationship between daytime sleepiness and
OSA severity, and the findings are equivocal (Engleman & Douglas, 2004). An early study showed that in patients with moderate to severe OSA (defined as sleep apnea index > 10 and minimum Sa02 < 80%), hypoxemia measures were the best predictors of daytime alertness or vigilance as measured by the four-choice reaction time test and sleepiness as measured by MSLT, while sleep disturbance (e.g., number of awakenings) only in part predicted alertness but not sleepiness (Bedard et al, 1991a). A few studies found only a weak association between respiratory disruptions measured by RDI/AHI and sleepiness (Baldwin et al, 2001; Mulgrew et al, 2007; Young et al, 1993). In a population study using two questions to measure excessive daytime sleepiness subjectively, it was found that sleepiness was more strongly predicted by depression,
BMI, age, subjective estimate of typical sleep duration, and smoking than by presence of
OSA, measured by AHI of more than 15 (Bixler et al, 2005). In contrast, in the Sleep
Heart Health Study, self-ratings on ESS had high concordance with RDI in classifying
OSA (Gottlieb et al, 1999). Along this line, several other studies have shown positive
correlations between RDI and daytime sleepiness (Johns, 1993). OSA severity level 10
categorized by AHI was associated with reduced sleep latency (objective sleepiness) in another study, which investigated the impact of OSA severity and short sleep duration (< five hours) on subjective and objective sleepiness, attention lapses, and tracking errors on
Divided Attention Driving Task in commercial drivers (Pack et al., 2006). More specifically, severe sleep apnea (i.e., AHI > 30) and short sleep duration were found to have similar effects on objective sleepiness in this sample. In a recent study, patients with severe OSA (mean AHI = 62) and excessive daytime sleepiness (ESS > 10 and
MSLT > 5 min) showed shorter sleep latency, increased sleep efficiency, and lower nocturnal oxygenation than patients with similar level of OSA severity but without excessive daytime sleepiness (Mediano et al, 2007). Thus, as suggested by different authors (Ancoli-Israel et al, 1999; Bedard et al, 1991b; Poceta et al, 1992), it could be that in patients with mild OSA, daytime sleepiness is accounted for primarily by the number of arousals during sleep or sleep disruption such as decreased REM sleep and slow-wave sleep, while sleepiness of patients with more severe OSA is more related to the breathing disruptions and the associated nocturnal hypoxemia.
One of the debilitating consequences of excessive daytime sleepiness is impaired driving. Retrospective studies have shown that drivers with OSA have two to seven times more traffic accidents than drivers without OSA (Barbe et al, 1998; Collop et al,
2006; Hartenbaum et al, 2006; Teran-Santos et al, 1999), and drivers with moderate to severe OSA (AHI > 15) are seven times more likely to have multiple accidents after controlling for age, mileage, alcohol use, BMI, and education as confounders (Young et al, 1997a). Teran-Santos and colleagues reported that drivers who received emergency treatment at hospitals in Spain after highway traffic accidents were four times more likely 11
to have OS A than age- and sex-matched patients randomly selected from primary health care centers. The association between OSA and traffic accidents remained significant after adjustment for potential confounding factors, including alcohol consumption, visual problems, age, body-mass index, driving experience, sleep schedule, use of drugs causing drowsiness, and history of traffic accidents. These findings are in line with previous retrospective studies correlating OSA and driving performance (Barbe et ah; Young et ah). In a recent study, Mazza and colleagues (2006) demonstrated that patients showed longer reaction times (RTs), longer stopping distances and more collisions on a naturalistic road safety platform. Their patients also showed divided attention deficits on a driving simulator test but did not show objective sleepiness or selective and sustained attention deficits on other laboratory tests. Similarly, population studies also did not find significant correlations between the risk of automobile accidents and sleepiness or any other clinical and physiological markers of disease severity (Barbe et ah; Young et ah).
These findings suggest that the link between OSA and driving impairments may extend beyond sleepiness.
A recent study demonstrated a relationship between excessive daytime sleepiness and self-reported limitations in work performance in a dose-dependent fashion in patients with OSA (Mulgrew et ah, 2007). Using ESS and the Work Limitations Questionnaire with patients diagnosed with OSA, the study found that every one-point increase on the
ESS was associated with an additional 1 % of time spent at suboptimal work performance for three out of four scales of work performance. 12
Mood
Patients with OSA frequently report or show symptoms of depression and impaired psychological well-being (Cheshire et al, 1992; Millman et al, 1989). In a telephone survey conducted between 1994 and 1999 in the general populations of the
United Kingdom, Germany, Italy, Portugal, and Spain, it was found that 18% of individuals with a major depressive disorder diagnosis also had breathing-related sleep disorders diagnosis (half being OSA) using the Diagnostic and Statistical Manual of
Mental Disorders, Fourth Edition (American Psychiatric Association, 1994), and 17.6% of individuals with a breathing-related sleep disorders diagnosis were also diagnosed with a major depressive disorder (Ohayon, 2003). The same study showed that the risk of having a DSM-IV breathing-related sleep disorders diagnosis was 5.33 higher in depressed respondents than in non-depressed respondents. Ohayon also reported that depression was strongly associated with diagnosis of a DSM-IV breathing-related sleep disorders even after relevant factors such as obesity and hypertension had been taken into account. Guilleminault and Dement (1977) found that 24% of their patients with OSA had seen a psychiatrist for depressive or anxiety symptoms, and 28% showed an elevated
score on the depression scale of the Minnesota Multiphasic Personality Inventory
(MMPI). Millman et al. reported that 45% of their patients with OSA scored above the
cut-off (i.e. score of 50) for depression on the Zung Self-Rating Depression Scale (SDS).
Using a questionnaire, Mosko et al. (1989) found that 58% of their group of patients with
OSA met the criteria for depression on the DSM. Similar results were found in the
narcolepsy and the periodic limb movement groups in the same study suggesting that
depressive symptoms may be common among patients with different types of sleep 13
disorders. Interestingly, only 26% of the patients reported they were currently depressed.
Danlof et al. (2000) reported that 34% of their patients with OSA met criteria for depression using clinical interview and the Comprehensive Psychiatric Rating Scale.
Yue et al. (2003) reported that their sample of patients with OSA showed elevated
General Severity Index (GSI) and higher scores on five subscales on the Symptoms
Checklist-90 (SCL-90) (Derogatis et ah, 1973), as compared with healthy controls. Kales et al. (1985) studied 50 patients with OSA and found significant depression (56%), hypochondriasis (35%), and conversion hysteria (29%) in the study sample. A group of elderly male patients with OSA also endorsed more depressive symptoms on the Geriatric
Depression Scale than healthy controls (Bliwise et al., 1986), although their scores still fell in the normal range and the female patients in the study did not show higher scores than control participants.
A recent study has provided some evidence regarding the association between
OSA and depression. In the Wisconsin Sleep Cohort Study, longitudinal data demonstrated a dose-response association between OSA and depression in a community sample of 1408 participants (Peppard et al, 2006). As the participants' OSA severity increased by one category (e.g., minimal to mild), the odds for development of depression increased 1.8-fold. In comparison with participants with no OSA, combined longitudinal and cross-sectional data showed that the odds for development of depression were increased by 1.6-fold for minimal OSA, by 2.0-fold for mild OSA, and 2.6-fold for moderate or worse (severe) OSA. These findings are consistent with a potential causal link between the two conditions. 14
Some other studies did not find an association between OSA and psychological problems (Lee, 1990; Phillips et al, 1996; Pillar & Lavie, 1998). Phillips et al. studied a group of older adults with mild OSA over five years and did not find any significant psychopathology. Using the Symptom Checklist-90, Pillar and Lavie also did not find an association between breathing disturbance and depression or other psychopathology in
2271 patients screened for OSA. Negative findings on the relationship between OSA and depression were also reported by Cassel (1993). While medical staff evaluated the patients as remarkably depressed and unmotivated, questionnaires filled out by patients did not indicate any personality change specific to sleep apnea. Cassel echoed the viewpoint of Lee (1990) that the so-called personality change or psychological
consequence of OSA is due to a misinterpretation of sleepiness by medical staff and the overlap of symptoms like fatigue between OSA and depression. The implication is that
symptoms of fatigue and depression have to be distinguished in studies investigating
mood in patients with OSA. Bardwell et al. (1999) concluded from their study that many
of the previously reported links between mood and OSA dissipate after controlling for
covariates such as age, BMI, and hypertension. Their data seem to provide some support
for a relationship between anger and OSA, with anger positively relating to the amount of
REM and hypoxemia in patients with OSA.
The relationship between OSA and psychological problems is still uncertain, and
the mechanisms and directionality of impairments are unknown. It has been reported that
depressive symptoms in patients with OSA appear to be related to the severity of the
illness as measured by oxygen desaturation and nocturnal hypoxemia (Cheshire et al,
1992; Kales et al, 1985; Millman et al, 1989). On the contrary, Yue et al. (2003) 15
proposed that psychological abnormalities in patients with OSA are related to sleep fragmentation and excessive daytime sleepiness but not nocturnal hypoxia. Millman et al. reported no correlation between the depressive scores on the Zung Self-Rating
Depression Scale (SDS) and the severity of OSA in their sample, but patients with SDS
scores in the clinical range showed higher respiratory event indices than those whose
SDS scores were below cut-off. It has also been shown that experimental sleep fragmentation produces mood symptoms as detrimental as those induced by total sleep
deprivation (Martin et al, 1996).
A number of authors have studied the impact of OSA treatment on psychological
functions (Engleman et al., 1994; Kribbs et al, 1993). Depressive symptoms were resolved in a group of patients with OSA after treatment with uvulopalatopharyngoplasty
(UPPP) (Klonoff et al., 1987). Dahlof et al. (2002) also reported that the percentage of
patients meeting criteria for depression dropped from 34% to 10% after UPPP. With
various corrective upper airway surgeries, Mosko (1989) also found a significant
reduction of depression scores on the Profile of Mood States (POMS) (McNair et al.,
1971). Also using the POMS, Derderian et al. (1988) found a reduction in depression
scores in seven patients with moderate to severe apnea. They also reported that the
improvement in depression correlated with increase in slow-wave sleep. Millman et al.
(1989) found that 10 out of 11 patients with SDS scores above cut-off showed
improvements in depressive symptoms after CPAP treatment, while the scores of seven
patients out of the 10 dropped below cut-off on the SDS. Flemons and Tsai (1997)
proposed that CPAP treatment is effective in improving depressive symptoms in at least
some patients with OSA, while the relationship between OSA and depression may be 16
multidimensional. In the study by Platon and Sierra (1992), modest and gradual improvement of depression scores and some other scales on the MMPI in 23 patients with
OSA was identified, with significant changes found in the third follow-up (11-14 months
after initiation of CPAP treatment). Sanchez (2001) also described improvements in depressive symptoms on the Beck Depression Inventory (BDI) after one and three months of CPAP.
Several placebo-controlled studies on the reversibility of psychopathology
(mainly depression) have been conducted and mixed results have been reported.
Engleman et al. (1999, 1997) found significant improvement in depression ratings on the
Hospital Anxiety and Depression Scale in patients with mild OSA (AHI between 5 and
15) after four weeks of CPAP treatment, but not after four weeks on oral placebo (lactose tablet). Other findings are less affirmative, however. Barnes et al. (2002) did not find
any difference in 28 patients with OSA (mean AHI = 12.9) between the effects of eight
weeks of CPAP treatment and those after eight weeks of oral placebo on the POMS or the
BDI. A randomized, placebo-controlled, partial cross-over study compared the effects of
properly titrated CPAP treatment and subtherapeutic CPAP on ratings of Geriatric
Depression Scale, among other measures for patients with mean AHI > 60 and did not
find any significant improvements or differences between the groups on their ratings of
symptoms (Henke et al, 2001). Yu et al. (1999) studied the effect of CPAP treatment on
mood states of patients with moderate OSA. They found that patients who used CPAP
and a placebo treatment both showed improvements in mood as measured by the POMS.
They therefore concluded that the effect of CPAP treatment on mood in patients with
OSA could be a placebo effect. A major limitation of this study was that patients were 17
only on CPAP and placebo treatments for one week and long-term effects on mood are unknown. Besides, the placebo group showed a 30% reduction in RDI, making it questionable as to whether the placebo treatment had indeed produced some level of therapeutic effect. Sateia (2003) concluded that while the clinical impression is that OSA may be related to psychopathology and that treatments can reverse the impairments, the literature does not provide unequivocal support for these associations.
Quality of Life
Sateia (2003) reviewed the studies investigating quality of life in patients with
OSA. He summarized that patients with moderate to severe OSA typically report diminished quality of life and marked subjective improvement with treatment. In terms of instruments, most studies have used the Short Form 36 (SF-36), a 36-item scale of the
Medical Outcomes Survey (MOS) (McHorney et al, 1993). Areas covered in the SF-36 include physical functioning, role limitations caused by physical and emotional difficulties, mental health, physical pain, vitality/energy, and general health perception.
While the particular area(s) of impairment may vary across studies, almost all studies identify impairment in one or more aspects of quality of life, as measured by the SF-36.
A dose-dependent relationship between the severity of illness and degree of functional
disturbance has been proposed, and improvements after treatment have been reported,
with the average impairment effect size of about 1 in all subscales compared with norms
(Engleman & Douglas, 2004). The Wisconsin Sleep Cohort Study examined 738 patients
using the SF-36 (Finn et al, 1998). A linear relationship between diminished general
health and apnea was reported. Increasing severity of OSA was also associated with
increasing impairments in physical function, mental health, role function associated with 18
physical problems, social role, and energy. In contrast, a large-scale study of 5816 patients (Sleep Heart Health Study) by Baldwin et al. (2001), also adopting the SF-36, found that energy/vitality was the only scale that showed a linear relationship with apnea.
Significant abnormalities in multiple SF-36 scales, namely physical and social function, vitality, and general health were associated with severe OSA. Impaired quality of life has been reported in several other studies of patients with mild to severe OSA defined by
RDI levels (Bennett et al, 1999; DAmbrosio et al, 1999; Gall et al, 1993; Kingshott et al, 2000).
Some authors argue that using generic instruments such as the SF-36 may be inadequate because the more subtle disease-specific effects on quality of life may not be detected (Flemons & Reimer, 1998; Lacasse et al, 2004). Flemons and Reimer developed the Calgary Sleep Apnea Quality of Life Index (SAQLI), which addresses four domains, namely daily function, social interaction, emotional function, and OSA-related symptoms chosen by individual patients. In a study of 62 patients who underwent four weeks of CPAP, Flemons and Reimer (2002) found that changes in scores on the SAQLI correlated strongly with change in RDI, global quality of life rating, and vitality and social function scales on the SF-36. Lacasse et al. (2002) have conducted independent validation of the SAQLI but they pointed out some shortcomings of the instrument. They remarked that as the SAQLI is interviewer-administered and its symptoms domain is individualized for each patient, a shorter, simpler, standardized, and self-administered quality of life instrument is needed. They therefore developed the Quebec Sleep
Questionnaire (Lacasse et al, 2004), which is a self-administered OSA-specific quality of life questionnaire. In a study of 60 patients, moderate to high correlations were found 19
between scores on the QSQ and other quality of life measures (SF-36; Symptom
Checklist-90), functional status (Functional Outcomes of Sleep Questionnaire, FOSQ), and symptom-based questionnaire (Epworth Sleepiness Scale). The QSQ was found to have good face validity, construct validity, as well as high reliability (Flemons, 2004).
Using the FOSQ in comparing healthy participants and those seeking medical attention for a sleep problem, large effect sizes of > 1 were reported in all but one subscales
(Weaver et al., 1997). In measuring global quality of life, some authors have advocated the use of a single question (de Boer et al., 2004). Eighty-three patients undergoing surgery were studied using a single-item quality of life visual analogue scale, the Medical
Outcomes Study Short-Form-20 and Rotterdam Symptom Check-List. The visual analogue scale was "a horizontal line of 100 mm ranging from 0 (worst imaginable quality of life) to 100 (perfect quality of life)". The scale was found to be of good validity, excellent reliability, moderate distribution-based responsiveness, and high anchor-based responsiveness compared to the multi-item questionnaires. Sanner et al.
(2000) also used a similar measure but in a verbal format for assessing quality of life in patients with OSA.
Studies have also shown that sexual function in patients with OSA was affected in both sexes (Koseoglu et al., 2007). A group of female patients with OSA showed
decreased sexual function in most domains as measured by the Sexual Function
Questionnaire Version 2 (SFQ-V2) with increasing severity of OSA. There was a
significant negative correlation between RDI and most SFQ-V2 domains. In addition,
Margel et al. (2004) demonstrated that male patients with severe OSA have increased
rates of impotence. 20
The literature has not provided a conclusive answer to the question of what predicts, or is associated with, quality of life in patients with OSA. There is some evidence that AHI correlates significantly with several SF-36 scales (Finn et al., 1998).
Baldwin et a/.(2001) found a linear relationship between RDI score and vitality. RDI also correlates with health distress, energy/fatigue, mobility, and social function. A weak relationship between pretreatment SF-36 scores and sleep fragmentation indices was reported by Bennett (1999). In some other studies, however, quality of life measures did not correlate with conventional nocturnal measures of severity of OSA; namely AHI and oxygen desaturation indices (D'Ambrosio et al, 1999; Sanner et al, 2000). Similarly,
Moore and colleagues (2001) proposed that health-related quality of life might improve along with nocturnal breathing and sleep as result of treatment but other factors such as total sleep time might be just as important. Using SF-36, Kawahara and colleagues (2005) found that quality of life was associated with depressive symptoms measured by SDS, but not with sleepiness measured by ESS. However, it should be noted that their patients did not show hypersomnolence before treatment (mean ESS score = 9.7), and that could produce a bias in the results.
Regarding the effects of treatment on quality of life, several studies reported improvement of quality of life after one to six months of CPAP (Bennett et al., 1999;
Bolitschek et al, 1998; Kingshott et al, 2000). Sanner et al. (2000) reported improvements in quality of life on some dimensions of the Nottingham Health Profile but not on the Verbal Analogue Scale after CPAP treatment for an average of nine months.
Improved areas included subjective impairment, emotional reactions, and energy. Along the same line, Nussbaumer et al. (2006) found that both constant CPAP and autoadjusted 21
CPAP improved vitality scores on SF-36 but not other scales. Improvements on SF-36 were found to correlate with minimum oxygen desaturation but not arousals in Kingshott et al.'s study. Consistent with Kingshott et al's findings, Kawahara et al. (2005) found significant improvements on SF-36 in patients after eight weeks of CPAP. In contrast,
D'Ambrosio et al. (1999) did not find an association between RDI severity prior to CPAP treatment and post-treatment improvements on SF-36. Further evidence for the effectiveness of CPAP in improving quality of life was provided by some placebo- controlled trials (Ballester et al., 1999; Engleman et al, 1999; Engleman et al., 1994;
Jenkinsone/a/., 1999).
OSA and Neurocognitive Consequences
Research in the past 20 years has highlighted numerous neuropsychological deficits associated with OSA, including problems with attention, memory, executive functions, and motor abilities (Beebe et al., 2003; Decary et al., 2000; Engleman et ah,
2000; Fulda & Schulz, 2003). Aloia et al. (2004) summarized findings from a review of
37 studies that investigated neurocognitive function of patients with OSA. A brief summary will be provided here while the literature in each cognitive area with tests commonly used will be described in detail later. Aloia et al. reported that general intellectual functioning (4/7 studies) and language function (3/3 studies) were generally spared. On the contrary, most studies found impairments in attention/vigilance (6/8 studies), executive functioning (6/9 studies), memory (7/11 studies), and construction and psychomotor function (e.g., fine manual dexterity) (4/5 studies). Fulda and Schulz reviewed 54 studies with a total of 1635 patients and 1737 control participants, and 22
conducted a meta-analysis of 28 studies that reported adequate statistics. They concluded that moderate to large reductions in mental flexibility, visual-delayed memory recall, and driving simulation were found in the literature, with pooled effect size estimates ranging from 0.61 to 0.72. Focused and sustained attention, verbal delayed recall, verbal fluency,
and various types of composite measures of general intellectual functioning were found
to demonstrate small to moderate reductions with pooled effect size estimates ranging
from 0.17 to 0.51. No difference was detected in areas like divided attention, concept
formation and reasoning, and verbal and visual immediate recall. The authors did not
conduct qualitative reviews or quantitative meta-analyses in areas of attention span and
motor functioning due to large between-study heterogeneity, and of perception, alertness,
and selective attention, vigilance, constructional performance, learning performance,
executive functions, and verbal and performance IQ measures due to insufficient data.
Nevertheless, they concluded that patients with OSA are impaired in their cognitive
function despite the variability across different functions and their subcomponents. In
another review, Engleman et al. (2000) reported large average values for effect sizes in
attention (1.0) and executive cognitive scores (0.9), and moderate effect sizes for memory
scores (0.6). As a general rule, little impairment was found in two community recruited
samples with a mean RDI/AHI of 17 and 19 respectively, and moderate to large
impairment effect sizes were obtained from clinical studies with a grand mean AHI of 30
(Engleman & Douglas, 2004).
Global Intellectual Functioning
Global cognitive functioning and language appears least affected, although
deficits in general intellectual functioning, usually assessed by the Wechsler Adult 23
Intelligence Scale-Revised (WAIS-R) have been reported (Bedard et al., 1991b; Cheshire et al., 1992; Findley et al, 1986; Greenberg et al, 1987). At the same time, most of these studies suggest an association between general intellectual function and nocturnal hypoxemia. An association between hypoxemia and general cognitive function has also been suggested by studies examining hypoxic patients with other medical conditions such as chronic obstructive pulmonary disease (Prigatano, 1983). An interesting study proposed a possible relationship between general intelligence and cognitive deficits in patients with OS A (Alchanatis et al, 2005). Forty-seven patients with OS A and 36 control participants were tested on the Raven's Progressive Matrices Intelligence Test and a computerized battery of attention/alertness tests. Participants were divided into the high-intelligence group (IQ > 90 percentile) and the normal intelligence group (50 < IQ
< 90 percentile). While patients in the high-intelligence group performed at a comparable level as intelligence-matched controls, patients in the normal-intelligence group showed a decline in attention/alertness compared with the intelligence-matched control participants.
The authors proposed that premorbid high intelligence may have a protective effect against OSA-related cognitive deterioration, owing to increased cognitive reserve. This theory awaits support from studies using a larger sample and a more comprehensive
assessment battery.
Attention and Concentration
Most studies have assessed and found deficits in attention/vigilance, as measured by the Trail Making Test (Bedard et al., 1991b; Cheshire et al., 1992; Dealberto et al,
1996; Findley et al, 1986), Stroop Color Test (Naegele et al, 1995), (Bedard et al.,
1991b; Cheshire et al, 1992), Paced Auditory Serial Addition Test (PASAT) (Cheshire et 24
ah, 1992; Findley et ah, 1986), Letter Cancellation (Bedard et al., 1991b; Greenberg et al.,
1987), Digit Symbol Substitution Test (Dealberto et ah), and Choice Reaction Time
(Bedard et al., 1991b). Vigilance has also been studied in patients with OS A using reaction time measures such as the Psychomotor Vigilance Test (PVT) (Dinges & Powell,
1985). Powell et al. (1999) compared patients with mild to moderate OSA to a group of healthy controls who were given alcohol on various parameters of the PVT. It was found that patients with OSA performed comparably to or worse than the healthy participants with elevated blood alcohol concentration levels (mean breath alcohol concentration =
0.057 grams/210 liters of breath), which was higher than the legal limits for driving in most American states. Using the computer program "Steer Clear", which measures vigilance performance in avoiding obstacles by pressing a button to move an automobile on the screen to another lane, Findley et al. (1991) found that older adults with OSA performed significantly worse than their age-matched controls. In contrast, Ingram et ah
(1994) did not find any differences between old adults with or without OSA on the same vigilance measure. Authors of the latter study attributed the discrepancies to the different sampling methods and hence subject characteristics. Ingram et al. studied community- dwelling individuals while Findley et al. recruited their participants from clinical referrals for evaluation of potential sleep apnea. Therefore, although the mean RDIs were similar for Findley et al.'s (50) and Ingram et al.'s (44) participants, Findley et ah's participants likely showed more symptoms and functional deficits.
Motor Functions
In the review by Fulda and Schulz (2003), motor functions were grouped into two categories, psychomotor speed and manual dexterity of speed according to the taxonomy 25
proposed by Lezak (1995). While study heterogeneity precluded meta-analyses, none of the nine studies using the Finger Tapping task to measure psychomotor speed found a difference between patients and controls (Sloan et al, 1986, 1989; Verstraeten et al,
1996). Using the digit copying task, Stone et al. (1994) also did not find any differences between patients and insomniac controls or the norms. The only study that found differences in psychomotor speed reported prolonged reaction time on a sensory motor task as compared to controls (Lee et al, 1999). More varied results were found in the manual dexterity and speed category. No difference was detected in the two studies using the Grooved Pegboard task (Kim et al., 1997). However, using the Purdue
Pegboard task, a few studies found differences between patients and normal controls
(Bedard et al., 1991a; Bedard et al., 1991b; Greenberg et al., 1987), snoring controls
(Verstraeten et al, 1997), or norms (Verstraeten et al, 2000). Two studies reported no
difference on the task as compared with insomniac controls (Verstraeten et al, 1996) and norms (Walsleben et al, 1989).
It should also be noted that the only Level 1 study1 in this area (Kim et al, 1997)
according to the criteria adopted by Fulda and Schulz (2003) showed reduction in
Grooved Pegboard performance in a population-based sample of 199 participants with
AHI score more than 5, as compared with the 642 participants with AHI score less than 5.
While evidence for the presence of neurocognitive impairments in general is
robust, findings on specific deficits can be inconsistent for some cognitive domains
1 Fulda and Schulz (2003) adopted the classification system of level of evidence developed by Clarke and Oxman (2000) for studies in evidence-based medicine. Level 1 studies are those with high external, internal, and statistical validity. 26
(Aloia et al, 2004). Executive functions and memory are less thoroughly studied and the findings are more inconsistent owing to the complexities of these constructs.
Memory
Deficits in short- and long-term memory have been reported in previous studies.
Patients with moderate (Apnea Index2 between 10 and 30) to severe (Apnea Index > 30)
OSA demonstrated deficits in non-verbal immediate recall and the severe group showed additional impairments in both verbal and nonverbal delayed recall (Bedard et al., 1991b).
These deficits were associated with vigilance decrements. Short-term memory deficits have also been observed in other studies (Findley et al, 1986; Greenberg et al, 1987;
Naegele et al, 1995). Consistent with the findings of Bedard et al, Findley et al. reported deficits in immediate recall of both verbal and nonverbal information.
Greenberg et a/.(1987) reported deficits in digit span in patients with OSA but no differences from controls on other subscales of the Wechsler Memory Scale (WMS).
Naegele et al. (1995) found impaired performance on both verbal and nonverbal delayed recall tasks in patients with OSA. However, looking more closely at their findings, they reported that the patients' deficits were attributable to learning of information, and not forgetting of learned information. Therefore, they contended that the pattern of impairments was indicative of dysexecutive functions, associated with frontal lesions, rather than the typical amnesic features in temporal lobe lesioned patients. In a more recent study, Naegele and colleagues investigated the specific memory processes affected in untreated individuals with moderate to severe OSA as measured by RDI > 15 (2006).
Performance of the OSA group was compared to an age- and education-matched control
2 Apnea Index was defined as the number of apneas per hour of sleep, excluding time awake. Only apneas lasting for more than 10 seconds were counted. 27
group on measures of episodic memory (list-learning), procedural memory (mirror tracing task), and working memory (maintenance and processing of information, a dual task). Findings of the working memory tasks are discussed under the executive function section below but those on the list-learning and mirror tracing tasks are described here.
Naegele et al. found that the OSA group performed worse than the control group in the learning trials of the list-learning task, with similar performance on the first trial, in free and cued delayed recall, and recognition. On the procedural memory task, the OSA group performed more poorly than the control group on all seven trials but showed learning across trials.
Decary et al. (2000) pointed out the lack of standardization of the use of a neuropsychological test battery, which may contribute to the contradictory findings across studies. They therefore proposed a standardized assessment battery that hopefully would eliminate the confusion. Use of this standardized battery would facilitate comparison of findings across studies, but it does not address the lack of sound theoretical neurocognitive frameworks to guide the choice and interpretation of neuropsychological tests (Verstraeten & Cluydts, 2004). Among the unresolved issues, the controversy over executive deficits of OSA patients deserves special attention and
further exploration.
Executive Function
Executive function is defined as the ability to respond to situations in an adaptive
manner and engaging successfully in independent, purposive and self-serving behavior
(Lezak, 1995). Executive function is not a unitary domain, but is composed of many
underlying functions. These functions form the basis for many cognitive, emotional, and 28
social skills. Impairment in executive function is frequently reported in studies of OSA patients, with large effect sizes, although the findings are not always consistent (Aloia et al., 2004; Engleman et al., 2000; Sateia, 2003; Verstraeten et al, 2004).
The literature seems to suggest that abnormalities with executive function are most evident in patients with severe OSA (Sateia, 2003). On standard measures of executive function, mild performance deficits on Wisconsin Card Sorting Test (WCST) and Tower of Toronto were reported (Naegele et al., 1995). Impaired performance on other tests that may have an executive component was reported in other studies. Findley et al. (1986) found deficits on the Paced Auditory Serial Addition Test (PASAT) in the more hypoxic group of his patients with OSA. Bedard et al. (1991b) reported diverse deficits in various tasks that involve executive function, including verbal fluency, planning, sequential thinking, constructional ability. The extent and severity of impairment appeared to vary with the severity of the illness. Consistent with this cluster of findings, Gale and Hopkins (2004) also found that executive function, as measured by
Trail Making B and verbal fluency, was the most commonly impaired area in their group of 14 patients with severe OSA with a mean RDI of 83.6. Compared to patients with multi-infarct dementia, dementia of Alzheimer's type, and chronic obstructive pulmonary disease, patients with OSA showed a distinctive profile of cognitive impairment, mainly involving inductive thinking measured by the Temporal Rule Induction subtest of the
Mental Deterioration Battery (MDB), deductive thinking measured by the analogies test, and constructional ability measured by copying geometric figures (Antonelli Incalzi et ah,
2004). On the contrary, other studies did not find significant differences between patients
and controls on executive tasks. For example, Salorio et al. (2002) found no difference 29
on Wisconsin Card Sorting Test (WCST), although results on verbal fluency measures were mixed with deficit in letter fluency but not semantic fluency. A few studies of patients with mild OSA also did not identify significant impairments in executive function as well as memory and other neurocognitive functions (Knight et ah, 1987;
Phillips et ah, 1996). These inconsistencies may be related to ambiguity in task classification, an absence of unifying explanations in terms of cognitive mechanisms, poorly defined concepts of frontal lobe function, executive function, memory and attention, and usage of tasks insensitive to mild deficit (Jones & Harrison, 2001).
Verstraeten and Cluydts (2004) noted that effects of attentional capacity deficits resulting from sleepiness on higher cognitive function have not been fully taken into account in interpreting findings of executive attentional dysfunction. They argued that executive performance should not be measured in isolation, but should be calculated by subtracting more basic components of capacity and/or speed from the executive measure through residual-based analyses to have a more specific measure of executive control.
For example, they advocated subtracting digit span forward (presumed to measure capacity) from digit span backward to measure manipulation of information without the bias of changes in capacity, or subtracting Trail Making A score (a measure of psychomotor speed and basic visual search) from Trail Making B to measure set shifting without the influence of motor speed. After controlling for basic attentional performance statistically in this way, Verstraeten et al. (2004) reported no specific indications for executive attentional deficits on Trail Making Test B, Digit Span Backward, Stroop incongruent condition, Five-point Test of design fluency, and the Flexibility Test subtest of the computerized Zimmermann-Fimm Test battery for Attentional Performance, in his 30
population of 36 patients, with a mean AHI of 60.5. Because of the apparent contribution of impaired basic attentional capacity to the deficits, they concluded that the deficits on executive function tasks appeared more related to sleep loss than to specific executive dysfunction that is associated with prefrontal damage. Their arguments are cogent, and it is imperative that basic attentional capacity be taken into account in testing and interpreting executive functions. However, their conclusions that untreated patients with
OSA present cognitive deficits similar to those of normal individuals with experimental sleep deprivation, and that sleep disruption only causes vigilance decrements and attentional capacity deficits, are inconsistent with the rich sleep deprivation literature which suggests that sleep deprivation is associated with some high-level cognitive dysfunction that is above-and-beyond impaired basic attentional capacity and processing
speed (Harrison & Home, 2000a). It should also be noted that the residual-based regression analysis requires a large sample size especially for the control group to yield a
stable estimate of the regression line (Salthouse & Hedden, 2002), which Verstraeten et
al.'s study might lack, although details of their statistical analyses were not described in the study. In a more recent article, Verstraeten (2007) proposed using two-way analysis
of variance in data analyses to delineate the effects of "lower-level" processes like
arousal, alertness, or basic attention and those of complex attention or executive function.
This is in agreement with the data analytic strategies of the present study.
In addition to the issue of basic capacity influencing executive test performance,
standard clinical neuropsychological tests used in most studies to date (such as Trail
Making or Stroop) are designed to detect major functional deficits in neurologically
impaired groups. It is well known that the sensitivity and specificity of these tests is 31
moderate at best, particularly to more subtle deficits (Lezak, 1995). Furthermore, tests like Wisconsin Card Sorting Test are multi-dimensional, with limited theoretical structure, making it difficult to isolate the specific underlying executive deficit (Camus et ah, 1999).
Theory-driven studies with sensitive and theoretically-defined experimental measures are called for to further delineate basic and executive attentional functions of OSA patients.
Two more recent studies have adopted a working memory model in investigating executive function (Lis et ah, 2008; Naegele et ah, 2006). Naegele and colleagues used three different dual tasks to measure allocation of attentional resources in performing: 1.
A digit span task and mirror tracing task, 2. a digit span task and a tracking task, and 3. A digit span task and a spatial span task. They used another set of tasks to measure the maintenance and processing components of working memory: 1. Auditory transformed span, which required recall of a series of digits presented auditorily after "transforming" the digits using specified arithmetic operations; 2. A modified Paced Auditory Serial
Addition Test, in which subjects were instructed to add each number presented to them to the preceding one under time pressure; 3. A self-ordered spatial memory tasks, in which subjects had to search for hidden targets by opening boxes presented on the screen. This task generated two measures, the self-ordered spatial memory, which was the number of times the subject reopened a wrong box, and the self-ordered spatial strategy, which was the degree of organization to search for targets. Results of this study showed that the
OSA group and the control group did not differ on any of the three dual tasks but
significantly worse performance on all the tasks requiring maintenance and processing was found in the OSA group. Lis and colleagues attempted to separate working memory
dysfunctions from additional processes involved in working memory tasks by adopting a 32
reaction-time-decomposition approach. Twenty patients with OSA with RDI of 57.9 were compared to 10 aged- and education-matched healthy controls on a simple reaction time task, stimulus discrimination task, a choice reaction task, a vigilance choice reaction task, a 1-back task, and a 2-back task. It was found that the OSA group has slower RTs than the healthy control group on the n-back tasks and also in the non-working memory tasks. Interestingly, lowered accuracy was only found in the 1-back and 2-back tasks.
The authors concluded that RT slowing in OSA group was attributed to deficits in more basic cognitive processes while reduced accuracy in working memory tasks indicated executive dysfunction.
Despite the inconsistencies of findings and limitations of study methodologies, the prominence of executive dysfunction in OSA brought about a recent review that specifically examined the evidence for executive deficits in OSA (Saunamaki &
Jehkonen, 2007). The review focused on the following aspects: the generalizability of study findings to the OSA population based on the studied patients' characteristics and the presence of a control group, the methods employed in evaluating executive functions, the common tests used and the respective domains of executive functions measured, the common executive deficits found, and the effects of CPAP treatment on executive functions. Reviewing 40 studies that satisfied their selection criteria (i.e., English, human subjects, adults, empirical studies, study of executive function), the authors concluded that the generalizability of the findings in the literature is limited by the variation in sample sizes (range from 8-199), the heterogeneity of patient groups in terms of severity of OSA, the overrepresentation of male patients (83% of studies dominated by >75% male patients, the lack of information on educational level (only half of studies specified 33
education), and the inaccuracies in measuring and defining the severity of OSA. Another problem was that half of the studies used only one or two tests in assessing executive function, which has multiple domains. Due to these issues, erroneous conclusions might therefore be drawn. Out of the 40 studies, only 12 had a healthy control group at baseline
Summarizing findings from these studies, the most frequently detected problems were found on Digit Span (forward and backward), Corsi's block-tapping test, the phonological fluency tasks, copy of the Rey-Osterrich Complex Figure Test, the Mazes, and the Wisconsin Card Sorting Test. In terms of domains of executive function, the most frequently reported deficits were working memory, phonological fluency, cognitive flexibility, and planning. Effects of CPAP on executive deficits are discussed in detail in
a later section.
Mechanisms of the Neurocognitive Dysfunction
Hypotheses regarding the pathogenesis of the cognitive deficits in OSA include
two main possible factors: The influence of excessive daytime sleepiness due to sleep
fragmentation (Bonnet, 1993; Martin et al., 1996) and the presence of brain dysfunction
due to the intermittent night-time blood-gas abnormality (i.e. hypoxemia, hypercarbia).
Several approaches have been adopted to investigate this question, as reviewed by
Valencia-Flores and colleagues (1996). One approach is to compare patients with OSA
to other patient groups manifesting excessive daytime sleepiness, such as patients with
narcolepsy or people with chronic sleep deprivation (Greenberg et al, 1987). The major
problem with this approach is that the hypersomnolence of patients with OSA may differ
from that of patients with narcolepsy or other conditions due to the different etiology and 34
sleep patterns. Another approach is to correlate measures of sleep disruption, such as arousals, and of hypoxemia, such as RDI or blood oxygen desaturation measures with cognitive deficits (Cheshire et ah, 1992). A concern with this approach is the often strong correlations between the measures of sleep disruption and hypoxemia, violating the assumptions of lack of colinearity in applying multiple regression. A third approach is to measure alertness and cognitive performance over the course of a day, with a view to demonstrate that variations in alertness across the day have a greater impact on some functions than others, which are theorized to be attributed to nocturnal hypoxemia
(Bedard et al., 1991b). Valencia-Flores and colleagues (1996) tackled this question with
a yet different approach. In their study, 37 patients with OSA were assessed on polysomnography (PSG), MSLT, and neuropsychological measures before and after two days of CPAP therapy. Subgroup analyses were conducted to tease out the differential
effects of alertness and hypoxemia. Patients without appreciable baseline hypoxemia, defined as greater than two desaturations below 80% per hour were divided into two
groups: one with increases in alertness after treatment and one with post-treatment
decreases in alertness, as measured by the MSLT. It was found that recall performance
on a list learning test improved for the group with increased alertness and worsened for
the group with reduced alertness, suggesting that alertness is related to auditory verbal
learning. To investigate the effects of hypoxemia, patients with improved alertness were
divided into the hypoxic group and the non-hypoxic group using the criterion outlined
above. The authors found that performance on a sustained attention task involving
repetitive arithmetic calculations improved in the non-hypoxic group but not the hypoxic
group. 35
Engleman et al. (2000) reviewed a series of case-control studies and described a positive relationship between effect sizes of cognitive impairment and hypoxemia in untreated individuals with OSA. Amongst the studies reviewed by Engleman, Findley et al. (1986) reported that patients with hypoxemia displayed more severe cognitive impairments than those without hypoxemia. The hypoxemic patients' performance was in the impaired range on measures of attention, concentration, complex problem-solving, and short-term recall of verbal and spatial information, while patients without hypoxemia performed in the normal range on all tests. The degree of hypoxemia during sleep and wakefulness also significantly correlated with the extent of overall cognitive impairment.
In contrast, measures of sleep fragmentation did not correlate with cognitive impairments identified in the study. Adams et al. (2001) studied the relationships between sleep variables and sleepiness and neuropsychological functions in a group of 100 patients with mild to moderate OSA using neuropsychological measures factor-analyzed into four
constructs. They concluded that factors like signal discrimination, declarative memory,
and working memory were predicted by RDI or hypoxemia in a dose-dependent fashion
while sleepiness was associated with vigilance. Other studies reported limited
differences between cognitive functions of patients with mild and moderate OSA and that
only patients with severe OSA are distinguished with more wide-spread and prominent
impairments. Antonelli Incalzi and colleagues (2004) reported a lack of association between hypoxemia indices and cognitive impairment in their group of patients with
mostly mild to moderate levels of OSA. A recent study of episodic, procedural, and
working memory in OSA did not investigate predictors of patients' performance per se,
but reported significant, albeit weak correlations between performance on a mirror 36
tracing task (procedural memory) and amount of time spent at Sa02 < 90% (r = -.28) and between scores on a self-ordered spatial memory task and RDI (r = .27) (Naegele et al,
2006).
It has been theorized that attentional deficits, and to some extent, memory are more related to sleepiness while hypoxemia causes the executive and psychomotor dysfunctions (Bedard et al, 1993; Jones & Harrison, 2001; Sateia, 2003). However, this is not a definite conclusion. For example, vigilance decrements were attributed to sleep fragmentation in patients with OSA in some studies (Bedard et al, 1991b; Montplaisir et al, 1992) but to hypoxemia (RDI/AHI) in others (Cheshire et al, 1992). Performance on executive function tests in OSA patients is often found to be better correlated with hypoxemia than measures of sleepiness (Bedard et al., 1991a; Bedard et al., 1991b;
Cheshire et al., 1992; Jones & Harrison, 2001; Kim et al., 1997; Naegele et al., 1995).
Naegele et al. examined 17 patients with OSA and found deficits in initiation, inhibition, perseveration, verbal and visual learning, and memory spans. Logistic regression analysis suggested that respiratory events were more related to memory deficits while hypoxemia was related to frontal lobe-related abnormalities. Studying untreated patients with OSA, Lis et al. (2008) reported significant correlations between subjective sleepiness and RTs in simple reaction time tasks and working memory tasks. Accuracy deficits in patients with OSA were correlated with objective sleepiness measured by the
Multiple Sleep Latency Test. Performance was not correlated with respiratory indices in this study. However, imitating the sleep fragmentation in OSA by inducing short repetitive microarousals in normal participants, Martin et al. (1996) found some executive-type deficits on tests of mental flexibility (Trail Making B) and PAS AT 37
(sustained attention, information processing with interference, inhibition). It should be noted that while the study was quite rigorously designed, there was no correction of
statistical significance for multiple comparisons, which rendered the significant finding
on the two tests (out of seven) questionable. A more recent study of experimentally
induced sleep fragmentation showed that performance on reaction time task and sustained
attention tasks did not change after two nights of subtle sleep fragmentation but
electroencephalographic and event-related potentials data suggested impairment in
information-processing capabilities associated with impaired arousal and attention (Cote
et ah, 2003). While this study did not address the effects of long-term sleep
fragmentation, the neurophysiologic deficits after subtle sleep fragmentation suggested
that the intermittent respiratory induced arousals in OSA likely caused deficits in
attention and information processing. Beebe and Gozal (2002) pointed out that while the
impact of sleepiness is most evident on long monotonous tasks, patients with OSA
showed difficulties on executive tasks that are relatively short. Further insight can also
be gained from studies on children with OSA. Gozal and colleagues (2001) reported that
prepubertal children with OSA are less likely to show daytime sleepiness unless their
OSA is moderately severe to severe, but executive dysfunction is often found in them,
suggesting the dissociation of sleepiness and executive function. In addition, executive
dysfunction is often found to remain after CPAP treatment, when sleep quality is restored
(Bedard et al., 1993; Naegele et al., 1995). The chronicity of executive dysfunction
despite normalization of sleep quality also suggests an underlying pathophysiological
mechanism independent of sleepiness or reduced alertness. Blunden and Beebe (2006)
reviewed the contribution of intermittent hypoxia, sleep debt and sleep disruption to 38
daytime performance deficits in children with respiratory and non-respiratory sleep disorders. They concluded that all these factors may be sufficient to cause daytime effects in children, and that they have additive effects.
Beebe (2005) reviewed a series of studies that investigated the mechanisms for neurobehavioral deficits in OSA on a cellular level. These studies mostly adopted a rodent model of OSA and focused on intermittent hypoxia. Row and colleagues (2003) reported that intermittent hypoxia is associated with oxidative stress and deficits in spatial learning in a water maze. Similar findings were reported by Payne and colleagues, who showed that electrophysiological properties of hippocampal neurons are disrupted by intermittent hypoxia during sleep (Payne et al, 2004). Up-regulation of pro inflammatory cytokines as well as excessive nitric oxide levels were also reported, which are associated with cortical and hippocampal neuronal apoptosis (Li et al, 2004). Li and colleagues reported that when the nitric oxide synthase-targeted gene was removed from mice, the increase in nitric oxide activity induced by intermittent hypoxic exposure was abolished and the related neurobehavioral spatial deficits were attenuated. In a follow-up study, Xu and colleagues (2004) extended these findings, reporting that mice showed attenuated apoptosis and spatial learning deficits after intermittent hypoxia if they had genetic mutations that result in reduced free radicals or nitric oxide or they received an anti-oxidant. Beebe also pointed out that there is accumulating evidence for adults, adolescents, and children with OSA showing elevated levels of inflammatory markers
(Larkin et al, 2005; Mills & Dimsdale, 2004). Mills and Dimsdale provided a review of findings on the relationship between sleep and cytokines, drawing links between 39
increased pro-inflammatory cytokines and OSA-related features such as day-time sleepiness, fatigue, depressive symptoms, and even neuropsychological functioning.
In addition to the hypoxia-induced cellular damage, there is evidence from animal research that recurrent hypoxia during postnatal development is associated with persistent, relatively long-standing abnormalities in respiratory functions (Moss et ah, 2006). This indicates long-standing effects of OSA, and that childhood OSA may compromise the ability to respond adequately to subsequent hypoxic events in adulthood. Kheirandish and colleagues (2005) investigated the effects of experimentally-induced intermittent hypoxia on spatial working memory , monoamines, and dendritic branching in rats. It was found that after 14 days of exposure to intermittent hypoxia starting at postnatal day
10, male rats but not female rats showed deficits in demonstrating memory of a previously learned escape location. The male rats also showed decreased dendritic branching in the frontal cortex and a lack of increased dopamine concentrations, which were found only in the female rats. The authors concluded that intermittent hypoxia is
associated with long-term alterations in frontal cortical dopaminergic pathways, leading
to neurobehavioral deficits. Kanaan and colleagues (2006) studied the effects of chronic
continuous hypoxia and chronic intermittent hypoxia in mice, and found that there was a
significant increase in capillary density in both cortex and hippocampus after two weeks
and four weeks of exposure to continuous and intermittent hypoxia. Upon re-
oxygenation, there was a reversal and capillary density was normalized in the continuous
3 Working memory as commonly used in the animal literature is defined differently from working memory as used in the human literature. A common working memory measure in the animal literature is the water maze. Paradigms usually involve rats locating a submerged platform hidden in the water maze. The reduction in latency and swimming distance in locating the platform in the delayed trial as compared to the acquisition trial is used as the working memory measure. In human literature, this type of measure is referred to as spatial delayed recall or spatial memory. 40
hypoxia group, the only condition studied with reoxygenation. With hypoxia in both conditions, demyelination was observed in the corpus callosum, and reoxygenation did not reverse the condition. The authors concluded that the potential irreversible dysmyelination in early life by conditions like OSA, could have long-term and devastating effects.
A few sleep deprivation studies have also reported evidence for inflammatory processes (Beebe, 2005). Healthy subjects showed increased concentration of C-reactive protein, an inflammatory marker of cardiovascular risk, after 88 hours of total sleep deprivation or 10 days of sleep restriction to 4.2 hours (Meier-Ewert et al., 2004), and elevated peripheral markers for inflammation along with increased sleepiness and reduced psychomotor vigilance after experimental sleep restriction from 8 to 6 hours per night for a week (Vgontzas et al., 2004). Rodents showed inflammatory cytokine production after 36 hours of sleep deprivation (Hu et al., 2003), and inhibition of hippocampal long-term potentiation and related learning deficits in contextual fear conditioning after 72 hours of primarily REM sleep deprivation (McDermott et al, 2003).
Research in the past two decades has shown that a dichotomous model with two pathways (i.e. sleep fragmentation and associated sleepiness, and hypoxemia) causing two groups of cognitive deficits is probably over-simplified. Attempts to understand the contributions of the two factors are further complicated by the correlation between sleep disruption and oxygen desaturation (Cheshire et al., 1992). While different researchers have pointed out that sleepiness alone cannot explain the daytime impairments and in particular the executive dysfunction displayed by patients with OSA, long-term sleep fragmentation and disruption may still have a role to play in producing persistent 41
executive dysfunction, particularly when one considers the fact that sleep abnormalities may have an onset much earlier than the full-blown OSA symptomatology and the long
waiting period for patients to see a sleep specialist for diagnosis of OSA. Another
complication is that evidence from animal research has shown that hypoxia-induced
arousals lead to more significant disruption in sleep architecture (reduced EEG delta
power of NREM sleep and total amount of REM sleep) than non-hypoxic arousals
(Polotsky et al., 2006). This implies that hypoxic exposure is closely associated with
sleep architecture deficits that are not explained by non-specific sleep fragmentation, and
that the effects of hypoxia on cognitive functions may be partially mediated by sleep
architecture disruptions, hence again demolishing the dichotomy. Research has also
shown a relationship between hypoxia and sleep architecture and hypersomnolence.
Veasey and colleagues (2004) demonstrated that long-term intermittent hypoxia results in
oxidative injuries in sleep-wake regions in the basal forebrain and brainstem, and these
biochemical changes are associated with hypersomnolence and heightened vulnerability
to acute sleep loss with little recovery after two weeks of normoxia. Beebe and Gozal
(2002) proposed a more integrative model to explain the physiological mechanisms that
cause the neurocognitive and psychological disturbances (Figure 1). This model
theorizes that OSA causes sleep disruption and intermittent hypoxia and hypercarbia.
Both of these factors contribute to disturbances of the restorative processes of sleep and
introduce cellular and biochemical stresses that result in disruption of functional
homeostasis and altered neuronal and glial viability in certain brain regions, particularly
the prefrontal cortex (PFC). While it may be difficult to separate out the contribution of
sleep disruption and blood gas abnormalities in executive dysfunction in patients with 42
OSA, investigating treated OSA patients may provide some special insights in dissociating the short-term impact of sleep deprivation and its associated excessive daytime sleepiness from the more long-term impact of sleep disruption and blood gas abnormalities. Such investigations would also allow evaluation of the individual contribution of the two possibly interacting mechanisms.
Lesions and neuroimaging studies have demonstrated that executive functioning involves the PFC and its rich interconnections with various areas of the brain (Foster et al, 1994; Kolb & Whishaw, 2003; Lawrence et al, 2003; Stuss et al, 1995). D'Esposito and colleagues (1998) reviewed functional Magnetic Resonance Imaging (fMRI) studies of working memory tasks specifically, and concluded that tasks involving only maintenance of information (i.e. phonological loop and visuospatial sketchpad's basic storage and rehearsal function) activate the ventral PFC while those requiring monitoring, updating, manipulating, or temporal tagging (i.e. the central executive), in addition to maintenance, activate the dorsal PFC. Using an event-related fMRI technique, the same group was able to provide further evidence for the double dissociation of the neural bases of the mental processes of memory storage and memory manipulation in working memory tasks (D'Esposito et al., 1999; Postle et al, 1999). Their data support the hypothesis that dorsal lateral PFC is sensitive to non-mnemonic executive control processes in working memory function and the left posterior perisylvian cortex mediates the mnemonic processes for storage on a verbal item-recognition working memory task.
A separate group reported parallel findings on spatial working memory (Owen et al,
1996; Owen et al, 1999). Using positron emission tomography (PET), the authors found that a spatial span task yields activation in the right mid-ventrolateral frontal cortex while 43
spatial 2-back as well as other monitoring tasks are associated with additional activations in mid-dorsal frontal regions. These conclusions seem to stand the test of time so far, except for a fMRI study that presents somewhat inconsistent findings (Rypma et al,
1999). The authors reported that activations in the middle and superior frontal gyri, the regions subserving executive function, were observed just by increasing the maintenance
load of the memory set in a delayed-response verbal working memory task, without overt
requirement to manipulate the stored information. It was proposed that the PFC is
activated when executive function is required as the demands of maintenance exceed the
capacity of the domain-specific slave systems. Replication of such findings is needed to
provide more definitive support for this proposal regarding the role of the central
executive in working memory maintenance. Reviewing evidence from functional
imaging and lesion studies, Mtiller and Knight (2006) summarized that the PFC has a
"non-mnemonic" role and is involved in attentional control of sensory processing,
integration of information from different domains, stimulus selection, and monitoring of
information maintained in memory.
Neuroimaging Findings on Patients with OSA
In recent years, brain imaging studies have also provided some support for the
theory of prefrontal damage as a cause of dysexecutive function in OSA specifically
(Beebe & Gozal, 2002). Sangal and Sangal (1997) examined pre- and post-treatment
changes in the P300, an event-related brain potential (ERP) reflecting brain electrical
activity during the performance of cognitive tasks that normally occurs about 300
milliseconds after presenting of a stimulus. The authors used P300 as a general index of 44
cognitive processing, and prolonged latency as an indicator of abnormal cognitive processing. In comparison to 10 healthy controls, their group of 31 patients with OSA showed prolonged P300 latency and did not improve post CPAP treatment, despite significant improvement in sleep and respiratory variables. They concluded that OSA may cause pathophysiological cortical changes that are unrelated to sleepiness and resistant to treatment. Similar conclusions were reached by another ERP study (Kotterba et al, 1998). Again, they found that P300 latency was prolonged in 31 patients with
OSA before treatment and that neuropsychological testing showed there were impairments on measures of alertness, selective attention, and continuous attention, which were correlated with oxygen desaturation. After CPAP treatment, alertness and continuous attention improved but problems remained for selective attention, divided attention, and vigilance, together with the persistent prolonged P300 latency. They proposed that residual deficits after CPAP may reflect irreversible cerebral damage due to hypoxia. Kotterba and colleagues argued that as shown in other studies that the PFC and some subcortical structures, like the limbic system, are involved in the generation of these ERPs, ERP changes in patients with OSA may reflect hypoxic lesions in these regions.
Localization of pathophysiological changes to the frontal cortex as well as other areas of patients with OSA has also been reported in brain structural/chemical studies
(Alchanatis et al., 2004; Macey et al, 2002). Macey et al. found in a sample of 21 patients a reduction in gray matter bilaterally in parietal, frontal, and temporal cortices,
and unilaterally in other areas of the parietal, frontal, temporal cortices, hippocampus,
anterior cingulate gyrus, and superficial and deep cerebellar cortex. The authors stated 45
that damage to the frontal and temporal regions may contribute to cognitive deficits commonly reported in patients with OSA, consistent with the model proposed by Beebe and Gozal (2002). Morrell et al (2003) studied seven patients and seven healthy controls using a voxel-based morphometry technique to characterize structural changes in MRI data, and reported a significantly lower gray matter concentration in left hippocampus in patients.
To investigate the effects of hypoxia in different medical conditions, Gale and
Hopkins (2004) studied two group of patients, those with carbon monoxide (CO) poisoning and OSA. Eleven males and nine females with moderate to severe accidental
CO poisoning were compared with 12 males and two females with severe OSA (mean
RDI of 83.6). Magnetic resonance imaging and neuropsychological testing were conducted for both groups but only OSA patients were reassessed after six months of
CPAP treatment. Magnetic resonance imaging focused on ventricle-to-brain ratio (VBR) and hippocampal volume. Both groups showed hippocampal atrophy compared to a normative database (30% of CO poisoning patients and 36% of OSA patients). No OSA patients demonstrated any significantly increased VBRs, contrasting with generalized brain atrophy shown in 35% of patients with CO poisoning. A significant correlation was found in the OSA patients between hippocampal volume and partial pressure of oxygen in arterial blood, a direct measure of oxygen saturation. Mixed findings were found regarding correlations between hippocampal volume and neurocognitive test performance.
In the OSA group, there was also a correlation between some memory test scores (verbal list learning and visual recall of a complex figure) and hippocampal volume. When
evaluating the composite scores of neurocognitive tests, hippocampal volume was 46
significantly correlated with the non-memory composite scores, but not with the memory composite scores in both groups.
Negative findings have also been reported in neuroimaging studies in this area, however. O'Donoghue et al. (2005) compared 27 patients with severe OSA to 24 age-
matched controls using a voxel-based morphometry technique. They did not find any
differences between the two groups in terms of gray matter volume or focal structural
changes. They proposed that previous morphological changes found by Macey et al.
(2002) could have been associated with the comorbidities of the patients such as
neurological and psychiatric disorders, as well as the use of uncorrected thresholds for
multiple statistical comparisons.
Davies et al. (2001) investigated the presence of cerebrovascular disease-related
changes; namely, lacunae, high signal foci in deep white matter, and periventricular white
matter abnormalities in patients with OSA. Looking at deep white matter and
periventricular hyperintensity, they found high frequencies of subclinical cerebrovascular
disease in both the OSA group (33%) and carefully-matched controls (36%), despite the
higher arterial blood pressure in the OSA group. Discrepant findings were reported in a
study of sleep-disordered breathing and prestroke cerebrovascular disease in 78 patients
with acute stroke (Harbison et al., 2003). Computed tomography (CT) scans showed that
prestroke cerebrovascular disease was present in 63% of acute-stroke patients and was
associated with more severe sleep-disordered breathing. Severity of white matter disease
correlated with AHI and age, with the strongest associations in frontal and basal ganglia
areas. The authors attributed the discrepancies with Davies et al.'s study to the
differences in the age range as elderly patients typically show more cardiovascular 47
complications. It should also be noted that home respiratory studies were used in
Harbison et al.'s study and hence the central vs. obstructive nature of the breathing disruptions were not verified. Also, the inter-rater agreement for the CT scores was only fair and the authors did not control for multiple statistical comparisons. An ancillary study of the Sleep Heart Health Study showed that the degree of white matter disease in the cortex indicated in repeated MRIs was associated with increasing degrees of central sleep apnea, but not OS A (Robbins et al, 2005). Based on the temporal order of the sleep study and MRIs over the years, the authors suggested that sleep-disordered breathing may play a causal role in central nervous system disease. The specific mechanisms of central sleep apnea that could link it to the white matter damage and the degree to which that mechanism might also apply to OSA were not discussed.
Single Photon Emission Computed Tomography (SPECT) was also used in studying OSA (Ficker et al, 1997). Fourteen patients were examined with tracer administration between 2am and 4am during stage 2 of sleep. Five of the patients showed frontal hyperperfusion and hypoperfusion in the left parietal region as defined by clinical criteria. Both of these effects were reversed after CPAP treatment. In another ongoing SPECT study (Koves et al, 2001), 11 out of 17 patients with OSA before treatment showed frontal hypoperfusion, one showed frontoparietal hypoperfusion, and another one showed parietal hypoperfusion. Six patients were examined after CPAP treatment and regional hypoperfusion was normalized.
Several studies used Magnetic Resonance Spectroscopy (MRS) to study neurochemistry of patients with OSA. The first study by Kamba et al. (1997) compared
23 patients with 15 healthy controls using two-dimensional chemical shift imaging. All 48
of the participants had previously undergone MRI and did not show any structural abnormalities. A significantly lower N-acetylaspartate/choline (NAA/Cho) ratio in the cerebral white matter was found in patients with moderate to severe OSA than in patients with mild OSA and normal controls. The authors theorized that the reduced NAA/Cho ratio was suggestive of the presence of cerebral damage in patients as NAA/Cho ratio is considered a sensitive marker for non-specific cerebral metabolic changes. Another study by the same group showed that AHI was negatively correlated with the NAA/Cho ratio for periventricular white matter but there was no significant correlation for the cerebral cortex (Kamba et al., 2001). The authors suggested that the findings indicated lactate production, implicating anaerobic glycolysis in the deep white matter caused by hypoxia. Further studies are needed to investigate the relationship between hypoxia, arterial oxygen desaturation and cerebral metabolic abnormalities, as well as the consequences of these processes for daytime functions.
A few other studies also found cerebral metabolic changes in patients with OSA.
Alchanatis et al. (2004) reported significant decrease in N-acetylaspartate/creatine and choline/creatine ratios and significantly lower absolute concentrations of NAA and Cho in frontal white matter of patients with OSA as compared to healthy controls. Bartlett et
al. (2004) found a decrease in creatine-containing compounds in the left hippocampal
area in a group of eight patients as compared with five age-matched controls, suggesting hypoxic damage in OSA. A recent study investigated brain metabolism in 14 patients without cardiovascular and cerebrovascular comorbidities using Proton MRS ('H-MRS)
(Tonon et al., 2007). Participants were tested before treatment and six months after
CPAP usage on single voxel' H-MRS in the parietal-occipital cortex, overnight 49
polysomnography, Multiple Sleep Latency Test (MSLT), and neuropsychological testing.
Ten control participants were also tested using 'H-MRS, which was used to measure absolute concentrations of NAA, creatine, and Cho. Tonon and colleagues found that cortical NAA was significantly lower in patients than in controls before treatment, and
NAA level was also positively correlated with minimum SpC>2 and MSLT scores in patients. After treatment, patients' breathing disruptions were resolved and their sleep structure was restored. Daytime sleepiness and neuropsychological difficulties were also reversed. However, the reduction in cortical NAA persisted after therapy. The authors concluded that the patients showed cortical metabolic changes consistent with neuronal loss, despite the absence of comorbidities in the sample, and the improvements in the illness, daytime sleepiness and neuropsychological functioning. These persistent changes were hypothesized to represent the consequences of hypoxic damage induced by the breathing interruptions prior to CPAP therapy.
Thomas and colleagues (2005) conducted a pioneering study on patients with
OSA using functional Magnetic Resonance Imaging (fMRI) and a working memory task before and after CPAP treatment. Their behavioral data showed that patients with OSA had reduced working memory performance as compared with normal controls and also had diminished activation of the dorsolateral prefrontal cortex. Both the reduction in performance and neuroimaging activation did not increase after CPAP treatment, which was effective in reversing subjective sleepiness. A second fMRI study examined cerebral response to a verbal learning task, which is associated with function that was presumably preserved in patients with OSA (Ayalon et al., 2006). Their hypothesis that intact performance in patients correlates with increased brain activation in areas involved in 50
compensatory recruitment on a verbal learning task in sleep deprivation studies was supported. Compared with age, BMI, blood pressure, and education-matched controls, patients with untreated OSA showed increased activation in bilateral inferior frontal and middle frontal gyri, cingulate gyrus, inferior parietal and superior temporal lobes, thalamus and cerebellum. The authors hypothesized that OSA, like aging, involves gradual structural and physiological changes in the brain, and therefore additional brain regions may be recruited to cope with cognitive challenges, suggesting global functional reorganization processes. The more long-term OSA-related changes may explain why the increased activation appeared to be more widespread than in studies of total sleep deprivation in healthy adults.
Prefrontal Cortex (PFC) and Sleep
Another line of evidence that implicates the PFC in changes in cognitive function due to OSA relates to the specific relationship between PFC and sleep. Distinct from other brain regions, PFC shows reduced activity in functional neuroimaging across all sleep stages (Beebe & Gozal, 2002). It is also reported to be functionally disconnected from its daytime interconnected brain regions during sleep (Braun et ah, 1998; Braun et ah, 1997; Hobson et al, 1998). Some authors have proposed that these unique phenomena of the PFC may be due to its need to recalibrate its circuits after a day of
"hard work" (Dahl, 1996; Harrison & Home, 2000a; Home, 1988). It is therefore proposed that PFC is one of the first brain regions to suffer from the sequelae of sleep disturbance. Consistent with this notion, EEG findings suggest that frontal regions are more sensitive to sleep deprivation and recovery sleep (Finelli et al, 2000). 51
Convergent evidence for the vulnerability of PFC to sleep disturbance can be found in sleep deprivation studies. Deficits in performance on cognitive tests pertaining to PFC have been shown in many sleep deprivation studies (Harrison & Home, 1999;
Harrison & Home, 2000a). For example, temporal memory, as measured by recency discrimination of two sets of faces presented at different time points was found to be specifically impaired after 36 hours of sleep deprivation while recognition was preserved
(Harrison & Home, 2000b). The finding that caffeine did not have a significant effect on the temporal memory performance was interpreted by the authors as evidence supporting the idea that sleep deprivation has more specific effects on cognitive functions, other than generalized reduction in arousal or increased sleepiness. Nilsson and colleagues (2005) examined executive function, psychomotor vigilance and verbal and visual-spatial working memory of healthy young adults after 31-32 hours of sleep deprivation. They reported that their participants showed significant impairments on an executive function test (Six Element Test), which was composed of storytelling, simple arithmetic, and object naming, but performance comparable to that of the control group on the measures of psychomotor vigilance and basic working memory as measured by short delayed recall of a list of 10 words (verbal) and of 16 locations of a circle presented consecutively on a screen (visual-spatial). Corroborative data for the vulnerability of executive function and the neurocircuitry subserving executive function also come from neuroimaging studies, which show changes in prefrontal metabolism and neurochemistry in healthy adults after sleep deprivation (Dorsey et al, 2000; Thomas et al, 1998; Thomas et al, 1993).
Studies using functional neuroimaging have also shown that sleep-deprived individuals showed poor performance on working memory tasks and reduced activation of PFC, a 52
region that subserves attention and higher cognitive functions, and the related cortico thalamic network, involved in alertness and attention (Drummond et al., 1999; Thomas et al., 2000). A more recent study found a stronger linear increase in activation to increasing task demands on a grammatical transformation task measuring logical reasoning after 36 hours of total sleep deprivation than the control condition. These
compensatory responses were found in bilateral inferior parietal lobes, bilateral temporal
cortex, and again, left inferior and dorsolateral PFC (Drummond et al, 2004).
Reversibility of Daytime Dysfunctions with Treatment of CPAP
CPAP is effective in improving subjective and objective measures of sleepiness in
patients with severe OS A (Patel et al, 2003). Therapeutic effects of CPAP on sleepiness
in OSA range from moderate to large (Engleman & Douglas, 2004). Using the Epworth
Sleepiness Scale (ESS), for example, improvement effect sizes range from 0.9 (Douglas,
1998) to 1.5 (Jenkinson et al, 1999) and even 2 (Ballester et al, 1999). The average
improvement for patients with severe OSA was 4.75 points on the ESS (Patel et al).
Large effect sizes were reported in a double-blind, randomized, controlled trial
comparing constant CPAP (cCPAP)4 and autoadjusted CPAP (aCPAP)5 (Nussbaumer et
al, 2006). The mean decrease in ESS score was 6.6 for both cCPAP and aCPAP, and the
effect sizes were 1.99 and 2.07 respectively. Hardinge and colleagues (1995) reported
that patients with severe OSA showed reduced daytime sleepiness after two months of
4 In constant CPAP, the therapeutic mask pressure is determined manually during in-laboratory titration. There is a tradeoff between pressure-related side effects and efficacy in abolishing airway obstruction. 5 Auto-adjusted CPAP is designed to address the variability of pressure requirement due to transient factors like body position, sleep stage, and more systemic factors like changes in body weight. Pressure is adjusted automatically by feedback control according to patterns of pressure, flow, and other signals. Essentially, pressure increases during obstructive events and decreases when breathing is normal. 53
CPAP treatment and the effects were sustained at one-year follow-up, as measured by the
ESS. A recent meta-analysis examined seven randomized controlled trials of the effects of CPAP on sleepiness in mild to moderate OS A (AHI: 5-30) (Marshall et al, 2006).
ESS improved by 1.2 points (effect size 0.27), MWT latencies improved by 2.1 minutes
(effect size 0.21), but no significant changes in MSLT latencies were detected. The authors concluded that the effect sizes were very small and might not be clinically relevant. A recent study reported a reliable change index for the ESS to be 5.89 as calculated by a formula with the mean change score, the standard deviation, and the reliability coefficient (Smith & Sullivan, 2007). Without an index for clinical significance to date, this reliable change index will be used in assessing the change of
ESS scores in our OSA group, in addition to the comparison with our control group.
While CPAP decreases sleepiness, it is not uncommon to find patients who continue to experience excessive daytime sleepiness with CPAP treatment (Kingshott et al., 2001). A study specifically selected this group of patients and investigated the effects of modafinil, which was shown to be effective in reducing sleepiness in patients with narcolepsy and in sleep-deprived individuals. This randomized, double-blind, placebo- controlled crossover trial showed that modafinil did not have effects superior to placebo on sleepiness (ESS and MSLT) but improved wakefulness (MWT) by an average of 1.7 minutes. The clinical significance of this improvement is unlikely. The authors concluded that modafinil may be of limited use for patients with OSA.
Cassel et al. (1996) reported a significant reduction in the accident rate of patients with OSA after one year of CPAP treatment. Mazza et al. (2006) reported that performance on a naturalistic driving test in patients with OSA were normalized after 54
CPAP treatment, except that patients still had a longer RT on the driving simulator. Hack et al. (2000) examined simulated steering performance in a randomized, double-blind, controlled trial of CPAP for patients with OSA and demonstrated that deviations from road positions, deterioration in steering position over time, and reaction time to target stimuli decreased after CPAP versus subtherapeutic control treatment. However, the authors did not find a significant correlation between decreases in sleepiness and improvements in steering performance in the therapeutic CPAP group, suggesting that reduced vigilance resulting from sleepiness may not be the sole factor that impairs driving performance. Mulgrew et al. (2007) showed that patients' self-reported work limitations were significantly reduced after CPAP treatment. However, due to the lack of follow-up measure of sleep and sleepiness, it was unclear whether the improvements in self-perceived work limitations were related to improved sleepiness, or a reduction of nighttime breathing events.
The effectiveness of CPAP in reversing cognitive deficits varies across different domains and across studies (Aloia et al, 2004; Jones & Harrison, 2001). In general, most studies report some improvement in attention/vigilance (Alchanatis et al, 2005; Bonnet,
1993; Douglas, 1998; Montplaisir et al, 1992; Sanchez et al, 2001; Schneider et al,
2004), executive functioning, and memory, although changes are often inconsistent between studies or within a study between tests even within a cognitive domain (e.g., executive function) (Saunamaki & Jehkonen, 2007; Weaver, 2001). Engleman et al.
(2000) found in a meta-analysis of 98 patients from six case control studies that while cognitive outcomes did improve on CPAP compared to placebo, the effect sizes were small, although this may have been related to the relatively mild nature of the disease in 55
the groups studied. The small effect sizes of therapeutic improvements in cognitive function were contrasted with the often substantial improvement in sleepiness and quality of life scores in patients with OSA (Engleman & Douglas, 2004). Several investigators have highlighted the resistance of executive dysfunction to CPAP treatment, and proposed that since these changes appear more related to the symptoms of hypoxia, executive dysfunction reflects the underlying more permanent pathophysiological damage due to hypoxemia as discussed below (Bedard et al., 1993; Ferini-Strambi et al.,
2003; Feuerstein et al, 1997; Naegele et al, 1998; Naegele et al., 1995). Bedard et al. studied 10 patients with moderate to severe OSA before treatment and after six months of
CPAP treatment. They found significant deficits in a variety of cognitive functions, which were mostly eliminated to the performance level of the 10 control participants after
CPAP treatment, with the exception of scores on some standard tests of executive function such as Trail Making Test and verbal fluency. In their 1998 study, Naegele et al. followed up 10 patients before and after CPAP and compared them to 10 control participants. The two groups differed significantly in multiple areas of neurocognitive function before treatment. All differences were normalized after four to six months of
CPAP treatment, except for the persistence of short-term memory impairment measured by digit and spatial spans. They proposed that there was frontal contribution to those memory impairments and that the impairments were maintained by hypoxemia and apneic episodes. Ferini-Strambi and colleagues investigated the effect of CPAP on cognitive function after both short (15 day) and longer (four month) periods of treatment in a group of patients with severe OSA. They reported that short-term treatment of CPAP improved sleepiness and cognitive functions including sustained attention capacities, 56
visuospatial learning, and motor abilities. However, executive function as measured by phonemic verbal fluency and the Stroop test remained impaired. This pattern of results persisted after four months of treatment. Another study compared the impact of CPAP and surgical treatment to conservative management (Lojander etal., 1999). Twenty- seven and 23 male patients with moderate OSA were considered as candidates for CPAP treatment and UPPP respectively. Within the groups, patients were randomly allocated to receive the active treatment or conservative management. Cognitive function and illness severity were assessed before treatment, and 3 and 12 months after treatment. They found that CPAP could alleviate both oxygen desaturation and daytime somnolence while
UPPP was partly successful in improving oxygen saturation and could effectively reduce somnolence. However, there was no correlation among somnolence, oxygen desaturation and cognitive functions, which did not improve after treatment.
A recent study specifically investigated working memory in OSA and the effects of CPAP treatment (Felver-Gant et ah, 2007). The authors reported that "high treatment adherers" (i.e., CPAP use more than four hours per night) demonstrated improved performance on a verbal 2-back task and on the Paced Auditory Serial Addition Test
(PASAT) (N=24) after three months of treatment, in comparison to "low adherers"
(N=27). They reported that there were no main effects for any other cognitive measures adopted in the study including the Hopkins Verbal Learning Test - Revised, the Trail
Making B, and the Grooved Pegboard test. However, there were several potential concerns with the study, and three major issues are discussed here. First, the conclusion that the high adherers demonstrated improvements on the working memory tasks was based on an interaction effect between treatment (pre vs. post) and adherence (high vs. 57
low). Taking a closer look at the interaction effects, while the high adherence group showed a 3% increase in accuracy on the 2-back task, the low adherence group showed a
7% decrease in their performance. The same pattern applied to the other working memory task they used. The high adherence group showed a 0.005% increase in accuracy on the PASAT and the low adherence group had a 0.03%. This pattern of performance raises at least two questions. The first one is whether the high adherence group really can be said to have had a significant improvement in their performance, especially given that the extent of reduction in performance in the low adherence group was larger than the improvement in the high adherence group. The second question, which the authors also identified, is the puzzling reduction of performance itself in the low adherence group after CPAP treatment. The authors suggested that it could be due to some unknown deleterious effects of CPAP treatment when it is used only partially. If this explanation is valid, the conclusion of this study that CPAP is effective in improving working memory in patients with OSA could be erroneous. Instead, the high adherence group might have just avoided the harmful effects of partial CPAP use on working memory performance. A more probable explanation might be that the high and low adherence groups differed on some undetected variables, which were associated with the reduction in performance on the working memory tasks in the low adherence group. The second major issue with the study was that they did not have a healthy control group and they mentioned that both of the CPAP adherence groups had a 2-back accuracy "within normal limits" at both baseline and three-month post treatment as compared with findings of another study (Naegele et ah, 2006). The authors failed to provide reasons for the better performance of their patients before and after treatment than healthy controls of 58
another study, and in the absence of a matched control group before or after treatment, it would be problematic to draw any conclusions on treatment effects if the patients did not show difficulties with the 2-back task in the first place. Another puzzling nature of the findings of this study was that there were no significant improvements of any other cognitive measures that were adopted to control for the component cognitive processes of the 2-back task. The authors simply concluded that working memory tasks were more sensitive to effects of treatment than the selected component cognitive tasks. Not only was this reasoning illogical, this is also contradictory to the accumulated evidence in the literature that executive dysfunction, if anything, is more resistant to treatment effects than the component processes like basic attention and memory in patients with OS A.
The current study attempted to study working memory in patients with OSA treated with
CPAP using a more sophisticated design and a matched healthy control group.
So far, only two studies have adopted a credible placebo control treatment
condition such as subtherapeutic or sham CPAP in evaluating impact of treatment on neurocognitive impairments. Henke, et al. (2001) compared CPAP treatment with
subtherapeutic CPAP-controlled treatment in a randomized cross-over study of 46 patients with a mean AHI of > 60. They found that the placebo treatment did not have an
effect on any of their neuropsychological outcome measures. When the two groups were
assessed after effective CPAP treatment for a minimum of 12 days, there were significant
improvements on neuropsychological measures including Digit Span Forward and
Backward, Complex Figure Recall, and Digit Symbol. However, when they compared
the two groups when one was receiving effective CPAP and the other was using
subtherapeutic CPAP, there were no significant differences between the groups on any 59
measure. At the end of the study, the CPAP group showed significant improvements on a few more neuropsychological measures than the placebo group that started effective
CPAP later, suggesting that the patients using CPAP for the whole study period benefited more from the longer treatment duration although this was not supported by a correlation between length of treatment and neuropsychological performance. The other placebo- controlled study reported that their CPAP group showed better overall cognitive functioning posttreatment than the sham CPAP group, although when considering individual tests, the two groups only showed significant differences on one (digit vigilance) out of 22 neuropsychological test scores (Bardwell et ah, 2001). Also, whether the sham treatment at an insufficient CPAP pressure was a valid placebo is questionable because the sham group's AHI was reduced from 44 to 28. Although it is above the clinical cut-off for OSA, the treatment might have indeed improved the condition to some extent. In fact, there appeared to be some improvements in all the sleep variables that were measured in the "placebo" group but unfortunately, the authors did not report analyses of the significance of these changes. These two placebo studies also have the caveat of a short duration of treatment, with one week for both groups in
Bardwell et a/.'s study and less than one month for the placebo group and two months for the CPAP group in Henke et a/.'s study. Despite all these limitations, these placebo studies did raise some valid challenges to the efficacy of CPAP treatment in restoring cognitive function (Sateia, 2003).
It has been noted, however, that the inability to see change with repeated measurements of some cognitive functions, especially executive function may be due to the nature of the tests themselves since they are designed to measure behavior under 60
conditions of novelty, and strategic planning. It seems possible that, with repeated exposure, these tests may lose sensitivity as executive function tests (Jones & Harrison,
2001). Thus, the investigation of the reversibility of executive function post-CPAP may require a different approach with more sensitive measures (Weaver, 2001). Lojander et al. (1999) argued that the neurocognitive tests are insufficiently sensitive to identify changes in patients with OSA, particularly among high-functioning individuals.
To summarize the findings on reversibility of executive deficits specifically, improvements were found in cognitive flexibility, speed, and non-verbal planning, based on five studies that have a healthy control group at baseline (Saunamaki & Jehkonen,
2007). Most studies relied only on pre- and post-treatment comparisons in assessing the effects of CPAP on cognitive function. This methodology requires repeated exposure of participants to the same neurocognitive tests, creating potential practice effects especially for executive tasks. The lack of a control group post-treatment also renders these findings questionable as the changes could be due to extraneous factors in addition to the practice effects. Comparing CPAP treatment with placebo or conservative treatment, only three out of nine studies found improvements in cognitive performance. Taken together, the literature to date suggests that working memory deficits persist after CPAP treatment (Saunamaki & Jehkonen), but further work that uses more sensitive and specific measures and a healthy control group is needed in order to draw meaningful conclusions about the residual cognitive deficits of individuals with OSA after treatment.
It is especially important to characterize residual executive deficits after CPAP treatment of sleep disruption, as it may help identify remaining chronic issues that continue to interfere with an individual's recovery and return to normal performance in 61
meaningful activities such as work and driving. Specific identification of these attention/executive function deficits is critical for the measurement of treatment outcome and the development of appropriate cognitive rehabilitation programs (Levine et al., 2000;
Sturm et al, 1997). An understanding of the cognitive deficits may also help elucidate associated neuropathological mechanisms that respond differentially to treatment.
Executive dysfunctions in OSA include deficits in behavioral inhibition, set-shifting, affect regulation, control of arousal, analysis/synthesis, contextual memory, and working memory (Beebe & Gozal, 2002). The current study attempted to clarify specific residual deficits of treated individuals with OSA using current concepts of attention and executive function as developed by Baddeley and colleagues (Baddeley, 1996a, 1996b; Baddeley &
Delia Sala, 1996) in the theoretical model of working memory.
Theory of Working Memory
Baddeley's theoretical concept of working memory proposes that it is a limited capacity store that temporarily maintains and processes information crucial for thought, planning and action (Baddeley & Hitch, 1994). The functioning of working memory has been linked to such complex processes as language comprehension and reading, learning and memory, counting and mental arithmetic, syllogistic reasoning, visual imagery, and dynamic perceptuo-motor control (Baddeley, 2003; Baddeley & Logie, 1999). In
Baddeley's model, working memory is not a unitary system, but comprises multiple specialized subcomponents, including systems for temporary maintenance and processing of verbal material (called the phonological loop), visual/spatial information (visuospatial sketchpad), a supra-modality attentional controlling mechanism, termed the central 62
executive, and an episodic buffer that provides a limited capacity multi-modal interface between systems (Baddeley, 2003). Each of these subcomponents can be fractionated into further subsystems (detailed below). Use of this working memory model has a number of advantages for this study. First, the model has been investigated extensively in the cognitive neuroscience community and there have been significant advances in specifying the cognitive processes and tasks to allow testable hypotheses to be formulated for patient populations. Secondly, considerable knowledge has also accumulated concerning the neurophysiological underpinnings of working memory (Smith & Jonides,
1997), which allows one to make specific predictions about the pathophysiological mechanisms based on neuro-cognitive associations. Finally, the functioning of working memory pervades all aspects of on-line cognition important for humans "to comprehend and mentally represent their immediate environment, to retain information about their immediate past experience, to support the acquisition of new knowledge, to solve problems, and to formulate, relate and act on current goals" (Baddeley & Logie, 1999).
Thus, investigation of the integrity of working memory would have potential significant relevance for patient functioning in their everyday environment.
As stated above, working memory comprises a number of subcomponent processes (Figure 2) (Baddeley, 2000). The phonological loop consists of two components: A limited store for verbal memory traces, and an articulatory rehearsal process that refreshes and maintains the contents of the store. Most studies of the verbal store assess the retention of item sequences, such as digits, letters or words. The visuospatial sketchpad contains a visual cache that functions to maintain visual and/or spatial information, as well as an "inner scribe" that is used for retrieval and rehearsal 63
(Logie, 1995). Typical tasks include immediate recall of visual and spatial information.
Finally, the central executive refers to the processes involved in the attentional control of the verbal and visual slave systems, making use of the episodic buffer. The episodic buffer is a relatively new addition to the model, and was formulated to take into account the need for the central executive to bind information from the separate storage areas into integrated episodes that are accessible to conscious awareness (Baddeley, 2000, 2001).
The episodic buffer acts as a global workspace, or a temporary storage component of the central executive. Baddeley has adopted the Norman and Shallice model of attentional control to more fully characterize the central executive. In this model, control is divided between two processes, one based on habit patterns or schemas, guided by environmental cues (routine intentions) and the other an attentionally demanding controller, the supervisory activating system (SAS) that intervenes when routine control is misguided, insufficient or inadequately specified (Shallice, 2002). The SAS thus functions as the attentional controller or central executive. Baddeley has postulated that the role of the central executive is to focus attention against potentially distracting irrelevant information, to switch attention between two or more sources of information or responses, to divide attention in order to perform two concomitant tasks, and to interface with long-term memory (Baddeley, 2001, 2002).
The generality and separability of these functions and subcomponents have been investigated and supported in cognitive paradigms with normal participants and in neuroimaging studies, as well as with different patient groups (Baddeley & Hitch, 1994).
Neuroimaging studies have provided consistent support for the existence of modality-
specific rehearsal and storage systems (Prabhakaran et al., 2000). The verbal "slave 64
systems" are found to be located in left inferior frontal Broca's area and premotor area
(rehearsal) and left posterior parietal cortex (storage), working in concert with a central executive component in the bilateral dorsolateral prefrontal cortex involved in manipulation and processing in a variety of different tasks (Becker et al, 1994; Petrides et al, 1993; Postle et al, 2005; Smith et al, 1998). Other studies have identified a parallel system for visuo-spatial rehearsal (anterior frontal) and storage (posterior parietal and temporal) components in the right hemisphere that recruit similar bilateral, but predominantly right dorsolateral prefrontal areas during tasks drawing upon central executive functions in handling increased demands of load, duration or manipulation (e.g. integration of information) on working memory (Goldberg et al, 1996; Smith & Jonides,
1997). This dorsolateral prefrontal area is recruited by diverse cognitive tasks, as would be expected given the variety of roles of the central executive (Duncan & Owen, 2000).
Neurological damage involving frontal systems is also associated with deficits in central executive function as measured with dual-task performance in individuals with
Alzheimer's disease (Baddeley et al, 1991; Baddeley et al, 1986), Parkinson's disease
(Dalrymple-Afford et al, 1994), and traumatic brain injury (McDowell et al, 1997).
Dual-task performance in Alzheimer's disease was shown to decline over time, despite the maintenance of single task performance (Baddeley et al, 1991) and was related to impairments in autobiographical memory as the disease progressed (Greene et al, 1995).
Further studies have also shown that the dual-task performance effect in Alzheimer's disease is not simply related to increased susceptibility to task difficulty and/or reduced speed of processing but that the decline in performance due to the requirement to perform two tasks simultaneously is a deficit qualitatively different from that shown by increased 65
difficulty of a single task (Baddeley, 2002). Central executive deficits have been shown to correlate with ratings of dysexecutive behaviour in patients with traumatic brain injury, and to predict response to rehabilitation (Baddeley, 2002; Baddeley & Delia Sala, 1996).
Purpose of the Present Study
The purpose of this study was to systematically investigate the executive function and psychosocial outcomes of individuals with OS A treated with CPAP for at least three months. This study emphasized the following: The use of a well-validated conceptual model of working memory to examine executive functioning; the use of tasks that reflect the isolable subcomponents of the working memory framework based on results from both the cognitive science and neuroimaging literature; and investigation of the association of these cognitive findings with everyday functioning. Beebe and Gozal
(2002) stated that dissociating effects on executive function from those on more basic functions would strengthen research in this area, and the working memory model serves this objective in the present study. Finally, working memory capacity has been linked to other cognitive tasks such as learning, reasoning and comprehension and thus these findings will have relevance for a broad range of functions in everyday activities.
I have chosen to study individuals already on stable CPAP treatment in comparison to normal controls, in order to eliminate, as much as possible, the influence of sleepiness on cognitive function, and thus focus on the underlying, presumed more permanent, pathophysiological sources of cognitive impairment. Given the strong link between these tasks and identified brain regions, these results also could have implications for further hypotheses regarding the localization of brain dysfunction due to 66
OS A. It is also imperative to fully understand the effectiveness and outcome of CPAP treatment in the long term, in order to identify other needed interventions (e.g., cognitive rehabilitation) and to provide more precise information to other health care providers and agencies, as well as to patients and families regarding long-term prognosis.
I also intentionally did not include pre-treatment neurocognitive testing because the focus of this study, executive function, is highly susceptible to learning effects
(Decary et al, 2000). In order to maintain the novelty and hence validity of the tests, all participants were chosen only if they had not been exposed to any of the assessment instruments. Widespread learning effects have been detected in other cognitive measures in patients with OSA, despite intensive familiarization efforts (Engleman et ah, 1994).
Also, our research question was whether patients with OSA under treatment with CPAP perform any differently from healthy individuals. Therefore, it was more important to compare treated patients with a healthy control group rather than with their pre-treatment condition.
Nevertheless, in order to ensure that the patients were indeed properly treated with CPAP, I collected data on their pre-treatment status (i.e., PSG data, sleepiness scores, subjective sleep quality) and compared them with post-treatment data. I also asked patients to rate their pre-treatment functional outcomes and quality of life retrospectively to understand if and to what extent they perceived CPAP helped in terms of daytime functions.
My hypotheses were:
1. CPAP will be effective in improving respiratory (RDI) and hypoxemia (mean and
minimum SpOa) indices, subjective sleep quality, daytime sleepiness, functional 67
outcomes, and quality of life, as indicated by improvements in these outcome
measures in post-CPAP evaluation as compared to pre-CPAP evaluation.
2. The performance of the stably treated OS A group will be comparable to normal
age-matched controls on tests of basic storage and rehearsal components in working
memory.
3. In contrast, the stably treated OS A group will show worse performance than normal
controls on tests that demand the involvement of the central executive.
4. Patients will show comparable performance to controls on neuropsychological
testing except for executive function.
5. Patients' executive function results will be predicted by their diagnostic respiratory
and hypoxemia indices before CPAP treatment.
6. The results of tests of executive function of patients will correlate with their
perceived cognitive efficiency and quality of life.
Exploratory analyses were also conducted to examine:
7. The predictors of patients' current psychosocial functioning. The outcome
variables will include the mood measures, subjective daytime functioning, and
quality of life. Predictors used will be age, BMI, diagnostic RDI, diagnostic
minimum Sp02, post-treatment sleep efficiency, current sleepiness (ESS), and
subjective sleep quality (PSQI).
8. The correlation of quality of life measures with other psychosocial variables 68
CHAPTER TWO: METHODS
Participants and Design
OSA Group
Since most of our tasks have not been used in this population before, a power analysis could only be calculated on an estimated effect size. For a medium effect size of executive function tasks, with power set at .80 (D = .02) and D at .05, a sample size of 30 was needed. Thirty-seven individuals with OSA treated with CPAP were recruited from the Sleep Disorders Clinic at the QEII Health Sciences Centre. Individuals who met the criteria were identified by the attending physicians who asked for permission to pass their contact information to the research team. Invitation letters were also sent out to individuals who had been treated at the clinic before. People who called us to express interest were screened for suitability on the phone. If they passed the telephone screening, they were asked to give us permission to review their chart at the Sleep Disorders Clinic to enable a decision as to their eligibility. Some participants in the OSA group also learned about our study through posters at the Sleep Disorders Clinic and respiratory companies.
Inclusion criteria included:
1. Diagnosis of moderate to severe OSA by physicians at the clinic using
polysomnography (PSG) with Apnea/Hypopnea Index (AHI) or by home studies
with Respiratory Disturbance Index (RDI) of at least 15 (per hour) and subjective
symptoms of non-restorative sleep as indicated by excessive daytime sleepiness,
fatigue, or functional impairment. It has been shown that home studies provide a
comparable measure of respiratory disturbance to in-laboratory PSG with a high 69
correlation between the RDIs obtained in the two methods in the same patients
(r=0.95) (Adams etal, 2001).
2. Treatment with CPAP for at least three months with compliance for at least four
hours per night on 80% of the week; compliance was monitored by a built-in time
counter and in cases of unavailability due to CPAP model constraints, by patients'
self-report.
Exclusion criteria included:
1. Concurrent diagnoses of other sleep pathologies or significant medical condition or
procedure (e.g., chronic obstructive pulmonary disease; sudden cardiac arrest;
coronary artery bypass graft (CABG) surgery, active phase of cancers) or
neurological disorder, past head injury with loss of consciousness, any other
neurological conditions which might cause cognitive impairment
2. Current alcohol or drug abuse (self-report)
3. History of severe psychiatric illness or current psychiatric illness that began more
than ten years before the diagnosis of OSA
4. Current use of medication that could affect cognitive function (e.g. psychotropics,
benzodiazepines)
5. Current OSA treatment other than CPAP.
Controls
Twenty-seven age- and education-matched healthy controls were recruited from the community.
Exclusion criteria included:
1. Evidence of sleep pathology or disorder based on clinical interview and the Sleep 70
Disorders Questionnaire (Douglass et al, 1994a; Douglass et al., 1994b), or PSG
after being recruited in the study.
2. Evidence of having high risk for OS A in the last month as indicated by responses on
the Berlin Questionnaire, using the cut-off score found to be sensitive and specific in
separating high and low-risk individuals (Netzer et al., 1999). Participants found to
have elevated RDIs (i.e., >15) were also excluded from the study.
3. Significant medical condition or procedure (e.g., chronic obstructive pulmonary
disease; sudden cardiac arrest; coronary artery bypass graft (CABG) surgery, active
phase of cancers) or neurological disorder, past head injury with loss of
consciousness, any other neurological conditions which might cause cognitive
impairment
4. Current alcohol or drug abuse
5. Current diagnosis of psychiatric illness
6. Current use of medication that could affect cognitive function (e.g. psychotropics,
benzodiazepines)
Design
The study was a between-group (OS A vs. healthy control) Analysis of Variance
(ANOVA) design, with some within-subject variables on the working memory tasks and neuropsychological tests with several conditions (i.e., Trail Making, Consonant Trigrams,
Digit Span, and Visual Span). The OSA group and controls were compared on neuropsychological tests, working memory tasks, as well as measures of sleep, mood, and functional outcomes. Relationships between the neurocognitive measures, psychosocial outcomes, and sleep variables as measured by questionnaires and PSG for patients were 71
also investigated. To ensure that CPAP was used effectively, pre-treatment data of PSG, ratings of sleepiness and of subjective sleep quality, and quality of life measures were compared to post-treatment data within the patient group.
Procedures
Approval of the research protocol was acquired from the Research Ethics Boards of the Capital District Health Authority and Dalhousie University. After the participants were identified and passed the initial screening on the phone and through chart review, they were invited to come into the laboratory to complete the consent procedure and questionnaires related to sleep, mood, and daily functioning. If control participants
showed high risk for presence of OSA on both the Sleep Disorders Questionnaire and the
Berlin Questionnaire, they were excluded from the study. There were two sessions of neurocognitive testing. Session one took place in the morning and involved the working memory experimental tasks. Depending on the performance and pace of the participant,
the morning session took about two to three hours. Session two was conducted in the
afternoon and consisted of the neuropsychological battery, which took about three hours.
The order of testing was fixed to control for circadian effects (Engleman et al, 1999;
Engleman et al, 1994). All participants also received an overnight PSG within eight
weeks of the testing. For patients, the procedure was to characterize the effectiveness of
CPAP treatment in terms of nighttime respiratory disturbances, SpC>2, and sleep
architecture. For controls, PSG was necessary to verify the absence of sleep disorders, as
recommended by Saunamaki and Jehkonen (2007). Participants were reminded not to
consume any alcohol or caffeinated products on the day of PSG. Nevertheless, in order
to measure participants' performance that reflects their daily functioning, there was no 72
caffeine restriction on the day of testing and participants were instructed to maintain as much of their daily routine as possible (e.g. bed-time the night prior to testing, wake-time on the day of testing, diet, etc). All sleep studies were scored by registered polysomnographic technologists using standard scoring criteria (Rechtschaffen & Kales,
1968), and reviewed by physicians at the clinic. In case of ambiguous studies, a scoring
from a second technologist was obtained. The scores endorsed by sleep physicians were used.
Protocols were developed to manage cases with elevated RDI or signs of other
sleep disorders in sleep studies. For individuals who had sub-optimal PSG even with their CPAP machine, their case physician at the Sleep Disorders Clinic was informed.
The physicians reviewed the study and determined follow-up actions with the patient.
Four individuals had post-treatment RDI > 5 but < 15, and one individual had post- treatment RDI > 15 and hence was excluded from the study. For control participants who
showed RDI > 5 (9 with RDI > 5 and < 15; and 7 had RDI > 15 and hence excluded from
the study), the study was reviewed by a sleep physician. The participant was then
informed of the major findings of the sleep study and the physician's recommendation.
With the participant's consent, the PSG report with the sleep physician's interpretation
and comments was sent to his or her family physician for clinical management.
Measures and Tests
Screening
Two questionnaires were used to screen for sleep pathology in control participants
(Appendix A). 73
Sleep Disorders Questionnaire (SDQ) (Douglass et al, 1994a)
The SDQ was used to screen control participants for presence of any sleep disorders. Participants considered each question as applying to the past six months6 on an anchored-scale ranging from 1 (never/strongly disagree), 2 (rarely/disagree), 3
(sometimes/not sure), 4 (usually/agree), to 5 (always/agree strongly). The questionnaire includes four diagnostic scales of 175 questions measuring Sleep Apnea, Narcolepsy,
Psychiatric Sleep Disorder, and Nocturnal Myoclonus respectively. High test-retest reliabilities of all four scales have been demonstrated, ranging from 0.753 to 0.848
(Douglass et al, 1994b). Specifically, it shows excellent concurrent validity with PSG in identifying OSA. The OSA patients also completed the SDQ for ruling out sleep disorders other than OSA.
Berlin Questionnaire (Netzer et al, 1999)
This is a validated measure to identify individuals with sleep apnea. It contains
10 items with three domains of questions, addressing the presence and frequency of
snoring behavior, waketime sleepiness or fatigue, and history of obesity or hypertension.
Individuals with persistent and frequent symptoms in any two of these three domains are
considered to be at high risk for sleep apnea. Netzer and colleagues showed that the
Berlin Questionnaire had high internal consistency, and its risk grouping was useful in predicting a respiratory disturbance index (RDI) of >5, with a sensitivity of 0.86, a
specificity of 0.77, and a positive prediction value of 0.89.
6 For people who started CPAP less than six months prior to enrollment in the study, I asked them to adjust the timeline from "the past six months" to "since CPAP" as they reported their recent and current condition. 74
Hospital Record
The researcher reviewed the medical charts of the patients. Demographic information as well as history of OSA, other relevant medical history and medications were recorded. Data from their pretreatment PSG were also gathered from the chart.
(Appendix B)
Clinical and Demographic Variables
Participants were interviewed about their general health condition by a trained interviewer at the screening session. All participants were asked about their years of education and vocational background, height, weight, major medical and neurological conditions (e.g. heart disease, cancer, thyroid problems, stroke, multiple sclerosis, epilepsy, Parkinson's Disease, Huntington's Disease), psychiatric illness, previous history of head injuries, unconsciousness, medication use, drug, alcohol, and caffeine consumption. Additional questions were asked of patients on duration of illness or symptoms of OSA (if known), and treatment duration, as well as compliance. Refer to
Appendix C for the interview schedule.
Sleep Questionnaires (Appendix D)
Epworth Sleepiness Scale (ESS) (Johns, 1991, 1992).
All participants completed the ESS. The ESS is a self-administered eight-item questionnaire for measuring daytime sleepiness by rating the chances of dozing off or falling asleep in eight different situations commonly encountered in daily life.
Participants rated their tendency to fall asleep in each situation from 0 (would never doze), 1 (slight chance of dozing), 2 (moderate chance of dozing), to 3 (high chance of
dozing) in recent times. Total scores could range from 0 (not sleepy at all) to 24 75
(extremely sleepy), with 9 being the upper limit for normal scores. It has been shown that total ESS scores significantly distinguished normal participants and patients in various diagnostic groups including OSA (Johns, 1992, 2000). An ESS score larger than
10 is conventionally considered as clinically significant (Hartenbaum et al, 2006). In patients with OSA, ESS scores also correlate significantly with the RDI and the minimum
Sa02, and ESS is a validated clinical and research tool in assessing daytime sleepiness
(Hardinge et al., 1995; Johns, 1993). In addition, its test-retest reliability (r = 0.82) and internal consistency (Cronbach's alpha = 0.88) have been established (Johns, 1991). In the present study, ESS scores of the control and patient groups were compared with each other, and were correlated with neurocognitive performance, as well as the sleep variables measured by PSG. In addition, current ESS scores were also compared with the pre-treatment ESS scores, which were collected routinely in clinical appointments and recorded in patients' chart.
Stanford Sleepiness Scale (SSS) (Hoddes et al, 1973)
The SSS is a seven-point anchored scale with descriptors of varying sleepiness levels from fully awake to fully asleep. Participants were asked to choose from the descriptors the level that best defines his/her current state from 1 ("feeling active and vital; alert; wide awake") to 7 ("almost in reverie; sleep onset soon; losing struggle to remain awake"). This provided an instant picture of the sleepiness level of the patient on the day of testing.
Pittsburgh Sleep Quality Index (PSQI) (Buysse et al, 1989)
The PSQI was used to assess sleep quality of all participants. The PSQI is a self- rated questionnaire which assesses sleep quality and disturbances over a one-month time 76
interval. Nineteen individual items generate seven "component" scores including subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. Component scores ranging from 0 to 3 are then summed to yield one global score (0-21), with higher scores indicating worse sleep quality. The global score is used to distinguish good and poor sleepers (>5 as poor sleepers) with a diagnostic sensitivity of 89.6% and specificity of
86.5%. Evidence of internal homogeneity (Cronbach's a = 0.83), consistency (test-retest reliability coefficient = 0.85) were reported by Buysse et al. (1989). The PSQI was also found to be valid in distinguishing different patient groups, and in comparison with sleep variables obtained by polysomnography.
Sleep Timing Questionnaire (STQ) (Monk et al., 2002)
The STQ is a 14-question single-administration instrument that examines sleep and wake timing and habits without using a diary. Test-retest reliability for the STQ was demonstrated for estimates of bedtime (r = .71) and waketime (r = .83), and convergent validity with wrist actigraphy (r = .59) was reported. It was also shown to be highly
correlated with formal two-week sleep diary (r average at about .8) (Monk et al.).
Mood Assessment
Depressive Symptoms
Since mood changes can accompany OS A and potentially affect cognitive
functions, depressive symptomatology was measured by the well-validated Beck
Depression Inventory (BDI) (Beck, 1987). A raw score of 14 to 19 indicates mild level
of depressive symptomatology. Moderate and severe levels of depressive
symptomatology are indicated by a raw score of 20 to 28, and 29 to 63 respectively. 77
Affective States
The Profile of Mood States (POMS) (McNair et al, 1971) consists of 65 adjective scales. Participants rate their feelings during the past week on a 5-point rating system from 0 (not at all) to 4 (extremely). The profile includes a total score and scores of six affective states: tension-anxiety, depression-dejection, anger-hostility, vigor-activity, fatigue-inertia, and confusion-bewilderment. These subscales separate out somatic symptoms (sleepiness, fatigue) from affective symptoms (anxiety, depression), providing useful evidence to validate the effectiveness of CPAP in improving sleep, without the confound of affect. The reliability and validity of the POMS has been demonstrated in numerous studies in a variety of normal as well as chronically-ill populations (McNair et al), including patients with OS A (Bardwell et al, 2006; Bardwell et al, 2003; Yu et al.,
1999).
Functional Outcomes and Quality of Life
Functional Outcomes
To elucidate the relationship of performance in the neurocognitive tasks and
everyday living, the Cognitive Failures Questionnaire (CFQ) (Broadbent et al, 1982;
Wagle et al, 1999) was used to measure self-reported failures in perception, memory,
and motor functions in daily life. Participants rated how often each of 25 minor mistakes
happened in the past six months on a 5-point scale, ranging from 0 (never) to 4 (very
often). The CFQ has been shown to correlate positively with other self-reported
measures of absent-mindedness, and memory or action slips. The authors also reported
that CFQ predicts psychiatric symptoms (Broadbent et al). No norms for CFQ are
available to our knowledge. 78
To evaluate the specific impact of excessive sleepiness or tiredness on multiple activities of everyday living, the Functional Outcomes of Sleep Questionnaire (FOSQ)
(Weaver et al., 1997) was used. The five factors measured by FOSQ include activity level, vigilance, intimacy, and sexual relationships, general productivity, and social outcome. A total score can be generated from these subscales. The FOSQ has been found to be capable of discriminating between normal participants and untreated sleep apnea patients (Weaver et al). It has very high content validity, test-retest reliability (r =
0.91) and internal consistency (Cronbach's alpha = 0.96). A total score of less than 18 is considered as clinically significant by the joint task force of the American College of
Chest Physicians, the American College of Occupational and Environmental Medicine, and the National Sleep Foundation (Hartenbaum et al, 2006). Participants were asked to fill out the questionnaires according to their current condition and then their condition before using CPAP. This is the "Then-Test" approach designed to eliminate treatment- induced response-shift effects, and is used to provide an unconfounded indication of treatment effects (Schwartz & Sprangers, 1999; Visser et al, 2005).
Quality of Life
The Quebec Sleep Questionnaire (QSQ) (Lacasse et al, 2004) was adopted to evaluate the quality of life of the patient group. It is a 32-item OSA-specific, self- administered quality of life questionnaire that has been standardized and validated. There are five domains: (1) daytime sleepiness; (2) diurnal symptoms; (3) nocturnal symptoms;
(4) emotions; and (5) social interactions. Each domain includes 4-7 items and each item is scored on a 7-point Likert scale. There are no published norms for the questionnaire. 79
Global quality of life was measured using a visual analogue scale (QOL).
Participants were asked to mark their perception of their quality of life on a vertical line of 100 mm ranging from 0 (worst imaginable quality of life) to 100 (best imaginable quality of life). They were asked to indicate their current quality of life, and then their quality of life before the use of CPAP. Controls did the same procedure except that they evaluated their quality of life three months ago, instead of before using CPAP. This measure is used to capture the subjective evaluation of quality of life and its change as defined and experienced by patients without the constraints of a structured questionnaire.
The use of a single item in measuring quality of life has been shown to be as valid, reliable, and responsive as measures containing multiple items (de Boer et al, 2004).
Polysomnography
All participants (OSA and control groups) underwent a one-night polysomnographic sleep study. Nocturnal PSG allowed for a precise measurement of
sleep disordered breathing with a number of physiological and sleep variables collected during the study from a multichannel recording over a minimum of six hours of overnight recording. The recording included four electroencephalograms (EEG), left and right
electroculogram (EOG), electromyogram (EMG) from submental and anterior tibialis muscles, electrocardiogram, respiratory effort measured by inductive plethysmography
(Respitrace) and airflow measured from the CPAP mask using pressure transducer
technology. Continuous oxygen saturation was recorded by pulse oximetry. A room
microphone was used to detect snoring and an infrared camera to monitor body position.
Rechtschaffen and Kales's (1968) standard criteria were used for scoring sleep studies.
Sleep measures included total sleep time (TST), sleep efficiency (defined as total sleep 80
time/total recording time x 100), sleep latency, number of REM periods, REM latency, time spent in respective sleep stages (Stages NREM 1-4 and Stage REM), Total Arousal
Index (total number of arousals per hour of sleep) and Arousal with Respiratory Index
(number of arousals associated with respiratory events per hour of sleep). An obstructive apnea was defined as airflow cessation of at least 10 seconds in the presence of continued respiratory effort (Kushida et al., 2005). An obstructive hypopnea was defined as a reduction in airflow of at least 50% of biocalibration, lasting a minimum of 10 seconds followed by an EEG arousal and/or fall in oxygen saturation of at least 4%. The AHI/RDI was calculated as: total number of apneas and hypopneas/total sleep time (hr). There were also separate indices for obstructive vs. central apneas and hypopneas, and a Mixed
Apnea Index for apneas that had both a central and an obstructive component. Blood oxygen saturation measured by pulse oximetry (SpCh) was quantified as overall mean, the minimum saturation level as well as % TST less than 90%, % TST less than 80%, and
% TST less than 70%.
Neuropsychological Variables
The neuropsychological testing took place in the afternoon on a day within eight weeks of the PSG. While the main purpose of this study was to investigate executive function, a baseline neuropsychological test battery was administered, including tests recommended by Decary et al. (2000) to examine factors that may influence experimental results between the groups (e.g., overall intellectual functioning) and to facilitate comparison of patient characteristics with other studies. The battery included standardized tests covering attentional and executive functioning, short-term and long- term memory, and a global evaluation of intellectual functioning (see Appendix E for a 81
description of the neuropsychological tests). All the tests were administered and scored according to standardized procedures. The testing session took about three hours in total.
Working Memory Components
Several tasks were chosen to measure the functioning of the components of working memory as outlined by Baddeley (2003) including the two storage systems, i.e., the phonological loop and visuospatial sketchpad, as well as the attentional controller, i.e., the central executive. Measurement of storage capacity is critical to the interpretation of the functioning of the central executive in OSA (Verstraeten & Cluydts, 2004;
Verstraeten et al., 2004). These tasks were based on previous methodology used by
Baddeley and neuroscientists in the delineation of the separability of the components and their neural localization.
Phonological Loop
Digit Span (Verbal Immediate Recall). Verbal storage in the phonological loop was measured by providing subjects with item lists of increasing length and asking for recall. Six lists at each list length were given. List length started at two digits, with increment by one until the subject failed to recall at least 2/6 lists at that list length. Lists were read aloud at a rate of 1 item/second. Recall was recorded verbatim and accuracy of the recall sequence was noted. Outcome measures are the total number of lists correctly recalled before the test was discontinued, and the Digit Span, which is the longest list length of the lists of which the participant recalled at least 5/6 correctly.
Verbal Memory Scanning Task (Item Maintenance and Recognition). The verbal
memory task was adapted from the memory scanning paradigm of Sternberg (1966,
1969). Each trial had four events (Figure 3): (1) a fixation cross presented for 500 ms; (2) 82
simultaneous presentation of a memory set of 2, 4, or 6 letters for 600 ms, 1200 ms, and
1800 ms respectively. The progressive presentation duration with 300 ms per item in the set was a methodology adapted from Ferraro & Balota (1999). The letters were presented in uppercase and were randomly chosen from a pool of 12 consonants (i.e., B, D, F, G, H,
J, K, M, N, Q, R, T); (3) a return to the fixation cross for a retention interval of 3000 ms; and (4) a probe consisting of a single lowercase target letter, which stayed on the screen until the participant responded. All letters appeared within 2.5 degrees from the centre of the screen. Our viewing distance was about 57 cm, but participants were allowed to adjust their seat to maximize viewing and comfort. Participants were instructed to indicate whether the target item was on the just-presented list, by pressing one of two buttons with one of two fingers of their dominant hand. The next trial would start as soon as the participant responded. Fifty percent of trials were positive. There were no more than three positive trials in succession. The number of trials varied as a function of set size in order to obtain stable estimates of response latency at the more error-prone larger set size. There were 24 trials of set size 2, 36 trials of set size 4, and 48 trials of set size 6.
Four, 6, and 8 practice trials of set size 2, 4, and 6 were given in a random order in the beginning. Performance was assessed by accuracy including hits (correct "yes") and false alarms (incorrect "yes"). Reaction times (RTs) were also recorded. The task took
about 12 minutes.
Visuo-Spatial Sketchpad
Visual Span (Spatial Immediate Recall). The visual-spatial memory span was measured by the Visual Span subtask of the Wechsler Memory Scale-Revised (WMS-R)
(Wechsler, 1987). Administration was very similar to that of the Digit Span task in terms 83
of number of trials and levels. The stimuli consisted of two cards, each with eight 1 cm circles arranged in a random order. The examiner tapped on the circles in specified sequences with increasing length starting with two. The participant's task was to imitate the sequences. Accuracy of the recall sequence was noted. Outcome measure was the total number of lists correctly recalled before the test was discontinued. Only the forward span was used as a measure of the capacity of the visuo-spatial sketchpad.
Spatial Memory Scanning Task (Item Maintenance and Recognition). The visual- spatial item recognition task (Figure 4) was adopted from the one developed by Gevins et al. (1996), Smith, Jonides, and Koeppe et al. (1995), and Smith, Jonides, and Koeppe
(1996). The trial sequence was (1) subjects fixated on a cross in the center of a screen for
500 ms; (2) the cross was followed by a set of 2, 4, or 6 target dots that were arrayed on
12 possible positions, each of which was randomly generated within a donut-shaped area within 1.5 to 4.5 degree of radius from the screen's centre. The memory set of 2, 4, and 6 dots remained in view for 600 ms, 1200 ms, and 1800 ms respectively; (3) next the fixation cross appeared alone for a retention interval of 3000 ms; (4) the retention interval was followed by a location probe, available for 1500 ms, that consisted of a single outline target circle that either encircled the location of one of the previous dots or did not. Our viewing distance was about 57 cm, but participants were allowed to adjust their seat to maximize viewing and comfort. Participants were instructed to indicate whether the target item was in the just-presented array, by pressing one of two buttons with one of two fingers of their dominant hand. The next trial started as soon as the participant responded. Fifty percent of trials were positive with no more than three positive trials in succession. As in the Verbal Memory Scanning Task, the number of trials varied as a 84
function of set size in order to obtain stable estimates of response latency. There were 24 trials of set size 2, 36 trials of set size 4, and 48 trials of set size 6. Four, 6, and 8 practice trials of set size 2,4, and 6 were given in a random order in the beginning. Performance was assessed by accuracy and RTs. The task took about 12 minutes.
Central Executive
In order to obtain a more stable and valid measure of the executive component of working memory, multiple measures were adopted in this study using different methodologies (Conway et al, 2005).
Verbal-Spatial Dual Task (Divided Attention). The capacity for successfully coordinating performance on two parallel tasks, which required the simultaneous operation of the phonological loop and the visuospatial sketchpad, was measured using a paradigm previously developed for investigating the central executive in patients with
Alzheimer's disease (Baddeley et al., 1986). A paper-and-pencil version of the dual task was proposed by Delia Sala et al. (1995) to overcome the technical difficulties of the computerized version of the task and to promote usage in clinical settings. They found that the paper and pencil version of the dual-task paradigm was suitable for use with cognitively impaired patients. The current study modified a revised version of the task developed by Delia Sala's laboratory (S. Delia Sala, personal communication, September,
01, 2005). In particular, three practice trials were introduced for the single tracking task based on the practice effects and variability of performance that was observed in pilot testing. Baddeley et al. also had a pretesting phase with their computerized tracking task to ensure stable performance. 85
Step 1: Determining a Participant's Digit Span. The first step of the dual task was to establish a participant's digit span. In order to ensure consistency of administration across participants, a tape was used to present a list of digits read aloud at a rate of 1 digit per second. The participants were asked to repeat the digits in the order of presentation.
There were six lists for each list length and the test terminated when the participant failed on more than one trial for a specific list length. A participant's digit span was the maximum length of digit lists of which the participant recalled at least five out of six without error. This is a more conservative measure than a regular digit span task administered in a clinical setting such as in the WAIS-R, but it was necessary to have a more stable measure of the capacity of a participant considering the nature of the later stages of the task. To shorten the administration time and reduce fatigue with the first trials presented, we started with the span of four and no participant had to reverse to shorter spans.
Step 2: The Single-task (Verbal Immediate Memory) Condition. After the participant's span was determined, lists of digits at their personal span length were presented, with recall of the items in order of presentation. The number of lists presented in a 1.5 minute period to a participant naturally varied, depending partly upon the participant's span. The performance measure, therefore, was the proportion of items correctly recalled.
Step 3: The Single-task (Visual-spatial) Condition. Participants were asked to use a pencil to trace a chain of circles (of size 1 cm square) which were linked to form a random path laid out on an 11-inch x 17-inch-size white paper. They were first given an example of a short, 20-circle path, to accustom them to the procedure. Our pilot testing 86
showed that participants improved over trials. In order to control for practice effect and to ensure stable performance, participants were given three practice trials on this task before they were tested on the verbal memory single task. In the testing phase proper participants were required, over a 1.5-minute interval, to follow a path on sheets that contained a total of 319 circles (see Figure 5). The participants were asked to start at one end of the chain and draw a line following the path as quickly as possible, thereby crossing out the circles arranged along the path of the maze. If all the circles on one sheet were traversed before the time limit had elapsed, the participant continued on a second sheet of circles. The score was the number of circles successfully marked by the participant in the 1.5 minute. The task was administered twice to achieve a more reliable estimate of the participants' performance. The tracking score of the single task was calculated as the average of the two trials.
Step 4: The Dual-task Condition. In the final phase, participants were presented, over a period of 1.5 minutes, with different lists of digits at span length, while at the same time performing the tracking task. As with the single-task condition, the verbal memory measure was the proportion of the presented lists that were correctly recalled, and the visuo-spatial measure was the number of circles successfully marked by the participants.
The proportional loss of memory performance under dual-task condition as compared with the single-task conditions was calculated for the two tasks separately, and then a measure (D), which combined the two equally-weighted measures and expressed an individual's dual-task performance as a percentage of single-task performance was calculated according to the methodology devised by Baddeley, Delia Sala, Gray, Papagno, and Spinnler (1997). The measure D was calculated by this formula: 87
D = [l-(pm+pt)/2] x 100 pm is the proportion of lists of digits correctly recalled in the dual condition subtracted from that in the single condition. ptis the difference between the number of circles traced in the single and the dual condition divided by the tracking score in the single condition.
N-back Task (Maintenance and On-line Processing). This continuous memory task was adapted by Smith, Jonides, and Koeppe (1996), from the one developed by
Gevins and Cutillo (1993) (see Figure 6).
Two-back Task (Maintenance and On-line Processing). In both verbal and spatial conditions, subjects were presented a continuous stream of single letters, each for 500 ms, with a 2500 ms interval between successive letters; each letter appeared at a randomly chosen location amongst 12 possible positions, each of which was randomly generated within a donut-shaped area within 1.5 to 4.5 degree of radius from the screen's centre, which matched with the memory scanning task. Thus, both the position and identity of the letters varied on each trial and the two stimulus streams used for the verbal memory condition and the spatial memory condition were identical, with the only difference between the two conditions being the response requirement (letter identity versus spatial location). In other words, the general presentation of the stimuli was the same for the two memory conditions, and they differed only with respect to whether the subject's task was to store and monitor verbal (letter identity) or spatial (letter position) information.
In the verbal 2-back condition, subjects had to decide whether or not each letter matched in identity, regardless of position, the one presented two back (i.e. not the previous letter, but the one before that). Subjects pressed one of two buttons with one of 88
two fingers of one hand to indicate positive or negative responses. In the spatial memory condition, subjects had to decide whether or not the position of each letter matched the position of the letters presented two back, regardless of letter identity.
In both conditions, slightly more than one-third of the letters (10 trials) provided matches that required a positive response. There were four matches one-back so that subjects could not use mere familiarity as the basis of their responses. The remaining 12 trials, with two in the beginning of each block, did not match with any of the two preceding letters. No more than three positive trials occurred in a row.
The viewing distance was about 57 cm, but again subjects were allowed to adjust their seat for optimal viewing and responding. Four blocks of 26 trials (including two void trials in the beginning of each block) each were presented for the verbal and then the spatial memory condition. The blocks were initiated by the experimenter when the participant was ready. One practice block for each condition was administered in the beginning of the respective condition. The entire task including the 2-back verbal and spatial conditions lasted about 15 minutes in total. Performance was assessed by accuracies and RTs.
0-back Control Condition. In order to control for alertness, perceptual and motor response processes as factors interfering with performance, the 0-back control condition was introduced. It was essentially an item recognition task, in which the participants were presented with a similar sequence of letters as in the 2-back condition, but were required to decide simply whether or not each letter matched a single target letter or spatial location specified at the beginning of each block. In both verbal and spatial conditions, subjects were presented a continuous stream of single letters, each for 500 ms, 89
with a 2500 ms interval between successive letters; each letter appeared at a randomly chosen location in the pool of the 12 possible locations as in the 2-back task. There were four blocks of 26 trials for each condition, with one practice block at the beginning of each condition. The first two void trials of each block were included for the purpose of balancing with the structure of the 2-back condition. A new letter or location was the target for each block, and they were not one of the 12 possible letters or locations used in the 2-back Task to avoid evaluated exposure of and attention drawn to a particular letter or location. The targets of the verbal condition were "P, S, W, Y, and Z". There were 10 matches in each block. The entire task, including the 0-back verbal and spatial conditions, took about 15 minutes in total. Both accuracies and RTs were assessed.
Working Memory Span (Episodic Buffer)
This task measures the capacity to hold and manipulate information in long-term memory, a process calling upon the episodic buffer of the central executive. This task was based on procedures developed by Daneman and Carpenter (1980), which requires the subject to both process and store information simultaneously. The procedures included presenting the subject with a series of sentences, each of which were verified
(probable or not) as well as requiring the recall of the last word of each sentence in order at the end of the series. In a comprehensive review of working memory span tasks,
Conway et al. (2005) reported that working memory span tasks generally have high reliability with coefficient alphas of internal consistency ranging from .70 to .90 and test- retest correlations of .70 to .80 across studies. Working memory span tasks also predict
other complex cognitive performance, such as reading comprehension, problem solving, 90
and reasoning, as well as emotional control and social cognition (Baddeley et al., 1985;
Daneman & Merikle, 1996; Feldman Barrett et al, 2004).
Since this can also be considered a dual-task procedure, combining word span and sentence verification, performance on each sub-task was assessed so that basic capacity on each task for each subject could be taken into account when the tasks were combined
(Duff & Logie, 2001). To construct the three tasks that measured Word Span, Sentence
Verification Span, and Working Memory Span, three lists of one-syllable nouns were
generated from the Medical Research Council (MRC) Psycholinguistic Database:
Machine Usable Dictionary, Version 2 (Coltheart, 1981a, 1981b; Wilson, 1988) (see
Appendix F). The three lists were balanced for their concreteness and familiarity ratings,
and were used as words to be recalled in the Word Span Task, the Sentence Verification
Task, and the Working Memory Span Task respectively.
Word Span. Performance was assessed as in standard digit span procedure by providing subjects with item lists of increasing length and asking for recall. Three lists at
each list length were given. Each word was presented for 2 s, followed by an inter-
stimulus-interval of 0.5 s. At the end of each list, participants were cued to recall by a
tone, accompanied by a blank screen. Each list was initiated by the experimenter. List
length started at two words, with increment by one until the participant failed on more
than one trial. Recall was recorded verbatim and accuracy of the recall sequence was
noted. Outcome measures were the total number of lists correctly recalled before the test
was discontinued, the proportion of trials correctly recalled over total numbers of trials
administered, and the Word Span, which was the longest list length of the lists of which
the participant recalled at least two out of three trials correctly. 91
Sentence Verification Span. This task was adapted from Daneman and Carpenter
(1980) and Duff and Logie (2001). Plausible and implausible sentences of three to six words were constructed, with words generated from the MRC Psycholinguistic Database serving as the last word of the sentences. The sentences were allocated randomly to lists varying in length from two to ten sentences, with each list having a mixture of plausible and implausible sentences. The sentences were presented one at a time on a computer screen and the participant's task was to respond whether it was plausible or not. Each list of sentences was presented at a predetermined rate over 10 s, so that at longer list lengths,
sentences were presented more rapidly (e.g., for two-sentence list length, sentences were presented for 5 s each; for four-sentence list length, they are presented for 2.5 s each).
List length increased by one until the participant could no longer correctly verify all of the sentences on two out of three trials for a given list length. Sentence verification span was then defined as the previous list length.
Working Memory Span. For the dual task procedure, the sentence verification procedures were conducted again in the same manner with different sentences and
participants were asked to both verify sentences and to memorize the final word of each
sentence for serial recall at the end of the sentence list. Testing continued until
participant failed to perform to the same criteria for both the memory and the verification
tasks. Depending on the participants' performance, the entire test took about 10 to 20
minutes. Working Memory Span was the mean number of words correctly recalled in the
last three correct trials. The Word Span and the Sentence Verification Span as well as the
total number as well as proportions of sentences correctly verified and words correctly 92
recalled in the dual task condition were also compared with the measures in the single task condition.
Statistical Analyses
To evaluate the effectiveness of CPAP treatment in improving breathing during sleep, oxygen saturation, subjective sleep quality, sleepiness, functional outcomes, and quality of life, paired sample t-tests were conducted comparing patients' pre- and post- treatment scores on RDI, Sp02, PSQI, ESS, and QOL. To understand the clinical significance of the changes, the percentages of patients and healthy controls whose scores were in the clinical range also were compared using chi-square analyses.
To examine the second hypothesis that the performance of the OSA group would be comparable to normal age-matched controls on tests of basic storage and rehearsal components in working memory, between-group independent t-tests were used to analyze the scores of the OSA group and healthy controls on the Digit Span test in the Dual Task procedure and Visual Span test of the WMS-R. Three-way Analysis of Variance
(ANOVA) was used to examine group differences in RTs and accuracies on the Verbal and the Spatial Memory Scanning Tasks.
To investigate the hypothesized worse performance in the OSA group on working memory tasks requiring the central executive, ANOVAs were used to analyze the Verbal and Spatial N-back Tasks, the Dual Task, and the Working Memory Span Task. For the
Verbal and the Spatial N-back Tasks, both RTs and accuracies were analyzed. For the
Dual Task, the OSA group and healthy controls were compared on the two task conditions (single vs. dual) on the individual task scores of List Memory and Tracking. 93
The two groups were also compared with an independent t-test on the Combined Dual
Task Score (D). Similar analyses were used for the Working Memory Span. ANOVAs were used to analyze the individual task scores (raw numbers and proportions) on the
Word Span and Sentence Verification Tasks in the single vs. dual condition. The composite score, Working Memory Span was analyzed using an independent t-test.
For the neuropsychological tests, factorial repeated measure ANOVAs were used for tests with more than one level of the outcome measure (e.g., Consonant Trigrams,
Trail Making, Digit Span, Visual Span). Independent t-tests were used to analyze group differences in all the other neuropsychological tests. Due to the large number of comparisons, a significance level of .01 was adopted for these ANOVAs and t-tests.
To examine the predictors of neurocognitive difficulties, neurocognitive tasks that demonstrated differences between patients and healthy controls were regressed on demographic variables (age, BMI), illness variables (diagnostic RDI, diagnostic minimum SpOa), post-treatment sleep efficiency, sleepiness (ESS), and subjective sleep quality (PSQI) using stepwise regression procedures with a significance level of .05.
BMI was included to rule out obesity itself as a predictor of performance, especially given that the OSA group had a higher BMI than the control group. Current respiratory and hypoxemia indices were not included as predictors owing to the small range of those variables in the post-treatment OSA group.
To explore the predictors of patients' psychosocial outcomes, mood measures
(BDI, POMS), subjective daytime functioning (FOSQ, CFQ), and quality of life measures (QOL, QSQ) were regressed on predictors, including age, BMI, diagnostic RDI, diagnostic minimum Sp02, post-treatment sleep efficiency, current sleepiness (ESS), and 94
subjective sleep quality (PSQI) using stepwise regression procedures with a significance level of .05.
To examine the relationship between performance on neurocognitive tasks that require executive function, and perceived cognitive efficiency, and quality of life, OSA group's scores on neurocognitive tasks that indicated worse performance compared to healthy controls were correlated with their scores on CFQ, QOL, and QSQ. Lastly, quality of life measures, namely QOL and QSQ were correlated with measures of mood
(i.e., BDI, POMS), and functional outcomes (i.e., FOSQ, CFQ) to explore the relationships among the psychosocial variables. Due to the large number of correlational analyses, a significance level of .01 was used for all correlational analyses to control for
Type I error. 95
CHAPTER THREE: RESULTS
Participants' Characteristics
Data from 37 individual with OSA and 27 healthy controls were included in the analyses. The numbers of patients and controls participating at each stage of the protocol are shown in a flowchart in Figure 8. Participants' age, sex, handedness, education level, and BMI are reported in Table 1. There were no significant differences between the patients with OSA and the controls on age and education level but, as expected, patient's mean BMI was significantly higher than that of the controls, t (62) = 4.83, p < .001.
Information about patients' OSA disorder is presented in Table 2. For 73% of the participants, compliance data were available from their CPAP machine. Compliance was based on self report for the remaining participants. Considering previous treatments for
OSA before using CPAP in the OSA group, thirteen (35%) had tonsillectomy, three had a nose surgery (8%), and one individual had a jaw surgery, a UPPP, and an uvulectomy, in addition to a tonsillectomy.
For 18 participants (4 controls, and 14 patients) testing was conducted on the same day as PSG. The average interval between PSG and testing was 8.7 days overall with a range of testing occurring from 1 to 50 days (average = 12.8 days) before PSG and
2 to 55 days (average = 11.96 days) after PSG.
Pre- and Post-Treatment Comparisons
In order to assess the effectiveness of CPAP in improving the OSA group's sleep apnea indicators, daytime sleepiness, and functional outcomes (Hypothesis 1), paired- sample t-tests were conducted on OSA group's data pre- and post-treatment, using 96
significance level of .01. Individuals with OSA treated with CPAP showed significant improvements on respiratory and oxygen saturation indices (RDI, minimum SpC>2, mean
SpC>2), sleepiness (ESS), sleep quality (PSQI), functional outcomes (FOSQ), and quality of life (QOL) (Table 3). To provide a comparison, the means of these variables for controls are also reported in Table 3 but the between-group comparisons will be discussed later.
For measures that have established clinical cut-offs, the percentage of individuals of the OSA group and that of the control group passing the thresholds for the clinically problematic range were shown in Figures 9a-d. The OSA group after CPAP treatment did not differ from the control group on minimum Sp02, mean Sp02, ESS, PSQI Global
Score, and FOSQ Total Score. The only difference in percentages in clinical range between the two groups was for RDI, with a higher percentage of controls (33%) having a RDI = or > 5 than the OSA group (18%), %\\, N = 6A) = 8.55, p < .01. It should be reiterated that the mean RDI of the control participants was under 5 and participants with a RDI > 15 were excluded from the study.
For 15 individuals in the OSA group who had a diagnostic PSG overnight study, their sleep efficiency (time in sleep/total time in bed), sleep latency (sleep onset measured in minutes), and "Arousal with Respiratory Index" were compared pre- and post-CPAP treatment. While their sleep latency did not change significantly after treatment (Mpre =
17.0, SD = 8.4; Mpo.st = 16.8, SD = 12.1), their sleep efficiency improved significantly
(Mpre = 62.8, SD = 13.0; Mpost = 79.2, SD = 10.5), t (14) = 4.15, p < .01. Their Arousal with Respiratory Index also improved significantly after treatment (Mpre = 15.4, SD =
15.5; MPoSt = 1-72, SD = 1.2), t (14) = 3.46, p < .01. 97
Between-Group Comparisons
Polysomnography
To further understand if sleep architecture of the OSA group was normalized after
CPAP treatment, the two groups were compared on all of the polysomnographic measures available. Independent samples t-tests were used to analyze total sleep time
(TST), sleep efficiency, sleep latency, REM latency and time spent in different sleep stages (Stages NREM 1-4 and Stage REM), obstructive apnea, obstructive hypopnea,
RDI, blood oxygen saturation (SpCh), and percentage of total sleep time (% TST) less than 90% saturation, % TST less than 80%, and less than 70%. In order to control for
Type II error, a significance level of .05 was adopted for these analyses. Significant differences were found on three variables, namely the Obstructive Hypopnea Index, the
RDI, and the Arousal with Respiratory Index. The OSA group using CPAP was in the more favorable direction as compared to the control group on all these three variables
(Table 4).
Working Memory Tasks
Overview
To compare the performance of the patients and the controls, analyses of variance
(ANOVAs) were used in analyzing the experimental tasks, and the significance level was set at .05 for a priori hypothesis testing. Reaction times for correct trials, accuracies (% of correct trials out of the total number of trials which included subthresholds and timeouts), errors (% of subthreshold responses (RT<200 ms) and % of time-out (>3000 ms) for the verbal tasks (i.e., Verbal Memory Scanning task, Verbal 0-back task, and 98
Verbal 2-back task) are presented in Table 5. Results from the parallel visual-spatial tests are summarized in Table 6. Table 7 shows results of the Dual task; performance, measured by D is expressed as a percentage of single-task performance, with performance on the tracking and digit span contributing equally to the score. Outcome measures of the
Working Memory Span task include working memory span composite measure, the total number of words recalled and sentences verified, as well as the proportion of words correctly recalled and the proportion of sentences correctly verified (Table 8).
Phonological Loop
Digit Span (Verbal Immediate Recall). Verbal storage capacity of the phonological loop as measured by digit span in the Dual Task procedure did not differ between the OS A group (M0SA=5.9, SD=1.3) and healthy controls (MHc=6.0, SD=1.2).
Verbal Memory Scanning Task. Reaction times and accuracies were analyzed using three-way ANOVA (3 Set Sizes (2, 4, 6) * 2 Target Types (present, absent) * 2
Groups (OSA, healthy controls) (Table 5). For RTs, there was a main effect of Set Size
(F (2, 124) = 157.73, p < .001), with RTs increasing with Set Size, and a main effect of
Target Type (F (1, 62) = 24.79, p <. 001), with target absent trials requiring longer RTs than target present trials. These main effects were qualified with a significant interaction between these two factors (F (2, 124) = 20.30, p < .001), with the difference between target present and target absent trials increasing with set size. There were no Group main effects or other interaction effects for RT.
Similar to the RT data, for accuracy there was a significant main effect of Set Size
(F (2, 124) = 101.14, p < 001), with larger set size associated with lower accuracies, and a main effect of Target Type (F (1, 62) = 22.52, p <= .001), with absent trials being less 99
accurate than present trials. There were no main effects of Group or any interaction effects for accuracy.
Visuo-Spatial Sketchpad
Visual Span (Spatial Immediate Recall). No differences were found between the
OSA group (M = 8.1, SD = 1.8) and controls (M = 9.3, SD = 1.9) on the Spatial Forward
Span raw score of WMS-R. (Table 9).
Spatial Memory Scanning Task. Reaction time and accuracy data were analyzed in the same way as the Verbal Memory Scanning task (Table 6). Three-way ANOVAs (3
Set Sizes * 2 Target Types * 2 Groups) of RTs showed a significant main effect of Set
Size (F (2, 124), p < .001), with larger set sizes taking longer RTs, and a main effect of
Target Type (F (1, 62) = 7.82, p < .01), with absent trials requiring longer RTs than present trials. No Group effects or interaction effects were found for RT.
Accuracy analyses showed a significant main effect of Set Size (F (2, 124) =
105.54, p <.001), with higher accuracies for smaller set sizes, and a significant main effect of Target (F (1, 62) = 4.03, p <.05), with higher accuracies for target present trials.
There was also a significant interaction between Set Size and Target (F (2, 124) = 35.9, p
< .001). While the accuracies of target present trials were stable across set sizes, accuracies of target absent trials decreased significantly as set size increased. There were no Group or other interaction effects for accuracies.
Central Executive
N-back Task
Verbal N-back Task. Reaction time data were analyzed using three-way ANOVAs with 2 Task Conditions (0-back vs. 2-back) * 2 Target Types (present vs. absent) * 2 100
Groups (OSA vs. controls) (Table 5). There were significant main effects of Task
Condition, with the 2-back condition showing longer RTs than the 0-back condition, F (1,
62) = 255.61, p < .001, and of Target Type, with absent trials showing longer RTs than present trials, F (1, 62) = 80.06, p < .001. There was a significant interaction between
Task Condition and Target Type, suggesting that absent trials were particularly slow in the 2-back condition), F (1, 62) = 83.79, p < .001. A significant 3-way interaction of
Task Condition * Target Type * Group was also found. Follow-up analyses showed that the control group (M = 630.6, SD = 98.8) had faster RTs than the OSA group (718.6, SD
= 123.9) only for absent trials in the 0-back condition (t (62) = 3.05, p < .01), and there were no significant differences between the groups in present trials for both conditions or in absent trials for the 2-back condition.
The same analyses were conducted with the accuracy data. The only significant main effect was of Task Condition, with lower accuracies on the 2-back compared to the
0-back conditions, F (1, 62) = 96.65, p < .001. The Task Condition * Group interaction approached significance (F (1, 62) = 3.30, p = .074), suggesting a trend for a bigger difference between the 0-back and 2-back conditions for patients than for controls.
Follow-up analyses showed that while the two groups did not show significant differences on Verbal 0-back accuracy (MOSA = 97.9, SD = 2.5, MHc = 98.4, SD = 2.1), the OSA group had a significantly lower accuracy in the 2-back condition than the controls (MOSA = 75.8, SD = 17.6, MHC = 83.6, SD = 10.9), t (62) = 2.17, p < .05)
Spatial N-back Task. Three-way ANOVAs with 2 Task Conditions (0-back vs. 2- back) * 2 Target Types (present vs. absent) * 2 Groups (OSA vs. controls) were used to analyze reaction time data (Table 6). There were significant main effects of Task 101
Condition, with slower RTs for the 2-back condition than the 0-back condition, F (1, 61)
= 240.7, p < .001. There was also a significant main effect of Target Type, with the absent trials showing longer RTs than the present trials, F (1, 61) = 25.05, p < .001.
There was a significant interaction between Task Condition and Target Type, indicating slower RTs in absent trials especially in the 2-back condition, F (1, 61) = 59.00, p < .001.
There was no Group main effect or other interaction effects.
Three-way ANOVAs with the same independent variables were conducted using the accuracy data. As with the Verbal N-back task, Task Condition, but not Group, showed a significant main effect, with lower accuracies in the 2-back than the 0-back condition, F (1, 61) = 34.13,/? < .001. The Task Condition * Group interaction was also significant, F (1, 61) = 1542.1,/? = <.01, indicating that while patients and controls did not differ on the 0-back condition (t (62) = 0.34, p = .74), patients were less accurate than controls in the 2-back condition (t (62) = 2.81, p < .01). This relationship is illustrated in
Figure 10.
Verbal-Spatial Dual Task. The two groups were compared on the Combined Dual Task
Score (D), and separately on the Digit List Memory performance and Tracking performance using two-way ANOVAs with 2 Task Conditions (single vs. dual) * 2
Groups (OSA vs. controls) (Table 7).
There was no significant group difference on the Combined Dual Task Score (D)
(MOSA = 90.8, SD = 9.5; MHC = 91.3, SD = 6.6).
On Digit List Memory, which is the proportion of lists of digits correctly recalled using the participant's predetermined span, there was a significant main effect of Task 102
Condition (F (1, 62) = 34.61, p < .001), with better performance in the single than the dual condition. There were no main effects of Group or interaction effects.
A similar pattern of results was found for Spatial Tracking. There was a significant main effect of Task Condition (F (1, 62) = 38.00, p < .001), with better performance in the single condition than in the dual condition, and no Group main effects or interactions.
Working Memory Span
The Working Memory Span was analyzed in a similar way as the Dual Task
(Table 8). Working Memory Span, is a composite score of a participant's performance when they are required to remember the last words of a series of sentences while verifying the sentences. It is derived from the mean number of words correctly recalled in the last three correct trials before the participant reached the discontinuing rule for both verbal and spatial components of the task. There was no significant difference between patients (M = 2.9, SD = 1.0) and controls (M = 3.3, SD = 1.0) on this measure. The total number of words recalled (Word Span) and sentences verified (Sentence Verification
Span), as well as the proportion of words correctly recalled and the proportion of sentences correctly verified were analyzed using 2 Task Condition (single vs. dual) * 2
Group (OSA vs. controls) ANOVAs.
For the Word Span, there was a main effect of Task Condition, F (1, 62) = 203.40, p < .001), with more words remembered in the single task condition than in the dual task condition. There was no main effect of Group or interaction effects. No main effects or interactions were found for the Sentence Verification Span. 103
Considering the proportions of words correctly recalled and sentences correctly verified, there was a significant main effect of Task Condition for Word Proportions (F
(1, 62) = 257.33,/? < .001), with a lower proportion of words being recalled in the dual condition than in the single condition. There were no main effects of Group or interactions. For the Sentence Verification Proportion, there were no significant main effects but a significant interaction between Task Condition and Group (F (1, 62) = 5.54, p < .05). Follow-up analyses demonstrated that while controls and patients verified a similar proportion of sentences in the single condition (t (62) = 0.38, p = .200), patients verified a significantly smaller proportion of sentences in the dual condition than controls,
(t (62) = 3.95,p < .001), (MOSA= .68, SD = .10; MHC= -78, SD = .10) (Figure 11).
Neuropsychological Tests
Patients and controls were compared on their performance on the neuropsychological battery. Means and standard deviations of raw scores are presented in Table 9, and the number of participants in each group showing impairments as defined by the standardized norms of individual tests is shown in Table 10. With the exception of the PVT, impairment was defined by a z-score less than -1.5, a T-score less than 35, a scaled score less than 6, a standard score less than 78, or less than the 7th percentile. For the PVT which does not have normative data, cut-offs were derived from a previous empirical study of sleep deprivation. Participants were considered impaired if they performed at a level worse than the mean of normal adults after seven days of sleep restriction to less than five hours nightly (Dinges et al., 1997).
Four measures that have several levels (i.e., Consonant Trigrams, Trail Making,
Digit Span (WAIS-R) and Visual Span (WMS-R)) were analyzed using factorial 104
ANOVAs, with significance level set at .01. The Consonant Trigrams was analyzed by 4
Delay Interval (0" vs. 9" vs. 18" vs. 36") * 2 Group (OSA vs. controls) ANOVA, with dependent variable being the number of letters recalled after a delay interval of counting backward by three's. There was a significant main effect of Delay Interval, F (3, 186) =
87.34, p < .001, with poorer recall following longer delays (M0-= 15.0, SD = 0.2; M9- =
11.3, SD = .2.6; Mi8»= 10.7, SD = 2.7; M36-= 10.0, SD = 2.6). There was no main effect of Group or interaction effect for Consonant Trigrams.
On the Trail Making test, ANOVA analysis with 2 Task Condition (Trail A vs.
Trail B) * 2 Group (patients vs. controls) was conducted, with the completion time in seconds as the outcome measure. There was a main effect of Task Condition, F(l, 62) =
224.41, p < .001, with shorter completion time for Trail Making A (M = 25.5, SD = 8.2) than Trail Making B (M = 69.0, SD = 24.9). There was also a trend for a main effect of
Group, F (1, 62) = 4.87, p = .031, with the OSA group taking longer to complete the tasks. Follow-up analyses showed that there was a trend for a significantly shorter completion time for controls on Trail Making B (t (62) = 2.04, p = .045), and that there was no difference between the two groups on Trail Making A (t (62) = 1.60, p = .116).
However, the interaction between the two variables was not significant, F (1, 62) = 2.65, p = A09.
Two way ANOVAs with 2 Task Condition (forward vs. backward) * 2 Group
(OSA vs. controls) were conducted for both Digit Span on WAIS-R and Visual Span on
WMS-R. For Digit Span, there was a trend for a significant main effect of Task
Condition, with longer forward span than backward span for both groups, F (1, 62) =
4.40, p = .040. There was no Group main effect nor interaction effect for Digit Span. On 105
the Visual Span, there was a significant main effect of Task Condtion, F (1, 62) = 9.17, p
< .01, with a shorter span for backward recall (M = 8,4, SD = 1.7) than forward recall (M
= 9.1, SD = 1.8). There was no Group main effect nor interaction effect for Visual Span.
Patients and controls were compared on all the other neuropsychological tests using independent samples t-tests (Table 9). In order to control type I errors due to the large number of tests in the battery, a significance level of .01 was adopted. Patients performed significantly worse than controls on measures of attention and concentration
(D2 Test of Attention - total minus errors and concentration performance; Digit Symbol), and executive functions (Stroop - interference score; Wisconsin Cord Sorting Test - perseverative errors, non-perseverative errors, and proportion of correct responses/total number of cards administered), and psychomotor dexterity and speed (Grooved
Pegboard - dominant hand).
Sleep Questionnaires
Findings on PSQI (post-treatment), STQ, ESS (post-treatment), and SSS, are summarized in Table 11. The OS A group and the controls did not differ on the questionnaires, at a significance level of .01. There was a trend for a difference between the two groups in the component score of daytime dysfunction on the PSQI, t (62) = 2.14, p = .036, with the OSA group reporting more sleep-related daytime dysfunction than the controls.
Mood Assessment
BDI and POMS results are reported in Table 12. No significant differences between the two groups were detected on the two mood measures at a significance level of .01. 106
Daily Functioning and Quality of Life
The OSA group and the control group were compared on their daily functioning on the CFQ and FOSQ (Table 13). The only difference found was on the subscale of
"Activity Level" on the FOSQ, with the OSA group having a lower activity level than controls, / (62) = 3.00,p < .01.
Data on QSQ and QOL are presented in Table 14. On the QOL, the OSA group reported significantly lower quality of life pre-treatment than controls (three months ago), t (57) = 5.74, p < .001). The OSA group also reported a greater difference in quality of life in the positive direction post-CPAP, compared to controls who reported minimal difference in their quality of life in the past three months, t (57) = 6.45, p < .001. There was no difference between the two groups in their current quality of life as measured by
QOL. There were no data on QSQ for controls owing to the OSA-specific nature of the instrument.
Regression Analyses
To elucidate the variables predicting the neurocognitive and psychosocial outcomes of individuals with OSA treated with CPAP, stepwise regression analyses were conducted on neurocognitive measures that differentiate the OSA group from healthy controls, and on all measures of mood, functional outcomes and quality of life (Table
15a). A significance level of .05 was used due to the exploratory nature of these analyses.
Working Memory Tasks
Verbal (Table 15b) and Spatial (Table 15c) 2-back RT and accuracy data were regressed on variables including age, BMI, education, diagnostic RDI, and diagnostic 107
minimum SpO^, post-treatment ESS, post-treatment PSQI - Global Score, and post- treatment sleep efficiency (%). Verbal 2-back RT was predicted by diagnostic minimum
SpC>2, with lower oxygen saturation associated with slower RT. Verbal 2-back accuracy was predicted by age and diagnostic minimum SpC>2, with younger individuals with higher oxygen saturation having higher accuracy. Spatial 2-back RT was predicted by diagnostic RDI, with higher RDI associated with longer RT. For Spatial 2-back accuracy, significant predictors included age, post-treatment ESS, and years of education. Higher
Spatial 2-back accuracy was associated with lower age, more education, and higher level of sleepiness. The same set of predictors was used in the regression model for Sentence
Verification (proportion) in the dual condition of the Working Memory Span (Table 15d).
Years of education and post-treatment sleep efficiency were significant predictors, with worse performance associated with fewer years of education and higher sleep efficiency.
Neuropsychological Tests
The same set of predictors was put into the regression analyses for neuropsychological tests that show a difference between patients and controls (Table 15a).
A significant regression model was found for Digit Symbol, with higher education associated with better performance (Table 15e). Perseverative errors in Wisconsin Card
Sorting Test was predicted by education, with higher age associated with fewer errors
(Table 15f). Faster Grooved Pegboard performance on the dominant hand was associated with younger age, higher diagnostic minimum Sp02, higher education, and higher PSQI -
Global Score (indicating poorer sleep quality) (Table 15g). Follow-up analyses were conducted to explore the relationship between PSQI subscales and performance on
Grooved Pegboard. It was found that Daytime Dysfunction was negatively correlated 108
with Grooved Pegboard performance, r (35) = -.43, p = .008, indicating that the less daytime dysfunction the individuals with OSA reported, the slower they were on the
Grooved Pegboard task.
Mood, Daily Functioning, and Quality of Life
The BDI, POMS, FOSQ, CFQ, QOL, and QSQ scores were analyzed in regression analyses with predictors including age, BMI, diagnostic RDI, diagnostic minimum Sa02, post-treatment sleep efficiency, post-treatment ESS, post-treatment
PSQI-Global Score (Table 16a).
BDI was predicted by post-treatment ESS score, with sleepier patients having higher level of depressive symptoms (Table 16b). Post-treatment ESS also predicted
POMS-Total Score, with worse affective states associated with sleepiness (Table 16c).
Individual subscales of POMS were regressed on the same set of variables. Post- treatment ESS predicted all subscales except for vigor-activity, which was predicted by
PSQI, with poorer sleep quality associated with lower vigor-activity. In addition to ESS,
BMI also predicted fatigue-inertia, with higher BMI associated with worse fatigue-inertia.
Post-treatment FOSQ-Total Score was predicted by post-treatment ESS score and post-treatment PSQI-Global Score. Better functional outcomes were associated with less sleepiness and better sleep quality (Table 16d). FOSQ - Activity level was associated with lower scores on the ESS, lower scores on the PSQI, and older age. FOSQ -
Vigilance was associated with lower ESS scores. FOSQ - Intimate relationships and sexual activity, General productivity, and Social outcomes were all associated with lower
PSQI scores. 109
CFQ score was predicted by diagnostic RDI. Interestingly, higher RDIs were associated with lower CFQ scores (i.e., fewer cognitive complaints).
The QOL was not predicted by any of the variables except for age, with older individuals reporting higher quality of life after CPAP treatment. QSQ - Daytime
Sleepiness was predicted by post-treatment ESS, which also predicted other QSQ scores, including Diurnal Symptoms, Nocturnal Symptoms, Emotions, and Social Interactions.
More sleepiness was associated with poorer quality of life on all of these QSQ subscales.
In addition to ESS, QSQ - Emotions subscale was also predicted by age, with older individuals having higher quality of life on emotions. QSQ - Social Interactions subscale was also predicted by age in the same direction, together with PSQI, with poorer sleep quality associated with lower quality of life in social aspects.
Correlational Analyses
To investigate our hypothesis that patients' executive function would correlate with their subjective cognitive complaints and quality of life, patients' scores on executive function measures that showed significantly worse performance than that of controls were correlated with their scores on the CFQ, QOL, and QSQ. Executive function measures included in the analyses were Verbal 2-back accuracy, Spatial 2-back accuracy, Sentence Verification Proportion in the Working Memory Span Task, Stroop
Interference score, and WCST - Perseverative and Nonperseverative errors. To control for type I error, a significance level of .01 was adopted.
None of the executive function measures correlated with CFQ, QOL or QSQ, suggesting that there was a discrepancy between objective cognitive functioning and 110
subjective cognitive failures, and that objective executive function was not associated with subjective quality of life. There were significant correlations between CFQ and all but one QSQ subscales, including Daytime Sleepiness (r (35) = -.45, p < .01), Diurnal
Symptoms (r (35) = -.49, p < .01), Emotions (r (35) = -.43, p < .01), and Social
Interactions (r (35) = -.49, p < .01), with more cognitive failures associated with lower quality of life.
As shown in the regression and correlational analyses, QSQ subscales were significantly correlated with post-treatment ESS scores and CFQ. To further elucidate what other factors were associated with quality of life, the QOL and the QSQ subscales were correlated with the post-treatment FOSQ - Total Score and each subscale, the BDI, and the POMS Total Score and each subscale. The QOL was not associated with the
FOSQ, the BDI, or the POMS. Results of correlations with the QSQ are presented in
Table 18. In general, the QSQ subscales were positively correlated with the FOSQ scores, indicating that lower quality of life was associated with poorer functional outcomes. The
QSQ was also negatively correlated with the BDI, as well as the POMS scores, except for the Activity-Vigor scale that is scored in the opposite direction (i.e. higher scores indicating better functioning). These findings showed that lower quality of life was associated with poorer functional outcomes, more depressive symptoms, and more negative affective states in individuals with OSA treated with CPAP. Ill
CHAPTER FOUR: DISCUSSION
This study aimed at investigating the neurocognitive function and psychosocial outcomes of individuals with OSA treated with CPAP for at least three months using the working memory model. The hypotheses and exploratory analyses were: 1) CPAP would be effective in improving respiratory (RDI) and hypoxemia (mean and minimum Sp02 indices), subjective sleep quality, daytime sleepiness, functional outcomes, and quality of life, as indicated by improvements in these outcome measures in post-CPAP evaluation as compared to pre-CPAP evaluation. 2) The performance of the stably treated OSA group would be comparable to normal age-matched controls on tests of basic storage and rehearsal components in working memory. 3) In contrast, the stably treated OSA group would show worse performance than normal controls on tests that demand the involvement of the central executive. 4) Patients would show comparable performance to controls on neuropsychological testing except for executive function. 5) Patients' executive function results would be predicted by their diagnostic respiratory and hypoxemia indices before CPAP treatment. 6) The results of tests of executive function of patients would correlate with their perceived cognitive efficiency and quality of life. 7)
The predictors of patients' current psychosocial functioning were investigated. 8)
Correlates of quality of life measures with other psychosocial variables were also explored.
Effectiveness of CPAP Treatment
I studied 37 CPAP-treated individuals with moderate to severe OSA with an overnight PSG while using CPAP. Since our main objective was to study the outcomes 112
of adequate CPAP treatment on cognitive function, I defined adequate treatment as use of
CPAP regularly for at least three months preceding the study, and an RDI < 15 on a current sleep study. CPAP compliance was verified in most cases (73%) by machine compliance data, and one individual was excluded from data analysis due to his low compliance as shown on the "smart card" data. Another participant who showed a RDI of 19.4 in the post-treatment PSG, was excluded as well. Thus, in comparison with their pre-treatment PSG or home study, all participants in the OSA group showed significant improvements in their respiration during sleep as measured by RDI, and their blood oxygen saturation as measured by their mean and minimum SpC<2. Pre- and post- treatment comparisons on subjective sleep quality (PSQI), daytime sleepiness (ESS), functional outcomes (FOSQ), and quality of life (QOL) also revealed significant improvements. Given that only one individual showed an RDI > 15, these findings supported our hypothesis that CPAP is effective in reducing apneas and hypopneas.
Further, the reduction in RDI was associated with improvements in a number of other night-time sleep variables, including increasing blood oxygen saturation during sleep.
These results are in line with previous studies investigating the improvement of sleep variables after CPAP treatment (Feuerstein et al, 1997; Grunstein, 2005; Patel et al,
2003). CPAP was also found to be effective in improving subjective sleep quality, daytime sleepiness, functional outcomes, and quality of life, consistent with findings of previous studies (Kingshott et al., 2000; Patel et al.; Sanner et al., 2000).
To further elucidate whether CPAP normalizes patients' breathing during sleep, sleep quality, and daytime outcomes, I also looked at established clinical cut-off scores for individual patients. The percentages of participants with clinically elevated scores in 113
the OSA group were compared to those in the control group, which was composed of 27 age- and education-matched controls.
Considering RDI, only one participant with OSA (out of 38) was found to have a
RDI greater than 15 with at least three months of treatment with CPAP suggesting that
CPAP was effective in eliminating breathing disruptions during sleep for most patients who were compliant with the treatment. While RDI > 15 was our criterion for inclusion in the analyses, most patients were better treated than that and showed a RDI < 5 (94.6%)
Surprisingly, even though we attempted to screen the controls carefully for the undetected presence of sleep apnea, 30% of the controls showed a RDI > 5. Given the high prevalence of mild breathing interruptions in our non-symptomatic control group, it seems questionable whether mild OSA (RDI between 5 and 15) is a meaningful category clinically and in research. Mild OSA is not treated in most patients due to a combination of reasons including the monetary costs, physical discomfort, and inconvenience of using
CPAP. In terms of research, a few studies of patients with mild OSA did not identify significant impairments in memory, executive function, or other neurocognitive functions
(Knight et al, 1987; Phillips et al, 1996). Phillips et al. (1996) also reported that elderly people with mild OSA do not show any psychopathology. None of our control participants reported consistent sleep-related difficulties or elevated symptoms of OSA in screening using the Sleep Disorders Questionnaire (SDQ), the Berlin Questionnaire, and a health interview, suggesting that the impact of mild OSA on people's lives may be relatively minimal, although longstanding mild sleep-disordered breathing may be associated with unspecified medical outcomes. Taken together, it seems reasonable to consider our OSA group to be adequately treated and normalized. It should be noted that 114
all patients showing a RDI greater than 5 were followed up clinically by physicians of the
QEII Sleep Disorder Laboratory to investigate if their CPAP treatment was optimized and if adjustments were necessary.
For self-reported sleepiness as measured by the Epworth Sleepiness Scale, the percentage of individuals in the OSA group with reports that suggested pathological sleepiness dropped from 76% to 30% after treatment, with a significant change in mean scores when analyzed using the reliable change index score (Smith & Sullivan, 2007). In addition, this percentage was not significantly different from the 15% of controls who reported symptoms in the pathological sleepiness range. It could be concluded that while sleepiness was significantly reduced in participants with OSA after CPAP treatment, some individuals continued to show excessive sleepiness during the day. While sleepiness was found to consistently predict functional outcomes in this study, as discussed later, there are no obvious explanations for the residual sleepiness found in treated individuals with OSA. With the lack of difference in self-reported sleepiness between the two groups, one could argue that the sleepiness in the OSA group could just be a normal age-related phenomenon.
A similar pattern of findings was shown in self-reported sleep quality pre- and post-treatment comparison as measured by the PSQI. The proportion of poor sleepers fell from 82% before treatment to 27% after treatment in the OSA group, as compared to
30% in the control group. As patients and controls did not differ on the PSQI (except for the trend for the OSA group having worse daytime dysfunction, which essentially taps on sleepiness), and the two groups were mostly comparable on PSG measures, it could be concluded that sleep of the OSA group was largely normalized. In fact, the OSA group 115
had fewer obstructive apneas and hypopneas and fewer arousals associated with respiratory events, suggesting that our participants diagnosed with OSA showed normal breathing during sleep.
While the percentage of individuals of the OSA group in the abnormal range on the FOSQ dropped by almost half after treatment, 43% of them were still in the clinically significant range. While this appears to be a rather alarming figure, 33% of our controls were also below the cut-off for good functional outcomes. It should be noted that this cut-off score was derived by systematic review of diverse sources of information regarding the relationship of OSA and daytime functioning specifically for the purpose of clinical management of commercial motor vehicle operators (Hartenbaum et al., 2006).
Given the specific purpose of this cut-off score, it is probably too "stringent" for individuals who are not in professions involving operation of heavy machinery. Another factor is that most of our participants were likely older than the subjects under review in
Hartenbaum et al.'s report and were engaged in less demanding activity, and hence did not experience significant functional deficits even though their FOSQ score might be elevated. In this light, it may be useful to develop clinical cut-off scores adjusted for different age groups given the potential impact of aging on sleep.
Another way to study the effectiveness of CPAP treatment was to investigate if and how it affected our participants' sleep architecture. Due to the fact that our sleep clinic mostly conducts "split-night" diagnostic studies, in which patients are titrated on
CPAP for the second half of the night if breathing obstruction was obvious in the first half, we only had valid pre-treatment data measured in split-night studies on three sleep architecture variables for 15 participants in the OSA group. Pre- and post-treatment 116
comparison of their sleep latency, sleep efficiency, and Arousal with Respiratory Index in those participants showed that their sleep latency did not change significantly, but their sleep efficiency improved significantly, and their Arousal with Respiratory Index significantly reduced after treatment. These findings provided support for the benefits of
CPAP in reducing sleep fragmentation and improving sleep efficiency.
To further investigate whether the individuals with OSA treated with CPAP had normalized sleep, their post-treatment PSG variables were compared to those of the control group. The OSA group did not differ from the controls on most sleep variables, and showed even better results in measures of sleep apnea, e.g., they showed a lower
Obstructive Hypopnea Index and Arousal with Respiratory Index, in addition to the lower
RDI as reported earlier. This provides further evidence for normalized breathing as well as sleep architecture for the OSA group. In view of the abnormality in sleep architecture commonly found in untreated patients with OSA in previous studies (Bardwell et al,
2000), our finding suggests that CPAP is effective not only in normalizing breathing during sleep, but also in improving sleep architecture. It should be noted that although the control group showed some elevated scores on a few PSG measures, the mean scores for the controls were all still low (0.3 for the Obstructive Apnea Index and 3.1 for the
Arousal with Respiratory Index) and probably did not indicate any clinical implications.
Performance of Individuals with OSA Treated with CPAP on Neurocognitive Measures
To test our second, third, and fourth hypotheses, 27 age- and education-matched healthy controls from the community were tested to compare with the OSA group on experimental working memory tasks and standardized neuropsychological tests. 117
Basic Storage and Rehearsal Components of Working Memory
In support of our second hypothesis that the OSA group would be comparable to controls on tests of basic storage and rehearsal components in working memory, I found no significant differences between the OSA group and the control group on measures of the phonological loop or on the visual-spatial sketchpad. The two groups showed similar forward Digit Span from the Verbal-Spatial Dual Task, and similar forward Visual Span from the WMS-R. The groups were also comparable in performance on the Verbal and
Spatial Memory Scanning Tasks. These findings suggest that treated participants with
OSA have normal maintenance capacity and basic rehearsal function of the phonological loop and the visual-spatial sketchpad.
Central Executive Function of Working Memory
Our data also provided evidence for our third hypothesis that participants with
OSA would have difficulties with tasks that involve executive function even after treatment, as compared with healthy controls. To measure the role of the central executive, we used the working memory model of Baddeley and colleagues, which suggests that the Central Executive is required in "attentionally demanding" situations when the tasks in hand call for more than standard maintenance and rehearsal of information. The Verbal and Spatial 2-back tasks were used as measures of on-line processing and updating of information, which required the ability to focus and divide attention. The Verbal-Spatial Dual Task was presumed to require focused and divided attention, and switching between responses. The Working Memory Span task required focused and divided attention and interface with verbal long-term memory. Each of these tests of the Central Executive was paired with control procedures in order to parcel out 118
the contribution of the more basic components, as suggested by Verstraeten and colleagues (Verstraeten & Cluydts, 2004; Verstraeten et al, 2004; Verstraeten et al,
2000). The two groups did not differ on most of the tests of the basic components (e.g. the Verbal and Spatial 0-back conditions, the Single conditions (List Memory and Spatial
Tracking) of the Verbal-Spatial Dual Task, and the Single conditions (Word Span and
Sentence Verification Span) of Working Memory Span task). The one exception where between-group differences were found was the slower RT in the OSA group on the target absent trials in the Verbal 0-back condition. In contrast to the basic storage and rehearsal function and in support of our hypothesis, the OSA group showed lower accuracy on both the Spatial and Verbal 2-back tasks, and worse performance on the Sentence Verification
Span in the Dual condition of the Working Memory Span task.
Findings on the N-back tasks provide direct evidence supporting our hypothesis that individuals with OSA treated with CPAP have difficulty with working memory tasks requiring the central executive in focusing attention, online processing and updating information, but not with the basic components of working memory. The slower verbal
RT in the OSA group on verbal target absent trials of the Verbal Memory Scanning Task might suggest a trade-off between slower processing and normal accuracy to compensate for an underlying psychomotor speed deficit. Alternatively, the slower RT could reflect a more cautious response bias on absent trials. Given that a similar RT slowing was not seen on the Spatial 0-back task, however, RT slowing or response bias may not be a general phenomenon and may simply reflect some variability in the OSA group's performance. 119
The central executive is also presumed to be needed when an individual is required to perform two tasks at the same time such as the Verbal-Spatial Dual Task. No significant between-group difference was found when performance was measured either by a relative measure of dual task ability, e.g. Combined Dual Task Score (D), or when the effects of combining the tasks were examined for each task separately (e.g., changes in the List Memory and the Tracking tasks in the Single condition compared to the Dual condition). The lack of significant differences between participants with OS A treated with CPAP and healthy controls in the present study could be due to the relatively low level of difficulty of this task. This task was developed for investigating the central executive functioning in patients with Alzheimer's disease (Baddeley et al., 1986). In comparison to patients with Alzheimer's disease, individuals with OS A, especially those treated with CPAP, are generally functioning at a much higher level, and the Verbal-
Spatial Dual Task may be insensitive to residual deficits in high functioning individuals.
Our laboratory is in the process of developing tools that are more sensitive. One main aspect to adjust is the level of engagement of the non-mnemonic component task.
The current tracking task has the advantage of being non-threatening in its presentation but participants can slow down or stop even just briefly and focus on recalling the digits, thereby possibly invalidating the presumed dual nature of the task.
Alternatively, it could be that the component tasks of the Verbal-Spatial Dual Task engage two different modalities (verbal/auditory and spatial/visual), which according to the domain-specific view of the working memory capacity are different systems and draw on separate resources/attention and therefore do not "compete" for attention (Shah &
Miyake, 1996). However, this view has been criticized and revised in the literature and 120
current evidence suggests that while domain-specific competencies may affect performance on executive function tasks, such tasks have a "commonality in their measurement of a domain-free ability to control attention" (Feldman Barrett et ah, 2004;
Miyake, 2001). In this light, it might be that the Verbal-Spatial Dual Task adopted here was just not demanding enough on the domain-free central executive. This explanation is consistent with the findings of a previous study, which to our knowledge was the only other study that has used Verbal-Spatial Dual Tasks in individuals with OSA (Naegele et ah, 2006). In fact, Naegele's group adopted three different dual tasks to study working memory in patients before treatment and still did not find a difference from controls.
While one of their dual tasks was very similar to ours (list memory and tracking), the other two were all verbal-spatial in nature and therefore did not provide further information as to whether the usage of different modalities (verbal/auditory and spatial/visual) was the reason behind the normal performance.
Findings from another test using the dual task-paradigm in our study may shed some light on questions of whether modalities of presentation (auditory vs. visual), and of materials (verbal vs. spatial) affect performance in individuals with OSA. The Working
Memory Span is presumed to require the episodic buffer of the central executive. It requires storage (words) and on-line processing (sentence verification) at the same time and is structured like a dual task. As such, it was analyzed similarly to the Vebal-Spatial
Dual Task, with a combined measure and then with the component measures in the
Single and the Dual conditions. There were no significant between-group differences on the combined measure, Working Memory Span, or any of the measures in the Single condition. In contrast, there was a disproportionate reduction in accuracy in sentence 121
verification in the Dual condition in the OSA group, suggesting that participants in the
OSA group had difficulty maintaining their performance on the sentence verification task when they were maintaining and rehearsing a series of words in mind. In other words, their performance was less accurate than controls' on a component task when they multitasked reflecting a central executive deficit. This task used the same modality for presentation (visual) and for material (verbal) in the two component tasks and hence probably posed a higher demand on the central executive in allocating and dividing attention than the Verbal-Spatial Dual Task and appears to be a more sensitive task for measuring divided attention. A slightly different but not mutually exclusive explanation for the difference in results between the two tasks was that the Working Memory Span required maintenance (words) and processing (sentence verification) simultaneously while the Verbal-Spatial Dual Task required maintenance (list memory) and basic visual attention and psychomotor speed (spatial tracking). While the tracking task was more like a distractor task, the sentence verification required processing of information. As such, the Working Memory Span task may capture the function of the central executive in dividing attention better than the Verbal-Spatial Dual Task, which measured the ability to focus under distraction.
Taken together, our data provided support for the hypothesis that while individuals with OSA treated with CPAP performed at a level comparable to healthy controls on basic working memory tasks, they showed difficulties on working memory tasks requiring the central executive. Two out of the three tasks showed evidence for sub-normal performance in the OSA group as compared with their age- and education- matched healthy controls. One of the tasks was visual-spatial in nature (Spatial N-back), 122
and the other one verbal (Working Memory Span), suggesting that the executive difficulties in patients with OSA are multimodal and beyond the basic functioning of the slave systems of the working memory model. The OSA group performed at a comparable level to the controls on the Verbal-Spatial Dual Task, suggesting normal ability to focus attention under distraction and to divide attention between tasks of different materials (verbal and spatial) and presentation modes (auditory and visual).
Their deficits were on tasks requiring simultaneous storage and on-line processing and updating of information, and interfacing with long-term memory store.
While three recent studies have investigated working memory more specifically and systematically, they had a different focus from this study and therefore did not provide the same information offered by this study. Naegele and colleagues (2006) studied episodic memory, procedural memory, and working memory and were able to identify specific processes that were affected in their patients in comparison to a matched control group. However, while they found impairments in tests requiring maintenance and processing, these tests did not have a control condition or a parallel test to control for the basic processes involved. In contrast, each of our measures of the central executive function had such control, enabling us to conclude what specific processes were affected.
In addition, our study targeted treated patients, who seemed to have been neglected in the literature as evidenced by the paucity of studies, likely because they were assumed to perform at a normal level. Another study that investigated working memory in untreated patients presented findings that were quite consistent with this study (Lis et ah, 2008).
They found that there were working memory deficits in untreated patients that were unlikely to be attributable to basic processes. Also consistent with this study, they 123
reported that RT slowing was likely to be associated with more elementary cognitive processes involved in working memory tasks, and deficits in accuracy were more specific to working memory. Up to date, there was only one other study that investigated specifically working memory in patients with OSA post-treatment (Felver-Gant et ah, 2007). They showed that performance on a verbal 2- back task correlated with the other executive function task they used, the PASAT.
However, this study had a number of methodological problems that made its findings very difficult to interpret, as discussed in the introduction section.
Neuropsychological Tests
I hypothesized that patients and controls would not differ on the neuropsychological measures that do not require executive function and controlled attention because most of the neurocognitive impairments should have been reversed by
CPAP and the remaining difficulties were likely to be subtle changes on more complex tasks (Lojander et ah, 1999). This hypothesis was supported as the two groups had comparable performance on measures of their general cognitive function (WAIS-R), verbal skills (Vocabulary), visual-spatial and constructional function (Block Design and
Rey-Osterreith Complex Figure copying), basic attention (Digit and Visual Spans), vigilance (Psychomotor Vigilance Test), and verbal and visual memory (California List
Learning Test, delayed recall of Rey-Osterreith Complex Figure).
In contrast, among the tests that require complex attention and executive function, the two groups differed on four tests measuring sustained attention, selective attention, inhibition, and complex problem solving. The OSA group showed worse performance on
Digit Symbol and on the D2 Test of Attention. Both of these tasks require sustained 124
attention, psychomotor speed, and working memory. On the Stroop test, the two groups did not differ on basic word reading and color naming speed but the OSA group showed greater interference in the color-word condition. This finding indicates that the OSA group took longer to process and produce their responses when inhibition of an automatic response was required. The OSA group also performed significantly worse than controls on the Wisconsin Cord Sorting Test in making more perseverative and nonperseverative errors. These errors could be interpreted as difficulties in strategic planning in response to feedback and mental flexibility in a problem-solving task while concept formation appeared to be intact. The deficits could also be viewed as associated with maintaining and updating information online.
On the Trail Making test, the two groups did not differ on Trail Making A but there was a trend for a significantly slower completion time on Trail Making B for the
OSA group. This pattern suggested a potential specific difficulty of the OSA group with the executive component (mental set-shifting, inhibiting an automatic sequence) of Trail
Making B above and beyond the basic skills common to Trail Making A and B. Our findings on the Trail Making test supported the argument of Verstraeten and colleagues that the "executive deficit" on some neuropsychological tests as reported in some previous studies might disappear when the basic processes were controlled (Verstraeten,
2007; Verstraeten & Cluydts, 2004; Verstraeten et al, 2004).
The two groups did not differ on a few tests that are conventionally regarded as complex attention or executive function tasks, including backward Digit and Visual
Spans, Consonant Trigrams, and Mazes. These findings showed that the OSA group performed normally on at least some standardized neuropsychological tests of working 125
memory, in contrast to their significantly worse performance on the experimental working memory tasks. The discrepancy might be due to the higher sensitivity of our experimental tasks in detecting more subtle differences. It might also be that the working memory functions measured in standardized neuropsychological tests were different from those of the experimental tasks. For example, the backward Digit and Spatial Span tests required focused attention and simple manipulation, but did not require simultaneous storage and online or complex processing of information that was found to be affected on the 2-back Task and the Working Memory Span task. The Consonant Trigrams required maintenance of letters under distraction, which was a similar ability to that required in the
Verbal-Spatial Dual Task, and both were found to be normal.
Lastly, the OSA group took a significantly longer time to complete the Grooved
Pegboard test for the dominant hand, suggesting compromised fine motor dexterity and speed even with CPAP treatment.
Findings on the performance of patients with OSA after treatment on standardized neuropsychological measures in comparison with healthy controls are limited and equivocal in the literature. While most studies found improvement in neurocognitive function after CPAP, some suggested enduring executive dysfunction using tasks like
Trail Making, Verbal Fluency, and Stroop (Bedard et ah, 1993; Ferini-Strambi et al,
2003; Feuerstein et al, 1997). The current study provided support for the conclusion that some of the neurocognitive deficits, as measured by standardized tests, of patients with
OSA are more resistant to change. Our finding on sustained attention (D2 Test of
Attention, Digit Symbol) is in contrast to an earlier study showing normalized performance on attention/concentration tasks (Digit Symbol, Letter Cancellation) after 126
treatment (Bedard et al., 1993). Interestingly, patients in Bedard et a/.'s study had a higher post-treatment mean Sleep Apnea Index (10.3) than our patients' RDI (1.8) and therefore the sustained attentional deficits in our patients could not be attributed to residual respiratory disturbance during sleep or suboptimal CPAP treatment. A possible explanation for the discrepancy was that the mean diagnostic Sleep Apnea Index in
Bedard et a/.'s study (65.4) was higher than our OSA group's diagnostic RDI (42.2), although diagnostic RDI was not a significant predictor of Digit Symbol performance in our study. Considering the possibility of differences in sleepiness of patients after treatment, both studies found significant improvement in sleepiness but our study used
ESS and their study used MSLT, rendering comparison difficult. However, it should be noted that while there were significant improvements in sleepiness in patients after treatment in Bedard et a/.'s study, the MSLT score was still worse than that of normal controls. As there was no significant difference between the ESS scores of patients and controls in the current study, and since sleepiness was not a significant predictor of the sustained attention tasks, it was unlikely that residual sleepiness was the reason for the differences on attention tasks. However, it is possible that there could be higher level of objective sleepiness in our OSA group, which was not measured in our study and could be discrepant from participants' subjective report. The only significant predictor of performance on Digit Symbol was years of education in our study but again, it would not explain the difference between our findings and Bedard et a/.'s as our OSA group (15.1) had more years of education on average than theirs (11.1).
While most studies have reported deficits in fine manual dexterity in patients before treatment (Aloia et a/., 2004), no study to our knowledge has reported post- 127
treatment performance on fine motor dexterity. Our finding that patients' performance on Digit Span and Spatial Span were normal was in contrast to the finding of previous studies that all working memory tests, including digit span and spatial span remained abnormal even after CPAP treatment (Feuerstein et al, 1997; Naegele et al, 1998).
Predictors of Residual Neurocognitive Deficits
To explore the predictors of patients' neurocognitive function, scores on working memory tasks that differentiated the OSA group and controls were regressed on a group of variables, including individual factors (age, BMI, education), measures of apnea/hypopnea severity (diagnostic RDI, diagnostic minimum Sp02), and measures of sleep and sleepiness (post-treatment sleep efficiency, current ESS and PSQI - Global
Score). Not surprisingly, older participants were less accurate on both 2-back tasks measuring maintenance of online processing and updating of information. They were also slower on a task of psychomotor speed and dexterity. Higher education level was associated with better performance on all neurocognitive measures that distinguished the two groups. On top of these demographic variables, higher diagnostic RDI was associated with slower RT on the Spatial 2-back task, and the lower the patients' diagnostic minimum Sp02, the worse their performance on the Verbal 2-back task and the fine motor dexterity task. Higher sleepiness and poorer sleep quality were not associated with any neurocognitive deficits.
Previous studies have mostly found executive function to better correlate with hypoxemia than sleepiness in patients before treatment (Bedard et al, 1991a; Bedard et al, 1991b; Cheshire et al, 1992; Jones & Harrison, 2001; Kim et al, 1997; Naegele et al, 128
2006). An exception was a recent study on working memory deficits in patients with
OSA, which reported a correlation between objective sleepiness and accuracy in working memory tasks and a correlation between subjective sleepiness and RT in working memory as well as other tasks such as target detection and vigilance (Lis et ah, 2008).
The small number of studies that reported post-treatment executive deficits in comparison with controls did not investigate their relationships with hypoxemia indices or sleepiness (Bedard et ah, 1993; Naegele et al., 1995). The only other study that used experimental tasks (Verbal 2-back task) to study working memory did not have a control group and did not investigate predictors of performance. Our findings were in line with
Beebe and Gozal (2002)'s proposed model that long-term sleep fragmentation induced by disrupted breathing and hypoxemia before treatment contribute to disturbances of sleep and introduce cellular and biochemical stresses resulting in damage to brain regions, particularly the prefrontal cortex, leading to persistent executive deficits after treatment.
These findings presented a pattern that seems to suggest a more chronic dose-dependent impact of pre-treatment breathing disruptions and oxygen desaturation in individuals who have been stabilized with CPAP treatment for at least three months.
It is also interesting to note that pre-treatment minimum SpC>2 level predicted individuals' fine motor dexterity and speed after treatment when their nocturnal oxygenation was improved and normalized. This finding again suggested a persistent impact of hypoxemia that was not completely resolved with CPAP. Similar to the experimental working memory tasks, no comparison with previous studies could be made as no studies to our knowledge have investigated correlates of residual cognitive deficits on neuropsychological tests after treatment. Nevertheless, our finding that diagnostic 129
SpC>2 predicted post-treatment fine motor dexterity and speed was in keeping with the idea that executive and psychomotor dysfunction are more related to hypoxemia and sleepiness is more associated with attentional and memory deficits (Bedard et al., 1993;
Jones & Harrison, 2001; Sateia, 2003). The finding that SpC>2 level was a predictor for several different measures, but RDI only predicted Spatial 2-back RT raises the question of whether SpC>2 level is a more sensitive indicator of hypoxemia and a more important predictor of daytime functioning. A clinical implication is that clinicians may have to consider desaturation level more closely when evaluating severity of OSA and it may be helpful to develop clinical criteria using SpC>2 indices to identify individuals at high risk for cognitive and potentially medical sequelae.
A few "unexpected" associations were found in our analyses. Higher sleep efficiency was associated with poorer performance on working memory span. While this might be a spurious finding, a possible explanation is that higher sleep efficiency could be an indicator of a stronger need for recovery sleep due to long-standing sleep disruption by OSA (Mediano et al., 2007). Sleepiness was associated with higher Spatial 2-back accuracy. One possibility is that patients who are sleepy in their daily lives are more used to fighting sleepiness and applying strategies to maintain alertness. As the 2-Back task is rather challenging, the normally sleepier patients can perform better due to heightened arousal. Another possibility is that the patients with higher ESS scores are more sensitive to their internal states and do not necessarily have reduced alertness in their daily lives as compared to patients with lower ratings on ESS. These individuals may have higher self-awareness and put forth more effort in doing the task, leading to better performance. Replication is needed to substantiate this finding and to further 130
investigate the relationship between sleepiness and on-line processing performance.
Lastly, poorer post-treatment subjective sleep quality was associated with better performance on fine motor dexterity. Again, this could be a spurious finding and future replication is needed to verify its significance.
To summarize, factors that predicted residual cognitive deficits in individuals with OSA treated with CPAP were older age, lower level of education, lower diagnostic minimum SpC>2, and higher diagnostic RDI. Self-reported sleepiness and sleep efficiency were found to have an unexpected relationship with Spatial 2-back accuracy and Working
Memory Span performance in that higher sleepiness and poorer sleep efficiency were associated with better performance on the Spatial 2-back and Working Memory Span, respectively. Potential explanations were discussed. Overall, it appears that independently of demographic factors of age and education, high diagnostic RDI and low diagnostic minimal SpC>2 were reliable predictors of long-term cognitive outcomes after treatment. The persistent impact of sleep-disordered breathing highlights the importance of early detection and treatment of OSA, and the significance of hypoxemia indices in diagnostic and treatment considerations. It appears that sleepiness, as defined by self- reported propensity to fall asleep in daily life does not contribute systematically to cognitive deficits in a research setting.
Predictors of Psychosocial Outcomes
Regarding the analyses of predictors of current psychosocial functioning of treated patients, BDI, POMS, FOSQ, CFQ, QOL, and QSQ scores were analyzed in stepwise regression analyses on variables including age, BMI, diagnostic RDI, diagnostic 131
minimum Sp02, post-treatment sleep efficiency, ESS, and PSQI-Global Score. Overall, our data showed that self-reported sleepiness was a potent predictor of mood, functional outcomes, as well as quality of life, while subjective sleep quality predicted functional outcomes and social aspects of quality of life. Diagnostic RDI predicted self-reported cognitive failures. Age was found to be a significant predictor of quality of life and activity level, while BMI was associated with fatigue. Results are discussed in detail below.
Emotional Functioning
In terms of emotional functioning, depressive symptoms and negative affective states were predicted by post-treatment ESS score. While the mean level of depressive symptoms (2.8) endorsed by patients on the BDI was within the minimal range (i.e., 0 to
13) and the percentage of patients with BDI scores higher than the clinical cut-off was only 5%, our findings showed that patients' mood was still associated with their daytime sleepiness as measured by their ESS scores. Our findings did not directly address the question regarding the association of OSA and mood as there was no pre-treatment mood measure. Nevertheless, the finding that mood was predicted by sleepiness suggests a relationship between pathological sleepiness associated with OSA and mood. This is in keeping with the Wisconsin Sleep Cohort Study (Peppard et al, 2006), which demonstrated a causal link between OSA and depression, although the current study does not address the directionality of the relationship.
Several studies reported a lack of correlation between OSA and psychological problems (Lee, 1990; Phillips et al., 1996; Pillar & Lavie, 1998). Some authors argue that sleepiness and fatigue are misinterpreted as depression (Cassel, 1993), and yet others 132
suggest that links between mood and OSA dissipate when covariates such as age, BMI, and hypertension are controlled for in the analyses (Bardwell et al., 1999). In our study, age and BMI were not significant predictors of BDI scores, suggesting that the association of sleepiness and mood is independent of these demographic factors or health conditions. In regard to the argument that sleepiness and fatigue in patients with OSA are misinterpreted as mood symptoms, our findings on the POMS may offer some insight.
The total score and scores of the six affective states measured (i.e., tension-anxiety, depression-dejection, anger-hostility, vigor-activity, fatigue-inertia, and confusion- bewilderment) separate out somatic symptoms (vigor, fatigue) from affective symptoms
(anxiety, depression). I found that sleepiness (ESS) was associated with five of the six subscales namely, tension-anxiety, depression-dejection, anger-hostility, fatigue-inertia, and confusion-bewilderment, and interestingly, not with vigor-activity. The relationships with all the affective subscales and not one of the physical scales (vigor-activity) may suggest that the mood symptoms and their associations with sleepiness in patients with
OSA are independent or above and beyond the manifestations of fatigue.
Daily Functioning
Post-treatment FOSQ - Total Score was predicted by post-treatment ESS score and post-treatment PSQI - Global Score, with better functional outcomes associated with less sleepiness and better subjective sleep quality. Looking more closely at the subscales of FOSQ, ESS predicted activity level and vigilance, and PSQI predicted intimate relationships and sexual activity, general productivity, social outcome, and activity level, which was also predicted by age, with an interesting association that older participants reporting higher activity level. FOSQ - Total Score was predicted by PSQI - Global 133
Score, and subscales including subjective sleep quality, sleep duration, and daytime dysfunction. FOSQ - Activity level was predicted by PSQI - Global Score, and subscales of sleep quality, sleep duration, and daytime dysfunction.
CFQ score was predicted by diagnostic RDI with higher RDIs being associated with lower CFQ, indicating that the more disrupted the breathing was before treatment, the fewer cognitive complaints patients reported after treatment. While it may seem counterintuitive that patients with more serious OSA have better cognitive functioning, it may be the case that patients with higher RDI experience a bigger change in their cognitive functioning after using CPAP and hence endorse fewer daily cognitive failures.
It is conceivable that patients with more serious OSA before treatment actually have more cognitive difficulties than patients with milder OSA but they report fewer day-to day cognitive slips due to the perception of more significant improvement in functioning.
Quality of Life
On the visual analogue scale (QOL), age was the only significant predictor. Older participants with OSA reported higher quality of life as measured by QOL. No other studies to our knowledge have reported similar findings. Our speculations are that older participants might have developed better coping with OSA over the years and therefore were not affected by it as much. It could also be that older participants were mostly retired and hence their occupational functioning would not be as affected.
More predictors were found to predict sleep-specific quality of life as measured by QSQ. Self-reported sleepiness (ESS) predicted not only quality of life related to daytime sleepiness, but also all the other component scales, namely Diurnal Symptoms,
Nocturnal Symptoms, Emotions, and Social Interactions. More sleepiness was associated 134
with poorer quality of life on all of these QSQ subscales. Our findings suggest that residual sleepiness in individuals with OSA treated with CPAP still affects their quality of life to a significant degree, independent of their pre-treatment illness severity.
Impact of Executive Difficulties on Perceived Cognitive Efficiency and Quality of Life
I hypothesized that executive difficulties would be related to patients' subjective cognitive efficiency and with quality of life. Our correlational analyses of neurocognitive measures that showed sub-normal performance (i.e. Spatial 2-back accuracy, Sentence
Verification Proportion in the Working Memory Span Task, Trail Making B, and
WCST - nonperseverative errors) with CFQ, QOL and QSQ did not show significant relationships. In other words, our data do not support an association between measures of executive functions and measures of perceived cognitive efficiency and quality of life.
While there has not been any study investigating the relationships among these variables,
I hypothesized that the executive deficits would be correlated with daily cognitive difficulties in the OSA group, as well as their quality of life because executive deficits would likely affect cognitive efficiency on a day-to-day basis, which in turn would affect quality of life.
The lack of such relationships between executive deficits and daily cognitive difficulties may be due to a number of reasons. It may be that the cognitive slips measured by the CFQ are not associated with the more complex executive functions.
There may also be a discrepancy between one's objective functioning and subjective cognitive performance, especially for patients who have had the disorder and the associated deficits for a long time and might subjectively experience a significant 135
improvement in cognitive performance despite residual difficulties when tested on experimental tasks. With regard to the lack of relationship between executive deficits and quality of life, one of the reasons may be that there are multiple factors that predict quality of life and executive deficits are not a dominant factor. It may also be that, because a lot of our participants were retired, their lives may not be as affected by their executive difficulties in the absence of significant pressure to perform on complex cognitive tasks.
Factors Relating to Quality of Life
In analyzing the relationship between executive dysfunction and subjective cognitive problems, and quality of life, it was found that QSQ scores were correlated with
CFQ. It was an interesting finding in that while objective cognitive difficulties on experimental tasks and neuropsychological tests did not have a significant association with quality of life, subjective difficulties, which also did not significantly correlate with objective neurocognitive performance, were in turn associated with quality of life. This highlighted the importance of subjective perception of one's cognitive function in self- evaluation of quality of life.
To further investigate factors associated with quality of life, in addition to sleepiness and subjective cognitive failures, the QOL and the QSQ subscales were correlated with the post-treatment FOSQ-Total Score and each subscale, the BDI, and the
POMS-Total Score and each subscale. The QOL scores were not associated with any of the scales. Findings on the QSQ indicated that lower quality of life was associated with 136
poorer functional outcomes, more depressive symptoms, and more negative affective states.
These findings raise several interesting points as well as questions. It is rather surprising that the quality of life measures, the QOL and the QSQ were not correlated with each other. The QOL has previously been studied and found to be a valid and reliable measure of quality of life (de Boer et al, 2004). It has the benefits of not limiting patients' self-evaluation to set criteria and therefore may have more ecological validity in terms of capturing patients' quality of life as they see it. However, a global rating may suffer from a lack of specificity as patients' quality of life may be affected by various different factors at any one point in their lives. It also makes inter-individual comparison difficult because different people approach the task differently. In the current study, QOL was not associated with any other variables, including the other quality of life measure, QSQ, while the QSQ was associated with a few other measures.
The discrepancy is probably due to the commonalities between the QSQ and the other instruments in terms of the format (i.e., multi-item questionnaires) and the scope (i.e., sleep-related), and the distinctiveness of the QOL, which was discussed above.
The associations of quality of life with self-reported sleepiness, cognitive failures, functional outcomes, and affective symptoms suggest that quality of life is predicted by multiple factors. Using different quality of life measures, some previous studies reported
RDI being a significant predictor (Finn et al., 1998) and some studies did not find a correlation (D'Ambrosio et al, 1999; Sanner et al., 2000). Our findings seem to support current measures of sleepiness, mood, and functional outcomes being stronger predictors than diagnostic measures of severity of OS A. Our findings are in contrast with a 137
previous study suggesting that quality of life is predicted by depressive symptoms and not sleepiness (Kawahara et ah, 2005). The discrepancy is probably due to the low ESS scores of their patients even before treatment (mean = 9.7). Our findings suggest that residual sleepiness is still a factor for quality of life in patients with OSA after treatment.
Daily cognitive failures and functional outcomes are also associated with quality of life after treatment, and depressive symptoms and affective states also predict quality of life.
As quality of life is an important outcome for treatment of OSA, treatment considerations should extend beyond elimination of breathing disruptions to include daytime sleepiness, functional outcomes, and mood.
Other Findings of Interests
Between-Group Comparison on Sleep Related Variables After Treatment
The OSA group and controls did not differ on subjective sleep quality (PSQI), sleep habits (STQ), habitual sleepiness on a day-to-day basis (ESS), and momentary sleepiness at three time-points on the day of testing (SSS). Again, these findings support the effectiveness of CPAP in restoring sleep. Regarding the trend for significantly worse daytime dysfunction for the OSA group than the control group, it appeared that there might be some subtle difficulties in the OSA group after treatment in performing their daytime activities. Looking at the two items composing this subscale, the OSA group reported more trouble staying awake during daily activities and keeping enough enthusiasm to get things done. Nevertheless, the average score of the OSA group on these two items indicated that they had these difficulties less than once a week, again 138
supporting that their daytime function was mostly normalized with only mild residual self-reported difficulties.
Between-Group Comparison on Emotional Functioning After Treatment
No significant differences between the two groups were detected on the two mood measures, BDI and POMS at a significance level of .01. The percentages of people with clinical level of depressive symptoms were also comparable between individuals with
OSA treated with CPAP and controls. The lack of evidence for depressed mood post- treatment suggested that the differences in cognitive functioning between the two groups could not be explained by mood. There was a trend toward significantly worse fatigue and inertia on the POMS in the OSA group as compared to the control group, suggesting a tendency for individuals with OSA to have a reduced energy level even after treatment with CPAP. Regression analyses revealed that residual fatigue and inertia as measured by POMS was predicted by BMI and ESS, both of which may deserve more attention targets for adjunctive treatment along with CPAP.
Between-Group Comparison on Daily Functioning and Quality of Life
On the FOSQ, the OSA group scored significantly lower on the subscale of
"activity level" than controls. Consistent with the trend toward higher fatigue shown on the POMS, individuals with OSA seemed to continue to have a compromised level of energy and activity level after treatment.
All participants filled out the QOL to assess their quality of life currently and before treatment for the OSA group or currently and three months ago for controls. Pre- treatment QOL score of the OSA group were significantly lower than controls three months ago. The OSA group also showed a greater improvement in their quality of life 139
after treatment as compared with controls, who reported minimal difference in their quality of life over the past three months. This showed that OSA significantly influenced individuals' quality of life before treatment, and patients experienced a significant improvement in their quality of life with CPAP, which was not explained by the mere passage of time as controls did not report any significant change in their quality of life.
Undetected OSA in Community Dwelling Individuals
The prevalence of mild and moderate OSA in individuals participating in the study as controls is both interesting and alarming. Nine out of 27 of the control participants (33%) had an RDI greater than 5, and an additional 7 persons were excluded due to elevated RDI (i.e., greater than 15). Out of those 7 people, 2 were female (29%) and 5 were male (71%), totaling 21% of individuals without a diagnosis of OSA screened with PSG in the study. It is difficult to evaluate the representativeness of this figure due to potential biases in recruitment such as the profile of people volunteering for a research study and the additional people (9 out of 73 screened) not recruited for the study due to high risk for sleep disorders as assessed with the telephone screening. Nevertheless, the number of individuals found to have elevated RDI on overnight PSG provided some support to the notion that many people in the community with significant nighttime breathing disruptions are undetected (Young et al., 1997a). Different methods to estimate prevalence of OSA were used in the literature but our figures were in keeping with the estimates in general. For example, 22% of participants were reported having a
RDI of more than 15 in the Sleep Heart Health Study (Gottlieb et al., 1999). An estimate of 32% was found in the National Sleep Foundation's Annual Sleep in America Poll
(Hiestand et al., 2006). While most of these undiagnosed individuals were probably 140
asymptomatic as we screened out anyone reporting sleep-related difficulties during the health interview or appeared to be of high risk of OSA using the Sleep Disorders
Questionnaire and the Berlin Questionnaire, it would be of interest to investigate their neurocognitive functioning to understand whether their breathing disturbances affect their cognitive status before they show symptoms of OSA itself. The incidental finding regarding the prevalence of sleep-disordered breathing in the control group also highlights the importance of using an overnight PSG or at least home monitoring for screening controls in research studies. In the few studies with a control group that investigated post-treatment neurocognitive function, usage of a full PSG was an exception rather than the norm (Bedard et ah, 1993). Most studies just relied on screening interview or questionnaires (Feuerstein et al, 1997; Naegele et ah, 1998).
Strengths and Limitations of the Current Study
This study adopted a well-validated theoretical model of working memory to investigate basic maintenance and attentional capacity, and central executive function in individuals with OSA treated with CPAP. Multiple tasks were used to map onto the model in assessing the basic capacity and executive function in both verbal and visuo- spatial domains, which allowed the distinction between underlying capacity vs. the use of the capacity in more complex cognitive processes, in different modalities - verbal/auditory vs. spatial/visual. The usage of this model is particularly fruitful for these individuals in distinguishing which neurocognitive effects are reversible with treatment and which are more resistant, while bearing in mind the assumptions made on baseline functioning when drawing inferences based on comparison with controls post-treatment. 141
Such distinctions have important implications for understanding the disorder, evaluating treatment efficacy, and developing strategies for patients to cope with persistent cognitive difficulties after treatment.
In addition to the theoretical model, the design of this study put it in a position to fill in several gaps in the literature. This study targeted post-treatment performance of individuals treated with CPAP, in comparison with an age- and education-matched healthy control group, which is often lacking in cognitive studies of treated OSA. As such, we were able to address the question of whether treated individuals function at a comparable level to those who do not have OSA or other sleep disorders. This is important information for clinicians and patients in considering the costs and benefits of
CPAP.
This study is unique in investigating predictors of post-treatment neurocognitive performance. We were able to provide some preliminary evidence as to selective predictors of residual neurocognitive deficits. While our findings would not enable us to draw definitive conclusions regarding mechanisms of neurocognitive deficits in OSA, this study offers some insights into the significance of sleep-disordered breathing and hypoxemia before treatment in predicting post-treatment neurocognitive performance.
Although psychosocial outcomes were not the primary focus of this study, we included multiple instruments for measuring mood, functional outcomes, and quality of life. Similar to post-treatment neurocognitive function, predictors of post-treatment psychosocial outcomes have not been adequately studied, and this study offers some preliminary evidence for future studies to build on. 142
Another strength of this study was that all participants including controls received an overnight PSG study. This screening procedure proved to be crucial in the current study given the prevalence of sleep-disordered breathing found in the control group. The full PSG ensured that the control group was well characterized in terms of sleep quality and breathing abnormalities allowing better comparisons of cognitive and psychosocial functions between groups.
This study has several limitations, one of which is the potential self-referral bias of our participants. This bias could potentially introduce extraneous variables to our sample and reduce the representativeness of both our OS A group and control group. One might argue that treated individuals with OSA with more cognitive difficulties might be more inclined to participate in order to receive a neurocognitive assessment. However, there was no evidence to support this contention and even if it were the case, the same might apply to our control participants, hence minimizing the selection bias. While our study aimed at investigating functioning of individuals compliant to treatment with CPAP and hence at least 26 individuals with compliance lower than our requirement were not recruited, discontinued, or were excluded from the data analyses, our OSA group could systematically differ from other patients with OSA who were less compliant to treatment with CPAP.
As the goal of this study was to investigate residual deficits of individuals with
OSA treated with CPAP, one patient with RDI greater than 15 was excluded because his
OSA was not "treated". The exclusion might create a limitation for this study in terms of generalizability of findings as most individuals may not receive a follow-up PSG to investigate whether their breathing disruption is successfully eliminated. As a result, the 143
findings of this study may not capture all the cognitive difficulties that a patient under standard treatment may experience. However, it was important for the purpose of this study to keep RDI as a screening criterion so that conclusions could be drawn regarding the neurocognitive and psychosocial outcomes of individuals effectively treated with
CPAP.
This was a cross-sectional study and hence no conclusions about causality could be made. I opted not to expose participants to testing before treatment because it was essential to maintain the novelty of the executive function tasks and repeated testing might have defeated the purpose. A possible partial solution is to develop alternative forms of executive function tests and validation studies will have to be done to ensure the executive elements of the tests are not affected by repeated testing.
Another limitation of the study was the lack of an acclimatization night for the polysomnography due simply to resource limitations. The control participants might have been systematically "disadvantaged" as none of them had had an overnight PSG before. Nevertheless, the PSG of both groups appeared to be largely normal, except for some relatively elevated breathing disruptions in the control group, which were probably not related to the lack of acclimation.
There were a large number of analyses in this study and multiple comparisons contribute to an increase in Type I error (i.e. the probability of making false positive errors). I attempted to control that type of error by setting up a priori hypotheses, which required only a small subset of analyses, and I adopted a significance level of .05 for all the ANOVAs of the working memory tasks and regression analyses. I used a significance level of .01 to reduce Type I error for the neuropsychological battery and for 144
more exploratory analyses. Between-group comparisons on post-treatment PSG were evaluated with a significance level of .05 because it was important to control for Type II error and to be conservative in evaluating whether the two groups were equalized in their sleep.
Lastly, the gender distribution of the patient group and the control group was not identical as I had more males in the patient group but more females in the control group.
Five male (in comparison to 2 female) volunteers were screened out due to their elevated
RDI (>15). The extensiveness of the protocol with a full day of testing and the overnight sleep study made it very difficult to recruit further. The potential impact of this difference was lessened by the fact that sex was not found to be a significant predictor of any of our outcome measures.
While our sample size was sufficient for detecting moderate effect sizes for our core analyses based on our a priori power analyses, a larger sample size would have given us more power in regression analyses with the number of predictors of interest.
Future Research Directions
This study demonstrated the value of Baddeley's working memory model for studying neurocognitive function in OSA. The model allows careful delineation of subcomponent processes of complex cognitive functions and offers an elegant framework for studies of treatment effects and residual deficits. To facilitate wider usage of the model, collecting norms for the working memory tasks would be one valuable next step to take. While it is ideal to have a matched control group in a study, it is not feasible in some situations. Not only can the norms be used in research studies, but they can also 145
potentially be applied to clinical work when an individual patient is tested on these working memory tasks. In addition to normal controls of different age groups, norms can be collected using different clinical populations such as patients with Alzheimer's disease, frontal-temporal dementia, and acquired brain injuries.
The working memory tasks can potentially be applied in studies of experimental sleep deprivation, shift work, and sleep fragmentation for professionals like on-call physicians and long-distance truck drivers. They can also be used for studies with patients with other sleep disorders like narcolepsy, periodic limb movements, and insomnia.
As found in this study, daytime sleepiness and subjective sleep quality were important predictors of functional outcomes, mood, and quality of life. It is important to explore specific strategies targeting such factors. For example, it would be valuable to study how interventions like sleep hygiene education affect sleep quality and daytime sleepiness, and the respective impact on daytime functioning, emotional functioning, and quality of life.
In terms of testing, it has always been a challenge for researchers of OSA and cognitive functions to compare studies and to generate meta-analyses because different studies use different tests, and tests are grouped under different categories of cognitive function. It is imperative to use a standard list of tests and to report which tests are used to measure each ability. In this study, the recommendations of Decary et al. (2000) were adopted in choosing the neuropsychological tests with some minor adjustments. The battery seemed sensitive in differentiating treated patients with OSA and healthy controls. 146
The literature will benefit from more studies using this standardized battery to facilitate cross-study comparison.
In attempting to evaluate the impact of OSA on individuals' lives, this study suggests a rather disappointing relationship between patients' subjective cognitive functioning and their objective performance in neurocognitive testing. In order to measure patients' daily function more objectively, it may be fruitful to explore the use of the decision making models which have been applied in the sleep deprivation literature
(Harrison & Home, 1999; Harrison & Home, 2000a). These paradigms, among other skills, require flexible thinking and planning in light of new information, and are proposed to associate with "real-world" decision making. It would be interesting to adapt the model and test individuals with OSA on the skills necessary in making decisions in more realistic life situations, and to investigate the associations between their performance on such tasks and working memory tests, as well as on neuropsychological tests.
Due to the goal of this study, which was to understand the functioning of patients treated with CPAP, and the respective design and methodology adopted, I cannot answer the question of Adams and colleagues (2001) regarding whether neuropsychological tests are useful in identifying individuals who are more likely to experience cognitive deficits and whether such deficits affect daily functioning. To address this important question, I would need to assess patients' functioning before treatment and investigate the relationship between neurocognitive functioning and daily functioning. I could also monitor those patients with deficits and assess whether their deficits persist after treatment with CPAP in comparison with sham CPAP or other types of placebo. 147
CHAPTER FIVE: CONCLUSIONS
This was a pioneering study in several ways in investigating neurocognitive function and psychosocial outcomes in individuals with OSA treated with CPAP. The use of Baddeley's working memory model in addition to the standardized neuropsychological tests provided us with a systematic theoretical framework for understanding executive function in our participants. It was found that individuals treated with CPAP had global neurocognitive functioning comparable to age- and education-matched healthy controls but still showed difficulties in working memory tasks requiring the central executive for online maintenance and updating of both verbal and spatial contents. They also showed significantly worse performance on standardized neuropsychological tests requiring more complex attention and executive function, as compared to the healthy controls, although only a small percentage of the OSA group scored below clinical cut-offs on the neuropsychological tests as compared to published norms. The implications of this difference in findings are two-fold. First, most of these individuals would probably show performance that is within normal limits when tested clinically using conventional neuropsychological tests. The question is whether the
"normal" performance may still represent a decline in functioning. While it is difficult to ascertain this given the current design, the differences found between the OSA and the control groups suggest that the treated individuals with OSA potentially perform at a level below their premorbid functioning. This potential drop in functioning is consistent with some anecdotal reports of treated patients that they are "just not the same" as before, even though their OSA per se was effectively treated with CPAP. The second implication is that it is important to have a healthy control group in this line of studies in 148
order to understand how individuals treated for OSA compare to their peers. Given the number of community dwelling volunteers who were found to have OSA in our study even after initial screening for symptoms, overnight PSG is needed to ascertain the absence of sleep-disordered breathing in the control group.
Our study was the first one that investigated predictors of neurocognitive and psychosocial outcomes in treated individuals with OSA. While the design of our study does not allow us to speak directly to the debate of the mechanisms of neurocognitive deficits in patients with OSA before treatment, our analyses of individuals' residual neurocognitive difficulties suggested that in addition to age and education, diagnostic minimum Sp02, followed by diagnostic RDI significantly predicted post-treatment cognitive functioning. On the other hand, current habitual sleepiness, followed by subjective sleep quality, appeared to be consistent predictors of psychosocial outcomes including mood, daytime function, and quality of life in individuals treated with CPAP.
While the OSA group did not differ from healthy controls in their current sleepiness rating as a group, one-third of them still reported sleepiness above the clinically significant range. In view of the potent associations between sleepiness and psychosocial outcomes, residual sleepiness in some patients deserves more attention in assessment and treatment of OSA.
Taken together, our study provides a piece of evidence for the residual neurocognitive effects of OSA on individuals under effective treatment with CPAP. Our findings suggested that individuals with more severe OSA tended to perform worse in the long term even when their disorder was treated and their sleep was normalized. While
CPAP treatment also seemed to be effective in reducing daytime sleepiness, some 149
individuals still reported excessive sleepiness, which predicted their mood, functional outcomes, and quality of life. Further research is needed to elucidate the predictors of post-treatment excessive sleepiness and treatment targeting improvement of sleepiness, psychosocial outcomes, and cognitive functioning should be explored to compliment the treatment of OS A itself. 150
Table 1 Group Demographic Characteristics
OSA Group (N=37) Controls (N=27) Range Range
Age (years) 57.9 (9.5)a 40-76 56.7 (10.5)a 33-85 Sex (F: M) 15:22 19:8 Handedness (R: L) 34:3 25:2 Education (years) 15.1 (3.6)a 10-24 15.7 (3.2)a 11-23 BMI 33.5 (7.4)a 24.0-56.6 25.5 (5.0)a 18.1-38.0
Mean (SD)
Table 2 OSA Group Medical History
Range
Onset of OSA (years) 13.6 (9.7)a 2-53 Time since diagnosis of OSA (months) 25.6(21.1) 6-120 Duration CPAP treatment (months) 17.8(11.4) 3-47 Usage of CPAP per week (hours) 51.4(6.7) 35-70 CPAP compliance* 96.1 (5.6) 81.0-100
*Percentage of days with usage >4 hours in the last 3 months. For 73% of the participants, compliance was determined by reading on the smart card of their CPAP machines. Self report of usage was used for the rest. aMean(SD) 151
Table 3 Pre- and Post-treatment Comparisons on Respiratory and Hypoxemia Indices, Sleep Questionnaires, Daytime Function, and Quality of Life
OSA OSA OSA Pre- vs. Post Pre-CPAP Post-CPAP t(df) P Controls (N = 37) (N = 37) (N = 27)
RDI 42.2 (24.9)a 1.8(1.5) 9.52 (32) .000 4.0 (3.4) Minimum SpC>2 80.2 (9.8) 90.5 (2.4) 5.61 (27) .000 88.6(3.3) Mean Sp02 93.7 (3.5) 95.7 (1.7) 3.17 (29) .004 95.7 (0.9) ESS 14.4 (5.2) 8.3 (4.5) 7.52 (36) .000 6.6 (4.7) PSQI Global Score 8.5 (3.3) 4.4 (2.4) 5.85 (27) .000 4.6 (2.8) FOSQ Total Score 13.6 (4.60) 17.8(2.1) 7.64 (32) .000 18.4(1.5) QOL 49.8 (22.8) 81.5 (14.5) 6.86 (32) .000 78.0(14.4)
RDI, respiratory disturbance index; Sp02, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSIQ, Pittsburg Sleep Quality Index; FOSQ, Functional Outcomes of Sleep Questionnaire; QOL, quality of life visual analogue scale. aMean(SD) 152
Table 4 Comparison between the OSA Group Treated with CPAP and the Control Group on Polysomnographic Variables
OSA Controls (N = 37) (N = 27) t(df) P
Total Sleep Period 449.9 (38.0)a 466.2 (62.0) 1.30(62) .198 Total Sleep Time 378.4(48.1) 384.7 (77.0) 0.37 (62) .711 Sleep Latency 15.8 (10.5) 15.8 (12.6) 0.01 (62) .989 REM Latency 118.5(70.6) 117.7(68.6) 0.05 (62) .964 Sleep Efficiency (%) 81.5 (9.7) 79.7 (10.5) 0.736 (62) .465 #REM 3.3 (1.2) 3.6(1.2) 1.00(62) .322 # Awakenings 24.6 (13.6) 25.7 (10.2) 0.36 (62) .722 Stage 1 Sleep (%) 12.2 (8.6) 12.6 (7.9) 0.22 (62) .831 Stage 2 Sleep (%) 61.3 (10.4) 61.0 (8.4) 0.10(62) .924 Stage 3 & 4 Sleep (%) 5.4 (7.8) 5.7 (8.2) 0.14(62) .887 REM (%) 21.1 (6.5) 20.6 (5.3) 0.33 (62) .742 Central Apnea Index 0.1 (0.4) 0.1 (0.2) 0.88 (62) .384 Obstructive Apnea Index 0.03 (0.1) 0.3 (0.7) 1.80 (62) .084 Mixed Apnea Index 0.01 (0.05) 0.0 (0.0) 1.7 (62) .096 Obstructive Hypopnea Index 1.0(1.3) 2.8 (3.1) 2.80 (62) .008** Central Hypopnea Index 0.3 (0.7) 0.3 (0.7) 0.08 (62) .935 Respiratory Disturbance Index 1.7(1.5) 4.0 (3.4) 3.25 (62) .003** Mean Sp02 95.7 (1.6) 95.7 (0.9) 0.09 (62) .929 Minimum Sp02 90.2 (3.6) 88.6 (3.3) 1.9(60) .062 Maximum Sp02 99.0(1.0) 99.6 (2.2) 1.5 (62) .127 % Sleep Time with Sp02 < 90% 1.7 (9.4) 0.1 (0.1) 0.88 (62) .382 % Sleep Time with Sp02 < 80% 0.03 (0.2) 0.0 (0.0) 8.52 (62) .397 Total Arousal Index 13.3 (6.5) 13.3 (5.6) 0.01 (62) .990 Arousal with Respiratory Index 1.5(1.4) 3.1 (2.9) 2.7 (62) .012*
Sp02, peripheral oxygen saturation aMean(SD) *p < .05 **p < .01 153
Table 5 Comparison between the OSA Group Treated with CPAP and the Control Group on Verbal Working Memory Tasks
OSA Controls t P (N = 37) (N = 27) (df=<52 )
Verbal Memory Scanning Reaction Time (ms) 1300.8 (207.7)a 1319.1 (273.9) 0.30 .763 Target Present 1233.8(191.3) 1249.4 (252.8) 0.28 .779 Target Absent 1367.1 (261.3) 1385.3 (314.0) 0.25 .801 Set size = 2 1077.2 (194.5) 1072.9 (213.5) 0.08 .933 Set size = 4 1274.2 (227.8) 1272.0(267.1) 0.03 .973 Set size = 6 1453.0(251.6) 1489.7 (345.4) 0.49 .625 Accuracies (%) 89.5 (5.8) 89.8 (5.3) 0.18 .857 Target Present 87.8 (6.8) 86.8 (8.7) 0.53 .600 Target Absent 91.2 (7.4) 92.8 (3.6) 1.12 .267 Set size = 2 94.1 (8.3) 94.1 (9.8) 0.36 .721 Set size = 4 99.3 (2.3) 98.5 (3.3) 0.69 .490 Set size = 6 93.5 (7.4) 91.3 (8.5) 0.63 .533 Sub-Threshold (%) 0.0 0.0 — — Time-out (%) 2.7 (2.9) 3.0 (3.4) 0.38 .705 Verbal 0-back Reaction Time(ms) 707.4 (121.7) 635.7 (96.7) 2.53 .014* Target Present 692.1(131.1) 624.6(98.1) 1.65 .104 Target Absent 718.6(123.9) 630.6(98.81) 3.05 .003** Accuracies (%) 97.9 (2.5) 98.4(2.1) 0.80 .424 Target Present 97.1 (4.9) 98.1 (2.8) 1.01 .319 Target Absent 98.4 (2.2) 99.1 (1.0) 1.38 .174 Sub-Threshold (%) 0.0 0.0 — — Time-out (%) 0.6(1.1) 0.2 (0.7) 1.63 .108 Verbal 2-back Reaction Time(ms) 1219.9 (234.9) 1213.4 (270.8) 0.10 .918 Target Present 1113.4(284.4) 1092.5 (233.9) 0.31 .755 Target Absent 1266.3 (289.6) 1297.1(311.5) 0.41 .685 Accuracies (%) 75.8 (17.6) 83.6(10.9) 2.17 .034* Target Present 76.6(17.1) 81.7(12.3) 1.32 .190 Target Absent 75.4 (20.0) 85.5 (12.5) 2.32 .024* Sub-Threshold (%) 0.8(1.1) 0.5 (0.8) 1.25 .217 Time-out (%) 9.2(13.8) 3.9(7.1) 1.85 .070 aMean(SD) *p < .05 **p < .01 154
Table 6 Comparison between the OSA Group Treated with CPAP and the Control Group on Spatial Working Memory Tasks
OSA Controls t P (N = 37) (N = 27) (df=(52 )
Spatial Memory Scanning Reaction Time(ms) 1316.8 (286.0)a 1232.5 (295.4) 1.15 .255 Target Present 1346.9 (302.4) 1275.6 (335.4) 0.89 .377 Target Absent 1287.9(291.1) 1193.6(298.8) 1.27 .210 Set size = 2 1140.8(285.3) 1104.1(290.1) 0.51 .615 Set size = 4 1316.0(296.8) 1249.6(314.1) 0.86 .391 Set size = 6 1421.8 (321.6) 1293.9(318.2) 1.58 .120 Accuracies (%) 82.0 (5.6) 84.5 (5.9) 1.70 .095 Target Present 83.7 (9.5) 88.2 (9.3) 1.87 .066 Target Absent 80.3 (8.9) 80.7(8.1) 0.19 .846 Set size = 2 92.1 (4.9) 91.7(6.1) 0.34 .733 Set size = 4 83.4 (7.0) 86.3 (7.4) 1.59 .117 Set size = 6 75.9 (8.7) 79.5 (6.8) 1.78 .081 Sub-Threshold (%) 0.0 0.0 — — Time-out (%) 3.9 (4.5) 2.6 (4.2) 1.24 .221 Spatial 0-back Reaction Time(ms) 748.3 (128.6) 715.3 (155.7) 0.93 .357 Target Present 743.8 (140.1) 742.0(181.1) 0.05 .963 Target Absent 733.1 (128.6) 697.1 (150.0) 1.03 .308 Accuracies (%) 92.4 (2.8) 92.1 (3.6) 0.34 .739 Target Present 95.3 (6.7) 94.4 (8.5) 0.44 .659 Target Absent 90.8 (2.9) 90.7(3.1) 0.04 .966 Sub-Threshold (%) 0.0 (0.2) 0.1 (0.3) 0.87 .387 Time-out (%) 0.7 (1.2) 0.6 (0.9) 0.14 .889 Spatial 2-back Reaction Time(ms) 1153.6(225.6) 1084.5 (229.6) 1.20 .235 Target Present 1058.2 (239.5) 1020.9 (232.9) 0.62 .536 Target Absent 1228.3 (235.4) 1131.3(242.5) 1.61 .113 Accuracies (%) 78.0 (16.9) 87.1 (8.7) 2.81 .007** Target Present 79.6(17.3) 86.3(11.5) 1.75 .085 Target Absent 77.5 (19.7) 87.7(9.1) 2.77 .008** Sub-Threshold (%) 0.6 (1.0) 0.2 (0.5) 1.79 .078 Time-out (%) 6.6(11.8) 1.6 (2.4) 2.19 .033 aMean(SD) *p < .05 **p < .01 155
Table 7 Comparison between the OSA Group Treated with CPAP and the Control Group on the Verbal-Spatial Dual Task
OSA Controls t P (N = 37) (N = 27) (df= 62)
Dual Task (u.)* 90.8 (9.5)a 91.3 (6.6) 0.21 .832 Single Digit List Memory .95 (.04) 0.97 (.04) 1.06 .291 Spatial Tracking 300.6 (54.2) 312.3 (54.2) 0.82 .415 Dual Digit List Memory .88 (.10) .89 (.10) 0.60 .548 Spatial Tracking 267.2(57.1) 277.7 (59.7) 0.71 .482
* Combined score of tracking and digit span in comparison to performance on the single tasks (higher score indicates better performance). aMean(SD)
Table 8 Comparison between the OSA Group Treated with CPAP and the Control Group on the Working Memory Span Task
OSA Controls t P (N = 37) (N = 27) (df=>62 )
Working Memory Span 2.9 (1.0)a 3.3 (1.0) 1.80 .076 Single Word Span 4.7(1.0) 5.1 (1.0) 1.53 .132 Sentence Verification Span 3.8(1.4) 4.2(1.5) 1.15 .254 Proportion of Words .83 (.07) .83 (.06) 0.48 .634 Proportion of Sentences .74 (.11) .78 (.12) 1.30 .200
Dual Word Span 3.0(1.0) 3.4 (1.0) 1.65 .104 Sentence Verification Span 3.5 (1.2) 4.1(1.1) 2.01 .049* Proportion of Words .56 (.13) .60 (.13) 1.30 .199 Proportion of Sentences .68 (.10) .78 (.10) 3.95 .000** aMean(SD) *p < .05 **p < .01 156
Table 9 Comparison between the OSA Group Treated with CPAP and the Control Group on Standardized Neuropsychological Tests (Raw Scores)
OSA Controls t P (N = 37) (N = 27) (df=( 52)
General Intelligence WAIS-R Vocabulary 56.6 (9.4)a 61.3 (5.9) 2.45 .017* WAIS-R Block Design 30.8 (8.3) 33.8(10.1) 1.32 .193 Estimated FSIQ 114.6(13.4) 121.0(11.6) 2.00 .050 Attention & Concentrationi D2 Test of Attention§ Total Items 412.9 (90.7) 467.9 (80.7) 2.74 .008** Total Errors 13.9 (10.7) 16.7(11.2) 1.02 .311 Concentration 157.6 (32.7) 185.2(41.6) 2.98 .004** Fluctuation Rate 11.5(4.3) 10.8 (3.2) 0.72 .473 Digit Symbol§ 53.1 (8.0) 60.7 (10.4) 3.3 .002** PVT Mean RT 280.3 (37.5) 261.4 (30.5) 2.09 .041 Slowest 10% RRT 2.4 (0.45) 2.61 (0.42) 2.06 .044 Lapses 2.3 (2.5) 1.0(1.8) 2.11 .038* WAIS-R Digit Span 16.1 (4.0) 16.9 (4.2) 0.72 .474 Forward 8.5 (2.3) 8.6(2.1) 0.26 .798 Backward 7.7 (2.5) 8.2 (2.5) 0.86 .393 WMS-R Visual Span 17.1 (3.1) 18.0 (2.9) 1.17 .246 Forward 9.1 (1.9) 9.3(1.9) 0.44 .663 Backward8 8.1 (1.8) 8.7 (1.6) 1.53 .131 Memory CVLT-II Total 48.6 (9.3) 52.7(9.1) 1.74 .086 Trial 1 5.6(1.9) 6.0(1.8) 0.79 .431 Trial 5 12.3 (2.2) 13.1(2.1) 1.42 .162 Trial B 5.6(1.5) 5.2 (2.0) 0.85 .397 S-D Free Recall 11.1 (2.7) 11.1 (3.3) 0.08 .935 S-D Cued Recall 12.5 (2.3) 12.6 (2.3) 0.25 .808 L-D Free Recall 12.1 (2.8) 12.4 (2.9) 0.40 .688 L-D Cued Recall 12.6 (2.7) 12.8 (2.8) 0.32 .752 Proactive Interference -0.6 (50.5) -1.3 (32.0) 0.06 .953 Rey-0 CFT Copy 30.7 (3.5) 31.3 (3.4) 0.68 .502 Recall 17.3 (6.1) 19.2 (6.4) 1.21 .229 % Recall/Copy 56.7(18.1) 60.6 (16.4) 0.87 .388 157
Table 9 (continued) Comparison between the OSA Group Treated with CPAP and the Control Group on Standardized Neuropsychological Tests(Raw Scores)
OSA Controls t P (N = 37) (N = 27) (df=i 52)
Executive Function ccc§ 9" 15.0 (0.2) 15.0 (0.2) .313 .755 18" 10.4 (2.8) 11.1(2.5) 1.37 .177 36" 9.5 (2.7) 10.6 (2.4) 1.07 .290 Total 45.1 (7.8) 46.6 (8.5) 1.68 .097 WCST§ Categories 4.8 (1.6) 5.6(1.1) 2.26 .028* Perseverative Errors 17.2(11.9) 9.5 (9.9) 2.71 .009** Nonpersev. Errors 16.3(11.5) 7.3 (5.5) 3.77 .000** Failure to Maintain 0.95 (1.45) 0.41 (1.2) 1.58 .119 Correct/Total 0.71 (0.14) .82(0.10) 3.28 .002** Trail-making Test A 26.9 (8.3) 23.6 (7.9) 1.76 .084 B§ 74.3 (26.1) 61.7(21.5) 2.04 .045* WISC-III - Mazes§ 21.6(4.3) 22.9 (4.8) 1.05 .298 S troop Word 102.6 (14.7) 97.7 (14.7) 1.31 .194 Color 71.9(11.9) 73.0 (8.4) 0.42 .673 Color-Word 39.0 (7.7) 42.5 (7.8) 1.77 .082 Interference -0.91 (5.8) 4.2 (5.9) 3.43 .001** Psychomotor speed Pegboard Dominant 73.0(11.0) 64.7(10.1) 3.04 .004** Non-Dominant 80.5 (15.6) 73.8 (12.6) 1.82 .074
§ Tests that involve executive function and controlled attention. aMean(SD) *p < .05 **p < .01 158
Table 10 Number of Participants and Within-group Percentages of the OSA Group Treated with CPAP and the Control Group with Scores Below Clinical Cut-offs Defined by Norms on Standardized Neuropsychological Tests
OSA Controls (N=37) (N=27)
General Intelligence WAIS-R Vocabulary (Scaled score) 0(0)a 0(0) WAIS-R Block Design (Scaled score) 0(0) 0(0) Estimated FSIQ (FSIQ < 70) 0(0) 0(0) Attention & Concentration D2 Test of Attention§ (Standard score) Total Items 0(0) 0(0) % Errors 0(0) 0(0) Concentration 0(0) 0(0) Fluctuation Rate 1 (2.7) 0(0) Digit Symbol§ (Scaled score) 0(0) 0(0) PVTb Slowest 10% RRT (< 2.3) 18 (48.6) 5(20) Lapses (>3) 7 (8.9) 2(8) WAIS-R Digit Span (Scaled score) 0(0) 0(0) Forward (Percentile) 1 (2.7) 0(0) Backward^ (Percentile) 0(0) 0(0) WMS-R Visual Span (Scaled score) 0(0) 0(0) Forward (Percentile) 0(0) 0(0) Backward (Percentile) 0(0) 0(0) Memory CVLT-II Total (T-score) 0(0) 0(0) Trial 1 (z-score) 4 (10.8) 2 (7.4) Trial 5 (z-score) 0(0) 0(0) Trial B (z-score) 2 (5.4) 0(0) S-D Free Recall (z-score) 0(0) 0(0) S-D Cued Recall (z-score) 0(0) 0(0) L-D Free Recall (z-score) 0(0) 0(0) L-D Cued Recall (z-score) 1 (2.7) 0(0) Proactive Interference (z-score) 2 (5.4) 1 (3.7) Rey-O CFT (z-score) Copy 4(10.8) 2 (7.4) Recall 0(0) 0(0) % Recall/Copy 0(0) 0(0) 159
Table 10 (continued) Number of Participants and Within-group Percentages of the OSA Group Treated with CPAP and the Control Group with Scores Below Clinical Cut-offs Defined by Norms on Standardized Neuropsychological Tests
OSA Controls (N=37) (N=27)
Executive Function ccc§ 9" (z-score) 6(16.2) 3(11.1) 18"(z-score) 4(10.8) 2 (7.4) 36" (z-score) 3(8.1) 1 (3.7) WCST§ Categories (Percentile) 2 (5.4) 0(0) Perseverative Errors (T-score) 3(8.1) 0(0) Nonpersev. Errors (T-score) 4 (10.8) 0(0) Failure to Maintain (Percentile) 1 (2.7) 0(0) Trail-making Test A (T-score) 0(0) 0(0) B (T-score) 0(0) 0(0) WISC-III - Mazes§ (z-score) 0(0) 1 (3.7) Stroop (T-score) Word 0(0) 0(0) Color 1 (2.8) 0(0) Color-Word 1 (2.8) 0(0) Interference 1 (2.8) 0(0) Psychomotor speed Pegboard (T-score) Dominant 0(0) 0(0) Non-Dominant 1 (2.7) 0(0)
§ Tests that involve executive function and controlled attention. Impairment was defined by a z-score less than -1.5, a T-score less than 35, a scaled score less than 6, a standard score less than 78, or less than the 7th percentile. a Number (%) of participants in the range of impaired performance. b Criteria based on performance of normal individuals after sleep restriction to less than 5 hours for 7 days (Dinges et al., 1997). 160
Table 11 Comparison between the OSA Group Treated with CPAP and the Control Group on Sleep Questionnaires
OSA Controls t P (N = 37) (N = 27) (df=<62 )
Epworth Sleepiness Scale (ESS) 8.3 (4.5)a 6.6 (4.7) 1.48 .144 Stanford Sleepiness Scale (SSS) Morning 1.9(0.9) 1.8(0.9) 0.37 .710 Noon 1.8 (0.8) 2.1 (0.6) 1.40 .166 Afternoon 2.6 (0.8) 2.2(1.0) 1.82 .073 Pittsburg Sleep Quality Index (PSQI) Global Score 4.4 (2.4) 4.6 (2.8) 0.29 .776 Subjective Sleep Quality 0.7 (0.6) 0.8 (0.7) 0.64 .526 Sleep Latency 0.4 (0.6) 0.7 (0.6) 1.87 .066 Sleep Duration 0.6 (0.8) 0.6 (0.7) 0.15 .880 Habitual Sleep Efficiency 0.3 (0.7) 0.6 (0.9) 1.38 .173 Sleep Disturbances 1.2 (0.5) 1.2(0.5) 0.47 .641 Use of Sleep Medication 0.2 (0.6) 0.1 (0.5) 0.29 .771 Daytime Dysfunction 0.9 (0.7) 0.6 (0.6) 2.14 .036* Sleep Timing Questionnaire (STQ) Sleep Latency (minutes) 11.3(12.5) 13.1 (8.6) 0.63 .529 Awakening (minutes) 18.3(31.3) 20.4(21.9) 0.30 .762
On both ESS and SSS, lower scores indicate less sleepiness; on PSQI, lower scores indicate better sleep quality. aMean(SD) *p < .05 **p < .01 161
Table 12 Comparison between the OSA Group Treated with CPAP and the Control Group on Mood Questionnaires
OSA Controls t P (N = 37) (N = 27) (df= 62)
Beck Depression Inventory (BDI)-•II 2.8 (4.7)a 2.9 (2.4) 0.01 .989 Profile of Mood States (POMS) Global Score 67.4 (24.0) 60.3 (18.2) 1.29 .204 Tension-anxiety 6.1 (5.0) 5.0 (4.7) 0.89 .375 Depression-dej ection 5.0(7.1) 5.2 (6.1) 0.11 .913 Anger-hostility 4.7 (6.2) 3.7 (4.0) 0.74 .465 Vigor-activity 17.5 (6.7) 19.0 (5.2) 0.96 .342 Fatigue-inertia 7.4 (6.2) 4.7 (4.9) 1.89 .064 Confusion-bewilderment 6.2 (4.6) 5.3 (4.5) 0.78 .436
Higher scores indicate more difficulty on both questionnaires. aMean(SD) 162
Table 13 Comparison between the OSA Group Treated with CPAP and the Control Group on Functional Outcome Measures
OSA Controls t P (N=37) (N=27) (df= 62)
Cognitive Failure Questionnaire (CFQ) 34.6 (15.2)a 33.0(13.2) 0.43 .668 Functional Outcomes of Sleep Q uestionnaire (FOSQ) Total score 17.6 (2.2) 18.4(1.5) 1.61 .113 Activity level 3.3 (0.6) 3.7 (0.3) 2.99 .004** Vigilance 3.4 (0.5) 3.5 (0.5) 0.96 .341 Intimacy 3.6 (0.6) 3.8 (0.5) 1.04 .304 General productivity 3.6 (0.5) 3.8(0.3) 1.48 .143 Social outcome 3.7 (0.5) 3.7 (0.5) 0.06 .953
Higher scores indicate more difficulty on CFQ and better functioning on FOSQ. aMean(SD) *p < .05 **p < .01 163
Table 14 Comparison between the OSA Group Treated with CPAP and the Control Group on Quality of Life Measures
OSA Controls t P (N=37) (N=27) (df= 62)
Quebec Sleep Questionnaire (QSQ) Daytime sleepiness 35.2 (7.52)a N/A Diurnal symptoms 55.2(13.8) N/A Nocturnal symptoms 43.2 (5.4) N/A Emotions 31.8(4.0) N/A Social interactions 25.9 (2.9) N/A Visual Analogue Scale (QOL)b Pre-CPAP 49.8 (22.8) 78.7 (13.3) 5.74 .000** Post-CPAP 81.5(14.5) 78.0 (14.4) 0.92 .364 QOL Difference 31.8 (26.6) -0.7 (10.0) 5.89 .000**
Higher scores indicate better quality of life for both measures. aMean(SD) bOn the QOL, we asked the controls to indicate their quality of life currently and three months ago. *p < .05 **p < .01 Table 15a Significant Predictors of Neurocognitive Outcomes Based on Stepwise Regressions in Participants with OSA Treated with CPAP
Verbal 2-back Spatial 2-back Working Digit Symbol Wisconsin Card Grooved Memory Span Sorting Test Pegboard RT Accuracy RT Accuracy Sentence Raw Score Perseverative Dominant Verification Errors hand Predictors Proportion Age -0.66 -0.50 0.68 Education 0.35 0.44 0.38 -0.43 -0.28 Diagnostic 0.45 RDI Diagnostic -0.40 0.33 -0.31 min SpC>2 Sleep -0.42 efficiency (post) ESS (post) 0.43 PSQI (post) -0.32 Values shown are standardized beta weights ifi) of significant predictors.
RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index. Table 15b Stepwise Regression Model of the Verbal 2-back Task in Participants with OSA Treated with CPAP
Verbal 2-back RT Verbal 2-back Accuracy 2 2 B(SE) fi R A t P B(SE) fi R A t P (df=26) (df=22)
Age -0.92(0.20) -0.66 .42 -4.59 .000 Education Diagnostic RDI Diagnostic min SpC>2 -8.87(4.03) -0.40 .16 -2.20 .037 0.42(0.18) 0.33 .11 2.29 .013 Sleep efficiency (post) ESS (post) PSQI (post)
Model R2 .16 .42 Adjusted R2 .13 .49 F(df) 4.85(1,26) 12.86(2,23) P .025 .000 RDI, respiratory disturbance index; SpCh, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index Table 15c Stepwise Regression Model of the Spatial 2-back Task in Participants with OSA Treated with CPAP
Spatial 2-back RT Spatial 2-back Accuracy 2 2 B(SE) fi R A t P B(SE) fi R A t P (df=24) (df=22)
Age -0.68(0.19) -0.50 .29 -3.48 .002 Education 1.34(0.55) 0.35 .12 2.44 .023 Diagnostic RDI 4.21(1.73) 0.45 .20 2.43 .023 Diagnostic min SpC>2 Sleep efficiency (post) ESS (post) 1.27(0.41) 0.43 .16 3.10 .005 PSQI (post)
Model R2 .20 .57 Adjusted R2 .17 .51 F(df) 5.92 (1,24) 9.76 (3,22) P .023 .000 RDI, respiratory disturbance index; SpCh, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index Table 15d Stepwise Regression Model of the Working Memory Span - Sentence Verification Span (Proportion) in Participants with OS A Treated with CPAP
Sentence Verification Span (Proportion) B(SE) fi R2A t(df=23) p
Age Education o.oi (0.004; .19 2.66 .014 Diagnostic RDI Diagnostic min SpC>2 Sleep efficiency (post) -0.004 (o.oo: -2.54 .018 ESS (post) PSQI (post)
Model R2 .37 Adjusted R2 .31 F(df) 6.68 (2,23) P .005 RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index
ON -J Table 15e Stepwise Regression Model of Digit Symbol in Participants with OSA Treated with CPAP
Digit Symbol (Raw Score) B(SE) fi R2A t(df=26) p
Age Education 1.04(0.50) 0.38 .14 2.08 .047 Diagnostic RDI Diagnostic min SpC>2 Sleep efficiency (post) ESS (post) PSQI (post)
Model R2 .14 Adjusted R2 .11 F(df) 4.33 (1, 26) P .047 RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index Table 15f Stepwise Regression Model of Wisconsin Card Sorting Test in Participants with OSA Treated with CPAP
Wisconsin Card Sorting Test (Perseverative Errors) B(SE) fi R2A t(df=26) p
Age Education -1.62(0.67) -2.43 .022 Diagnostic RDI Diagnostic min Sp02 Sleep efficiency (post) ESS (post) PSQI (post)
Model R2 .19 Adjusted R2 .15 F(df) 5.92 (1, 26) P .022 RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index Table 15g Stepwise Regression Model of Grooved Pegboard Test in Participants with OSA Treated with CPAP
Grooved Pej2boar d (Dominant Hand) 2 B(SE) fi R A t(df=23) P
Age 0.84(0.12) 0.68 .49 7.14 .000 Education -1.04 (0.35) -0.28 .07 -2.93 .007 Diagnostic RDI Diagnostic min SpC^ -.37(0.11) -0.31 .15 -3.25 .004 Sleep efficiency (post) ESS (post) PSQI (post) -1.45 (0.43) -0.32 .10 -3.35 .003
Model R2 .80 Adjusted R2 .77 F(df) 23.25 (4, 23) P .000 RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index Table 16a Significant Predictors of Psychosocial Outcomes Based on Stepwise Regressions in Participants with OSA Treated with CPAP
Mood Functional Quality of Life Outcomes BDI-II POMS FOSQ CFQ QOL QSQ - QSQ - QSQ - QSQ - QSQ - Social Sleepiness Diurnal Nocturnal Emotions Interactions
Age 0.40 0.43 0.42 Education Diagnostic -0.45 RDI Diagnostic min Sp02 Sleep efficiency (post) ESS (post) 0.45 0.54 -0.42 -0.60 -0.50 -0.44 -0.41 -0.40 PSQI(post) -0.37 -0.31 Values shown are standardized beta weights (fi) of significant predictors.
RDI, respiratory disturbance index; Sp02, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index; BDI, Beck Depression Inventory; POMS, Profile of Mood States; FOSQ, Functional Outcomes of Sleep Questionnaire; CFQ, Cognitive Failures Questionnaire; QOL, quality of life visual analogue scale; QSQ, Quebec Sleep Questionnaire. Table 16b Stepwise Regression Model of the Beck Depression Inventory in Participants with OS A Treated with CPAP
Beck Depression Inventory 2 B(SE) fi R A t(df=27) P
Age Education Diagnostic RDI Diagnostic min sPo2 Sleep efficiency (post) ESS (post) 0.50(0.19) 0.45 .20 2.59 .015 PSQI (post)
Model R2 .20 Adjusted R2 .17 F(df) 6.71(1,27) P .015 RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index; BDI, Beck Depression Inventory. Table 16c Stepwise Regression Model of the Profile of Mood States in Participants with OS A Treated with CPAP
POMS - Total Score POMS - Depression-Dejection 2 2 B(SE) fi R t(df=27) P B(SE) fi R A t(df=27) P A
Age Education Diagnostic RDI Diagnostic min Sp02 Sleep efficiency (post) ESS (post) 2.08 (0.62) 0.54 .29 3.33 .002 0.52(0.24) 0.39 0.16 2.23 .035 PSQI (post)
Model R2 .29 .15 Adjusted R2 .27 .12 F(df) 11.12(1,27) 4.95(1,27) P .002 .035 RDI, respiratory disturbance index; Sp02, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index; POMS, Profile of Mood States. Table 16c (continued) Stepwise Regression Model of the Profile of Mood States in Participants with OS A Treated with CPAP
POMS - Anger-Hostility POMS - Vigor-Activity B(SE) fi R2 t(df=27) p B(SE) fi R2 A t(df=27)
Age Education Diagnostic RDI Diagnostic min Sp02 Sleep efficiency (post) ESS (post) 0.34(0.13) 0.46 .21 2.70 .012 PSQI(post) -1.58(0.43) -0.58 .33 -3.66 .001
Model R2 .22 .33 Adjusted R2 .18 .31 F(df) 7.28(1,27) 13.39(1,27) p .012 .001 RDI, respiratory disturbance index; Sp02, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index; POMS, Profile of Mood States. Table 16c (continued) Stepwise Regression Model of the Profile of Mood States in Participants with OS A Treated with CPAP
POMS - Fatigue-Inertia POMS - Confusion-Bewilderment B(SE) fi R2 t(df=26) p B(SE) fi R2 A t(df=27)
Age BMI 0.41(0.16) 0.38 .14 3.06 .005 Education Diagnostic RDI Diagnostic min sPo2 Sleep efficiency (post) ESS (post) 0.66 (0.22) 0.47 .27 3.06 .005 0.48(0.17) 0.47 0.22 2.78 .010 PSQI (post)
Model R2 .41 .22 Adjusted R2 .37 .19 F(df) 9.14(2,26) 7.70(1,27) P .001 .010
BMI, Body Mass Index; RDI, respiratory disturbance index; Sp02, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index; POMS, Profile of Mood States. Table 16c (continued) Stepwise Regression Model of the Profile of Mood States in Participants with OS A Treated with CPAP
POMS - Anger-Hostility POMS - Vigor-Activity B(SE) fi R2 t(df=27) p B(SE) fi R2 A t(df=27)
Age Education Diagnostic RDI Diagnostic min Sp02 Sleep efficiency (post) ESS (post) 0.34(0.13) 0.46 .21 2.70 .012 PSQI (post) -1.58(0.43) -0.58 .33 -3.66 .001
Model R2 .22 .33 Adjusted R2 .18 .31 F(df) 7.28(1, 27) 13.39(1,27) P .012 .001 RDI, respiratory disturbance index; Sp02, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index; POMS, Profile of Mood States. Table 16d Stepwise Regression Model of Functional Outcomes of Sleep Questionnaire in Participants with OS A Treated with CPAP
FOSO-- Total Score FOSO-- Activity Level 2 2 B(SE) fi R t(df=26) P B(SE) fi R A t(df=25) P A
Age 0.21 (0.01) 0.33 .12 2.31 .030 Education Diagnostic RDI Diagnostic min sPo2 Sleep efficiency (post) ESS (post) -0.21 (0.08) -0.42 .29 -2.67 .013 -0.04 (0.02) -0.32 0.09 -2.12 .044 PSQI (post) -0.33 (0.14) -0.37 .13 -2.36 .026 -0.10(0.04) -0.43 0.28 -2.86 .009
Model R2 .42 .49 Adjusted R2 .37 .43 F(df) 9.31(2,26) 8.13(3,25) P .001 .001 RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index; FOSQ, Functional Outcomes of Sleep Questionnaire. Table 16d (continued) Stepwise Regression Model of Functional Outcomes of Sleep Questionnaire in Participants with OS A Treated with CPAP
FOSO-- Vigilance FOSQ - Intimate Relationships & Sexual Activity B(SE) fi R2 t(df=27) B(SE) fi R2A t(df=23) p A
Age Education Diagnostic RDI Diagnostic min sPo2 Sleep efficiency (post) ESS (post) -0.08 (0.02) -0.66 .43 -4.52 .000 PSQI (post) -0.09(0.04) -0.43 0.19 -2.30 .031
Model R2 .43 .19 Adjusted R2 .41 .15 F(df) 20.41(1, 27) 5.29(1,23) P .000 .031 RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index; FOSQ, Functional Outcomes of Sleep Questionnaire. Table 16d (continued) Stepwise Regression Model of Functional Outcomes of Sleep Questionnaire in Participants with OS A Treated with CPAP
FOSQ - General Productivity FOSQ - Social Outcomes B(SE) fi R2 t(df=27) p B(SE) fi R2 A t(df=27)
Age Education Diagnostic RDI Diagnostic min Sp02 Sleep efficiency (post) ESS (post) PSQI (post) -0.08 (0.04) -2.28 .031 -0.12(0.03) -0.61 0.37 -3.95 .001
Model R2 .43 .37 Adjusted R2 .41 .34 F(df) 20.41(1,27) 15.61(1,27) P .000 .001 RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index; FOSQ, Functional Outcomes of Sleep Questionnaire. Table 16e Stepwise Regression Model of the Cognitive Failures Questionnaire in Participants with OS A Treated with CPAP
Cognitive Failures Questionnaire B(SE) fi R2A t(df=27) p
Age Education Diagnostic RDI -0.29(0.11) .20 -2.58 .016 Diagnostic min SpC>2 Sleep efficiency (post) ESS (post) PSQI (post)
Model R2 .20 Adjusted R2 .17 F(df) 6.66 (1, 27) P .016 RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index Table 16f Stepwise Regression Model of the Quebec Sleep Questionnaire in Participants with OSA Treated with CPAP
oso- Sleepiines s QSQ - Diurnal Symptoms B(SE) fi R2 t(df=27) p B(SE) Ji R2A t(df=27) A
Age 0.54(0.24) 0.35 .12 2.29 .030 Education Diagnostic RDI Diagnostic min sPo2 Sleep efficiency (post) ESS (post) -1.04(0.27) -0.60 .35 -3.85 .001 -1.67(0.51) -0.50 0.28 -3.30 .003 PSQI (post)
Model R2 .35 .40 Adjusted R2 .33 .35 F(df) 14.80(1,27) 8.60(2, 26)
P .001 .001 RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index; QSQ, Quebec Sleep Questionnaire Table 16f (continued) Stepwise Regression Model of the Quebec Sleep Questionnaire in Participants with OS A Treated with CPAP
OSO - Nocturnal Symptoms QSQ - Emotions B(SE) fi R2 t(df=27) p B(SE) fi R2 A t(df=26)
Age Education Diagnostic RDI Diagnostic min sPo2 Sleep efficiency (post) ESS (post) -0.56(0.22) -0.44 .19 -2.52 .018 -0.36(0.14) -0.41 .17 -2.64 .014 PSQI (post)
Model R2 .19 .38 Adjusted R2 .16 .33 F(df) 6.36(1,27) 7.8(2, 26) P .018 .002 RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index; QSQ, Quebec Sleep Questionnaire Table 16f (continued) Stepwise Regression Model of the Quebec Sleep Questionnaire in Participants with OS A Treated with CPAP
QSQ-- Social Interactions 2 B(SE) fi R A t(df=25) P
Age 0.12(0.04) 0.04 .17 3.06 .005 Education Diagnostic RDI Diagnostic min SpC>2 Sleep efficiency (post) ESS (post) -0.24(0.09) -0.40 0.28 -2.77 .010 PSQI (post) -0.32(0.15) -0.31 0.09 -2.15 .041
Model R2 .53 Adjusted R2 .48 F(df) 9.49 (3, 25) P .000 RDI, respiratory disturbance index; SpC>2, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index; QSQ, Quebec Sleep Questionnaire Table 16g Stepwise Regression Model of the Quality of Life - Visual Analogue Scale in Participants with OS A Treated with CPAP
Visual Analogue Scale (OOP B(SE) fi R2A t(df=25) p
Age 0.65(0.31) 0.40 .16 2.12 .045 Education Diagnostic RDI Diagnostic min SpC>2 Sleep efficiency (post) ESS (post) PSQI (post)
Model/?2 .16 Adjusted/?2 .12 F(df) 4.48(1,24) p .045
RDI, respiratory disturbance index; Sp02, peripheral oxygen saturation; ESS, Epworth Sleepiness Scale; PSQI, Pittsburg Sleep Quality Index Table 17 Correlations of Quebec Sleep Questionnaire with Functional Outcomes of Sleep Questionnaire, Beck Depression Inventory, and Profile of Mood States
Quebec Sleep Questionnaire (QSQ) Daytime Diurnal Nocturnal Emotions Social Sleepiness Symptoms Symptoms Interactions
Functional Outcomes of Sleep Questionnaire (FOSQ) Total .89 (.000)a .79 (.000)** .47 (.004)** .52 (.001)** .69 (.000)** Activity Level .81 (.000)** .86 (.000)** .50 (.002)** .53 (.001)** .65 (.000)** Vigilance .76 (.000)** .64 (.000)** .38 (.022)* .50 (.002)** .51 (.001)** Intimate Relationships .66 (.000)** .51 (.003)** .25 (.178) .33 (.074) .47 (.007)** General Productivity .89 (.000)** .79 (.000)** .45 (.006)** .53 (.001)** .65 (.000)** Social Outcomes .61 (.000)** .51 (.001)** .37 (.023)* .28 (.095) .62 (.000)** Beck Depression Inventory (BDI) -.62 (.000)** -.70 (.000)** -.41 (.000)** -.82 (.000)** -.55 (.000)** Profile of Mood States (POMS) Total -.34 (.039)* -.49 (.002)** -.15 (.391) -.67 (.000)** -.69 (.000)** Tension-Anxiety -.45 (.005)** -.57 (.000)** -.28 (.090) -.64 (.000)** -.66 (.000)** Depression-Dejection -.41 (.011)* -.56 (.000)** -.21 (.207) -.75 (.000)** -.76 (.000)** Anger-Hostility -.27 (.010)* -.39 (.018)* -.11 (.519) -.59 (.000)** -.63 (.000)** Vigor-Activityb .51 (.000)** .62 (.000)** .47 (.003)* .31 (.061) .35 (.036)* Fatigue-Inertia -.57 (.000)** -.80 (.000)** -.43 (.008)** -.75 (.000)** -.68 (.000)** Confusion-B ewilderment -.52 (.001)** -.65 (.000)** -.28 (.092) -.61 (.000)** -.64 (.000)**
Higher scores indicated favorable outcomes on the QSQ and the FOSQ. Higher scores indicated worse emotional functioning on the BDI and the POMS. ar(p-value),d/=28 *p < .05 **p<.0l bVigor-activity is scored in the opposite direction as the other subscales on the POM oo 186 Figure 1. Prefrontal Model of OS A Deficits. (Beebe & Gozal, 2002, p.3; reprinted with permission from corresponding author as copyright holder)
Sleep Intermittent hypoxia disruption ' and hypercarbia
Disruption of restorative Disruption of cellular or chemical features of sleep homeostasis • t Prefrontal cortical dysfunction V
Dysfunction of cognitive executive svstem
Behavioral Set Self regulation of Working Analysis/ Contextual inhibition shifting affect and arousal memory synthesis memory I Adverse dav time effects ^^IBM Problems in mentally manipulating information Poor planning and haphazard execution of plans Disorganization Poor judgement/deciskm-making Rigid thinking Difficulty in maintaining attention and motivation Emotional lability ('mood swings') Overactivity/impulsiviry (especially in children)
Figure I. The proposed prefrontal model. In this model. OSA-related sleep disruption and intermittent hypoxemia and hypercarbia alter the efficacy of restorative processes occurring during sleep, and disrupt the functional homeostasis and neuronal and glial viability within particular brain regions. Subsequent dysfunction of prefrontal cortical regions, manifested in dysfunction of a muliifaccted cognitive 'executive system', alters the functional recruitment of more primary cognitive abilities, thereby resulting in maladaptive daytime behaviors. The dotted arrow in the model signifies that the executive system is an eptphenomenoti arising from the activity of the PFC, rather than a true 'effect'. Figure 2. Working Memory Model. (Baddeley, 2000, p.21, reprinted with permission from publisher, Elsevier Limited)
Vlsuospatial Episodic Phonological sketchpad buffer loop -J— -t- visual Episoac Language serrarics LTM
trsntis in Cognitive Sciences Fig. 1. The current version of the multi-component work ing memory mocfol. The episodic buffer is assumed to be capa ble of storing information in a multi-dimensional code. It thus provides a temporary interface between the slave systems (the phonological loop and the visuospatial sketchpad) and LTM. It is assumed to be controlled by the central executive, which is re sponsible for binding information from a number of sources into coherent episodes. Such episodes are assumed to be retrievable consciously. The buffer serves as a modelling space that is sep arate from LTM, but whkh forms an important stage in long- term episodic learning. Shaded areas represent 'crystallized' cog nitive systems capable of accumulating long-term knowledge, and unshaded areas represent 'fluid' capacities (such as atten tion and temporary storage), themselves unchanged by learning. 188 Figure 3. Sequence of Events of the Verbal Memory Scanning Task.
500 ms 1200 ms 1800ms Q s M 600ms + + H z 3000 ms
Until responded 2 blocks X (TARGET!) each set New trial size 189 Figure 4. Sequence of Events in the Spatial Memory Scanning Task.
500ms 12 possible locations 600ms (2), 1200ms (4), 1800ms f 6}
3000ms
Until responded 2 blocks (TARGET!) X each set New trial size 190 Figure 5. Schematic of the Tracking Component of the Verbal-Spatial Dual Task. 191 Figure 6. Sequence of Events of the Verbal 2-Back Task.
12 possible locations 500ms \
2500ms
500ms 3 blocks X 24 trails 2500ms
500ms
TARGET! 192 Figure 7. Sequence of Events of the Spatial 2-Back Task
12 possible locations 500 ms \
2500ms 1, I /
500ms 3 blocks X 24 trails 2500ms
500ms
TARGET! Figure 8. The Number of Participants at Each Stage of the Protocol. OSA Henlthv Controls
Telephone Screening/Chart Review Telephone Screening n = 141(M=Sl.F=60) n = 73(M=31.F=42)
X Passed Screening Did not Pass Screening Passed Screening Did not Pass Screening n = 73 n= 63 (M=35, F=2S) n = 64 n = 9 (U=5. F=4) (M=46, F=32) Major Reasons: (M=26, F=3S) Major Reasons: -54 did not meet inclusion/ - ° did not meet exclusion criteria (18 mild inclusion/exclusion criteria OSA, 13 not oil' compliant to (3 had potential sleep CPAP. etc.) disorders. 2 were too - 2 became deceased young, 1 on antidepressant Written Consent - 6 insufficient diagnostic Written Consent potentially interfering information cognitive function, etc. I -1 diagnosed at another clinic and diagnostic information unavailable
Did not Sign Consent Signed Consent Did not Sign Consent n = 31 (M=17.F=14) n = 51 n= 13(M=4.F=0) Major Reasons. 1 Not Complete Complete Not Complete n=S(M=6,F=2) n= 3? n=16(M=S,F=8) Major- Reasons.: (M=14,F=21) Major Reasons: -1 could not be contacted -5 could not be contacted -3 excluded (diagnostic sleep -5 excluded (one male data not available, insufficient unable to perform diagnostic sleep data, bypass neurocogtutive tasks, eye. •surgery) injury, chronic fatigue, bladder issues, mood/sleep Current data analysis -4 withdrew (No longer Current data analysis interested, too many medical problems) issues to deal with, discontinued - 6 withdrew (5 did not CPAP use) have enough time to complete study, 1 left) 1 Included Excluded Included Excluded ti = 37 n = 2{M=l.F=l) n = 27 n = 3 (M=6. F=2) (M=22,F=15) Reason: (M=S.F=19) Reason: 7 (5 males, 1 post-treatment RDI >15 2 females) RDI >15 1 had compliance < 80% as and 1 male was later shown on smart card discovered to have had bypass surgery Figure 9a. Percentages of Participants with OSA Pre- and Post-Treatment and of Healthy Controls Scoring in the Clinically-Significant Range on the Respiratory Disturbance Index (RDI). Respiratory Disturbance Index >5 100 100 90 80 70 60 % Within 50 Group 40 29.6 30 20 10 5A.~- 0 OSA{Pre-treatment) OSA {Post-treatment) HC Figure 9b. Percentages of Participants with OSA Pre- and Post-Treatment and of Healthy Controls Scoring in the Clinically-Significant Range on the Epworth Sleepiness Scale (ESS). Epworth Sleepiness Scale > 10 100 90 80 75.7 70 60 % Within 50 Group 40 29.7 30 20 JLO. 10 0 OSA(Pre-treatment) OSA (Post-treatment) HC 196 Figure 9c. Percentages of Individuals with OSA Pre- and Post-Treatment and of Healthy Controls Scoring in the Clinically-Significant Range on the Pittsburg Sleep Quality Index (PSQI-Gobal Score). Pittsburg Sleep Quality Index > 5 100 90 82:1 80 70 60 % Within 50 Group 40 27 29.6 30 20 10 0 OSA(Pre-treatment) OSA (Post-treatment) HC 197 Figure 9d. Percentages of Participants with OSA Pre- and Post-Treatment and of Controls Scoring in the Clinically-Significant Range on the Functional Outcomes of Sleep Questionnaire (FOSQ-Total Score). Functional Outcomes of Sleep Questionnaire < 18 100 90 .JifeL-ci-. 80 70 60 % Within 50 ...43,2... Group 40 -33- 30 20 10 0 OSA(Pre-treatmervt) OSA (Post-treatment) HC 1 Figure 10. Interaction Effects between Task Condition and Group on Spatial N-Back Task: the OSA Group Deteriorated Disproportionately in Accuracies in the 2-back Condition as Compared to the Healthy Control Group. Spatial N-back Task - Accuracies 100 95 90 85 80 I 75 u 70 < 65 60 55 50 0-back 2-bac* "-HC 92.55 88.15 — OSA 93.01 78.55 Error bars represent standard error of each data point. Figure 11. Interaction Effects between Task Condition and Group on Working Memory Span: the OSA Group Deteriorated Disproportionately on the Measure of the Proportion of Sentences Verified in the Dual Condition as Compared to the Healthy Control Group. Working Memory Span 0.85 >CD 0.8 en Error bars represent standard error of each data point. 200 REFERENCES Adams, N., Strauss, M., Schluchter, M., & Redline, S. (2001). Relation of measures of sleep-disordered breathing to neuropsychological functioning. American Respiratory Critical Care Medicine, 163, 1626-1631. Alchanatis, M., Deligirogis, N., Zias, N., Amfilochiou, A., Gotsis, E., Karakatsani, A., et al. (2004). Frontal brain lobe impairment in obstructive sleep apnoea: A Proton MR Spectroscopy study. European Respiratory Journal, 24, 980-986. Alchanatis, M., Zias, N., Deligiorgis, N., Amfilochiou, A., Dionellis, G., & Orphanidou, D. (2005). Sleep apnea-related cognitive deficits and intelligence: An implication of cognitive reserve theory. Journal of Sleep Research, 14, 69-75. Aloia, M. S., Arnedt, J. T., Davis, J. D., Riggs, R. L., & Byrd, D. (2004). Neuropsychological sequelae of obstructive sleep apnea-hypopnea syndrome: A critical review. Journal of the International Neuropsychological Society, 10, 772- 785. Aloia, M. S., Ilniczky, N., Pasquelina, D. D., Perlis, M. L., Greenblatt, D. W., & Giles, D. E. (2003). Neuropsychological changes and treatment compliance in older adults with sleep apnea. Journal of Psychosomatic Research, 54, 71-76. American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Disorders (4th ed.). Washington, DC: American Psychiatric Association. Ancoli-Israel, S., Kripke, D. F., Ronald, J., Manfreda, J., Roos, L., & Kryger, M. H. (1999). Sleep-disordered breathing in community dwelling elderly. Sleep, 14, 486-495. Antonelli Incalzi, R., Marra, C, Lorena Salvigni, B., Petrone, A., Gemma, A., Selvaggio, D., et al. (2004). Does cognitive dysfunction conform to a distinctive pattern in obstructive sleep apnea syndrome? Journal of Sleep Research, 13(1), 79-86. Arias, M. A., Alonso-Fernandez, A., & Garcia-Rio, F. (2006). Obstructive sleep apnea as an independent risk factor for stroke and mortality [letter to the editor]. Stroke, 37, 1150. Ayalon, L., Ancoli-Israel, S., Klemfuss, Z., Shalauta, M. D., & Drummond, S. P. A. (2006). Increased brain activation during verbal learning in obstructive sleep apnea. Neurolmage, 31, 1817-1825. 201 Babu, A. R., Herdegen, J., Fogelfeld, L., Shott, S., & Mazzone, T. (2005). Type 2 diabetes, glycemic control, and continuous positive airway pressure in obstructive sleep apnea. Archives of Internal Medicine, 165, 447-452. Baddeley, A., Logie, R., Nimmo-Smith, I., & Brereton, N. (1985). Components of fluent reading. Journal of Memory and Language, 24, 119-131. Baddeley, A. D. (1996a). Exploring the central executive. The Quarterly Journal of Experimental Psychology, 49A(1), 5-28. Baddeley, A. D. (1996b). The fractionation of working memory. Proceedings of the National Academy of Sciences of the United States of America, 93, 13468-13472. Baddeley, A. D. (2000). The episodic buffer: A new component of working memory? Trends in Cognitive Sciences, 4, 417-423. Baddeley, A. D. (2001). Is working memory still working? American Psychologist, 56(11), 864-878. Baddeley, A. D. (2003). Working memory: Looking back and looking forward. Nature Reviews Neuroscience, 4, 829-839. Baddeley, A. D. (Ed.). (2002). Fractionating the central executive. Oxford: Oxford University Press. Baddeley, A. D., Bressi, S., Delia Sala, S., Logie, R., & Spinnler, H. (1991). The decline of working memory in alzheimer's disease. Brain, 114, 2521-2542. Baddeley, A. D., & Delia Sala, S. (1996). Working memory and executive control. Philosophical Transactions of the Royal Society of London. B., 351, 1397-1404. Baddeley, A. D., Delia Sala, S., Gray, C, Pagagno, C, & Spinnler, H. (1997). Testing central executive functioning with a pencil-and-paper test. In P. Rabbitt (Ed.), Methodology of frontal and executive function (pp. 61-80). East Sussex: Psychology Press, Publishers. Baddeley, A. D., & Hitch, G. J. (1994). Developments in the concept of working memory. Neuropsychology, 8(4), 485-493. Baddeley, A. D., Logie, R., Bressi, S., Delia Sala, S., & Spinnler, H. (1986). Dementia and working memory. The Quarterly Journal of Experimental Psychology, 38A, 603-618. 202 Baddeley, A. D., & Logie, R. H. (1999). Working memory. The multi-component model. In A. Miyake & P. Shah (Eds.), Models of working memory. Mechanisms of active maintenance and executive control (pp. 28-61). Cambridge: Cambridge University Press. Baldwin, C. M., Griffith, K. A., Nieto, J., O'Connor, G. T., Walsleben, J. A., & Redline, S. (2001). The association of sleep-disodered breathing and sleep symptoms with quality of life in the sleep heart health study. Sleep, 24(96-105). Ballester, E., Badia, J. R., Hernandez, L., Carrasco, E., Pablo, J., Fornas, C, et al. (1999). Evidence of the effectiveness of continuous positive airway pressure in the treatment of sleep apnea/hypopnea syndrome. American Journal of Respiratory and Critical Care Medicine, 159, 495-501. Barbe, F., Pericas, J., Munoz, A., Findley, L., Anto, M. J., Agusti, A. G. N., et al. (1998). Automobile accidents in patients with sleep apnea syndrome. American Journal of Respiratory Critical Care Medicine, 158, 18-22. Barcelo, A., Barbe, F., de la Peha, M., Vila, M., Perez, G., Pierola, J., et al. (2006). Antioxidant status in patients with sleep apnoea and impact of continuous positive airway pressure treatment. European Respiratory Journal, 27, 756-760. Bardwell, W. A., Ancoli-Israel, S., & Berry, C. C. (2001). Neuropsychological effects of one-week continuous positive airway pressure treatment in patients with obstructive sleep apnea: A placebo-controlled study. Psychosomatic Medicine, 63(4), 579-584. Bardwell, W. A., Ancoli-Israel, S., & Dimsdale, J. E. (2006). Comparison of the effects of depressive symptoms and apnea severity on fatigue in patients with obstructive sleep apnea: A replication study. Journal of Affective Disorders. Bardwell, W. A., Berry, C. C, Ancoli-Israel, S., & Dimsdale, J. E. (1999). Psychological correlates of sleep apnea. Journal of Psychosomatic Research, 47(6), 583-596. Bardwell, W. A., Moore, P., Ancoli-Israel, S., & Dimsdale, J. E. (2000). Does obstructive sleep apnea confound sleep architechture findings in subjects with depressive symptoms? Biological Psychiatry, 48(\0), 1001-1009. Bardwell, W. A., Moore, P., Ancoli-Israel, S., & Dimsdale, J. E. (2003). Fatigue in obstructive sleep apnea: Driven by depressive symptoms instead of apnea severity? American journal of Psychiatry, 160(2), 350-355. 203 Barnes, M., Houston, D., Worsnop, C. J., Neill, A. M., Mykytyn, I. J., Kay, A., et al. (2002). A randomized controlled trial of continuous positive airway pressure in mild obstructive sleep apnea. American Journal of Respiratory Critical Care Medicine, 165, 773-780. Bartlett, D. J., Rae, C, Thompson, C. H., Byth, K., Joffe, D. A., Enright, T., et al. (2004). Hippocampal area metabolites relate to severity and cognitive function in obstructive sleep apnea. Sleep Medicine, 5, 593-596. Beck, A. T. (1987). Beck depression inventory. San Antonio, TX: The Psychological Corporation. Becker, H. F., Jerrentrup, A., Ploch, T., Grote, L., Penzel, T., Sullivan, C. E., et al. (2003). Effect of nasal continuous positive airway pressure treatment on blood pressure in patients with obstructive sleep apnea. Circulation, 107, 68-73. Becker, J. T., Mintun, M. A., Diehl, D. J., Dobkin, J., Martidis, A., Madoff, D. C, et al. (1994). Functional neuroanatomy of verbal free recall: A replication study. Human Brain Mapping, 1(4), 284-292. Bedard, M. A., Montplaisir, J., Malo, J., Richer, F., & Rouleau, I. (1993). Persistent neuropsychological deficits and vigilance impairments in sleep apnea syndrome after treatment with continuous positive airway pressure (CPAP). Journal of Clinical and Experimental Neuropsychology, 15, 330-341. Bedard, M. A., Montplaisir, J., Richer, F., & Malo, J. (1991a). Nocturnal hypoxemia as a determinant of vigilance impairment in sleep apnea syndrome. Chest, 11(2), 367- 370. Bedard, M. A., Montplaisir, J., Richer, F., Rouleau, I., & Malo, J. (1991b). Obstructive sleep apnea syndrome: Pathogenesis of neuropsychological deficits. Journal of Clinical and Experimental Neuropsychology, 13(6), 950-964. Beebe, D. W. (2005). Neurobehavioral effects of obstructive sleep apnea: An overview and heuristic model. Current Opinion in Pulmonary Medicine, 11, 494-500. Beebe, D. W., & Gozal, D. (2002). 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, 11, 1-16. Beebe, D. W., Groesz, L., Wells, C., Nichols, A., & McGee, K. (2003). The neuropsychological effects of obstructive sleep apnea: A meta-analysis of norm- referenced and case-controlled data. Sleep, 26(3), 298-307. 204 Bennett, L. S., Barbour, C, Langford, B., Stradling, J. R., & Davies, R. J. (1999). Health status in obstructive sleep apnea: Relationship with sleep fragmentation and daytine sleepiness, and effects of continuous positive airway pressure treatment. American Journal of Respiratory and Critical Care Medicine, 159(6), 1884-1890. Bixler, E. O., Vgontzas, A. N., Lin, H.-M., Calhoun, S. L., Vela-Bueno, A., & Kales, A. (2005). Excessive daytime sleepiness in a general population sample: The role of sleep apnea, age obesity, diabetes, and depression. Journal of Clinical Endocrinology & Metabolism, 90(8), 4510-4515. Bliwise, D. (2002). Sleep apnea, apoe4 and Alzheimer"s disease 20 years and counting? Journal of Psychosomatic Research, 55(1), 539-546. Bliwise, D. L., Yesavage, J. A., Sink, J., Widrow, L., & Dement, W. C. (1986). Depressive symptoms and impaired respiration in sleep. Journal of Consulting and Clinical Psychology., 54(5), 734-735. Blunden, S. L., & Beebe, D. W. (2006). The contribution of intermittent hypoxia, sleep debt and sleep disruption to daytime performance deficits in children: Consideration of respiratory and non-respiratory sleep disorders. Sleep Medicine Reviews, 10(2), 109-118. Bolitschek, J., Schmeiser-Rieder, A., Schobersberger, R., Rosenberger, A., Kunze, M., & Aigner, K. (1998). Impact of nasal continuous positive airway pressure treatment on quality of life in patients with obstructive sleep apnoea. European Respiratory Journal, 11(A), 890-894. Bonnet, M. H. (1993). Vigilance/sleepiness sequelae of sleep disordered breathing and sleep apnea: Cognitive effects of sleep and sleep fragmentation. Sleep, 16, S65- S67. Braun, A. R., Balkin, T. J., Wesensten, N. J., Gwadry, F., Carson, R. E., Varga, M., et al. (1998). Dissociated pattern of activity in visual cortices and their projections during human rapid eye movement sleep. Science., 279(5341), 91-95. Braun, A. R., Balkin, T. J., Wesenten, N. J., Carson, R. E., Varga, M., Baldwin, P., et al. (1997). Regional cerebral blood flow throughout the sleep-wake cycle. An H2(15)0 PET study. Brain Research, 120(Pt 7), 1173-1197. Broadbent, D. E., Cooper, P. F., FitzGerald, P., & Parkes, K. R. (1982). The cognitive failures questionnaire (cfq) and its correlates. British Journal of Clinical Psychology, 21, 1-16. 205 Buysse, D. J., Reynolds III, C. R, Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The Pittsburge Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Research, 28, 193-213. Camus, J.-F., Baque, & Fleury, B. (1999). Evaluation of attentional processing impairment in obstructive sleep apnea syndrome by the means of a dichotic listening task. Current Psychology of Cognition, 18(1). Carrasco, E., Santamaria, J., Iranzo, A., Pintor, L., de Pablo, J., Solanas, A., et al. (2006). Changes in dreaming induced by CPAP in severe obstructive sleep apnea syndrome patients. Journal of Sleep Research, 15, 430-436. Cassel, W. (1993). Cognitive effects and daytime sleepiness (a) psychosocial sequelae of sleep disordered breathing: Sleep apnea and personality. Sleep, 16 (suppl), S56- S58. Cassel, W., Ploch, T., Becker, C, Dungnus, D., Peter, J. H., & Wichert, v. (1996). Risk of traffic accidents in patients with sleep-disordered breathing: Reduction with nasal CPAP. European Respiratory Journal, 9, 2606-2611. Cheshire, K., Engleman, H. M., Deary, I., Shapiro, C, & Douglas, N. J. (1992). Factors impairing daytime performance in patients with sleep apnea/hypopnea syndrome. Archives of Internal Medicine, 152, 538-541. Clarke, M., & Oxman, A. D. (Eds.). (2000). Cochrane reviewers' handbook 4.1. [updated June 2000]. In: Review Manager (RevMan) [Computer program]. Version 4.1. Oxford, England: The Cochrane Collaboration. Collop, N., Hartenbaum, N., Rosen, I., & Phillips, B. (2006). Paying attention to at-risk commercial vehicle operators. Chest, 130, 637-639. Coltheart, M. (1981a). Mrc psycholinguistic database user manual: Version 1., from http://www.psych.rl.ac.uk/User_Manual_vl_0.html Coltheart, M. (1981b). The mrc psycholinguistic database. Quarterly Journal of Experimental Psychology, 33(A), 497-505. Conway, A. R. A., Kane, M. J., Bunting, M. F., Hambrick, D. Z., & Wilhelm, R. W. (2005). Working memory span tasks: A methodological review and user's guide. Psychonomic Bulletin & Review, 12, 769-786. 206 Cote, K. A., Milner, C. E., Osip, S. L., Ray, L. B., & Baxler, K. D. (2003). Waking quantitative electroencephalogram and auditory event-related potentials following experimentally indcued sleep fragmentation. Sleep, 26(6), 687-694. D'Ambrosio, C, Bowman, T., & Mohsenin, V. (1999). Quality of life in patients with obstructive sleep apnea: Effect of nasal continuous positive airway pressure—a prospective study. Chest, 115(1), 123-129. D'Esposito, M., Aguirre, G. K., Zarahn, E., Ballard, D., Shin, R. K., & Lease, J. (1998). Functional MRI studies of spatial and nonspaital working memory. Cognitive Brain Research, 7, 1-13. D'Esposito, M., Postle, R. B., Ballard, D., & Lease, J. (1999). Maintenance versus manipulation of information held in working memory: An event-related fmri study. Brain and Cognition, 41, 66-86. Dahl, R. E. (1996). The regulation of sleep and arousal: Development and psychopathology. Development and Psychopathology, 8, 3-27. Dahlof, P., Ejnell, H., Hallstrom, T., & Hedner, J. (2000). Surgical treatment of the sleep apnea syndrome reduces associated major depression. International Journal of Behavioral Medicine, 7(1), 73-88. Dalrymple-Alford, J. C, Kalders, A. S., Jones, R. D., & Watson, R. W. (1994). A central executive deficit in patients with parkinson's disease. Journal of Neurology, Neurosurgery and Psychiatry, 57, 360-367. Daneman, M., & Carpenter, P. A. (1980). Individual differences in working memory and reading. Journal of Verbal Learning and Verbal Behavior, 19, 450-466. Daneman, M., & Merikle, P. M. (1996). Working memory and language comprehension: A meta-analysis. Psychonomic Bulletin & Review, 3(A), 422-433. Davies, C. W. H., Crosby, J. H., Muffins, R. L., Traill, Z. C, Anslow, P., Davis, R. J. O., et al. (2001). Case control study of cerebrovascular damage defined by magnetic resonance imaging in patients with OSA and normal matched control subjects. Sleep, 24(6), 715-720. de Boer, A. G. E. M., van Lanschot, J. J. B., Stalmeier, P. R. M., van Sandick, J. W., Hulscher, J. B. F., de Haes, J. C. J. M., et al. (2004). Is a single-item visual analogue scale as valid, reliable and responsive as multi-item scales in measuring quality of life? Quality of Life Research, 13, 311-320. 207 Dealberto, M.-J., Pajot, N., Courbon, D., & Alperovitch, A. (1996). Breathing disorders during sleep and cognitive performance in an older community sample: The EVA study. Journal of American Geriatric Society, 44, 1287-1294. Decary, A., Rouleau, I., & Montplaisir, J. (2000). Cognitive deficits associated with sleep apnea syndrome: A proposed neuropsychological test battery. Sleep, 23(3), 369- 381. Delia Sala, S., Baddeley, A., Papagno, C, & Spinnler, H. (1995). Dual-task paradigm: A means to examine the central executive. In J. Grafman, K. J. Holyoak & F. Boiler (Eds.), Structure and functions of the human frontal cortex (pp. 161-171). New York: Annals of the New York Academy of Sciences. Derogatis, L. R., Lopman, R. S., & Covi, L. (1973). Scl-90: An outpatient psychiatric rating scale preoliminary report. Psychophamacology, 9, 13-28. Dinges, D. F., Pack, F., Williams, K., Gillen, K. A., Powell, J. W., Ott, G. E., et al. (1997). Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4-5 hours per night. Sleep, 20(4), 267-277. Dinges, D. F., & Powell, J. W. (1985). Microcomputer analyses of performance on a portable, simple visual rt task during sustained operations. Behavior Research Methods, Instruments, & Computers, 17(6), 652-655. Dorsey, C. M., Moore, C. M., McCarley, R. W., Brown, R., Epstein, L. J., Tartarini, W. L., et al. (2000). Spectroscopy before and after total sleep deprivation in healthy adult men. Sleep, 23(Suppl. 2), A357-A358. Douglas, N. J. (1998). Systematic review of the efficacy of CPAP. Thorax, 53, 414-415. Douglas, N. J., & Polo, O. (1994). Pathogenesis of sleep apnea/hypopnoea syndrome. Lancet, 344, 653-655. Douglass, A. B., Bornstein, R., Nino-Muscia, G., Keenan, S., Miles, L., Zarcone Jr., V. P., et al. (1994a). The sleep disorders questionnaire i: Creation and multivariate structure of sdq. Sleep Research, 17, 160-167. Douglass, A. B., Bornstein, R., Nino-Muscia, G., Keenan, S., Miles, L., Zarcone Jr., V. P., et al. (1994b). Test-retest reliability of the seep disorders questionnaire (sdq). Sleep Research, 19,215. 208 Drummond, S. A., Brown, G. G., Strieker, J. L., Buxton, R. B., Wong, E. C, & Gillin, J. C. (1999). Sleep deprivation-induced reduction in cortical functional response to serial subtraction. NeuroReport, 10(18), 3745-3748. Drummond, S. P. A., Brown, G. G., Salamat, J. S., & Gillin, C. (2004). Increasing task difficulty facilitates the cerebral compensatory response to total sleep deprivation. Sleep, 27(3), 445-451. Duff, S. C, & Logie, R. H. (2001). Processing and storage in working memory span. The Quarterly Journal of Experimental Psychology, 54A(\), 31-48. Duncan, J., & Owen, A. M. (2000). Common regions of the human frontal lobe recruited by diverse cognitive demands. Trends Neuroscience, 23, 475-483. El Solh, A. A., Saliba, R., Bosinski, T., Grant, B. J. B., Berbary, E., & Miller, N. (2006). Allopurinol improves endothelial function in sleep apnoea: A randomized controlled study. European Respiratory Journal, 27, 997-1002. Engleman, H. M., & Douglas, N. J. (2004). Sleep • 4: Sleepiness, cognitive function, and quality of life in obstructive sleep apnoea/hypopnoea syndrome. Thorax, 59, 618- 622. Engleman, H. M., Kingshott, R. N., Martin, S. E., & Douglas, N. J. (2000). Cognitive function in the sleep apnea/hypopnea syndrome (SAHS). Sleep, 23(4), S102-108. Engleman, H. M., Kingshott, R. N., Wraith, P. K., Mackay, T. W., Deary, I. J., & Douglas, N. J. (1999). Randomized placebo-controlled crossover trial of continuous positive airway pressure for mild sleep apnea/hypopnea syndrome. American Journal of Respiratory Critical Care Medicine, 159, 461-467. Engleman, H. M., Martin, S. E., Deary, I. J., & Douglas, N. J. (1994). Effect of continuous positive airway pressure treatment on daytime function in sleep apnoea/hypopnea syndrome. Lancet, 343, 572-575. Engleman, H. M., Martin, S. E., Deary, I. J., & Douglas, N. J. (1997). Effect of CPAP therapy on daytime function in patients with mild sleep apnoea/hypopnoea syndrome. Thorax, 52, 114-119. Faccenda, J. F., Mackay, T. W., Boon, N. A., & Douglas, N. J. (2001). Randomized placebo controlled trial of continuous positive airway pressure treatment on blood pressure in the sleep apnea-hypopnea syndrome. American Journal of Respiratory and Critical Care Medicine, 163, 344-348. 209 Feldman Barrett, L., Tugade, M. M., & Engle, R. W. (2004). Individual differences in working memory capacity and dual-process theories of the mind. Psychological Bulletin, 130(4), 553-573. Felver-Gant, J. C, Bruce, A. S., Zimmerman, M., Sweet, L. H., Millman, R. P., & Aloia, M. S. (2007). Working memory in obstructive sleep apnea: Construct validity and treatment effects. Journal of Clinical Sleep Medicine, 3(6), 589-594. Ferini-Strambi, L., Baietto, C, Di Giolia, M. R., Castaldi, P., Castronovo, C, Zucconi, M., et al. (2003). Cognitive dysfunction in patients with obstructive sleep apnea (OSA): Partial reversibility after continuous positive airway pressure (CPAP). Brain Research Bulletin, 61, 87-92. Ferraro, F. R., & Balota, D. A. (1999). Memory scanning performance in healthy young adults, healthy older adults, and individuals with dementia of the Alzheimer type. Aging, Neuropsychology, and Cognition, 6(4), 260-272. Feuerstein, C, Naegele, B., Pepin, J.-L., & Levy, P. (1997). Frontal lobe-related cognitive functions in patients with sleep apnea syndrome before and after treatment. Acta Neurologica Beligica, 97, 96-107. Ficker, J. H., Feistel, H., Moller, C, Merkl, M., Dertinger, S., Siegfried, W., et al. (1997). Changes in regional ens perfusion in obstructive sleep apnea syndrome: Initial spect studies with injected nocturnal 99mtc-hmpao]. Pneumologie, 51(9), 926-930. Findley, L. J., Barth, J. T., Powers, D. C, Wilhoit, S. C, Boyd, D. G., & Surratt, P. M. (1986). Cognitive impairment in patients with obstructive sleep apnea and associated hypoxemia. Chest, 90(5), 686-690. Findley, L. J., Presty, S. K., Barth, J. T., & Suratt, P. M. (1991). Impaired cognition and vigilance in elderly subjects with sleep apnea. In S. T. Kuna, P. M. Suratt & J. E. Remmers (Eds.), Sleep and respiration in aging adults (pp. 259-266). New York: Elsevier. Finelli, L. A., Landolt, H. P., Buck, A., Roth, C, Berthold, T., Borbely, A. A., et al. (2000). Functional neuroanatomy of human sleep states after Zolpidem and placebo: A H2150-PET study. Journal of Sleep Research, 9(2), 161-173. Finn, L., Young, T., Palta, M., & Fryback, D. G. (1998). Sleep-disordered breathing and self-reported general health status in the Wisconsin sleep cohort study. Sleep, 21(1), 701-706. 210 Flemons, W. W. (2004). Measuring quality of life in patients with sleep apnoea: Whose life is it anyway? Thorax, 59, 457-458. Flemons, W. W., & Reimer, M. A. (1998). Deveopment of a disease-specific health related quality of life questionnaire for sleep apnea. American Journal of Respiratory Critical Care Medicine, 158, 494-503. Flemons, W. W., & Reimer, M. A. (2002). Measurement properties of the calgary sleep apnea quality of life index. American Journal of Respiratory and Critical Care Medicine, 165(2), 159-164. Foster, G. E., Hanly, P. J., Ostrowski, M., & Poulin, M. J. (2007). Effects of CPAP on cerebral vascular response to hypoxia in obstructive sleep apnea patients. American Journal of Respiratory and Critical Care Medicine, 175, 720-725. Foster, J. K., Eskes, G. A., & Stuss, D. T. (1994). The cognitive neuropsychology of attention: A frontal lobe perspective. Cognitive Neuropsychology, 11(2), 133-147. Fulda, S., & Schulz, H. (2003). Cognitive dysfunction in sleep-related breathing disorders: A meta-analysis. Sleep Research Online, 5(1), 19-51. Gale, S. D., & Hopkins, R. O. (2004). Effects of hypoxia on the brain: Neuroimaging and neuropsychological findings following carbon monoxide poisoning and obstructive sleep apnea. Journal of the International Neuropsychological Society, 10, 60-71. Gall, R., Isaac, L., & Kryger, M. (1993). Quality of life in mild obstructive sleep apnea. Sleep, i6(8), S59-61. Gevins, A., Smith, M. E., Le, J., Leong, H., Bennett, J., Martin, N., et al. (1996). High resolution evoked potential imaging of the cortical dynamics of human working memory. Electroencephalography and Clinical Neurophysiology, 98, 327-348. Gevins, A. S., & Cutillo, B. A. (1993). Spatiotemporal dynamics of component processes in human working memory. Electroencephalography and Clinical Neurophysiology,^!), 128-143. Goldberg, T. E., Berman, K. F., Randolph, C., Gold, J. M., & Weinberger, D. R. (1996). Isolating the mnemonic component in spatial delayed response: A controlled pet 15o-labeled water regional cerebral blood flow study in normal humans. Neurolmage, 3, 69-78. 211 Gottlieb, D. J., De Stephano, A. L., Foley, D. J., Mignot, E., Redline, S., Givelber, R. J., et al. (2004). Apoe U4 is associated with obstructive sleep apnea/hypopnea: The sleep heart health study. Neurology, 63, 664-668. Gottlieb, D. J., Whitney, C. W., Bonekat, W., H., Iber, C, James, G., Lebowitz, M., et al. (1999). Relation of sleepiness to respiratory disturbance index: The sleep heart health study. American Journal of Respiratory and Critical Care Medicine, 159, 502-507. Gozal, D., Wang, M., & Pope, D. W. J. (2001). Objective sleepiness measures in pediatric obstructive sleep apnea. Pediatrics, 108(3), 693-697. Greenberg, G. D., Watson, R. K., & Deptula, D. (1987). Neuropsychological dysfunction in sleep apnea. Sleep, 10(3), 254-262. Greene, J. D. W., Hodges, J. R., & Baddeley, A. D. (1995). Autobiographical memory and executive function in early dementia of Alzheimer type. Neuropsychologia, 33(12), 1647-1670. Grunstein, R. (2005). Continuous positive airway pressure treatment for obstructive sleep apnea-hypopnea syndrome. In M. H. Kryger, T. Roth & W. C. Dement (Eds.), Principles and practice of sleep medicine (4th ed., pp. 1066-1080). Philadelphia: Elsevier Saunders. Guilleminault, C., & Bassiri, A. G. (2005). Clinical features and evaluation of obstructive sleep apnea-hypopnea syndrome and the upper airway resistance syndrome. In M.H. Kryger, T. Roth & W. C. Dement (Eds.), Principles and practice of sleep medicine (4th ed., pp. 1043-1052). Philadelphia: Elsevier Saunders. Guilleminault, C, & Dement, W. C. (1977). 235 cases of excessive daytime sleepiness. Diagnosis and tentative classification. Journal of Neurological Sciences, 31(1), 13-27. Hack, M., Davies, R. J. O., Muffins, R., Choi, S. J., Ramdassingh-Dow, S., Jenkinson, C, et al. (2000). Randomized prospective parallel trial of therapeutic versus subtherapeutic nasal continuous positive airway pressure on simulated steering performance in patients with obstructive sleep apnoea. Thorax(55), 224-231. Harbison, J., Gibson, G. J., Birchall, D., Zammit-Maempel, I., & Ford, G. A. (2003). White matter disease and sleep-disordered breathing after acute stroke. Neurology, 61, 959-963. 212 Hardinge, F. M., Pitson, D. J., & Stradling, J. R. (1995). Use of the Epworth Sleepiness Scale to demonstrate response to treatment with nasal continuous positive airways pressure in patients with obstructive sleep apnoea. Respiratory Medicine, 89, 617- 620. Harrison, Y., & Home, J. A. (1999). One night of sleep loss impairs innovative thinking and flexible decision making. Organizational Behavior and Human Decision Processes, 78(2), 128-145. Harrison, Y., & Home, J. A. (2000a). The impact of sleep deprivation on decision making: A review. Journal of Experimental Psychology: Applied, 6(3), 236-249. Harrison, Y., & Home, J. A. (2000b). Sleep loss and temporal memory. The Quarterly Journal of Experimental Psychology, 53A(l), 271-279. Hartenbaum, N., Collop, N., Rosen, I. M., Phillips, B., George, C. F., Rowley, J. A., et al. (2006). Sleep apnea and commercial motor vehicle operators: Statement from the joint task force of the american college of chest physicians, the american college of occupational and environmental medicine, and the national sleep foundation. Chest, 130, 902-905. He, J., Kryger, M. H., Zorick, F. J., Conway, W., & Roth, T. (1988). Mortality and apnea index in obstructive sleep apnea. Experience in 385 male patients. Chest, 94, 9-14. Henke, K. G., Grady, J. J., & Kuna, S. T. (2001). Effect of nasal continuous positive airway pressure on neuropsychological function in sleep apnea-hypopnea syndrome. A randomized, placebo-controlled trial. American Journal of Respiratory and Critical Care Medicine, 163(4), 911-917. Hiestand, D. M., Britz, P., Goldman, M., & Phillips, B. (2006). Prevalence of symptoms and risk of sleep apnea in the us population: Results from the National Sleep Foundation Sleep in America 2005 Poll. Chest, 130, 780-786. Hobson, J. A., Stickgold, R., & Pace-Schott, E. F. (1998). The neuropsychology of REM sleep dreaming. Neuroreport., 9(3), Rl-14. Hoddes, E., Zarcone, V., Smythe, H., Phillips, R., & Dement, W. C. (1973). Quantification of sleepiness: A new approach. Psychophysiology, 10, 431-436. Home, J. A. (1988). Sleep loss and "divergent" thinking ability. Sleep, 11(6), 528-536. 213 Hu, J., Chen, Z., Gorczynski, C. P., Gorczynski, L. Y., Kai, Y., Lee, L., et al. (2003). Sleep-deprived mice show altered cytokine production manifest by perturbations in serum il-lra, tnfa, and il-6 levels. Brain, Behavior, and Immunity, 77(6), 498- 504. Hudgel, D. W. (1989). Neuropsychiatry manifestations of obstructive sleep apnea: A review. International Journal of Psychiatry in Medicine, 19(1), 11-22. Ingram, F., Henke, K. G., Levin, H. S., Fishel Ingram, P. T., & Kuna, S. T. (1994). Sleep apnea and vigilance performance in a community-dwelling older sample. Sleep, 77(3), 248-252. Jenkinson, C., Davies, R. J., Mullins, R., & Stradling, J. R. (1999). Comparison of therapeutic and sub therapeutic nasal continuous positive airway pressure for obstructive sleep apnea: A randomized prospective parallel trial. Lancet, 353, 2100-2105. Johns, M. W. (1991). A new method for measuring daytime sleepiness: The Epworth Sleepiness Scale. Sleep Research, 14, 540-545. Johns, M. W. (1992). Reliability and factor analysis of the Epworth Sleepiness Scale. Sleep Research, 75(4), 376-381. Johns, M. W. (1993). Daytime sleepiness, snoring, and obstructive sleep apnea. The Epworth Sleepiness Scale. Chest, 103, 30-36. Johns, M. W. (2000). Sensitivity and specificity of the multiple sleep latency test (mslt), the maintenance of wakefulness test and the epworth sleepiness scale: Failure of the mslt as a gold standard. Journal of Sleep Research, 9(1), 5-11. Jones, K., & Harrison, Y. (2001). Frontal lobe function, sleep loss and fragmented sleep. Sleep Medicine Reviews, 5(6), 463-475. Kadotani, H., Kadotani, T., Young, T., Peppard, P. E., Finn, L., Colrain, I. M., et al. (2001). Association between apolipoprotein e D4 and sleep-disordered breathing in adults. Journal of American Medical Association, 285(22), 2888-2890. Kales, A., Cadieux, R. J., Bixler, E. O., Soldatos, C. R., Vela-Bueno, A., Misoul, C. A., et al. (1985). Severe obstructive sleep apnea—1: Onset, clinical course, and characteristics. Journal of Chronic Diseases, 38(5), 419-425. 214 Kamba, M., Inoue, Y., Higami, S., Suto, Y., Ogawa, T„ & Chen, W. (2001). Cerebral metoabolic impairment in patients with obstructive sleep apnoea: An independent association of obstructive sleep apnoea with white matter change. Journal of Neurology Neurosurgery Psychiatry, 71, 334-339. Kamba, M., Suto, Y., Inoue, Y., & Matsuda, E. (1997). Cerebral metabolism in sleep apnea. Evaluation by magnetic resonance spectroscopy. American Journal of Respiratory Critical Care Medicine, 156, 296-298. Kanaan, A., Farahani, R., Douglas, R. M., LaManna, J. C, & Haddad, G. G. (2006). Effect of chronic continuous or intermittent hypoxia and reoxygenation on cerebral capillary density and myelination. American Journal of Physiology - Regulatory Integrative and Comparative Physiology, 290, Rl 105-R1114. Kaneko, Y., Floras, J. S., Usui, K., Plante, J., Tkacova, R., Kubo, T., et al. (2003). Cardiovascular effects of continuous positive airway pressure in patients with heart failure and obstructive sleep apnea. The New England Journal of Medicine, 348, 1233-1241. Kawahara, S., Akashiba, T., Akahoshi, T., & Horie, T. (2005). Nasal CPAP improves the quality of life and lessens the depressive symptoms in patients with obstructive sleep apnea syndrome. Internal Medicine, 44(5), ATI-All. Kheirandish, L., Gozal, D., Pequignot, J.-M., Pequignot, J., & Row, B. W. (2005). Intermittent hypoxia during development induces long-term alterations in spatial working memory, monoamines, and dendritic branching in rat frontal cortex. Pediatric Research, 58, 594-599. Kim, H. C, Young, T., Matthews, C. G., Weber, S. M., Woodard, A. R., & Palta, M. (1997). Sleep-disordered breathing and neuropsychological deficits. American Journal of Respiratory Critical Care Medicine, 156, 1813-1819. Kingshott, R. N., Vennelle, M., Coleman, E. L., Engleman, H. M., Mackay, T. W., & Douglas, N. J. (2001). Randomized, double-blind, placebo-controlled crossover trial of modafinil in the treatment of residual excessive daytime sleepiness in the sleep apnea/hypopnea syndrome. American Journal of Respiratory Critical Care Medicine, 163, 918-923. Kingshott, R. N., Vennelle, M., Hoy, C. J., Engleman, H. M., Deary, I. J., & Douglas, N. J. (2000). Predictors of improvements in daytime function outcomes with CPAP therapy. American Journal of Respiratory Care Medicine, 161, 866-871. 215 Knight, H., Millman, R. P., Gur, R. C, Saykin, A. J., Doherty, J. U., & Pack, A. I. (1987). Clinical significance of sleep apnea in the elderly. The American Review of Respiratory Disease, 136(4), 845-850. Kobayashi, K., Nishimura, Y., Shimada, T., Yoshimura, S., Funada, Y., Satouchi, M., et al. (2006). Effect of continuous positive airway pressure on soluble cd40 ligand in patients with obstructive sleep apnea syndrome. Chest, 129, 632-637'. Kolb, B., & Whishaw, I. Q. (2003). Fundamentals of human neuropsychology (5th ed.). New York: Worth Publishers. Koseoglu, N., Koseglu, H., Itil, O., Baklan, B., Ikiz, A. O., & Esen, A. A. (2007). Sexual function status in women with obstructive sleep apnea. The Journal of Sexual Medicine, 4(5), 1352-1357. Kotterba, S., Rasche, K., Widding, W., Duscha, C, Blombach, S., Schultze-Werninghaus, G., et al. (1998). Neuropsychological investigations and event-related potentials in obstructive sleep apnea syndrome before and during CPAP-therapy. Journal of the Neurological Sciences, 159, 45-50. Koves, P., Szakacs, Z., Bernath, I., & Molnar, M. (2001). Cognitive deficits in obstructive sleep apnea. Sleep, 24(Abstract Supplement), A287. Kribbs, N. B., Pack, A. I., Kline, L. R., Getsy, J. E., Schuett, J. S., Henry, J. N., et al. (1993). Effects of one night without nasal CPAP treatment on sleep and sleepiness in patients with obstructive sleep apnea. American Review of Respiratory Disorders, 147, 1162-1168. Kuniyoshi, F. H. S., & Somers, V. K. (2006). Sleep apnea in hypertension: When, how, and why should we treat? Hypertension, 47, 818-819. Kushida, C. A., Littner, M., Morgenthaler, T., Alessi, C, Bailey, D., Coleman, J. J., et al. (2005). Practice parameters for the indications for polysomnography and related procedures: An update for 2005. Sleep, 28(A), 299-521. Lacasse, Y., Bureau, M. P., & Series, F. (2004). A new standardized and self- administered quality of life questionnaire specific to obstructive sleep apnoea. Thorax, 59(6), 494-499. Lacasse, Y., Godbout, C, & Series, F. (2002). Independent validation of the sleep apnoea quality of life index. Thorax, 57, 483-488. 216 Larkin, E. K., Rosen, C. L., Kirchner, H. L., Storfer-Isser, A., Emancipator, J. L., Johnson, N. L., et al. (2005). Variation of C-reactive protein levels in adolescents: Association with sleep-disordered breathing and sleep duration. Circulation, 111, 1978-1984. Lavie, L. (2003). Obstructive sleep apnoea syndrome: An oxidative stress disorder. Sleep Medicine Reviews, 7(1), 35-51. Lawrence, N. S., Ross, T. J., Hoffmann, R., Garavan, H., & Stein, E. A. (2003). Multiple neuronal networks mediate sustained attention. Journal of Cognitive Neuroscience, 15(7), 1028-1038. Lee, M. M., Strauss, M. E., Adams, N., & Redline, S. (1999). Executive functions in persons with sleep apnea. Sleep and Breathing, 3, 13-16. Lee, S. (1990). Depression in sleep apnea: A different view. Journal of Clinical Psychiatry, 57(7), 309-310. Levine, B., Robertson, I. H., Clare, L., Carter, G., Hong, J., Wilson, B. A., et al. (2000). Rehabilitation of executive functioning: An experimental-clinical validation of goal management training. Journal of the International Neuropsychological Society, 6(3), 299-312. Lezak, M. D. (1995). Neuropsychological assessment (3rd ed.). New York: Plenum Press. Li, R. C, Row, B. W., Kheirandish, L., Brittian, K. R., Gozal, E., Guo, S. Z., et al. (2004). Nitric oxide synthase and intermittent hypoxia-induced spatial learning deficits in the rat. Neurobiology of Disease, 17(1), 44-53. Lis, S., Krieger, S., Hennig, D., Christian, R., Kirsch, P., Seeger, W., et al. (2008). Executive function and cognitive subprocesses in patients with obstructive sleep apnoea. Journal of Sleep Research, electronic publication (ahead of print), 1-10. Logie, R. H. (1995). Visuo-spatial working memory. Hove: Lawrence Erlbaum Associates. Lojander, J., Kajaste, S., Maasilta, P., & Partinen, M. (1999). Cognitive function and treatment of obstructive sleep apnea syndrome. Journal of Sleep Research, 8, 71- 76. Macey, P. M., Henderson, L. A., Macey, K. E., Alger, J. R., Frysinger, R. C, Woo, M. A., et al. (2002). Brain morphology associated obstructive sleep apnea. American Journal of Respiratory Critical Care Medicine, 166, 1382-1387. 217 Margel, D., Cohen, M., Livne, P. M., & Pillar, G. (2004). Severe, but not mild, obstructive sleep apnea syndrome is associated with erectile dysfunction. Urology, 63(3), 545-549. Marin, D. B., Breuer, B., Marin, M. L., Silverman, J., Schmeidler, J., Greenberg, D., et al. (1998). The relationship between apolipoprotein e, dementia, and vascular illness. Atherosclerosis, 140, 173-180. Marin, J. M., Carrizo, S. J., Vicente, E., & Agusti, A. G. N. (2005). Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuous positive airway pressure: An observational study. Lancet, 365, 1046-1053. Marshall, N. S., Barnes, M., Travier, N., Campbell, A. J., Pierce, R. J., McEvoy, R. D., et al. (2006). Continuous positive airway pressure reduces daytime sleepiness in mild to moderate obstructive sleep apnoea: A meta-analysis. Thorax, 61, 430-434. Martin, S. E., Engleman, H. M., Deary, I. J., & Douglas, N. J. (1996). The effect of sleep fragmentation on daytime function. American Journal of Respiratory and Critical Care Medicine, 153, 1328-1332. Mazza, S., Pepin, J.-L., Bernadette, N., Rauch, E., Deschaux, C, Ficheux, P., et al. (2006). Driving ability in sleep apnoea patients before and after CPAP treatment: Evaluation on a road safety platform. European Respiratory Journal, 28, 1020- 1028. McDermott, C. M., LaHoste, G. J., Chen, C, Musto, A., Bazan, N. G., & Magee, J. C. (2003). Sleep deprivation causes behavioral, synaptic, and membrane excitability alterations in hippocampal neurons. The Journal of Neuroscience, 23(29), 9687- 9695. McDowell, S., Whyte, J., & D'Esposito, M. (1997). Working memory impairments in traumatic brain injury: Evidence from a dual-task paradigm. Neuropsychologia, 35(10), 1341-1353. McHorney, C. A., Ware, J. E. J., & Raczek, A. E. (1993). The mos 36-item short form health survey (sf-36):Ii. Psychometric and clinical tests of validity in measuirng physical and mental health constructs. Medical Care, 31, 835-843. McNair, D. M., Lorr, M., & Druppleman, L. F. (1971). Eits manual for the profile of mood states. San Diego: Education and Industrial Test Services. 218 Mediano, O., Barcelo, A., de la Pena, M., Gonzal, D., Agusti, A., & Barbe, F. (2007). Daytime sleepiness and polysomnography variables in sleep apnea patients. European Respiratory Journal, 30, 110-113. Meier-Ewert, H. K., Ridker, P. M., Rifai, N., Regan, M. M., Price, N. J., Dinges, D. F., et al. (2004). Effect of sleep loss on c-reactive protein, an inflammatory marker of cardiovascular risk. Journal of the American College of Cardiology, 43(4), 678- 683. Millman, R. P., Fogel, B. S., McNamara, M. E., & Carlisle, C. C. (1989). Depression as a manifestation of obstructive sleep apnea: Reversal with nasal contrinuous positive airway pressure. Journal of Clinical Psychiatry, 50(9), 348-351. Mills, P. J., & Dimsdale, J. E. (2004). Sleep apnea: A model for studying cytokines, sleep, and sleep disruption. Brain, Behavior, and Immunity, 18, 298-303. Mitler, M. M., Gujavarty, K. S., & Browman, C. P. (1982). Maintenance of wakefulness test: A polysomnographic technique for evaluation treatment efficacy in patients with excessive somnolence. Electroencephalography and Clinical Neurophysiology., 53(6), 658-661. Miyake, A. (2001). Individual differences in working memory: Introduction to the special section. Journal of Experimental Psychology: General, 130(2), 163-168. Monk, T. H., Buysse, D. J., Kennedy, K. S., Potts, J. M., DeGrazia, J. M., & Miewald, J. M. (2002). Measuring sleep habits without using a diary: The Sleep Timing Questionnaire. Sleep, 26(2), 209-211. Montplaisir, J., Bedard, M. A., Richer, F., & Rouleau, I. (1992). Neurobehavioral manifestations in obstructive sleep apnea syndrome before and after treatment with continuous positive airway pressure. Sleep, 15(6 Suppl), S17-19. Moore, P., Bardwell, W. A., & Ancoli-Israel, S. (2001). Association between polysomnographic sleep measures and health-related quality of life in obstructive sleep apnea. Journal of Sleep Research, 10(4), 303-308. Morrell, M. J., McRobbie, D. W., Quest, R. A., Cummin, A. R. C, Ghiassi, R., & Corfield, D. R. (2003). Changes in brain morphology associated with obstructive sleep apnea. Sleep Medicine, 4, 451-454. Mosko, S., Zetin, M., Glen, S., Garber, D., Deantonio, M., Sassin, J., et al. (1989). Self- reported depressive symptomatology, mood ratings, and treatment outcome in sleep disorders patients. Journal of Clinical Psychology, 45(1), 51-60. 219 Moss, I. R., Belisle, M., & Laferriere, A. (2006). Long-term recurrent hypoxia in developing rat attenuates respiratory responses to subsequent acute hypoxia. Pediatric Research, 59, 525-530. Mulgrew, A. T., Ryan, C. F., Fleetham, J. A., Cheema, R., Fox, N., Koehoorn, M., et al. (2007). The impact of obstructive sleep apnea and daytime sleepiness on work limitation. Sleep Medicine, 9(1), 42-53. Miiller, N. G., & Knight, R. T. (2006). The functional neuroanatomy of working memory: Contributions of human brain lesion studies. Neuroscience, 139, 51-58. Myers, R. H., Schaefer, E. J., & Wilson, W. M. (1996). Apolipoprotein e element 4 association with dementia in a population-based study: The farmington study. Neurology, 46, 673-677. Naegele, B., Launois, S. H., Mazza, S., Feuerstein, C, Pepin, J.-L., & Levy, P. (2006). Which memory processes are affected in patients with obstructive sleep apnea? An evaluation of 3 types of memory. Sleep, 29, 533-544. Naegele, B., Pepin, J. L., Levy, P., Bonnet, C, Pellat, J., & Feuerstein, C. (1998). Cognitive executive dysfunction in patients with obstructive sleep apnea syndrome (OSAS) after CPAP treatment. Sleep, 21(4), 392-397. Naegele, B., Thouvard, V., Pepin, J. L., Levy, P., Bonnet, C, Perret, J. E., et al. (1995). Deficits of cognitive executive functions in patients with sleep apnea syndrome. Sleep, 18(1), 43-52. Netzer, N. C, Stoohs, R. A., Netzer, C. M., Clark, K., & Strohl, K. P. (1999). Using berlin questionnaire to identify patients at risk for the sleep apnea syndrome. Annals of Internal Medicine, 131, 485-491. Nieto, F. J., Young, T. B., Lind, B. K., Shahar, E., Samet, J., Redline, S., et al. (2000). Association of sleep disordered breathing, sleep apnes, and hypertension in a large community-based study. Journal of the American Medical Association, 283, 1829-1836. o Nilsson, J. P., Soderstrom, M., Karlsson, A. U., Lekander, M., Akerstedt, T., Lindroth, N. E., et al. (2005). Less effective executive functioning after one night's sleep deprivation. Journal of Sleep Research, 14 (1), 1-6. Norman, D., Loredo, J. S., Nelesen, R. A., Ancoli-Israel, S., Mills, P. J., Ziegler, M. G., et al. (2006). Effects of continuous positive airway pressure versus supplemental oxygen on 24-hour ambulatory blood pressure. Hypertension, 47, 840-845. 220 Nowak, M., Kornhuber, J., & Meyrer, R. (2006). Daytime impairment and neurodegeneration in OSAS. Sleep, 29(12), 1521-1530. Nussbaumer, Y., Bloch, K. E., Genser, T., & Thurnheer, R. (2006). Equivalence of autoadjusted and constant continuous positive airway pressure in home treatment of sleep apnea. Chest, 129, 638-643. O'Donoghue, F. J., Briellmann, R. S., Rochford, P. D., Abbott, D. F., Pell, G. S., Chan, C. H. P., et al. (2005). Cerebral structural changes in severe obstructive sleep apnea. American Journal of Respiratory Critical Care Medicine, 171, 1185-1190. Ohayon, M. M. (2003). The effects of breathing-related sleep disorders on mood disturbances in the general population. Journal of Clinical Psychiatry, 64, 1195- 1200. Owen, A. M., Evans, A. C, & Petrides, M. (1996). Evidence for a two-stage model of spatial working memory processing within the lateral frontal cortex: A positron emission tomography study. Cerebral Cortex, 6(31-38). Owen, A. M., Herrod, N. J., Menon, D. K., Clark, J. C, Downey, S. P. M. J., Carpenter, T. A., et al. (1999). Redefining the functional organization of working memory processes within human lateral prefrontal cortex. European Journal of Neuroscience, 11(2), 567-574. Pack, A. I., Maislin, G., Staley, B., Pack, F. M., Rogers, W. C, George, C. F. P., et al. (2006). Impaired performance in commercial drivers: Role of sleep apnea and short sleep duration. American Journal of Respiratory Critical Care Medicine, 174, 446-454. Partinen, M., & Hublin, C. (2005). Epidemiology of sleep disorders. In M. H. Kryger, T. Roth & W. C. Dement (Eds.), Principles and practice of sleep medicine (4th ed., pp. 626-647). Philadelphia: Elsevier Saunders. Patel, S. R., White, D. P., Malhotra, A., Stanchina, M. L., & Ayas, N. T. (2003). Continuous positive airway pressure therapy for treating sleepiness in a diverse population with obstructive sleep apnea. Archives of Internal Medicine, 163, 565- 571. Payne, R. S., Goldbart, A., Gozal, D., & Schurr, A. (2004). Effect of intermittent hypoxia on long-term potentiation in rat hippocampal slices. Brain Research, 1029, 195- 199. 221 Peker, Y., Hedner, J., Norum, J., Kraiczi, H., & Carlson, J. (2002). Increased incidence of cardiovascular disease in middle-aged men with obstructive sleep apnea: A 7 year follow-up. American Journal of Respiratory and Critical Care Medicine, 166, 159-165. Peppard, P. E., Szklo-Coxe, M., Hla, K. M., & Young, T. (2006). Longitudinal association of sleep-related breathing disorder and depression. Archives of Internal Medicine., 166, 1709-1715. Peppard, P. E., Young, T., Palta, M., & Skatrud, J. (2000). Prospective study of the association between sleep-disordered breathing and hypertension. The New England Journal of Medicine, 342, 1378-1384. Petrides, M., Alivisatos, B., Meyer, E., & Evans, A. C. (1993). Functional activation of the human frontal cortex during the performance of verbal working memory tasks. Proceedings of the National Academy of Science of the United States of America, 90, 878-882. Phillips, B. A., Berry, D. T., & Lipke-Molby, T. C. (1996). Sleep-disordered breathing in healthy, aged persons. Fifth and final year follow-up. Chest, 110(3), 654-658. Phillips, C. L., & Grunstein, R. R. (2006). Obstructive sleep apnea: Time for a radical change? European Respiratory Journal, 27, 671-673. Pillar, G., & Lavie, P. (1998). Psychiatric symptoms in sleep apnea syndrome: Effects of gender and respiratory disturbance index. Chest, 114(3), 697-703. Poceta, J. S., Timms, R. M., Jeong, D. U., Ho, S. L., Erman, M. K., & Mitler, M. M. (1992). Maintenance of wakefulness test in obstructive sleep apnea syndrome. Chest, 101, 893 - 897. Polotsky, V. Y., Rubin, A. E., Balbir, A., Dean, T., Smith, P. L., Schwartz, A. R., et al. (2006). Intermittent hypoxia causes REM sleep deficits and decreases EEG delta power in NREM sleep in the C57BL/6J mouse. Sleep Medicine, 7(1), 7-16. Postle, B. R., Berger, J. S., & D'Esposito, M. (1999). Functional neuroanatomical double dissociation of mnemonic and executive control processes contributing to working memory performance. Proceedings of National Academy of Sciences of the United States of America, 96(22), 12959-12964. Postle, B. R., D'Esposito, M., & Corkin, S. (2005). Effects of verbal and nonverbal interference on spatial and object visual working memory. Memory & Cognition, 33(2), 203-212. 222 Powell, N., Riley, R., Schechtman, K. B., Blumen, M., Dinges, D. F., & Guilleminault, C. (1999). A comparative model: Reaction time performance in sleep-disordered breathing versus alcohol-impaired controls. The Laryngoscope, 109(10), 1648- 1654. Prabhakaran, V., Narayanan, K., Zhao, Z., & Gabrieli, J. D. (2000). Integration of diverse information in working memory within the frontal lobe. Nature Neuroscience, 3(1), 85-90. Prigatano, G. P. (1983). Neuropsychological test performance in mildly hypoxemic patients with chronic obstructive pulmonary disease. Rechtschaffen, A., & Kales, A. (1968). A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Los Angeles: UCLA Brain Information Service/Brain Research Institute. Richardson, G. S., Carskadon, M. A., Flagg, W., Van den Hoed, J., Dement, W. C, & Mitler, M. M. (1978). Excessive daytime sleepiness in man: Multiple sleep latency measurement in narcoleptic and control subjects. Electroencephalography and Clinical Neurophysiology., 45(5), 621-627. Robbins, J., Redline, S., Ervin, A., Walsleben, J. A., Ding, J., & Nieto, J. (2005). Associations of sleep-disordered breathing and cerebral changes on MRI. Journal of Clinical Sleep Medicine, 1, 159-165. Row, B. W., Liu, R., Xu, W., Kheirandish, L., & Gozal, D. (2003). Intermittent hypoxia is associated with oxidative stress and spatial learning deficits in the rat. American Journal of Respiratory Critical Care Medicine, 167, 1548-1553. Rypma, B., Prabhakaran, V., Desmond, J. E., Glover, G. H., & Gabrieli, J. D. (1999). Load-dependent roles of frontal brain regions in the maintenance of working memory. Neurolmage, 9, 216-226. Salorio, C. F., White, D. A., Piccirillo, J., Duntley, S. P., & Uhles, M. L. (2002). Learning, memory, and executive control in individuals with obstructive sleep apnea syndrome. Journal of Clinical and Experimental Neuropsychology, 24(1), 93-100. Salthouse, T. A., & Hedden, T. (2002). Interpreting reaction time measures in between- group comparisons. Journal of Clinical and Experimental Neuropsychology, 24(1), 858 - 872. 223 Sanchez, A. I., Buela-Casal, G., Bermudez, M. P., & Casas-Maldonado, F. (2001). The effects of continuous positive air pressure treatment on anxiety and depression levels in apnea patients. Psychiatry and Clinical Neuroscience, 55, 641-646. Sangal, R. B., & Sangal, J. M. (1997). Abnormal visual p300 latency in obstructive sleep apnea does not change acutely upon treatment with CPAP. Sleep, 20, 702-704. Sanner, B. M., Klewer, J., Trumm, A., Randerath, W., Kreuzer, I., & Zidek, W. (2000). Long-term treatment with continuous positive airway pressure improves quality of life in obstructive sleep apnoea syndrome. European Respiratory Journal, 16, 118-122. Sanner, B. M., & Tepel, M. (2006). Sleep is more than a break from life. European Heart Journal, 27, 1016-1017. Sateia, M. J. (2003). Neuropsychological impairment and quality of life in obstructive sleep apnea. Clinical Chest Medicine, 24, 249-259. Saunamaki, T., & Jehkonen, M. (2007). A review of executive functions in obstructive sleep apnea syndrome. Acta Neurologica Scandinavica, 115, 1-11. Schneider, C, Fulda, S., & Schulz, H. (2004). Daytime variation in performance and tiredness/sleepiness ratings in patients with insomnia, narcolepsy, sleep apnea and normal controls. Journal of Sleep Research, 13, 373-383. Schwartz, C. E., & Sprangers, M. A. G. (1999). Methodological approaches for assessing response shift in longitudinal health-related quality-of-life research. Social Science & Medicine, 48, 1531-1548. Shah, P., & Miyake, A. (1996). The separability of working memory resources for spatial thinking and language processing: An individual differences approach. Journal of Experimental Psychology: General, 125, 4-27. Shallice, T. (Ed.). (2002). Fractionation of the supervisory system. Oxford: Oxford University Press. Shivalkar, B., Van De Heyning, C, Kerremans, M., Rinkevich, D., Verbraecken, J., De Backer, W., et al. (2006). Obstructive sleep apnea syndrome: More insights on structural and functional cardiac alterations, and the effects of treatment with continuous positive airway pressure. Journal of American College of Cardiology, 47, 1433-1439. 224 Sloan, K., Craft, S., & Walsh, J. K. (1986). Neuropsychology function in obstructive sleep apnea with and without hypoxemia. Sleep Research, 18, 304. Sloan, K., Craft, S., & Walsh, J. K. (1989). Neuropsychology function in obstructive sleep apnea with and without hypoxemia. Sleep Research, 18, 304. Smith, E. E., & Jonides, J. (1997). Working memory: A view from neuroimaging. Cognitive Psychology, 33, 5-42. Smith, E. E., Jonides, J., & Koeppe, R. A. (1996). Dissociating verbal and spatial working memory using pet. Cerebral Cortex, 6, 11-20. Smith, E. E., Jonides, J., Koeppe, R. A., Awh, E., Schumacher, E. H., & Minoshima, S. (1995). Spatial versus object working memory: Pet investigations. Journal of Cognitive Neuroscience, 7, 337-356. Smith, E. E., Jonides, J., Marshuetz, C, & Koeppe, R. A. (1998). Components of verbal working memory: Evidence from neuroimaging. Proceedings of the National Academy of Sciences of the United States of America, 95, 870-882. Smith, S., & Sullivan, K. A. (2007). A reliable change index (rci) for the epworth sleepiness scale (ess). Sleep Medicine, 9(1), 102. Sternberg, S. (1966). High-speed scanning in human memory. Science, 153, 652-654. Sternberg, S. (1969). Memory-scanning: Mental processes revealed by reaction-time experiments. American Scientist, 57(4), 421-457. Stone, J., Morin, C. M., Hart, R. P., Remsberg, S., & Mercer, J. (1994). Neuropsychological functioning in older insomniacs with or without obstructive sleep apnea. Psychology and Aging, 2, 231-236. Strittmatter, W. J., Saunders, A. M., Schmechel, D., Vance, M. P., Enghild, J., Salvensen, G. S., et al. (1993). Apolipoprotein e: High-avidity binding to b-amyloid and inceased frequency of type 4 allele in late-onset familial Alzheimer disease. Proceedings of the National Academy of Sciences of the United States of America, 90, 1977-1981. Sturm, W., Willmes, K., Orgass, B., & Hartje, W. (1997). Do specific attention deficits need specific training? Neuropsychological Rehabilitation, 7(2), 81-103. 225 Stuss, D. T., Shallice, T., Alexander, M. P., & Picton, T. W. (1995). A multidisciplinary approach to anterior attentional functions. Structure and functions of the human prefrontal cortex. The Annals of the New York Academy of Sciences, 769, 191-211. Svatikova, A., Wolk, R., Garni, A. S., Pohanka, M., & Somers, V. K. (2005). Interactions between obstructive sleep apnea and the metabolic syndrome. Current Diabetes Reports, 5(1), 53-58. Tarasiuk, A., Greenberg-Dotan, S., Simon, T., Tal, A., Oksenberg, A., & Reuveni, H. (2006). Low socioeconomic status is a risk factor for cardiovascular disease among adult obstructive sleep apnea syndrome patients requiring treatment. Chest, 130, 766-773. Teran-Santos, J., Jimenez-Gomez, A., Cordero-Guevara, J., & Cooperative Group Burgos-Santander. (1999). The association between sleep apnea and the risk of traffic accidents. New England Journal of Medicine, 340(11), 847-851. Thomas, M., Sing, H., Belenky, G., Holcomb, H., Mayberg, H., Dannals, R., et al. (2000). Neural basis of alertness and cognitive performance impairments during sleepiness. I. Effects of 24 h of sleep deprivation on waking human regional brain activity. Journal of Sleep Research, 9(4), 335-352. Thomas, M. L., Balkin, T., Sing, H., Wesensten, N., & Belenky, G. (1998). Pet imaging studies of sleep deprivation and sleep: Implications for behavior and sleep function. Journal of Sleep Research, 7(Suppl. 2), 274. Thomas, M. L., Sing, H. C., Belenky, G., Holcomb, H. H., Dannals, R. R, Wagner, H. N., et al. (1993). Cerebral glucose utilization during task performance and prolonged sleep loss. Journal of Cerebral Blood Flow and Metabolism, 73(Suppl. 1), S531. Thomas, R. J., Rosen, B. R., Stern, C. E., Weiss, W., & Kwong, K. K. (2005). Functional imaging of working memory in obstructive sleep-disordered breathing. Journal of Applied Physiology, 98, 2226-2234. Tishler, P. V., Larkin, E. K., Schluchter, M. D., & Redline, S. (2003). Incidence of sleep- disordered breathing in an urban adult population: The relative importance of risk factors in the development of sleep-disordered breathing. Journal of the American Medical Association, 289, 2230-2237. Tonon, C, Vetrugno, R., Lodi, R., Gallassi, R., Provini, F., Iotti, S., et al. (2007). Proton magnetic resonance spectroscopy study of brain metabolism in obstructive sleep apnoea syndrome before and after continuous positive airway pressure treatment. Sleep, 30(3), 305-311. 226 Ulfberg, J., Carter, N., Talback, M., & Edling, C. (1996). Excessive daytime sleepiness at work and subjective work performance in the general population and among heavy snorers and patients with obstructive sleep apnea. Chest, 110, 659-663. Valencia-Flores, M., Bliwise, D. L., Guilleminault, C, Cilveti, R., & Clerk, A. (1996). 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, 18(2), 197-210. Veasey, S. C, Davis, C. W., Fenik, P., Zhan, G., Hsu, Y.-J., Pratico, D., et al. (2004). Long-term intermittent hypoxia in mice: Protracted hypersomnolence with oxidative injury to sleep-wake brain regions. Sleep, 27, 194-201. Verstraeten, E. (2007). Neurocognitive effects in obstructive sleep apnea syndrome. Current Neurology and Neuroscience Reports, 7, 161-166. Verstraeten, E., & Cluydts, R. (2004). Executive control of attention in sleep apnea patients: Theoretical concepts and methodological considerations. Sleep Medicine Review, 8, 257-267. Verstraeten, E., Cluydts, R., Pevernagie, D., & Hoffmannm, G. (2004). Executive function in sleep apnea: Controlling for attentional capacity in assessing executive attention. Sleep, 27(4), 685-693. Verstraeten, E., Cluydts, R., Verbraecken, J., & De Roeck, J. (1996). Neuropsychological functioning and determinants of morning alertness in patients with obstructive sleep apnea syndrome. Journal of the International Neuropsychological Society., 2, 306-314. Verstraeten, E., Cluydts, R., Verbraecken, J., & De Roeck, J. (1997). Psychomotor and cognitive performance in nonapneic snorers: Preliminary findings. Perceptual and Motor Skills, 84(3), 1211-1222. Verstraeten, E., Hoffmann, G., & Cluydts, R. (2000). Attentional capacity and/or executive control deficits in obstructive sleep apnea. Journal of Sleep Research, 9(Suppl. 1), 201. Vgontzas, A. N., Zoumakis, E., Bixler, E. O., Lin, H.-M., Follett, H., Kales, A., et al. (2004). Adverse effects of modest sleep restriction on sleepiness, performance, and inflammatory cytokines. The Journal of Clinical Endocrinology & Metabolism, 89(5), 2119-2126. 227 Visser, M. R. M., Oort, F. J., & Sprangers, M. A. G. (2005). Methods to detect response shifts in quality of life data: A convergent validity study. Quality of Life Research, 14, 629-639. Wagle, A. C, Berrios, G. E., & Ho, L. (1999). The cognitive failures questionnaire in psychiatry. Comprehensive Psychiatry, 40(6), 478-484. Walsleben, J. A., Squires, N. K., & Rothenberger, V. L. (1989). Auditory event-related potentials and brain dysfunction in sleep apnea. Electroencephalography and Clinical Neurophysiology, 74(4), 297-311. Weaver, T. E. (2001). Outcome measurement in sleep medicine practice and research. Part 2: Assessment of neurobehavioral performance and mood. Sleep Medicine Reviews, 5(3), 223-236. Weaver, T. E., & George, C. F. P. (2005). Cognition and performance in patients with obstructive sleep apnea. In M. H. Kryger, T. Roth & W. C. Dement (Eds.), Principles and practice of sleep medicine (4th ed., pp. 1023-1033). Philadelphia: Elsevier Saunder. Weaver, T. E., Laizner, A. M., Evans, L. K., Maislin, G., Chugh, D. K., Lyon, K., et al. (1997). An instrument to measure functional outcomes for disorders of excessive sleepiness. Sleep, 20(10), 835-843. Wechsler, D. (1987). Wechsler memory scale-revised manual. San Diego, TX: The Psychological Corporation. Wieber, S. J. (2005). The cardiac consequences of the obstructive sleep apnea-hypopnea syndrome. The Mount Sinai Journal of Medicine, 72(1), 10-12. Wilson, M. D. (1988). The mrc psycholinguistic database: Machine readable dictionary, version 2. Behavioural Research Methods, Instruments and Computers, 20, 6-11. Xu, W., Chi, L., Row, B. W., Xu, R., Ke, Y., Xu, B., et al. (2004). Increased oxidative stress is associated with chronic intermittent hypoxia-mediated brain cortical neuronal cell apoptosis in a mouse model of sleep apnea. Neuroscience, 126, 313- 323. Yaggi, H., & Mohsenin, V. (2004). Obstructive sleep apnoea and stroke. Lancet Neurology, 3, 333-342. 228 Yaggi, H. K., Concato, J., Kernan, W. N., Lichtman, J., Brass, L., & Mohsenin, V. (2005). Obstructive sleep apnea as a risk factor for stroke and death. The New England Journal of Medicine, 343, 2034-2041. Ye, J. Y., Li, Y. R., Wang, X. Y., Kou, Y., Zhang, Y. H., Xiu, D., et al. (2005). Significance of apnea hypopnea time index in obstructive sleep apnea hypopnea syndrome patients and possible correlation between polysomnographic parameters and severity of the disease. National Medical Journal of China, 85(32), 2274-2278. Young, T., Blustein, J., Finn, L., & Palta, M. (1997a). Sleep-disordered breathing and motor vehicle accidents in a population-based sample of employed adults. Sleep, 20, 608-613. Young, T., Evans, L., Finn, L., & Palta, M. (1997b). Estimation of the clinically diagnosed proportion of sleep apnea syndrome in middle-aged men and women. Sleep, 20, 705-706. Young, T., Palta, M., Dempsey, J., Skatrud, J., Weber, S., & Badr, S. (1993). The occurrence of sleep-disordered breathing among middle-aged adults. The New England Journal of Medicine, 328, 1230-1235. Young, T., Peppard, P. E., & Gottlieb, D. J. (2002). Epidemiology of obstructive sleep apnea: A population health perspective. American Journal of Respiratory and Critical Care Medicine, 165, 1217-1239. Yu, B.-H., Ancoli-Israel, S., & Dimsdale, J. E. (1999). Effect of CPAP treatment on mood states in patients with sleep apnea. Journal of Psychiatric Research, 33, 427-432. Yue, W., Hao, W., Liu, P., Liu, T., Ni, M., & Guo, Q. (2003). A case control study on psychological apnea-hypopnea syndrome. Canadian Journal of Psychiatry, 48(5), 318-323. 229 APPENDIX A - SCREENING QUESTIONNAIRES SDQ Sleep disorders Questionnaire (version 2.01) Copyright© 1986-1989, 1991, 1992 created by Alan Douglass MD, Robert Bornstein PhD, German Nino-Murcia MD. Derived from a pool of questions (Sleep Questionnaire and Assessment of Wakefulness, "SQAW") created at Standford University Sleep Disorders Clinic by Drs. Laughton Miles, Christian Guilleminault, Vincent Zarcone, and William Dement. The SQAW was copyrighted by Dr. Miles, 1979, and is used here by permission. The SDQ is copyrighted by the seven above-named persons. Dept. of Psychiatry, B2951 CFOB, University of Michigan, 1500 E. medical Center Drive, ann arbor, MI, 48109-0704 (send correspondence to this address). Department of Psychiatry, Ohio state University, Columbus, OH Sleep Medicine and Neuroscience Institute, Palo Alto, CA Appendix A - Screening Questionnaires (continued) 230 - page 2 - Instructions: This questionnaire will give your doctor a good understanding about your problems with sleeping and waking. It is very important to answer every question, because some disorders show up as a pattern of answers to different questions. In answering the questions, consider each question as applying to the past six months of your life, unless you have been told differently by the persong who gave you this booklet. Some people work night shift, or rotating shifts. Others have a very changeable bedtime. For these people, questions which ask about "day, daytime, morning, etc." will mean the time when they wake from their longest sleep of the day and become active. Similarly, "night, nighttime, bedtime, nocturnal" would refer to whatever time of day it is that they are having their longest sleep of the day. Most of the questions are simple statements. You answer by circling a number from 1 to 5. If you strongly disagree with the statement, or if it never happens to you, answer" 1". If the statement is always true in your case, or you agree strongly with it, answer "5". You may also choose "2 rarely", "3 sometimes", or "4 usually" as your answer. Notice that an "answer key" appears at the bottom of each page to remind you what is meant by the numbers. Please answer all of the questions. Here is an example of how to fill out a question: 1. How often does it snow in Florida in July? 12 3 4 5 IF YOU A CERTAIN THAT A QUESTION DOES NOT APPLY TO YOU, LEAVE IT BLANK. But try to answer every question if at all possible. This is important. Notice that answer "1" can mean that the things in the question never happen to you. If you are using the computerized answer sheet, blacken the space which corresponds to your answer, "1 to 5", instead of circling the answer in the booklet. ?l» rfc J^> ^j> ^ *jC ^C ^ tfc 5|* »j£ »jC »p *j> 5JC JjC 5j» *p £fC *jC Jf» 2|C PfC •jC *j£ Sj* *fs *jC ^JC l^^\7 T/"^Y* d Tl C \H7f-^"t*C ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ 1 2 3 4 5 NEVER RARELY SOMETIMES USUALLY ALWAYS (strongly disagree) (disagree) (not sure) (agree) (agree strongly) Appendix A - Screening Questionnaires (continued) 231 - page 3 - 1.1 get too little sleep at night 12 3 4 5 2.1 often have a poor night's sleep 12 3 4 5 3.1 have trouble getting to sleep at night 12 3 4 5 4.1 wake up often during the night 12 3 4 5 5. My bed time varies a lot 12 3 4 5 6. At bedtime, thoughts race through my mind 12 3 4 5 7. At bedtime, I feel sad and depressed 12 3 4 5 8. At bedtime, I worry about things 12 3 4 5 9. At bedtime, I feel muscular tension 12 3 4 5 10. At bedtime, I'm afraid of not being able to go to sleep 12 3 4 5 11. When falling asleep, I feel paralyzed (unable to move) 12 3 4 5 12. When falling asleep, I have "restless legs" ( a feeling of crawling, aching or inability to keep legs still) 12 3 4 5 13. After waking at night, I fear I will not be able to get back to sleep 12 3 4 5 14. My night sleep is restless and disrupted 12 3 4 5 15. At night, my sleep disturbs my bed partner's sleep 12 3 4 5 16. My night sleep is disturbed by light 12 3 4 5 17. My night sleep is disturbed by noise 12 3 4 5 18. My sleep is disturbed by severe heartburn and choking ("regurgitation" bringing up bitter stomach fluid) 12 3 4 5 *T* "T* *T^ *T* *f* ^T* N^ *K 'I* *K ^ ^ *I^ ^1* 'K **^ ^ *T* *T* *I* ^ *l* *T^ "** *I* *3* *f^ *l^ *l* jrpiT T^~\"t, OT"tC\17^T*C ^ ^" ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ *^ ^ ^ ^ ^ ^ **^ ^ ^ ^* *** ^ ^ 1 2 3 4 5 NEVER RARELY SOMETIMES USUALLY ALWAYS (strongly disagree) (disagree) (not sure) (agree) (agree strongly) Appendix A - Screening Questionnaires (continued) 232 - page 4 - 19.1 often wake up because I am hungry 12 3 4 5 20.1 snore in my sleep 12 3 4 5 21.1 am told I snore loudly and bother other 12 3 4 5 22.1 am told I stop breathing ("hold my breath") in sleep 12 3 4 5 23.1 awake suddenly gasping for breath, unable to breathe 12 3 4 5 24. At night my heart pounds, beats rapidly, or beats irregularly 12 3 4 5 ("palpitations") 25.1 sweat a great deal at night 12 3 4 5 26.1 walk in my sleep 12 3 4 5 27.1 grind my teeth while I sleep 12 3 4 5 28.1 wake from sleep screaming, confused and at times violent 12 3 4 5 ("night terrors") 29. My sleep is disrupted because of pain in the neck, 12 3 4 5 back, muscles, joints, legs or arms 30. My sleep is disrupted by chest pain (not angina) 12 3 4 5 31. My sleep is disrupted by "restless legs" 12 3 4 5 (a feeling of crawling, aching, inability to keep legs still) 32. My sleep is disrupted by thoughts racing through my mind 12 3 4 5 33. My sleep is disrupted by sadness or depression 12 3 4 5 34. My sleep is disrupted by worrying about things 12 3 4 5 *K *l^ *i^ *T* *T* *f* *I* "J^ *K *T* *P *T* *T* *t* *J* *$* *J* *j* *T* *T^ *T* *1* *l* "f* *¥* *f* *P *F *i^ K"£*\7 T/~\"I* d Y"l t! \17^'1*C **^ *^ ^ ^ ^ ^ *** *** *** "** ^ ^ ^* ^* *** *•* *•* ^ ^ ^* *** "•* *•* *•* *•* ^ ^ ^ ^* *^ 1 2 3 4 5 NEVER RARELY SOMETIMES USUALLY ALWAYS (strongly disagree) (disagree) (not sure) (agree) (agree strongly) Appendix A - Screening Questionnaires (continued) 233 - page 5 - 35. My sleep is disrupted by muscular tension 12 3 4 5 36. My sleep is disrupted by fears that I might not be able to get back to sleep if I should wake up 12 3 4 5 37.1 often have a night full of intense vivid dreams 12 3 4 5 38.1 have a lot of nightmares (frightening dreams) 12 3 4 5 39.1 feel unable to move (paralyzed) after a nap 12 3 4 5 40.1 have dream-like images (hallucinations) when I awaken in the morning even though I know I am not asleep 12 3 4 5 41.1 am sometimes very sleepy in the daytime, and this seems to go in cycles at regular intervals 12 3 4 5 42.1 have slept for several days at a time, or at least I have been overwhelmingly sleepy for that long 12 3 4 5 43.1 have been unable to sleep at all for several days 12 3 4 5 44.1 feel my sleep is abnormal 12 3 4 5 45.1 feel that I have insomnia 12 3 4 5 46. As a child, I had difficulty waking up in the morning 12 3 4 5 47. As a child, I had sleepiness during the day 12 3 4 5 48.1 have a problem because of headaches while sleeping 12 3 4 5 49. As a child, I was fatigued during the day 12 3 4 5 50. As a child, I rocked myself to get to sleep 12 3 4 5 •r* *T* *K *1> *1* *T* *l* •!* 'P 'I* 'f* ^ "I^ *** *1^ ^* *I* *P **• M^ *f* *J^ M* *t^ *f* **^ *t* ^ *1* |^^\7 T^^T* CI Tl C \X/f-*T*C ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ *^ ^ 1 2 3 4 5 NEVER RARELY SOMETIMES USUALLY ALWAYS (strongly disagree) (disagree) (not sure) (agree) (agree strongly) Appendix A - Screening Questionnaires (continued) 234 - page 6 - 51.1 used to bang my head as a child 12 3 4 5 52.1 used to sleepwalk in childhood 12 3 4 5 53. As a child, I had convulsions (seizures) during sleep 12 3 4 5 54. As a child, I would grind my teeth while asleep 12 3 4 5 55. Now, I am very sleepy during the day and I struggle to stay awake 12 3 4 5 56. In the past 6 months, I have fallen asleep accidentally in some of these situations: eating a meal, talking on the phone, talking to someone, riding in a bus or car, watching TV, at a theater, reading a book, at a lecture 12 3 4 5 57.1 got bad grades in school because I was too sleepy 12 3 4 5 58.1 now have trouble doing my job because of sleepiness or fatigue 12 3 4 5 59.1 often have to let someone else drive the car because I am too sleepy to do it 12 3 4 5 60.1 see vivid dream-like images (hallucinations) either just before or just after a daytime nap, yet I am awake when they happen 12 3 4 5 61.1 have vivid dreams during my daytime naps 12 3 4 5 62.1 am often unable to move (paralyzed) when I am waking up in the morning 12 3 4 5 63. Sometimes I realize I have driven my car to the wrong place, and I can't remember how I did it 12345 64.1 find myself doing things which make no sense, such as writing nonsense instead of notes, or mixing together chocolate and gravy 12 3 4 5 1 2 3 4 5 NEVER RARELY SOMETIMES USUALLY ALWAYS (strongly disagree) (disagree) (not sure) (agree) (agree strongly) Appendix A - Screening Questionnaires (continued) 235 - page 7 - 65. People tell me that I act strangely at times, and yet I was not aware of it when it happened 12 3 4 5 66.1 get "week knees" when I laugh 12 3 4 5 67.1 get sudden muscular weakness (or a brief period of paralysis, being unable to move) when laughing, angry, or in situations of strong emotion 12 3 4 5 68.1 am excessively sleepy during the daytime 12 3 4 5 69.1 have at some time had trouble with my bladder 12 3 4 5 70.1 have had problems with tonsils or adenoids 12 3 4 5 71.1 have high blood pressure (or once had it) 12 3 4 5 72. My tonsils and/ or adenoids have been removed 12 3 4 5 73.1 get pains in my abdomen (stomach) 12 3 4 5 74.1 have had a head injury 12 3 4 5 75.1 have been knocked unconscious (knocked out) 12 3 4 5 76.1 suffer from dizzy spells 12 3 4 5 77.1 have seizures ("fits", convulsions, epilepsy) 12 3 4 5 78.1 have problems with clumsiness, incoordination 12 3 4 5 79.1 feel that I have a sexual problem 12 3 4 5 80. My desire or interest in sex is less than it used to be 12 3 4 5 81.1 have pain or discomfort during sexual intercourse 12 3 4 5 *p »j£ *jC *j* *j£ *fl *j» Jj> -^C ^ ^i* »j» *^C *j£ *jC *jC *j* ^ *ft *j£ *p *p *jC »[£ »jC »jC *|» 2jC »f» 1^^"\ j i f\X* OT1 C \H/£*T*C ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^* ^* ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ 1 2 3 4 5 NEVER RARELY SOMETIMES USUALLY ALWAYS (strongly disagree) (disagree) (not sure) (agree) (agree strongly) Appendix A - Screening Questionnaires (continued) 236 - page 8 - 82.1 sleep better after having sex 12 3 4 5 83.1 am unhappy about my social life 12 3 4 5 84.1 am unhappy about loving relationships in my life 12 3 4 5 85.1 am unhappy about my sex life 12 345 86.1 am dissatisfied with my job 12 3 4 5 87.1 have a problem with my sleep 12 3 4 5 88.1 wake up in the morning with a headache 12 3 4 5 89.1 have considered or attempted suicide 12 3 4 5 90.1 feel I am useful and needed 12 3 4 5 91.1 am sleeping more than I used to 12 3 4 5 92. Someone in my immediate family has trouble with insomnia (brother/sister, father/mother, son/daughter, gradparent) 12 3 4 5 93. Someone in my immediate family is very sleepy during the day 12 3 4 5 94. Someone in my immediate family has psychiatric or emotional illness (EG: depression, alcoholism) 12 3 4 5 95. Some of my other relatives have trouble with insomnia (uncles, aunts, cousins) 12 3 4 5 96. Some of my other relatives are very sleepy during the day 12 3 4 5 97. Some of my other relatives have psychiatric illness 12 3 4 5 98. Some family member has died suddenly in their sleep 12 3 4 5 *r* ^ ^ *I^ *T* *i* *T* *I* *l* ^ *i* *t* 'I* "j^ ^ *i* *r* ^ *T^ 'I* *T^ *T* *i^ 'P *I* ^ *T* *i* *T* l^O"\7 T^^Y* C\ n C\X/^T*C ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ 1 2 3 4 5 NEVER RARELY SOMETIMES USUALLY ALWAYS (strongly disagree) (disagree) (not sure) (agree) (agree strongly) Appendix A - Screening Questionnaires (continued) 237 - page 9 - 99. Some family member has "restless legs" while sleeping (a feeling of crawling, aching, inability to keep the legs still) 12 3 4 5 100. A child in my family died from "crib death" (sudden infant death syndrome, SIDS) 12 3 4 5 101. Someone in my family has been hospitalized for a psychiatric illness or "nervous breakdown" 12 3 4 5 102. People in my family seem to be worriers 12 3 4 5 103. Someone in my family has diabetes 12 3 4 5 104. Someone in my family has had a stroke ("apoplexy") 12 3 4 5 105.1 often use alcohol in order to get to sleep 12 3 4 5 106.1 use alcohol to steady my nerves 12 3 4 5 107. While drinking alcohol, I have carried out actions without being aware of them, and not remembered them the next day 12 3 4 5 108.1 smoke tobacco within two hours of bedtime 12 3 4 5 109.1 have used "street drugs" (marijuana, "uppers", "downers", narcotics, hallucinogens, cocaine) 12 3 4 5 110.1 have used tobacco to help me go to sleep 12 3 4 5 111.1 have used marijuana to help me go to sleep 12 3 4 5 112.1 currently take a non-prescription drug from the pharmacy in order to help me sleep 12 3 4 5 113.1 currently take a non-prescription drug to stop me being so sleepy and fatigued in the daytime 12 3 4 5 1 2 3 4 5 NEVER RARELY SOMETIMES USUALLY ALWAYS (strongly disagree) (disagree) (not sure) (agree) (agree strongly) Appendix A - Screening Questionnaires (continued) 238 - page 10- 114.1 take a prescription drug which the doctor gave me mainly to help me sleep ( sleeping pills, anti-depressants, tranquilizers) 12 3 4 5 115.1 take a prescription drug which the doctor gave me mainly to keep me awake during the day (EG: ritalin) 12 3 4 5 116.1 take some drugs at night for my other illnesses, not related to sleep, yet I find they help me sleep 12 3 4 5 117.1 have taken drugs for my heart 12345 118.1 use relaxation techniques or mental imagery (EG: counting sheep) to help me sleep 12 3 4 5 119.1 use non-drug therapies in order to get to sleep (EG: biofeedback, acupuncture, electrosleep) 12 3 4 5 120.1 exercise regularly 12 3 4 5 121.1 was born as part of a multiple birth (twins, or triplets, etc. Includes cases where the others died at birth or afterwards) 12 3 4 5 122. My family was emotionally close in my childhood 12 3 4 5 123.1 got along well with my parents while growing up 12 3 4 5 124.1 am currently unemployed 12 3 4 5 125.1 am working at a job with rotating shifts 12 3 4 5 126.1 have had a job where I worked unusual times 12 3 4 5 127.1 am presently living in a house 12 3 4 5 128.1 get along with my husband/wife/friend, who is currently living with me 12 3 4 5 *t* ^t* *t* M^ *J* ^ *i* M* *T^ *I* "I* M* *I* "t* *t* *Tt* *%* *K *i^ *t* *T* *I* ^P *P *t* *I^ *P *P *3^ K"^\7 I"(~\1<* *I\ WC\X7£*T*C ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^* ^ ^ ^* ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ 1 2 3 4 5 NEVER RARELY SOMETIMES USUALLY ALWAYS (strongly disagree) (disagree) (not sure) (agree) (agree strongly) Appendix A - Screening Questionnaires (continued) 239 - page 11- 129. Coffee, tea, or cola drinks seem to worsen my sleep 12 3 4 5 130. Mental stress, worry, or anxiety worsens my sleep 12 3 4 5 131. Physical exercise helps my sleep 12 3 4 5 132. A daytime nap worsens my nighttime sleep 12 3 4 5 133. Mental stress, worry, or anxiety makes me feel sleepy during the day 12 3 4 5 134. After a nap, I feel less sleepy in the daytime 12 3 4 5 135. Hot weather makes me sleepy during the day 12 3 4 5 136. When doing shift work, I am sleepy during the day 12 3 4 5 137.1 have a small jaw, or other abnormality of the bones in my head or neck. 12 3 4 5 138.1 have a chronic chest disease (bronchitis, asthma, emphysema) 12 3 4 5 139.1 have a problem with my nose blocking up when I am trying to sleep (allergies, infections) 12 3 4 5 140.1 wake up with "attacks" which are different from those described anywhere else in this questionnaire 12 3 4 5 141. My snoring or my breathing problem is much worse if I sleep on my back 12 3 4 5 142. My snoring or my breathing problem is much worse if I fall asleep right after drinking alcohol 12 3 4 5 143. My snoring or my breathing problem is much worse when I have an allergy or infection in the nose, throat, or chest. 12 3 4 5 **• *T* *p *t* *f* *»* •** *T* *l* *T* ^K T* *p "l^ "l^ "T* *P *1* •!* *i* *K *P *I^ *f* 'H 'I* *I^ H^ *T^ K"^\^ Ti^"!* d Y"l C\T7£^"f,C ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ 1 2 3 4 5 NEVER RARELY SOMETIMES USUALLY ALWAYS (strongly disagree) (disagree) (not sure) (agree) (agree strongly) Appendix A - Screening Questionnaires (continued) 240 - page 12 - THE FOLLOWING QUESTIONS ARE FOR WOMEN ONLY: 144.1 have gone through the menopause ("change of life") 12 3 4 5 145. My sleep at night is affected by my menstrual cycle 12 3 4 5 146. My daytime sleepiness worsens with pregnancy 12 3 4 5 147. My daytime sleepiness is worse since my menopause 12 3 4 5 THE FOLLOWING QUESTIONS ARE FOR MEN ONLY: 148.1 often have problems getting an erection 12 3 4 5 149.1 have trouble maintaining an erection 12 3 4 5 150.1 have trouble with ejaculation (either I can't do it at all, or it happens too soon) 12 3 4 5 151. My erections are physically distorted 12 3 4 5 152.1 often awaken with an erection during the night or in the morning 12 3 4 5 1 2 3 4 5 NEVER RARELY SOMETIMES USUALLY ALWAYS (strongly disagree) (disagree) (not sure) (agree) (agree strongly) Appendix A - Screening Questionnaires (continued) 241 - page 13 - IN THE NEXT SECTION, PLEASE CIRCLE THE ITEM (NUMBERED 1-5) WHICH BEST MATCHES YOUR ANSWER. 153. How many hours of sleep do you get at night, not including time spent awake in bed? 1) Less than4hrs. 2) Four to 5 hrs. 3)Sixhrs. 4)Sevenhrs. 5) Eight or more hrs. 154. How long is your longest wake period at night? 1) Less than 5 min. 2) Six to 19 min. 3) 20 to 59 min. 4) One to 2 hrs. 5) More than 2 hrs. 155. How many times in a night do you get up to urinate? l)None. 2) One time. 3) Two times. 4) Three times. 5) Four or more times 156. How many work accidents have you had as a result of sleepiness or fatigue? l)None 2) One 3) Two 4) Three 5) Four or more 157. How many car accidents or "near misses" have you had because of excessive sleepiness? l)None 2) One 3) Two 4) Three 5) Four or more 158. How many daytime naps (asleep for 5 minutes or more) do you take on an average working day? l)None 2) One 3) Two 4) Three or four 5) Four or more 159. How many rest periods do you take on an average working day (but do not sleep during them)? 1) None 2) One 3) Two or three 4) Four or five 5) Six or more Appendix A - Screening Questionnaires (continued) 242 - page 14 - 160. How many times, in an average working day, do you try to nap but find that you can't fall asleep? 1) None 2) One 3) Two 4) Three 5) Four or more 161. How long do you remain restored (refreshed, alert) after a daytime nap? 1) Less than 1 hour 2) One to 2 hours 3) Three hours 4) Four or 5 hours 5) Six hours or more 162. How long do you remain restored after a rest? 1) Less than 30 min. 2) 30-59 minutes 3) One to 2 hrs. 4) Three to 4 hr. 5) Five hours or more 163. What is your current weight (in lb.) ? 1) 134 lb. or less 2) 135-159 lb. 3) 160-183 lb. 4) 184-209 lb. 5) 210 lb. or more 164. What was your weight six months ago? 1) 134 lb. or less 2) 135-159 lb. 3) 160-183 lb. 4) 184-209 lb. 5) 210 lb. or more 165. What was your weight at age 20? 1) 125 lb. or less 2) 126-139 lb. 3) 140-155 lb. 4) 156-175 lb. 5) 176 lb. or more 166. How many cups of regular coffee do you have in a day? 1) None 2) One cup 3) Two cups 4) 3 to 5 cups 5) Six cups or more 167. How many of the coffees are within 2 hrs. of bedtime? 1) None 2) One cup 3) Two cups 4) 3 to 5 cups 5) Six cups or more 168. How many glasses/cans of cola drinks do you have in a day (do not include decaffeinated types)? 1) None 2) One can 3) Two cans 4) 3 to 5 cans 5) Six cans or more Appendix A - Screening Questionnaires (continued) 243 - page 15 - 169. How many of these colas are within 2 hrs. of bedtime? l)None 2) One can 3) Two cans 4) 3 to 5 cans 5) Six cans or more 170. How many years were you a smoker? 1) None 2) One year 3) 2 to 12 years 4) 13 to 25 years 5) 26 years or more 171. How long does it take you to adjust after traveling across time zones (especially 4 or more zones)? 1) No time at all 2) One day 3) Two days 4) three to 4 days 5) Five or more days 172. How tall are you? 1) 63 inches or less 2) 64 to 66 in. 3) 67 to 69.5 in. 4) 70 to 71 in. 5) 71.5 inches or taller 173. How old are you now? 1)25 or under 2) 26-35 yr. 3) 36-44 yr. 4) 45-50 yr. 5) 51 yr. or older 174. How many years did you go to school? Include year of college and university too. l)4yr. or less 2)5-llyr. 3)12yr. 4) 13-14 yr. 5) 15 yr. or more 175. Before this visit, how many "therapists" (doctor, psychiatrist, psychologist, nurse, counselor, osteopath, chiropractor) have you ever seen about a problem of sleeping too much or too little? 1) None 2) One only 3) Two 4) Three or 4 5) Five or more If you are using the computerized sheet, please check that you put your name, sex, and birthdate on that sheet. Also, please remember to fill in the circles under these items. Thank you. Appendix A - Screening Questionnaires (continued) 244 Berlin Questionnaire Age: Weight: (kg) Height: (cm) BMI: Code #:. Sex: Neck circumference: Ethnicity: Date: Please check ('V") the answer the best describes you. 1. Has your weight changed? Increased Decreased No change 2. Do you snore? Yes No Do not know 3. Snoring loudness Loud as breathing Loud as talking Louder than talking Very loud 4. Snoring frequency Almost every day 3-4 times/wk 1-2 times/wk 1-2 times/mo Never or almost never 5. Does your snoring bother other people? Yes No (P9 1 of 2) Appendix A - Screening Questionnaires (continued) 245 Berlin Questionnaire (continued) 6. How often have your breathing pauses been noticed? Almost every day 3-4 times/wk 1-2 times/wk 1-2 times/mo Never or almost never 7. Are you tired after sleeping? Almost every day 3-4 times/wk 1-2 times/wk 1-2 times/mo Never or almost never 8. Are you tired during waketime? Almost every day 3-4 times/wk 1-2 times/wk 1-2 times/mo Never or almost never 9. Have you ever fallen asleep while driving? Yes No 10. Do you have high blood pressure? Yes No Do not know (pg2of2) -End- 246 APPENDIX B - MEDICAL RECORD SUMMARY & PRE-TREATMENT SLEEP STUDY FORM Code# Date: Medical Record Summary Demoqraphics: Date of Birth: Age: Gender: Handedness: Height (cm): Weight (kg): Occupation: Phone Number: OSA Historv: OSA symptoms (onset: Years) Diagnosis of OSA (No. of Years AHI (PSG)/RDI (home study) =_ Respiratory company: CPAP machine model: Other Sleep Measures and Results: Duration of CPAP treatment since: (No. of Months Any history or indication of other OSA treatments Yes/No Any history or indication of other sleep disorders? (Specify) Other Relevant Medical Historv: Medical illnesses: Psychiatric illnesses: Medications: Remarks: Appendix B - Medical Record Summary and Pre-treatment Sleep Study Form (continued) 247 Date: Pre-Treatment Sleep Study Form Participant Code: Date of Birth: Date of Study: / / year month day Type of Study: • NPSG D Home study D Oximetry Interpreting Physician: D Dr. Morrison D Dr. Rajda D Dr. Kukreja Diag Treat Diag Treat Diag Treat TRT (min) Stage2(%) RDI TST (min) Stage3&4(%) MeanSa02(%) SleepLat REM% MinSA02(%) RemLat CentAp MaxSa02(%) SleepEff(%) ObsAp SaO2<90%(%TST) #REM Mixed Ap SaO2<80%(%TST) Awakenings ObsHypop TotArousal Ind Stage 1(%) CentHypop Resp Arousal Ind Remarks: Data entered in computer: D on 248 APPENDIX C HEALTH INTERVIEW Code #: Date: Demographics: Age: Date of Birth Gender: Education (years; repeated grade; LDs): Handedness: First Language: Height (cm): Weight (kg): Occupation: Phone Number: Medical Information: How would you describe your excellent very good good fair poor general state of health? Have you ever been diagnosed or treated for any of the following: If YES: Details • Head injury with loss of consciousness? YES NO • Headaches? YES NO • Seizures? YES NO • Sleep problems? YES NO • Diabetes? YES NO • Thyroid problems? YES NO • High Blood Pressure? YES NO • Stroke? YES NO • Heart problems? YES NO Uncontrolled irregular heartbeats? YES NO Bypass? YES NO Pacemaker? YES NO • Surgery? YES NO • Cancer? YES NO • Multiple Sclerosis? YES NO • Parkinson's Disease? YES NO • Huntington's Disease? YES NO • Rheumatoid Arthritis? YES NO (pg1of2) Appendix C - Health Interview (continued) 249 Health Interview (continued) Psychiatric Illness? YES NO PVD (poor circulation in legs)? YES NO Visual Problems? YES NO Glasses? YES NO If yes, are they present for testing? YES NO Colour Blindness? YES NO Do you have any trouble seeing blue or red? YES NO Trouble Hearing? YES NO Pregnancy? YES NO you drink alcohol? YES NO How much and for how long? Do you smoke? YES NO How much and for how long? How much caffeine do you consume on a regular day? (e.g. Coffee, tea, cola, chocolate) For OS A Patients: Onset of OSA symptoms Date Diagnosis of OSA Date Duration of CPAP treatment Date started Usage of CPAP Nights/week Hours/night Are there any other medical issues that we haven't asked about? Any hospitalizations (when, what, any complications)? Medications: (note any recent use of benzodiazepines, opiates, anti-psychotics; and sleep pills & allergy meds for the past 7 days) -'End- (pg 2 of 2) 250 APPENDIX D - SLEEP AND PSYCHOSOCIAL QUESTIONNAIRES Epworth Sleepiness Scale Code# Date: Your age: Your sex: How likely are you to doze off or fall asleep in the following situations, in contrast to feeling just tired? This refers to your usual way of life in recent times. Even if you have not done some of these things recently try to work out how they would have affected you. Use the following scale to choose the most appropriate number for each situation: 0 = would never doze 1 = slight chance of dozing 2 = moderate chance of dozing 3 = high chance of dozing Situation Chance of dozing Sitting and reading Watching TV Sitting, inactive in a public place (e.g. a theatre or a meeting) As a passenger in a car for an hour without a break Lying down to rest in the afternoon when circumstances permit Sitting and talking to someone Sitting quietly after a lunch without alcohol In a car, while stopped for a few minutes in the traffic Thank you for your cooperation! Appendix D - Sleep and Psychosocial Questionnaires (continued) 251 Stanford Sleepiness Scale Code #: Date: Please read the following statements. Then, circle the number corresponding to the statement that best describes your state of sleepiness AT THIS TIME. 1. Feeling active, vital; wide awake. 2. Functioning at a high level; but not at peak; able to concentrate. 3. Relaxed; awake; not at full alertness; responsive. 4. A little foggy; not at peak; let down. 5. Fogginess; beginning to lose interest in remaining awake; slowed down. 6. Sleepiness; prefer to be lying down; fighting sleep; woozy. 7. Almost in reverie; sleep onset soon; lost struggle to remain awake. Appendix D - Sleep and Psychosocial Questionnaires (continued) 252 Cognitive Failure Questionnaire Code #: Date: The following questions are about minor mistakes which everyone makes from time to time, but some of which happen more often than others, we want to know how often these things have happened to you in the last six months: Please circle the appropriate number. Very Quite Very Often Often Occasionally Rarely Never 1. Do you read something and find you haven't been thinking about it and must read it again? 4 3 2 10 2. Do you find you forget why you went from one part of the house to the other? 0 3. Do you fail to notice road signs? 4. Do you find you confuse right and left when giving directions? 5. Do you bump into people? 0 6. Do you find you forget whether you've turned off a light or the stove or locked the door? 7. Do you fail to listen to people's names when you are meeting them? 8. Do you say something and realize afterwards that it might be taken as insulting? 0 9. Do you fail to hear people speaking to you when you are doing something else? Appendix D - Sleep and Psychosocial Questionnaires (continued) 253 CFQ Continued Very Quite Very Often Often Occasionally Rarely Never 10. Do you lose your temper and regret it? 4 3 2 10 11. Do you leave important letters unanswered for days? 4 3 2 10 12. Do you find you forget which way to turn on a road you know well but rarely use? 4 3 2 10 13. Do you fail to see what you want in a supermarket (although it's there)? 43 2 10 14. Do you find yourself suddenly wondering whether you've used a word correctly? 4 3 2 10 15. Do you have trouble making up your mind? 4 3 2 10 16. Do you find you forget appointments? 4 3 2 10 17. Do you forget where you put something like a newspaper or a box? 4 3 2 10 18. Do you find you accidentally throw away the thing you want and keep what you meant to throw away - as in the example of throwing away the matchbook and putting the used match in your pocket? 4 3 2 10 19. Do you daydream when you ought to be listening to something ? 4 3 2 10 20. Do you find you forget people's names? 4 3 2 10 Appendix D - Sleep and Psychosocial Questionnaires (continued) 254 CFQ Continued Very Quite Very Often Often Occasionally Rarely Never 21. Do you start doing one thing at home and get distracted into doing something else (unintentionally)? 4 3 2 10 22. Do you find you can't quite remember something although it's on the tip of your tongue? 4 3 2 10 23. Do you find you forget what you came to the store to buy? 4 3 2 10 24. Do you drop things? 4 3 2 10 25. Do you find you can't think of anything to say? 4 3 2 10 Appendix D - Sleep and Psychosocial Questionnaires (continued) 255 FUNCTIONAL OUTCOMES OF SLEEP QUESTIONNAIRE (Canadian English version of the FOSQ) Some people have difficulty performing everyday activities when they feel tired or sleepy. The purpose of this questionnaire is to find out if you generally have difficulty carrying out certain activities because you are too sleepy or tired. In this questionnaire, when the words "sleepy" or "tired" are used, it means the feeling that you can't keep your eyes open, your head is droopy, that you want to "nod off, or that you feel the urge to take a nap. These words do not refer to the tired or fatigued feeling you may have after you have exercised. DIRECTIONS: Please put an (X) in the box for your answer to each question. Select only one answer for each question. Please try to be as accurate as possible. All information will be kept confidential. (0) (4) (3) (2) (1) I don't do No Yes, Yes, Yes, this activity difficulty a little moderate extreme for other difficulty difficulty difficulty reasons 1. Do you have difficulty concentrating on the things you do • • • • because you are sleepy or tired? 2. Do you generally have difficulty remembering things, n • n n because you are sleepy or tired? 3. Do you have difficulty finishing a meal because you become n • • • sleepy or tired? 4. Do you have difficulty working on a hobby (for example, • • n • • sewing, collecting, gardening) because you are sleepy or tired? ©Weaver, September 1996 Functional Outcomes of Sleep Questionnaire (FOSQ) Appendix D - Sleep and Psychosocial Questionnaires (continued) 256 (0) (4) (3) (2) (1) I don't do No Yes, Yes, Yes, this activity difficulty a little moderate extreme for other difficulty difficulty difficulty reasons 5. Do you have difficulty doing work around the house (for • • • • • example, cleaning house, doing laundry, taking out the trash, repair work) because you are sleepy or tired? 6. Do you have difficulty operating a motor vehicle for a short n • n • • time (less than 1 hour) because you become sleepy or tired? 7. Do you have difficulty operating a motor vehicle for a long • • • • • time (more than 1 hour) because you become sleepy or tired? 8. Do you have difficulty getting things done because you are too • • • • • sleepy or tired to drive or take public transportation? 9. Do you have difficulty taking care of financial affairs and doing • • • • • paperwork (for example, writing checks, paying bills, keeping financial records, filling out tax forms, etc.) because you are sleepy or tired? ©Weaver, September 1996 Functional Outcomes of Sleep Questionnaire (FOSQ) Appendix D - Sleep and Psychosocial Questionnaires (continued) 257 (0) (4) (3) (2) (1) I don't do No Yes, Yes, Yes, this activity difficulty a little moderate extreme for other difficulty difficulty difficulty reasons 10. Do you have difficulty performing employed or volunteer work because you are sleepy or • tired? 11. Do you have difficulty • • maintaining a telephone n conversation, because you become sleepy or tired? 12. Do you have difficulty visiting with your family or friends n n in your home because you become • sleepy or tired? 13. Do you have difficulty visiting with your family or friends • n in their home because you become sleepy or tired? 14. Do you have difficulty doing things for your family or friends • • because you are too sleepy or tired? (4) (3) (2) (1) No Yes, Yes, Yes, a little moderately extremely 15. Has your relationship with fH family, friends or work colleagues • • • been affected because you are sleepy or tired? In what way has your relationship been affected? ©Weaver, September 1996 Functional Outcomes of Sleep Questionnaire (FOSQ) Appendix D - Sleep and Psychosocial Questionnaires (continued) 258 (0) (4) (3) (2) (1) I don't do No Yes, Yes, Yes, this activity difficulty a little moderate extreme for other difficulty difficulty difficulty reasons 16. Do you have difficulty exercising or participating in a • • • • • sporting activity because you are too sleepy or tired? 17. Do you have difficulty watching a movie or videotape • • • • • because you become sleepy or tired? 18. Do you have difficulty enjoying the theater or a lecture • n • • • because you become sleepy or tired? 19. Do you have difficulty enjoying a concert because you • • • • • become sleepy or tired? 20. Do you have difficulty watching TV because you are sleepy • • • • • or tired? 21. Do you have difficulty participating in religious services, n • • • n meetings or a group or club, because you are sleepy or tired? 22. Do you have difficulty being as active as you want to be in the • • • • evening because you are sleepy or tired? 23. Do you have difficulty being as active as you want to be in the • n n • morning because you are sleepy or tired? ©Weaver, September 1996 Functional Outcomes of Sleep Questionnaire (FOSQ) Appendix D - Sleep and Psychosocial Questionnaires (continued) 259 (0) (4) (3) (2) (1) I don't do No Yes, Yes, Yes, this for other difficulty a moderate extreme reasons little difficulty difficulty difficulty 24. Do you have difficulty being as active as you want to be in the • • • • afternoon because you are sleepy or tired? 25. Do you have difficulty keeping up with others your own age n • • n because you are sleepy or tired? (1) (2) (3) (4) Very Low Low Medium High 26. How would you rate your general level of activity? • n • • (0) (4) (3) (2) (1) No No Yes, Yes, Yes, intimate or a moderately extremely sexual little relationshi P 27. Has your relationship with your intimate or sexual partner been • • • • • overall affected because you are sleepy or tired? IF NO RELATIONSHIP STOP HERE!! IF NO RELATIONSHIP STOP HERE!! (0) (4) (3) (2) (1) No No Yes, Yes, Yes, intimate or a moderately extremely sexual little relationshi P 28. Has your desire for intimacy or sex been affected because you are • • • • sleepy or tired? Appendix D - Sleep and Psychosocial Questionnaires (continued) 260 (4) (3) (2) (1) No Yes, Yes, Yes a moderately extremely little 29. Has your ability to become sexually aroused been affected because you are sleepy or tired? n • n • 30. Has your ability to "come" (have an orgasm) been affected because you are sleepy or tired? n n • n ©Weaver, September 1996 Functional Outcomes of Sleep Questionnaire (FOSQ) Appendix D - Sleep and Psychosocial Questionnaires (continued) QUEBEC SLEEP QUESTIONNAIRE This questionnaire has been designed to find out how you have been doing and feeling over the last 4 weeks. You will be questioned about the impact that sleep apnea may have had on your daily activities, your emotional functioning, and your social interactions, and at>out any symptoms it might have caused. A large A moderate A moderate A small to A small amount of to large amount of moderate amount of All the Bine Not at all During the last 4 weeks : the time amount of the time amount of the time the time the time 1. Have you had to force yourself to do your 1 2 3 4 5 6 7 activities? 2. Have you disturbed everyone at night while 1 2 3 4 5 6 7 staying with friends? 3. Have you felt like not wanting to do things 1 2 3 4 5 6 7 together with your partner, children or friends? 4. Have you woken up more than once per night 1 2 3 4 5 6 7 to urinate? 5. Have you been feeling depressed? 1 2 3 4 5 6 7 6. Have you been feeling anxious or fearful 1 2 3 4 5 6 7 about what was wrong? 7. Have you needed to map during the day? 1 2 3 4 5 6 7 Page 1 of 4 Appendix D - Sleep and Psychosocial Questionnaires (continued) A large A moderate A moderate A small to A small amount of to large amount of moderate amount of AB the time Not at at During the last 4 weeks : the time amount of the time amount of the time the time the time 6. Have you teen feeling Impatient? 1 2 3 4 5 6 7 9. Have you woken up often {more than twice) 1 2 3 4 5 6 7 during the night? A very large A large A moderate A moderate A small to A small amount amount to large: amount moderate amount None During the last 4 weeks.: amount amount 10. Have you had difficulty with trying to 1 2 3 4 5 6 7 remember things? 11. Have you had difficulty with trying to 1 2 3 4 5 6 7 concentrate? 12. Have you been upset about being told that 1 2 3 4 5 6 7 your snoring was bothersome or irritating? 13. Have you felt guilty about your relationship with family members or close personal 1 2 3 4 5 6 7 friends? 14. Have you noticed a decrease in your 1 2 3 4 5 6 7 performance at work? 15. Have you been concerned about heart problems or premature death? 1 2 3 4 5 8 7 Page 2: of 4 Appendix D - Sleep and Psychosocial Questionnaires (continued) During the last 4 weeks, how much of a problem A. very large A large A moderate A moderate A small to A small problem problem to large moderate problem have you had with : problem No problem problem problem 18. Having to fight to stay awake during She day? 2 3 4 5 6 7 17. Feeling decreased energy? 2 3 4 5 6 7 18. Feeling excessive fatigue? 2 3 4 5 6 7 13. Feeling that ordinary activities require an extra 2 3 4 5 6 7 effort to perform or complete? 20. Falling asleep if not stimulated or active? 2 3 4 5 6 7 21. Difficulty with a dry or sore mouth/throat upon 2 3 4 5 6 7 awakening? 22. Difficulty returning to sleep if you wake up in 2 3 4 5 6 7 the night? 23. Feeling that you lack energy? 2 3 4 5 8 7 Page 3 of 4 Appendix D - Sleep and Psychosocial Questionnaires (continued) During the last 4 weeks, how much of a problem A very large A large A moderate A moderate A small to A small No problem have you had with : problem problem to large problem moderate problem problem problem 24. Concern about the times you stop breathing at 1 2 3 4 5 6 7 night? 25. Loud snoring? 1 2 3 4 5 6 7 26. Difficulties with attention? 1 2 3 4 5 6 7 27. Falling asleep suddenly? 1 2 3 4 5 6 7 28. Waking up at night feeling like you were 1 2 3 4 5 6 7 choking? 29. Waking up in the morning feeling unrefreshed 1 2 3 4 5 6 7 and/or tired? 30. A feeiing that your steep is restless? 1 2 3 4 5 6 7 31. Difficulty staying awake while reading? 1 2 3 4 5 6 7 32. Fighting the urge to fall asleep while driving? 1 2 3 4 5 6 7 © Universite Laval 2002 — tous droits reserves Page 4 of 4 Appendix D - Sleep and Psychosocial Questionnaires (continued) 265 Participant Code: Date: Visual Analogue Scale Quality of Life (OSA) Current Before CPAP Best Imaginable Best Imaginable 100 100 0 Worst Imaginable Worst Imaginable Appendix D - Sleep and Psychosocial Questionnaires (continued) 266 Participant Code:. Date: Visual Analogue Scale Quality of Life (Control) Current 3 Months Ago Best Imaginable Best Imaginable 100 100 0 Worst Imaginable Worst Imaginable 267 APPENDIX E - NEUROPSYCHOLOGICAL TESTS Neuropsychological Test Battery Test Function Time of Completion Wechsler Adult Intelligence Scale (WAIS-R) (short form General intellectual functioning 30 minutes 2) - Vocabulary & Block Design) D2 Test of Attention! Selective attention 8 minutes Concentration/Psychomotor Digit Symbol (WAIS-R)§ 90 seconds slowing Psychomotor Vigilance Test Vigilance 10 minutes Wechsler Memory Scale (WMS-R) -Visual Span Short-term/working memory 10 minutes WAIS-R - Digit Span Forward, Backward5 Brown-Peterson - Consonant Short-term/working memory 10 minutes Trigrams (CCC)§ California Verbal Learning Verbal learning 10-15 minutes Test II (CVLT II) Rey-Osterreith Complex Visuoconstructive abilities; Figure (RCFT) - 10-15 minutes visual memory immediate recall & delay Wisconsin Card Sorting Test Executive functioning 15-30 minutes (WCST)§ Trail Making Test (A, B§) Sequencing, mental flexibility 5-10 minutes Wechsler Intelligence Scale for Children (WISC-III)- Planning 5 minutes Mazes§ Mental flexibility, response Stroop5 5 minutes inhibition Grooved Pegboard Test Manual dexterity 5 minutes § Tests that involve executive function and controlled attention Appendix E - Neuropsychological Tests (continued) 268 Neuropsychological Test Summary - Profile Domain Test Within Border Impaired Normal line Limits General WAIS-R Vocabulary Intellectual WAIS-R Block Design functioning Selective D2 Test of Attention attention Vigilance Psychomotor Vigilance Test Concentration WAIS-R Digit Symbol (& Psychomotor speed) Short-term WAIS-R Digit Span Memory WMS-R Visual Span Verbal Memory CVLT-II Overall Learning (Total) Proactive Interference Long-delay Free Recall Visual Memory (& Complex Figure Test visuo- construction abilities) Planning WISC-III Mazes Sequencing, Trailmaking A mental flexibility Trailmaking B Working Memory Consonant Trigrams Mental flexibility, Stroop Test response inhibition Executive Wisconsin Card Sorting Test function, problem solving Manual dexterity Grooved Pegboard Test administration by: Additional comments: Appendix E - Neuropsychological Tests (continued) 269 General Intellectual Functioning Wechsler Adult Intelligence Scale-Revised (WAIS-R) (Wechsler. 1981). General intellectual functioning was measured by the Vocabulary and Block Design subtests from the WAIS-R. The two subtests show the highest correlation with the verbal and performance IQ scores and an estimated Full-Scale IQ can be derived from the combined score of the two subtests (Silverstein, 1982). The WAIS scores have also been shown to be affected by hypoxemia induced in chronic obstructive pulmonary disease (Prigatano, 1983). Selective Attention d2 Test of Attention (Brickenkamp & Zillmer, 1998). The d2 is a timed test of selective attention, measuring processing speed and concentration performance in discriminating similar visual stimuli. Participants have to pick out target letters, interspersed with distractors in 14 successive trials. Diagnostic utility and construct validity of the test have been demonstrated in both European and American populations (Bates & Lemay, 2004; Brickenkamp & Zillmer). Measures include the total number of items processed, errors, concentration performance (total items of correctly cancelled items minus incorrectly cancelled). Vigilance Psychomotor Vigilance Test (PVT). The PVT was adopted as a measure of sustained attention (Dinges & Powell, 1985). It measures reaction times to visual stimuli presented at a random inter-stimulus interval of 2 to 10 s. It is widely used in sleep studies and has been validated as an effective instrument to measure several important aspects of attention and is sensitive to performance Appendix E - Neuropsychological Tests (continued) 270 deficits induced to sleep deprivation (Belenky etal., 2003; Dinges, 2001; Dinges etal., 1997; Graw etal., 2004; Van Dongen etal., 2003). It also has the advantages of being free from practice effects (no or little learning curve) and cumulative impairments caused by fatigue in performing the task (Dinges & Powell, 1985; Van Dongen et al., 2003). Overall speed is measured by the mean response latency (or its reciprocal) and the median. Mean of the fastest 10% of response times is taken as measure of optimum response capability in a trial (i.e. 10-minute administration). Lapses of attention are measured by the slowest 10% of response times, as well as the built-in measure of lapse, which is preset to be the number of reaction times greater than 500 ms. Minute-by-minute analysis provides a slope showing response slowing or habituation, and indicates intra- trial fluctuation of performance. Concentration and Psychomotor Speed WAIS-R: Digit Symbol (Wechsler, 1981). Participants are given 1.5 minutes to substitute symbols for digits. This task requires visual attention and is found to be the most sensitive WAIS-R subtest to detect various kinds of brain damage (Lezak, 1995). Short-term Memory WAIS-R: Digit Span ((Wechsler, 1981). Participants are instructed to repeat digits presented to them in the same order (Forward) or in reverse order (Backward). The Forward Digit Span is a measure of short-term storage capacity. The Backward Digit Span has a manipulation component in it and thus requires working memory. Appendix E - Neuropsychological Tests (continued) 271 Wechsler Memory Scale (WMS-R): Visual Span (Wechsler, 1987). The Visual Span is a visual analogue of the Digit Span, and offers measures of short-term memory capacity (Forward) and working memory (Backward) in the visual domain. Learning and Memory California Verbal Learning Test - Second Edition (CVLT-II) (Delis et a/., 2000). The CVLT is one of the five most used tests by clinical neuropsychologist in North America (Rabin et a/., 2005). It involves learning a list of 16 items, which belong to one of four semantic categories over five presentation trials. Measures include the learning performance, s well as retention after short delay with a distractor task and a 20-minute long delay. There are also recognition trials and forced-choice items. Computerized scoring with norms is used. Rev-Osterrieth Complex Figure Test (Rev-O) (Osterrieth, 1944; Rey, 1941). The Rey-O has two procedures, namely copy and recall. In copying, participants are instructed to copy a figure as carefully and completely as they can. This is used as a measure of general visuo-spatial perception and constructional skills. The recall procedure is conducted about 30-40 minutes after the initial copy. Participants are asked to draw the figure from memory expectedly. This is used as a measure of incidental visual learning and memory. We adopted the scoring criteria devised by L. Taylor and cited in Spreen and Strauss (Spreen & Strauss, 1998). Appendix E - Neuropsychological Tests (continued) 272 Executive function Wechsler Intelligence Scales for Children (WISC-III): Mazes (Wechsler, 1991). The Mazes subtest of the WISC-III is adopted as a test of planning ability. Although it is a test designed for children, the difficulty level of comparable to other lengthier planning tests. The original norms cover to 15 years 10 months, which give us an estimate of the level of performance of adults (Lezak, 1995). There are also published norms for adults that we adopted in the study (Spreen & Strauss, 1998). Trail Making Test: Parts A and B (TMT) (Reitan & Davison, 1974). The TMT A and B are widely-used tests of complex visual scanning, sequencing, visuomotor tracking, and for Part B only, set-shifting or mental flexibility. Outcome measure is the time of completion of the task. Errors are taken into account by the examiner pointing out errors as they occur so that participants can always complete the task error-free and completion time can be used as the only measure. Age, sex, and education-corrected norms are used (Heaton etal., 1991). Consonant Triqrams (also called Brown-Peterson Technique) (Peterson, 1966; Peterson & Peterson, 1959). The Consonant Trigrams is a divided attention task that requires the participants to retain information while performing a distractor task. We adopted the procedure developed by Edith Kaplan (Lezak, 1995), in which a group of three letters are presented, followed by a number. The participant has to count backward from that number by three's until signaled Appendix E - Neuropsychological Tests (continued) 273 to stop counting and then to report the letters. The retention intervals are 0, 9, 18, and 36 seconds. Aged norms are adopted from Spreen and Strauss (1998). Stroop Color and Word Test (Golden, 1975; Golden, 1978). The Stroop Test measures one's speed of word reading, color naming, and color naming under interference. It consists of three pages of color words, patches of colors, and color words printed in incongruent colored ink (e.g. the word "green" printed in red ink) respectively. Examinees' score is determined by the number of items processed in 45 seconds. In a typical individual, color naming in the third condition is impeded by word reading, which is the more preponderant response. An interference score can be calculated by subtracting the predicted score (i.e. dividing the product of the word reading and color naming score by the sum of the two scores) from the actual colored-word naming score. A positive score indicates resistance to interference and a negative score shows a stronger than predicted interference effect. Wisconsin Card Sorting Test (WCST) (Heaton, 1981). The WCST is a problem- solving test, in which examinees are required to match two decks of cards to four index cards according to strategies they devise on the basis of feedback given by the examiner to their placement of cards. Scores generated include the number of categories completed (ranging from 0 to 6), the number and percentage of perseverative errors, the number and percentage of nonperseverative errors, and failure to maintain set. The percentage of perseverative errors is of particular interests in discriminating patients with frontal lesions and control participants (Lezak, 1995). Norms are available from the manual. Appendix E - Neuropsychological Tests (continued) 274 Psychomotor Speed Grooved Peqboard (Reitan & Davison, 1974). This is a test of manual dexterity. Participants are instructed to place 25 grooved pegs into slotted holes on a board using only one hand, with the dominant first and followed by the non-dominant hand. The outcome measure is time in seconds for each hand. We use the norms by Heaton et al. (1991). Appendix E - Neuropsychological Tests (continued) 275 REFERENCES (for Neuropsychological Tests) Bates, M. E., & Lemay, E. P., Jr. (2004). The d2 test of attention: Construct validity and extensions in scoring techniques. Journal of the International Neuropsychological Society, 10, 394-400. Belenky, G., Wesensten, N. J., Thome, D. R., Thomas, M. L, Sing, H. C, Redmond, D. P., et al. (2003). Patterns of performacne degradation and restoration during sleep restriction and subsequent recovery: A sleep dose-response study. Journal of Sleep Research, 12, 1-12. Brickenkamp, R., & Zillmer, E. (1998). The d2 test of attention. Seattle, Washington: Hogrefe & Huber Publishers. Delis, D. C, Kramer, J. H., Kaplan, E., & Ober, B. A. (2000). California verbal learning test - second edition. Adult version. Manual. San Antonio, TX: Psychological Corporation. Dinges, D. F. (2001). Stress, fatigue, and behavioral energy. Nutrition Reviews, 59(1), S30-32. Dinges, D. F., Pack, F., Williams, K., Gillen, K. A., Powell, J. W., Ott, G. E., et al. (1997). Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4-5 hours per night. Sleep, 20(4), 267-277. Dinges, D. F., & Powell, J. W. (1985). Microcomputer analyses of performance on a portable, simple visual rt task during sustained operations. Behavior Research Methods, Instruments, & Computers, 17(6), 652-655. Golden, C. J. (1975). A group form of the stroop color and word test. Journal of Personality Assessment, 39, 384-386. Golden, C. J. (1978). Stroop Color and Word Test. Chicago, IL: Stoelting. Graw, P., Krauchi, K., Knoblauch, V., Wirz-Justice, A., & Cajochen, C. (2004). Circadian and wake-dependent modulation of fastest and slowest reaction times during the psychomotor vigilance task. Physiology and Behavior, 80, 695-701. Heaton, R. K. (1981). Wisconson Card Sorting Test. Manual. Odessa, FL: Psychoogical Assessment Resources. Heaton, R. K., Grant, I., & Matthews, C. G. (1991). Comprehensive norms for an expanded Halstead-Reitan Battery. Florida: Psychological Assessment Resources. Appendix E - Neuropsychological Tests (continued) 276 Lezak, M. D. (1995). Neuropsychological assessment (3rd e<±). New York: Plenum Press. Osterrieth, P. A. (1944). Le test de copie d'une figure complex: Contribution a I'etude de la perception et de la memoire. Archives de Psychologie, 30, 286-356. Peterson, L. R. (1966). Short-term memory. Scientific American, 215, 90-95. Peterson, L. R., & Peterson, M. J. (1959). Short-term retention of individual verbal items. Journal of Experimental Psychology, 58(3), 193-198. Prigatano, G. P. (1983). Neuropsychological test performance in mildly hypoxemic patients with chronic obstructive pulmonary disease. Rabin, L. A., Barr, W. B., & Burton, L. A. (2005). Assessment practices of clinical neuropsychologists in the united states and canada: A survey of ins, nan, and apa division 40 members. Archives of Clinical Neuropsychology, 20, 33-65. Reitan, R. M., & Davison, L. A. (1974). Clinical neuropsychology: Current status and applications. New York: Wiley. Rey, A. (1941). L'examen psychologique dans less cas d'encephalophathie traumatique. Archives de Psychologie, 28, 286-340. Silverstein, A. B. (1982). Two- and four-subtest short forms of the Wechsler Adult Intelligence Scale-Revised. Journal of Consulting and Clinical Psychology, 50(3), 415-418. Spreen, O., & Strauss, E. (1998). A compendium of neuropsychological tests: Administration, norms, and commentary (2nd ed.). New York: Oxford University Press. Van Dongen, H. P. A., Maislin, G., Mullington, J. M., & Dinges, D. F. (2003). The cumulative cost of additional wakefulness: Dose-response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep, 26(2), 117-126. Wechsler, D. (1981). Manual for the Wechsler Adult Intelligence Scale-Revised. New York: The Psychological Corporation. Wechsler, D. (1987). Wechsler memory scale-revised manual. San Diego, TX: The Psychological Corporation. Wechsler, D. (1991). Wechsler Intelligence Scale for Children (3rd ed.). San Antonio, TX: The Psychological Corporation. APPENDIX F - WORKING MEMORY SPAN Word Span (>2/3 correct trials; + 0.5 for 1 correct trial): Total # Words Correctly Recalled/Admin: I Sentence Verification Span (>2/3 correct trials; + 0.5 for 1 correct trial): Total # Sentences Correctly Verified/Admin: Working Memory Span (mean of last 3 correctly recalled word lists): Word Span: (Total: __/_) Sentence Span: (Total: __/__) WORD Sentence Verification P LAST WORD Working Memory Span P BAG She finished the last boss. 2 She wrote her daughter a card. 1 1 RISK He worries about the next step. 1 Everyone brings the hard sell. 2 LAND She dresses in good style. 1 He conducted himself with grace. 1 2 MOUTH The actor plays the main gin. 2 The coach sang the soccer team. 2 WAGE He poured water in the cope. 2 She gets commission on each sale. 1 3 SAUCE She belongs to the school rice. 2 The student handed in his brain. 2 | 2-wor d # of Trials: 6 BLOCK The moon revolves around the earth. 1 He likes to eat flow. 2 1 HEALTH The sailor stands on the stern. 1 The plant frequently cuts its root. 2 QUEEN That man assassinated the King. 1 She cannot finish her speech. 1 MEAN That deer is a male. 1 He slept through the whole cross. 2 2 DRESS Nobody tells the pond. 2 He was scolded by a ball. 2 COLD The patient lies in wish. 2 The lady combed her foot. 2 BOY The shepherd holds a stick. 1 The cat drives the truck. 2 3 AUNT She did not leave a know. 2 She will get there by fat. 2 STAR The brother cleaned with a brush. 1 The children play in the snow. 1 | 3-wor d # of Trials: 15 Appendix F - Working Memory Span (continued) CAMP The brick is made of smoke. 2 Listen carefully with your throat. 2 GIRL They left the countryside in droves. 1 These vegetables are grown from seed. 1 1 SHELL The clouds swam in the pool. 2 He wrote his family a mate. 2 BOMB The girl eagerly open that join. 2 She changed the hairdryer's plug. 1 LOCK The mattress is made of foam. 1 He worries about his height. 1 BEEF She smells the aim. 2 The dog is chasing the fort. 2 2 THREAT He throws the die. 1 The tank is filled with main. 2 CHILD She finds him a face. 2 There are many types of beer. 1 BEACH He cannot hear the cent. 2 He enjoyed the parachute jump. 1 TALE The storekeeper carefully packed the smile. 2 He confessed his sin. 1 3 FOOL My favorite season is the fall. 1 The boy wears a year. 2 STAY They took picture on the green hand. 2 Please tighten your seat belt. 1 1 4-wor d # of Trials: 27 WILD This sandwich is made of group. 2 The fridge made a dive. 2 RISE He disproved the myth. 1 He covered the mud with cape. 1 1 CRIME They follow the teacher's tough. 2 The sky is covered in mud. 2 5-wor d | SWEAT That vase has a special staff. 2 He has no room for his guest. 1 VEIN Her speech pacified the crowd. 1 The man puffed on his pipe. 1 Appendix F - Working Memory Span (continued) SWIFT The babies gathered in the tax. 2 The lake soon came into view. 1 ROUGH The lion has great wheel. 2 The crying baby woke the house. 1 2 GOD The mayor visits the power plant. 1 The throne crowns the prince. 2 LOVE The boy jumps the hope. 2 The books marched to the front 2 TOWN The agent is processing a claim. 1 The poor families are in want. 1 PILE The shirt prepared the brief. 2 His clothes are covered in dirt. 1 ACT 1 skipped lunch. 1 Detectives carried out a thorough search. 1 3 COAT The dress is made of proof. 2 She combs her hair with hide. 2 SHIP The parents stopped the fight. 1 She believes there is a cure. 1 JOY She rushed into her room. 1 She cannot have peace of mind. 1 # of Trials: 42 TAIL They found the wounded bear. 1 Fire is beaten up by gang. 2 NONE He greased the pan with oil. 1 She likes watching the rising sun. CHANCE His speech has no net. 2 He won't finish at this rate. 1 Yesterday they decorated the Christmas BEAT I only like the least. 2 6-wor d | tree. SOUTH We frequently play golf. 1 The plane headed to the west. REAR That is a challenging task. 1 I was the only person on deck. Appendix F - Working Memory Span (continued) COURT Ice gives out cold heat. 2 He cut meat with a miss. 2 TYPE He is satisfied with a peace. 2 The town is covered in fog. 1 HEN The tree needs a trim. 1 He is excited to meet steel. 2 2 DATE The prisoners hate the guard. 1 She mixed the flour with tape. 2 MASS The car came to a stop. 1 I like silver more than gold. END The son paid for the bill. 1 We talked along the coast. CASE The meat weighs one march. 2 I wonder which way is east. KNIFE The maid carefully ironed the youth. 2 She is always dressed in blue. SLAVE She wears a gold farm. 2 The students greeted the harm. 2 3 BLONDE The elephant stepped on the law. 2 The ship has vanished without trace. POPE 1 try to make more curve. 2 It's time to swallow your pride. DANCE She ran to answer the door. 1 She washes her face with shoot. 2 # of Trials: 60 FAIL The father gave him a trend. 2 She puts her money in mile. 2 BLIND The climber hung on the edge. 1 This dish has an unusual taste. 1 LAUGH The baby cannot finish the milk. 1 She showers in the grade. 2 1 PLAN The party is in the yard. 1 There's the smell of rotting flesh. 1 7-wor d I RACE She yearns for her parents' fill. 2 What time does the hair close? 2 STORE He fell off the horse. 1 They walked on the steep plain. 2 CREW The soldier lost some safe. 2 He poured water into the work. 2 Appendix F - Working Memory Span (continued) HAT Cover the fish with ring. 2 He's happy to join the firm. SUITE He measures the dress's length. 1 I'd like to have a drink. PICK The worker finished the difficult part. 1 He felt better after a cry. 2 ROOF The dog recognizes its owner's bid. 2 We rested under the coconut palm. ICE Her voice shows anger and form. 2 He brushes his teeth with still. 2 TRY The sauce gave a good talk. 2 She finally repaid her student loan. DEAL We finished the first phase. 1 They celebrated the final win. FLASH The girl laughed with the scale. 2 He escaped death by an inch. LEFT The election needs a second count. 1 The cook baked the light. 2 PORT She shows a look of lane. 2 She stuffed the cushion with bit. 2 3 COW The children swim in the truth. 2 He listens with a square. 2 SPOKE She is a bright kid. 1 Payments can be made in cash. 1 BANK That is a scenic trail. 1 She boiled the wall. 2 WOOD The customer chats with the shape. 2 He added salt into the plot. 2 # of Trials: 81 Appendix F - Working Memory Span (continued) TREAT He often doubts his own strength. 1 I broke my leg. 1 DRUG Laughter and applause followed that joke. 1 She put makeup on the gift. 2 FRAME Someone has abandoned that little dog. 1 She likes dipping fruits in coal. 2 SIGN She drank a glass of role. 2 He turned off her nose. 2 1 GRAVE She quickly filled out the hurt. 2 His life is at stake. 1 8-wor d | VOICE The committee changed the rule. 1 He broke the record of while. 2 AID She put her career on hold. 1 The boat flows into the pale. 2 KNEE He replaced the left front might. 2 The pen doesn't like the feel. 2 PLAY He knows how to enjoy life. 1 He had to pay a fine. 1 BOND He finished another coat of paint. 1 The audience enjoyed the show. 1 GLASS I ordered soup and a roll. 1 He likes playing with the lack. 2 TONGUE The chair broke at the box. 2 That jacket was a good buy. 1 2 COOL The child is shaking in fear. 1 Love is a terrible vice. 2 SHORT She barely passed that calculus course. 1 She talks to the rush. 2 LOSS I try to achieve the goal. 1 He reached for the medicine chest. 1 RED The dog was gnawing a bone. 1 The little girl nodded her head. 1 Appendix F - Working Memory Span (continued) THROW The priest was wearing a sink. 2 They hugged the park. 2 REACH She reports daily to her piece. 2 The balloons were taught the craft. 2 HEART He ate three bowls of band. 2 The man honked his car horn. 1 PAST She whispered something in his ear. 1 It was worth the wait. 1 3 LONG The story is just a lie. 1 The fishermen start work at dawn. 1 SAVE She is in the tip. 2 They met at the book fair. 1 FLOOR She is swimming in the moon. 2 The shoes took a bath. 2 CLAY 1 have a really busy term. 1 He works at the gold mine. 1 # of Trials: 105 LAST The horse has a white mark. 1 He's trying to deny the fact. 1 DEAR He gave that matter careful thought. 1 Flowers bloom in the spring. 1 EYE The boy dropped one shame. 2 She gladly accepted the wedding gun. 2 KEEP It was damaged by the storm. 1 It is pouring with rain. 1 1 NECK Being famous is her bed. 2 He bought a green phrase. 2 9-wor d 1 WRONG He smells with his hand. 2 The band played mold. 2 GRASS This act contradicts the original soap. 2 Today we'll have a fire drill. 1 MOOD A golden watch wore the chief. 2 The gardener dug a deep hole. 1 FORCE They drilled into the rock. 1 Bananas are sold by weight. 1 Appendix F - Working Memory Span (continued) DRAW 1 sent the letter by post. 1 The pig asked for more cream. 2 BIRD He prefers giving donations in kind. 1 The phone has a leg wound. 2 SEAT She chose her words with care. 1 She left for Berlin by plane. 1 WEEK 1 will not betray your trust. 1 He drives a flying train. 2 2 RENT Let's go for a walk. 1 The newborn unlocked the gate. 2 ARC She pulled into the slow hate. 2 I have finished the first page. 1 JUDGE She made a mental note. 1 The watch listens to the arm. 2 GUILT The gardener cut off that branch. 1 I need to get some sleep. 1 TOOL Their marriage is under great strain. 1 Rivers are built across the bridge. 2 PRIZE The cup repeated its point. 2 We stopped for a well-earned rest. 1 SCENE The prisoner returned to his cell. 1 He fries pan in the meat. 2 WASH Tears ran down his bore. 2 He stepped on the gas. 1 GROUND They have given up line. 2 She said that out of spite. 1 3 CAR They keep animals in the barn. 1 Her nails bit her tooth. 2 QUICK The car emits black wealth. 2 The soap took a break. 2 TIE The claim is capable of salt. 2 The ant was dressed in white. 2 EASE The house has taken the lead. 2 She moved into the family home. 1 MAN The rock waited for a pause. 2 She is my best friend. 1 # of Trials: 132 Appendix F - Working Memory Span (continued) BOARD He has contributed the lean. 2 The tiger killed the loop. 2 FUN She has been released from jail. 1 He declined his daughter's help. 1 DESK I don't like this horror film. 1 The divers cooked the wave. 2 LIFT He swam across the bay. 1 Everyone was sad about the news. 1 CORPS No one likes to pay hall. 2 He just checked the TV guide. 1 1 THEME She always wears a cost. 2 The cups administer the test. 2 AIR The protesters went on a smell. 2 Winners get into the final round. 1 SHOT She kept valuables in the blood. 2 The captain sailed around the world. 1 JACK These flowers are past their prime. 1 The soldier shot with a month. 2 NEED She gracefully put on the foil. 2 The boy made a birthday core. 2 GAIN Mistakes gave him a bad name. 1 Wipe your eyes with a jet. 2 10-wor d \ LOOK Fire releases fierce meet. 2 The driver found an alternative route. 1 PORCH The negotiators tried to make pass. 2 I can hear some food. 2 DAY She is married to a lord. 1 The play is written in verse. 1 RIGHT He planted seeds in the soil. 1 Grandmother is making water with tea. 2 2 TENT He traveled everywhere in his meal. 2 I prefer black to gray. 1 BIRTH We should prevent unnecessary waste. 1 Air is impressed by his charm. 2 MAIL He owns a 200-acre chain. 2 The exam put on a skirt. 2 NEST She belongs to the tennis club. 1 He struck the last match. 1 HAIR The singer has set a pat. 2 She bought a new address book. 1 Appendix F - Working Memory Span (continued) CLOTH The book lost its job. 2 A bear chose the cast. 2 LUCK Desks have good sense of smell. 2 the table took a deep breath. 2 SCHOOL These tires have a good grip. 1 A rabbit moved the hill. 2 TOP Animals are bred on the ranch. 1 The flower forgets its age. 2 DREAM He lost control on a space. 2 The trees walk on the shore. 2 3 WEAR She kept looking at her watch. 1 The bath made a good guess. 2 ART They always hunt by night. 1 Lions feed on their kill. 1 BENCH They entertain on a large dust. 2 The grass initiates the move. 2 HALF The guests have eaten their touch. 2 She catches nets with fish. 2 FEET The boy waited for his turn. 1 Stir the spoon with a pot. L. # of Trials: 162